v>EPA
United States
Environmental Protection
Agency
Municipal Environmental Research EPA-600/2-80-028
Laboratory March 1980
Cincinnati OH 45268 C » f
Research and Development
Water Treatment
Process
Modifications for
Trihalomethane
Control and Organic
Substances in the
Ohio River
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-80-028
March 1980
WATER TREATMENT PROCESS MODIFICATIONS
FOR TRIHALOMETHANE CONTROL
AND ORGANIC SUBSTANCES IN THE OHIO RIVER
By
Ohio River Valley Water Sanitation Commission
Cincinnati, Ohio 45202
Grant No. R-804615
Project Officers
Walter A. Feige
Jack DeMarco
Physical and Chemical Contaminants Removal Branch
Drinking Water Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U. S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U. S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
ii
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health and
welfare of the American people. Noxious air, foul water, and spoiled land are
tragic testimony to the deterioration of our natural environment. The com-
plexity of that environment and the interplay between its components require
a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem solution
and it involves defining the problem, measuring its impact, and searching for
solutions. The Municipal Environmental Research Laboratory develops new and
improved technology and systems for the prevention, treatment, and management
of wastewater and solid and hazardous waste pollutant discharges from munici-
pal and community sources, for the preservation and treatment of public drink-
ing water supplies, and to minimize the adverse economic, social, health, and
aesthetic effects of pollution. This publication is one of the products of
that research; a most vital communications link between the researcher and the
user community.
This report describes the results of studies to evaluate several treat-
ment modifications for the control of trihalomethane levels at seven water
supply utilities in the Ohio River Valley. Examination of within-plant and
finished waters was made to ensure bacteriological integrity. In addition,
the levels of trihalomethanes and other selected organic compounds were deter-
mined in the raw and finished water of eleven utilities for a one-year period.
Francis T. Mayo, Director
Municipal Environmental Research Laboratory
iii
-------
ABSTRACT
Plant-scale studies at seven water utilities using the Ohio, Allegheny,
Beaver, and Monongahela Rivers as their source of supply evaluated various
water treatment process modifications for both the control of trihalomethane
levels and the modifications' impact on bacteriological quality of the fin-
ished water. Process modifications studied, based on comprehensive organic
analysis, included relocation of the chlorine application point, chlorina-
tion/ammoniation, partial or complete substitution of chlorine dioxide for
chlorine, and placement of four different types of virgin granular activated
carbons in filter beds. Supplemental studies included organic analysis of
monthly raw and finished water samples collected for a one-year period at
each of 11 participating water utilities. In addition to providing plant
facilities and personnel, the 11 utilities joined USEPA in funding this pro-
ject, which was conducted by the Ohio River Valley Water Sanitation
Commission.
This report was prepared in fulfillment of USEPA Grant R-804615 for pro-
ject activities for the period October 1976 to August 1979.
iv
-------
CONTENTS
Foreword iii
Abstract iv
Figures . vii
Tables ix
Acknowledgements. xiv
1. Introduction. . . 1
2. Conclusions 5
3. Areas for Further Study 10
4. Project Organic Compounds 11
5. Analytical Procedures and Quality Assurance ,v 15
Organic Contract Laboratory . 15
General Laboratory Controls .. 15
Analytical Procedure for Purgeable Halocarbons 15
Quality Assurance for Purgeable Halocarbons 19
Analytical Procedure for Base-Neutral Extractable
Compounds 22
Quality Assurance for Base-Neutral Extractable Compounds. 30
Attempted Analysis of Base-Neutral Extractable
Nitrogen-Containing Hydrocarbons 34
Mass Spectrometer Analytical Procedures 34
Utility Laboratories 35
6. Trihalomethane Treatability Studies 37
General 37
The Effect of Chlorine Application Points
on Trihalomethane Formation 38
Pittsburgh Department of Water 39
Cincinnati Water Works. 45
Wheeling Water Department 50
The Effect of Ammoniation on Trihalomethane Formation .... 55
Louisville Water Company 56
The Effect of Chlorine Dioxide on Trihalomethane Formation. . 63
Western Pennsylvania Water Company 64
The Effect of Granular Activated Carbon
Adsorption/Filtration on Trihalomethane Control ..... 75
Huntington Water Corporation 75
Beaver Falls Authority 90
v
-------
CONTENTS (Continued)
Conclusions from Trihalomethane Treatability Studies 107
7. Organic Compound Survey 109
General 109
Survey for Purgeable Halocarbons 109
Survey for Base-Neutral Extractable Halocarbons 159
Survey for Base-Neutral Extractable Non-Halogenated
Hydrocarbons 206
Organic Compounds Not Designated as Priority Pollutants ... 214
References 219
Appendices 221
A. General Organic Laboratory Procedures 221
B. Equipment and Analytical Procedures for Purgeable Halocarbon
Priority Pollutants 224
C. Quality Assurance Data for Purgeable Halocarbons 226
D. Equipment and Analytical Procedures for Base-Neutral Extractable
Hydrocarbons 253
E. Quality Assurance Data for Extractable Halocarbons 255
F. Quality Assurance Data for Non-Halogenated Extractable
Hydrocarbons 276
G. Solvent Impurities and Halogenated By-Products of
Solvent Impurities 280
H. Attempted Anlaysis of Base-Neutral Extractable Organo-Nitrogen
Compounds 281
I. Mass Spectrometry Equipment and Analytical Procedures 285
J. Organic Sampling Procedures 286
K. Procedure and Medium Formula for a Membrane Filter-Standard
Plate Count 289
vi
-------
FIGURES
Number Page
1 Utility locations . 2
2 Graphical representation of trihalomethane parameters 12
3 Typical gas chromatogram of purgeable halocarbon Priority
Pollutants calibration standard using Hall detector 17
4 Typical gas chromatogram of purgeable system
blank using Hall detector 18
5 Typical gas chromatogram of purgeable sample using Hall detector. . . 18
6 Typical gas chromatogram of base-neutral extractable halogenated
Priority Pollutants calibration standard using Hall detector. . . 23
7 Typical gas chromatogram of base-neutral extractable
solvent blank using Hall detector ..... 24
8 Typical gas chromatogram of base-neutral extractable
sample using Hall detector ..25
9 Typical gas chromatogram of base-neutral extractable
Priority Pollutants calibration standard using flame
ionization detector 27
10 Typical gas chromatogram of base-neutral extractable
solvent blank using flame ionization detector 28
11 Typical gas chromatogram of base-neutral extractable
sample using flame ionization detector 29
12 Treatment at Pittsburgh Department of Water 40
13 Trihalomethane formation at Pittsburgh Department of Water .41
14 Treatment at Cincinnati Water Works 46
15 Trihalomethane formation at Cincinnati Water Works 47
16 Treatment at Wheeling Water Department 52
17 Trihalomethane formation at Wheeling Water Department 53
18 Treatment at Louisville Water Company 57
19 Trihalomethane formation at Louisville Water Company 58
20 Effect of pH on trihalomethane formation 59
21 Trihalomethane formation at Louisville Water Company 60
22 Trihalomethane formation at Louisville Water Company 62
23 Treatment at Western Pennsylvania Water Company 65
24 Chlorine dioxide generation -at
Western Pennsylvania Water Company 65
25 Trihalomethane formation at Western Pennsylvania Water Company. ... 66
26 Trihalomethane formation at Western Pennsylvania Water Company. ... 69
27 Trihalomethane formation at Western Pennsylvania Water Company. ... 70
28 Trihalomethane formation at Western Pennsylvania Water Company. ... 73
29 Treatment at Huntington Water Corporation 77
30 Trihalomethane formation at Huntington Water Corporation 79
vii
-------
FIGURES (Continued)
Number
31 Trihalomethane removal by granular activated carbon
at Huntington Water Corporation. . 80
32 Trihalomethane removal by granular activated carbon
at Huntington Water Corporation 81
33 Treatment at Beaver Falls Authority 91
34 Trihalomethane removal by granular activated carbon
at Beaver Falls Authority .95
35 Trihalomethane removal by granular activated carbon
at Beaver Falls Authority 96
36 Trihalomethane removal by granular activated carbon
at Beaver Falls Authority 97
37 Trihalomethane removal by granular activated carbon
at Beaver Falls Authority .98
38 Trihalomethane removal by granular activated carbon
at Beaver Falls Authority 99
39 Trihalomethane removal by granular activated carbon
at Beaver Falls Authority .100
40 Treatment at West View Water Authority 110
41 Treatment at Evansville Water Department .111
42 Treatment at Fox Chapel Authority 112
43 Treatment at Wilkinsburg-Penn Joint Water Authority 112
44 Raw water THMFP variation 146
C-l Precision of instantaneous chloroform data 228
C-2 Precision of instantaneous chloroform data 229
C-3 Precision of terminal chloroform data 230
C-4 Precision of instantaneous bromodichloromethane data 232
C-5 Precision of terminal bromodichloromethane data 233
C-6 Precision of instantaneous dibromochloromethane data 235
C-7 Precision of terminal dibromochloromethane data. 236
C-8 Precision of instantaneous carbon tetrachloride data 239
C-9 Precision of instantaneous bromoform data 239
C-10 Precision of terminal bromoform data 240
C-ll Precision of instantaneous total trihalomethane data ....... .251
C-12 Precision of terminal total trihalomethane data 252
H-l Typical gas chromatogram of base-neutral extractable
Priority Pollutants calibration standard using
alkali flame ionization detector .283
H-2 Typical gas chromatogram of base-neutral extractable
sample using alkali flame ionization detector .284
viii
-------
TABLES
Number Page
1 Project Organic Compounds, Purgeable Halocarbons, GC/Hall Detector . 13
2 Project Organic Compounds, Base-Neutral Extractable
Halocarbons, GC/Hall Detector. ..... 13
3 Project Organic Compounds, Base-Neutral Extractable
Halocarbons, GC/Flame lonization Detector 13
4 Purgeable Halocarbons, GC/Hall Detector 16
5 Halogenated Base-Neutral Extractable Priority Pollutants,
GC/Hall Detector and 3,000 Concentration Factor 26
6 Non-Halogenated Base-Neutral Extractable Priority Pollutants,
GC/Flame lonization Detector and 3,000 Concentration Factor. . . 30
7 Ratio of Individual Trihalomethanes to Total Trihalomethanes in
the Clear Well, Pittsburgh Department of Water ......... 43
8 Tetrachloroethylene Concentrations, Pittsburgh Department of Water . 43
9 TTHM Concentrations, Pittsburgh Department of Water 44
10 Ratio of Individual Trihalomethanes to Total Trihalomethanes
in the Clear Well, Cincinnati Water Works 49
11 Ratio of Individual Trihalomethanes to Total Trihalomethanes
in the Clear Well, Wheeling Water Department 54
12 Ratio of Individual Trihalomethanes to Total Trihalomethanes
in the Clear Well, Louisville Water Company. . 63
13 Terminal TTHM Concentrations at Western Pennsylvania Water Company . 67
14 Ratio of Individual Trihalomethanes to Total Trihalomethanes
in the Clear Well, Western Pennsylvania Water Company 72
15 Water Quality Data at Huntington Water Corporation 78
16 Removal of Trihalomethanes by Granular Activated Carbon
at Huntington Water Corporation 82
17 Removal of Bromoform by Granular Activated Carbon
at Huntington Water Corporation 84
18 Removal of Carbon Tetrachloride by Virgin Granular
Activated Carbon at Huntington Water Corporation 85
19 Removal of 1,4-Dichlorobenzene by Virgin Granular
Activated Carbon at Huntington Water Corporation 86
20 Removal of an Unidentified Base-Neutral Extractable Halocarbon by
Virgin Granular Activated Carbon at Huntington Water Corporation 86
21 Removal of Unidentified Base-Neutral Extractable Halocarbons by
Virgin Granular Activated Carbon at Huntington Water Corporation 87
22 Removal of Carbon Tetrachloride by Older Granular
Activated Carbon at Huntington Water Corporation . . 88
23 Hydraulic Data at Beaver Falls Authority 91
24 Water Quality Data at Beaver Falls Authority 92
25 Water Quality Data at Beaver Falls Authority 93
ix
-------
TABLES (Continued)
Number
26 Carbon Tetrachloride Data at Beaver Falls Authority 101
27 Removal of 1,4-Dichlorobenzene by Virgin Granular
Activated Carbons at Beaver Falls Authority 102
28 Removal of an Unidentified Base-Neutral Extractable Halocarbon
by Granular Activated Carbon at Beaver Falls Authority 103
29 Removal of an Unidentified Base-Neutral Extractable Halocarbon
by Granular Activated Carbon at Beaver Falls Authority 104
30 Water Quality Data at Beaver Falls Authority . . 105
31 Chlorobenzene Levels at Louisville Water Company 117
32 Raw Water Chloroform Data 120
33 Finished Water Chloroform Data 121
34 Finished Water Chloroform Levels 122
35 Raw Water Bromodichloromethane Data 123
36 Finished Water Bromodichloromethane Data 124
37 Finished Water Bromodichloromethane Levels 125
38 Raw Water Dibromochloromethane Data 126
39 Finished Water Dibromochloromethane Data 127
40 Finished Water Dibromochloromethane Levels 128
41 Raw Water Bromoform Data 129
42 Finished Water Bromoform Data. .. 130
43 Finished Water Bromoform Levels .131
44 Raw Water Dichloroiodomethane Data 132
45 Finished Water Dichloroiodomethane Data 133
46 Finished Water Dichloroiodomethane Levels 134
47 Finished Water Total Trihalomethane Levels 135
48 Trihalomethane Formation Potential Data
for Huntington Water Corporation 136
49 Trihalomethane Formation Potential Data
for Fox Chapel Authority 137
50 Trihalomethane Formation Potential Data
for Wilkinsburg-Penn Joint Water Authority 138
51 Trihalomethane Formation Potential Data
for Pittsburgh Department of Water 139
52 Trihalomethane Formation Potential Data
for Western Pennsylvania Water Company . . .140
53 Trihalomethane Formation Potential Data
for Beaver Falls Authority 141
54 Trihalomethane Formation Potential Data
for Wheeling Water Department . .142
55 Trihalomethane Formation Potential Data
for Cincinnati Water Works ..... ..... .143
56 Trihalomethane Formation Potential Data
for Louisville Water Company ........ .144
57 Trihalomethane Formation Potential Data
for Evansville Water Department. . 145
58 Raw Water Carbon Tetrachloride Data 147
59 Finished Water Carbon Tetrachloride Data 148
-------
TABLES (Continued)
Number Page
60 Raw Water Chlorobenzene data 149
61 Finished Water Chlorobenzene Data 150
62 Raw Water 1,1-Dichloroethane Data 151
63 Finished Water 1,1-Dichloroethane Data ... 152
64 Raw Water 1,2-Dichloroethane Data 153
65 Finished Water 1,2-Dichloroethane Data 154
66 Raw Water 1,2-Dichloropropane Data 155
67 Finished Water 1,2-Dichloropropane Data 156
68 Raw Water trans-l,3-Dichloropropene Data 157
69 Finished Water trans-l,3-Dichloropropene Data 158
70 Raw Water 1,4-Dichlorobenzene Data 166
71 Finished Water 1,4-Dichlorobenzene Data 167
72 Raw Water 1,3-Dichlorobenzene Data , 168
73 Finished Water 1,3-Dichlorobenzene Data. .... .169
74 Raw Water Data for 1,2-Dichlorobenzene and/or Hexachloroethane . . .170
75 Finished Water Data for
1,2-Dichlorobenzene and/or Hexachloroethane. . 171
76 Raw Water Data for
1,2,4-Trichlorobenzene and/or Hexachlorobutadiene 172
77 Finished Water Data for
1,2,4-Trichlorobenzene and/or Hexachlorobutadiene 173
78 Raw Water Data for
bis(2-Chloroethyl) Ether and/or bis(2-Chloroisopropyl) Ether . .174
79 Finished Water Data for
bis(2-Chloroethyl) Ether and/or bis(2-Chloroisopropyl) Ether . .175
80 Raw Water bis(2-Chloroethoxy) Methane Data 176
81 Finished Water bis(2-Chloroethoxy) Methane Data 177
82 Raw Water Hexachlorocyclopentadiene Data 178
83 Finished Water Hexachlorocyclopentadiene Data 179
84 Raw Water 2-Chloronaphthalene Data 180
85 Finished Water 2-Chloronaphthalene Data 181
86 Raw Water 4-Chlorophenyl Phenyl Ether Data 182
87 Finished Water 4-Chlorophenyl Phenyl Ether Data 183
88 Raw Water Data for 4-Bromophenyl Phenyl Ether and/or a-BHC 184
89 Finished Water Data for 4-Bromophenyl Phenyl Ether and/or a-BHC. . .185
90 Raw Water Data for Jf-BHC (Lindane) and/or S-BHC. . .186
91 Finished Water Data for *-BHC (Lindane) and/or S-BHC 187
92 Raw Water Data for Heptachlor and/or p-BHC 188
93 Finished Water Data for Heptachlor and/or p-BHC. . 189
94 Raw Water Aldrin Data. . . ; 190
95 Finished Water Aldrin Data 191
96 Raw Water Heptachlor Epoxide Data 192
97 Finished Water Heptachlor Epoxide Data 193
98 Raw Water a-Endosulfan Data .194
99 Finished Water a-Endosulfan Data . .195
100 Raw Water DDT Data 196
101 Finished Water DDT Data 197
xi
-------
TABLES (Continued)
Number Page
102 Raw Water Data for Dieldrin and/or DDE 198
103 Finished Water Data for Dieldrin and/or DDE '.'.'. '.199
104 Raw Water Endrin Data 200
105 Finished Water Endrin Data '.201
106 Raw Water Data for ODD and/or 3-endosulfan 202
107 Finished Water Data for ODD and/or 3-endosulfan. '. .203
108 Raw Water Methoxychlor Data 204
109 Finished Water Methoxychlor Data 205
110 GC/MS-SIM Confirmation of Polynuclear Aromatic Hydrocarbons 209
111 GC/MS-SIM Confirmation of Polynuclear Aromatic Hydrocarbons. . . . .210
112 GC/MS-SIM Confirmation of Polynuclear Aromatic Hydrocarbons
at Western Pennsylvania Water Company 211
113 GC/MS-SIM Confirmation of Polynuclear Aromatic Hydrocarbons
at Huntington Water Corporation. . .v 212
114 GC/MS-SIM Confirmation of Polynuclear Aromatic Hydrocarbons
at Beaver Falls Authority 213
115 Unidentified Purgeable Halocarbon Data
at Western Pennsylvania Water Company 215
116 Unidentified Base-Neutral Extractable Halocarbon Data 216
117 Unidentified Base-Neutral Extractable Halocarbon Data 217
118 Unidentified Base-Neutral Extractable Halocarbon Data. ...... .218
C-l Significance of Chloroform Data. . . 227
C-2 Significance of Bromodichloromethane Data .231
C-3 Significance of Data for Dibromochloromethane and/or
cis-l,3-Dichloropropene and/or 1,1,2-Trichloroethane 234
C-4 Significance of Bromoform Data 237
C-5 Significance of Carbon Tetrachloride Data. 238
C-6 Significance of Dichloroiodomethane Data 241
C-7 Significance of Chlorobenzene Data 242
C-8 Significance of 1,1-Dichloroethane Data 243
C-9 Significance of 1,2-Dichloroethane Data 244
C-10 Significance of 1,2-Dichloroethane Data 245
C-ll Significance of 1,2-Dichloropropane Data 246
C-12 Significance of trans-l,3-Dichloropropene Data 247
C-13 Significance of 1,1,1-Trichloroethane Data 248
C-14 Significance of Trichloroethylene Data 249
C-15 Significance of Data for
1,1,2,2-Tetrachloroethane and/or Tetrachloroethylene 250
E-l Significance of 1,4-Dichlorobenzene Data 256
E-2 Significance of 1,3-Dichlorobenzene Data 257
E-3 Significance of 1,2-Dichlorobenzene and/or Hexachloroethane Data . .258
E-4 Significance of
1,2,4-Trichlorobenzene and/or Hexachlorobutadiene Data 259
E-5 Significance of bis(2-Chloroisopropyl) Ether and/or
bis(2-Chloroethyl) Ether Data 260
E-6 Significance of bis(2-Chloroethoxy) Methane Data 261
E-7 Significance of Hexachlorocyclopentadiene Data . 262
xii
-------
TABLES (Continued)
Number
E-8 Significance of 2-Chloronaphthalene Data 263
E-9 Significance of 4-Chlorophenyl Phenyl Ether Data 264
E-10 Significance of 4-Bromophenyl Phenyl Ether and/or a-BHC Data . . . .265
E-ll Significance of 5T-BHC (Lindane) and/or S-BHC Data 266
E-12 Significance of Heptachlor and/or P-BHC Data ... .267
E-13 Significance of Aldrin Data. . 268
E-14 Significance of Heptachlor Epoxide Data 269
E-15 Significance of a-Endosulfan Data. 270
E-16 Significance of DDT Data . . 271
E-17 Significance of Dieldrin and DDE Data. . .272
E-18 Significance of Endrin Data 273
E-19 Significance of ODD and p-Endosulfan Data 274
E-20 Significance of Methoxychlor Data. . 275
F-l Extraction Recoveries of Non-Halogenated Base-Neutral Standards. . .277
F-2 Reproducibility of Non-Halogenated Base-Neutral Standards 278
F-3 Reproducibility of Non-Halogenated Base-Neutral Standards 279
H-l Extraction Recoveries and Detection Levels of Nitrogen
Containing Base-Neutral Compounds. 281
xiii
-------
ACKNOWLEDGMENTS
The Ohio River Valley Water Sanitation Commission is especially apprecia-
tive of the efforts of the superintendents, directors and managers of the par-
ticipating water utilities, who assisted in development, financing and conduct
of the project. Special thanks is accorded the water utility personnel, who
devoted many hours to the operation of the project.
Fox Chapel Authority
Melvin Hook; Reginald Adams and Thomas Stehle
(Reginald Adams Laboratory, Pittsburgh, Pennsylvania)
Wilkinsburg-Penn Joint Water Authority
Harold McFarland, Dennis Beck
Pittsburgh Department of Water
John Miller, John Beck and staff
Western Pennsylvania Water Company
William Neuman, Michael Burns and staff
West View Water Authority
Joseph Dinkel
Beaver Falls Authority
Frank Richter and staff
Wheeling Water Department
Albert Campbell and staff
Huntington Water Corporation
Thomas Holbrook and staff
Cincinnati Water Works
Richard Miller, Edward Kispert and staff
Louisville Water Company
Frank Campbell, Don Duke and staff
Evansville Water Works
Mahlon Henderson and Matthew Rexing
The Commission gratefully acknowledges the participation and contribu-
tions which led to the successful completion of this project. The project
staff was responsible for conducting the project:
Robert J. Boes, Project Director
Richard J. Miltner, Principal Investigator (November 1977 - April 1979)
, Project Engineer (November 1976 - October 1977)
Bill G. Razor, Principal Investigator (November 1976 - October 1977)
Bonnie Barger Cummins, Project Scientist
Sarah B. Dirr, Project Secretary
Robert C. Kroner, Consultant
The comprehensive task of preparing the initial draft report was carried out
xiv
-------
by Richard J. Miltner and Bonnie Barger Cummins under the general direction of
the Project Director, Robert J. Boes.
The cooperation of Radian Corporation of Austin, Texas, contributed to
the success of the project. Individuals who merit special acknowledgment are
Dr. Donald Rosebrodk and Dr. Lawrence Keith; Dr. Kenneth Lee,( David Present
and the gas chromatography staff; and Dr. Robert Spraggins and the mass spec-
trometry staff.
Guidance in the start-up phase of the project was provided by the
Steering Committee:
F. T. Bess, Union Carbide, ORSANCO Chemical Industry Committee
Don T. Duke, Louisville Water Company, ORSANCO Water Users Committee
James Erb, Pennsylvania Department of Environmental Resources,
Public Water Supply Agencies of Commission member states
Michael J. Taras, American Water Works Association Research Foundation
Jim Finger, USEPA Region TV
Edward C. Kispert, Cincinnati Water Works, ORSANCO Water Users Committee
Dr. Pasquale Scarpino, University of Cincinnati
Jack DeMarco, USEPA, Municipal Environmental Research Laboratory
The USEPA in Cincinnati provided technical assistance:
Dr. Harry D. Nash and Alan Stevens, Municipal Environmental Research
Laboratory
Dr. Herbert Brass, Office of Drinking Water
xv
-------
-------
SECTION 1
INTRODUCTION
BACKGROUND
In 1974 and 1975, surveys and studies reported the identification of tri-
halomethanes and other organic compounds in the public drinking water supplies
in the Ohio River Valley and nationwide.^~^ Some compounds were present in
rivers that were the water sources for water utilities, and trihalomethanes
and other compounds were formed during the water treatment process.
Because of increasing concern about these organic compounds, the Ohio
River Valley Water Sanitation Commission (ORSANCO) and its Water Users
Committee, representatives of public and industrial water supply systems using
the Ohio River and major tributaries as their source, developed a cooperative
project to evaluate treatment process modifications for the control of tri-
halomethanes and analyze the utilities' raw and finished waters for organic
substances. The project established a program to be operated by the
Commission with the assistance of eleven water utilities, who pledged both
financial support and use of their water treatment facilities and personnel.
The U. S. Environmental Protection Agency (USEPA) awarded the Commission a
research grant for the project in October 1976.
PARTICIPATING UTILITIES
The project utilities (Figure 1) were:
Evansville Water Department, Indiana
Louisville Water Company, Kentucky
Cincinnati Water Works, Ohio
Huntington Water Corporation, West Virginia
Wheeling Water Department, West Virginia
Beaver Falls Authority, Pennsylvania
Municipal Authority of the Borough of West View, Pennsylvania
Western Pennsylvania Water Company, Pennsylvania
Pittsburgh Department of Water, Pennsylvania
Wilkinsburg-Penn Joint Water Authority, Pennsylvania
Fox Chapel Authority, Pennsylvania
OBJECTIVES
The first of two major objectives was the investigation and evaluation of
modifications of water treatment practices for the control of trihalomethanes.
-------
FOX CHAPEL
WILKINSBURG
BEAVER
FALLS
PITTSBURGH
WHEELING
OHIO
CINCINNATI
WEST
VIRGINIA
INDIANA
EVANSVILLE
LOUISVILLE
KENTUCKY
Figure 1. Utility locations.
-------
These control studies were based on bench scale and pilot plant studies done
by USEPA to investigate, sample for and control trihalomethanes.^""' This ob-
jective also included an investigation of bacteriological levels to ensure
that treatment modifications designed to lower trihalomethane concentrations
were not compromising finished water quality.
The second major objective was the determination of the levels of tri-
halomethanes and other selected organic compounds in raw and finished waters
at all project utilities for one year. Other compounds for investigation were
selected from a list designated by USEPA as organic Priority Pollutants for
which an analytical protocol was available.
CONTRACT LABORATORY
A laboratory service contract was awarded to the Radian Corporation,
Austin, Texas, after a review of proposals from several private laboratories
detailing analytical costs and capabilities for performing gas chromatography
(GC) and gas chromatography/mass spectrometry (GC/MS) analyses for selected
organic Priority Pollutants.
SCOPE OF WORK
Early in the project, members of the staff visited each participating
water utility to study its treatment practices and to determine the level of
participation by each utility. Minimum participation included monthly samp-
ling for organic analyses of raw and finished waters for one year, and
measurement and reporting of several background water quality parameters.
Participation in trihalomethane control studies included: sampling of raw,
in-plant, and finished waters for organic analysis several times a week for
periods ranging from four weeks to several months; determination of levels of
routine physical, chemical, and bacteriological water quality parameters for
each sampling location; and reporting of hydraulic, maintenance, and operation
data during routine and modified treatment (Sections 5 and 6).
Monthly sampling began at all 11 utilities in July 1977 and continued
through June 1978. Trihalomethane control studies at seven of the utilities
began in July 1977 and concluded in November 1978. The project staff worked
with each utility to coordinate sampling schedules and shipment to the con-
tract laboratory and to follow the progress at those utilities involved in
trihalomethane control studies.
The staff worked with Radian Corporation personnel to develop GC and GC/
MS quality control programs, coordinate organic analyses and shipment of sam-
ple bottles to the utilities, and review the progress of organic analyses.
This review led to changes in some analytical procedures and the implementa-
tion of a more rigorous quality assurance program. (Laboratory procedures and
quality assurance programs are described in Section 5 and Appendices A, B, D,
G and I.)
The project staff reviewed, interpreted and compiled all organic data
received from the contract laboratory and all data received from the utility
laboratories (Sections 6 and 7). Utility personnel collected a total of 3,446
-------
samples for organic analyses of which 2,950 produced usable chromatograms or
mass spectra. Data from about 500 samples were not available because of dam-
age in shipment, damage at the contract laboratory, headspace development in
volatile samples, samples not analyzed, and data not usable for reasons
including occasional loss of GC sensitivity or deviation from routine GC oper-
ating conditions.
-------
SECTION 2
CONCLUSIONS
The following conclusions are based on findings summarized in this sec-
tion. They apply to raw and finished water in the treatment plant but not to
the water in the distribution system.
1. Trihalomethanes are formed during the treatment of surface water when
free chlorine is present for significant periods of time.
2. Modifications of the chlorination process which may be viable trihal-
omethane control methods include: relocation of the initial chlorine applica-
tion to a location where treatment has reduced the precursor concentration;
ammoniation to convert free to combined chlorine; and chlorine dioxide as an
alternative to chlorine as the initial disinfectant.
3. Granular activated carbon (GAG) used in place of sand in the gravity
filters (filtration/adsorption) may be an effective trihalomethane control
process for approximately two months; however, periodic GAC reactivation is
necessary if GAC is to be used for trihalomethane control for extended periods
of time.
4. Evaluation of the effectiveness of treatment process modification for
trihalomethane control should include determination of instantaneous and ter-
minal trihalomethane concentrations and the trihalomethane formation potential
(a measure of precursor concentration) to aid in defining changing precursor
levels in the raw water and in determining the effects of treatment on precur-
sor removal and trihalomethane formation.
5. Total coliform and standard plate count levels should be determined
routinely on in-process and finished water samples to ensure that process
modification for trihalomethane control has not adversely affected bacterio-
logical levels in the treated water.
6. Process modification for trihalomethane control should extend over a
period of time adequate to determine short-term, seasonal and other variations
in raw water precursor concentrations, bacterial levels, and other water
quality parameters, and to evaluate the effects of these variations on the
quality of the treated water.
7. For the evaluation of raw, in-process, and finished water quality, a
complete and continuing quality assurance program is necessary to ensure the
accuracy and precision of the analytical procedures and the resulting data for
trihalomethanes and other organic compounds.
-------
8. Chloroform and other trihalomethanes were detected in many raw and
all treated surface water samples. At most utilities, the reaction between
precursor and free chlorine resulted in significant increases in trihalome-
thane concentrations. Other compounds occasionally present in raw and treated
water samples included carbon tetrachloride, dichlorobenzene isomers, 1,2,4-
trichlorobenzene, 1,2-dichloroethane and several polyaromatic hydrocarbons.
9. Analytical procedures more sensitive than those employed for project
samples (lower detection levels generally 0.1 to 0.2 ug/L) would be necessary
to evaluate the removal of organic compounds, other than trihalomethanes, by
normal or modified water treatment processes.
SUMMARY OF FINDINGS
The following summarizes the results of the treatment process modifica-
tion studies and the analysis of raw and finished water monthly samples.
Trihalomethanes
Chloroform was present in the majority of untreated surface water samples
at levels generally less than 1 ug/L; bromodichloromethane and dibromochloro-
methane were present less frequently, with most levels below 0.1 ug/L; bromo-
form and dichloroiodomethane were not present above 0.1 ug/L.
Trihalomethanes were formed during water treatment in the presence of
free chlorine. Trihalomethane levels in treated water (clear well effluent)
varied seasonally, with the lowest levels occurring during the winter and the
highest levels during the summer. The levels also varied with each utility's
treatment. Total trihalomethane (TTHM) levels for finished surface waters
ranged from 2 ug/L at one utility in February to 240 ug/L at another utility
in August. Finished water total trihalomethane levels at West View, a ground-
water source, did not exceed 2 ug/L. For ten utilities treating surface
water, trihalomethane levels in finished waters were:
Concentration, ug/L
Mean Annual Maximum
Chloroform
Bromodichloromethane
Dibromochloromethane
Bromoform
Dichloroiodomethane
Total trihalomethanes
35
13
5.6
0.4
0.1
54
180
54
33
4.4
1.0
—
Relatively higher concentrations of brominated trihalomethanes resulted in
finished water when the in-plant reaction time with free chlorine was reduced.
All finished waters contained unreacted trihalomethane precursor as mea-
sured by trihalomethane formation potential (THMFP). Data averaged from ten
utilities treating surface water indicated that 23% of raw water THMFP was
converted to total trihalomethane during treatment, 37% of raw water THMFP was
removed by treatment, and 40% of raw water THMFP was passed into the distribu-
-------
tion system. Reduction in terminal TTHM concentrations generally coincided
with reduction in turbidity levels.
Trihalomethane Treatability
Moving the point of initial chlorine application to a location where
treatment had reduced precursor levels resulted in decreased instantaneous
trihalomethane concentrations in the finished water, because a better quality
water, in terms of reduced THMFP, was chlorinated. The reduction of precur-
sor-chlorine reaction time was also a factor in the decreased trihalomethane
formation.
In studies at Pittsburgh and Wheeling, significant reduction in bacterial
densities occurred in unchlorinated waters when potassium permanganate was fed
with other chemicals prior to flocculation and settling.
At Pittsburgh, Wheeling and Cincinnati, moving the initial chlorine
application point caused a delay in reduction x>f bacterial densities, but the
bacterial quality of the finished waters was maintained.
The Louisville study showed that when sufficient ammonia was applied to
in-plant waters to convert free chlorine to combined chlorine, little or no
further trihalomethane formation resulted. The bacterial quality of the fin-
ished water was satisfactory; ammoniation followed three hours of free
chlorine contact time. At the Western Pennsylvania Water Company's Hays Mine
Plant, only very low levels of trihalomethane were formed when raw water
ammonia levels were such that no free chlorine resulted from raw water
chlorination.
The study at the Western Pennsylvania Water Company also showed that
little or no trihalomethanes were formed when chlorine dioxide was fed to the
raw water in place of chlorine. Although 1.5 mg/L chlorine dioxide was not as
effective as 2.6 mg/L chlorine in reducing raw water bacteria levels, clear
well chlorination provided adequate disinfection. Chlorine dioxide was gener-
ated from sodium chlorite and hydrochloric acid at an 80% yield and with only
limited formation (less than 5%) of free chlorine. Although 60 to 70% of the
chlorine dioxide reacted with substances in the water forming chlorite ion,
flocculation, settling and filtration through two-and-one-half year old GAG
reduced the residual chlorite concentration in the treated water to less than
0.1 mg/L.
The effects of individual treatment materials, including powdered acti-
vated carbon (PAC), potassium permanganate or chlorine dioxide, on precursor
levels could not be determined, because all of the chemicals are generally
added at a single point prior to flocculation and settling.
During summer months at Huntington and Beaver Falls, virgin granular
activated carbon (GAG) operated in the filtration/adsorption mode in beds
designed for sand filtration was exhausted for the removal of chloroform at
seven to 15 weeks of operation, for bromodichloromethane at eight to 15 weeks
of operation, for dibromochloromethane at eight to 15 weeks of operation, for
total trihalomethane at seven to 15 weeks of operation, and for THMFP at seven
-------
to 12 weeks of operation. Time to exhaustion was different for each utility
and type of GAG used. GAG filter/adsorbers passed carbon tetrachloride at
concentrations that could not be differentiated from influent concentrations
after four to seven months of operation, and 1,4-dichlorobenzene at concentra-
tions that could not be differentiated from influent concentrations after five
to 12 weeks of operation.
At Huntington and the Western Pennsylvania Water Company, GAG filter/
adsorbers which had been in service for one to two-and-one-half years were
exhausted for the removal of chloroform, bromodichloromethane, dibromochloro-
methane, and instantaneous TTHM.
Desorption from GAG filter/adsorbers was observed. GAG in use for one to
two-and-one-half years at Huntington desorbed carbon tetrachloride. When GAG
influent trihalomethane concentrations were significantly reduced, two-and-
one-half year old GAG desorbed trihalomethanes at the Western Pennsylvania
Water Company, and GACs in service for five months desorbed trihalomethanes
at Beaver Falls.
In three studies (Huntington, Beaver Falls and Western Pennsylvania Water
Company) bacterial densities in GAG effluent waters exceeded densities in GAG
influent waters when water temperatures exceeded 10°C. The bacterial quality
of the finished waters was satisfactory with clear well chlorination.
Other Organic Compounds
Carbon tetrachloride was occasionally present at concentrations from 0.1
to 0.6 ug/L in raw water at and downstream from Huntington. Carbon tetra-
chloride was occasionally present at 0.1 to 6 ug/L concentrations in finished
surface waters at all of the utilities. Its presence in finished waters was
probably attributable to contamination of chlorine used for disinfection.
Chlorobenzene was occasionally present in Huntington's raw and treated
water at concentrations up to 1 ug/L. It was not found in untreated or fin-
ished waters upstream from Huntington. It was frequently found in West View's
untreated groundwater at concentrations reaching 3.9 ug/L. After a reported
upstream spill, chlorobenzene was found at 8.5 ug/L in a finished surface
water.
During the winter months, polyaromatic hydrocarbons (PAHs)—naphthalene,
acenaphthylene, acenaphthene, fluorene, fluoranthene, pyrene, and phenanthrene
and/or anthracene—were present in raw and finished waters at concentrations
above 0.1 ug/L. Some GAG filter/adsorbers appeared to be effective in removal
of the PAHs.
Dichlorobenzene isomers were occasionally present in raw and finished
waters at levels above 0.2 ug/L. They were more frequently detected at and
downstream from Huntington. During a reported upstream spill, 1,4-dichloro-
benzene was found in a treated surface water at a concentration of approxi-
mately 11 ug/L.
1,2,4-Trichlorobenzene was occasionally present in raw and finished
-------
waters at levels greater than 0.2 ug/L. It was more frequently found at and
downstream from Cincinnati.
Unidentified halocarbons were detected in chlorinated waters but these
compounds were rarely found in raw waters. These may have been chlorination
products or may have resulted from contamination of chlorine used for
disinfection.
1,2-Dichloroethane, 1,2-dichloropropane, and 1,1-dichloroethane were
occasionally present in raw and finished waters at concentrations of 0.1 to 1
ug/L.
Tetrachloroethylene was found in Allegheny River water at approximately
60 ug/L as a result of what appeared to be an upstream spill.
Other specific organic Priority Pollutants were not present or were
rarely present at or above their lower detection levels in raw and finished
waters.
-------
SECTION 3
AREAS FOR FURTHER STUDY
During the winter months, several polyaromatic hydrocarbons were identi-
fied in raw and finished waters at most project utilities. Further research
into the presence and concentration of these .compounds and effective treatment
methods for their removal is needed.
Several Priority Pollutant halocarbons were identifiedxat and downstream
from Huntington, West Virginia. Organic analyses of Kanawha River samples
collected for another project indicated that these halocarbons in the Ohio
River at Huntington originated from the Kanawha River. A comprehensive point
source and river survey for these and other organic compounds in the industri-
alized section of the Kanawha River would provide information on specific
organic compounds to be considered in renewal of NPDES permits.
Carbon tetrachloride and unidentified halocarbons may have been intro-
duced to treated waters as a result of chlorine contamination. Chlorine manu-
facturing processes should be investigated and procedures for control of con-
tamination by carbon tetrachloride and possibly by other halocarbons should be
considered.
Unidentified halocarbons were found in chlorinated waters that were
rarely found in raw waters and may be chlorination products. Continuing
research to identify chlorination products other than trihalomethanes is
needed.
When water temperatures exceeded 10°C, bacterial densities in GAG filter
effluents were higher than in GAG influents at three utilities using GAG for
filtration/adsorption. Comprehensive studies of the nature of this increase
in bacterial densities and the development of methods to control bacterial
levels in GAG effluent are suggested.
Project utilities typically feed powdered activated carbon and potassium
permanganate during treatment. This project was not able to evaluate the
full-scale effects of these chemicals on trihalomethane control but their
effects at typical feed rates should be studied.
This project was not able to evaluate the full-scale effect of applied
chlorine dioxide on precursor levels. Further study of the effect of reason-
able feed rates of chlorine dioxide on the resulting chlorine .species and the
nature of resulting organic compounds is needed.
10
-------
SECTION 4
PROJECT ORGANIC COMPOUNDS
TRIHALOMETHANES
Five individual trihalomethane (THM) compounds were qualified and quanti-
fied in utility waters. They were chloroform, bromodichloromethane, dibromo-
chloromethane, bromoform and dichloroiodomethane. In order to facilitate the
investigation of trihalomethanes and their control, other parameters were also
utilized. Although these parameters are discussed elsewhere? they will be
defined here as they applied to this project.
1. Total trihalomethane (TTHM) concentration is the summation of the
concentrations of five individual THMs in a sample. Example: 42 ug/L CHC13 +
12 ug/L CHBrCla + 8 ug/L CHBr2Cl + 1 ug/L CHBr3 + 1 ug/L CHIC12 = 64 ug/L
TTHM.
2. Instantaneous TTHM (inst TTHM) is the concentration of TTHM in the
water at the time the sample is collected.
3. Terminal TTHM (term TTHM) is the sum of TTHM present in the water at
the moment of sampling and TTHM subsequently formed during additional reaction
time under defined conditions. During the project, the reaction was driven
toward completion by adding chlorine to exhaust the precursor. The sample was
stored at finished water pH and temperature for seven days, i.e., beyond the
normal detention time in the distribution system of the utilities, with suffi-
cient free chlorine added to satisfy demand. After seven days under storage
conditions, a concentration was reached that was assumed to represent a com-
pleted reaction. For the project, that concentration was defined as terminal
TTHM.
4. Trihalomethane formation potential (THMFP) is the difference between
the terminal TTHM and the instantaneous TTHM (term TTHM - inst TTHM = THMFP),
an indirect measure of the unreacted precursor in the water sampled. It is
the increase in the TTHM concentration that occurred during the storage period
for the determination of the terminal TTHM concentration. The unreacted pre-
cursor has the potential to further increase TTHM concentrations in the pre-
sence of free chlorine.
These parameters, illustrated in Figure 2, were used to evaluate trihalo-
methane concentration and control in Sections 6 and 7.
11
-------
THMFP =
indirect measure of
the concentration of
unreacted precursor
present at the time
water is sampled
inst TTHM =
TTHM concentration
at the time water
is sampled
(reacted precursor)
term TTHM =
TTHM concentration
possible for the
water sampled under
defined conditions.*
*buffered to finished water pH, 15 mg/L chlorine added,
stored for seven days at finished water temperature.
Figure 2. Graphical representation of trihalomethane parameters.
OTHER PRIORITY POLLUTANTS
Analyses for numerous other organic compounds were performed throughout
the study. These compounds were chosen from USEPA's Priority Pollutants list8
on the basis of three criteria: they were of known or suspected health con-
cern; their occurrence in the waters of the Ohio Valley was a possibility
because of their association with industrial discharges or agricultural run-
off; and USEPA had proposed a GC/MS analytical procedure for analyses for
these compounds in water.
Consideration of project objectives, available funds, and analytical
costs and capabilities led to a decision to analyze for some, but not all, of
the Priority Pollutants. Tables 1, 2 and 3 list the organic compounds for
which analyses were performed.
Analyses were not performed for three groups of organic compounds: vola-
tile hydrocarbons by GC/flame ionization detection (toluene, benzene and ethyl
benzene), because of unacceptable detection levels; acid extractable halocar-
bons by GC/Hall detection (chlorophenols), because of unacceptable detection
levels; and base/neutral extractable nitrocarbons (benzidine, nitrotoluenes,
12
-------
etc.), because of detection levels and GC/MS sensitivity (Section 5 and
Appendix H).
TABLE 1. PROJECT ORGANIC COMPOUNDS
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
Chloroform 1,2-Dichloropropane
Bromodichloromethane trans-1,3-Dichloropropene
Dibromochloromethane Trichloroethylene
Bromoform cis-1,3-Dichloropropene
Dichloroiodomethane 1,1,2-Trichloroethane
1,1-Dichloroethane 1,1,2,2-Tetrachloroethane
1,2-Dichloroethane Tetrachloroethylene
1,1,1-Trichloroethane Chlorobenzene
Carbon Tetrachloride
TABLE 2. PROJECT ORGANIC COMPOUNDS
BASE-NEUTRAL EXTRACTABLE HALOCARBONS, GC/HALL DETECTOR
1,3-Dichlorobenzene tf-BHC (Lindane)
1,4-Dichlorobenzene o-BHC
Hexachloroethane Heptachlor
1,2-Dichlorobenzene 3-BHC
bis(2-Chloroiosopropyl) ether Aldrin
bis(2-Chloroethyl) ether Heptachlor epoxide
1,2,4-Trichlorobenzene a-Endosulfan
Hexachlorobutadiene Dieldrin
bis(2-Chloroethoxy) methane DDE
Hexachlorocyclopentadiene Endrin
2-Chloronaphthalene ODD
4-Chlorophenyl phenyl ether 3-Endosulfan
4-Bromophenyl phenyl ether DDT
a-BHC Methoxychlor
TABLE 3. PROJECT ORGANIC COMPOUNDS
BASE-NEUTRAL EXTRACTABLE HYDROCARBONS, GC/FLAME IONIZATION DETECTOR
Naphthalene Butyl benzyl phthalate
Acenaphthylene bis(2-Ethylhexyl) phthalate
Acenaphthene 1,2-Benzanthracene
Dimethyl phthalate Chrysene
Fluorene 3,4-Benzofluoranthene
Diethyl phthalate 11,12-Benzofluoranthene
Phenanthrene Benzo(a)pyrene
Anthracene Indeno(l,2:C,D)pyrene
Di-n-butyl phthalate l,2:5,6-Dibenzanthracene
Fluoranthene 1,12-Benzoperylene
Pyrene ___
13
-------
GAS CHROMATOGRAPHY VERSUS MASS SPECTROMETRY
The USEPA protocol for the organic Priority Pollutants is based on GC/MS
analysis.8 A decision was made to analyze all samples by GC and the Hall or
other detectors to provide presumptive identification of organic compounds,
because the cost of GC/MS procedures would limit the number of samples which
could be analyzed. GC/MS analyses were used to provide positive or negative
confirmation of presumptive identifications. For individual organic compounds
there were significant differences between the lower detection levels by
GC/detector and GC/MS. Specific examples are discussed in Section 7.
14
-------
SECTION 5
ANALYTICAL PROCEDURES AND QUALITY ASSURANCE
ORGANIC CONTRACT LABORATORY
At submicrogram and microgram per liter (ug/L) levels of analysis for
organic compounds, a comprehensive quality assurance program must accompany
all aspects of sample handling and analysis. The program is necessary for
two reasons: GC reports of an organic compound should be the result of the
presence of the compound in the water at the time it was sampled and not the
result of procedural contamination; and the significance (accuracy and preci-
sion) of the data must be known before interpretation. The following sub-
sections and their related appendices describe the quality assurance program.
General Laboratory Controls
Extensive laboratory control procedures were necessary to ensure that
interferences were definable at acceptably low concentrations. General lab-
oratory control procedures involved the following: the cleaning, preparation
and handling of bottles for sample collection and of laboratory glassware used
in the analysis of project samples; the preparation of low organic water for
purgeable blank analyses, preparation of purgeable standards, rinsing of
glassware, recovery tests for extractable compounds, and preparation of
buffers; the identification and control of interferences from materials such
as solvents and gases for purging and chromatography; and the storage of pro-
ject samples to maintain integrity prior to analysis. These control proce-
dures are detailed in Appendix A.
The effectiveness of these controls was routinely evaluated by the labo-
ratory. At the same time project samples were analyzed, system blanks were
analyzed to detect interferences. When an unacceptable interference was
observed in system blanks, sample analyses were discontinued until the inter-
ference was identified and/or controlled.
Analytical Procedure for Purgeable Halocarbons
The purgeable halocarbon Priority Pollutants for which routine analysis
was performed and the approximate lower detection levels are listed in Table
4.
Q
The USEPA Priority Pollutant Protocol for analysis of halocarbons by
purge, trap, desorption and gas chromatography/mass spectrometry (GC/MS) was
revised by the laboratory^ to enable analysis by purge, trap, desorption and
15
-------
TABLE 4. PURGEABLE HALOCARBONS, GC/HALL DETECTOR
Approximate Lower
Detection Level
Compound ug/L
1, 1-Dichloroethane
Chloroform
1, 2-Dichloroethane
1, 1, 1-Trichloroethane
Carbon Tetrachloride
Bromodichloromethane
1, 2-Dichloropropane
trans-1, 3-Dichloropropene
Trichloroethylene
cis-1, 3-Dichloropropene
1, 1, 2-Trichloroethane
Dibromochloromethane
Dichloroiodomethane
Bromoform
1,1,2, 2-Tetrachloroethane
Tetrachloroethylene
Chlorobenzene
0.1
0.1
0.1
0.6 - 2.6a
0.1
0.1
0.1
0.1
0.5 - 1.9a
0.1
o.ib
0.1
a
1.0 - 3.4
0.1
, Laboratory contamination; see Section 7
Quantification relative to 1,4-dichlorobutane
GC/Hall detection with occasional GC/MS verification. A detailed description
of the purge, trap, desorption and GC/Hall detector equipment and analytical
procedures as used by the laboratory is given in Appendix B.
Qualitative and quantitative determinations of the purgeable halocarbons
were based on a calibration standard of these compounds (excluding dichloro-
iodomethane) and an internal standard of 1,4-dichlorobutane added to calibra-
tion standards and project samples. These determinations were automatically
performed by a Hewlett Packard 3380A programmable integratorlO and were
reviewed in each chromatogram by the project staff. Qualification (identifi-
cation) of peaks in sample chromatograms was based on relative retention time
(RRT) matching within ± 5% of RRT of standard peaks in calibration chromato-
grams. Quantification was based on a comparison of the response of a compound
and the internal standard in the calibration. Figure 3 represents a typical
chromatogram of a calibration standard, Figure 4, a typical system blank
chromatogram and Figure 5. a typical chromatogram of a project sample.
A stable calibration standard of dichloroiodomethane could not be main-
tained. Therefore, its relative retention time was obtained only once and it
was not a component of the purgeable halocarbon standard. Qualification in
field samples was based on this relative retention time. The GC/MS labora-
tory confirmed dichloroiodomethane GC identified in this manner. Routine
quantification was relative to 1,4-dichlorobutane.
Both qualitative and quantitative data produced by GC/Hall analyses are
presumptive. However, validity of the GC/Hall procedure for purgeable com-
16
-------
1,1- DICHl-QgQ ETHANE
••—1,2- DICHLOROETHANE
_____ •— I, I, I- TR1CHLOROETHANE
CARBON TETRA CHLORIDE^ ^-SROMO PI CH LQ R O METH A ME
jf A.
TRICHLOROETHVLENE
V (Dl
X-
-------
- CHLOROFORM
- 1,1, I - TRICHLOROETHANE
•TRICHLOROETHYLENE
RA.CHLOKO ETHYL EM E
\ 1,1,2, 2 - TETfcACHLOROETHANE
-INTERNAL STANDARD
Figure 4. Typical gas chromatogram of
purgeable system blank using Hall detector.
CHLOROFORM
•BLAvNK
CARBON TETRACHLORIDE
BRONCO METHANE
1— UNKNOWN
-»-DIBROMOCHLORON/lETH AME
-DICHLOROIO DO METHANE
- BROMOFORM
r—BLAWK .r-INTERNAL STANDARD-
Figure 5. Typical gas chromatogram of purgeable sample using Hall detector.
18
-------
pounds was maintained through the use of daily calibration standards, USEPA
reference standards, an internal standard for qualification and quantification
and occasional GC/MS verification.
The trihalomethane compounds in terminal level samples were also evalua-
ted by purge, trap, desorption and GC/Hall detection. The calibration stan-
dard contained only chloroform, bromodichloromethane, dibromochloromethane,
bromoform, and the internal standard, 1,4-dichlorobutane. The equipment and
analytical procedures used were the same as for other purgeable halocarbons,
with the exception of temperature programming. Details are given in Appendix
B.
Quality Assurance for Purgeable Halocarbons
In order to ensure that GC reports of a compound were not a result of
interference and to provide sufficient data to define the accuracy and preci-
sion of GC data, laboratory analyses were supplemented by a comprehensive
quality assurance program.
Periodic Quality Assurance—
The laboratory established a concentration above which the purgeable
halocarbons could be routinely detected in project samples by the method of
analysis detailed in Appendix B. The lower detection level was defined as an
integrable peak greater than an arbitrary area count of 1000 units and was
determined by diluting the calibration standards by factors of two until inte-
gration could not occur. These levels are listed in Table 4. For most of the
purgeable halocarbons, the approximate lower detection level was 0.1 ug/L.
This level appeared to have good validity when compared to GC/MS verification
of GC/Hall detector data close to the reported detection level and when tested
by periodic analyses of calibration standards at 0.1 ug/L.
Because the HP 3380A integrator assumes linearity of the Hall detector
response when quantifying, the linear relationship between the amount of com-
pound purged and the Hall detector response was tested. A least squares
regression analysis, assuming a linear model, was done using detector response
as the dependent variable. Concentrations expected in project samples were
evaluated, i.e., chloroform ranging from 1.0 ug/L to 200 ug/L, bromoform
ranging from 1.0 ug/L to 10 ug/L. Correlation coefficients of 0.98 verified
the linearity of the Hall detector over the range of concentrations in project
samples.
The variability of standard analyses at several concentrations was eval-
uated periodically. Appendix C, Tables C-l to C-15, contains compiled data
on the reproducibility of laboratory standards by purge, trap, desorption and
Hall-detection over a range of concentrations. The data indicate that concen-
trations were significant to two figures from 0.1 ug/L to 200 ug/L. This
level of significance was applied to project sample data.
Routine Quality Assurance—
Daily control criteria and limits were established by the project and
laboratory staffs. If control limits were exceeded, sample analyses were dis-
continued until conditions were again within the limits. Control criteria
19
-------
data were also accumulated for determination of the significance of project
sample data.
The daily control program involved an initial analysis of a 16-component
calibration standard containing the 1,4-dichlorobutane internal standard.
This analysis was used to program the integrator for relative retention times
and response factors. Interspersed with subsequent project sample analyses
were the following: USEPA reference sample analyzed daily as an unknown
against the calibration standard; low organic water analyzed periodically
through the day as a system blank to determine possible interference from the
syringe, purge, trap, desorption, GC/Hall system or laboratory air; each day,
a previously analyzed sample was reanalyzed for comparative evaluations of
day-to-day analytical conditions; and calibration standard analyzed approxi-
mately every six hours as an unknown to determine stability of the system for
RRT and response factors. In addition to the laboratory control program,
approximately 12 per cent of project field samples were submitted in
replicate.
The background concentrations defined by system blanks were used to
correct data by one of two methods. An interference detected on only one
analytical day was subtracted from all sample data produced that day. A re-
curring interference was evaluated over the period of occurrence and statisti-
cally weighted (mean interference plus two standard deviations) to reflect the
interference over that period. This statistical correction was subtracted
from all sample data produced over that period.
Application of Quality Assurance Data
for Purgeable Halocarbons to Sample Data—
Accumulated quality assurance data from analyses of USEPA reference sam-
ples, calibration standards handled as unknowns, replicate field samples, and
reanalysis of single field samples are presented in Appendix C for the purge-
able halocarbons. These data defined the significance of the sample data.
The following examples demonstrate the application of these quality assurance
data to sample data.
Quality assurance data for chloroform are presented in Appendix C, Table
C-l and Figures C-l to C-3. An examination of these data provides a measure
of both the accuracy and precision that must be considered in interpretation
of chloroform data. Data were compiled from analyses of replicate field sets
and from replicate analyses of single field samples. The mean concentration
of each data set was plotted versus the deviation of the set about the mean.
(For example, a pair of field duplicates were analyzed for instantaneous
chloroform. Concentrations obtained for the pair were 88 ug/L and 72 ug/L
producing a mean value of 80 ug/L.and a relative average deviation of ± 10%.
For this set, the mean of 80 ug/L was plotted versus the relative average
deviation of ± 10%. If more than two field replicates were analyzed for
instantaneous chloroform and concentrations were 41 ug/L, 45 ug/L and 46 ug/L,
a mean value of 44 ug/L and a relative standard deviation of ± 6%, the mean of
44 ug/L was plotted versus the relative standard deviation of ± 6%.)
Instantaneous chloroform data obtained from the replicate sets were
plotted in concentration ranges. See Figure C-l. In the concentration range
20
-------
of 5.0 to 140 ug/L, chloroform replicated within 19% within a set, 95% of the
time. Thus, if the concentration of chloroform in a sample was determined to
be 42 ug/L, reanalysis of the sample or analysis of a duplicate field sample
produced a concentration within ± 19% of 42 ug/L, 95% of the time. Therefore,
concentrations of 36 ug/L and 47 ug/L could not be differentiated.
For instantaneous chloroform data in the concentration range of 1.0 to
5.0 ug/L, chloroform replicated within 23% within a set, 95% of the time, as
shown in Figure C-2. As the concentrations of instantaneous chloroform
decreased below 1.0 ug/L and approached the approximate detection limit of
0.1 ug/L, variability increased greatly. Figure C-2 shows that the vari-
ability approached ± 100% at the detection limit. Therefore, concentrations
of 0.1 ug/L and 0.2 ug/L could not be differentiated.
Terminal chloroform data were also plotted for sets of field samples and
are shown in Figure C-3. In the concentration range of 5.0 to 325 ug/L,
chloroform replicated within 20% within a set, 95% of the time, not unlike the
± 19% variability for instantaneous chloroform data in a similar concentration
range.
In addition to quality assurance data from field samples, data from
reproducibility of USEPA reference standards and laboratory calibration stan-
dards were compiled as shown for chloroform in Table C-l. At concentrations
for which a large number of standards were analyzed, data indicate variability
similar to that shown in field data in the same concentration range. USEPA
reference standards containing chloroform at 68.5 ug/L were analyzed 83 times
as part of the routine quality assurance program. The data were blank
corrected. A mean value of 70.9 ug/L with a relative standard deviation of
± 14% resulted. The mean represented a relative error of + 4% from the true
value as reported by USEPA. Calibration standards containing chloroform at
10 ug/L were analyzed 57 times as unknown samples by comparison to the pro-
grammed calibration standard as part of the routine quality assurance program.
The data produced were blank corrected. A mean value of 9.4 ug/L with a rela-
tive standard deviation of ± 20% resulted. The mean represented a relative
error of - 6% from the true value reported by the laboratory. These data
indicate that quantification of chloroform standards at or above 10 ug/L were
accurate within ± 6%. Repeatability (precision) of analyses was within ± 20%.
Quality assurance data from the analyses of pure compounds in low organic
water (Table C-l) only suggest the significance of data produced from the
analyses of field samples. Quality assurance data from replicate analyses of
field samples (Tables C-l to C-3) are more meaningfully applied in determining
the significance of sample data.
As a second example, quality assurance data in Table C-7 for chloroben-
zene illustrate the significance applicable to data as concentrations approach
the detection level. Analyses of 19 sets of field replicate samples indicate
increasing variability of data with decreasing chlorobenzene concentrations.
Six samples within sets producing chlorobenzene data in the 1.4 to 2.9 ug/L
range, replicated within ± 29%. The variability of replication increased to
± 59% in six sets of samples producing data in a lower range of concentrations
from 0.1 to 0.8 ug/L. Seven sets of samples producing chlorobenzene concen-
trations less than 0.1 ug/L varied ± 100% in replication. Thus, chlorobenzene
21
-------
concentrations of 1.0 ug/L and 2.0 ug/L in project samples could be differ-
entiated, but concentrations of 0.1 ug/L and 0.2 ug/L could not be
differentiated.
A comparison of the field quality assurance data to the data on precision
of chlorobenzene from analyses of laboratory standards at concentrations below
1.0 ug/L indicates less variability in laboratory than in field samples; how-
ever, the evaluations at low concentrations were based on a small number of
analyses of pure compounds in low organic water. When calibration standards
containing chlorobenzene at 10 ug/L were analyzed 57 times, as part of the
routine quality assurance program, a precision of ± 37% was obtained, a value
similar to the ± 29% obtained for field samples in the concentration range of
1.4 to 2.9 ug/L.
Application of the significance of quality assurance data to total tri-
halomethane (TTHM) values must also be made for interpretation of instantan-
eous and terminal TTHM project data. Instantaneous and terminal TTHM data
were compiled from analyses of replicate field sets and from replicate
analyses of single field samples. The mean TTHM concentration of each data
set was plotted versus the relative deviation of the set about the mean. The
resulting levels of precision for 95% of the sample sets were ± 20% for
instantaneous TTHM and ± 16% for terminal TTHM, as illustrated in Figures C-ll
and C-12, respectively. These levels of variability generally agree with
levels from replicate data sets of individual trihalomethane compounds at con-
centrations greater than 1.0 ug/L. These data indicate that sample instan-
taneous TTHM concentrations of 40 ug/L and 65 ug/L can be differentiated but
instantaneous concentrations of 80 ug/L and 86 ug/L cannot.
Quality assurance data from analyses of field samples and from analyses
of standards and blanks are presented for each purgeable compound in Appendix
C. These data must be carefully evaluated and applied to the interpretation
of project sample data for each of the purgeable compounds.
Analytical Procedure for Base-Neutral Extractable Compounds
The basic and neutral organic Priority Pollutants extracted from a sample
with methylene chloride under alkaline conditions are referred to within this
report as base-neutral extractable compounds. The extraction procedure, as
described in USEPA's Protocol,8 was used with several laboratory modifications
as detailed in Appendix D.^
Two groups of compounds were analyzed from an extracted and concentrated
sample. One group, extractable halocarbons including specific pesticides,
was analyzed by GC/Hall detector (GC/Hall). Individual compounds and their
approximate lower levels of detection are listed in Table 5. Figure 6 is a
representative GC/Hall chromatogram for a direct injection analysis of cali-
bration standards, Figure 7 a chromatogram of a system blank, and Figure 8 a
chromatogram of an extracted and concentrated sample.
The second group, non-halogenated extractable hydrocarbons, was analyzed
by GC/flame ionization detector (GC/FID). Individual compounds and their
approximate lower levels of detection are listed in Table 6. Figures 9, 10,
22
-------
note: for reference code, see Table 5
J = hexachlorobenzene (internal standard)
Figure 6. Typical gas chromatogram of base-neutral extractable halogenated
Priority Pollutants calibration standard using Hall detector.
23
-------
solvent peaks
internal standard
Figure 7. Typical gas chrpmatogram of base-neutral extractable
solvent blank using Hall detector.
24
-------
solvent peaks
-1,4-dichlorobenzene
-1,2-dichlorobenzene and/or hexachloroethane
-blank
—1,2,4-trichlorobenzene and/or hexachlorobutadiene
-bis(2-chloroethoxy) methane
-unknown
^unknown
internal standard
heptachlor and/or g-BHC
- unknown
*—- o-endosulfan
Figure 8. Typical gas chromatogram of base-neutral
extractable sample using Hall detector.
25
-------
TABLE 5. HALOGENATED BASE-NEUTRAL EXTRACTABLE PRIORITY POLLUTANTS
GC/HALL DETECTOR AND 3.000 CONCENTRATION FACTOR
Reference
Code
A
B
Compound
F
G
H
I
K
M
N
P
Q
R
U
V
1,3-Dichlorobenzene
1,4-Dichlorobenzene
Hexachloroethane
J.,2-Dichlorobenzene
bis(2-Chloroisopropyl) Ether
_bis(2-Chloroethyl) Ether
1,2,4-Trichlorobenzene
Jexachlorobutadiene
bis(2-Chloroethoxy)methane
Hexachlorocyclopentadiene
2-Chloronaphthalene
__4-Chlorophenyl Phenyl Ether
4-Bromophenyl Phenyl Ether
jx-BHC
Jf-BHC (Lindane)
5-BHC
"Heptachlor
3-BHC
"Aldrin
Heptachlor Epoxide
a-Endosulfan
JDieldrin
[DDE
Endrin
[~DDD
[_3-Endosulfan
DDT
Methoxychlor
Approximate Lower
Detection Level3
ug/L
0.1
0.1
0.1
0.2
0.1
0.1 - 0.2
0.1 - 0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1 - 0.2
a = not corrected for extraction losses
and 11 are chromatograms of a direct injection calibration standard, a system
blank, and an extracted and concentrated sample, respectively.
The method of qualitative determination differed for the two groups of
extractable compounds. Compound identification by GC/Hall analysis was based
on relative retention time match within ± 5% RRT of the corresponding compound
in the calibration standard and an extracted internal standard of hexachloro-
benzene in each sample. When the sample was analyzed by GC/FID, qualification
was based on absolute retention time match within ± 5% of absolute retention
times of standard peaks in the calibration chromatograms. Although the hexa-
chlorobenzene internal standard did not elicit a sufficient response on the
flame ionization detector for internal standard qualification, it did cause a
small, integrable response that was used as an internal standard for relative
retention time matching when chromatograms were reviewed by the project staff.
The recovery of these compounds by extraction was variable; therefore,
26
-------
solvent peak
naphthalene
-acenapthene
dimethyl phthalate
fluorene
— diethyl phthalate
phenanthrene and anthracene
di-n-butyl phthalate
fluoranthene
-« pyrene
butyl benzyl phthalate
fbis(2-ethylhexyl)phthalate
^1,2-benzanthracene
|_chrysene
benzo(a)pyrene
indeno(1,2:C,D)pyrene
1,2:5,6-dibenzanthracene
1,12-benzoperylene
Figure 9. Typical gas chromatogram of base-neutral extractable
Priority Pollutants calibration standard using flame ionization detector.
27
-------
•solvent peak
•dimethyl phthalate
-hexachlorobenzene (internal standard)
•di-n-butyl phthalate
bis(2-ethylhexyl) phthalate
Figure 10. Typical gas chromatogram of base-neutral
extractable solvent blank using flame ionization detector.
_J
28
-------
—-acenaphthylene acenaphthene
Lank
rluorene
-hexachlorobenzene (internal standard)
•phenanthrene and/or anthracene
—blank
-fluoranthene
— pyrene
•* blank
-* blank
note: other peaks are unknovms
Figure 11. Typical gas chromatogram of base-neutral
extractable sample using flame ionization detector.
29
-------
TABLE 6. NON-HALOGENATED BASE-NEUTRAL EXTRACTABLE PRIORITY POLLUTANTS
GC/FLAME IONIZATION DETECTOR AND 3,000 CONCENTRATION FACTOR
Compound
Approximate Lower
Detection Level3
ug/L
Naphthalene
Acenaphthylene
Acenaphthene
Dimethyl Phthalate
Fluorene
Diethyl Phthalate
Phenanthrene
[Anthracene
Di-n-butyl Phthalate
Fluoranthene
Pyrene
^Butyl Benzyl Phthalate
T"bis(2-Ethylhexyl) Phthalate
1,2-Benzanthracene
_Chrysene
3,4-Benzofluoranthene
_11,12-Benzof luoranthene
Benzo(a)pyrene
_Indeno (1,2: C, D) pyrene
1,2:5,6-Dibenzanthracene
1,12-Benzoperylene
j
C
o.
0.
1.
5.
0.5
2.0
1.0
0.5
1.0
0.5
2.0
1.0
5.0
5.0
10.0
10.0
a = not corrected for extraction losses
this procedure for base-neutral extractable Priority Pollutants must be con-
sidered semi-quantitative. Quantification was based on a comparison of the
response of corresponding peaks in the concentrated sample extract and cali-
bration chromatograms, and the concentration factor. The concentrations were
not corrected for extraction losses. Both qualification and quantification
wer| automatically handled by a Hewlett Packard 3380A programmable integra-
tor and were reviewed in each sample chromatogram by the project staff.
GC data generated for the base-neutral extractable compounds with the
Hall and FI detectors are presumptive. In order to determine the validity of
data produced by GC only, GC/MS confirmation attempts were essential. Section
7 discusses comparative GC and GC/MS data for each compound.
Quality Assurance for Base-Neutral Extractable Compounds
An extensive quality assurance program was necessary to ensure the signi-
ficance and validity of the data.
Periodic Quality Assurance—
Approximate lower detection levels were established for routine analysis
of the extractable compounds by direct injection of calibration standard com-
pounds diluted by factors of two until an arbitrary area count fell below
1,000 units.
30
-------
For the halogenated compounds, the lower detection levels by GC/Hall
detection varied throughout the project and ranged from approximately 0.1 ug/L
to 0.2 ug/L depending on the particular compound (Table 5). Validation of
detection levels in this range was supplied by extraction recovery tests of
calibration standard compounds analyzed by GC/Hall, and by GC/MS confirmation
of GC/Hall data at the lower levels of detection for some, but not all, of the
halogenated compounds.
For the non-halogenated compounds analyzed by GC/FID, the levels ranged
from 0.5 ug/L to 10 ug/L depending on the particular compound (Table 6).
Further validation of these levels was supplied by direct injection and
extraction recovery tests of calibration standard compounds analyzed by
GC/FID, and by GC/MS confirmation of GC/FID data at the lower detection
levels.
Extraction recoveries of the base-neutral extractable Priority Pollutants
at several concentrations were determined by spiking calibration standard com-
pounds in methanol into three liters of low organic distilled water. Extrac-
tion was evaluated by averaging the recoveries of triplicate extraction and
concentration tests. Values were corrected for interferences that occurred in
blanks representative of three liters of low organic distilled water extracted
and concentrated in an identical manner. Percent recoveries for each halo-
genated compound are given in Appendix E, Tables E-l through E-20 and for the
non-halogenated compounds in Appendix F, Table F-l.
Recoveries were based on extraction of calibration standard compounds
from low organic distilled water and only suggest that similar recoveries oc-
curred when extracting Priority Pollutants from field samples representing
varied and complex waters. While extraction recovery tests from selected
field waters rather than from distilled water would have been more representa-
tive, there was no assurance that a relatively small number of such recovery
tests would have been representative of the hundreds of samples analyzed dur-
ing the project.
The accuracy and precision of standards analyzed at several concentra-
tions by direct injection were evaluated periodically and as part of a routine
quality assurance program. The data are compiled in Appendix E, Tables E-l
through E-20, and Appendix F, Tables F-2 and F-3, for the halogenated and the
non-halogenated base-neutral extractable Priority Pollutants, respectively.
The results of the data indicate that concentrations were significant to two
figures at the ug/L level. This level of significance was applied to field
data.
Routine Quality Assurance—
A routine quality assurance program was found to be particularly impor-
tant in the preparation of samples for the analysis of the extractable
Priority Pollutants. Extraction of compounds from samples into solvent and
concentration of the solvent were found to introduce significant levels of
impurities causing interference in the GC/FID analyses. The purity of
solvents was routinely evaluated as part of an evaluation of the entire analy-
tical procedure that included glassware cleaning, solvent extraction, concen-
tration, storage and analysis. This evaluation was conducted by analysis of
31
-------
solvent blanks handled in a manner identical to samples, i.e., volumes of sol-
vent as specified in the procedure were introduced to extraction glassware,
concentrated, exchanged for a second solvent, concentrated, stored and ana-
lyzed by both GC/Hall and GC/FID.
Solvent blank analyses identified an interference in the analysis of bis-
chloroethers. This problem is detailed in Appendix G.
The daily quality assurance program for base-neutral extractable analyses
was based on a group analysis concept. One bottle of methylene chloride con-
tained sufficient volume for six extractions utilizing 550 mL each. Four
samples and two control blanks were extracted from each bottle of solvent.
Initial analyses of extracted and concentrated groups indicated that variabil-
ity of interferences between the two blanks within a group was often high.
Further, variability of blanks among groups was often high. Thus, the fre-
quency of two blanks per extraction group was maintained in order to charac-
terize the purity of each bottle of solvent and all analytical conditions
associated with the procedure. Data was corrected in groups for solvent blank
interferences specific only to a group. Data was statistically corrected for
several groups for solvent blank interferences that occurred consistently.
Samples were extracted, concentrated, stored and analyzed in groups with
associated solvent blanks. The daily GC/Hall and GC/FID analysis included the
following components per group: four field samples, a direct injection cali-
bration standard used to program the integrator for relative retention times
and response factors, two solvent blanks, a previously analyzed field extract
for comparative evaluations of day-to-day analytical conditions, and a direct
injection calibration standard handled as an unknown to determine stability
of the system for RRT and response factors. In addition, approximately ten
percent of the field samples were submitted in replicate.
All quality assurance data were used daily to ensure that analytical con-
ditions were within established control limits. All data were also compiled
for determination of the significance of project sample data.
Application of Quality Assurance Data to Sample Data—
Accumulated quality assurance data from analyses of standard compounds
extracted from distilled water, direct injected standard compounds handled as
unknowns, replicate field samples, and replicate analyses of single field
samples are presented in Appendices E and F for base-neutral extractable
Priority Pollutants. These data define the significance of the project sam-
ple data.
Application of Quality Assurance Data for Base-Neutral Extractable Halo-
carbons—Two examples demonstrate the significance of this data.
The quality assurance data for 1,3-dichlorobenzene are presented in
Appendix E, Table E-2. These data indicate that approximately 60% of the com-
pound was extracted from distilled water; however, extraction recovery data
only suggest that recovery of the compound by extraction from raw and treated
field waters was similar. For example, when a concentrated extract of a
field sample was analyzed for 1,3-dichlorobenzene, the indication is that the
32
-------
quantification represented approximately 60% of the field concentration.
Further, the precision obtained from analyses of extracts from field replicate
samples and from replicate analyses of extracts from single field samples in-
dicates that concentrations of 1,3-dichlorobenzene above 0.1 ug/L reported in
field extracts may be ± 58% to ± 100%. Thus, when 0.4 ug/L of 1,3-dichloro-
benzene was detected in a field extract, extraction recovery data suggest that
0.6 ug/L to 0.7 ug/L may have been in the sample, and precision data indicate
that an extract concentration of 0.4 ug/L could not be differentiated from
extract concentrations of 0.3 ug/L or 0.6 ug/L.
The second example illustrates the implications of co-eluting compounds.
The quality assurance data for co-eluting 1,2,4-trichlorobenzene and hexa-
chlorobutadiene are presented in Appendix E, Table E-4. Because of co-
elution, GC/Hall quantification was based on the assumption that both com-
pounds were equally present. This assumption was valid for laboratory extrac-
tion and reproducibility tests, but not for analyses of sample extracts. Only
GC/MS analyses of extracts determined whether one or both compounds were pre-
sent. Thus, when 0.3 ug/L of 1,2,4-trichlorobenzene and/or hexachlorobuta-
diene were detected in a sample extract, and compiled GC/MS data consistently
identified the Hall-detected peak as 1,2,4-trichlorobenzene and not as hexa-
chlorobutadiene, the quantification at 0.3 ug/L, based on the assumption that
both compounds were present, was only an estimated value. The true concentra-
tion of 1,2,4-trichlorobenzene could not be determined.
The importance of quality assurance data and its application to project
data cannot be overemphasized. These data must be evaluated and applied to
the interpretation of project sample data for any of the base-neutral extract-
able halocarbons.
Application of Quality Assurance Data for Non-Halogenated Base-Neutral
Extractable Hydrocarbons—Quality assurance data were obtained from standard
compounds analyzed by direct injection and from standard compounds extracted
from distilled water. Quality assurance data from sets of field sample
extracts were not obtained because GC/FID analyses of sample extracts pro-
duced little data above the approximate lower detection levels. Quality
assurance data produced from standard compounds injected at 5.0 ug/L and 10
ug/L and analyzed as unknown samples are contained in Appendix F, Tables F-2
and F-3.
Data produced from analysis of standard compounds at 1.5 ug/L and 10 ug/L
extracted from distilled water are presented in Appendix F, Table F-l.
Although variability of extraction recoveries for standard compounds analyzed
in triplicate on any one day was low, variability between tests run on dif-
ferent days during the project was very high. These data are not sufficient
to establish the relationship between the levels of compounds in field
extracts and the levels in field waters. The variability of recoveries only
suggests that extraction recoveries of project samples were also highly
variable.
33
-------
Attempted Analysis of Base-Neutral Extractable
Nitrogen-Containing Hydrocarbons
Analyses were attempted for nitrogen-containing base-neutral extractable
Priority Pollutants by' GC/alkali detector. This analytical task was aban-
doned, however, because GC/alkali detector data could not be supported by
GC/MS. Appendix H details the attempted analyses of these compounds.
Mass Spectrometer Analytical Procedures
Gas chromatography/mass spectrometry (GC/MS) verification of GC/Hall or
GC/FID data was done using the USEPA Protocol.8 Details of the laboratory's
MS equipment and procedures are given in Appendix I.
GC/MS support of GC data was used in several ways. Requests for GC/MS
confirmation were based on -the need to define the validity of GC/Hall or GC/
FID presumptive identification of Priority Pollutants. GC/MS confirmations of
these compounds at concentrations close to the GC/Hall and GC/FID approximate
lower levels of detection were frequently made. As a quality control measure,
GC/MS searches were also conducted for compounds not identified by GC/detector.
GC/MS was used to identify non-halogenated, base-neutral extracted hydrocarbon
Priority Pollutants at concentrations below the GC/FID lower level of detec-
tion. For the halogenated base-neutral extractable compounds, however, the
GC/MS and GC/Hall lower levels of detection were approximately the same.
Although the characterization of selected organic compounds in project
samples was the primary objective, GC/MS identification of frequently occur-
ring unknown compounds was also attempted.
Qualitative and Quantitative Determination—
Characteristic masses or mass ranges as given in the USEPA Protocol were
used for qualitative and quantitative determinations of project compounds.
Generally, in support of GC identifications at concentrations in excess of one
ug/L, extracted ion current profiles (EICP) were obtained in the scanning mode
for GC/MS confirmation or quantification of GC/Hall or GC/FID data. An EICP
is defined as a reduction of GC/MS data obtained from continuous, repetitive
measurement of spectra by plotting the change in relative abundance of the
primary or secondary ions as a function of time. A positive GC/MS confirma-
tion was based on the following conditions as recommended in the Protocol: the
time at which the peak occurred was within a retention time match of ± 1 min-
ute; a characteristic primary and secondary ion for a compound were found to
maximize in the same spectrum; and the ratio of the primary and secondary ion
agreed with relative intensities established for the compound.
In support of GC identifications at concentrations below one ug/L, GC/MS
selected ion monitoring (SIM) was used. SIM is defined as a measurement of
the GC/MS response at one or several characteristic masses in real time.
Again, a primary and secondary ion were used for confirmation in the SIM mode.
GC/MS-SIM was the approach most often used in support of GC data for pro-
ject compounds, other than the trihalomethanes in in-plant or finished water
samples, because GC data were often in the 0.1 to 1.0 ug/L range of concentra-
34
-------
tions. Identification of a recurring unknown peak in project samples was
attempted only when a concentration of approximately one ug/L was present,
because the GC/MS-scanning mode was needed for generation of a total ion
current profile.
UTILITY LABORATORIES
Sample Scheduling
Schedules were established for all organic, inorganic and bacteriological
sampling. Early in the project, each utility was visited, plant hydraulics
were discussed, sample locations were selected, and sample collection was
scheduled. Sample collection times were designed to follow the flow of a
theoretical plug of water through the plant. If dictated by changing hydrau-
lics, utility personnel modified pre-scheduled sample collection times.
Organic Sampling and Handling
The collection and handling of samples for organic analysis were done by
utility personnel using procedures specified by the project staff and sample
bottles prepared and shipped by the contract laboratory. Sample bottles were
stored at the utility in shipping containers until used. Samples were
collected according to the procedures detailed in Appendix J. Purgeable and
extractable samples were refrigerated in the dark until shipment. After addi-
tion of excess chlorine, terminal level purgeable samples were stored in the
dark for seven days at a temperature approximating that of the utility's
finished water, quenched with thiosulfate, and refrigerated in the dark until
shipment. Time in .refrigeration for all samples at the utility ranged from
one to seven days. All samples were shipped in insulated containers with
frozen ice packs via air transport to the contract laboratory. Time in
transit between the utility and the laboratory was typically one or two days
but occasionally as long as four days.
Inorganic Water Quality Analyses
At each organic sample location, waters were sampled by utility personnel
for analyses of background water quality parameters. All utilities analyzed
for physical and chemical parameters known to affect the THM reaction, i.e.,
pH, temperature, chlorine residuals. Utilities participating in THM control
studies performed additional sampling and analyses for other parameters
necessary for evaluation of the control, i.e., ammonia, turbidity, taste,
odor, iron, manganese, chlorine dioxide, etc. Methods used for measurement of
those parameters were those routinely used by the utility and detailed in
Standard Methods. •*••*• The only exception was the utilization of an analytical
procedure-*-7 for the measurement of chlorine, chlorine dioxide and chlorite in
sample waters.
Bacteriological Water Quality Analyses
Bacteriological monitoring was done by utility personnel during each THM
control study to ensure that the quality of the finished water was not compro-
mised by the treatment modification being studied. At each organic sample
35
-------
location, waters were sampled for bacteriological analyses. Total coliform
(TC) and standard plate count (SPC) analyses were performed according to
Standard Methods.11
Tests were also conducted to evaluate a membrane filter procedure using
m-SPC agar for all treated samples in which low standard plate count densities
were expected. This procedure13 permitted the examination of sample volumes
greater than one mL, the sample limitation of the SPC pour plate technique.
The procedure is detailed in Appendix K. A USEPA microbiologist visited the
utilities performing these analyses to review bacteriological procedures and
to familiarize utility personnel with the membrane filter SPC procedure.
Water quality parameters that affect disinfection conditions (turbidity,
temperature, pH, ammonia) and residual concentrations of disinfectants (chlo-
rine, chlorine dioxide) were evaluated for each bacteriological sample.
Operational Data
During THM control studies, utility personnel provided the operational
data necessary for evaluation of the control, i.e., chemical feed rates,
filter/adsorber hydraulics, filter/adsorber backwashing history, etc.
36
-------
SECTION 6
TRIHALOMETHANE TREATABILITY STUDIES
GENERAL
One project objective was to evaluate existing and modified utility water
treatment practices to control trihalomethane concentrations. Trihalomethanes
result from the reactions-*:
C12 + precursor -» CHC13
C12 + precursor + Br~ + I~ -> other THMs
To control THMs, three approaches are possible. The reaction can be allowed
to proceed with the subsequent removal of the THMs, steps can be taken to pro-
hibit the reaction from proceeding, or both approaches can be employed.
USEPA examined such controls on pilot plant and bench scales. This pro-
ject studied full scale applications of those controls to reduce TTHM concen-
trations in clear well effluents. Another aspect of the control studies was
to investigate the effect of treatment on precursor levels as measured by the
parameters THMFP and terminal TTHM. The modification implemented to control
THMs at a utility was the decision of the project staff and the utility per-
sonnel after studying the adaptability of the utility's treatment to
modification.
There were other aspects of the THM treatability studies. Evaluations of
treatment modifications were made to ensure that treatment changes did not
compromise the bacteriological integrity of the finished water. Evaluations
of halocarbons other than THMs were conducted to assess the effect of existing
and modified treatment on these compounds. Bromide and iodide concentrations
were not determined.
Finally, it was expected that water quality parameters (pH, temperature
and chlorine levels) and chemical application rates (chlorine, powdered acti-
vated carbon and chlorine dioxide) that can affect the THM reaction^'6 would
vary during the study period. Water quality data and chemical application
rates are discussed only when their variation may have had a significant
effect on THM formation.
DATA INTERPRETATION
To evaluate a treatment modification for THM control or to evaluate the
control of other Priority Pollutants, comparisons were made of the means
37
-------
of data sets, and of data from individual samples. Such comparisons were
based on statistical evaluations which determined means or individual data to
be different or to be non-differentiable.
Comparison of Mean Data
To evaluate comparatively routine and modified treatment for control of
trihalomethanes, the significance of the statistical parameters used in the
evaluation was defined.
A comparison of data from two periods of treatment, i.e., finished water
TTHM during routine and modified treatment, was based on mean values obtained
from averaging data representing the study periods. The significance of each
mean value was dependent on the variability of the set of data used in its
calculation. To establish whether the means of two distributions (study per-
iods) were different, a 90% confidence interval for the difference between
the means was calculated using a "t" distribution. The confidence interval
was established at a 90% level rather than at some greater level. Calculation
of the interval for the difference between means is based on three factors:
the number of samples representing the distributions, the variation in sample
data within each distribution, and the level of confidence at which a state-
ment of difference is to be made. Each treatment period was represented by a
relatively small number of samples. Cost and time demands for increasing the
number of samples and analyses were prohibitive. Variability of raw water
precursor over a sampling period could not be predicted or controlled. There-
fore, in order to differentiate between mean values within the design of the
study, a 90% confidence interval was chosen.
On the basis of the calculated interval, it was established for each com-
parison of means whether the values were statistically different.
Comparison of Data from Individual Samples
A detailed discussion of significance applicable to interpretation of
data produced from individual samples is presented in Section 5, pages 20 to
22. As stated in that section, a comparison of data from single samples,
i.e., adsorber influent and effluent samples collected in plug flow sequence,
was based on the significance of data obtained from analyses of numerous sets
of field replicates and of replicate analyses of single samples. The signi-
ficance of data varied for different compounds and for different concentration
ranges.
THE EFFECT OF CHLORINE APPLICATION POINTS ON TRIHALOMETHANE FORMATION
General
An examination of the THM reaction
C12 + precursor + Br~ + I~ -^ THMs
indicates that if the chlorination practice were discontinued, the reaction
would not proceed. Unless an equally effective alternative disinfectant is
38
-------
used, elimination of chlorination for THM control is not acceptable. However,
reduction of precursor levels prior to chlorination is a viable approach to
THM control. USEPA has demonstrated on the pilot plant scale that coagulation
and settling reduced precursor levels. At three project utilities, the ini-
tial chlorine application point was moved further into the treatment process
in order to reduce precursor levels prior to chlorination and to reduce the
in-plant THM reaction time. This means of control was studied at the
Pittsburgh Department of Water, the Cincinnati Water Works and the Wheeling
Water Department.
At each utility, raw, in-plant, and finished waters were sampled two to
four times weekly for periods of one to two weeks during both routine and
modified treatment studies. For each sample day, waters were sampled follow-
ing a theoretical plug from raw water through the plant to the clear well.
Pittsburgh Department of Water
Routine and Modified Treatment—
Pittsburgh routinely chlorinated untreated Allegheny River water. For
THM control, the chlorine application was moved to a point immediately follow-
ing coagulation and clarification. The utility's treatment scheme and water
quality data representative of two weeks of sampling during routine treatment
and two weeks of sampling during modified treatment are presented in Figure
12.
During the study period, 75% of the clarified water received 13 hours of
settling and the remaining 25% bypassed settling. These two waters were mixed
prior to filtration. During modification, water influent to the filter was a
mix of chlorinated settled water and unchlorinated clarified water.
Evaluation of Trihalomethane Control—
Instantaneous and terminal TTHM concentrations based on data from two
weeks of sampling with raw water chlorination and from two weeks with clari-
fied water chlorination are illustrated in Figure 13.
A statistical comparison of mean terminal TTHM concentrations indicates
that raw water precursor levels could not be differentiated during the two
study periods.
For both study periods, raw and clarified mean terminal TTHM concentra-
tions could be differentiated, but clarified and finished mean terminal TTHM
concentrations could not be differentiated. Thus, coagulation and clarifica-
tion reduced precursor levels but subsequent treatment likely did not.
Figure 12 indicates that mean raw water turbidity levels of 7.2 and 7.1
NTU were comparable over the two study periods and that coagulation and clari-
fication reduced turbidities to mean levels of 0.8 and 1.0 NTU. These data
show that as coagulation and clarification reduced turbidity, it also reduced
precursor levels; however, when turbidities fell below 1.0 NTU, further reduc-
tion in precursor levels could not be observed.
Mean raw water turbidity and raw water terminal TTHM concentrations were
39
-------
.e-
o
I i L-\*y r*
III
I
u
_J
<
-1 (1) MIX' AN
^| t MIX CLAR
,G -@ «^»
D ' ' 1' — FiiTirc1
IFY- SETTLE zT5VJ nLTCR
1 1 'il—..,. 1 ^x75o/
(CHLORINE) LIME /Vui /U,^,- \
[ROUTINE ^LUM (CHLORINE)
TREATMENT] R PAC [^R^K>ENT]
FeCI3
POLYMER
-
PARAMETER (T)
TIME.HR O
TEMP,°C 24/21
TURB.NTU 7.2/7.1 O
pH 7.1 /7J 8
FREE CI2l PPM - 0.
TOTAL CI2, PPM - 0.
TC/lOOmL 62OO/fc3OO
© ©
2 15
_
8/1. 0 O.fe/O.^
.8/8.8 8.8/8.8
2/<0.ltt 0.2/0-3
6/0-2°- 0.3/O.fe
-------
-T75
2
o
h
2
UJ
o
z
o
u
UJ
- •
f\J
II
o.
U-
5
I
h
LEFT= ROUTINE TREATMENT
~
- ••
• ••—
27 RIGHT = MODIFIED TREATMENT
222 223
Rfcr—
—
J
IO
-• •*
_j —
•^
_ j
• "
.— —
w
11
0-
U-
z
I
h
2(5
31
,- —
• -~
-"Z?
n^"^
$
2O2
__—
jf ^
30
2.8
_ —
^^
^^**
^^
^-*
—
i —
=T2
2O3
2.1
_ —
^»
i ^
^--"
^--'
^
^
^
_ —
r '
^
^~
^
'
^
2OT
= THMFP
= INST TTHM
[""I = TERM TTHM
RAW
t
COAG AND
CLARIFIED
SETTLED
1.2 PPM CI2
O.4 PPM PAC
G
r t t
4.8 PPM PAC 0.5 PPM CI2
FILTERED FINISHED
t
2.4 PPM Clj,] ROUTINE TREATMENT
2.1 PPM CI21 MODIFIED TREATMENT
Figure 13. Trihalomethane formation (mean values), Pittsburgh Department of Water,
228,000 cu m/day (60 MGD), September - October 1978.
-------
comparable during the two study periods; however, on a day-to-day basis, both
fluctuated and not always in the same direction.
As shown in Figure 13, chlorination of raw water with a mean THMFP of 274
ug/L (275 ug/L term TTHM - 1 ug/L inst TTHM = 274 ug/L THMFP) resulted in 56
ug/L mean instantaneous TTHM in the finished water. Chlorination of clarified
water with a mean THMFP of 214 ug/L resulted in 26 ug/L finished water TTHM.
There was a reduction in the percent formation of finished water TTHM from
available raw water precursor. Of the THMFP available in the raw water, 20%
reacted to form TTHM in the finished water during routine treatment (56 ug/L
finished water inst TTHM/274 ug/L raw water THMFP). During modified treat-
ments, 10% of the available THMFP in the raw water reacted to form finished
water TTHM. Moving the chlorine application point to a location with reduced
THMFP resulted in significantly lower finished water trihalomethane
concentrations.
Other factors that may have contributed to lowered TTHM formation were a
reduction of two hours in available THM reaction time and a slight reduction
in chlorine feed. The effect of mixing 75% chlorinated settled water with
25% unchlorinated clarified water could not be evaluated because the resulting
instantaneous TTHM concentrations were low and could not be differentiated.
Moving the chlorine application point to a water with reduced chlorine
demand also resulted in a savings in chlorine feed (1.2 mg/.L to 0.5 mg/L) when
attempting to maintain 0.1 mg/L free chlorine in the water applied to the
filters.
A taste and odor incident related to the source water necessitated an
increase in the use of powdered activated carbon (PAC) during the period when
chlorine was applied to clarified water. This increase (0.4 mg/L mean to 4.8
mg/L mean) apparently contributed little to the reduction in finished water
TTHM concentration. Although 4.8 mg/L was the mean PAC feed for the period,
the range was 2.0 to 6.7 mg/L. These varying PAC feeds had no significant
effect on instantaneous TTHM concentrations (2.7 ug/L mean) for filtered
water, i.e., the last process location where water-PAC contact could have
occurred. The effect of increased PAC (0.4 mg/L mean to 4.8 mg/L mean) on
precursor levels could not be assessed. Reduction in precursor levels was, to
a great extent, attributable to coagulation and clarification; the effect, if
any, of PAC could not be evaluated.
As shown in Table 7, a change in the chlorine application point had no
significant effect on the ratio of individual THM compounds found in the
finished water.
Evaluation of Other Priority Pollutants—
This study was conducted in September of 1978 following the year during
which monthly sampling was conducted at all utilities. Monthly data had indi-
cated that other volatile and extractable halocarbons (Tables 1 and 2)
occurred infrequently at Pittsburgh at concentrations where precision of field
data was highly variable (Appendices C and E); therefore, analyses for these
compounds were not conducted during this study.
42
-------
TABLE 7. RATIO OF INDIVIDUAL TRIHALOMETHANES TO TOTAL TRIHALOMETHANES
IN THE CLEAR WELL (%), PITTSBURGH DEPARTMENT OF WATER
(INSTANTANEOUS MEAN VALUES) =======
Treatment
Routine Modified
(raw water chlorination) (clarified water chlorination)
Compound (September 6-16. 1978) (September 20-October 2, 1978)
Chloroform 19% 22%
Bromodichloromethane 29% 27%
Dibromochloromethane 37% 38%
Bromoform 15%
56
26
us/L
aGC/Hall detector
For a short period in September, however, a volatile halocarbon which was
GC/MS confirmed as tetrachloroethylene was reported at high concentrations.
These data, shown in Table 8, indicate a passing slug of the compound in the
river and show a reduction in concentration when comparing raw and finished
waters. However, the variable nature of a slug, the precision of tetrachloro-
ethylene concentrations at the reported levels (± 25% to ± 32%, Appendix C,
Table C-15), and the fact that data exist for only two sample sequences
suggest caution in concluding that treatment lowered tetrachloroethylene
concentrat ions.
TABLE 8. TETRACHLOROETHYLENE CONCENTRATIONS
PITTSBURGH DEPARTMENT OF WATER
Concentration^, ug/L
Water
Raw
Clarified
Settled
Filtered
Clear Well
Sep 6-13 Sep 14-15
<1.0 64b
20
27
29
22
Sep 15-16
17
27
11
11
8.2
Sep 20-Oct 2
<1.0
aGC/Hall detector
t"GC/MS confirmed as tetrachloroethylene
Evaluation of Trihalomethanes in Finished Water Open Reservoir—
Pittsburgh has three large open reservoirs in the distribution system.
One of these was sampled for this study. It has been described in detail by
other researchers.14 The irregular shape and numerous effluent points made it
difficult to select sampling times based on plug flow through the reservoir.
The tetrachloroethylene incident demonstrated that the selected times for re-
servoir sampling relative to times for clear well sampling were in error. Two
phenomena may have affected instantaneous TTHM levels in the open reservoir:
increased formation from periodic chlorination and volatilization of THMs to
the atmosphere. Instantaneous and terminal TTHM concentrations for the clear
well and reservoir are given in Table 9. Statistical comparison of these mean
43
-------
TTHM concentrations indicates no difference between clear well and open reser-
voir waters but it should be noted that sample times for these two locations
were not in plug flow agreement.
TABLE 9. TTHM CONCENTRATIONS,41 ug/L, PITTSBURGH DEPARTMENT OF WATER
(MEAN VALUES)
Water
Parameter Clear Well
inst TTHM
term TTHM
inst TTHM
term TTHM
56
203
26
207
Open Reservoir Treatment
53
197
27
210
Routine (raw water chlorination,
September 6-September 19)
Modified (clarified water chlorination,
Bacteriological Evaluation—
A comparison of the bacteriological conditions during the two periods of
sampling was made to ensure that treatment modifications did not compromise
the bacteriological integrity of the finished water. Total coliform and
standard plate count densities obtained for both periods are presented in
Figure 12. These data indicate that raw water chlorination resulted in a re-
duction of the mean raw water total coliform density from 6,200/100 m'L to
<1/100 mL after clarification. A similar reduction of raw water total coli-
form density from mean values of 6,300/100 mL to <1/100 mL is indicated after
clarification without raw water chlorination. Thus, clarification in combina-
tion with application of powdered activated carbon and permanganate was as
effective in coliform reduction as raw water chlorination and clarification in
combination with PAC and permanganate application. Although permanganate was
applied at approximately 1 mg/L for manganese control during the study, it
probably contributed to disinfection.
The chlorine disinfection conditions were more favorable during modified
treatment because chlorine was applied to a clarified water of one turbidity
unit as compared to the routine application of chlorine to a more turbid raw
water.
The delay in chlorine application caused a parallel delay in reduction of
the general bacterial population as measured by the standard plate count.
After the processes of chlorination and clarification, the mean standard plate
count density was 31 bacteria/mL; after clarification alone, the mean density
was 230/mL.
The quality of the finished water was not altered by the delay in chlor-
ination. During both periods of study, bacteriological conditions in the
finished water were satisfactory, i.e., total coliform and standard plate
count densities complied with the 1975 USEPA Interim Drinking Water Standard
of ^l coliform colony/100 mL and the recommended limit for the standard plate
count of <500 organisms/mL.
Findings—
1. Trihalomethanes were formed during treatment after chlorine was
44
1 c
-------
applied.
2. As clarification reduced turbidity to 1.0 NTU, it also reduced pre-
cursor levels. When turbidities fell below 1.0 NTU, further reduction in pre-
cursor levels could not be observed.
3. Moving the chlorine application point from raw water to clarified
water resulted in chlorinating a water of lower THMFP.
4. Moving the chlorine application point to a better quality water in
terms of reduced THMFP resulted in significantly lower finished water trihalo-
methane concentrations.
5. Moving the chlorine application point from raw water to clarified
water resulted in a savings in chlorine feed.
6. Moving the chlorine application point reduced the in-plant THM reac-
tion time 6% and had no significant effect on the ratio of individual THM
compounds found in finished water.
7. A tetrachloroethylene spill was observed on the Allegheny River with
concentrations in the plant reaching 60 ug/L.
8. Permanganate, flocculant, and PAC application followed by clarifica-
tion were as effective in coliform reduction as chlorine applied with the
other materials prior to clarification.
9. Moving the chlorine application point caused a delay in reduction of
the general bacterial population as measured by the standard plate count, but
the bacterial quality of the finished water was not altered.
Cincinnati Water Works
Routine and Modified Treatment—
The city of Cincinnati stores Ohio River water in a large, open reservoir
where it is treated with a coagulant. Other treatment chemicals and chlorine
are routinely added ahead of in-plant treatment processes. Relocation of this
chlorine application point to the effluent from the settling basins was
studied. The treatment schematic and water quality data representing two
weeks of routine treatment sampling and two weeks of modified treatment samp-
ling are presented in Figure 14.
Evaluation of Trihalomethane Control—
A problem at the contract laboratory resulted in a considerable loss of
project samples collected during September and October 1977—the time of this
study. Consequently, instantaneous and terminal TTHM data presented in
Figure 15 are mean values for 80% of the samples collected during routine
treatment and 60% of the samples collected during modified treatment.
A statistical comparison of mean terminal TTHM concentrations indicated
a difference in raw water precursor levels between routine and modified treat-
ment study periods. During the two-week period when reservoir settled raw
45
-------
[ROUTINE TREATMENT]
(CHLORINE)
COAG AND
SETTLE
(CHLORINE)
[MODIFIED
TREATMENT]
PARAMETER
TIME.HR -48
TEMP, °C |8/22
TURB, NTU 32/14
pH 7.3/7.6
FREECI2(PPM
TOTAL C \2 , PPM
TC/JOOmL 960O/84OOO
SPC/mL
O
i.o/o.s
1.0/7.2
22O/24OO
LEGEND
ROUTINE TREATMENT/MODIFIED TREATMENT
Q= SAMPLE POINT
(OPTIONAL FEED)
I.2//.0
8.5/8.1
w -
-------
Z
o
Z
UJ
o
2
O
o
UJ
343
355
$309
338
215
224-
—.232
IO&
40
65
RAW , RESERVOIR SETTLED FILTERED
f SETTLED f
4.8 PPM PAC 4.8 PPM PAC 1
CIZJ
FINISHED
ROUTINE TREATMENT
r 1
|4.8PPMPAC 4.8PPM PAC 3.3 PPM C»2 MODIFIED
I J ' »^ t r^ I I
LEFT= ROUTINE TREATMENT
R.ISHT « MODIFIED TREATMENT
= INST TTHM
EAT ME NT
=THMFP
= TERM TTHM
Figure 15. Trihalomethane formation (mean values), Cincinnati Water Works,
560,000 cu m/day (150 MGD), September - October 1977.
47
-------
water was chlorinated, the mean raw water terminal TTHM concentration was 508
ug/L and mean raw water turbidity was 32 NTU. During the two-week period when
in-plant.settled raw water was chlorinated, the mean raw water terminal TTHM
concentration was 309 ug/L and the mean raw water turbidity was 14 NTU.
During the four-week period, reservoir settling reduced turbidity to
levels of approximately 1.0 NTU. At the same time, reservoir settling reduced
precursor levels an average of 31% (mean terminal TTHM from 508 ug/L to 343
ug/L during the two-week period of routine treatment and mean terminal TTHM
from 309 ug/L to 215 ug/L during the two-week modified treatment period).
During the four-week period, subsequent treatment, including in-plant coagula-
tion and settling, did not significantly reduce precursor levels. During the
routine treatment period, mean terminal TTHM concentrations of 343 ug/L and
338 ug/L could not be differentiated. During modified treatment, mean term-
inal TTHM concentrations of 215 ug/L and 232 ug/L could not be differentiated.
Thus, 48 hours of alum enhanced reservoir settling reduced precursor levels
but subsequent treatment, including in-plant coagulation and settling, had
little, if any, effect on precursor levels. These data suggest that as reser-
voir settling reduced turbidity it also reduced precursor levels but that when
turbidities had been reduced to levels of approximately 1.0 NTU, further re-
duction in precursor levels could not be observed.
Figure 15 indicates that when chlorinating reservoir settled water with
a mean THMFP concentration of 342 ug/L, a mean of 106 ug/L instantaneous TTHM
resulted in the finished water. When chlorinating in-plant settled water with
a mean THMFP concentration of 223 ug/L, a mean of 65 ug/L TTHM resulted in the
finished water. While it appears that moving the chlorine application point
to a better quality water in terms of THMFP resulted in reduced finished water
TTHM concentrations (106 ug/L to 65 ug/L), an inspection of the percent forma-
tion of finished water instantaneous TTHM from available raw water precursor
indicates that a reduction did not likely result. Of the 507 ug/L THMFP
available in the raw water during the period of routine operation, 21% formed
finished water instantaneous TTHM (106 ug/L finished water inst TTHM/507 ug/L
raw water THMTP). Of the 308 ug/L raw water THMFP available during the
period of modified treatment, 21% again reacted to form TTHM in the finished
water. These data suggest that the reduction in finished water instantaneous
TTHM during modified treatment was attributable to significantly lower raw
water precursor levels during that period. During both routine and modified
treatment, significant reduction in precursor did not occur beyond reservoir
settling. Thus, moving the chlorine application point to an in-plant settled
water resulted in chlorinating a water of lower THMFP only because precursor
levels were significantly lower during that time. The decrease in THM reac-
tion time from 7% to 3^ hours had no apparent effect in limiting THM formation
because per cent formation relative to raw water precursor was unchanged.
These data demonstrate the importance of the terminal TTHM and THMFP
parameters in evaluating trihalomethane control and suggest the need for
further investigation to understand the effect of both the variability of raw
water precursor levels and treatment processes on finished water TTHM levels.
Moving the chlorine application point resulted in a slight savings in
chlorine feed (3.6 mg/L to 3.3 mg/L) when attempting to maintain 1.5 mg/L free
48
-------
chlorine effluent from the plant.
The change in the chlorine application point had an effect on the ratio
of THMs found in finished water. Brominated THMs were relatively more preva-
lent when chlorinating in-plant settled water. This was probably attributable
to the difference in THM reaction time. Other factors include the variable
nature and concentration of the precursor, the effect of unknown raw water
bromide concentrations, and the uncertain role of bromine in the THM reaction.
Table 10 shows individual compounds as percentages of finished water TTHM.
TABLE 10. RATIO OF INDIVIDUAL TRIHALOMETHANES TO TOTAL TRIHALOMETHANES
IN THE CLEAR WELL (%), CINCINNATI WATER WORKS
(INSTANTANEOUS MEAN VALUES)
Treatment
Routine Modified
(chlorination of reservoir (chlorination of coagulated
Compound settled raw water) and settled water)
Chloroform
Bromodichloromethane
Dibromochloromethane
Bromof orm
Dichloro iodomethane
inst TTHMa
59%
28%
12%
<1%
<1%
106 ug/L
41%
33%
22%
3%
<1%
65 ug/L
^GC/Hall detector
Evaluation of Other Priority Pollutants—
For this study analyses were performed for volatile halocarbons other
than THMs and for base-neutral extractable halocarbons. These compounds were
found infrequently at Cincinnati and at low concentrations where precision of
field data was highly variable. An evaluation of the effect of the change in
chlorine application point on these compounds could not be made. These com-
pounds will be discussed as a part of the year-long survey for Priority
Pollutants in Section 7.
Bacteriological Evaluation—
Bacteriological data were evaluated during both routine and modified
treatment periods and are presented in Figure 14. During both periods of the
study, 48-hour, alum enhanced, reservoir settling effectively reduced total
coliform densities 97%. However, an evaluation of the treatment modification
involved a comparison of bacteriological data obtained from reservoir settled
water with data obtained from in-plant settled water.
During routine chlorination of reservoir settled water, the mean total
coliform density was reduced from a reservoir settled value of 220/100 mL to
<1/100 mL in the in-plant settled water. During the modified period when
chlorine was applied after in-plant settling, the mean coliform density was
reduced from a reservoir settled value of 2,400/100 mL to 1,400/100 mL in the
in-plant settled water; a mean density of <1/100 mL was not obtained until the
in-plant settled water was chlorinated and filtered.
A similar delay in the reduction in standard plate count densities
49
-------
occurred with the delay in chlorination. A mean density of 5,500 bacteria/mL
in the in-plant settled water without chlorination compared with a mean den-
sity of 500/mL in the in-plant settled water when chlorinated.
The delay in chlorination resulted in a parallel delay in reduction of
bacterial densities until chlorine was applied. This delay resulted in no
significant change in the bacterial quality of the finished water and resulted
in no apparent in-plant problems.
Findings—
1. Trihalomethanes were formed during treatment after chlorine was
applied.
2. Forty-eight hours of alum coagulated, reservoir settling reduced
turbidity to 1.0 NTU, and also reduced precursor levels. When turbidities
fell below 1.0 NTU, further reduction in precursor levels could not be
observed.
3. Raw water precursor levels were significantly lower during modified
treatment than during routine treatment. Because reduction in precursor
levels could not be observed following reservoir settling, moving the chlorine
application point from reservoir settled water to in-plant settled water
resulted in chlorinating a water of lower THMFP only because precursor levels
were lower during that period.
4. Significantly lower finished water trihalomethane concentrations
resulted during modified treatment presumably because precursor levels were
lower during that period.
5. Moving the chlorine application point resulted in some savings in
chlorine feed.
6. Moving the chlorine application point reduced the in-plant THM reac-
tion time 53% and had a significant effect on the ratio of individual THM
compounds found in finished water; brominated THM concentrations were rela-
tively higher.
7. Forty-eight hours of alum coagulated, reservoir settling reduced
coliform densities 97%.
8. Moving the chlorine application point caused a delay in reduction of
bacterial densities, but the bacterial quality of the finished water was not
altered.
Wheeling Water Department '
Routine and Modified Treatment—
Wheeling routinely chlorinated a gravity settled Ohio River water. For
purposes of THM control, the chlorination point was moved to coagulated and
settled water. Iron and manganese removal was accomplished by chlorine oxi-
dation, coagulation, settling and -filtration during routine treatment. When
treatment was modified, the utility added permanganate as a substitute oxidant
50
-------
for chlorine. Water quality data representative of two weeks of routine
treatment and two weeks of modified treatment are presented in Figure 16 with
the treatment schematic. Figure 17 presents mean instantaneous and terminal
TTHM data for both periods of study.
Evaluation of Trihalomethane Control—
The trend of individual terminal TTHM data indicated raw water precursor
levels were lower during routine treatment than during modified treatment.
During either study period, a statistical comparison of mean values indicated
that raw water terminal TTHM and gravity settled terminal TTHM could not be
differentiated; therefore, one hour of gravity settling did not reduce pre-
cursor levels. Gravity settling did not reduce turbidity levels.
During routine treatment, gravity settled and coagulated and settled mean
terminal TTHM concentrations (325 ug/L and 265 ug/L, respectively) could be
differentiated. Mean terminal TTHM concentrations in coagulated and settled
and finished water (265 ug/L and 273 ug/L, respectively) could not be differ-
entiated. Thus, coagulation and settling reduced precursor levels but sub-
sequent treatment likely did not. Turbidity levels were reduced by coagula-
tion and settling and by filtration.
During modified treatment, gravity settled and coagulated and settled
mean terminal TTHM concentrations (371 ug/L and 347 ug/L, respectively) could
not be differentiated but gravity settled and finished water mean terminal
TTHM concentrations (371 ug/L and 324 ug/L, respectively) were different.
Thus, coagulation and settling was not as effective for precursor removal
during modified treatment. The reason for this is not known. Turbidity
levels were reduced by coagulation and settling and by filtration.
Because coagulation and settling was not as effective in lowering pre-
cursor levels during modified treatment and because raw water precursor levels
during that period were somewhat higher, moving the application point did not
result in chlorinating a water with lower THMFP (324 ug/L and 346 ug/L could
not be differentiated).
However, lower instantaneous TTHM were formed in the finished water dur-
ing the modification (152 ug/L compared to the modified value of 104 ug/L).
This was a significant reduction in the percentage formation of TTHM from
raw water precursor; 47% during routine treatment compared to 28% during the
modification. Thus, moving the chlorine application point resulted in lowered
finished water TTHM, not because a better quality water was chlorinated, but
because the THM in-plant reaction time was decreased from 4% to 1% hours.
Although pH levels ranging from 8.9 to 9.7 were a major factor in the
formation of 104 ug/L TTHM in only 1% hours, other factors, such as chlorine
application rate, species of residual chlorine, and the nature and concentra-
tion of precursor, may have affected the reaction rate.
The change in the chlorine application point increased the percentages
of the brominated THMs with a corresponding decrease in chloroform formation
(Table 11). This was probably attributable to a reduction in the THM reaction
time. Other factors include the variable nature and concentration of the pre-
51
-------
Ul
N5
[MODIFIED TREATMENT]
<* -/TV GRAVITY _/5vL-
^T-Vi/" SETTLE :Wf=
£ '
TlNjIF- °
MIX T AERATE
1
HR
COAG AND _/Tv_
SETTLE v^T1
PAC AIR (CHLORINE) 1
o LIME [ROUTINE (CHLORI
i Fe2(so4)3 TREATMENT] [MODIFIE
O TREAT N/
PARAMETER (T)
TIME , HR 0
TEMP,°C M/|2
TURB, NTU 12/^.q
pH 7-4/7-3
FREE CI2, PPM
TOTAL C I2 , PPM
©
-
IO/9.6
7.5/7.4
-
-
TC/IOOmL 75OO/77OO 8IOO/6TOO
SPC /m L
Fe, PPM 0.8/0.7
Mn, PPM 0.8/0.8
I2OOO/8IOO
-
(D
5
-,
4.2/7.O
^.t/'l.T
O.4-/0.la
2.1/0.1*
1 /88O
O.6> /O.<0
O.4/O-3
PLUORI DE
: FILTER =(4)=i= Cwi
^px CHLORINE
ENT] s
0
5.5
-
0.3/0.5
^.6/9.4
O.3/<0.
1.7/2.1
/
1 /2
.
-
EAR
=1LL
TIME =
•- 6.25 HR
UCTION _/-\
WELL =V3>=
©
65
_
0.4/0. 5
q.3 /g.
-------
2
0
l-
Z
UJ
U
z
0
u
2
-------
cursor, the effect of unknown raw water bromide concentrations, and the uncer-
tain role of bromine in the THM reaction.
TABLE 11. RATIO OF INDIVIDUAL TRIHALOMETHANES TO TOTAL TRIHALOMETHANES
IN THE C'LEAR WELL (%) , WHEELING WATER DEPARTMENT
(INSTANTANEOUS MEAN LEVELS)
Treatment
Routine Modified
(chlorination of gravity (chlorination of coagulated
Compound settled raw water) and settled water)
Chloroform 36% 23%
Bromodichloromethane 30% 31%
Dibromochloromethane 25% 34%
Bromoform 9% 12%
Dichloroiodomethane <1% <1%
inst TTHMa 152 ug/L 104 ug/L
detector
Chlorine application was based on maintaining a 0.3 mg/L free chlorine
residual onto the filters and a 2.0 mg/L finished water residual. No savings
in total chlorine application resulted from the modification.
The data indicate that modified treatment with oxidation by permanganate
was as effective for iron and manganese control as routine treatment with oxi-
dation by chlorine. The effect, if any, of permanganate on precursor could
not be separated from the effect of coagulation and settling.
Evaluation of Other Priority Pollutants—
This study was conducted in November of 1978 following the year-long
period of monthly sampling. Annual data indicated infrequent and low level
occurrence of other halocarbons. For this reason, analyses of these compounds
were not performed during this THM control study.
Bacteriological Evaluation—
Bacteriological levels were evaluated during both routine and modified
treatment periods and are presented in Figure 16.
The data indicate that chlorination of gravity settled raw water resulted
in a complete reduction of the mean total coliform density from 8,100/100 mL
in the gravity settled raw water to <1/100 mL in the coagulated and settled
water. During modified treatment a significant reduction also occurred. A
mean total coliform density of 6,700/100 mL in the gravity settled raw water
was reduced to 12/100 mL in the coagulated and settled water without chlorina-
tion, when one hour raw water gravity settling and application of permanganate
preceeded four hours of coagulation and settling. This combination of pro-
cesses resulted in a significant reduction of coliform organisms; however,
reduction to <1/100 mL was achieved only after chlorine was applied to the
coagulated and settled water.
The delay in chlorine application during modified treatment caused a
parallel delay in reduction of the standard plate count. After chlorination
54
-------
of gravity settled raw water, the mean standard plate count density was 9
bacteria/mL in the coagulated and settled water; after coagulation and sett-
ling without chlorination, the mean density was 880/mL. However, the mean SPC
density was effectively reduced to 2 bacteria/mL after chlorination of the
coagulated and settled water.
The quality of the finished water was not altered by the delay in chlor-
ination during modified treatment. During both periods of the study, bacter-
ial densities in the finished water complied with USEPA Interim Drinking Water
Standards.
Findings—
1. Trihalomethanes were formed during treatment after chlorine was
applied.
2. Raw water precursor levels were higher during modified treatment than
during routine treatment.
3. One hour of gravity settling did not reduce precursor levels. Coagu-
lation and settling were more effective for precursor removal during routine
treatment than during modified treatment.
4. One hour of gravity settling did not reduce turbidity levels. Turbi-
dity levels were reduced by coagulation, settling and filtration.
5. Moving the chlorine application point from gravity settled water to
coagulated settled water did not result in chlorinating a water of lower THMFP
because raw water precursor levels were higher during that period.
6. Significantly lower finished water trihalomethane concentrations
resulted during modified treatment presumably because THM in-plant reaction
time was reduced 67%.
7. Moving the chlorine application point and reducing the in-plant THM
reaction time 67% had a significant effect on the ratio of individual THM com-
pounds found in finished water; brominated THM concentrations were relatively
higher.
8. Moving the chlorine application point caused a delay in the reduction
of bacterial densities, but the bacterial quality of the finished water was
not altered.
9. Coagulation, settling and permanganate application significantly
reduced coliform and standard plate count densities.
THE EFFECT OF AMMONIATION ON TRIHALOMETHANE FORMATION
General
Bench scale studies have shown that combined chlorine species form tri-
halomethanes at a much slower rate than do free chlorine species.5 Conversion
of free chlorine to combined chlorine was a THM control evaluated full scale
55
-------
at the Louisville Water Company by adding ammonia as a treatment modification.
Raw, in-plant and finished waters were sampled two or three times weekly
for periods of one to two weeks during both routine and modified treatment
studies. For each sample day, sampling followed theoretical plug flow through
the plant.
Louisville Water Company
Routine and Modified Treatment—
Chlorine was routinely applied to gravity settled raw water and to the
clear well. Modified treatment evaluated the application of ammonia first to
the clear well and second to the "softening" basins. Lime-soda softening was
practiced during periods when raw water total hardness exceeded 140 mg/L.
During the period when routine treatment was studied, softening was
practiced. During the period when ammonia was applied to the clear well,
softening was practiced intermittently. Softening was off-line during the
final period of study when ammonia was applied to the softening basins. The
treatment schematic is presented in Figure 18. Each ammonia application point
was preceeded by a chlorine application point so that chloramines were not a
primary disinfectant.
Evaluation of Trihalomethane Control—
TTHM concentrations and water quality data presented in Figure 19 are
representative of the period when softening was practiced and ammonia was not
applied. Mean instantaneous TTHM data indicate formation of trihalomethane
resulting from chlorination and enhanced by an increase in pH in the softening
basins.
Significant reduction in precursor levels was not observed in-plant when
mean terminal TTHM concentrations were evaluated. Evaluation of terminal
level TTHM data should be made cautiously when finished water pH is lower
than the pH of some in-plant waters. Waters stored for the determination of
the terminal TTHM parameter were buffered to pH 8.3 to maintain finished water
pH. Softened and filtered water samples collected for TTHM determinations
represented several hours of instantaneous TTHM formation at pH 9.2. The rate
of THM formation is pH dependent.5 It is, therefore, possible for the term-
inal TTHM concentrations of softened and filtered waters to exceed the term-
inal TTHM concentration of settled water because of the instantaneous TTHM
formed at the accelerated rate during treatment.5 This difference in reaction
rate as a function of pH was demonstrated for the utility's settled water
(Figure 20).
Water quality data and TTHM concentrations representative of the period
when softening was practiced intermittently and ammonia was applied to the
clear well are presented in Figure 21. Mean instantaneous TTHM data indicate
formation of trihalomethane resulting from chlorination and enhanced by an
increase in pH in the softening basins. Statistical comparison of means indi-
cated that softened, filtered and finished instantaneous TTHM levels could not
be differentiated. Thus, there was no significant increase in THM formation
during the one half hour through the filter and no significant increase in the
56
-------
<£
UJ
>
t£
i
0
(CHLORINE)
1 /•""" ^^
f~^ 1 / OPEN V — x-x x COAG AND —
"^— ^ i ^i?f=e>Fl?\/ni £ 1 — 4 4. SETTLE
TV y i i
(pAvC) CHLORINE (PAvC)
x ALUNA
(POLYMER)
(C02)
1 — MIX — SETTLE: — i
(50 FT EN) Lry-
J— I W _
LirJtC 67% ^SOFTEN^
(SODA ASH) (cta)
(NH3) t
O=4
V^ '
1 1 i i r\ O r^» C
r LUC/ K Ut
u f. , ^^ o
Fl l "T" P (? { 1 i • vK/crl 1
i v. i i— !-. ^^/ » VV C. 1— U
i
., x T
(NH3) I
\ O/ 1
1
CHLORIb
.EGEND
Q SAMPLE POIMT
FEED)
Figure 18. Treatment at Louisville Water Company,
473,000 cu m/day (125 MGD), July - October, 1977,
-------
Ln
00
ENTRAT
n-T
?£
u -*-
2
<
UJ
S
_— -•
— —
OOT
r
^
355
-
— -1
,"*
k\\\\\\
3t4
~7cj
.^-^
r — --
x
1
328
14-
-——
^
^
^
X
X
X
rx
*3yL A
O^T^
:
»2O
^— '
.-— -
x
X
rx
X
X
X
x
x^
jS
"? 1^
O*0
-
I O Q
RAW
PARAMETER
TIME, HR
TEMP, °C
TURB, WTU
pH
FREE CI£lPPM
TOTAL CI2(PPM
NH3( PPM
TC /|00 mL
SPC mL
^i^v'ITY 1 COAG AND.
SETTLED I SETTLED f
CHLORIME LIME
SODA ASH
SOFTENED
FILTERED ^ FINISHED
CHLORINE
5.5
•b.-ZS
6.5
7-5
2.1
7-*2
2.5
3.0
-
5.O
*».-z
O-fe
0.7
-
0.-2
9.1
0-3
0-4
O.I
0.2
8.5
(-a
£.0
O.I
10
I
< I
II 4- |
INST TTHM | | = THMFP
= TERM TTHM.
Figure 19. Trihalomethane formation (mean values), water quality data (mean values), Louisville
Water Company, 473,000 cu m/day (125 MGD), no ammoniation, softening on-line, July 1977.
-------
300.
250-
2 200
o
2
IU
O
z
o
I
h
IOO-
50 .
r i
03
REACTION! COMOITIOMS'-
A = SETTLED WATER
ADDED CHLORINE = 15 mg/L
TEMPERATURE = I6°C
• pH * 8.3
a pH = 9.2
24 48 12
REACTION TIME , HOURS
144
Figure 20. Effect of pH on trihalomethane formation.
59
-------
2
O
t-
1-
z
UJ _j
o T*
o <
2
IU
- — •
TT
315
^ w
— —
J— •—
Z92
r^ * n
--—
.
x'x
X
£1<0O
-
15
--—
. —
X
X
x
-X1
2e(4
^
-
141
RAW GRAV TY A COAG, AND i SOFTENED
SETTLED 1 SETTLED |
PARAMETER
CHLORINE , LIME
(SODA ASH)
TIME, HR -8-25 0 3 ^"~«^ 5
TEMP, °C
TURB, MTU IS 23 4-1 3-8
pH 1-3 7-5 l-O q.2
FREE Ct^, PPM - - 2.6> O-6>
TOTALCI2tPPM - - 2-8 0.1
NH-5, PPM O.2 0.3
TC/ioOmL 32OO 49OO <\ <\
5PC /ro L - - < | 5i
LEGEND
(FEED OM-UNE/OFF LINE)
X
= IM6
>T
^•>
x^
215
0 / /-
t-
-------
clear well after ammonia had been applied.
The trend of terminal TTHM data represented by mean concentrations in
Figure 21 indicated reduction in precursor between raw and finished water.
Data representative of the period when softening was not practiced and
ammonia was applied to the softening basins are presented in Figure 22.
A problem at the contract laboratory resulted in a considerable loss of
project samples collected during October of 1977—the time of this last phase
of ammonia application. Consequently, THM data presented in Figure 22 repre-
sent 60%-80% of the samples collected.
Detention time in the open reservoir was longer during this period of
"softening" basin ammoniation than during previous periods (22 hours compared
to eight hours), because part of the reservoir had earlier been off-line.
The reservoirs were chlorinated intermittently during this period for
algal control resulting in 9.6 ug/L mean instantaneous TTHM. Chlorination of
settling basins increased TTHM to 65 ug/L. Sufficient ammonia was applied to
two-thirds of the "softening" basins to carry an ammonia residual to the dis-
tribution system. Because one-third of the basins were not ammoniated, the
THM reaction proceeded until these waters were mixed. Mean "softened" water
TTHM therefore reached 84 ug/L. On the non-ammoniated side, the pH was 7.9;
on the ammoniated side, it was 9.3. No further THM formation was observed
across the filter. A statistical comparison indicated that a mean of 83 ug/L
TTHM in the filtered water and a mean of 94 ug/L TTHM in the finished water
could not be differentiated. Thus, the TTHM formation proceeded in the plant
as a result of chlorination. However, little further increase in TTHM
resulted in waters subsequently treated with ammonia.
Comparisons of mean terminal TTHM concentrations (Figure 22) indicated
that raw and gravity settled mean concentrations were different, that gravity
settled and coagulated settled mean concentrations could not be differentia-
ted, and that coagulated settled and finished mean concentrations could not
be differentiated. Thus, 22 hours of gravity settling reduced precursor
levels but subsequent treatment probably did not.
During the three periods of study, significant precursor level reduction
was observed only during 22-hour gravity settling. Turbidity reduction, how-
ever, occurred during coagulated settling, not during gravity settling. The
relationship between turbidity levels and precursor levels suggested by other
utility studies was not supported during this study.
Ammoniation had no significant effect on the ratio of individual THM
compounds in the finished water. Table 12 shows individual compounds as per-
centages of instantaneous TTHM.
Evaluation of Other Priority Pollutants—
For this study analyses were performed for volatile halocarbons other
than THMs and for base-neutral extractable halocarbons. These compounds were
found infrequently at Louisville and typically at low concentrations where
61
-------
.244
TIME, HR
TEMP °C
5.5
, MTU
pH
FREE C!2(PPM
TOTAL C (2( PPM
-I PPM
TC / 10 O m L
5 PC /ml
15
1-5
O.I
4000
18
0. t
1100
KJOTES
a RANGE -
-------
TABLE 12. RATIO OF INDIVIDUAL TRIHALOMETHANES TO TOTAL TRIHALOMETHANES
IN THE CLEAR WELL (%), LOUISVILLE WATER COMPANY
(INSTANTANEOUS MEAN VALUES)
Treatment
Routine Modified Modified
(clear well (ammoniation of
ammoniation) softening basins)
Chloroform
Bromodichloromethane
Dibromochloromethane
Bromof orm
Dichloroiodome thane
inst TTHMa
53%
30%
17%
<1%
<1%
129 ug/L
57%
28%
14%
<1%
<1%
149 ug/L
69%
24%
6%
<1%
<1%
94 ug/L
aGC/Hall detector
precision of field data was highly variable. An evaluation of the effect of
ammoniation on these compounds could not be made. These compounds will be
discussed as a part of the year-long survey for Priority Pollutants in
Section 7.
BactPviological Evaluation—
A comparison of the bacteriological conditions during the three periods
of study was made. During each period, the application of chlorine to gravity
settled raw water effected a complete reduction in both total coliform and
standard plate count densities. Densities remained low in all subsequent in-
plant samples. With clear well chlorination, with clear well ammoniation and
chlorination, and with ammoniation of softening basins and clear well chlori-
nation, the bacteriological quality of the finished water was satisfactory.
Findings—
1. Trihalomethanes were formed during treatment after chlorine was
applied.
2. When ammonia was applied to in-plant waters sufficient to convert
free chlorine to combined chlorine, little or no further trihalomethane forma-
tion resulted.
3. Precursor levels were reduced by 22-hour gravity settling. Turbidity
levels were not reduced by gravity settling but were reduced by coagulation
and settling.
4. The bacteriological quality of the finished water was satisfactory
when ammoniation followed three hours of free chlorine disinfection.
THE EFFECT OF CHLORINE DIOXIDE ON TRIHALOMETHANE FORMATION
General
An examination of the THM reaction
C12 + precursor + Br~ + I~ •> THMs
63
-------
indicates that if the chlorination practice were discontinued, the reaction
would not proceed. This would be an acceptable means of trihalomethane con-
trol only if an equally effective disinfectant were substituted. USEPA has
demonstrated on the pilot scale and bench scale that chlorine dioxide (C102)
reacts with precursor to form little or no tr-ihalomethanes and reacts to lower
precursor concentration. Chlorine dioxide was studied as a THM control at
the Western Pennsylvania Water Company.
Western Pennsylvania Water Company
Routine and Modified Treatment—
At the company's Hays Mine plant, routine treatment included chlorina-
tion of Monongahela River water. For THM control, chlorine dioxide was sub-
stituted for chlorine as the raw water disinfectant. The treatment schematic
for this utility is presented in Figure 23. Raw water flow was split inside
the plant and each stream was treated separately. For this study, only one
side of the plant was sampled and modified. Two and one-half year old
Filtrasorb 400 granular activated carbon (GAC) served as a filter/adsorber
in the plant.
The utility's raw, in-plant and finished waters were sampled two to four
times weekly during routine and modified treatment. For each sample day, the
sample collection schedule followed the time of travel of a theoretical plug
of raw water through the plant to the clear well.
During any full scale study, significant changes in raw water quality
could necessitate treatment modification and/or affect the quality of in-plant
waters. Such changes affected THM control studies at this utility when
unusually high precursor and ammonia concentrations occurred. The following
discussions address four THM study periods. While they represent routine
(raw water chlorination) and modified (raw water chlorine dioxide disinfec-
tion) treatment, they are probably not representative of typical THM forma-
tion and precursor control at the utility.
Raw Water Chlorination—
Evaluation of Trihalomethane Control—Chlorine was applied to raw water
at 2.6 mg/L for two weeks in July 1978. Water quality data and instantaneous
TTHM concentrations are presented in Figure 25. Raw water ammonia concentra-
tions were low (0.1 mg/L mean) during this period. Trihalomethane formation
resulted from the application of chlorine to the raw water and further for-
mation resulted from chlorine application to the clear well.
Precursor levels were found to be unusually high during this July 1978
period. The utility's raw water .was sampled for determination of terminal
TTHM once a month between July 1977 and May 1978. It was also sampled fre-
quently in September and October 1978. Raw water terminal TTHM concentrations
ranging from 200 ug/L to 250 ug/L were typical. During this July 1978 period,
however, raw water terminal TTHM concentrations exceed 1,200 ug/L. These were
the highest levels detected during the project, but the reason for these
unusually high precursor levels is not known. These data are presented in
Table 13. The ratio of terminal level chloroform relative to terminal level
brominated THMs was unusually high. The concentrations of terminal level
64
-------
Ul
OJ
Qi
-i Tv-y
T/
<
2
(CHLORINE DIOXIDE)
i i 1
12 MQD
1 COAG
| MIX CLARIFY ^^-
1
I~ (CHLORINE) LIME
t — f
2.2. MGC
YMER)
O / Figure 23
5/
ALUM
> KMnO^
ni n c \ F
SETTLE =Tjp GAC =O=T= wi
riLl £R T '
i 1 |
CHLORINE
FLUORIDE
LEGEND
(2)= SAMPLE POINT
(OPTIONAL FEED)
EAR
ELL =
Treatment at Western Pennsylvania Water Company 129,000 cu m/day (34 MGD) .
K/tAk-p-l IP V^/A-nrp
(FROM
WAT
ELE\
STOP
FINISHED VACUUM
=.R AT ^^
/ATED 1 M
rtu
5 y— -
PRESSURE n^r:
REGULATOR r^^
1 PRESSURE
I ^ GAGE II
X-) /N 1 ' JTT. ' l^
Pv -C) ^> — *• — •-
• ^ ^— ^
<^>^ t
ETERING VALVE
(TYPICAL)
, "2.
J=- HCl
- — AIR
j 1 6?) ^ DEHVERN
1 i i i <_y —
t U (TO MIX T
/
ANKS)
VALVE f!^/e VENTURI SiGHT HOAMPLF
(TYPICAL) METER TUBE J D^^TC
=o
Figure 24. ORENCO* chlorine dioxide generator, Western Pennsylvania Water Company.
*Rio Linda Chemical Co., Rio Linda, CA
-------
TIME, HR
TEMP, °C
TU
pH
FR
TOT
NH3| PPM
TC/
SPC
2
5°
h <
tjj<
o
. , 34.
X > 31e 7-
/ X
'x 22 X .x
/ -o X
X
X
X
X
LED F.LTERED f F'NIiHED
!-( PPM Clg,
g ° 0-5 3-75 I2.S 13.5 14-75
C
2-2
JTU 51 38 5.-7 S.5 o. 0-2
7-2 7.1 7-3 -M
2, PPM - 0.4 <0.l
-------
brominated THMs were similar to those observed at other times at the utility,
indicating that high terminal TTHM concentrations were attributable to unusu-
ally high raw water precursor levels and not to unusually high river bromide
concentrations.
TABLE 13.
WESTERN
Water
Raw TTHM
Finished TTHM -
CHC13
CHBrCla
CHBr2Cl
CHBr3
TERMINAL TTHM CONCENTRATION
PENNSYLVANIA WATER COMPANY
Concentration,51 ug/L (Mean Values)
July 5-7, 1978 July 10-14, 1978
>450 >1200
171 --
150
".2 >106° "
. 0.4
>1030
19
5.7
1.6
aGC/Hall detector
Bacteriological Evaluation—Bacteriological data, presented in Figure 25,
indicate that a significant reduction in total coliform and standard plate
count densities resulted from raw water chlorination. However, a slight
increase in both TC and SPG densities occurred through the GAG filter/
adsorber. Chlorine application at the clear well further reduced bacterial
densities. Total coliform and standard plate count densities in the finished
water complied with the 1975 USEPA Interim'Drinking Water Standards.
Raw Water Application of Chlorine Dioxide—
Chlorine Dioxide Generation—Chlorine dioxide (C102) was evaluated as a
modification to treatment in September 1978. Problems with the control of
C102 generation in July 1978 prevented evaluation at that time. Alterations
to the generator by the manufacturer resulted in the configuration shown in
Figure 24.
Chlorine dioxide was generated by reacting sodium chlorite with hydro-
chloric acid thereby allowing the utility to take raw water chlorinators off
line. An analytical procedure was employed to measure C102, chlorite, free
chlorine and total chlorine in generator effluent samples and in in-plant
waters.12 xhe generator was found to produce chlorine dioxide and little or
no free chlorine. The generator's yield of C102 (mg/L C102 produced per mg/L
chlorite consumed) was approximately 80%. The yield of free chlorine was 5%
or less. The generator may have produced no free chlorine. Dilution factors
and the sensitivity of the analytical procedure below 0.1 mg/L did not allow
accurate free chlorine determination. Unreacted chlorite was not found in the
generator's effluent. The application rate of C102 to raw water was 1.5 mg/L
and the accompanying free chlorine application rate was less than 0.1 mg/L.
The C102 application rate did not exceed 1.5 mg/L for economic reasons. USEPA
has proposed a 1.0 mg/L limit.
Evaluation of Trihalomethane Control—Water quality data and TTHM concen-
trations representing this treatment period are given in Figure 26. As a
result of treating raw water with 1.5 mg/L chlorine dioxide and less than 0.1
mg/L free chlorine, low instantaneous TTHM concentrations were found in
settled water. The increase in TTHM through the filter/adsorber was likely a
result of desorption of TTHM from the three-year-old GAG. Post-chlorination
67
-------
further increased TTHM concentration in the clear well. Thus, generated in
the manner described, chlorine dioxide formed little trihalomethane; TTHM
found in the finished water was attributable to clear well chlorination and to
desorption from GAG.
Raw water ammonia concentrations were unusually high (1.2 mg/L mean) and
variable (0.5 mg/L to 1.9 mg/L) during this period. Chlorine dioxide does not
react with ammonia.17 With chlorine dioxide generated as described, little or
no free chlorine was applied to the raw water. Therefore, it is assumed that
these ammonia concentrations had no effect on instantaneous TTHM formation.
High ammonia concentrations did interfere, however, in maintaining a free
chlorine residual in samples for the determination of terminal level TTHM con-
centrations. As a result, the terminal TTHM concentrations presented in
Figure 26 represent only 50%-75% of the samples collected for the determina-
tion of this parameter. These data suggest little, if any, precursor removal
by treatment because mean concentrations of 206 ug/L and 181 ug/L could not
be differentiated. The effect of C102, and settling, and of permanganate on
precursor levels could not be separated.
Chloro-species Evaluation—Data presented in Figure 26 indicate that 1.5
mg/L C102 applied to raw water was consumed in several hours. One end product
was chlorite; its concentration decreased through the plant (0.9 mg/L in clar-
ified water to less than 0.1 mg/L in finished water), with most of the
decrease occurring across the GAG filter/adsorber (0.6 mg/L to 0.1 mg/L). No
attempt was made to measure other chlorine dioxide end products.
Bacteriological Evaluation—Bacteriological data presented in Figure 26
indicate that 1.5 mg/L C102 application was not as effective a raw water dis-
infectant as 2.6 mg/L chlorine. During raw water chlorination, mean total
coliform and standard plate count densities in the GAG filter/adsorber influ-
ent were 1/100 mL and 50/mL, respectively (Figure 25). During C102 applica-
tion to raw water, however, mean bacterial densities in the GAG influent were
43/100 mL for total coliforms and 7,100/mL for standard plate count organisms.
With chlorine disinfection at the clear well during this period of study,
finished water bacterial densities were satisfactory.
Raw Water Application of Chlorine and Chlorine Dioxide with High Background
Ammonia Levels—
Because 1.5 mg/L ClOz was not an acceptable control for filter/adsorber
bacterial densities, a treatment modification was evaluated in which the C102
feed was reduced to 1.0 mg/L and raw water chlorinators were brought on-line
at 1.2 mg/L. Data for this period are presented in Figure 27.
Raw water ammonia concentrations during this period remained unusually
high (0.6 mg/L mean). Ammonia concentrations measured in-plant fluctuated
widely (up to 4.0 mg/L).
Evaluation of Trihalomethane Control—TTHM formation was dependent on
the concentration of ammonia present. Chlorine applied at 1.2 mg/L was rapid-
ly converted to the combined chlorine species—which drive the THM reaction at
a very slow rate.5 Therefore, low instantaneous TTHM concentrations were
found in settled water. The TTHM increase through the GAG filter/adsorber was
68
-------
- -9/v "?C\~1
i | XCyvo r" ~. — I ^-^ '
"2.
O
.f A/ —
^5\
U oo
i^
0
RAW RAW |
i — 208
• * *-* , . 1 / S\
IfoZ \la^J
" ,- II r-
„" I'2
COAG AND SETTLED _. ,(lAr<~,F n f
CLARIFIED FILTERED
101
"^Tao
^
/^ >*
FINISHED
1.5 PPM CIO,, 1.4 PPM C12
-------
z
o
O (T5
O
218
185
2.2
17
12
7-1
RAW |
U2 PPM
PARAMETER
TIME , HR O
TEMP, °C
TURB,NTU
pH
FREE CI2,PPM
TOTAL CI2, PPM
CIO2, PPINA
CIO^", PPM
NH3 , PPM
TC/)OO^U
SPC/mU
K10T E S
p. RANGE: 5.2 TO SO
b RANG E = ND TO 0.4-5
c RANGE = 1.6 TO
-------
likely attributable to desorption. With post-chlorination, further formation
of TTHM varied inversely with the concentration of ammonia in the clear well.
When clear well ammonia was less than 0.1 mg/L, free chlorine was 0.45 mg/L
and finished water TTHM reached 50 ug/L. When clear well ammonia was 1.6
mg/L, no free chlorine was detected and finished water TTHM reached only 5.2
ug/L—a level that could not be differentiated from the filter/adsorber
effluent TTHM concentration. Thus with high levels of background ammonia
present, THM formation was essentially halted. Because of the presence of
ammonia, the combined effects of C102 and chlorine on TTHM formation could not
be evaluated.
High ammonia concentrations interfered with free chlorine added to sam-
ples for the determination of terminal level TTHM concentrations. Therefore,
terminal TTHM concentrations presented in Figure 27 represent 0% to 75% of the
samples collected for the determination of this parameter. Comparisons of
mean terminal TTHM concentrations indicated reduction of precursor level
between the raw water and filtered water sample points. The effect of CIOz
on precursor could not be separated from the effect of coagulation and
settling.
Chloro-species Evaluation—Demand for C102 consumed the 1.0 mg/L applied
to raw water and chlorite was found as an end product. GAG filtration/
adsorption accounted for most of the removal of chlorite during treatment
(0.7 mg/L after clarifiaction, 0.5 mg/L after settling, and less than 0.1 mg/L
after filtration/adsorption).
Bacteriological Evaluation—Bacteriological data presented in Figure 27
indicate that pre-disinfection with chlorine and C102 was satisfactory for
control of bacterial densities in the GAG influent. Chlorine applied to the
raw water was rapidly converted to combined chlorine forms because of high
ammonia levels in the raw water during this time period. A complete reduc-
tion in bacterial densities did not occur immediately upon chlorination.
However, densities in the GAG influent were satisfactory with <1/100 mL for
total colifonn bacteria and 33/mL for standard plate count bacteria. Again,
bacterial densities increased through the GAG filter/adsorber. GAG effluent
densities were 2/100 mL and 440/mL for the total coliform and standard plate
count bacteria, respectively. With application of chlorine at the clear well,
finished water bacterial densities were satisfactory.
Raw Water Chlorination with High Background Ammonia Levels—
Chlorination of raw water was again evaluated in October 1978, when raw
water ammonia concentrations were unusually high (1.5 mg/L mean). TTHM con-
centrations and water quality data for this period are presented in Figure 28.
Evaluation of Trihalomethane Control—The applied chlorine (2.2 mg/L)
was rapidly converted to the combined chlorine species. With little,or no
free chlorine present, only low concentrations of instantaneous TTHM were de-
tected in settled water. The slight increase in TTHM through the GAG filter/
adsorber was probably attributable to desorption. Further formation of TTHM
in the clear well resulted from post-chlorination only if ammonia concentra-
tions were low. With 0.1 mg/L ammonia in the clear well, the free chlorine
concentration was 0.6 mg/L resulting in 43 ug/L TTHM. With 1.5 mg/L ammonia
71
-------
in the clear well, no free chlorine was detected and only 7.1 ug/L TTHM
resulted in the finished water—a level that could not be differentiated from
the filter/adsorber effluent. Thus, with sufficient levels of background
ammonia present to convert free chlorine to combined chlorine, only low con-
centrations of instantaneous TTHM resulted.
Comparisons of mean terminal TTHM data indicated reduction in precursor
levels by coagulation, clarification and settling. These data are based on
67% of the samples collected for determination of this parameter. High ammo-
nia concentrations interfered with free chlorine added to samples for the
determination of terminal TTHM.
Bacterial Evaluation—Bacteriological data presented in Figure 28 indi-
cate that predisinfection with 2.2 mg/L chlorine was satisfactory during this
period when raw water ammonia levels were in excess of 1 mg/L. An increase
in standard plate count densities again occurred through the GAG filter/
adsorber. However, with chlorine application at the clear well, the total
coliform and standard plate count densities were satisfactory in the finished
water.
Ratio of THM Compounds—
Data presented in Table 14 indicate differences in the ratio of indivi-
dual THMs found in finished water during the four study periods. Relatively
higher concentrations of CHC13 were found when free chlorine residuals were
carried through the entire treatment process (raw water chlorination in July).
Relatively higher concentrations of brominated THMs were found when free
chlorine residuals were observed only in the clear well (treatment with C102
and/or sufficient ammonia to convert pre-chlorine disinfectant to combined
species in September and October). Other than the difference in reaction time
with free chlorine, possible causative factors include the variable nature and
concentration of the precursor from July to October, the effect of unknown raw
water bromide concentrations, and the uncertain role of bromine in forming
THMs.
TABLE 14. RATIO OF INDIVIDUAL TRIHALOMETHANES TO TOTAL TRIHALOMETHANES
IN THE CLEAR WELL (%), WESTERN PENNSYLVANIA WATER COMPANY
(INSTANTANEOUS MEAN VALUES)
Pre-Treatment
Compound
CHC13
CHBrCl2
CHBraCl
CHBr3
CHIC12
inst TTHM3
Routine
(raw water
chlorination,
no background
ammonia)
(July 1978)
71%
20%
8%
<1%
<1%
42 ug/L
Modified
(C102 to
raw water)
(Sep 1978)
26%
28%
36%
10%
<1%
20 ug/L
Modified
(CIO 2 and
chlorine to
raw water,
background
ammonia)
(Sep 1978)
23%
33%
36%
8%
<1%
17 ug/L
Routine
(raw water
chlorination,
background
ammonia)
(Oct 1978)
20%
32%
39%
9%
<1%
22 ug/L
aGC/Hall detector
72
-------
Z 2<
g
UJ Y~ 1^—T-
—•7 i ~
SlU\
"7 i
o
u
RAW
2-2
PARAMETER
TIME, HR O
TEMP, °C
TURB, NTU IO
pH -7.!
FREE Cl^.PPKA
TOTAL Cl PPM
NH3 , PPM US
TC/lOOmL 25OOO
e.pc /.VN i
^ < ^- / pr* L-
ex. RANGE = 7.1 TO 43
b RAMG E- Kl D TO O-
r RANGE - US TO 4-3
^
1 c/-) 1 4-z
1 1 i>SX 1 ^ O
1. JZI
"i r ~r~~~\i>'?0^
s
y.
f { S
y* to-*o SS
^ %
t RAW I C°LAl.^E^ SETTLED F1L^CRBD t F.N.5HED
PPKA C(2 Q-4PPM KMnO^.
0.5 3-75 12. 5
_ -
8-& 4.1 2.1
7.2 8.1 -7-4
< O- 1 M D
-------
Evaluation of Other Priority Pollutants—
These studies were conducted from July through October 1978 following the
year-long period of monthly sampling. Annual data indicated infrequent and
low level occurrence of other halocarbons; therefore, analyses of these com-
pounds were not performed during these studies.
Findings—
1. Trihalomethanes were formed during treatment after chlorine was
applied.
2. Little or no trihalomethanes were formed when only chlorine dioxide
was applied to raw water.
3. With background ammonia concentrations sufficient to convert free
chlorine to combined chlorine, little or no trihalomethane formation resulted.
4. When applied to raw water with sufficient demand, chlorine dioxide
was consumed. An end product measured was chlorite. In three hours on a mg/L
basis, 60%-70% of the applied C102 went to chlorite.
5. Settling and GAC filtration/adsorption decreased chlorite concentra-
tions to less than 0.1 mg/L in the finished water.
6. When applied to raw water, 1.5 mg/L C102 was not as effective a dis-
infectant as 2.6 mg/L chlorine.
7. When applied to raw water, the combination of 1.0 mg/L C102 and 1.2
mg/L chlorine was as effective a disinfectant as 2.6 mg/L chlorine.
8. With temperatures above 22°C, total coliform and standard plate count
densities increased through GAC filtration/adsorption.
9. The bacterial quality of the finished water was satisfactory with
chlorine post-disinfection.
10. Chlorine dioxide generation by chlorite and hydrochloric acid had an
80% yield (mg/L C102 produced per mg/L ClOl consumed). The yield of free
chlorine was less than 5%.
11. Ammonia and precursor conditions on the Monongahela River varied
considerably. The effects of routine and modified treatment on precursor
levels could not be evaluated.
12. Two-and-one-half year old GACs receiving chlorinated and settled
water in the filtration/adsorption mode in beds designed for sand filtration
were exhausted for the removal of CHC13, CHBrCl2, CHBr2Cl, CHBr3 and instan-
taneous TTHM. With a significant decrease in influent instantaneous TTHM con-
centrations, instantaneous TTHM was likely desorbed from the GAC.
74
-------
THE EFFECT OF GRANULAR ACTIVATED CARBON ADSORPTION/FILTRATION
ON TRIHALOMETHANE CONTROL
General
An adsorber can control trihalomethanes in two ways. An examination of
the THM reaction
C12 + precursor + Br~ + I~ -> THMs
indicates that a reduction in THM formation would result if precursor levels
were reduced or if THMs were formed and subsequently removed. Granular acti-
vated carbon (GAG) has been shown to adsorb both precursor and trihalomethanes
in pilot scale operation.6 This means of control was examined full scale at
two project utilities: the Huntington Water Corporation and the Beaver Falls
Authority. These two studies investigated the adsorptive capacity of virgin
GAG in the filtration/adsorption mode over time.
At each utility, raw, finished, GAG influent and GAG effluent waters were
sampled one or more times weekly to define exhaustion of GAG for the removal
of THMFP and instantaneous TTHM and to evaluate GAG filtration/adsorption for
a period of time following exhaustion. For each sample day, waters were sam-
pled following a theoretical plug from raw water through the plant to the
clear well.
GAG Evaluation
GAG evaluation for this project was based on exhaustion. Exhaustion was
determined by a point in time when effluent concentrations of a compound or
group of compounds equaled or first exceeded influent concentrations.
Appendix C indicates that variability of a reported instantaneous TTHM concen-
tration can approach ± 20%. This variability was considered in determining
when influent and effluent concentrations were likely equal. In a hypotheti-
cal case, apparent exhaustion of a GAG for the removal of TTHM was defined at
10 weeks when the effluent concentration of 20 ug/L exceeded the influent con-
centration of 17 ug/L. If, however, at nine weeks, the influent concentration
was 31 ug/L and the effluent level was 26 ug/L, exhaustion may have occurred.
Given ± 20% variability of the data, these concentrations could have been 25
ug/L and 31 ug/L, respectively, indicating earlier exhaustion. Thus, trend
should also be considered when defining exhaustion. The data following the
point of apparent exhaustion should indicate influent and effluent concentra-
tions within 20% of each other or should indicate effluent concentrations
generally exceeding influent concentrations. The exhaustion of GAG, as dis-
cussed in this report, is consistent with such trends.
Breakthrough was determined by a point in time when a compound was first
detected in the GAG effluent.
Huntington Water Corporation
Background— .
At Huntington a virgin GAG bed was evaluated for adsorption of influent
75
-------
instantaneous trihalomethanes and influent unreacted precursor (THMFP). West-
vaco's WVW 14x40 GAG was evaluated. The selection of GAG was based on its
history of effective taste and odor control at the utility. The virgin GAG
replaced taste and odor exhausted GAG. It was operated in the filtration/
adsorption mode in a bed originally designed for sand filtration. No previous
pilot scale studies had been conducted to determine optimum selection of GAG
or bed depth for organics control.
The bed was placed with 76 cm (30 inches) of GAG on top of 30 cm (12
inches) of sand and gravel. After placement, the bed was backwashed several
times to remove fine particulates. When the bed was placed in operation, it
received chlorinated, coagulated and settled water. Treatment is illustrated
in Figure 29. Backwashing frequency was based on head loss and effluent tur-
bidity levels. The bed was backwashed 16 times the first week and 14 times
the second week and an average of eight times per week thereafter. Hydraulic
data provided by the utility demonstrated a mean loading rate of 6.1 m/hr
(2.6 gpm/ft^) and a mean empty bed contact time (EBCT) of 7.2 minutes. Water
quality data for the utility are given in Table 15.
The virgin GAG bed represented only 8% of the plant capacity. Periodi-
cally, influent and effluent waters for older WVW 14x40 GAG beds were sampled
to evaluate performance after long periods of time in operation.
Trihalomethane Adsorption by Virgin GAG—
Figure 30 illustrates removal of TTHM by virgin GAG after varying lengths
of time in operation. Breakthrough of THMFP and instantaneous TTHM was obser-
ved during the first week as both were detected in the bed's effluent. By
the fourth week of operation, the percent removal of THMFP and instantaneous
TTHM by the GAG bed was decreasing with time. After 22 weeks of operation,
influent and effluent concentrations could not be differentiated, indicating
that exhaustion had occurred on or before that time.
Figure 31 is a plot of the removal of instantaneous TTHM by GAG adsorp-
tion for the first 45 weeks of operation of the virgin bed showing that the
GAG was exhausted for the removal of instantaneous TTHM at seven to eight
weeks of operation. (Prior to that time, influent concentrations exceeded
effluent concentrations by at least 20%. Following that time, effluent con-
centrations exceeded influent concentrations, or influent and effluent concen-
trations were within 20% of one another, and thus could not be differentia-
ted.) The GAG was exhausted for the removal of THMFP at seven to ten weeks of
operation as illustrated by Figure 31.
The adsorption of individual instantaneous THMs by virgin GAG is plotted
in Figure 32. These data indicate that the GAG was exhausted for the removal
of chloroform at seven to eight weeks of operation. Exhaustion for the
removal of bromodichloromethane and dibromochloromethane occurred at 11 to 14
weeks of operation.
Data presented in Table 16 indicate that the virgin GAG was not exhausted
for bromoform removal at 12 weeks of operation. Beyond that time, influent
and effluent concentrations were low and could not be differentiated. Appen-
dix C, Figure C-9, indicates that the precision of field data for instantan-
76
-------
UJ
o
o
33%
T*O HR5
T= I HR
<*-l°/o LIME CHLORINE
FeS04
(PAC)
(POLYMER) (KM*O4)
(Cu.504) (POLYMER)
LEGEND
SAMPLE POINT
(OPTIONAL FEED)
HYDRO
TREATER
OLD
WVW I4*4O
GAC
FILTERS
30%
SETTLING
T=3.25 MRS
VIRGIKI
WVW 14x40
GAC
FILTER
OLD
WVW 14*4O
GAC
FILTERS
4,2 °/o
O
T=3-5 MRS
CLEAR
WELL
T» 5-25 MRS
CHLORINE
Figure 29. Treatment at Huntington Water Corporation,
64,000 cu m/day (17 MGD), July 1977-March 1978.
-------
00
TABLE 15.
Week of
Virgin GAC
Operation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
20
22
23
25
27
31
WATER QUALITY DATA (MEAN VALUES) HUNTINGTON WATER CORPORATION JULY 1977-MARCH 1978
Raw Water
Mean pH = 7.
Temp, °C
27
28
28
28
28
27
27
27
27
27
26
24
19
19
14
15
15
11
8
5
3
2
2
Turbb
14
21
26
13
15
80
37
34
17
18
25
24
47
98
34
22
18
42
240
160
24
30
34
5
TCC
1,600
1,200
910
870
1,500
3,000
5,300
2,300
1,400
970
1,100
1,700
3,100
4,300
3,900
2,600
2,800
3,900
1,400
26,000
2,800
5,900
610
GAC Influent (Settled)
Mean pH = 8.9
Turbb
2.0
4.6
4.9
4.4
6.5
5.8
3.8
5.9
3.3
4.6
7.9
4.4
8.7
4.3
16
9.1
10
5.5
9.8
8.0
7.0
9.0
14
Chlorine3
Free
0.8
0.3
1.8
0.5
0.6
0.5
0.4
0.3
0.5
0.5
0.4
0.5
0.5
0.5
0.5
0.5
0.4
0.6
0.2
0.5
ND
0.9
0.3
Total TC
1.4 <1
0.4 <1
3.7 <1
0.7 <1
0.9 <1
0.8 <1
0.7 <1
0.6 <1
0.7 <1
0.7 <1
0.6 <1
0.7 <1
0.8 <1
0.9 <1
0.7 <1
0.9 <1
0.9 <1
0.8 <1
0.3 <1
0.6 <1
0.7 <1
1.1 <1
0.9 <1
SPC
4
52
42
7
18
28
17
22
24
26
28
28
31
—
34
39
18
200
55
36
30
—
—
Virgin GAC Effluent
Mean pH = 8.7
Turbb
0.2
1.5
1.4
1.7
1.8
1.1
0.5
1.5
0.3
3.2
1.7
0.4
0.4
0.4
0.4
1.0
0.5
0.7
12
0.8
0.5
0.2
0.3
Chlorines
Free
ND
ND
0.1
TR
0.4
0.1
TR
0.2
TR
TR
TR
TR
TR
0.1
—
TR
TR
0.3
0.2
0.2
ND
0.3
0.4
Total TC
ND <1
TR <1
0.3 6
0.3 8
0.6 5
0.2 <1
0.1 <1
0.4 2
0.2 2
0.2 <1
0.2 <1
TR 1
0.2 <1
0.4 <1
<1
0.5 <1
0.6 <1
0.4 <1
0.4 <1
0.3 <1
0.5 <1
0.5 <1
0.8 <1
SPCd
100
53
12
41
18
13
3
25
46
140
23
12
30
2
10
2
4
11
3
<1
—
.Chlorine, mg/L
Turbidity, NTU
^Total coliform/100 mL
Standard plate count/mL
-------
30O -
< -
^
- 20O-
2
O
H
M
2
ui 100 -
o
2
O
O
X^
'/
V
^
^
•
•
•
f
.
-
'/
X
X
X
X
-
•
.
s
X
X
-
-
^
^
RAW PRE" P°ST- RAW PRE- POST- PAW PRE~
GAG GAG GAG GAG GAG
•4-
1
/
_ TERM
TTHM
THMFP
MST TTMM
>:
GAG
WEEK I
WEEK 22
Figure 30. Trihalomethane formation, Huntington Water Corporation.
-------
30O
20 O
IOO-
2
0
h
4
&.
t-
Z
UJ
o
2
o
u
INFLUENT
THMFP
-*-EFFLUENT THMFP
IOO
5O -
10 15 20 25 30 35 4O
HUMTIN6TOM WATER CORP.
!4x4-O
DEPTH * 76 CM (30 INCHES) GAG
LOADING RATE -6.1 M/HR(2.
EBCT s -7.1 MIUUTES
INJFLUEKJT TTHM
EFFLUEKJT TTHM
15 20 25 30 35 40
TIME IN OPERATIOW, WEEKS
28 28 21 14- n 3 2 4
TEMPERATURE, °C
—r~
13
14
Figure 31. Trihalomethane removal by granular activated carbon.
80
-------
ISO
too -
50 ,
J
(T)
z
o
4-0 -
h
B
z
o
o
INFLUENT
CHCI3
^EFFLUENT CHCI3
HUNTINGDON WATER CORP.
GA.C = WV W 14x40
DEPTH - -76 CM ( 3O INCHES) GAG
LOADINJG I2ATE -- 6-1 M/HR(2.
15 20 25 30 35 4O
TIME IN OPERATION, WEEKS
28
28
27
13
45
14 II 3 2 4-
TEMPERATURE. °C
Figure 32. Trihalomethane removal by granular activated carbon
14
81
-------
ecus bromoform could be ± 15% for concentrations above 1.0 ug/L, ± 40% near
0.5 ug/L, and ± 100% below 0.2 ug/L.
TABLE 16. REMOVAL OF TRIHALOMETHANES BY GRANULAR ACTIVATED CARBON3
HUNTINGTON WATER CORPORATION. JULY 1977-MAY 1978
Week of
Operation
1
2
3
4
5
6
7
8
9
10
11
12
15
16
17
19
21
22
35
39
42
45
Concentrat ion
Bromoform
Influent
1.6
4.4
1.2
0.3
0.2
0.2
0.1
0.5
0.6
1.5
1.6
1.9
<0.1
0.1
0.1
0.2
ND
ND
ND
ND
ND
0.5
Effluent
ND
<0.1
<0.1
ND
<0.1
ND
ND
<0.1
0.1
0.2
0.2
0.2
0.1
<0.1
0.1
<0.1
ND
<0.1
ND
<0.1
ND
0.2
,b ug/L
Dichloroiodomethane
Influent
<0.1
0.1
0.3
0.2
0.2
0.4
0.7
0.6
0.4
0.2
0.2
0.1
<0.1
<0.1
0.1
<0.1
ND
ND
ND
0.1
<0.1
0.2
Effluent
ND
ND
ND
<0.1
<0.1
<0.1
<0.1
<0.1
ND
<0.1
<0.1
0.1
<0.1
<0.1
<0.1
ND
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
Bed depth = 76 cm (30 inches) GAC
Loading rate =6.1 m/hr (2.6 gpm/ft )
EBCT =7.2 minutes
bGC/Hall detector, approximate lower detection level =0.1 ug/L
ND = not detected
Data presented in Table 16 indicate that the virgin GAC was not exhausted
for the removal of dichloroiodomethane at 11 weeks of operation. Beyond that
time, influent and effluent concentrations were low and could not be differ-
entiated. Appendix C, Table C-6 indicates that the precision of field data
for instantaneous dichloroiodomethane could be ± 40% below 0.2 ug/L and ± 100%
below 0.1 ug/L.
The data in Table 16 do not show that the GAC was exhausted for the re-
moval of bromoform or dichloroiodomethane after the llth or 12th week of oper-
ation because, after that time, influent and effluent concentrations were too
low for interpretation. During later operation, when temperatures and influ-
ent concentrations increased, further adsorption may have occurred. Figure 31
indicates that influent TTHM concentrations generally varied with temperature.
82
-------
Trihalomethane Adsorption by Older GAG—
Periodically older WVW 14x40 GAG bed effluent waters were sampled. These
beds were of identical geometry and similar hydraulics and received the same
water as the virgin GAG bed.
One bed was sampled during its 9th, llth, 13th and 14th months of opera-
tion. It was found to be exhausted for the removal of THMFP, chloroform,
bromodichloromethane and dibromochloromethane and instantaneous TTHM, however,
it was not exhausted for the removal of bromoform after 11 months of opera-
tion. At 13-14 months of operation, with lower temperatures, influent bromo-
form concentrations were low and could not be differentiated from effluent
concentrations. These data are presented in Table 17.
Another bed was sampled during its 27th, 28th, 29th, 31st and 32nd month
of operation. It was found to be exhausted for the removal of THMPF, chloro-
form, bromodichloromethane, dibromochloromethane, instantaneous TTHM and,
possibly, bromoform. The precision of field data for bromoform (Appendix C,
Figure C-9) indicates that the influent and effluent bromoform concentrations
cannot be differentiated. These bromoform data are presented in Table 17.
Adsorption of Priority Pollutants and Other Compounds by Virgin GAG—
Analyses were performed for compounds other than trihalomethanes in GAG
influent and effluent waters to determine their adsorption by virgin GAG when
present. Purgeable halocarbons and base-neutral extractable halocarbons were
detected infrequently, and, when detected, their concentrations were low and
in ranges where precision of the field data indicates that influent and efflu-
ent concentrations could not be differentiated, i.e., the compounds were typi-
cally detected at or below 0.2 ug/L. Until more sensitive analytical proce-
dures are employed, the adsorptive capacity of GAG for these compounds at low
concentrations cannot be evaluated. There were exceptions, however.
Carbon Tetrachloride—Carbon tetrachloride occurred frequently in
Huntington's raw and GAG influent waters. Table 18 presents influent and
effluent data for the virgin GAG. Appendix C, Figure C-8, indicates that the
precision of carbon tetrachloride data below 0.2 ug/L may be ± 100%; there-
fore, influent and effluent concentrations below 0.2 ug/L were too low to be
differentiated. The data in Table 18 indicate adsorption occurred during
weeks 5, 10 and 12, for example, but influent and effluent concentrations
could not be differentiated during weeks 14 or 42. These data indicate that
virgin GAG was an effective barrier when higher influent concentrations
occurred (week 10); that in the first two months of operation, it adsorbed the
influent load (breakthrough was not observed until week 9); but that after
several months of operation, it was not acting as a barrier to the routine
influent loading. During weeks 16 and 42 the compound was detected in the
effluent at concentrations that could not be differentiated from influent con-
centrations. This does not imply that exhaustion had occurred after several
months of operation or that the GAG would not act as an effective barrier to
a higher influent load at a later time.
Chlorobenzene—Chlorobenzene was detected infrequently in GAG influent
waters^However, when detected, data indicate that Chlorobenzene was
adsorbed. During the 6th week of operation, the influent concentration was
83
-------
TABLE 17. REMOVAL OF BROMOFORM BY GRANULAR ACTIVATED CARBON3, HUNTINGTON WATER CORPORATION
Bromoform Concentration, ug/Lb
co
Raw Water
Temp (°C)
28
28
27
26
17
9
5
Virgin GAG
Month of
Operation
1
1 1/2
2
3
4
5
6
Placed
Influent
1.9
0.2
0.3
1.4
0.1
0.1
ND
July 1977°
Effluent
0.1
0.1
0.1
0.2
0.1
0.1
0.1
GAG Placed October 1976
Month of
Operation
9
9 1/2
11
13
14
Influent
2.0
4.4
1.9
0.2
ND
Effluent
0.4
2.0
1.1
0.2
0.1
GAG PI
Month of
Operation
27
27 1/2
28
29
31
32
aced April
Influent
2.0
4.4
0.1
1.8
0.2
ND
1975
Effluent
1.0
4.6
0.3
1.4
0.4
ND
GAG = WVW 14x40
Bed depth = 76 cm (30 inches) GAC
Loading rate =6.1 m/hr (2.6 gpm/ft ) for virgin bed
EBCT =7.2 minutes for virgin bed
bGC/Hall detector, approximate lower detection level =0.1 ug/L
CData taken from Table 16.
ND = not detected
-------
TABLE 18. REMOVAL OF CARBON TETRACHLORIDE BY VIRGIN GRANULAR ACTIVATED CARBONa
HUNTINGTON WATER CORPORATION, JULY 1977-MAY 1978
Week of
Operation
1
2
3
4
5
6
7
8
9
10
11
Concentration, b ug/L
Influent
<0.ic
ND
NFB
0.4C
0.6C
O.lc
0.1
O.lc
0.3
13+
0.4
Effluent
ND
ND
NFB
ND
ND
NFB
NFB
NFB
<0.1
0.4+
0.1
Week of
Operation
12
14
15
16
18
22
35
39
42
46
Concentration,0 ug/L
Influent
0.5
0.2
0.2
0.3
0.2
<0.1
<0.1
<0.1
0.1
0.3
Effluent
0.1
<0.1
0.1
0.3
<0.1
<0.1
0.1
0.1
0.2
0.2
aGAC = WVW 14x40
Bed depth = 76 cm (30 inches) GAG
Loading rate = 6.1 m/hr (2.6 gpm/ft2)
EBCT =7.2 minutes
^GC/Hall detector, approximate lower detection level =0.1 ug/L
cCo-elution with 1,1,1-trichloroethane
ND = not detected
NFB = not found after blank correction
+ = GC/MS confirmed as carbon tetrachloride
1.0 ug/L and the compound was not detected in the effluent. During the 10th
week, the influent concentration was 0.8 ug/L (GC/MS confirmed) and the
effluent concentration was 0.4 ug/L (GC/MS confirmed). During the 35th week,
the influent concentration was 0.5 ug/L and the compound was not detected in
the effluent. The precision of field data for chlorobenzene (Appendix C,
Table C-7) indicates these influent and effluent concentrations are different.
1,4-Dichlorobenzene—1,4-dichlorobenzene was found with some frequency in
the GAG influent. Evaluation of adsorption of this and other base-neutral
extractable compounds was complicated by the losses during extraction (Section
5, page 30). 1,4-dichlorobenzene adsorption data not corrected for extraction
losses are presented in Table 19. Concentrations in the waters sampled are,
therefore, somewhat higher than those presented. Further, precision of field
data for the compound indicates that the variability for the data presented in
Table 19 may be ± 70% (Appendix E, Table E-l); therefore, influent and efflu-
ent concentrations of 1,4-dichlorobenzene cannot be differentiated. These
data do not imply exhaustion. They indicate, however, the GC/MS confirmed
presence of 1,4-dichlorobenzene in the GAC effluent, at concentrations that
cannot be differentiated from those influent, as early as the 5th week of
operation.
Unidentified Base-Neutral Extractable Halocarbons—Adsorption data for an
unknown base-neutral extractable halocarbon are presented in Table 20. When
using the procedure described in Appendix D, the compound has the same elution
time as aldrin; however, the compound is not believed to be aldrin because re-
peated GC/MS confirmation attempts for aldrin proved negative. Further, the
85
-------
TABLE 19. REMOVAL OF 1,4-DICHLOROBENZENE BY VIRGIN GRANULAR ACTIVATED CARBON3
Week of
Operation
1
2
3
5
6
7
8
10
Concentration
Influent
ND
ND
ND
0.8
1.0
0.1
0.2
1.2+
,b,c ug/L
Effluent
ND
ND
ND
1.0+
0.7
ND
0.7
0.5+
Week of
Operation
11
12
13
14
22
31
35
Concentration
Influent
<0.1
0.6
<0.1
0.4
1.4+
<0.1
0.2
,b>c ug/L
Effluent
<0.1
ND
ND
0.1
ND
ND
ND
Bed depth = 76 cm (30 inches) GAG
Loading rate =6.1 m/hr (2.6 gpra/ft )
EBCT = 7.2 minutes
^Base-neutral extraction, GC/Hall detector,
approximate lower detection level =0.1 ug/L
CNOT CORRECTED FOR EXTRACTION LOSSES.
ND = not detected
+ = GC/MS confirmed as 1,4-dichlorobenzene
compound could not be GC/MS identified (Section 7, page 163). The extraction
recovery of the compound is not known because its identity is not known. The
precision of the data presented in Table 20 may be ± 20% above 0.1 ug/L
(Appendix E, Table E-13). These data do indicate adsorption during the first
two months of operation (breakthrough was not observed until week 10) and
suggest adsorption beyond that time.
TABLE 20. REMOVAL OF AN UNIDENTIFIED BASE-NEUTRAL EXTRACTABLE HALOCARBONa
BY VIRGIN GRANULAR ACTIVATED CARBON^, HUNTINGTON WATER CORPORATION
JULY 1977-MARCH 1978
Week of Concentration,0 ug/L
Operation Influent Effluent
1 0.4 ND
2 0.2 ND
3 0.2 ND
5 ND ND
6 0.1 ND
7 0.2" ND
8 <0.1 ND
9 0.2 ND
10 0.4 ND
Week of
Operation
11
12
13
14
15
22
31
35
aUsing procedure described in Appendix D,
compound has same elution time as aldrin.
bGAC = WVW 14x40
Bed depth = 76 cm (30 inches) GAG
Loading rate = 6.1 m/hr (2.6 gpm/ft2)
EBCT = 7.2 minutes
CNOT CORRECTED FOR EXTRACTION LOSSES.
Concentration,0 ug/L
Influent Effluent
0.1 0.1
3.5~ <0.1
0.7 0.1
0.2 0.3
0.2 <0.1
ND ND
ND ND
ND ND
ND = not detected
~ = Found not to be
aldrin by GC/MS
86
-------
At Huntington and at other utilities, base-neutral extractable halocar-
bons were occasionally detected in finished waters but were rarely found in
raw waters. As discussed in Section 7, these may be products of chlorinaticn
or may be contaminants in the chlorine supply. At Huntington, one such halo-
carbon was not detected in raw water but was detected 12 of 19 times in fin-
ished water. Another such halocarbon was not detected in raw water but was
detected 8 of 19 times in finished waters. When detected, concentrations in
GAG influent waters were lower than concentrations in finished waters (Table
21). Although the influent concentrations were low and detection was infre-
quent, the data suggest that the halocarbons were adsorbed.
TABLE 21. REMOVAL OF UNIDENTIFIED BASE-NEUTRAL EXTRACTABLE HALOCARBONS
BY VIRGIN GRANULAR ACTIVATED CARBON,a HUNTINGTON WATER CORPORATION
JULY 1977-MARCH 1978
Concentration, b ug/L
Week of
Operation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
18
20
22
26
31
35
Halocarbonc
,d
Influent Effluent
<0.1
<0.1
<0.1
<0.1
ND
ND
0.2
ND
0.1
ND
<0.1
<0.1
ND
<0.1
ND
—
—
<0.1
—
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
<0.1
ND
ND
Halocarbon" > e
Influent
0.4
ND
ND
<0.1
ND
<0.1
0.5
ND
0.6
ND
<0.1
ND
ND
ND
ND
—
—
ND
—
ND
<0.1
Effluent
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
aGAC = WVW 14x40
Bed depth = 76 cm (30 inches) GAG
Loading rate =6.1 m/hr (2.6 gpm/ft )
EBCT = 7.2 minutes
bNOT CORRECTED FOR EXTRACTION LOSSES
cUsing procedure described in Appendix D, compound has retention
time of approximately 0.75 relative to hexachlorobenzene.
dQuantification based on hexachlorobenzene.
eUsing procedure described in Appendix D, compound has retention
time of approximately 0.77 relative to hexachlorobenzene.
ND = not detected
Carbon Tetrachloride Desorption from Older GAG—
When effluents from the older GAG filter/adsorbers were sampled, concen-
87
-------
trations of carbon tetrachloride were found to be higher than influent con-
centrations. These data are presented in Table 22. These GAC beds were in
place in February 1977 when a large carbon tetrachloride spill (raw water
concentrations in excess of 100 ug/L) moved through the Huntington plant.
These data indicate desorption of carbon tetrachloride that had been earlier
adsorbed by the GAC. USEPA reported desorption of carbon tetrachloride from
GAC for a period of nine months following extremely high influent carbon tet-
rachloride loading.1°
TABLE 22. REMOVAL OF CARBON TETRACHLORIDE BY
GAC placed October 1976
Week of
Operation
9
9%
11
13
14
Concentration,13 ug/L
Influent
ND
ND
0.5
0.2
0.1
Effluent
2.2C
2.3C
1.0
0.4
0.3
GAC placed April
Week of
Operation
27
27%
28
29
31
32
1975
Concentration,13 ug/L
Influent
ND
ND
0.1
0.2
0.2
0.1
Effluent
1.5C
0.7C>+
0.6
0.7
0.3
Bed depth = 76 cm (30 inches) GAC
Loading rate = approximately 6.1 m/hr (2.6 gpm/ft2)
EBCT = approximately 7.2 minutes
GC/Hall detector, approximate lower detection level =0.1 ug/L
cCo-elution with 1,1,1-trichloroethane
+ = GC/MS confirmed as carbon tetrachloride
ND = not detected
Bacteriological Evaluation—
Microbiological characteristics of the Ohio River raw water and the GAC
influent and effluent waters are presented in Table 15. These data indicate
that during the 31-week study, the mean density of total coliforms in the
Ohio River raw water was 3,400/100 mL. After the processes of chlorination,
coagulation and settling, the density of total coliforms in the GAC influent
water was always <1/100 mL. Coliform densities were apparent in the GAC
effluent and seem to be related to source water temperatures. During weeks
three through nine, when the raw water temperatures were 26-28°C (79-82°F),
the total coliform densities in the GAC effluent ranged from <1 to 8/100 mL.
During the remainder of the study period, the water temperatures declined
from 27°C to 2°C (80°F to 35°F) and the GAC effluent coliform densities were
always <1/100 mL with the exception of a density of 1/100 mL during week 12.
A similar occurrence was observed in the general bacterial population
data. During the first ten weeks 'of the study, the data indicate that GAC
effluent standard plate count bacterial densities occasionally exceeded in-
fluent densities. After ten weeks, GAC effluent bacterial densities were con-
sistently lower than influent densities.
The higher densities of coliforms and of the general bacterial population
in the effluent water during the first ten weeks do not seem to correlate with
either raw water turbidity or raw water total coliform densities during that
88
-------
time. These parameters had lower values during weeks one through ten than
the 31-week mean value. The raw water temperatures during the first ten weeks
were in a range that may have favored regrowth of bacteria on the carbon bed.
Other growth conditions may have been favorable on the GAG with the reduction
of free chlorine on the carbon, the provision of a large surface area and the
possible accumulation of nutrients.
Finished water quality was adequately maintained during the study at a
total coliform density of <1/100 mL and a standard plate count density of
<500/mL with the application of chlorine following GAG adsorption/filtration.
Findings—
1. Trihalomethane formation occurred during treatment after chlorine
application and generally varied with water temperature.
2. During summer months, virgin WVW 14x40 GAG receiving chlorinated,
settled water and operating in the filtration/adsorption mode in a bed
designed for sand filtration was exhausted for the removal of:
a. chloroform at seven to eight weeks of operation.
b. bromodichloromethane at eleven to 14 weeks of operation.
c. dibromochloromethane at eleven to 14 weeks of operation.
d. instantaneous TTHM at seven to eight weeks of operation.
e. THMFP at seven to ten weeks of operation.
3. WVW 14x40 GAG receiving chlorinated and settled water in the filtra-
tion/adsorption mode in a bed designed for sand filtration was not exhausted
for the removal of bromoform for periods of from one to two years.
4. WVW 14x40 GAG operated in the filtration/adsorption mode in beds
designed for sand filtration:
a. was an effective barrier for high influent concentrations
(13 ug/L) of carbon tetrachloride.
b. did not reach breakthrough for carbon tetrachloride for
nine weeks.
c. was passing carbon tetrachloride at concentrations (0.1-
0.3 ug/L) that could not be differentiated from influent
concentrations after four months of operation.
d. was passing 1,4-dichlorobenzene at concentrations that
could not be differentiated from influent concentrations
after five weeks of operation.
5. One and two-and-one-half year old WVW 14x40 GACs receiving chlorina-
ted and settled water in the filtration/adsorption mode in a bed designed for
89
-------
sand filtration were exhausted for the removal of chloroform, bromodichloro-
methane, dibromochloromethane, instantaneous TTHM and THMFP.
6. One and two-and-one-half year old WVW 14x40 GACs operated in the
filtration/adsorption.mode in beds designed for sand filtration desorbed car-
bon tetrachloride.
7. With temperatures in excess of 10°C, total coliform densities and
standard plate count densities in GAG effluent waters occasionally exceeded
densities in GAG influent waters.
8. The bacterial quality of the finished water was satisfactory with
clear well chlorination.
Beaver Falls Authority
Background—
Three virgin GAG beds were evaluated for adsorption of influent instan-
taneous trihalomethanes and influent unreacted precursor (THMFP). The utility
had conducted pilot column studies with several GACs for taste and odor con-
trol but not to determine optimum selection of GAG or bed depth for oreanics
control.
The GACs replaced sand. One bed was filled with 61 cm (24 inches) of
Calgon's Filtrasorb 400 on top of 30 cm (12 inches) of sand and gravel, back-
washed several times to remove fine particulates, and held static under fin-
ished water for six days. A second bed was filled with 61 cm of Calgon's
Filtrasorb C on top of 30 cm of sand and gravel, backwashed several times,
and then held static under finished water for one day. Filtrasorb C was a
Calgon research product designed for adsorption of trihalomethanes. A third
bed was filled with 61 cm of ICl's Hydrodarco 8x16 on top of 30 cm of sand and
backwashed several times. All three beds were placed in service
simultaneously.
The same chlorinated, coagulated and settled water was applied to the
three GAG filter/adsorbers. The filters were geometrically identical except
that the Calgon filters had tile bottoms while the ICI filter had a porous
plate bottom. Although the beds were chosen so that their hydraulic operation
would be identical, the hydraulic data collected during the study indicated
that the bed containing Filtrasorb C had passed approximately 10 percent more
volume than did the other beds. These data are presented in Table 23. The
ICI carbon required less frequent backwashing than did the Calgon carbons.
The ICI carbon was backwashed one to two times weekly throughout the study.
The Calgon carbons were backwashed two to five times weekly during the first
21 weeks and one to three times weekly thereafter.
Treatment is illustrated in Figure 33. Water quality data for the util-
ity are presented in Tables 24 and 25. The virgin GAG beds represented only
30% of the plant capacity.
A problem at the contract laboratory resulted in a significant loss of
samples collected during the first several weeks of the study. Thus, THM data
90
-------
TABLE 23. HYDRAULIC DATA (MEAN VALUES). BEAVER FALLS AUTHORITY
GAG
Parameter
Filtrasorb 400 Filtrasorb C Hydrodarco 8x16
Loading rate, m/hr
(gpm/ft2)
Empty bed contact time,
minutes
GAG depth, cm
(inches)
sand and gravel depth, cm
(inches)
3.1
(1.3)
11.3
61
(24)
30
(12)
3.5
(1.5)
10.1
61
(24)
30
(12)
3.1
(1.3)
11.4
61
(24)
30
(12)
Mlxt=^SETTLE|=y:
MIX
SETTLE
(CHLORINE)
UESEMP
O = SAMPLE POIWT
(OPTIOMAL FEED)
LIME (PAC) 1
CHLORINE (PAC)
T= 9 HR5
II
FILT
4OO
SAC
FILTER
FILT C
&AC
FILTER
HD8XI6
GAC
FILTER
II II
n= 10 HRS
CHLORINE
10%
10%
10%
T« \Z HRS
Figure 33. Water treatment scheme, Beaver Falls Authority,
Eastvale Plant, 17,000 cu m/day (4.5 MGD).
91
-------
TABLE 24. WATER QUALITY DATA (MEAN VALUES) BEAVER FALLS AUTHORITY SEPTEMBER 1977-APRIL 1978
VO
to
Week of
Operation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
17
18
21
23
25
27
29
32
o
Raw
Mean pH = 7.2
Temp, °C
21
21
15
11
16
16
16
10
10
8
6
3
4
2
1
1
1
1
4
4
7
10
11
Turbidity
44
28
22
9.5
7.5
9
10
9
16
10
14
10
22
10
10
12
8
14
10
150
12
8
6
TC° x 103
98
71
140
150
39
190
80
98
220
120
120
69
89
75
65
48
27
6
23
84
13
24
8.4
GAG Influent (Settled)
Mean pH - 7.4
Chlorine'3
Free
2.0
1.7
1.3
1.1
1.2
1.4
1.1
1.0
1.3
1.0
1.4
1.0
1.2
1.3
1.0
1.4
1.0
0.4
0.3
TR
, —
0.2
0.2
Total
1.7
1.4
1.3
1.4
1.6
1.2
1.0
1.6
1.3
1.6
1.1
1.7
1.5
1.2
1.7
1.1
1.6
1.6
1.4
1.4
1.1
1.6
Turbidityb
5.6
4.8
2.3
2.9
2.5
3.3
3.6
3.2
4.6
4.5
3.7
5.9
4.6
6.6
4.8
5.9
5.5
6.4
5.8
6.6
6.3
1.7
1.9
TCC SPCd
<1
<1
< ]_
<1 . 100
<1 800
<1 350
<1 10
<1 42
2 110
<1 33
<1 95
1 360
<1 660
1 200
<1 120
<1 150
<1 33
<1 30
<1 24
<1 38
<1 58
<1 33
<1 17
.Chlorine, mg/L
Turbidity, NTU
,Total coliform/100 inL
Standard plate count/mL
TR = trace
-------
TABLE 25. WATER QUALITY DATA (MEAN VALUES) BEAVER FALLS AUTHORITY SEPTEMBER 1977-APRIL 1978
Week of
Operation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
17
18
21
23
25
27
29
32
Raw
Water
Temp,°C
21
21
15
11
16
16
16
10
10
8
6
3
4
2
1
1
1
1
4
4
7
10
11
Filtrasorb 400
Mean pH = 7.3
Chlorine3
Free
ND
ND
ND
ND
ND
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
TR
__
—
Total
ND
ND
ND
ND
ND
<0 1
<0.1
TR
<0.1
<0.1
<0.1
TR
<0.1
<0.1
<0.1
0.3
0.4
0.2
0.1
0.1
0.3
Turbb
0.4
0.4
0.3
0.3
0.4
0.6
0.3
0.4
0.5
0.3
0.6
0.6
0.4
0.1
0.4
0.4
0.5
0.6
0.6
0.6
0.3
0.3
0.3
THC
64
75
98
45
34
42
78
22
13
12
2
1
<1
1
<1
<1
<1
<1
<1
<1
<1
<1
-------
presented for the Beaver Falls study represent approximately 60% of the sam-
ples collected during the first six weeks the virgin beds were in operation.
Further, no GAG influent data representing the first four weeks of operation
are presented because of a sampling problem.
Trihalomethane Adsorption by Filtrasorb 400—
Figure 34 is a plot of the removal of instantaneous TTHM by Filtrasorb
400 during the first 32 weeks of operation of the virgin bed. The data indi-
cate that the GAG was exhausted for the removal of instantaneous TTHM at nine
to ten weeks of operation.
The GAG was exhausted for the removal of THMFP at approximately eleven
weeks of operation as illustrated in Figure 34. The expected variability of
an instantaneous TTHM concentration may be within ± 20%; the expected varia-
bility of a terminal TTHM concentration may be within ± 16% (Appendix C,
Figure C-ll and C-12). Therefore, the expected variability of the THMFP
concentration may be greater than ± 20%. Beyond the eleventh week of opera-
tion, influent and effluent THMFP concentrations were within ± 20% of one
another and thus could not be differentiated.
The adsorption of individual instantaneous THMs is plotted in Figure 35.
These data indicate that the GAG was exhausted for the removal of chloroform
at nine to ten weeks of operation. In the same manner, Figure 35 indicates
that the GAG was exhausted for the removal of bromodichloromethane at eight to
ten weeks of operation, and exhausted for the removal of dibromochloromethane
at ten to 14 weeks of operation.
Trihalomethane Adsorption by Hydrodarco 8x16—
Figure 36 indicates that HD 8x16 was exhausted for the removal of TTHM at
eight to ten weeks of operation and exhausted for the removal of THMFP at
approximately eleven weeks of operation. Figure 37 indicates that the GAG was
exhausted for the removal of chloroform, bromodichloromethane and dibromo-
chloromethane at eight to ten weeks of operation.
Trihalomethane Adsorption by Filtrasorb C—
Data presented in Figures 38 and 39 indicate that Filtrasorb C was in
operation several weeks longer than were the other GACs before reaching
exhaustion for the removal of instantaneous trihalomethanes. As illustrated
by Figure 38, Filtrasorb C was exhausted for the removal of instantaneous
TTHM at 12 to 15 weeks of operation. Although exhaustion was not apparent
until the 15th week of operation, influent and effluent concentrations were
within 20% of one another beyond the 12th week of operation and could not be
differentiated. (Other GACs were exhausted for TTHM removal at seven to eight
weeks of operation.) Figure 39 indicates that the Filtrasorb C was exhausted
for the removal of chloroform and bromodichloromethane at 12 to 15 weeks and
dibromochloromethane at 14 to 15 weeks. As shown in Figure 38, the GAG was
exhausted for the removal of THMFP at approximately 12 weeks of operation.
Bromoform and Dichloroiodomethane—
The adsorptive capacity of the three GACs for bromoform and dichloroiodo-
methane could not be evaluated because influent and effluent concentrations,
when found, were typically at or below 0.1 ug/L where precision of field data
94
-------
3OO-
200-
vO
2
0
IOO-
INFLUENT
' THMFP
EFFLUEWT THMFP
—i—
IO
—r~
15
2.0
25
100
u
2
O
U
5O -
INFLUENT
TTHM
\
BE-A.VER. FA,L.LS AUTHORITY
GAG - FJLTRASORB 4OO
DEPTH » <°\ CM ( 24 INCHES) GA.C
LOADING RATE«3.1 M/HR (l.3GPM/FT2)
EBCT i 11.3 Ml MUTES
EFFLUEKJT TTHM
5 IO 15 20 25
TIME IN OPERATION, WEEKS
21
IO
TEMPERATURE, ° C
Figure 34. Trihalomethane removal by granular activated carbon.
95
-------
40.
20 .
_l
\
CD
BEAVER FALLS AUTHORITY
GAC = FILTRASORB 4OO
DEPTH = CP\ CM ( 24- INCHES ) GAG
LOAOIMG RATE. 3.1 M/HR (l.SGPM/FT2)
EBCT= 11.3 MIM.
NJFLUEMT
CHCI3
10
15
20
INFLUENT CHBrCl2
EFFLUENT
CHCI3
IO j
5 .
kxaoo
EFFLUENT
5 10 15 20 25
TIME IN OPERATION , WEEKS
30
21
TEMPERATURE, °C
10
Figure 35. Trihalomethane removal by granular activated carbon.
96
-------
too -
5O -
INFLUENT
T T H M v
\O
15
20
25
30
!c
ti
\-
2.
UJ
^200.
0
u
IOO
INFLUENT
THMFP
BEAVER FALLS AUTHORITY
GAC •- HD 8* 16
DEPTH « GI CM (24 INCHES) GAC
LOADING RATE - 3.1 M/HR
EBCT - 11.4
•EFFLUENT THMFP
10 .15 20 25
TIME IN OPERATION, WEEKS
30
q I | 4 10
TEMPERATURE , °C
Figure 36. Trihalomethane removal by granular activated carbon.
97
-------
40.
20 .
EFFLUENT
CHCI3
BEAVER FALLS AUTHORITY
GAG * HD 8 X 16
OEPTH=
-------
too -
50 •
INFLUENT
TTHM
\O
15
20
25
30
2300
Of
h
UJ
^200
O
u
1OO.
EFFLUENT
THMFP
BEAVER FALLS AUTHORITY
GAC * FILTRASORB c
DEPTH -
-------
40.
20 .
BEAVER FALLS AUTHORITY
GAC- FILTASORB C
DEPTH = 61 CM ( 24 INCHES) GAC
LOADING RATE = 3.5 M/HR (\.5 GPM/FT*)
EBCT = IO. I M IN.
EFFLUENT CHCI3
INFLUENT CHCI3
10
15
—i—
20
25
30
220-
h
2 10 -\
hi
O
2
O
1NFLUENJT" CHBrCI2
EFFLUENT CHBrCI?
IO
15
2O
25
30
IO .
5 .
/-INFLUENT CHBr2CI
5 10 15 20 25
TIME IN OPERATION , WEEKS
r
30
21
ft.
9 I I
TEMPERATURE. °C
IO
Figure 39. Trihalomethane removal by granular activated carbon.
100
-------
(Appendix C, Figure C-9 and Table C-6) indicates that they could not be
differentiated.
Desorption of Trihalomethanes—
Near the 21st week of the study, high chlorine demand caused the utility
to stop the practice of breakpoint chlorination. Figures 34 through 39 indi-
cate that influent concentrations of individual THMs and of TTHM decreased
sharply with little or no free chlorine present. These data indicate that
effluent concentrations were significantly higher than influent concentra-
tions, i.e., expected variability of ± 19% to ± 26% (Appendix C, Figures C-l,
2, 4, 6 and 11) would not explain the difference, beyond the 21st week of
operation. It is likely that the three GACs were desorbing THMs beyond the
21st week of operation.
Adsorption of Priority Pollutants and Other Compounds—
Compounds other than trihalomethanes were searched for in GAG influent
and effluent waters to determine their presence or absence and, if present,
their adsorption by virgin GAG. Purgeable halocarbons and base-neutral
extractable halocarbons were detected infrequently. When detected, their
concentrations were low and in ranges where precision of the field data indi-
cates that influent and effluent concentrations could not be differentiated,
i.e., the compounds were typically detected at or below 0.2 ug/L. Until more
sensitive analytical procedures are employed, the adsorptive capacity of GAG
for these compounds at low concentrations cannot be evaluated; however, some
data at low concentration proved informative.
Carbon Tetrachloride—Carbon tetrachloride was not detected in raw water,
but was occasionally detected in treated waters. Its presence likely resulted
from contamination of the chlorine supply. When detected, concentrations were
typically below 0.2 ug/L where precision can be ± 100%. Carbon tetrachloride
data for one sample day are presented in Table 26. These data indicate intro-
duction of carbon tetrachloride during treatment and demonstrate the presence
of the compound in the GAC effluent at concentrations that could not be
differentiated from those in the GAC influent. Thus, the carbons were not
acting as a barrier to routine influent loading after seven months of opera-
tion. This does not imply that exhaustion had occurred or that the carbons
would not act as an effective barrier to a higher influent load.
TABLE 26. CARBON TETRACHLORIDE DATA
BEAVER FALLS AUTHORITY - APRIL 26. 1978
GAC GAC Effluent3
Water Raw Influent F400 FC ICI Finished
Concentration.^ ug/L ND~ 0.3+ <0.1 0.2+ 0.2 0.2
aGAC in operation for seven months. Hydraulic data in Table 23.
t>GC/Hall detector, approximate lower detection level = 0.1 ug/L
ND = not detected
+ = GC/MS confirmed as carbon tetrachloride
- = Carbon tetrachloride not detected by GC/MS at 0.1 ug/L
1,4-Dichlorobenzene—1,4-dichlorobenzene was found occasionally in the
GAC influent. Evaluation of adsorption of this and other base-neutral
extractable compounds was complicated by the losses during extraction (Section
101
-------
5, page 30). 1,4-dichlorobenzene adsorption data, not corrected for extrac-
tion losses, are presented in Table 27. Concentrations in the waters sampled
are somewhat higher than those presented. Further, precision of field data
for the compound indicates that the variability for the data presented in
Table 27 can be ± 70% (Appendix E, Table E-l); therefore, influent and efflu-
ent concentrations of 1,4-dichlorobenzene cannot be differentiated. These
data do not imply exhaustion but indicate the GC/MS confirmed presence of 1,4-
dichlorobenzene in GAG effluents at concentrations that cannot be differen-
tiated from those in the influent after three months of operation.
TABLE 27. REMOVAL OF 1,4-DICHLOROBENZENE BY VIRGIN GRANULAR ACTIVATED CARBONS
BEAVER FALLS AUTHORITY, SEPTEMBER 1977-MARCH 1978
Concentration, a'° ug/L
Week of
Operation
1
2
3
4
5
6
7
9
10
11
12
13
15
18
21
23
27
Effluent0
Influent
—
—
—
— -
ND
—
<0.1
—
ND
0.3
0.2
0.1+
ND
<0.1
<0.1
<0.1
—
F400
ND
ND
<0.1
ND
ND
ND
—
<0.1
ND
ND
0.3
ND
ND
<0.1
<0.1
<0.1
—
FC
<0.1
ND
__
ND
—
0.1
ND
ND
ND
0.2-
0.5+
ND+
ND
<0.1
<0.1
ND
<0.1
ICI
ND
—
_
ND
ND
ND
ND
__
ND
__
0.2
<0.1
ND
0.1+
0.1+
<0.1
ND
aBase-neutral extraction, GC/Hall detector,
approximate lower detection level =0.1 ug/L
NOT CORRECTED FOR EXTRACTION LOSSES
GHydraulic data in Table 23.
ND = not detected
+ = GC/MS confirmed as 1,4-dichlorobenzene
- = 1,4-dichlorobenzene not detected by GC/MS
at approximately 0.15 ug/L
Unidentified Base-Neutral Extractable Halocarbons—At Beaver Falls, base-
neutral extractable halocarbons were occasionally detected in finished waters
but rarely found in raw waters. They are believed to be products of chlorina-
tion or contaminants in the chlorine supply (Section 7). At Beaver Falls, one
such halocarbon was not detected in raw water but was detected 13 of 20 times
in finished water. Another such halocarbon was detected two of 18 times in
raw water but was detected 12 of 18 times in finished waters. When detected,
concentrations in GAG influent waters were lower than concentrations in
finished waters. Adsorption data for these halocarbons are presented in
Tables 28 and 29. Data presented in Table 28 suggest that the halocarbon was
102
-------
present in GAG effluents at concentrations that cannot be differentiated from
those influent after three months of operation. Data presented in Table 29
suggest that Filtrasorb 400 better adsorbed the halocarbon in the first four
months of operation than did the other GACs. However, after four months of
operation, the halocarbon was present in GAG effluents at concentrations that
could not be differentiated from GAG influent concentrations.
TABLE 28. REMOVAL OF UNIDENTIFIED BASE-NEUTRAL EXTRACTABLE HALOCARBONa
BY VIRGIN GRANULAR ACTIVATED CARBON
BEAVER FALLS AUTHORITY, SEPTEMBER 1977-MARCH 1979
Concentration,3''5 ug/L
Week of
Operation
1
2
3
4
5
6
7
9
10
11
12
13
15
18
21
23
27
Effluentc
Influent
— —
—
—
—
0.6
—
<0.1
—
0.2
0.3~
1.2
0.4
NQ
0.1
0.1
ND
—
F400
ND
ND
ND
ND
ND
ND
—
ND
0.1
0.2
0.9
0.3
0.4
NQ
0.2
ND
—
FC
ND
ND
—
ND
—
0.6
<0.1
0.1
0.1
0.8~
1.7
0.4~
NQ
0.2
ND
ND
ND
ICI
ND
—
—
ND
<0.1
0.2
<0.1
—
0.2
—
1.0
0.2
NQ
0.3~
0.1
ND
ND
aUsing procedure described in Appendix D, com-
pound has elution time of 2-chloronaphthalene.
Quantification based on 2-chloronaphthalene.
bNOT CORRECTED FOR EXTRACTION LOSSES
ND = not detected
NQ = Present but not quantified
- = Found not to be 2-chloronaphthalene by GC/MS
Bacteriological Evaluation—
The microbiological characteristics of the raw water and the GAG influent
water during the 32-week study are presented in Table 24 and the data for the
GAG effluent waters are presented in Table 25. The Beaver River raw water was
characterized during weeks one through 32 by a mean total coliform density of
91,000 organisms/100 mL.
A comparison of the total coliform bacterial data in Tables 24 and 25
indicates that the densities in the GAG effluent were in excess of influent
densities during weeks one through 12. The GAG influent coliform densities
were <1/100 mL during the entire study with three exceptions of ^2/100 mL.
During the first twelve weeks, mean coliform densities in the three GAG efflu-
ent waters were: 45/100 mL from Filtrasorb 400; 42/100 mL from Filtrasorb C;
103
-------
TABLE 29. REMOVAL OF UNIDENTIFIED BASE-NEUTRAL EXTRACTABLE HALOCARBON3
BY VIRGIN GRANULAR ACTIVATED CARBON
BEAVER FALLS AUTHORITY. SEPTEMBER 1977-MARCH 1978
Week of
Operation
1
2
3
4
5
6
7
9
10
11
12
13
15
18
21
23
27
Concentration,
Influent F400
ND
ND
ND
ND
<0.1 ND
ND
0.1
ND
<0.1 ND
<0.1 ND
<0.1 ND
<0.1 ND
<0.1 ND
<0.1 <0.1
<0.1 <0.1
<0.1 <0.1
—
a»D ug/L
Effluent^
FC
ND
ND
ND
—
<0.1
<0.1
<0.1
<0.1
ND
ND
<0.1
<0.1
<0.1
<0.1
-------
TABLE 30. WATER QUALITY DATA BEAVER FALLS AUTHORITY SEPTEMBER 1978-DECEMBER 1978
o
Ul
GAG Influent
(Settled) Filtrasorb
Week of
Operation
53
54
55
56
57
58
59
60
61
62
63
64
Raw
Temp, °C
26
23
22
19
14
12
14
13
11
9
8
6
TCa
18,000
10,000
22,000
9,200
31,000
10,000
8,700
19,000
5,000
12,000
82,000
8,000
Free
Chlorine
1.4
1.2
1.6
1.6
1.4
1.1
1.4
1.3
1.5
0.8
1.2
1.0
Free ,
TCa Chlorine
<1 TR
<1 TR
<1 TR
<1 TR
<1 TR
<1 TR
<1 TR
<1 TR
<1 TR
<1 TR
<1 TR
<1 TR
400
TC3
100
120
230
470
62
44
30
8
—
1
<1
-------
indicate that effluent total coliform densities from all three GAG beds again
exceeded influent densities of <1/100 mL when temperatures were above 10°C.
As the temperatures dropped below 10°C, effluent total coliform densities
from all three GAG beds measured <1/100 mL.
Rate of reproduction of bacteria in the GAG beds was the probable cause
of higher GAG effluent bacterial densities when temperatures exceeded 10°C.
Other conditions that may have favored growth on the GAG were the reduction
of free chlorine, the large surface area, and the possible accumulation of
nutrients.
Finished water quality was adequately maintained during the study at a
total coliform density of <1/100 mL and a standard plate count density of
<500/mL with the application of chlorine following GAG adsorption/filtration.
Findings—
1. Trihalomethane formation occurred during treatment following chlorine
application and generally varied with water temperature.
2. Virgin GAG receiving chlorinated, settled water and operating in the
filtration/adsorption mode in beds designed for sand filtration during warmer
months was exhausted for the removal of:
Weeks to Exhaustion
Chloroform
Bromodichloromethane
Dibromochloromethane
Inst TTHM
THMFP
Filtrasorb Filtrasorb
400 C
9-10 12-15
8-10 12-15
10 - 14 14 - 15
9-10 12-15
12 12
Hydrodarco
8x16
8-10
8-10
8-10
8-10
11
3. When breakpoint chlorination was discontinued, resulting in signifi-
cant reduction of GAG influent trihalomethane concentrations, five-month-old
GACs desorbed trihalomethanes.
4. GACs operated in the filtration/adsorption mode in beds designed for
sand filtration: /
a. passed carbon tetrachloride at concentrations (0.1-0.3 ug/L)
that could not be differentiated from influent concentrations
after seven months of operation.
b. passed 1,4-dichlorobenzene at concentrations that could not be
differentiated from influent concentrations after three months
of operation.
5. With temperatures in excess of 10°C, total coliform densities and
standard plate count densities in GAG effluent waters greatly exceeded den-
sities in GAG influent waters.
106
-------
6. The bacterial quality of the finished water was satisfactory with
clear well chlorination.
CONCLUSIONS FROM TRIHALOMETHANE TREATABILITY STUDIES
1. A change in the chlorine application point to a better quality water
was a viable approach to trihalomethane control.
2. Moving the point of chlorine application resulted in lower finished
water instantaneous trihalomethanes because a better quality water in terms of
reduced THMFP was chlorinated and/or because in-plant THM reaction time was
reduced.
3. The use of chlorine dioxide as an alternative disinfectant to chlo-
rine was a viable approach to trihalomethane control.
4. Ammoniation was a viable approach to trihalomethane control.
5. Relatively higher concentrations of brominated THMs resulted in fin-
ished water when the in-plant reaction time with free chlorine was reduced.
6. Granular activated carbon was effective for trihalomethane control
for short periods of time but would not be effective for long periods of time
without reactivation.
7. The extent to which a utility can lower its trihalomethane levels
will depend on its physical plant, its adaptability to these and other changes
in treatment, and its financial capability.
8. Any modification to treatment should not be evaluated by instantane-
ous trihalomethane concentrations alone. Terminal trihalomethane concentra-
tions and THMFP can define the changing levels of precursor in raw water and
can define the effects of treatment on precursor levels. An understanding of
precursor is necessary for an evaluation of the modification.
9. Raw water precursor levels, as measured by terminal level trihalo-
methane concentrations, can vary significantly over short periods of time. A
better evaluation of changing levels of raw water precursor and of the effects
of treatment on precursor levels will be made as the number of instantaneous
and terminal level trihalomethane samples increases.
10. Treatment modifications should not be evaluated without monitoring
the bacterial quality of in-plant and finished waters.
11. Any modification to treatment should be studied over a long period of
time. Seasonal effects in bacterial densities and trihalomethane formation
should be evaluated. Changes in raw water precursor levels should be evalua-
ted. Other changes in water quality may affect results.
12. The effect of PAC, permanganate or chlorine dioxide on precursor
could not be determined because raw water precursor levels varied significant-
ly over a short time period, feed of these materials preceeded coagulation and
107
-------
settling, and settling normally reduced precursor levels.
13. Reduction in terminal TTHM concentrations generally coincided with
reduction in turbidity levels.
108
-------
SECTION 7
ORGANIC COMPOUND SURVEY
GENERAL
Project activities included sampling for analysis for selected organic
Priority Pollutants in raw and finished waters at all project utilities once
a month from July 1977 to June 1978. In-plant waters were not sampled as a
part of this survey. Raw and finished waters were sampled following theoreti-
cal plug flow through the plant. Although the raw and finished waters at a
given utility could be compared, similar comparisons between utilities were of
limited value.
Schematic treatment diagrams representative of routine treatment at the
project utilities during the sample year are given in Figures 12, 14, 16, 18,
23, 29 and 33 in Section 6 and Figures 40 to 43 in this section. Although
those diagrams presented in Section 6 are representative of treatment at the
time trihalomethane control studies were conducted, they also describe treat-
ment representative of the sample year.
All utilities treating surface waters practiced chlorination. The reac-
tion between chlorine and precursor, discussed in Section 6, resulted in tri-
halomethane formation during treatment at these utilities. The extent of
trihalomethane formation at each utility depended upon its treatment
processes, pH levels, chlorine feed rates, ammonia levels, in-plant THM reac-
tion time, etc.
SURVEY FOR PURGEABLE HALOCARBONS
Discussion of purgeable halocarbons is based on GC/Hall and GC/MS analy-
ses of project samples and on accumulated purgeable halocarbon quality assur-
ance data (Appendix C). The following discussions are based on the quality
assurance procedures and methods of interpretation discussed in Section 5.
Chloroform (Raw water data: Table 32. Finished water data: Tables 33 and 34.
Quality assurance data: Table C-l and Figures C-l and C-2.)
Chloroform was detected in 139 of 198 raw water samples and in 169 of 170
finished water samples. Mean raw water chloroform concentration, when
detected, was 0.8 ug/L. Mean annual finished water chloroform concentration
was 35 ug/L for treated surface waters and 0.9 ug/L for West View's treated
ground water.
Chloroform was found in 100% of chlorinated surface waters. Finished
109
-------
pHO.4-
T=0 MRS
GROUND WATER Q =
CHLORIWE
(PAC)
ION
EXCHANGE
37%
FERRO-
SANO
LEGEND
CHLORINE
SAMPLE POINT
(OPTIONAL FEED)
T= '/e MRS
HR5
t
NaOH
(CHLORINE)
(CI02)
CLEAR
WELL
= 8-0
Figure 40. Treatment at West View Water Authority, 57,000 cu m/day (15 MGD).
-------
CHLORIWE.* 1-4 PPM
ALUM
(PAC)
(KMnO4)
(POLYMER)
to % 1
c/
UJ
£
o
I
lo
F<>^
: MIX
CLARI
FY
LIME
(PAC)
FLUORIDE
I
MIX
FILTER
T«
4V2 MRS
T*OHRS
pH«7.1
40% \ L
i
K A 1 V
IvuX
^*i
v^L
A O 1 CV
.f\^l r \
\
MIV
CLARIFY
LIME
CHLORINE=I-4PPM /o^rN
A i i i K A 1 r r^^< i
\
FILTER
FREE CHLORIME= 1-5 PPM
• > . 1 J V (~\
_ CLEAR. _^X
1-^/2 -
-------
CHLORINE = |-2 PPM
CHLORINE«0.3 PPM
°H 1
UJ \
!>) * i
z
Ul
I
UJ
J
-"
'Ul
I
U)
U)
_J
CHLORINE= 1-2 PPM CHLORIM E « O.2 PPM
1 Cl°2
1
^ t 1 , | 1 _j 1, 1 riFAP
=O i: MIX SETTLE — CLAR.IFY -FILTER — -vX/ELL ^O^
T=0 MRS 4 'T=3HRS
LIME ASH TOTAL CHLORINE = A
i /\O
ALUM LEGEND
A./- -ri\/ A-r- cr-\ cii i^- A
FLUORIDE
O SAMPLE POINT
A= O-3 PPM AT 4°C
1-0 PPM AT 28° C
Figure 43. Treatment at Wilkinsburg-Penn Joint Water Authority, 95,000 cu m/day (25 MGD).
-------
water chloroform concentrations were typically lower at utilities attempting
to minimize chlorine feed rates, i.e., Wilkinsburg, and typically higher at
utilities carrying finished water free chlorine residuals at or above 1.5
ug/L, i.e., Wheeling, Louisville or Evansville. Finished water chloroform
concentrations were typically higher where finished water pH was high, i.e.,
Wheeling. Finished water chloroform concentrations were lower in the coldest
months of the year and higher in the warmest months of the year. When West
View's ground water was chlorinated, trihalomethane formation did not exceed
1.2 ug/L and no seasonal pattern was apparent.
Chloroform levels reaching the consumer will be higher than levels pre-
sented in Tables 33 and 34 if a free chlorine residual persists in the distri-
bution system.
Bromodichloromethane (Raw water data: Table 35. Finished water data: Tables
36 and 37. Quality assurance data: Table C-2 and Figure C-4.)
Bromodichloromethane was detected in 84 of 200 raw water samples and in
all 170 finished water samples. The mean raw water bromodichloromethane con-
centration, when detected, was 0.3 ug/L. The mean annual finished water
bromodichloromethane concentration was 13 ug/L for treated surface waters and
0.4 ug/L for treated ground water. As with chloroform, the formation of
bromodichloromethane resulted from in-plant chlorination, varied with seasonal
temperature (except for the ground water) and was different for each utility's
treatment.
Dibromochloromethane (Raw water data: Table 38. Finished water data: Tables
39 and 40. Quality assurance data: Table C-3 and Figure C-6.)
Dibromochloromethane was detected in 33 of 200 raw waters and in 168 of
170 finished waters. Mean raw water concentration, when detected, was 0.2
ug/L. Mean annual finished water concentration was 5.6 ug/L for treated
surface waters and 0.3 ug/L for treated ground water. As with chloroform, the
formation of dibromochloromethane resulted from in-plant chlorination, varied
with seasonal temperature (except for ground water) and was different for each
utility's treatment.
Bromoform (Raw water data: Table 41. Finished water data: Tables 42 and 43.
Quality assurance data: Table C-4 and Figure C-9.)
Bromoform was detected in 8 of 200 raw waters and in 114 of 170 finished
waters. Raw water concentrations did not exceed 0.1 ug/L. Finished water
concentrations, when detected, averaged 0.8 ug/L in treated surface waters and
0.1 ug/L in treated ground water. As with chloroform, the formation of bromo-
form resulted from in-plant chlorination, varied with seasonal temperature
(except for ground water) and was different for each utility's treatment.
Dichloroiodomethane (Raw water data: Table 44. Finished water data: Tables
45 and 46. Quality assurance data: Table C-6.)
Dichloroiodomethane was rarely detected (frequency = 1/200) in raw water
and was detected in 81 of 170 finished water samples. Raw water concentra-
113
-------
tions did not exceed 0.1 ug/L. Finished water concentrations, when detected,
averaged 0.2 ug/L in treated surface waters and were less than 0.1 ug/L in
treated ground water. As with chloroform, the formation of dichloroiodome-
thane resulted from in-plant chlorination and generally varied with seasonal
temperature (except for ground water). Because the precision of dichloroio-
domethane data below 0.2 ug/L may be ± 100%, caution is suggested in conclud-
ing that this compound was absent in Evansville's waters or that it occurred
infrequently in other utility waters.
Total Trihalomethane (Finished water data: Table 47. Quality assurance data:
Figure C-ll.)
As with the individual trihalomethane compounds, finished water TTHM con-
centrations varied with seasonal temperatures and were different for each
utility's treatment. The seasonal trend was not apparent at West View where
ground water is chlorinated. TTHM levels reaching the consumer will be higher
than the levels presented in Table 47 if a free chlorine residual persists in
the distribution system because finished waters contain trihalomethane for-
mation potential.
Trihalomethane Formation Potential (THMFP)
Once a month, or more frequently if THM control studies were conducted,
waters were sampled for analysis of instantaneous level THMs and terminal
level THMs. Instantaneous level THM data are presented in Tables 32 through
47. As explained in Section 4, pages 11 and 12, terminal level THM data can
be used to evaluate precursor levels. Such data for raw and finished water,
then, allow the evaluation of THM formation and reduction of precursor levels
in-plant, as shown in Tables 48 through 57.
Table 48 presents these data for Huntington. In July, for example, at
Huntington, the raw water mean terminal TTHM concentration for several sample
days was 327 ug/L. Because the mean instantaneous TTHM concentration was <1
ug/L, the mean raw water THMFP was 326 ug/L. Finished water mean concentra-
tions were 232 ug/L terminal, 112 ug/L instantaneous and 120 ug/L THMFP.
Thus, treatment affected raw water THMFP, or raw water unreacted precursor,
in several ways. Chlorine reacted to form 112 ug/L TTHM, accounting for 34%
of the raw water THMFP. Treatment, principally coagulation and settling,
removed 29% of the raw water THMFP. Thus, 37% of the raw water THMFP remained
after treatment and had the potential to form an additional 120 ug/L TTHM in
the distribution system.
Less than 120 ug/L TTHM may have been formed in the distribution system
because system detention time was less than the seven-day storage period for
the terminal level parameter, distribution system free chlorine residuals were
less than the 15 mg/L free chlorine added to drive the THM reaction during the
storage period, and storage conditions for determination of the terminal level
parameter (headspace free in clean glassware) are unlike distribution system
conduit and storage tanks. Nevertheless, the finished water had the potential
to form further THMs in the distribution system.
When these Huntington data were evaluated over a one-year period, they
114
-------
indicated that 30% of the raw water precursor formed TTHM, 29% of the precur-
sor was removed by treatment, and 41% entered the distribution system with the
potential for further THM formation.
Averaging data from the ten utilities treating surface water indicated
that 23% of the raw water THMFP was converted to TTHM during treatment, 37% of
the raw water THMFP was removed by treatment, and 40% of the raw water THMFP
was discharged to the distribution system. Thus, trihalomethane formation
will continue in the distribution system if a free chlorine residual is
present.
Such percentages are presented in an attempt to evaluate treatment. The
significance of these percentages cannot be defined. It is known that the
expected variability of an instantaneous TTHM concentration may be ± 20%
(Figure C-ll) and that the expected variability of a terminal TTHM concentra-
tion may be ± 16% (Figure C-12), but the expected variability of the differ-
ence of these, i.e., THMFP, or a ratio of these, i.e., percentage, cannot be
defined.
Comparison of these data for several utilities should be made cautiously
for several reasons: chlorine application rates can vary from month to month
within a utility and do vary among utilities; in-plant THM reaction times vary
among utilities; coagulants and their effectiveness vary among utilities; pH
varies among utilities, etc.; the significance of such data cannot be defined.
Raw water THMFP concentrations were evaluated to determine if precursor
varied seasonally. Because the storage temperature of samples for determina-
tion of the terminal level TTHM was at or near the finished water temperature,
it was expected that raw water THMFP concentrations would be lowest when water
temperatures were coldest. For all ten utilities, Figure 44 presents monthly
mean storage temperature data and raw water THMFP concentrations plotted
against time. Initially, terminal level samples were stored at room tempera-
ture. When water temperature began falling, terminal level samples were
stored at or near finished water temperature. Figure 44 presents mean storage
temperature and mean raw water temperature for the initial months of the
study.
These data indicate that from October through June, temperature and raw
water THMFP concentrations generally varied in the same direction. However,
from August through October, raw water THMFP concentrations increased while
storage temperatures remained constant and raw water temperature decreased.
This suggests that precursor levels were higher between August and October
than at other times of the year.
Seasonal variation in raw water THMFP data and data for the fate of raw
water THMFP are not presented for West View's ground water. These data were
highly variable both in the terminal TTHM concentrations formed and in the
amounts of chlorine consumed during storage for determination of this para-
meter. Finished water instantaneous TTHM concentrations for this utility,
however, demonstrate that the ground water precursor differed from the surface
water precursor because West View finished water total trihalomethanes never
exceeded 2 ug/L.
US
-------
Carbon Tetrachloride (Raw water data: Table 58. Finished water data: Table
59. Quality assurance data: Table C-5 and Figure C-8.)
With one exception, carbon tetrachloride was not detected in untreated
surface waters upstream from Huntington. The frequency of detecting carbon
tetrachloride in untreated surface waters was highest at Huntington and
decreased with increasing distance downstream. On one occasion, the compound
was GC/MS confirmed in the Allegheny River.
In another ORSANCO project utilizing the same analytical procedure and
laboratory, carbon tetrachloride was present at 83% frequency in the Kanawha
River at concentrations up to 1.9
Carbon tetrachloride was occasionally detected in finished waters at all
utilities except in treated ground water at West View. The presence of this
compound in finished waters may be attributed to low level carbon tetrachlor-
ide contamination of chlorine used for disinfection. Periodic chlorine con-
tamination is suggested by one-time finished water carbon tetrachloride con-
centrations at Louisville and Evansville of 1.3 ug/L and 6 ug/L, respectively.
At Huntington, finished water carbon tetrachloride levels were signifi-
cantly higher than levels found in untreated surface water, i.e., the preci-
sion of the data indicates that the levels could not be the same. In addition
to the possibility of carbon tetrachloride contamination of the chlorine
supply, the increase was attributed to desorption of carbon tetrachloride from
the one to two-year-old GAC filter/adsorbers in place at the utility (Section
6, Table 22). Carbon tetrachloride was detected 47% of the time (23/49) in
Huntington' s raw water but was detected 100% of the time in its finished
water.
Chlorobenzene (Raw water data: Table 60. Finished water data: Table 61.
Quality assurance data: Table C-7.)
The presence of chlorobenzene was GC/MS confirmed in untreated surface
waters at Huntington and in untreated ground waters at West View. Accompany-
ing finished waters at both locations also contained chlorobenzene. The
frequency and concentrations of the data at Huntington are similar for raw
and finished waters. At West View, however, all nine finished water samples
contained chlorobenzene, while it was detected in only five of eleven raw
water samples. The reason for the difference in frequency of data in raw and
finished waters is not known.
In late March and early April 1978, Louisville was asked by project staff
to increase once-a-month sampling frequency when ORSANCO was notified of a
chlorobenzene spill. The resultant data (Table 31) indicate that chloroben-
zene concentrations reached 8.5 ug/L in the finished water and suggest that
conventional treatment at Louisville (raw water chlorination, settling, PAC,
filtration and post-chlorination) was not effective for chlorobenzene removal.
1, 1-Dichloroethane (Raw water data: Table 62. Finished water data: Table 63.
Quality assurance data: Table C-8.)
116
-------
TABLE 31.
CHLOROBENZENE LEVELS, LOUISVILLE WATER COMPANY3
chlorobenzene,t> ug/L
day
March 29
March 30
March 31
March 31
April 1
time
afternoon
morning
morning
afternoon
morning^
raw water
0.8
1.6
5.0
2.1
0.1
finished water
—
1.1
2.5
8.5
5.3
aPlant detention time typically 30 hours
'•'GC/Hall detector, approximate lower
detection level 0.1 ug/L
The presence of 1,1-dichloroethane was presumptively reported at several
utility locations in both raw and finished waters at concentrations less than
1.0 ug/L. Its presence was GC/MS confirmed only in raw water at Wilkinsburg
on one occasion and in raw and finished ground water at West View. There was
no significant difference in the frequency and concentration of 1,1-dichloro-
ethane for raw and finished water at West View.
1,2-Dichloroethane (Raw water data: Table 64. Finished water data: Table 65.
Quality assurance data: Tables C-9 and C-10.)
1,2-dichloroethane was detected in the raw waters of eight project utili-
ties with the frequency of detection increasing at and downstream from
Huntington. The presence of 1,2-dichloroethane was GC/MS confirmed in raw
waters at seven of those utilities. In finished waters, 1,2-dichloroethane
was detected at four utilities only and GC/MS confirmed at two of those
locations.
Review of project data for 1,2-dichloroethane indicated that the presence
of large chloroform peaks eluting immediately ahead of this compound in pro-
ject samples interfered with both its detection and quantification. The con-
centrations of 1,2-dichloroethane when found in raw waters were typically at
or below 0.5 ug/L. Chloroform concentrations in raw water were typically at
or below 1.0 ug/L and thus did not cause interference. In chlorinated waters,
however, where chloroform concentrations were much higher and where 1,2-
dichloroethane was found in the accompanying raw water, the compound was not
detected. The chromatograms gave the visual appearance of a small deviation
in the smooth tailing edge of the chloroform peak, a deviation that had
insufficient slope change to cause integration (qualification and quantifica-
tion). The difference in frequency of detection of 1,2-dichloroethane in
project raw and finished samples is likely related to such chloroform
interferences.
1,2-Dichloropropane (Raw water data: Table 66. Finished water data: Table
67. Quality assurance data: Table C-ll.)
1,2-dichloropropane was detected infrequently in raw water samples from
seven project utilities; the presence was GC/MS confirmed at two of those
locations. In finished water samples, 1,2-dichloropropane was detected infre-
quently at ten utilities and GC/MS confirmed at two of those locations. Con-
centrations in both raw and finished waters never exceeded 0.2 ug/L.
117
-------
trans-1,3-Dichloropropene (Raw water data: Table 68. Finished water data:
Table 69. Quality assurance data: Table C-12.)
Trans-1,3-dichloropropene was detected only once and was of insufficient
concentration for GC/MS confirmation. The compound was not found in project
raw or finished waters at concentrations above 0.1 ug/L.
cis-1,3-Dichloropropene and/or 1,1,2-Trichloroethane
The compounds cis-1,3-dichloropropene and 1,1,2-trichloroethane co-elute
with dibromochloromethane. Data presented in Table 38 indicate that detection
at 0.1 ug/L of the co-eluters was infrequent in untreated surface waters and
concentrations never exceeded 0.7 ug/L. GC/MS confirmation attempts for
dibromochloromethane in untreated surface water were positive. One GC/MS
confirmation attempt for cis-1,3-dichloropropene in untreated surface water
proved negative.
The co-eluting compounds were detected in all chlorinated, finished sur-
face water samples (Table 39), lending support to the presence of the dibromo-
chloromethane. GC/MS confirmation attempts for dibromochloromethane in
finished surface waters were positive; whereas, GC/MS confirmation attempts
for cis-1,3-dichloropropene and 1,1,2-trichloroethane in finished surface
waters were negative. It is believed that cis-1,3-dichloropropene and/or
1,1,2-trichloroethane rarely occurred in raw and finished surface waters.
Cis-1,3-dichloropropene and/or 1,1,2-trichloroethane were presumptively
identified on two occasions in untreated and finished ground water at West
View. GC/MS confirmation was not possible.
1,1,1-Trichloroethane, Trichloroethylene, and 1,1,2,2-Tetrachloroethane
and/or Tetrachloroethylene (Quality assurance data: Tables C-13 to C-15.)
Constantly occurring interferences in all system blanks and project
samples were apparent at the relative retention times of 1,1,1-trichloro-
ethane, trichloroethylene, and 1,1,2,2-tetrachloroethane and/or tetrachloro-
ethylene (Figure 4 and 5), and were GC/MS confirmed as being those compounds.
An extensive investigation was conducted by the laboratory to determine the
source of contamination and to eliminate or control it at acceptable concen-
trations. It was determined that laboratory air was probably the source of
contamination. System exposure to laboratory air was minimized and the con-
centrations of contaminants were reduced.
The concentrations of contamination in system blanks were evaluated over
a period of occurrence and statistically weighted (mean concentration plus two
standard deviations) to reflect the interference for that period. This sta-
tistical correction was then subtracted from all sample data produced during
that period. When the level of interference in a daily system blank exceeded
the statistical correction, the daily blank correction was subtracted from
all sample data produced that day.
A review of the resulting data after blank correction led to the conclu-
sion that the presence of these compounds in project samples could not be
118
-------
reported. The resulting data reflected the highly variable nature of the con-
taminants and may have falsely suggested the absence of a compound. Thus,
while the GC/Hall detection levels of these compounds were approximately 0.1
ug/L, they could not be reported below the following: 2.6 ug/L for 1,1,1-
trichloroethane, 1.9 ug/L for trichloroethylene, and 3.4 ug/L for 1,1,2,2-
tetrachloroethane and/or tetrachloroethylene. It is likely that these
compounds were not present in project raw or finished waters above those con-
centrations. However, as mentioned in Section 6, page 43, high tetrachloro-
ethylene concentrations (up to 60 ug/L) were observed and GC/MS confirmed in
the Allegheny River. (Text continues on page 159.)
119
-------
TABLE 32. CHLOROFORM RAW WATER DATA, JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Utility3
i
Fox Chapel
iWilkinsburg
i • —
Pittsburgh
WPW/Hays Mineb
[West View
Beaver Falls
Wheeling
iHuntington
Cincinnati
i ,
Louisville
Evansville
Total or Mean
West View
0
U-l
o
£
u
t-t
0)
CO
rH
iH
CO
PC
CO
CU
•H
H
11
12
11
6
11
29
8
49
17
22
11
187
11
h-3
00
o •
<4H O
iH
cfl
PC
CO
cu
B
•H
H
5
4
7
5
1
25
5
38
14
13
10
127
2
r-3
OO
-0 3
C
3 rH
0 •
4-1 O
rH
CO
EC
CO
cu
H
1
1
1
1
1
1
1
3
0
0
0
10
0
B
O
•H
cfl ,.4
4-1 OO
B 3
CU
O rH
C •
0 0
°A\
rH
rH C
CO
O 4-> O
o a c
LO 0) II
Cfl CO i-H
0) cfl
B EC
H
§
rC T3
S CU
•o a
cu cu
H CU
B 4->
o o
o B
C/) II
CO rH
CU CO
B EC
H
i
b = Western Pennsylvania Water Co., Hays Mine Plant
c = Ohio River at West View
d = Ground water supply
120
-------
cr
n
n
o s; co o
l-l ft) CD H
o co ro ft
C rt cu
3 fD **j H
&. H H-
3 00 S!
0 CfD
jB hd i-j I—'
Pt ft) fD H1
co
3
CO
m
l-h
i-h
I-1
(B
3
05
P>
rt
ft)
i-l
O
O
s?
3
fD
>-o
M
3
«
n>
CO
rt
<
H-
§
CL
VO
oo
0
o
•
^0
M
•
to
H
O
rt
W
M
0
H
!
o
M
ON
M
h-1
C^
H1
0
w
Ul
H
00
0
Ul
Ln
Evansville
(-1
r-o
M
NJ
O
CTv
O
oo
c^
Louisville
N3
M
ro
h-1
O
Ul
(-•
H1
O
0
tN3
(S3
Cincinnati
M
VO
i-1
*J3
O
U)
Ul
-^1
a\
M
H1
Hunt ing ton
S3
JN
N5
J^-
o
J>
to
M
00
o
Wheeling
M
M
M
I-1
O
Ui
VO
M
N5
O
I
Beaver Falls
S3
•-J
to
^i
0
-P-
H
^O
K3
WPW/Hays Mine°
h-1
h-»
M
h-1
O
M
oo
U1
h-»
Pittsburgh
M
N3
H"
to
O
to
O
O>
OJ
H-1
M
Wilkinsburg
H
N3
M
10
O
OO
w
10
Ul
>*
O
X
n
Co
T3
o
(-»
M
N3
M
(S3
O
OO
•
\->
t*3
VO
M
M
C
rt
K-
I—1
!-••
rt
"^
cr
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall >Q.I ug/L
Times MS confirmation
attempted when
Hall <0.1 ue/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
TABLE 33. CHLOROFORM FINISHED3 WATER DATA, JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
-------
TABLE 34. FINISHED WATER3 CHLOROFORM LEVELS 1977-1978 GC/HALL DETECTOR
Utility
Wilkinsburg3
Fox Chapel
Pittsburgh3
WPW/Hays Mineb
Beaver Falls
Wheeling
Huntington
Cincinnati
Louisville
Evansville
West View11
Jul
8.2
3.7
38+
9.3
—
120
38
46+
67*
33
ND
Aug
25
29
34
23
40
—
78
49
85g
82
0.5
Sep
22
29
63
11
69
87
62
62e
—
86
—
Mean Concentration
Oct
14
11
33
24
39
100
46
64e
65§
86
1.0
Nov
5.6
2.4
40
13
50
66
26
20
77+
70
1.1
Dec
3.2
1.2
36
18
40
49
22
22
48
61
—
Jan
1.6
1.4
4.4
9.0
38
24
—
26
27
36
1.1
ug/L
Feb
1.4
1.4
1.7
6.5
32^
16
15
9.1
33
17
1.2
Mar
2.4
2.7+
4.1
15
6.5d
38
20
20
45
39
0.8
Apr
2.1
3.3
32
51
7.6
39
40
26
12
58
—
May
6.1
6.1
19
22
—
47
51
27
35
71
0.4
Jun
6.4
6.9
3.5
__
92
62
59
' 52
57
84
0.8
Annual
Mean
8.3
8.1
26
18
41
59
42
35
51
60
0.8
NO
N5
a = Clear well sample
b = Western Pennsylvania Water Co./Hays Mine Plant
c = February 1-15
d = February 21-March 31
e = Normal operation only. Not representative of treatment modification reported in Section 6.
f = MS confirmed in one sample. Others not MS attempted.
g - Treatment modification reported in Section 6.
h = Ground water supply
+ = MS confirmed
—No data available
ND = not detected
-------
NJ
Co
o. n
II II
O O
H 3*
O H-
c o
3
CL ?d
* 5"
to n>
rt n
m
i-i to
rt
CO
C S3
•O CO
M rt
<
H-
fD
O4
II
S!
(o
CO
rt
n>
>t
S
fD
3
CO
V)
R
n
o
CO
2
H-
3
m
to
to
n
CO
fD
fD
OQ
ft!
I-1
£,
fD
CO
rt
H-
P
I-"
N3
O
t-0
O
*
S3
H
O
rt
to
o
i-i
3?
§
00
VO
Co
Ul
o
o
Co
ON
N5
M
Ul
CO
Evansville
H1
CO
CO
o
o
Ul
H
-
H
-
Louisville
N3
*-
*
O
N3
O
.p-
Cincinnati
5
CO
*.
o
S3
p
-
o
Huntington
-p-
VO
H
P
O
CO
O
-
-
-
M
Wheeling
oo
o
CO
Beaver Falls
N3
CO
ro
o
o
N3
-
o
as
fD
co
rt
H-
P
O
O
jWPW/Hays Mineb
oo
M
0
O
o
i Pittsburgh
P
CO
-
o
H
1
Wilkinsburg
S3
O
M
H"
h-1
O
n
-o
fD
P
0
CO
rt
H-
f-1
H-
rt
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
^1
*-?
>
hH
M
2
3
r4
*3ti
0
a
H
M
n
H
M
O
z
1
H
OQ
w
2I
t»
O
*j M
M
t"1 3>
O £3
*l
O M
M £d
H
M H
££
^E
11 3
oo
-------
ro
G. o
n n
o S3
>"< ro
o co
Q n
D ro
ag
ft ro
ro 3
H 3
co
co •
BJ'
(D
i-l
n
o
ffi
Co
CD
S
3'
ro
at
3
rt
a = Clear well effluent
b = see Figure 1
ro
CO
n
p.
o.
-0
»
o
o
•
p
oo
H
O
rt
B>
t-1
O
ro
BJ
3
M
0-v
h-i
0
M
Ul
Ui
Ui
Evansville
M
ro
ro
o
to
Ul
-p-
Louisville
ro
ro
o
VD
00
H
^
Cincinnati
H
5
0
M
00
ro
h-1
H
Hun ting ton
ro
ro
o
M
-
Wheeling
P
M
H
0
LO
UJ
W
CD
05
M
h- "
CO
ro
3
o
M
ro
VO
ro
ro
WPW/Hays Mine°
M
M
I-1
O
ro
M
Pittsburgh
ro
M
ro
o
M
(jO
ro
-
M
S!
H-
M
H-
3
CO
0"
OQ
N3
(-"
ro
o
OJ
M
O
T!
O
X
n
T3
ro
ro
M
ro
o
K3
ID
H
N3
C
rt
H-
I—1
H-
rt
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 UE/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
GC/HALL E
GC/W
ETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
S, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
w
I
§
H
fc
M
O
>
r
a
H
vo
—i
00
-------
TABLE 37. FINISHED WATERa BROMODICHLOROMETHANE LEVELS 1977-1978 GC/HALL DETECTOR
Utility
Fox Chapel
Wilkinsburg3
Pittsburgh3
WPW/Hays Mineb
Beaver Falls
Wheeling
Hunt ing ton
Cincinnati
Louisville
Evansville
West View11
Mean Concentration, ug/L
Jul
3.5
5.9
16+
3.1
—
33
29
36+
40f
35
0.4
Aug
12
10
26
14
29
—
30
42
42§
54
0.1
Sep
7.9
7.6
27
7.4
18
33
27
30e
—
27
—
Oct
3.2
5.4
20
8.6
15
22
18
30e
23g
29
0.4
Nov
0.7
1.7
16
3.6
16*
14
11
14
20
25
0.6
Dec
0.3
1.9
13
5.4
11
4.3
6.5
11
12
9.7
—
Jan
0.6
1.5
1.5
2.7
11
2.9
—
13
9.5
8.5
0.8
Feb
0.8
0.9
0.5
2.2
10C
2.6
6.1
2.9
12
6.8
0.5
Mar
0.7
1.4
1.1
2.4
2.4d
3.6
5.3
5.9
9.9
13
0.4
Apr
0.8
0.7
17
10
3.1
7.0
14
12
6.6
13
—
May
1.7
2.4
14
9.1
—
10
13
15
12
14
0.4
Jun
3.1
3.3
3.7
—
26
14
25
24
22
24
0.4
Annual
Mean
2.9
3.6
13
6.2
14
13
17
18
19
22
0.4
Ul
a = Clear well sample
b = Western Pennsylvania Water Co./Hays Mine Plant
c = February 1-15
d = February 21-March 31
e = Normal operation only. Not representative of treatment modification reported in Section 6.
f = MS confirmed in one sample. Others not MS attempted.
g = Treatment modification reported in Section 6.
h = Ground water supply
+ = MS confirmed
—No data available
ND = not detected
-------
II II
to
a = Tabled GC/Hall data represents dibromi
eis-l,3-dichloropropene and/or 1,1,
b = Tabled GC/MS data represents dibromoc]
c = See Figure 1.
d = Western Pennsylvania Water Co., Hays 1
e = Ohio River at West View.
Dchloromethan
2-trichloroet
lloromethane
4ine Plant.
O P4 fD
•
O
to
ff
0
H
O
rt
P
0
l-t
f
00
to
to
o
o
to
o
to
to
-
-
Evansville
h- >
-
0
O
CO
O
Co
Louisville
to
M
•P-
M
O
to
o
-
H1
-
-
Cincinnati
5
-
to
o
o
H-l
Huntington
•P**
\£)
N5
CO
O
to
0
a-.
-
-
Wheeling
oo
o
o
Beaver Falls
10
VD
H
H>
O
to
O
to
fD
CO
rt
H-
fD
fD
M
O
O
WPW/Hays Mined
GO
O
O
Pittsburgh
p
to
to
o
o
Wilkinsburg
to
M
O
O
CO
o
CO
o
X
n
P
fD
M
H1
o
H1
G
rt
H-
H1
H
rt
n
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
o
o
o E<
n r1
S o
- H
w
£> O
TOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
PPROX.IMATE LOWER DETECTION LEVEL =0.1 ug/L
H
15
-------
TABLE 39. DIBROMOCHLOROMETHANEa'b 'FINISHED WATER DATA, JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Utilityc
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mine
Seaver Falls
Wheeling
Hun ting ton
Cincinnati
Louisville
Evansville
Total or Mean
West Viewg
S-J
0
L searched
,
CO
33
en
CU
B
•H
H
12
12
12
11
27
11
24
19
21
12
161
9
60
TJ 3
0 •
IH O
, — |
cfl
33
CO
cu
B
•H
H
10
12
12
11
27
11
24
19
21
12
159
7h
rf
OD
T3 3
§rH
0 •
in O
rHV
Cfl
33
en
CU
0
•H
H
2
0
0
0
0
0
0
0
0
0
2
0
c
o
4J
concentra
£0.1 ug/L
i
rH P
CO CU
P
CO
cu
0.9
1.0
6.5
4.0
4.6
4.6
9.4
11
7.2
6.7
5.6
0.3*
0-0
3
all
ntration,
at cu
ximum
cone
g
3.0
2.7
16
15
13
19
25
26
33
24
33
0.4h
o
conf irmati
pted when
£0.1 ug/L
B
CQ CU rH
S 4-1 rH
4J CO
co co 33
cu
H
1
2
1
1
5
cu
S
confirmed
>0.1 ug/L
«3
cu
H
1
2
1
1
5
c
o
conf irmati
pted when
<0.1 UR/L
s
C/3 CU rH
4-1 Ct
cn cfl 33
cu
H
c
CU
3
confirmed
<0.1 ug/L
C/3 rH
CO 33
CU
|
H
0 CU
H 4J
4J C3 U
0 CU CU
U 3 cu
H id
2 01 4-i
O 4J O
0 CX C
B
en cu II
S 4-1
4-1 rH
CO CO rH
CU CO
B 33
•H
•
c
0)
rC T3
3 cu
confirmed
not detect
C/1 II
rH
CO rH
CU Cfl
B 33
•H
H
a = Tabled GC/Hall data represents dibromochloromethane and/or
cis-1,3-dichloropropene and/or 1,1,2-trichloroethane unless noted.
b = Tabled GC/MS data represents dibromochloromethane only.
c = See Figure 1.
d = Western Pennsylvania Water Co., Hays Mine Plant.
e = One time GC/MS confirmation for cis-1,3-dichloropropene proved negative.
f = One time GC/MS confirmation for 1,1,2-trichloroethane proved negative.
g = Ground water supply.
h = Does not represent one time GC/Hall report of cis-1,3-dichloropropene
and/or 1,1,2-trichloroethane at 0.5 ug/L.
127
-------
TABLE 40. FINISHED WATER3 DIBROMOCHLOROMETHANE LEVELS 1977-1978
GC/HALL DETECTOR
Utility
Fox Chapel
Wilkinsburg3
Pittsburgh3
WPW/Hays Mineb
Beaver Falls
Wheeling
Hun ting ton
Cincinnati
Louisville
Evansville
West View11
Jul
1.5
2.7
9.2+
3.1
—
19
25
26+
23f
16
0.3
Mean Concentration, ug/L
Aug
3.0
2.3
16
15
13
—
20
24
20§
24
ND
Sep
1.6
2.0
11
7.9
4.4
11
14
17e
—
7.7
—
Oct
0.6
1.2
8.6
5.1
5.5
5.2
6.8
13e
5.4*
6.4
ND
Nov
0.2
0.3
4.7
2.8
4.4f
3.0
6.0
7.3
3.2
5.5
0.2
Dec
<0.1
0.6
3.3
2.0
2.6*
0.4
2.5
4.9
1.3
1.3
—
Jan
0.1
0.4
0.2
0.6
2.7
0.4
—
5.5
3.2
1.6
0.4
Feb
0.2
0.4
0.2
0.9
4.1C
0.4
2.8
0.8
4.0
3.0
0.2
Mar
<0.1
0.4
0.2
0.6
0.6d
0.5
2.3
3.3
2.2
4.8
0.2
Apr
0.2
0.1
7.2
2.7
0.8
1.7
7.6
6.2
4.5
2.3
—
May
0.3
0.8
11
3.8
—
3.7
3.7
6.6
3.9
2.3
0.3
Jun
1.4
1.5
6.6
—
7.9
5.2
13
13
8.7
5.3
0.3
Annual
Mean
0.8
1.0
6.5
4.0
4.6
4.6
9.4
11
7.2
-6.7
0.2 '
00
a = Clear well sample
b = Western Pennsylvania Water Co./Hays Mine Plant
c = February 1-15
d = February 21-March 31
e = Normal operation only. Not representative of treatment modification reported in Section 6.
f = MS confirmed in one sample. Others not MS attempted.
g = Treatment modification reported in Section 6.
h = Ground water supply
+ = MS confirmed
—No data available
ND = not detected
-------
TABLE 41. BROMOFORM RAW WATER DATA, JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Utility3
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mine
West View0
Beaver Falls
Wheeling
Huntington
Cincinnati
Louisville
Evansville
Total or Mean
West View
>-i
o
14-4
T>
Ol
42
O
H
cd
0)
CO
rH
rH
CO
«
CO
0)
H
11
12
11
8
11
29
8
49
17
22
11
189
11
*J
^v^
too
TJ 3
§ .H
O •
<4H O
rn*\
rH
Cfl
EC
CO
0)
.§
H
0
0
1
0
0
0
0
0
0
0
0
1
0
^
•""^
toO
"0 3
C
3 rH
o -
IH O
rH V
rH
cfl
re
CO
0)
H
1
0
1
0
0
0
0
0
4
0
0
6
1
c
o
•H
4-J
cfl i-3
^ -•»
4-1 oo
C 3
Ol
0 rH
c •
0 0
UA\
rH
•-t C
cfl 0)
K -C
&
c
5
0)
sa
0.1
0.1
rJ
CO)
3
#*
C
O
•H
4J
CO
rH M
rH -U
fd W
O CX
C/3 5 rH
g ^ -H
4-1 Cfl
w cfl ffi
0)
H
H
C
a)
J3
5
iJ
TJ ~~.
01 00
e 3
•H rH
M-l
c o
sv
W rH
5C I — 1
cfl
co ec
0)
.§
H
c -a
O CU
•H -U
•U C O
cfl 0) 01
B ^ 4-1
W & 0)
•H 13
IH TJ
C 0) 4-1
O 4J O
o a c
CO Q) II
S w
4-1 rH
CO CO rH
QJ Cfl
•S ^
H
2
2
c
0)
J3 13
& 0)
4-J
T3 U
OJ 0)
€4-1
ai
•H T3
IH
C 4-1
o o
U C
CO II
^
rH
CO rH
0) Cfl
e«
H
2
2
a = see Figure 1
b = Western Pennsylvania Water Co.
c = Ohio River at West View
d = Ground water supply
Hays Mine Plant
129
-------
a = Clear well
b = see Figure
c = Western Pen
d = Ground wate
i-i 3 M (D
CO Hi
CO VJ Ml
-§ IT £*
TJ 03 (D
1-3 3
^ H* rt
rt
fD
i-j
n
o
.
w
ffi
CO
S
H-
3
fD
•D
M
3
rt
fD
CO
rt
H-
(D
CL
VO
-
NJ
O
O
H
H
O
rt
03
O
i-i
3
03
3
£
VO
0
NJ
M
O
00
-P-
J>
u,
•*
-
o
Evansville
K
co
-
o
Co
O
oo
Louisville
NJ
00
-
o
Ui
NJ
^
H
^
Cincinnati
VO
i-1
00
0
M
O
NJ
^
-
-
Hunt ing ton
NJ
-P-
N3
I-1
-
HU
H
JN
**
Wheeling
P
-
M
O
•P-
O
VO
Beaver Falls
NJ
-vl
00
VO
o
CO
0
ON
NJ
NJ
-
O
IWPW/Hays Mine°
P
-
NJ
M
O
co
H-
Pittsburgh
NJ
-
NJ
I--
O
co
oo
H
O
Wilkinsburg
i-1
NJ
O
M
o
X
n
3-
fD
h-i
NJ
M
O
O
NJ
0
NJ
G
rt
H-
I—1
H-
rt
cr
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 UR/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
n
r^
«•
1
H
W
t-1
9
w
o
H
m
n
H
0
f
M
<
II
C
^
I
w
-------
TABLE 43. FINISHED WATER BROMOFORM LEVELS 1977-1978
GC/HALL DETECTOR
Utility
Fox Chapel
Wilkinsburg3
Pittsburgh3
WPW/Hays Mineb
Beaver Falls
Wheeling
Hun ting ton
Cincinnati
Louisville
Evansville
West Viewh
Jul
ND
ND
0.2~
0.3
—
0.9
2.0
2.1+
1.1
0.4
ND
Aug
ND
ND
0.7
3.1
0.4
—
1.3
1.2
0.9§
0.8
ND
Sep
ND
ND
0.4
2.6
<0.1
0.6
1.4
1.6e
—
<0.1
—
I
Oct
ND
ND
0.3
0.8
0.3
0.2
0.5
0.8e
0.28
<0.1
ND
lean Cc
Nov
ND
ND
<0.1
0.3
o.if
0.1
0.6
0.2
ND
<0.1
ND
mcenti
Dec
ND
ND
<0.1
<0.1
-------
H
U)
*Quant if icat ion
c =
d =
J^
H-
O
O4
P3
3
to
Ohio River
a - see Figure !
b = Western Penr
fU
PS
rt
(D
H
Q
•
w
PC
CO
S
H-
3
tt>
I—1
(a
a
rt
s:
n>
co
rt
H-
h-1
h-1
O
o
H
O
rt
O
H
oo
o
-
Evansville
P
o
0
Louisville
N>
O
O
Cincinnati
--j
0
0
Huntington
VO
0
0
Wheeling
00
o
M
Beaver Falls
S
0
0
s:
to
co
rt
H-
«0
p
o
o
WPW/Hays Mineb
oo
o
o
Pittsburgh
P
0
0
Wilkinsburg
M
NJ
O
0
o
o
(D
M
P
0
0
c
rt
H-
rt
CU
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
H
O tt
o t-1 -P-
n t-1 •
>|l
x" H
i ^l
^_ K*N ^^
» wg
o r1 ^
W O g-j
2 w ti
^ ^ [t*
~* ri j>
PS H Jr^
< M
W 0 (_,
C-1 2 ^
II f1 E<
M
PI VD
»-• t-1 -S
C II ]
» C-l
C M
OQ vo
r1 oo
-------
c II
P)
BS
H- O
Hi C
H- 13
O Qj
Pi
rt 5;
H. m
2 <*•
0 ro
M *
P.g
£•§
rr "o
n
n
it
o
p4
|
rt
3
ro
s; CD n
ro ro M
CD ro ro
rt P)
ro 'TI H
SH-
OP s!
e ro
ro ro M
3
3 M ro
CD Hi
EJ
H-
CO
C?
n
o
33
Co
3
ro
ro
rt
«
ro
CO
rt
H-
ro
*
o
Cn
-
I-1
rt
P>
H"
0
ft
S
P>'
3
H
h-1
CO
10
-p-
o
10
M
O
-P-
-
to
M
Evansville
H
to
O
O
Louisville
to
-
CO
o
•C-
M
0
Cincinnati
M
Co
10
O
M
0
M
-
H
M
O
Hunt ing ton
to
P
so
o
10
o
-p-
-
-
Wheeling
H1
-
Co
O
Co
M
O
to
to
Beaver Falls
5
H
Ui
O
to
o
to
WPW/Hays Mine°
^
M
OS
O
to
0
Cn
Pittsburgh
i-1
to
to
Co
0
Co
O
Ov
-
H
Wilkinsburg
to
M
H
O
H-
O
H
O
n
T)
ro
h-1
to
M
10
O
to
0
to
rt
H-
1— '
H-
rt
cr
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 UE/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
o
>
o
M
H
M
f
O
&
m
H
tr1
<
n
o
c
-------
TABLE 46. FINISHED WATER3 DICHLOROIODOMETHANE LEVELS 1977-1978
GC/HALL DETECTOR
Utility
Fox Chapel
Wilkinsburga
Pittsburgh3
WPW/Hays Mineb
Beaver Falls
Wheeling
Hun ting ton
Cincinnati
Louisville
Evansville
West Viewh
Mean Concentration, ug/L
Jul
ND
ND
0.1
<0.1
—
0.2
<0.1
ND
<0.1
ND
ND
Aug
0.1
ND
0.6
0.5
0.2
—
0.2
ND
0.38
ND
<0.1
Sep
ND
<0.1
ND
ND
ND
<0.1
0.1
NDe
—
ND
—
Oct
ND
ND
ND
0.2
ND
0.1+
0.2
<0.l|
0.18
ND
ND
Nov
ND
ND
ND
<0.1
ND
0.1
<0.1
ND
<0.1
ND
<0.1
Dec
<0.1
ND
<0.1
ND
<0.1
<0.1
<0.1
ND
<0.1
ND
—
Jan
ND
ND
ND
<0.1
<0.1
<0.1
—
ND
<0.1
ND
<0.1
Feb
<0.1
ND
ND
<0.1
NDC
ND
<0.1
ND
<0.1
ND
<0.1+
Mar
ND
ND
ND
ND
<0.1d
0.1+
ND
ND
<0.1
ND
ND
Apr
ND
ND
<0.1+
<0.1
ND
0.4
0.2
ND
<0.1
ND
—
May
ND
ND
<0.1
0.1
—
0.4
0.1
ND
<0.1
ND
ND
Jun
ND
0.1
ND
—
ND
1.0
0.3+
ND
0.1
ND
<0.1
Annual
Mean
<0.1
<0.1
<0.1
<0.1
<0.1
0.2
0.1
<0.1
<0.1
ND
<0.1
a = Clear well sample
b = Western Pennsylvania Water Co./Hays Mine Plant
c - February 1-15
d = February 21-March 31
e = Normal operation only. Not representative of treatment modification reported in Section 6.
f = MS confirmed in one sample. Others not MS attempted.
g = Treatment modification reported in Section 6.
h = Ground water supply
+ = MS confirmed
—No data available
ND = not detected
-------
TABLE 47. FINISHED WATER TOTAL TRIHALOMETHANE LEVELS. 1977-1978. GC/HALL DETECTOR
Utility '
Q
Wilkinsburg
Fox Chapel
Pittsburgh3
WPW/Hays Mineb
Beaver Falls
Wheeling
Hunt ing ton
Cincinnati
Louisville
Evansville
West Viewg
Mean Concentration, ug/L
Jul
17
9
63
16
—
173
94
109
129
84
1
Aug
27
44
77
56
83
—
129
116
149f
161
1
Sep
32
39
101
29
91
132
106
llle
--
121
—
Oct
21
15
62
38
60
128
72
106e
94f
121
2
Nov
8
3
61
19
71
83
44
42
100
101
2
Dec
6
2
52
25
54
54
31
38
61
72
—
Jan
4
2
5
12
52
27
—
45
40
47
—
Feb
3
2
2
10
47C
19
24
13
49
27
2
Mar
4
3
6
18
iod
42
28
30
57
57
—
Apr
3
4
56
64
12
48
62
45
23
73
—
May
9
8
45
35
—
61
68
49
51
87
1
Jun
11
12
18
—
126
82
98
89
87
113
2
Annual
Mean
13
12
46
29
60
77
69
66
76
89
2
, Clear well sample
Western Pennsylvania Water Co./Hays Mine Plant
.February 1-15
February 21-March 31
^Normal operation only,, Not representative of treatment modification reported in Section 6.
Treatment modification reported in Section 6.
Ground water supply
—No data available.
-------
TABLE 48. TRIHALOMETHANE FORMATION POTENTIAL (THMFP) - GC/HALL DETECTOR
HUNTINGTON WATER CORPORATION 1977-1978
U)
Month3
Juld
Augd
Sepd
Octd
Nov
Decd
Jan
Feb
Mar
Apr
May
Jun
Storageb
Temp, °C
room
room
room
room
9
4
3
3
4
6
2
21
pH
—
8.3
8.3
8.4
8.4
8.4
8.4
8.3
7.9
8.0
8.3
Mean Concentration, ug/L
Rawc
THMFP
A
326
355
219
—
140
225
90
79
150
180
220
350
Clear Well
inst
TTHM
B
112
89
106
63
44
31
—
24
28
62
68
98
term
TTHM
C
232
286
202
215
130
91
—
56
81
90
150
300
THMFP
D=C-B
120
197
96
152
86
60
—
32
53
28
82
202
mean
Fate of Raw Water THMFP
% that
formed
inst TTHM
B/A
34
25
48
—
32
14
—
30
19
34
31
28
30
% removed
by
treatment
(A-C)/A
29
19
7.7
—
7.1
59
—
29
46
50
32
14
29
%
remaining
D/A
37
56
44
—
61
27
—
41
35
16
37
58
41
»3
,one sample day per month.
15 mg/1 chlorine added. 7-day storage.
draw water inst TTHM^l ug/L.
mean of two to four sample days per month.
—data not available.
-------
TABLE 49. TRIHALOMETHANE FORMATION POTENTIAL (THMFP) - GC/HALL DETECTOR
FOX CHAPEL AUTHORITY 1977-1978
a
Month
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Storage
Temp, °C
room
room
room
room
4
9
10
1
13
18
26
20
pH
8.0
7.8
7.7
7.8
7.8
7.8
7.8
7.8
7.8
7.8
7.8
Mean Concentration, ug/L
Rawc
THMFP
A
133
265
285
282
119
92
76
71
176
130
226
193
Clear Well
inst
TTHM
B
8.7
44
38
15
3.3
1.6
2.1
2.4
3.4
4.3
8.1
12
term
TTHM
C
86
158
182
136
62
50
32
40
63
64
80
103
THMFP
D=C-B
77
114
144
121
59
48
30
38
60
59
72
91
mean
Fate of Raw Water THMFP
% that
formed
inst TTHM
B/A
6.5
17
13
5.3
2.8
1.7
2.6
3.4
1.9
3.3
3.6
6.2
5.6
% removed
by
treatment
(A-C) /A
35
40
36
51
48
46
58
44
64
51
65
47
49
7
to
remaining
D/A
58
43
50
43
50
52
39
54
34
39
32
47
45
GJ
, one sample day per month.
15 mg/1 chlorine added. 7-day storage.
°raw water inst TTHM^l ug/L.
—data not available.
-------
TABLE 50. TRIHALOMETHANE FORMATION POTENTIAL (THMFP) - GC/HALL DETECTOR
WILKINSBURG-PENN JOINT WATER AUTHORITY 1977-1978
Month a
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Storage1*
Temp, °C
room
room
room
room
6
4
2
4
4
10
16
24
pH
—
8.0
8.0
8.0
8.0
8.0
8.1
8.1
7.8
8.1
8.1
8.1
Mean Concentration, ug/L
Rawc
THMFP
A
—
160
303
216
171
134
—
76
124
126
170
255
Clear Well
inst
TTHM
B
17
37
32
21
7.6
5.8
3.5
2.8
4.2
2.8
9.3
11
term
TTHM
C
110
120
165
172
97
68
70
42
68
78
99
195
THMFP
D=C-B
93
83
133
151
90
62
66
39
64
75
90
184
mean
Fate of Raw Water THMFP
% that
formed
inst TTHM
B/A
— •
23
10
10
4.4
4.3
—
3.7
3.4
2.2
5.5
4.3
7.1
% removed
by
treatment
(A-C) /A
—
25
46
20
43
49
—
45
45
38
42
24
38
%
remaining
D/A
—
52
44
70
53
46
51
52
60
53
72
55
00
.one sample day per month.
15 mg/1 chlorine added. 7-day storage.
raw water inst TTHM^l ug/L.
—data not available.
-------
TABLE 51. TRIHALOMETHANE FORMATION POTENTIAL (THMFP) - GC/HALL DETECTOR
PITTSBURGH DEPARTMENT OF WATER 1977-1978
Month3
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Storage b
Temp, °C
room
room
room
room
11
7
12
7
6
12
14
22
PH
7.0
7.8
8.0
8.6
8.5
8.6
8.6
8.7
8.4
8.4
8.4
8.5
Mean Concentration, ug/L
Raw0
THMFP
A
482
344
207
£.279
219
294
158
118
282
209
196
260
Clear Well
inst
TTHM
B
63
77
101
62
62
52
6.1
2.5
5.4
56
45
18
term
TTHM
C
271
235
—
197
109
136
102
89
107
136
181
132
THMFP
D=C-B
208
158
—
135
47
84
97
86
102
80
135
114
mean
Fate of Raw Water THMFP
% that
formed
inst TTHM
B/A
13
22
49
22
28
18
3.9
2.1
1.9
27
23
6.9
15
% removed
by
treatment
(A-C) /A
44
32
—
29
50
54
35
24
62
35
7.6
49
38
7
/o
remaining
D/A
43
45
—
48
21
28
61
73
36
38
69
44
46
VD
, one sample day per month.
15 mg/1 chlorine added. 7-day storage.
°raw water inst TTHM^l ug/L.
— data not available.
-------
TABLE 52. TRIHALOMETHANE FORMATION POTENTIAL (THMFP) - GC/HALL DETECTOR
WESTERN PENNSYLVANIA WATER COMPANY (HAYS MINE PLANT) 1977-1978
Month3
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jund
Stora£eb
Temp, °C
room
room
room
room
7
7
2
3
7
14
15
—
PH
—
7.7
7.5
7.6
7.8
7.8
8.0
8.0
7.4
7.4
7.4
—
Mean Concentration, ug/L
RawL
THMFP
A
191
264
289
—
98
77
127
79
£189
83
—
—
Clear Well
inst
TTHM
B
16
55
29
39
19
26
12
9.6
18
64
35
—
term
TTHM
C
—
122
153
—
98
—
74
63
58
73
82
—
THMFP
D=C-B
—
67
124
—
79
—
62
53
41
9
47
—
mean
Fate of Raw Water THMFP
% that
formed
inst TTHM
B/A
8.4
21
10
—
19
34
9.4
12
9.5
78
—
—
23
% removed
by
treatment
(A-O/A
—
54
47
—
0
—
42
20
69
12
—
—
35
%
remaining
D/A
—
25
43
—
81
—
49
• 67
22 .
11
—
—
42
.one sample day per month.
15 mg/1 chlorine added. 7-day storage.
,raw water inst TTHM^l ug/L.
no samples collected.
—data not available.
-------
TABLE 53. TRIHALOMETHANE FORMATION POTENTIAL (THMFP) - GC/HALL DETECTOR
BEAVER FALLS MUNICIPAL AUTHORITY 1977-1978
Month3
Jul
Aug
Sep
Oct
Novd
Decd
Jan
Feb l-15d
Feb 21- d,e
Mar 31
Apr
May
Jun
Storage
Temp, °C
room
room
room
room
10
4
2
22
4
11
16
21
pH
7.5
7.3
7.2
7.4
7.4
7.5
7.2
7.3
7.2
7.4
7.5
Mean Concentration, ug/L
Rawc
THMFP
A
143
180
383
£245
268
175
143
119
151
138
—
189
Clear Well
inst
TTHM
B
— .
83
84
59
70
54
54
47
10
11
—
126
term
TTHM
C
—
150
178
183
226
112
101
94
50
80
—
136
THMFP
D=C-B
—
67
94
124
156
58
47
47
40
69
—
10
mean
Fate of Raw Water THMFP
% that
formed
inst TTHM
B/A
—
46
22
24
26
31
38
39
6.6
8.0
—
67
31
% removed
by
treatment
(A-C) /A
—
17
53
25
16
36
29
21
67
42
—
28
33
%
remaining
D/A
—
37
25
51
58
33
33
40
27
50
—
5.3
36
, one sample day per month.
15 mg/1 chlorine added. 7-day storage.
^raw water inst TTHM^l ug/L.
mean of two to four sample days per month.
no breakpoint chlorination.
—data not available.
-------
TABLE 54. TRIHALOMETHANE FORMATION POTENTIAL (THMFP) - GC/HALL DETECTOR
WHEELING WATER DEPARTMENT 1977-1978'
Month3
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Storageb
Temp, °C
room
room
room
room
7
4
6
7
4
9
20
23
PH
—
9.4
9.1
9.4
9.3
9.3
9.2
9.3
9.3
9.3
9.2
9.1
Mean Concentration, ug/L
Rawc
THMFP
A
247
323
—
260
232
240
—
125
154
98
676
342
Clear Well
inst
TTHM
B
173
—
132
127
83 ^
54
27
19
42
48
61
81
term
TTHM
C
—
>391
157
225
115
132
89
93
107
75
—
275
THMFP
D=C-B
—
—
25
98
32
78
62
75
65
27
—
194
mean
Fate of Raw Water THMFP
% that
formed
inst TTHM
B/A
70
—
—
49
36
22
—
14
27
49
9.0
24
32
% removed
by
treatment
(A-O/A
—
-21
—
13
50
45
—
26
30
23
—
20
30
%
remaining
D/A
—
—
38
14
33
—
60
42
28
—
57
39
Q
.one sample day per month.
15 mg/1 chlorine added. 7-day storage.
raw water inst TTHM^l ug/L.
— data not available.
-------
TABLE 55. TRIHALOMETHANE FORMATION POTENTIAL (THMFP) - GC/HALL DETECTOR
CINCINNATI WATER WORKS 197^-1978
Month a
Jul
Aug
Sep
Octd
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Storage
Temp, °C
room
room
25
18
17
7
7
2
4
16
18
24
PH
8.2
8.4
8.3
8.2
8.3
8.2
8.3
8.4
8.5
8.5
8.1
Mean Concentration, ug/L
Rawc
THMFP
A
—
202
£305
£508
£230
£321
194
125
266
—
373
£379
Clear Well
inst
TTHM
B
109
116
111
106
42
38
45
13
30
45
49
89
term
TTHM
C
287
121
165
338
119
98
89
61
83
115
133
243
THMFP
D=C-B
178
5
54
232
77
60
45
48
54
70
84
154
mean
Fate of Raw Water THMFP
% that
formed
inst TTHM
B/A
—
57
36
21
18
12
23
10
11
—
13
23
22
% removed
by
treatment
(A-O/A
—
40
46
33
48
69
54
51
69
—
64
36
51
7
/o
remaining
D/A
—
2.5
18
46
33
19
23
38
20
—
22
41
26
-t-
OJ
, one sample day per month.
15 mg/1 chlorine added. 7-day storage.
^raw water inst TTHM^l ug/L.
mean of two to four sample days per month.
—data not available.
-------
TABLE 56. TRIHALOMETHANE FORMATION POTENTIAL (THMFP) - GC/HALL DETECTOR
LOUISVILLE WATER COMPANY 1977-1978
Month3
Jule
Auge
Sep
Octe
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
S tor age b
Temp, °C
room
room
—
room
room
11
6
4
8
15
22
20
PH
—
—
—
—
8.3
8.4
8.4
8.4
8.4
8.3
8.3
8.1
Mean Concentration, ug/L
Raw0
THMFP
A
339
315
—
325
252
245
91
80
185
240
192
269
Clear Well
Inst
TTHM
B
129
149
—
94
100
61
40
49
57
23
51
87
term
TTHM
c
316
245
—
244
160
112
74
78
120
100
148
192
THMFP
D=C-B
187
96
—
150
59
51
34
28
63
77
97
105
mean
Fate of Raw Water THMFP
% that
formed
inst TTHM
B/A
38
47
—
28
40
25
44
62
31
10
26
32
35
% removed
by
treatment
(A-C) /A
6.8
22
__
25
36
54
19
2.5
35
58
23
29
28
%
remaining
D/A
55
30
__
46
23
21
37
35
34
32
50
39
36
a
, one sample day per month.
15 mg/1 chlorine added. 7-day storage.
,raw water inst TTHM<1 ug/L.
no samples collected.
mean of two to four sample days per month.
—data not available.
-------
TABLE 57. TRIHALOMETHANE FORMATION POTENTIAL (THMFP) - GC/HALL DETECTOR
EVANSVILLE WATER DEPARTMENT 1977-1978
a
Month
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Storage*5
Temp, °C
room
room
room
room
12
4
<1
1
6
15
16
26
r PH
—
7.8
7.9
8.2
8.1
8.1
7.8
8.1
8.1
8.3
8.0
8.1
Mean Concentration, ug/L
Raw0
THMFP
A
285
324
437
£573
266
248
208
113
173
213
331
379
Clear Well
inst
TTHM
B
84
161
121
121
100
71
46
26
57
73
88
113
term
TTHM
C
259
218
308
—
147
112
82
—
84
225
152
259
THMFP
D=C-B
175
57
187
—
47
41
36
—
27
152
64
146
mean
Fate of Raw Water THMFP
% that
formed
inst TTHM
B/A
29
35
28
21
38
29
22
23
33
34
26
30
30
% removed
by
treatment
(A-C)/A
9.1
33
29
45
55
61
—
51
- 5.6
54
32
36
7
/o
remaining
D/A
61
18
43
—
18
16
17
—
16
71
19
38
32
L/l
.one sample day per month.
15 mg/1 chlorine added. 7-day storage.
raw water inst TTHM ^1 ug/L.
—data not available.
-------
IL
H
tt
UJ
,, <
350-
300-
_ 250-
2
£ o
(V <
2 »- 2OO
u in
2
8
»5O-
IOO -
25-
-80
15 -
10--
B-
-V V- MEAN RAW WATEJ^ TEMP
UJ
a
.3
5
70 (^
til
0.
PA
fcO W I
O I
^ '
* I
UJ
-40
\ T
32
JUL AUG SEP OCT MOV DEC JAN FEB MAP APR MAY JUM
wn
Figure 44. Raw water THMFP variation (mean of project surface waters).
-------
a = see Figure 1
b = Western Pennsylvania Water Co., Hays Mine Plant
c = Ohio River at West View
d = Ground water supply
s;
re
co
rt
H-
re
i-1
0
0
H
O
rt
t-1
O
if
8?
B
oo
oo
to
Cn
K
o
o
o^
^o
VO
-
H
CO
0
Evansville
H
H
O
0
H1
O
h-1
-
H
Louisville
Ni
CO
Co
o
o
h-1
-
IS3
n
3"
o
H-
6)
rt
H-
5
^
t-o
O
0
N3
CO
CO
-
0
Huntington
VO
5
O^
o
o
Co
CO
Wheeling
oo
o
o
Beaver Falls
N3
0
0
M
O
s:
re
CO
rt
H-
M
O
0
WPW/Hays Mineb
^
o
o
Pittsburgh
P
o
M
M
-
Wilkinsburg
M
o
o
o
X
o
13
H"
P
O
O
C
rt
H-
I-1
H-
rt
P
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
GC/HALL DETECTOR, APPROXIMATE LOWE
GC/MS, APPROXIMATE LOWER PETE
R DETECTION LEVEL = 0.1 ugA-
CTION LEVEL =0.1 ugA,
TABLE 58
C
ON TETRACHLORIDE RAW
LATER' DATA
JULY
-vo
~J
~J
I
CH
W
VO
•^J
00
-------
TABLE 59. CARBON TETRACHLORIDE FINISHED3 WATER DATA, JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL = 0.1 ug/L
Utility
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mine°
Beaver Falls
Wheeling
Hunt ing ton
Cincinnati
Louisville
Evansville
Total or Mean
West View6
t-i
o
4-1
0)
X!
O
J-4
CO
cu
CO
rH
CO
S3
co
cu
E
H
12
12
12
11
27
11
22
19
21
12
L59
9
60
T3 3
C
3 >H
O
4-4 O
/*\
r •(
cO
S3
CO
cu
B
H
0
0
0
0
6
I
21
3
k
4
39
0
60
rO ^
fz3
3 rH
O
4-1 O
v
r1 1
rH
CO
S3
CO
01
B
H
2
1
1
1
2
1
1
5
2
0
16
0
c
o
•H
CO ,_}
4-J 60
C 3
CU
a iH
c •
0 0
u A\
rH
•H C
UH rd
C
CO
S
0.1
0.1
0.5
0.2
0.4f
0.2d
0.4
rJ~
1 — -^
60
3
c"
O
•H
4-1
CO
1 — 1 >-l
rH 4-1
CO C
S3 CU
o
B C
3 0
•S °
r*S
s
0.2
0.1
1.8
0.3
1.3
6.0
6.0
1
c
o
4-> C <
CO CU 00
B rC 3
^ 15
•H rH
C 0) O
O 4-J sA
B
CO 0) rH
S 4-1 rH
4-1 CO
W cfl S3
0)
§
•H
H
1
1
2
c
0)
X!
s
0) 60
E 3
•H rH
C 0
O y&
O 'x
S rH1
CO
to S3
0)
B
H
1
1
2
c
o
•H rj
4J c <:
g cu t>
j-< ^
-H i-
4-1 TD
C 01 O
O 4-1 \/
o a.
rH
5^ 4_) |—
4-J CI.
CD CO S3
0)
E
•H
H
1
1
1
1
1
5
C3
01
£j
-a -^
CU 60
E 3
•H r-t
4-1 .
C3 O
o \/
QJ
CO rH
S rH
Cfl
co S3
cu
B
•H
H
1
1
1
0
1
4
C3 rO
o cu
4-1 C U
CO 01 CU
>-< S CU
•H T3
U-< T3
q cu 4-1
O 4-> O
o a. c
co cu u
S 4-1
CO CO rH
CU CO
B S3
•H
H
p
CU
x; -n
S cu
T3 O
01 CU
B 4-1
}-i Ol
•H 13
4-1
C 4-1
O O
o c
CO ||
J 1
CO rH
0) CO
B S3
•H
H
a = Clear well effluent
b = see Figure 1
c = Western Pennsylvania Water Co., Hays Mine Plant
d = Excluding the maximum concentration of 6 ug/L
e = Ground water supply
f = Excluding the maximum concentration of 1.3 ug/L
148
-------
-C-
vo
n> CL
O 2!
H O
O rt
cu 3
n
3 M
S C
rt &
(D H-
i-i 3
OQ
CO
C O
cr to
II II
O S3 co
3* fl> (D
H- CO (D
O rt
(t> ^
H- 3 OQ
TJ H
I-! (D (D
CO 3 I-1
rt co
O
H
O
cr
(D
3
N
§
tc
fl> <
CO Co
rt 3
H-
< CO
H-
g S3
rt
fD
o
o
S
H-
3
S3
CD
CO
rt
H-
h-1
*
-
£
(jO
M
-
-
O
H
O
rt
CO
O
?
oo
*•
OJ
o
H
O
S3
1-1
-
M
!
Evansville
p
o
o
1 J
jLouisville
N)
M
O
O
o
M
h-1
O
i
Cincinnati
5
o
M
I
(Hunt ing ton
I
f-
LO
M
O
00
M
0
I-1
M
M
I-1
i
(wheeling
00
o
o
I
'Beaver Falls
ho
VO
O
O
S3
(D
CO
rt
H-
«„
P
0
o
WPW/Hays Mineb
-4
O
0
1
iPittsburgh
L
p
o
o
Wilkinsburg
to
o
0
n
T)
fi>
o
o
o
d
rt
H-
t-1
H-
rt
Co
Times Hall searched for
Times Hall found
>0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
Q
° >
££
O t-1 M
n f
55 M •
" H
W
25 Hffi
"3 o P
§.»§
x o
i_g £Q N
M O ^
(^ . |
o t-1 g
w o S
H s: t3
w w H
c~j pd ^
H ,_,
M G g
o w s
2 H g
_ w >
tr1 n
< M ^
^i§
11 ^s
p < Zj
_ „ t_i
c n a
» g
~-- o g
tr1 '
OQ oo
f
-------
a = Clear well effluent
b = see Figure 1
c = Western Pennsylvania Water Co., Hays Mine Plant
d = Not including chlorobenzene spill data.
e = Ground water supply
f = Field replicates: 2.8 ug/L and MS confirmed;
not detected by GC/Hall and MS negative.
sT~
fD
CO
rt
<
P-
ro
€
CD
^D
VO
0
I-1
•
VD
N)
•
VO
-P-
OJ
(->
Ml
0
H
O
rt
(u
M
O
1-1
C?
to
3
M
C^
0
-P-
W
O
(jO
0
-P-
M
I-1
H
!-•
i Evansville
H
to
O
O
f
o
H-
5
H-
M
M
CD
CL
to
M
O
M
j Cincinnati
M
U3
l_l
M
O
•
K3
O
•
to
M
M
| Hunt ing ton
to
*-
U)
i-1
0
•
U)
0
•
-p-
M
H
Wheeling
M
M
0
O
1
Beaver Falls
S3
•^1
o
0
;WPW/Hays Mine°
M
h-"
O
0
i Pittsburgh
(-»
to
o
o
! Wilkinsburg
M
to
O
O
51
X
o
sT
•o
fD
M
H"
M
O
O
c
rt
H-
(-•
H-
rr
*<
a-
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 uE/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
^i
pr M
o^£
ri r* r
r? O
C/3 M f~.
- sp
532
IOBENZENE FINISHED3 WATER DATA, JULY 1977-JUNE 1978
OR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
PROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
-------
&* o
II II
o o
a. po
» ro
rt H
(!)
hi |k)
rt
CO
C £
"o ro
•a co
I-1 rt
§
a = see Figure 1
b = Western Pennsylvania Water Co., Hays Mine Plant
z:
CO
rt
H-
*
Cn
M
O
S3
O
-P-
-
-
-
-
H
O
I-1
O
n
1
Ui
Ui
ON
o
to
o
-
o
-
-
NJ
o
Evansville
o
o
o
JLouisville
ro
tO
o
o
o
5-
o
H-
01
rt
H-
ot
-
O
-
o
rt
H-
3
OQ
rt
O
3
CO
ro
-
O
to
o
Wheeling
^
o
o
Beaver Falls
Ui
M
O
O
o
ro
co
rt
H-
«„
CXI
o
o
iwPW/Hays Mineb
ON
O
o
Pittsburgh
h-1
O
H
CO
O
O
to
-
o
Wilkinsburg
o
o
H
M
-
-
O
X
n
cr
T)
ro
vo
O
H-
C
rt
H-
H'
rt
OJ
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ug/L
Times MS confirmed when
Hall >0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
TABLE 62. 1 , 1-DICHLOROETHANE RAW W
GC/HALL DETECTOR, APPROXIMATE LOWE
GC/MS, APPROXIMATE LOWER DETE
ATER DATA, JULY 1977-JUNE 1978
R DETECTION LEVEL =0.1 ug/L
CTION LEVEL =0.1 ug/L
-------
II
O
i-(
O
rt
(D
1-1
en
C
o cr
II II
S3 en
m rt>
en n>
rt
(D »il
if H-
3 00
^d PI
(D fl>
3
3 M
en
rt
ft)
n
o
as
to
H-
3
m
13
M
cu
3
Co
II
n
M
n>
cu
n>
o>
HI
HI
rr
3
rt
n
CO
rt
H-
m
Cu
VO
u,
0
p
LO
O
S3
ro
rt
O
f
3
-P-
•O
oo
u,
o
(jj
0
OJ
o
ISJ
o
Evansville
N3
o
o
Louisville
I-1
00
M
0
0
O
Cincinnati
VO
-
M
H1
0
Hunt ing ton
N3
M
-
K3
O
O
ls>
; Wheeling
H-1
0
0
O
i Beaver Falls
NJ
S3
N3
-
0
IS5
0
H
O
WPW/Hays Mine°
i
M
O
0
^Pittsburgh
h-1
O
I—1
O
O
o
I-1
o
| Wilkinsburg
M
O
NJ
O
O
O
ON
-
O
3
X
n
fu
13
ft)
M
O
-
-
0
rt
H-
I—1
H-
rt
cr
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ue/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
w
O M
n
o P
n t-1
"-«. H4
S U"
C/3 M M
- H 1
M 0
> 0 M
TJ H 0
V 0 EC
P0 PO t-1
° " 2
M |> O
3; 1-5 M
^> ^T^ ^"3
WOK
b< r^H
I-1 M M
OS..
S 3> ^
^ r-j j~j
po trd g
ppj ^ O
WATER
R DETEC
CTION L
W H 0
n r"
M C_,
° W ^
PT?^J
05 §
00
-------
Ln
LO
a- o
n n
o o
H CT
O H-
C 0
3
O- SO
P] fD
rt i-(
fD
i-t ti)
rt
CO
c z:
13 fD
T3 01
t-1 ft
'•<
a = see Figure
b = Western Pen
P t->
01
<
3
H-
P>
Pi
rt
fD
O
O
w
1
CO
H-
3
fD
I-1
3
rt
fD
CO
rt
H-
M
S3
H
O
Ui
O
oo
t->
o
H
0
rt
I-1
O
i-i
P?
oo
oo
M
t;
o
N3
O
O
oo
UJ
I-1
Evansville
P
H
H
O
0
-
-
Louisville
S3
Ln
M
O
O
<_n
u,
U)
Cincinnati
h->
N)
OJ
O
N3
0
*
M
-
-
O
Huntington
VO
*.
Ul
o
o
M
o
-
o
Wheeling
oo
o
o
Beaver Falls
N3
S3
U,
O
I—1
0
I-1
I—1
h-1
CO
ft
H-
fD
P
O
0
I k
jWPW/Hays Mine
-
o
0
!
j Pittsburgh
M
I-1
O
0
O
OS
-
-
jWilkinsburg
M
S3
M
M
O
N3
O
-
-
O
n
fD
I-1
h-1
O
I-1
I-1
I-1
a
rt
H-
H-1
H-
rt
(u
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall ^0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
GC/HALL E
GC/fr
>!
S §
li
tr1 M
O 2
S H
5O M
M W
n po
H
M O
O M
2 H
M H
m o
f Z
II tr1
m
o <
• w
M r1
3Q
h-1
00
W
fc
>
VD
^4
W
00
-------
Ui
O- O CT1 03
II II II II
O SI co n
H ID ID M
O CO ID ID
3 ID *T) i-t
p. l"t H-
3 OQ
m
rt ID ID
ID 3
CO
CO ^
C hr'-f
T( <
13 0>
M 3
*^ H*
DI
rt
ID
O
O
X
s
H-
3
ID
§
Si
ID
CO
rt
<
>-"•
§
a.
VO
M
U>
o
(J1
o
Ui
I-"
o
H
O
rt
P>
M
O
^
ff
So
3
M
O\
O
-P-
U)
o
•
H
O
N>
N>
KJ
M
O
Ui
O
Evansville
H
NJ
O
O
Louisville
N>
H
O
0
H
O
•
Cincinnati
M
v£>
O
H>
(S3
O
Hunt ing ton
1-0
-f>
O
O
M
O
Wheeling
i-1
M
O
O
Beaver Falls
ro
o>
N3
M
O
•
M
O
•
S3
M
M
M
O
WPW/Hays Mine°
i-1
(-»
o
o
Pittsburgh
M
S3
M
O
A
H1
A
M
M
H
Wilkinsburg
HI
t-o
M
M
O
•
I-"
O
•
M
M
O
"3
O
X
o
pi
13
n>
M
h->
NJ
O
o
a
rt
H-
h-"
H-
rt
<<
D4
Times Hall searched for
Times Hall found
£0.1 ug/L,
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
H
J
0? ^
S <-n
OF '
O f
K OJT
-MS N»
M JL
^n"
^ o H
?d f^ M
9* o
X Hj
M t> P;
)ETHANE FINISHED3 WATER DATA, JULY 1977-JUNE 1978
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
MATE LOWER DETECTION LEVEL =0.1 ug/L
-------
a- o
H ii
o o s; to
M 3* fD n>
o H- co n>
§
O rt
Ui
a to H H-
H- 3 TO
p) fD hi H
rt i-t fD (D
fD 3
rt 0)
to ^
c s: M
*O fD ^
T3 OJ 0)
t-" rt 3
«
-------
Oi
ex o cr
II II II
o s: co
•i ro ro
p to ro
§
fC he)
3 OQ
P3 hd i-j
rt ro ro
ro 3
•t 3 H-
co
co •
i-J
n
o
ffi
Cu
S
H-
ro
(U
3
a = Clear well effluent
CO
rt
H-
CL
VO
O
O
H
O
rt
O
S
(!)
(to
3
H
OH
u,
o
•
I-1
p
rO
M
-
M
Evansville
K
M
o
o
•
o
•
H
-
Louisville
ro
H
H
O
O
•
H
0
H
Cincinnati
HI
o
0
M
H
Huntington
rO
ro
-
0
•
O
Wheeling
H
0
H-
W
rt>
B)
ro
(13
CO
rO
H
O
o
M
O
ro
H
O
s:
&>
CO
S
ro
o
P
o
H-
Pittsburgh
M
O
H1
Wilkinsburg
ro
^
0
o
o
o
X
n
BJ
ro
K
o
M
rt
H-
H-
ft
cr
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall >0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
o w
1?
n f1
n f ,_,
s" at,
OT M ^
> n S
TJ H G
PO §^
M > 8
-------
&. o
II II
o o
O H-
« <
CO (D
r*f ^»
(D
H Co
rt
CO
C S3
*O (C
T3 CO
O- CO
II II
S3 co
(c n>
ca n>
rt
3 TO
CD
3
rt
S3
(D
CO
rt
H-
M
O
0
H
O
rt
Co
O
i-l
i?
B
oo
oo
o
o
Evansville
p
o
o
Louisville
NJ
M
O
0
'
Cincinnati
5
O
o
Huntington
-P-
o
o
Wheeling
00
o
o
w
n>
Co
ro
M
t-1
CO
NJ
O
O
S3
n>
CO
rt
H-
p
o
o
WPW/Hays Mine
-^j
o
o
Pittsburgh
h->
o
o
Wilkinsburg
M
S3
O
0
TJ
O
n
T!
(D
I-1
h-1
M
O
0
c
rt
H-
I-1
H-
rt
CO
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
TABLE 68. TRANS-1,3-DICHLOROPROPENE RAW WATER DATA, JULY
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =
nr./MS. APPROXIMATE LOWER DETECTION LEVEL = 0.1 v.
1-1 i
c ^
oo g
i-1
CO
-------
Ul
00
CL o u4 a>
o S3 co n
H fD fD
O CO fD
C rt
3 fD IT)
a. i-j H-
3 OQ
S! C
fD ._
n 3
co
CO **d
C I-1
-d <
T) BJ
M 3
"^ )-J.
BJ
rt
fD
n
o
W
BJ
co
S
H-
3
(D
n>
kr
3
rt
a = Clear well effluent
S3
fD
CO
rt
H-
fD
*
O
O
H
O
rt
BJ
t-"
O
i-l
fD
BJ'
Oi
H
O
M
3
CO
H-
fD
K
o
0
Louisville
M
0
H
Cincinnati
S
0
o
Hunt ing ton
NJ
0
o
Wheeling
H1
O
0
w
fD
BJ
fD
i-l
BJ
M
CO
s
O
0
"0
S3
1
CO
(-••
3
fD
n
M
H
O
0
Pittsburgh
M
O
O
Wilkinsburg
M
N3
O
O
n
BJ
T3
fD
M
N>
0
O
C
rt
H-
rt
CT1
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall >0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ue/L
Times MS confirmed when
Hall <0.1 ug/L
Times MS confirmation
attempted when
Hall = not detected
Times MS confirmed when
Hall = not detected
H
C/} M i
*~U c") i i
f& ^3 |Mt_j
x" 8
M a> K
.OROPROPENE FINISHED3 WATER DATA, JULY 1977- JUNE 197E
.PPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L,
MATE LOWER DETECTION LEVEL =0.1 ug/L
-------
SURVEY FOR BASE-NEUTRAL EXTRACTABLE HALOCARBONS
Discussions of extractable halocarbons are based on GC/Hall and GC/MS
analyses of project samples and on accumulated extractable halocarbon quality
assurance data (Appendix E). The application of quality assurance data for
extraction recovery, analyses of replicate samples, and replicate analyses of
sample extracts to the interpretation of project sample data was discussed in
Section 5.
1,4-Dichlorobenzene (Raw water data: Table 70. Finished water data: Table
71. Quality assurance data: Table E-l.)
1,4-dichlorobenzene (p-dichlorobenzene) was detected in 55 of 150 raw
water extracts and 62 of 154 finished water extracts. GC/MS confirmation
attempts for 1,4-dichlorobenzene were positive 85% of the time. Therefore,
1,4-dichlorobenzene was present in project raw and finished waters.
1,4-dichlorobenzene was detected more frequently at and downstream from
Huntington than upstream from Huntington. 'Further support for the presence
of the compound in this section of the Ohio River occurred in March 1978 when
a dichlorobenzene spill was reported on the Kanawha River. 1,4-dichloroben-
zene was GC/Hall detected and GC/MS confirmed in Louisville waters when flow
forecasts predicted the spill would pass.
Application of extraction recovery data suggests the following: when
detected in project extracts, 1,4-dichlorobenzene was present in project raw
and finished waters at concentrations not exceeding 3.1 ug/L (maximum concen-
tration in extract =1.9 ug/L, extraction recovery approximately 62%, there-
fore, 1.9/0.62 = 3.1 ug/L in water); following a reported 1,4-dichlorobenzene
spill on the Kanawha River, 1,4-dichlorobenzene was present in Louisville
waters at approximately 11 ug/L; when not detected in project extracts, 1,4-
dichlorobenzene was not present in project raw and finished waters above 0.2
ug/L.
1,4-dichlorobenzene was found in the extracts of raw and finished waters
from all project utilities. The precision of project field data for 1,4-
dichlorobenzene indicates that raw and finished water concentrations could
not be differentiated.
In another ORSANCO project utilizing the same analytical procedure and
laboratory, 1,4-dichlorobenzene was present in 80% of the samples from the
Kanawha River.1'
1,3-Dichlorobenzene (Raw water data: Table 72. Finished water data: Table
73. Quality assurance data: Table E-2.)
1,3-dichlorobenzene (m-dichlorobenzene) is a coproduct in the production
of 1,4-dichlorobenzene.20 1,3-dichlorobenzene was presumptively identified
in 12 of 146 raw water extracts and 14 of 151 finished water extracts.
Using the analytical procedure described in Appendix D, 1,3-dichloroben-
zene and 1,4-dichlorobenzene elute closely together and were sometimes not
159
-------
well resolved. GC/MS confirmation attempts for 1,3-dichlorobenzene were made
on four of the 12 presumptive raw water GC identifications; two of the four
confirmed. However, GC/MS confirmation attempts were also made on six raw
water extracts when the compound was not GC detected; 1,3-dichlorobenzene was
identified in three of the six samples.
1,3-dichlorobenzene was detected more frequently at and downstream from
Huntington than upstream from Huntington. Application of extraction recovery
data and GC/MS data suggest the following: when presumptively detected in
Huntington extracts, 1,3-dichlorobenzene may have been present in Huntington
waters at concentrations not exceeding 6.9 ug/L; when presumptively detected
in samples from other utilities, 1,3-dichlorobenzene may have been present in
those utilities' waters at concentrations not exceeding 1.2 ug/L; when not
detected in sample extracts, 1,3-dichlorobenzene was not present in raw and
finished waters above 0.2 ug/L; the frequency in which 1,3-dichlorobenzene was
identified may be other than that described by Tables 72 and 73.
In another ORSANCO project utilizing the same analytical procedure and
laboratory, 1,3-dichlorobenzene was detected in 40% of the samples from the
Kanawha River.
1,2-Dichlorobenzene and/or Hexachloroethane (Raw water data: Table 74. Fin-
ished water data: Table 75. Quality assurance data: Table E-3.)
1,2-dichlorobenzene (o-dichlorobenzene) is a coproduct in the production
of 1,4-dichlorobenzene.20 1,2-dichlorobenzene and/or hexachloroethane were
detected in 29 of 149 raw water extracts and 39 of 148 finished water
extracts. GC/MS confirmation attempts of presumptive identifications of 1,2-
dichlorobenzene were positive 67% of the time and GC/MS confirmation attempts
for hexachloroethane were positive 20% of the time.
Because of the GC/MS confirmation frequency and because this compound
was detected more frequently at and downstream from Huntington (similar to the
frequency of detection of 1,3-dichlorobenzene and 1,4-dichlorobenzene), it is
believed that 1,2-dichlorobenzene was more likely to have been present than
hexachloroethane. Further support for the presence of 1,2-dichlorobenzene in
this section of the Ohio River occurred in March 1978 when a dichlorobenzene
spill was reported on the Kanawha River. 1,2-dichlorobenzene was GC/Hall
detected in Louisville waters when flow forecasts predicted the spill would
pass.
Application of extraction recovery data suggests that: when detected in
project extracts, 1,2-dichlorobenzene was present in project raw and finished
waters at concentrations not exceeding 1.5 ug/L; when not detected in project
extracts, 1,2-dichlorobenzene was not present in project raw or finished
waters above 0.2 ug/L.
The precision of project field data indicates that raw and finished
water concentrations at and downstream from Huntington could not be
differentiated.
In another ORSANCO project utilizing the same analytical procedure and
160
-------
laboratory, 1,2-dichlorobenzene and/or hexachloroethane were detected In all
samples from the Kanawha River. Both 1,2-dichlorobenzene and hexachloroethane
were GC/MS confirmed in that river. "
GC/MS confirmation of hexachloroethane in finished waters of the Western
Pennsylvania Water Company (Monongahela River) and in the Kanawha River demon-
strates the presence of this compound.
1,2,4-Trichlorobenzene and/or Hexachlorobutadiene (Raw water data: Table 76.
Finished water data: Table 77.Quality assurance data: Table E-4.)
1,2,4-trichlorobenzene and/or hexachlorobutadiene were detected in 23 of
150 raw water extracts and in 20 of 120 finished water extracts. GC/MS con-
firmations of 1,2,4-trichlorobenzene were positive 89% of the time. GC/MS
confirmations of hexachlorobutadiene proved negative. Based on GC/MS fre-
quency, the compound detected was 1,2,4-trichlorobenzene.
The compound was rarely detected upstream from Cincinnati. The presence
of project field data indicates that raw and finished water concentrations at
and downstream from Cincinnati could not be differentiated.
Application of extraction recovery data suggests that: when detected in
project extracts at Cincinnati, Louisville and Evansville, 1,2,4-trichloro-
benzene was present in the raw and finished waters of those utilities at con-
centrations ranging from 0.2 ug/L to 1.0 ug/L; when not detected in project
extracts, 1,2,4-trichlorobenzene was not present in project raw and finished
waters above 0.2 ug/L.
Other Halocarbons
Information on the following base-neutral extractable halocarbons is
less definitive. The compounds were not detected or were detected in only a
few samples at low concentrations. GC/MS confirmation attempts on a limited
number of samples for a given compound were always negative. Extraction
efficiencies were highly variable.
Following the project data evaluation procedures, limiting concentra-
tions are suggested. These upper limit values apply to the specific analyti-
cal procedures used during this study. Data for the following compounds
should be used only after reference to the tabulated information.
bis(2-Chloroethyl) Ether and/or bis(2-Chloroisopropyl) Ether—
(Raw water data: Table 78. Finished water data: Table 79. Quality
assurance data: Table E-5.)
Detection of these compounds was complicated by interference from
dichlorocyclohexane as described in Appendix G. After statistical blank cor-
rection of sample chromatograms, the co-eluting compounds were presumptively
present in 4 of 267 project extracts; however, the concentrations were too low
for GC/MS analyses. Application of extraction recovery data suggests that
bis(2-chloroethyl) ether and bis(2-chloroisopropyl) ether were not found in
project raw or finished water at concentrations above 0.4 ug/L.
161
-------
bis(2-Chloroethoxy) Methane—
(Raw water data: Table 80. Finished water data: Table 81. Quality
assurance data: Table E-6.)
This compound was infrequently presumptively identified in project
extracts (frequency = 27/243). Most of these presumptive data were of insuf-
ficient concentration to attempt GC/MS confirmation. The presumptive GC
report of highest concentration proved negative by GC/MS.
Extraction recovery data for bis(2-chloroethoxy) methane at low levels
were extremely variable. The variability prohibits suggestion of a concentra-
tion at which bis(2-chloroethoxy) methane could be reported in project raw and
finished waters.
Hexachlorocyclopentadiene—
(Raw water data: Table 82. Finished water data: Table 83. Quality
assurance data: Table E-7.)
Hexachlorocyclopentadiene was infrequently presumptively identified in
project extracts (frequency = 17/260). When detected by GC/Hall, concentra-
tions were too low for GC/MS confirmation. Extraction recovery data for
hexachlorocyclbpentadiene at low levels were variable. This variability pro-
hibits suggestion of a concentration at which hexachlorocyclopentadiene could
be reported in project raw and finished waters.
2-Chloronaphthalene—•
(Raw water data: Table 84. Finished water data: Table 85. Quality
assurance data: Table E-8.)
2-chloronaphthalene was presumptively identified in 4 of 150 raw water
extracts and in 30 of 120 finished water extracts. GC/MS confirmation proved
negative in four of these finished water extracts. GC/MS confirmation
attempts of several chlorinated, in-plant waters also proved negative. The
compound is not believed to be 2-chloronaphthalene. The compound could not be
GC/MS identified. Because of difference in detection frequency and in concen-
tration between raw and finished water extracts, the unidentified compound may
be a chlorination product or may be a contaminant in chlorine used for
disinfection.
Application of extraction recovery data suggests that when not detected
in project extracts, 2-chloronaphthalene was not present in project raw and
finished waters above 0.2 ug/L.
4-Chlorophenyl Phenyl Ether—
(Raw water data: Table 86. Finished water data: Table 87. Quality
assurance data: Table E-9.)
4-chlorophenyl phenyl ether was rarely presumptively identified in pro-
ject extracts (4 of 150 raw water extracts and 8 of 155 finished water
extracts). Presumptive GC/Hall reports of higher concentrations proved GC/MS
negative. Application of extraction recovery data suggests the following:
when the compound was not detected in project extracts, 4-chlorophenyl phenyl
162
-------
ether was not present in project raw and finished waters above 0.2 ug/L; when
the compound was presumptively identified in project extracts at higher con-
centrations and GC/MS confirmation was not'attempted (frequency = 2/305), the
compound may have been present in project waters at approximately 1.0 ug/L.
4-Bromophenyl Phenyl Ether and/or a-BHC—
(Raw water data: Table 88. Finished water data: Table 89. Quality
assurance data: Table E-10.)
4-bromophenyl phenyl ether and/or a-BHC were rarely presumptively iden-
tified in project extracts (frequency = 4/304). These detections were of
insufficient concentration to attempt GC/MS confirmation. Application of
extraction recovery data suggests that these compounds were not present in
project raw and finished waters above 0.2 ug/L.
#-BHC (Lindane) and/or S-BHC--
(Raw Water data: Table 90. Finished water data: Table 91. Quality
assurance data: Table E-ll).
Lindane and S-BHC were presumptively identified in 4 of 149 raw water
extracts and in 20 of 155 finished water extracts. Concentrations of these
presumptively identified compounds were too low for GC/MS confirmation.
Application of extraction recovery data suggests the following: when not
detected in project extracts, these compounds were not present in project raw
or finished waters above 0.2 ug/L; when presumptively identified in project
extracts, the compounds may have been present in project finished waters at
0.4 ug/L. The USEPA interim standard for lindane in finished water is 4
ug/L.15
Heptachlor and/or g-BHC—
(Raw water data: Table 92. Finished water data: Table 93. Quality
assurance data: Table E-12.)
Heptachlor and/or p-BHC were presumptively identified in 42 of 149 raw
water extracts and in 43 of 155 finished water extracts. When concentrations
were sufficient for GC/MS analysis, the presence of neither compound could be
confirmed. Other GC/Hall reports remain presumptive.
The compounds were detected more frequently at Beaver Falls and at and
downstream from Huntington than at other utilities. The precision of field
data indicates that the concentrations in raw and finished water extracts
could not be differentiated.
Application recovery data suggests the following: when not detected in
project extracts, heptachlor and 3-BHC were not present in project raw or fin-
ished waters above 0.2 ug/L; when presumptively identified in project
extracts, heptachlor and/or 3-BHC may have been present in project raw and
finished waters at 0.2-1.5 ug/L.
Aldrin—
(Raw water data: Table 94. Finished water data: Table 95. Quality
assurance data: Table E-13.)
163
-------
Aldrin was presumptively identified in 32 of 149 raw water extracts and
in 45 of 155 finished water extracts. GC/MS confirmation proved negative in
five of these extracts. GC/MS confirmation attempts of several in-plant
waters also proved negative. The compound is not believed to be aldrin. The
compound could not be GC/MS identified.
The unidentified compound appeared with greatest frequency at and down-
stream from Huntington. The precision of field data indicates that the con-
centrations of the unidentified halocarbon in raw and finished waters could
not be differentiated.
Application of extraction recovery data suggests that when not detected
in project extracts, aldrin was not present in project raw and finished waters
above 0.2 ug/L.
Heptachlor Epoxide—
(Raw water data: Table 96. Finished water data: Table 97. Quality
assurance data: Table E-14.)
Heptachlor epoxide appears in the environment as a metabolite of hepta-
chlor.20 Heptachlor epoxide was rarely detected (frequency = 7/303) in pro-
ject extracts. Application of extraction recovery data suggests the follow-
ing: heptachlor epoxide was not present, with one exception, in project raw
and finished waters at 0.2 ug/L; on one occasion, the compound may have been
present at 0.3 ug/L.
a-Endo sulfan—
(Raw water data: Table 98. Finished water data: Table 99. Quality
assurance data: Table E-15.)
a-Endosulfan was presumptively identified in 35 of 149 raw water
extracts and in 24 of 154 finished water extracts. Presumptive GC/Hall
reports at higher concentrations proved GC/MS negative. It is not known
whether other GC/Hall reports of lower concentration (extract concentrations
of 0.3 ug/L or lower) were a-endosulfan.
Extraction recovery data indicate low recovery of a-endosulfan and sug-
gest the following: a-endosulfan was not present in project raw or finished
waters above 3.0 ug/L; when not detected in project extracts, a-endosulfan
was not present in project raw and finished waters above 1.0 ug/L.
DDT—
(Raw Water data: Table 100. Finished water data: Table 101. Quality
assurance data: Table E-16.)
DDT was presumptively identified in 6 of 303 extracts of project samples.
The GC/Hall report of highest concentration proved negative by GC/MS. Appli-
cation of extraction recovery data suggests that DDT was not present in pro-
ject raw or finished waters above 0.2 ug/L.
Dieldrin and/or DDE—
(Raw water data: Table 102. Finished water data: Table 103. Quality
164
-------
assurance data: Table E-17.)
20
Dieldrin appears in the environment as a metabolite of aldrin and DDE
as a metabolite of DDT.^0 Dieldrin and/or DDE were rarely presumptively
identified (frequency = 6/303) in the extracts of project samples. Applica-
tion of extraction recovery data suggests that dieldrin and DDE were not
present in project raw or finished waters above 0.2 ug/L.
Endr in—
(Raw water data: Table 104. Finished water data: Table 105. Quality
assurance data: Table E-18.)
Endrin was presumptively identified in 1 of 303 extracts of project sam-
ples. Application of extraction recovery data suggests that endrin was not
present in project raw or finished waters above 0.2 ug/L. The USEPA interim
standard for endrin in finished water is 0.2 ug/L.l-*
ODD—
(Raw water data: Table 106. Finished water data: Table 107. Quality
assurance data: Table E-19.)
ODD appears in the environment as a metabolite of DDT.^0 it was not
detected in the extracts of project samples. Application of extraction recov-
ery data suggests that DDD was not present in project raw or finished waters
above 0.3 ug/L.
3-Endosulfan—
(Raw water data: Table 106. Finished water data: Table 107. Quality
assurance data: Table E-19.)
3-endosulfan was not detected in the extracts of project samples. Appli-
cation of extraction recovery data suggests that 3-endosulfan was not present
in project raw and finished waters above 0.3 ug/L.
Methoxychlor—
(Raw water data: Table 108. Finished water data: Table 109. Quality
assurance data: Table E-20.)
Methoxychlor was not detected in the extracts of project samples. Appli-
cation of extraction recovery data suggests that methoxychlor was not present
in project raw or finished waters above 0.2 ug/L. The USEPA interim standard
for methoxychlor in finished water is 100 ug/L.15 (Text continues on page
206.)
165
-------
ON
ON
* fD O- o CT* B>
g O -~J O S3 co
21 O O H- CO n>
H C O rt
>3 3 C fD >rj
&> &TO pd 3 H-
(D
O pj
W fD C 3
_. i-i H PJ 3
g H- rt CO
O co 3 ^
H C TO S3 M
n •
O
n
H
M
o
M
3
i
H
i
CL co PJ
H- rt 3
O p.
P4 < ft)
o ro' s;
H S| PJ
O rt
W fD
fD
ro
n
o
ft?
rt
s:
co
rt
P-
m
«
fD
|__|
M
M
to
H1
ON
M
ON
M
M
H
O
rt
PJ
H'
O
£
re
P>
3
M
U>
to
to
CO
0
O
ON
M
VO
^1
ON
CO
to
P>
3
co
P-
P*
P1
(D
p»
**
to
O
ON
P>
i
CO
to
£
P-
CO
H-
(-»
p>
n>
1— 1
CO
0
0
00
M
M
CU
M
M
n
H-
3
O
H-
I
n
P.
• j
P*
*"
*
0
Ul
M
^
PC
§
rt
5"
00
rt
O
3
to
*-'
*
0
ON
!-•
Ul
10
to
M
S"
n
n
p.
00
ij
to
0
*
M
CD
2
fD
Tl
p)
I—1
CO
I .
00
o
Ul
z:
0>
co
rt
H-
fD
s
1 .
M
O
CO
s;
6s
V!
co
•v
H-
3
a"
. .
to
to
o
o
co
0
ON
h- i
O
rt
rt
CO
cr
TO
. .
M
M
to
o
ON
O
ON
H-
H- >
H-
3
CO
a*
w
TO
VO
I--
M
M
VO
M
VO
M
I-1
C?
x
n
IT
•o
fD
, ,
to
O
M
G
rt
i .
H-
rt
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £.0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall >0.1 ug/L
Times MS confirmed when
Hall £-0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
1-3
a w
t-|H
iv» "^j
O tr1 O
or1*
CO M h- '
M •^"*"
fe 2 o
'"o ^^ "^
srg
Ij hrl te
M CD W
g M |
W M ^
Ggl
!?d m g
M H S
O M ftf
3 M O
f 0 ^
"i *
II M ^
P ^ K
KV§
0 O |
• H*
c 3
TO M
^S
oo
-------
* d.
o
§"
n o
M H
2; o
H C
W 3
?> &
H
M S|
Sft
w m
2!
O CO
H e
b
c
O M
I
o
s: co
m fD
CO (D
rt
fD Tj
H H-
3 OQ
^^
(D fD
3
3 M
CO
VJ
ft)
3
X
O
H
i
w
M
fD
H
n
o
PI
^
CD
ft)
3
rt
V
II
o
M
fD
ft)
n
s!
(D
M
M
efflue
3
rt
(D
CO
rt
H-
fD
M
to
M
(•0
O
ON
O
ON
-
-
rt
H
O
f?
f»
3
H
JN
NJ
JS
-
-P-
H
H"
KJ
-
M
Pittsburgh
-
o
M
Wilkinsburg
o
o
o
%
T3
ft
H"
M
O
,_,
*
M
*
M
O
G
rt
H-
rt
Times Hall searched for
Times Hall found
^.0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ue/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
H
n f i-1
» H O
M H
H > W
rg| m hd
f H M
§ l> ^
H
O t-1 W
n fa ft
O M t>
21 H H
M O >
" ET*
PJ *P
o <
• W LH
II ^
€?5
c 7"
00 ^
I-1
VO
00
-------
c»
n
i
o a4 fu
II II II
CD O S3 CD
i-f 3* m ro
o H- co n>
£ O rt
3 fD "rl
ei. pa M H.
H- 3 (TO
O 03
'Z rt
CO (t)
n
n
H
H
i
c-1
o
CO
CO
hd i-(
(B (D
§ H
CO
i—3 C ^ i j
T3 (D <
O "Tj CO P
O H- rt 3
S"^ H'
2 sj
S! 0>
rt
a
o
o
EC
pj
X
co
S
fi
ft
03
rt
<
H.
1
M
I-"
M
O
O
•
w
O
•
co
M
O
H
O
ft
P>
H
O
H
I
0
H
CO
01
Cn
O*
O
•
VO
CO
•
00
I-"
M
00
.p»
Evansville
(-•
H
M
to
0
•
I-"
O
•
M
(->
M
M
0
Louisville
H
H
O
H
M
M
Cincinnati
M
(-•
O
(->
sc
rt
H-
3
OQ
rt
O
3
N9
H
CO
O
(-•
•
*•
CO
•
00
CO
CO
Wheeling
(-•
o
O
o
M
O
Beaver Falls
M
00
(->
0
o
•
to
0
•
M
rt)
co
rt
<
H-
IB
O
M
M
M
l->
O
•
N)
O
•
Ni
WPW/Hays Mineb
H
H
0
O
M
O
Pittsburgh
M
M
O
O
Wilkinsburg
00
0
o
h-1
o
*}
o
X
n
CU
T3
n>
i-1
M
NJ
O
O
C
rt
H-
I-1
H-
ft
1>
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^.0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall >0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
s
0 f
n M
W ^j
£?
o t-<
n
^M-M
W H Co
» M 1
0 O
> H M
ni O O
ITJ M aj
pd « f
o o
X > £J
M hi O
S !D W
t> Sd M
NZENE RAW WATER DATA,* JULY 1977- JUNE 1978
OXIMATE LOWER DETECTION LEVEL =0.1 ug/L
TE LOWER DETECTION LEVEL = 0.15 ug/L
-------
TABLE 73. 1,3-DICHLOROBENZENE FINISHED3 WATER DATA,* JULY 1977-JUNE
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.15 ug/L
1978
Utility0
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mine0
Beaver Falls
Wheeling
Huntington
Cincinnati
Louisville
Evansville
Total or Mean
West Viewd
t-4
O
T3
at
f.
u
O>
tn
CO
PC
(fl
0)
B
•H
H
11
10
10
12
20
11
23
16
15
11
139
12
j
oc
'X3 3
c
3 i— 1
O •
4-1 O
/fc
i
CO
PC
en
0)
e
•H
H
0
0
0
0
1
0
4
2
1
0
8
0
fj
co
•0 3
C
3 i-l
O •
H-l O
!_) V
CO
CC
0)
e
H
1
0
0
1
0
0
1
0
2
1
6
0
o
•rt
CO >J
•u oo
C 3
01
O iH
C •
0 O
^»
•-I C
CO O)
*•§
c
cfl
0.6
0.9
0.5
0.2
0.7
oo
3
C
0
4-1
cfl
cO C
ffi O)
O
1 §
!*!
ss
0.6
2T5
0.7
0.2
2.5
0
o
•H (-3
co 01 w
eJ= 3
S
U-t T3
q a> o
O 4-1 /\
CJ P.
f
in CO
en co PC
01
jjj
H
1
1
2
4
1
c
01
.c
3
ij
TD - —
0) OO
e 3
C 0
o v
CJ
cy: T— I
CO
CO PC
ai
•H
H
0
1
0
1
0
a = Clear well effluent
b = see Figure 1
c = Western Pennsylvania Water Co., Hays Mine Plant
d = Ground water supply
*CONCENTRATIONS NOT CORRECTED FOR EXTRACTION LOSSES.
169
-------
*
o
i
o
JQ
fD
O O*
H 00 Hi-
H
O II II
2
WOO
o o
2! 3 3
O Hi Hi
H H- H-
O (B CD
S H- H-
pa o o
033
H
pa o o
O Hi Hi
*3
o
«M
CU I
O CL
H1 o'
O 3*
•f M M
n o o
rt O
pa
3
H
H
O
Cfl
CO
Cfl
3 3
fD N
II II II II
ff.
o «~.
§c
&> 00
O O Si
fti
fD
CO
rt
CD
H
(B fD ^
rt CX H fD
fD C 3
H H CO 3
£ OQ ^ (—•*
*T3 fD ^
1j Q_, (/) fu
H H- rt 3
*
P
-
O
o
NJ
O
NJ
H
O
rt
0)
O
2
fD
01
3
M
CO
oo
-
p
p
-p-
o
VO
w\
X
\
\
Evansville
P
-
-
o
NJ
O
H\
o \
\
Louisville
H
O
a.
*
P.
p.
Cincinnati
o
i-1
-
o
p
NJ
\
\
Hunt ing ton
NJ
.p-
Ln
O
OJ
O
00
X
X
\
\
Si
fD
(D
H-
3
00
NJ
O
O
w
fD
(B
(D
(0
H1
CO
oo
o
o
Si
fD
CO
rt
H-
fD
O
P
O
-
Si
CO
S
H-
fD
H1
NJ
h-"
O
O
O
•
\
X
Pittsburgh
M
O
NJ
Si
3
CO
cr
i-!
00
VO
0
-
o
X
*o
fD
h-1
r-1
NJ
O
O
H-
ft
Cu
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall >0.1 ug/L
Times MS confirmed when
Hall £.0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
o
n
EC
o F
s o
63 M
" pi
CACHLOROETHANE JULY 19 7 7- JUNE 1978
R, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
ROXIMATE LOWER DETECTION LEVEL = 0.15 ug/L
pa
3
-------
a. o cr os
n n n n n
Q o S3 co o
l-t • CD (T> K->
O Cn CD fD CD
§rt 03
c n>
s;
(D
01
rt
<
H-
§
srt>
i-1
to
O
H
H
O
rt
f»
H
O
l-t
i?
S3
3
M
CO
O>
CO
Co
O
o
*
CO
M
•
M
^
X
X
X
Evansville
H
O
M
Ul
O
•
*>
O
•
VO
X
X
Louisville
M
-p-
o
&
CO
a.
a.
X
X
Cincinnati
M
-P-
H-1
»-J
O
•
(-•
O
•
\->
X
X
Huntington
ro
10
co
*-
o
•
to
O
•
j^
Wheeling
t->
10
O
M
Beaver Falls
to
o
M
CO
o
•
H*
O
*
M
WPW/Hays Minec
M
Co
M
N3
M
•
h-1
h-1
•
^
X
x
Pittsburgh
M
M
O
S3
Wilkinsburg
VO
o
to
3
X
o
S"
Ca
T3
(0
(-•
h-1
I-1
O
I-1
C
rt
H-
I-1
I—
rt
^!
cr1
Times Hall searched for
Times Hall found
>0.l ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ue/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
TABLE 75. FINISHED3 WATER DATA FOR 1,2-DICHLOROB
HEXACHLOROETHANE, JULY 19 7 7- JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEV
GC/MS, APPROXIMATE LOWER DETECTION LEVEL = 0.
i-1 w H
CHI- g
c H S
J>? H
[__1 t^
c o
£. 0
p &
-------
O H- H-
£d o o
ftf 3 3
13 co cu
M rt 3
< to'
O
El
Ml
O
l-t
tx| fD »
O X to
XI PJ »
O -f>
M p- |
X M rt
H O hj
ta hj H-
> o n
n C71 3*
H C H
M rt O
O BJ h|
2! CU 0
H- CT
f fD fD
O 3 3
C/3 fD N
(/l fD
M 3
C/3 rt>
n
O
„
pr*
Pi
CO
^g*
H-
3
fD
t-^
^_j
to
5
rt
ID
co
rt
H-
fD
M
0
-
H
O
rt
I-1
O
hi
S
fD
3
h-»
Co
vo
H
oo
O
CO
o
Ln
V
C0\
\
Evansville
M
M
-
-P-
O
CO
o
\M
M\
Louisville
M
-
to
O
Co
O
Ul
V
N\
N
Cincinnati
H
H
CO
fO
O
H
O
H
\
\ *"*
Hun ting ton
tO
h-1
O
o
Wheeling
e
0
0
Beaver Falls
oo
o
o
s:
fD
CO
rt
H-
O
-
O
o
S3
(U
cn
H-
0)
cr
K
o
o
Pittsburgh
u
0
0
Wilkinsburg
*
o
0
o
X
n
ft)
h-1
-
0
o
rt
H-
h-1
H-
rt
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
o
o H
n n
M M
- H
i-d o
)R, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
ROXIMATE LOWER DETECTION- LEVEL = 0.15 ug/L
w
EC
Sg
§3
W ?d
H O
£g
H £>
g *
W *"d
o
en1"
^r
»-< ro
V
I-1 4^
vp I
^J H
«J &
I H
t-i n
OROBENZENE
E 978
OR
-------
TABLE 77. FINISHED3 WATER DATA* FOR 1,2,4-TRICHLOROBENZENE
AND/OR HEXACHLOROBUTADIENE, JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL = 0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.15 ug/L
Utility0
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Minec
Beaver Falls
Wheeling
Hunt ing ton
Cincinnati
Louisville
Evansville
Total or Mean
West Viewd
1-1
0
TD
0)
O
l-i
to
OJ
en
rH
PS
CO
01
H
11
9
8
10
18
10
11
12
11
8
108
12
60
13 3
C
3 rH
o •
«4-l O
rH
cn
QJ
•H
H
0
0
0
0
0
0
0
1
4
3
8
0
00
pj
3 rH
O •
<4-l O
CO
CO
CU
H
0
0
1
0
0
1
0
4
3
2
11
1
c
o
•H
l-l •>»
4-> 00
C 3
0)
O rH
°A\
rH
rH C
CO 01
CO
cu
0.3
0.3
0.3
0.3
5
00
G
0
•H
4-1
(fl
rH 4->
a] (S
PC 53
a
§ §
B o
•H
K
1
0.3
0.3
0.6
0.6
o
o
S CU 01
c ^
C
j3
^
6 3
C 0
OA\
CQ rH
tO
CO PS
s
H
2 s'
^0
1 ./
/o
x^
-4
o
H nJ
4-1 C -»-
Cfl QJ 00
3 "S '")
C ^ 0
O 4-1 V
0 P.
C/3 01 rH
IS 4-1 rH
4-1 CO
CO CO PS
-1
1 ^S
S^l.
^
0)
^ J
01 00
3 3
•H rH
a o
8V
C/J rH
x3
.§
H
O^x^
^•^o
>^
a = Clear well effluent
b = see Figure 1
c = Western Pennsylvania Water Co., Hays Mine Plant
d = Ground water supply
e = confirmation for 1,2,4-trichlorobenzene
f = confirmation for hexachlorobutadiene
*CONCENTRATIONS NOT CORRECTED FOR EXTRACTION LOSSES.
173
-------
TABLE 78.
BIS(2-CHLOROIOSOPROPYL)ETHER AND/OR BIS(2-CHLOROETHYL)ETHER
RAW WATER DATA,* JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL = n 9 /T
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =02 '
Utility3
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mineb
West View0
Beaver Falls
Wheeling
Hun ting ton
Cincinnati
Louisville
Evansville
Total or Mean
West Viewd
Times Hall searched for
12
9
11
12
11
18
12
21
10
11
11
138
11
*i
00
•g s
3
o •
"4-1 O
M V
.H
£
(0
4-l >
C 0
SA\
CO iH
*1
SK
J
H
Times MS confirmation
attempted when
Hall <0. 2 ug/L
1
1
Times MS confirmed when
Hall <0.2 ug/L
0
0
a = see Figure 1
b = Western Pennsylvania Water Co., Hays Mine Plant
c = Ohio River at West View
d = Ground water supply
*CONCENTRATIONS NOT CORRECTED FOR EXTRACTION LOSSES BUT
ARE BLANK CORRECTED. SEE APPENDIX G.
174
-------
TABLE 79. BIS(2-CHLOROIOSOPROPYL)ETHER AND/OR BIS(2-CHLOROETHYL)ETHER
FINISHED3 WATER DATA,* JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.2 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL = 0.2 ug/L
Utilityb
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Minec
Beaver Falls
Wheeling
Hunt ing ton
Cincinnati
Louisville
Evansville
Total or Mean
d
West View
o
U-l
•o
01
f
U
CO
0)
(0
iH
CO
EC
(0
01
H
11
9
8
10
17
10
11
11
11
8
106
12
,J
00
tJ 3
in O
CO
0)
01
.9
H
0
0
0
0
0
0
0
0
0
0
0
0
,J
00
o •
*l~j ^^
CO
Ed
oi
0)
H
0
0
0
0
0
0
0
0
1
0
1
0
c
0
^
4J
CO i-J
4-1 00
g =•
O CN
U ^
•H C
CO 0)
CO
0>
00
§
•H
4-1
CO
!H M
iH 4-»
£ g
Maximum
cone
c
o
^ e jJ
CO (U 01
H C-
3 "S CD
C« 0) iH
SB 4-> r-t
4-1 C!
U) CO EC
0)
^
C
•H CN
>4-( «
£2 O
o /\
0
(U
H
^
o
4-1 C --^
01 0) 00
: "S d
o w v
00.
2 *-> rH
4-1 CO
CO CO EC
D
H
H
01
.C
-
•o • —
4J 00
e 3
•H CM
<4-l
c o
8V
s: --i
CO
H
a = Clear well effluent
b = see Figure 1
c = Western Pennsylvania Water Co., Hays Mine Plant
d = Ground water supply
CONCENTRATIONS NOT CORRECTED FOR EXTRACTION LOSSES BUT
ARE BLANK CORRECTED. SEE APPENDIX G.
175
-------
x- CL n D
O II II II
a
n o o s
M n cr ro
!2! O H- CO
1-3 C O rt
SO 3 (D
5> (X !*j H
I c <
i pi fo
rt
3
§
CO fD
H (SI .
3 rt CO
O CO •-<
T3 ft) <
nta co pi
O M rt 3
n
H
M
U
M
§2-
n
H
M
§
o
C/3
CO
M
CO
rt
ro
n
o
PC
(U
s
3
3
rt
Q)
n
co
(15
to
H-
1
O
(D
0)
rt
H-
(D
-
0
O
H
0
rt
O
H
(0
Co
D
M
VD
I-"
h-1
O
00
O
*
00
-
0
-
o
Evans ville
i-1
o
o
w
M
0
f
O
C
H-
M
I-1
CD
P
M
ON
O
•
00
0
•
00
M
o
Cincinnati
p
o
o
Hun ting ton
o
o
o
Wheeling
H-1
O
M
»
I
01
I-1
co
-
o
-
n>
CO
rt
(D
€
n
P
o
o
WPW/Hays Mineb
(-•
o
0
Pittsburgh
-
o
0
Wilkinsburg
VO
o
o
o
X
n
0)
ro
t-1
I-1
Nl
O
O
c
rr
I—1
H-
rt
ftl
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
H
&
0 tr1
O M
t""1 *
a ^
CO M s~>
- n M
H 1
5§R
OROETHOXY) METHANE RAW WATER DATA,* JULY 1977- JUNE 1978
APPROXIMATE LOWER DETECTION LEVEL = 0.1-0.2 ug/L
ROXIMATE LOWER DETECTION LEVEL =0.25 ug/L
-------
o
o
o, o
n n
tr
n
o S3 to n
H tt> fo (->
O CD (D (D
0 rt (B
p (D >TJ ^
CL If H-
S OQ Si
$ C i?
fU ^ H H
ft tD (D H
n> 3
H 3 M ft)
en hh
en *
Louisville
P
0
CO
Cincinnati
i-1
o
ts3
Huntington
VO
-
O
o
o
10
Wheeling
o
h- '
N>
O
1-"
o
I-1
Beaver Falls
S3
I-"
O
o
I-"
o
H"
WPW/Hays Mine0
o
o
o
Pittsburgh
oo
o
-
Wilkinsburg
VO
o
o
o
X
n
T3
fD
I-"
O
o
ft
h—
1— •
H-
rt
CT4
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall >0.1 UE/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
GC
GC
AP
DETECTOR, APPROXIMATE LOWER DETECTION LEV
MS, PROXIMATE LOWER DETECTION LEVEL =
LE 81. BIS(2-CHLOROETHOXY
JULY 19 7 7
E
T
EL
to
Ui
VD W
^J
00 Hrj
H
2!
M
Cfl
g
O
to
I
w
f o
.
ro
oo
o
15
-------
Cu O
II II
CO
|
n o o si CD
cr
II
o H-
§ °
CX. td
ro n>
CD (D
rt f-j fD
H
M
O
to fD
SI
O M
H C
n xi co fo
O M rt 0
o
H
fd
M
i
O
W
CO
M
en
CO
. "
rt
(D
i-i
O
O
W
(U
V!
CO
S
6
rt
03
rt
H-
It
M
O
O
H
0
rt
BJ
0
H
fl>
3
H
VO
-
•vj
O
I-1
p
I-1
Evansville
M
o
-
Louisville
1-1
o
Isi
Cincinnati
M
O
O
Hun ting ton
H1
O
to
Wheeling
to
-
0
p
o
Beaver Falls
00
o
-
s:
n
en
rt
H-
n>
o
M
o
o
WPW/Hays Mineb
to
0
o
Pittsburgh
M
o
-
Wilkinsburg
VO
O
o
13
o
X
r>
Cu
T)
(6
I-i
N)
O
O
C
rt
H-
H-
rt
P)
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £.0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall >0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
CD
O
O P
rsi o
55 M
• H
> o
§8
o -
H !>
H Jti
M O
t-H
O S
0 f
M O
M O
is
M
f n
M H
W O
n rli
LJ tr1
Ln
fl HJ
1
0
N)
&
i
f
.00
K3
n
S
hd
H
*
Pd
WATER
VO
•^1
I
W
VO
00
-------
Jk
TABLE 83. HEXACHLOROCYCLOPENTADIENE FINISHED3 WATER DATA, JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL = 0.1-0.2 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL = 0.35 ug/L
Utilityb
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mine°
Beaver Falls
Wheeling
Hunt ing ton
Cincinnati
Louisville
Evansville
Total or Mean
West View
M
O
^~*
t)
01
u
(-1
CO
0)
CO
rH
cfl
pa
0)
e
•H
H
11
9
8
10
18
10
11
12
11
8
108
12
j
00
•0 3
c
3 rH
O •
'_[__[ £2
rH
CO
pa
CO
01
g
H
0
0
0
0
0
0
0
1
1
0
2
0
^
00
-0 3
C
3 rH
M-l O
__, V
rH
CO
pa
CO
0)
.§
H
0
0
1
0
2
1
1
0
1
1
7
0
c
iH
4J
4-1 OO
St 3
0)
O rH
c •
0 0
rH
rH C
CO 01
pa x
<0
£
0.1
0.2
0.2
jj
00
3
c
o
CO
rH r-l
rH 4J
m C
iximum H
conce
£
0.1
0.2
0.2
rj
3
CO 0) 01
H_| rrj .
c o) o
° w A\
1 1 ^^
0) 0) rH
SS 4J rH
•M Ct
w co pa
01
.g
0)
^
•o *i
0) 00
c o
OAK
co pa
0)
.9
H
c
O
£j c jj
CO 0) 00
3 •£ 3
H rH
C 0) O
o -u v
u ex
CO Ol rH
JE -U rH
4-1 CO
co co pa
01
B
H
H
C
QJ
rC
^
-0*3
01 Oi
B 3
•H rH
C 0
o V
CA i— 1
CO
to pa
0)
•H
H
a = Clear well effluent
b = see Figure 1
c = Western Pennsylvania Water Co., Hays Mine Plant
d = Ground water supply
*CONCENTRATIONS NOT CORRECTED FOR EXTRACTION LOSSES
179
-------
oo
o
i
S
n n n
o o
i-f rr
S3 co
CD n>
CO (D
o rt
(D
, <
o cu ro
H H
P
21 rt i-i CD 0)
en fD 3
2! rt CO
^D W *^
H e s^ M
t3 (B <
nt3 co (u
O r-1 rt 3
n
tr1
O
co
01
< fu
rt
fD
n
o
CO
g
5'
ru
nd
I-1
(a
y
s:
(D
to
n
<
H-
(t>
*L
M
M
O
.O
g
rr
P>
M
O
11
3
ft
CD
3
H1
U>
VO
S3
N>
O
•
ro
0
•
N)
Evansville
M
I-1
0
H>
Louisville
M
(->
O
M
Cincinnati
(-•
M
O
O
Hun ting ton
N5
M
O
O
Wheeling
i-1
KJ
O
O
Beaver Falls
M
oo
rs:
o
o
•
NJ
O
•
NJ
s:
n>
co
rt
<
H-
(D
O
I-1
l->
O
O
WPW/Hays Mineb
(->
N5
O
O
Pittsburgh
i->
M
O
O
Wilkinsburg
«3
o
o
•fl
o
X
n
^r
(U
X)
ft
H-1
I-1
to
O
o
c
rf
H-
M
H-
rt
•<
fa
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £.0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall >0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
TABLE 84. 2-CHLORONAPHTHALENE RAW WATER DATA,* JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.15 ug/L
-------
CO
I
a
O 0*
O SI CO
f-f CD (D
o co n>
(i rt
(T) *rl
n
a.
o P3
SS rt
C/3 (D
H
2!
O CD
(D
RJ
O
H
M
n
H
H
i
f
O
CO
CO
M
en
•"S
hd i-t
fD (D H
0
0 M (D
CO Hi
3
rt
rt
(D
i-i
n
O
n
0)
n
H-
1
N)
O
0
H
O
rt
H
O
M
1
9
0
00
s
-
0
H
o
-
o
Evansville
00
ro
-
0
CO
o
CO
-
o
Louisville
i-"
CO
o
o
co
0
Cn
H
o
Cincinnati
H
ro
en
o
p
NJ
O
CO
Hunt ing ton
H*
H
M
-
O
en
o
Cn
Wheeling
0
10
-
o
CO
0
CO
-
0
Beaver Falls
oo
i->
N3
0
O
en
H"
O
-
O
WPW/Hays Minec
0
o
o
Pittsburgh
oo
-
H
O
CO
0
CO
Wilkinsburg
\o
0
o
o
X
n
g"
•o
re
H1
I-"
0
O
c
ft
H>
H-
rt
cr1
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
Times MS confirmation
attempted when
Hall £0.1 ue/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
o
hd
h-l
.OXIMAT
H
>Tl
. DETECT
H
O
M
<(
f
II
M
Ln
C
OQ
o
n oo
^ *Ji
O f K>
I
O
O
§
.»fc
si
§
W
H M
M s
H
f CO
>d.
§
H
n
M
§
f
w
I
I
v
*
c-i
OQ
V£5
^J
^J
VO
•^J
co
-------
oo
* CU O
I • •
n o o
a4 P
n n
s! co
fD fD
CO ft)
O rt
3* 5"
O P fD
25 rt hj
hj (D
-S5 rt
O co
t-i C 5^
n T3 co
O h-1 rt
rs
O fD
H f.
M
O
n> in
fD fD
o
*J
w
a
I
H
M
§
C/3
M
cn
rt
fD
H
O
O
•3
CO
3
fD
M
0)
3
rt
s:
fD
co
rt
H-
fD
1
1 .
"""
O
0
H
0
rt
P)
H"
0
2
n>
5
VO
I-1
u>
o
10
o
,10
M
o
M
P
3
01
H-
(D*
• .
M
O
O
r-1
0
H-
co
H-
M
(t
. ,
M
O
0
n
H-
O
H-
3
g
H-
0
o
PC
3
rt
H-
OQ
rt
O
3
O
M
f.
ff
fD
f-^
H-
3
OQ
10
O
0
n
P
^
fD
TJ
03
I-1
CO
00
0
o
y.
co
rt
H-
fD
0
M
M
M
0
O
N3
H-
0
y.
s:
PC
P
CO
S
H'
3
fD
cr1
10
0
o
^
rt
rt
CG
cr
c
00
M
o
M
p.
H-
q—
(-"•
3
(D
cr
c
n
TO
VO
0
O
T1
O
n
g"
•c
fD
I—1
10
O
O
^
rt
H- >
H*
rt
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £.0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
H
&
tr1
M
O 00
o
sT
t^
O t"1 -f^1
n f i
~-~ o
SOW
• HO
p§ n S
x J<
M O M
i i ^
S H W
SS^
H s: ..
H M g
O M g
f n M
Pd H ?®
< M ._.
M O O
II f t1*
? M *
H f ii
Ln G
II t"1
c K!
^ O j_,
c •>"
OQ 1
^ 1
s
oo
-------
oo
OJ
a = Clear well
b = see Figure
c = Western Pei
d = Ground wat(
H 3 M ro
CO Hi
CO •< hh
£ M M
T3
-
O
o
p
H*
O
Pittsburgh
H1
h-1
O
O
Wilkinsburg
h-1
O
O
O
TJ
0
X
n
CB
ro
*-•
i-1
o
o
G
rt
H-
H-
rt
Times Hall searched for
Times Hall found
^0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ue/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
o
o
ffi
s»
o F
n t-1
•**^
S G
CO W
- H
M
^3
ho o
§,*"
X
gs
H
15
f
w
CO
VJ
•
*~
o
p
o
hd
-------
00
H- CL O cr1 (U
O II II II ||
o o o s! co
M M p" fD fD
2| O H- CO fD
H d O rt
» 3 (D hrj
!> a. pd H H>
H H- 3 00
M 5| < e
O 03 fD hd l-J
25 rt l-J fD fD
00 fD 3
l-J P3 3 M
25 rt CD
Hd S3 M
O 13 CO [U
O M rt 3
2 <3 fu
n rj> 23
H 3 (a
o n>
o n
po o
x
o 'd
H S
O H.
i*£< 3
tr1 ro
O m
CO l_i
00 03
3
5
(C
co
rt
H-
It
M
t-1
O
O
H
rt
0>
O
X,
3
U)
00
H
NJ
i
o
M
0
H1
n
vansville
M
o
o
r
ouisville
M
O
O
n
in cinnati
M
o
*-
X
un ting ton
to
~
M
0
M
O
M
s
heeling
N5
o
0
Dd
eaver Fal.
CO
.00
o
o
s
0!
co
rt
(C
n
i-4
o
o
s
PC
tu
co
H-
3
fD
cr
i-1
o
o
TJ
ittsburgh
M
O
O
SI
ilkinsbur
OQ
VO
o
o
»fl
o
X
n
cr
03
T3
ft
O
O
C
rt
H-
I—1
H-
rt
(B
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
!"imes MS confirmed when
Hall <0.1 ug/L
8
tCT*
H
n t-1
io M
" M
f? H
' 0 tj pO
pd pd p>
O w S]
^>^3
K% >
j pd td
S! ^ (-3
id H <• >;.
H f f-t
H ^ M
o w ^
25 H ^J
M I
-1 o c_,
?J O M
tr1 21
M
1 tr1 vo
o
00
oo
Q
ta
PC
n
-------
00
Ul
&• o
II II
n n
*
n
i
Q O SI CD O
W i-i fD fD h->
O CO (D fD
rt PJ
fD trj H
<; tu g H-
H 3 OQ
H S £
O PJ hrf hj
3 rt fD fD
W fD 3
w
g
O CO H
H c
V
H1
tsJ
0
0
3
rt
»
M
O
H
a?
to
3
H
|£
o,
M
Evansville
M
t-1
o
o
Louisville
H*
Ui
O
0
Cincinnati
M
OS
0
I-1
Hun ting ton
N>
*-
o
o
Wheeling
!->
ISi
O
O
Beaver Falls
to
o
o
o
WPW/Hays Minec
M
U>
0
O
Pittsburgh
M
(-•
O
0
Wilkinsburg
M
O
O
O
*j
o
x
n
»
•o
m
M
H
H-*
o
o
c:
rt
H-
M
H-
rt
*<
cr
Times Hall searched for
Times Hall found
^0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 UE/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
TABLE 89. 4-BROMOPHENYL PHENY]
FINISHED3 WATER DATA? JU
GC/HALL DETECTOR, APPROXIMATE LOWER
GC/MS, APPROXIMATE LOWER DETEC
l-i IT1 IT1
H a ~s
O M M
^Ei'sa
tr1 n -vi w
M H ^J ?d
<^ M 1
t3 o <-t >
f a g g
II t-1 td ^
w o
0 < M Pd
• m v£>
h-» f ^J 9
<_n 00 1
II td
C K
OQ o n
€
t-r
-------
oo
* a. n a4 tu
0 II II II II
n o o S3 CD
30 H- co a>
H g O rt
C? £» H_l Lj ij
H H- 3 OQ
M 5! < C
O B> fl> ^d i-j
2J rt H (T) fl>
CO fl> 3
h| CO 3 H"
2! rt co
O CO •-<
H C S3 M
*O (D <
cTd co fu
O M rt 3
pdV! H-
O CD S3
H 3 Co
0 0
!?d o
M
X
w Co
n *c1
M 3
0 Hi
25 3
fl>
O '"'d
M 3
00 rt
s
CO
rt
H-
ID
£-
H1
l~1
O
O
H
O
rt
O
o
PC
(u
g
u>
00
o
w
B>
9
5
H-
M
I-1
A
H
M
O
^
f
r1
o
c
H-
§
H-
I-1
M
n
i-1
H
0
o
n
£•
o
&
g
rt
H-
^
O
O
K
§
rt
5-
00
rt
O
3
to
O
to
z:
re
H-
3
00
i-i
to
o
o
03
I
n
0)
CO
oo
o
M
y.
n>
co
rt
p.
^
o
I-1
o
o
f
s
(u
co
s
5-
°^
t-1
o
o
T3
H-
rt
tsbur
CM
sr
M
O
O
s
?r
H-
CR
cr
c
l-<
OQ
VO
O
O
IT]
O
X
O
T3
(6
to
O
O
rt
I—1
H.
rt
03
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.l ug/L
Mean Hall concentration
when £0.1 ug/L
teximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
H
%
M
O vo
0 0
"N-* •
W
Q F °<
0 f 1
sow
" M
'"tf O 2
^ po o
bd E^
s *5 S
H ?d ^
X G
f M 5.
Q S O
g i> (sd
?3 H
ta M on
O f W
MOW
-3 g O
H g
[™IS
3 H ^
M >
f 0 H
M H M
I-1 f1
Ui C_i
11 p
i O ^
VO
C ^*J
OQ "JJ
^^
£rj
VO
oo
-------
TABLE 91. tf-BHC (LINDANE) AND/OR S-BHC FINISHED WATER DATA*
JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL = 0.15 ug/L
Utilityb
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mine0
Beaver Falls
Wheeling
Huntington
Cincinnati
Louisville
Evansville
Total or Mean
West View6
o
"O
0)
CJ
Ki
cd
0)
CO
cd
nc
M
01
a
H
11
10
11
13
20
12
24
16
15
11
143
12
i-4
00
•0 3
C
3 i-l
0 •
u-i O
I
cd
33
CO
0)
a
H
0
0
0
0
3
0
1
0
0
1
5
0
»J
00
•0 3
e
3 iH
O •
4-1 O
t— l
cd
JB
CO
01
a
H
0
0
0
0
5
1
4
2
1
2
15
0
c
0
•H
4->
4J 00
C 3
01
O «-l
c •
0 0
CC
co cd JC
(U
a
•H
H
0)
u
•tJ -^
01 00
g!3
p* CD
t/3 ^
a
H
O
4-» q ^
CO 0) 00
3 rC 3
5 w v
u ex v
r-4
4-1 Cd
U) cd DC
0)
s
ld
ld
c
01
s
TJ --
0) OO
c o
o v
O
cd
en 3:
0)
a
H
od
od
a = Clear well effluent
b = see Figure 1
c = Western Pennsylvania Water Co., Hays Mine Plant
d = Both compounds
e = Ground water supply
*CONCENTRATIONS NOT CORRECTED FOR EXTRACTION LOSSES.
187
-------
00
00
O
O II
S!
no
M l-j
30
HC
O ft)
CO fD
3^
O CO
OT3
O
a. o
n n
o
H-
O
H-
n>
n
u
n
H
H
i
f
O
CO
CO
M
CO
S3
ro
CO
H-
§
cr ft)
u n
S3 co
ro ro
co ro
rt
CD >rj
9 CK)
ro fD
9
CD
H
ft)
9
£
rt
fD
H
n
0
•
03
ft)
CD
^
H-
3
M
ft)
9
rt
s:
ro
co
rt
ro
M
o
H
H
O
rt
0)
)— *
O
i-l
fl>
ft)
9
j_j
00
H
ro
vo
0
M
O
VO
ro
O
CU
<
ft)
9
0)
H-
M
fD
P
U)
O
M
0
ro
o.
o
f
o
c
H-
CO
<
H-
M
(t
M
M
OJ
0
ro
O
N)
CL
0
n
H-
9
O
g
rt
H-
M
ro
«
O
ro
O
*
PB
9
rt
H-
9
TO
rt
0
9
ro
>
-
O
M
O
vo
:r
ro
ft!
H-
9
TO
C
O
I-1
bo
fi>
fi>
^
fD
i-l
ft)
H"
CO
oo
M
vo
O
ro
0
*
ro
ro
CD
ft
H-
0
t-1
0
ro
a
s:
T*
CO
S
H-
3
ro
cr
M
O
O
rt
CO
cr
TO
9"
"
O
-
jj----
9
CO
cr
C
i-f
TO
VO
O
O
Q
f-^
rr
•o
ro
K
o
o
G
'
rt
ft)
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £.0.1 ug/L
feximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall >0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
"imes MS confirmed when
Hall <0.1 ug/L
H
ttf
oS
n
^- VO
0 ^^
o t-1
s" K
C^ f^3 '"U
" ^^
^3S
hd 0 P
pa &d o
S " ^
S§i
H Jtf ->>.
MOO
O S ^co
3 H W
]xJ UJ U-l
o
O t~*
I?d o jjd
M M £J
o £d
Qtrrl h-^
3 H M
M W
«^ M t>
M 0 H
t-1 S! t>
II t-1 *
M
M f< P*
Ul ^
II
c t-
00 p ^0
|— 1 "*»J
1
C t_i
TO G
tr* M
I-"
VO
00
-------
n a4 pi
00
8
n o w s; co n
rt i-i o ro n> M
t* O rt M (D fi>
3 C P4 rt P
S 0 fD IT] H
Sgl^ll
g ft O CD (D M
W H B "*
O
o
o
3
W
•n
o
X
H
n
H
H
O
Cfi
en
w
CO
CO
CO
I
H-
Pi
rt
fD
t-t
O
O
I
CO
S
p
3
rt
CD
Hi
H,
M
i
rt
c
n>
01
rt
<
H.
I
M
N)
0
O
H
O
rt
H
M
0
M
?
(1)
P
H
*-
W
H
H
CO
N)
O
•
to
O
•
*-
(-•
&•
°0,
^
<&.
Evansville
M
(->
tO
CO
O
*
M
0
*
to
Louisville
M
Cn
o
*-
MP.
°0.
Cincinnati
M
ON
CO
*»
0
•
NJ
0
•
CO
Hunt ing ton
to
*»
en
^j
O
•
M
O
*
to
Wheeling
H1
to
0
K>
Beaver Falls
to
0
CO
00
o
•
to
o
•
*~
-a.
°^
WPW/Hays Mine°
M
CO
O
to
Pittsburgh
i-*
M
O
M
Wilkinsburg
i-1
o
o
i-^
•^
o
X
n
tf
T3
rc>
M
H
(-•
O
O
c
rt
H-
M
H-
rf
><
CT1
Times Hall searched for
Times Hall found
^0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ue/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
TABLE 93. HEPTACHLOR AND/ OR 3-BHC FINISHED3 WATER DATA,*
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVE]
GC/MS, APPROXIMATE LOWER DETECTION LEVEL = 0.
!-• IT1
Ln C_H
* " §
(W OK!
f M-M
C ^J
00 ^J
^a
w
M
VO
^J
00
-------
H-CL O CT1 03
O H II II II
OO O S3 co
M •"( IT* fD fD
SO H- CO fD
dS ° £ _
po 3 fD ^
t>cu pa i-i H-
H H- 3 OQ
MS;
a
CO
CO
M
CO
n>'
3
H-
EB
£
rt
ro
n
o
o
ffi
P
fu
3
rt
S2
fD
CO
rt
«-
fD
t!
CU
!_,
M
I-1
I—1
0
to
o
ro
H
O
rt
B>
M
O
H
S
3
00
M
VO
M
M
O
*•
ro
o
ro
0
«
<
P>
3
CD
H-
M
I-"
(t
|— i
M
ro
H
O
*•
0
M
O
r1
o
c
H-
5
H-
t-1
M
fD
M
^
U>
O
u.
0
Ui
M
o
o
H-
3
r>
3"
6
rt
H-
.
M
0>
O
O
Wl
M
U)
SB
C
3
rt
5-
OQ
n
o
3
M
O
ro
o
*>
ro
o
c
y
fD
fD
H-
3
OQ
ro
o
M
03
(D
Q)
fD
^
0)
M
co
oo
0
M
z:
fD
CO
rt
,.,
H-
fD
€
o
M
O
1-1
,.
^
S5
en
S
H-
3
fO
CT
M
O
O
^
j_i.
rt
rt
CO
CT
it
OQ
sr
M
o
M
a
P.
?r
H-
3
co
O"
c
H
OQ
VO
O
O
*T|
o
x
n
sr
a>
T3
fD
H-
1— '
ro
o
M
c:
f—i
H-
CD
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
o
o
"**Y
S t-3
O P >
nt-fe
SUM
CO M
- HVD
X O
M >>£d
2; t-5 M
t-3 ^d
?d O &3
X >>
S3 K S3
a H >
0 tH {fl
H |^
HJt>
D W *
Z Sc^
-1 o ^
< H «J
M SH
VO
1 S ^
M ^
-> tr1 S
Ln &
II W
-H i - ^J
' rr\
C
OQ
-------
x. &. o a" to
o H n n n
n o £3 co o
H H fD fD h-1
ZJ O CO fl> (T>
H g rt PJ
[> & K H-
H 3 OP Si
MS! c n>
fD fD
2| CO
*"O ^
f^ *"0 (U
O I- 3
pdV! H-
n
H
M
tfl
X
n
H
M
i
s
CO
00
w
rt
CO
i-f
cu
•
to
M
ro
0
M
O
H
&
n
»
H
O
11
*
0>
9
H
*>
W
ro
H
to
M
0
*-
I-1
to
u>
o
H«
O
Evansville
M
M
to
K>
O
to
O
W
Louisville
i-1
i*
*>
ro
O
f-
O
00
M
o
Cincinnati
M
o>
~j
*-
0
4>-
O
VO
to
O
Huntington
ro
4»
00
vo
0
*>
i-1
to
Wheeling
\->
ro
0
to
Beaver Falls
to
o
o
M
t-1
o
WPW/Hays Mine0
M
OJ
O
M
Pittsburgh
M
I-"
O
O
Wilkinsburg
M
O
0
0
•n
o
X
n
cu
•o
m
M
t->
I-1
o
o
c
n
H-
I—1
H-
n
^<
cr
Times Hall searched for
Times Hall found
^0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
«vVkAT> ^O 1 11O/T
wnen ^u.x ug/jj
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ueA,
Times MS confirmed when
Hall ^0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
o
"g
£$
ot^w
o tr1
~~- vo
- H
W
^^^
i-d OO
fO ^ fti
O- M
%t*
g ^ hrj
> h3 H
H ?dZ
ISHEDa
OXIMATE
E LOWER
o t-1 >
M O H
rt S^ M
M M !*3
S^«
M O >
O M H
2! H >
W»
f n *
33C-,
M Q Cl
^a^
11 KM
0 < VO
. M »-J
H f ^J
Ln 1
II t-i
|?1
C ^
is
-------
I
D- o cr &>
n n M n
o o s; CD
H 3* fl> (D
O H- CO (B
O rt
(B
' -j H
f> &•
H <*
§ft
C/3 fB
S!
pd 3 H-
H- 0 TO
Oi 3
rt CD
CO P
O H-1 rt 0
^ <£
n
H
O
W
O
o
f
o
a>
ai
M
Crt
ft>
rt
N
V
ffi
CO
I
(B
OJ
3
rt
rt)
CO
rt
H-
(B
H
O
O
H
O
rt
P>
M
O
i-(
S
(B
P)
M
00
M
M
O
to
p
Evansville
M
0
O
r
o
c
H-
co
H-
H
(B
. H
M
0
O
to
p
to
Cincinnati
M
o
o
Hun ting ton
*"
o
o
Wheeling
to
o
o
Beaver Falls
00
o
o
fB
CO
rt
H-
(B
n
M
o
o
WPW/Hays Mine
M
o
o
Pittsburgh
i— '
o
-
sc
3
CD
cr
c
00
^
o
o
o
X
1
(B
to
o
o
c
M
rt
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £.0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1ug/L
Times MS confirmed when
Hall <0.1 ug/L
n f
n t-1 ?^
S a
?*• O 1-3
^ 0 >
Pd pd cd
O - p
X o
POXIDE
ROXIMAT]
TE LOWE
72 PJ td
O tr1 s;
MO
H s: s:
M M J>
ss^
3^^
f n H
M HJ>
^ O *
t-i
ii f d
M f1 H
Ln vo
II ~J
0 0 1
C W
cw
tr1 ^o
OO
-------
TABLE 97. HEPTACHLOR EPOXIDE FINISHED WATER DATA,* JULY 1977-JUNE
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.15 ug/L
1978
b
Utility
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mine°
Beaver Falls
Wheeling
Huntington
Cincinnati
Louisville
Evansville
Total or Mean
West Viewd
o
<4-l
•o
0*
o
CO
01
00
j
CO
00
oi
6
H
11
10
11
13
20
12
23
16
15
11
142
12
-i
«iv'
oc
T3 3
C
3 .-1
o •
*4-4 O
rH ^
( i
0)
ac
(0
01
B
H
0
0
0
0
0
0
0
0
0
0
0
0
I-J
00
•o 3
c
o •
•4-1 O
1
CO
ffi
00
QJ
_Ej
H
0
0
0
0
0
1
0
1
2
1
5
0
c
0
•H
4-1
CO tJ
4-1 00
C 3
HI
O f-H
go
p~i
•-I C
CO 0)
' ^
C
3
j
— ^
00
3
M
O
•H
4J
(0
(0 P
33 0)
a
1 §
•H
X
cfl
c
o
•H |J
4-* C ~*^
(0 Ol 0(
M !3
•H tH
M-i T3
C O) O
o 4->A\
CJ O.
e
CO Ol i-H
S: 4-> r-l
4-1 CO
U) CO 33
01
e
•H
H
.11 1 .•!• 1 II
C
01
tC
3
h-3
T3 *~>
01 OC
B 3
t-4
•H r-\
c o
OAX
c/3 i— I
"cO
(fl pd
0)
.§
H
C
0
H h- J
^J d '"*•»»
CO Ol OO
M 3
14-1 -a
C 0) O
O 4-1 W
o a. v
E
en 01 1-1
4-1 CO
en co nc
Ol
S
•H
H
C
Ol
^
3
I— 3
'O "*^-
d) OO
B 3
W
C 0
8 v
cy. i-i
co
w n;
01
B
•H
H
a = Clear well effluent
b = see Figure 1
c - Western Pennsylvania Water Co., Hays Mine Plant
d = Ground water supply
*CONCENTRATIONS NOT CORRECTED FOR EXTRACTION LOSSES,
193
-------
!•
o
P3
2; o
n
H
M
o
§
w
a
9
H
i
n n n
O £3 co
D* ro ro
H- co ro
O rt
ro *TJ
pd ft H-
H- 3 oo
< c
ro ""d t^
i-i ro ro
3
rt co
S*
CO fu
rt 3
H-
< pj
ST.!
s! PJ
rt
ro
o
o
6?
•
en
rt
<
H-
0>
*h.
M
H>
0
M
H
O
ft
0>
M
O
d
3S
(I>
(>>
3
H
UJ
00
Ul
K>
VO
O
•
**
O
•
VO
N>
0
Evansville
t-1
H
O
N>
Louisville
M
M
O
ISJ
Cincinnati
M
M
0
U>
Hun ting ton
N>
M
(-•
00
O
•
N>
O
•
N>
Wheeling
M
K3
H
N>
O
•
M
0
•
M
Beaver Falls
H
00
N)
O\
O
•
0>
0
•
VO
M
0
01
n
<
H-
(D
n
M
H
M
O
O
•
ro
o
•
N5
M
o
WPW/Hays Mine
M
h->
0
S3
Pittsburgh
M
H1
0
N3
Wilkinsburg
VO
0
o
•n
o
X
n
03
T>
ro
H-
M
N>
O
N3
c
rt
K-
M
H-
rf
V!
(U
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^.0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall >0.1 ug/L
Times MS confirmed when
Hall £.0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
TABLE 98. a-ENDOSULFAN RAW WATER DATA,* JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.15 ug/L
-------
Oi
«. P-» O U"1 P3
o n n n ii
(—\ o 5} cn o
W n ro ro M
2 o cn ro ro
H § ro trj S
S &. n H-
f-i ^ 3 era g
MS! c ro
2 rt ro ro M
n
H
M
O
f«
w
o
H
H
i
cn
•
01
rt
<
H-
L
M
to
0
O
H
O
rt
»
(-•
O
i-<
•8?
0>
3
I-1
*-
to
co
ro
M
O
•
IsJ
O
•
to
I-"
o
Evansville
M
M
!-•
O
O
•
M
0
•
I-"
Louisville
i-»
cn
H1
CO
O
•
to
0
*
10
M
O
Cincinnati
M
(T>
0
NJ
Hunt ing ton
to
Co
(->
ON
O
•
ro
o
•
to
Wheeling
t->
to
0
CO
Beaver Falls
to
0
o
*-
WPW/Hays Mine0
M
CO
O
t-1
Pittsburgh
i-1
h-1
O
10
Wilkinsburg
M
O
O
O
I?
X
n
m
•a
n>
t-1
M
l->
O
O
c
rt
H-
I—1
H-
rt
v;
cr
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall ^0.1 ue/L
Times MS confirmed when
Hall ^0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
TABLE 99. a-ENDOSULFAN FINISHED3 WATER DAW
GC/HALL DETECTOR, APPROXIMATE LOWER DETECT
GC/MS, APPROXIMATE LOWER DETECTION LI
"
TO 0 1
t*M|j
C W
^M
f vo
^J
oo
-------
iff
23 °
H 6
<£-
S*!
o 9>
15
q
M
H
M
i
O
O)
M
en
a = see Figure 1
b = Western Pennsylvania Water Co., Hays Mine Pla
c = Ohio River at West View
o
rt
£
ft
en
rt
<
H-
ft
«k
H
M
0
Q
H
0
rt
0)
I-1
O
n
3
fD
(U
0
M
OJ
oo
o
hO
Evansville
H>
M
O
O
Louisville
M
M
O
0
Cincinnati
i-"
M
O
O
Hun ting ton
N)
M
O
N3
Wheeling
M
to
O
O
Beaver Falls
H
oo
o
o
fD
CD
ft
<
H-
ft
0
I-1
M
O
0
WPW/Hays Mineb
i-1
M
O
O
Pittsburgh
M
(-1
O
O
Wilkinsburg
VO
o
o
TJ
0
X
n
g"
•o
ft
t-1
M
Isi
O
O
C
rt
H-
M
H-
rt
><
&)
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £.0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
0
n
EC
O Ir1 H
^&
Sdt1'
en M M
' SM
%3§
HC) o •
g.W
^fe«
f^H
RAW WATER DATA,* JULY 19 7 7- JUNE 1978
ROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
TE LOWER DETECTION LEVEL = 0.15 ue/L
-------
* &. n a4 fu
o
O II II II ||
&
o cr> 53 CD n
w H rt> (D M
3 O CO (D (D
H C rt fo
5d 3 (D hcj hi
?> QJ H '-'•
H rf
§rt
. . (D (D
W 3
N 3 H
2 co
O CO ^
H C M
O *"U Q}
9 "-1 °
?d *< H'
Q ^
H (!)
M rt
a n>
M
a
H
i
en
W
n
o
S
S'
m
S
§
rt
c
CO
rt
H-
tt>
H1
to
O
O
rt
CD
O
H
I
M
JS-
to
H
«
O
to
p
to
H
O
Evansville
M
O
M
Louisville
t->
0
-
Cincinnati
s^
-
0
0
Is)
O
to
-
O
Huntington
to
u>
o
o
Wheeling
to
o
o
Beaver Falls
to
o
o
o
WPW/Hays Minec
u>
o
-
Pittsburgh
i-1
o
o
Wilkinsburg
i-1
o
o
o
o3
X
n
rr
T3
-
O
O
rt
rt
CT1
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ue/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
TABLE 101. DDT FINISHED3 WATER DATA,* J
GC/HALL DETECTOR, APPROXIMATE LOWER DETECT!
GC/MS, APPROXIMATE LOWER DETECTION LE^
M O ^
" ^ M
h- * f* i
01 C-j
OQ O 2
M H
C! ^O
W3 ^
-------
oo
I
o
2
O
O
H
M
e>
»^j
o
fri O CT" (U
II II II II
O O S3 co
t-< CJ" fD (D
O H- CO ft)
C O rt
3 fD ^
a. fa i-t H'
H- 3 OQ
c < e
B3 fD fti t-i
rt hi fD fD
fD <
CO (B
rt 3
s! to
rt
fD
H.
n
o
en
en
fD
I-1
rt
fD
Cfi
rt
H-
fD
M
O
0
H
O
It
0)
H*
O
3
00
o
I-1
Evansville
H
O
O
Louisville
p
o
o
Cincinnati
H1
O
O
Hun ting ton
tsJ
o
-
Wheeling
K
o
0
00
n>
n
0)
en
M
00
o
o
£3
n>
co
rt
H-
0)
P
O
o
S3
ff"
CO
s
H-
"a"
P
o
0
Pittsburgh
(-1
o
o
Wilkinsburg
VO
o
0
X
n
•a
fD
H-
to
o
0
C
rt
H-
M
H-
rt
01
Times Hall searched for
Tiroes Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £.0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
TABLE 102. DIELDRIN AND/OR DDE RAW WATER DATA,* JULY 19 77- JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.15 tis/T,
-------
TABLE 103. DIELDRIN AND/OR DDE FINISHED WATER DATA,* JULY 1977-JUNE
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.15 ug/L
1978
Utilityb
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Minec
Beaver Falls
Wheeling
Hunt ing ton
Cincinnati
Louisville
Evansville
Total or Mean
West V±ewd
—
*o
U
H
cO
(U
CO
,—1
r- 4
CO
EC
CO
CU
H
11
10
11
13
20
12
23
16
15
11
142
12
-i
^^
00
^J £j
e
3 i-H
O •
IH O
,— 1
CO
PC
(0
Q
H
0
0
0
0
0
0
0
0
0
0
0
0
_q
00
T3 9
B
3 «H
VH O
r-H
CO
»
co
A\
r~t
iH C
CO Hi
c >
CO
£
|J
*Xx*
W)
p
A
c
o
•H
4J
CO
•H w
CO C
33 cu
CJ
1 §
a °
«
s
o
o -,
•H H
4-1 C "->•
CO CU W
E "» °
•H l-t
14-1 -O .
C r-H
4-1 CO
CO CO PC
CU
£3
H
C
cu
|£
)_4
ro **^
0) 00
6 3
•H r-l
C 0
OA\
^5
CU
fi
H
C
O
H jj|
4J C ---.
(0 CU 00
ef 3
>
c i-t
4J CO
CO CO PC
cu
B
•H
H
c
cu
x:
|jC
uj
*X3 ^^^
-------
O
O
o
o
n
O
M
3
M
1 J
sz
n
M
o
f
o
CO
CO
M
CO
D-t O O"1 fD
II II II II
o o s! co
o H- co ro
C o rt
A* pd i-i p.
H- P 00
Pi (T> hd i*f
rt h{ fD fD
fD P
i-i (B P M
CO V!
T) co Co
h-1 rt 3
la?
rt
fD
4
O
Q
•
**
W
03
CD
P'
fD
M
rt
g
fD
CO
rt
H-
fl>
*b.
M
1-1
o
o
H
O
rt
03
0
s
3
1 ,
00
00
0
M
i
M
^
Co
P
CO
H-
H
H-"
ft)
M
M
O
O
,_,
Q
c
H-
§
H-
H
M
fD
t-i
M
0
0
n
H*
P
n
H-
p
to
rt
p.
M
M
O
O
PC
c
p
rt
5-
00
rt
O
p
N)
M
O
-
5,
3-
ft!
H-
00
(_J
ro
O
O
to
n>
^
fD
P)
H1
CO
(_1
oo
o
0
s:
fD
CO
H-
fD
C
n
i_j
1-1
0
o
•c
s:
Co
CO
f-i.
a4
(->
l-»
O
o
•n
H-
rt
rt
CO
cr
if
00
13"
j_,
M
O
O
y.
H.
(-"
^
H-
P
CO
cr
c
i-j
oo
VD
O
O
^
o
X
n
g"
•o
fD
t-1
.,
N3
O
O
C
H-
rt
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
o
o
EC i-3
O. tr1 W
n t-1 t-1
CO H |— >
» HO
> 0 f"
^d O
t-g \-£ HH
O " tZS
X O
IT1 M i
o r*^
S H >
?d W H
O f fd
M o
M G »
O M >J-
•< M Kj
a o
n t-1 *j
o ^ ^
II M
1-1 cxi
00
-------
TABLE 105. ENDRIN FINISHED3 WATER DATA,* JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/£
GC/MS, APPROXIMATE LOWER DETECTION LEVEL = 0.15 ug/L
Utility13
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mine0
Beaver Falls
Wheeling
Hunt ing ton
Cincinnati
Louisville
Evansville
Total or Mean
West Viewd
T
0
o
)»)
cd
01
CO
, — |
CO
CA
CU
H
11
10
11
13
20
12
23
16
15
11
142
12
-i
*j
00
T3 3
3 _j
o •
>4-i O
,—(
CO
CO
CU
H
0
0
0
0
0
0
0
0
0
0
0
0
T
^~*
00
TJ 3
3 r— I
O •
>4-4 O
i-H
CO
a
(A
Ol
H
0
0
0
0
0
0
0
0
0
0
0
0
a
o
•H
CO H-l
4-1 00
C 3
0)
O rH
c •
0 0
o A\
rH
iH C
C
x
J
00
3
C
0
•H
4->
rH 1-1
rH 4-1
CO C
a
-------
o
S3
a = see Figure 1
b = Western Pennsylvania Water Co., Hays Mine Plant
c = Ohio River at West View
d = Ground water supply
CONCENTRATIONS NOT CORRECTED FOR EXTRACTION LOSSES.
S3
(D
09
rt
<
H-
ID
V
I-J
H
O
O
H
O
rt
03
M
O
H
3
n>
5
M
u>
00
o
o
r
Evansville
M
(-•
O
O
h
Louisville
H
M
0
O
Cincinnati
H1
M
O
O
Hun ting ton
N>
H
O
O
Wheeling
H*
NJ
0
o
Beaver Falls
M
00
o
o
(t>
CO
rf
<
H-
ID
«
O
M
I-1
0
O
WPW/Hays Mineb
M
M
O
0
Pittsburgh
M
M
O
O
Wilkinsburg
VO
0
O
^
o
X
o
=r
03
•a
(D
M
(-•
ro
o
o
G
rt
H-
(-<
H-
rt
vi
0)
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when ^.0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall >0.1 ug/L
Times MS confirmed when
Hall £.0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
i
OM
gS
oPr
n f
sao
co MO
" ^°
> r>h>
^ H§
^ O t)
/OR 3-ENDOSULFAN RAW WATER DATA,* JULY 1977- JUNE 1978
R, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/Ik
ROXIMATE LOWER DETECTION LEVEL =0.15 ue/L
-------
TABLE 107. DDD AND/OR 3-ENDOSULFAN FINISHED WATER DATA,* JULY 1977-JUNE 1978
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
GC/MS, APPROXIMATE LOWER DETECTION LEVEL =0.15 ug/L
v,
Utility0
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Minec
Beaver Falls
Wheeling
Hunt ing ton
Cincinnati
Louisville
Evansville
Total or Mean
West Viewd
—
ft
U
l-t
co
0)
CO
, — 1
, — 1
cfl
05
0)
a
H
11
10
11
13
20
12
23
16
15
11
142
12
.
^-4
60
TO 3
C
3 r-H
O •
14-1 O
i f\ *
,— |
to
DC
CO
0)
H
0
0
0
0
0
0
0
0
0
0
0
0
•1
i,
60
•0 3
e
3 i-H
O •
14-4 O
1 "
r- 4
tfl
EC
to
0)
.§
H
0
0
0
0
0
0
0
0
0
0
0
0
c
o
•H
(0 t-3
4-1 60
g a
O r-t
C •
0 0
O A\
t— |
rH C
CO (U
C
X
5
«»^
60
3
gt
C
O
•H
4-1
(0
iH l-i
rH 4-1
PC <1)
U
§ §
•1 °
X
£
c
o
•H >-4
4J C ->.
(0 0) W
a-§a
14-1 T3 .
C 0) 0
O 4-1 y^t
{j Q-
0
C/J 0) rH
S 4-t r-t
4-1 CO
to CO EC
0)
H
C
0)
A
3
n4
•O ~-~
0) 60
e s
c o
o /Cv
CJ
CO rH
CO
(0 (C
(U
H
cs
O
H 4-3
co at 60
e j= 3
C s
U-* 'D •
C <1> O
C
01
.£
3
T3 •*-
01 60
£3 3
C 0
O 4-> W O V
OP.
^j *~!
4-1 CO
CO CO EC
0)
H
O
C/} t-H
CO
to EC
ai
B
•H
H
a = Clear well effluent
b = see Figure 1
c = Western Pennsylvania Water Co., Hays Mine Plant
d = Ground water supply
CONCENTRATIONS NOT CORRECTED FOR EXTRACTION LOSSES.
203
-------
o cr1
II II
H- 3 Qt}
< C
fD !-d H
i-i 0>
3
s:
n>
01
rt
<
H-
ft)
S
M
M
O
O
H
O
n
K
l->
O
11
3K
n>
PI
3
M
ui
-«j
o
o
Evansville
M
M
O
O
Louisville
H
M
O
O
Cincinnati
M
M
o
o
Hun ting ton
NJ
M
0
O
Wheeling
M
N5
O
O
Beaver Falls
M
00
O
O
s
ft
U)
rr
' <
H-
ft)
n
M
M
o
o
WPW/Hays Mineb
M
M
O
O
Pittsburgh
H
I-"
O
0
Wilkinsburg
<£>
o
o
T]
O
X
n
cu
"O
ft
i— •
M
M
O
O
C
rr
H-
h-1
H-
rt
v;
W
Times Hall searched for
Times Hall found
£0.1 ug/L
Times Hall found
<0.1 ug/L
Mean Hall concentration
when £0.1 ug/L
Maximum Hall
concentration, ug/L
Times MS confirmation
attempted when
Hall £0.1 ug/L
Times MS confirmed when
Hall £0.1 ug/L
Times MS confirmation
attempted when
Hall <0.1 ug/L
Times MS confirmed when
Hall <0.1 ug/L
TABLE 108. METHOXYCHLOR RAW WATER DATA,* .
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTIOI
GC/MS, APPROXIMATE LOWER DETECTION LE1
•=H •& <— 1
M c!
tr1 f f
M Kj
M l-i
o f ^o
•~J
M II -^1
Ui 1
O CH
Q M 3
-•* 1 M
f 0
• M
N> VO
~»J
C 00
00
t-""
-------
TABLE 109. METHOXYCHLOR FINISHED WATER DATA,* JULY
GC/HALL DETECTOR, APPROXIMATE LOWER DETECTION LEVEL
GC/MS, APPROXIMATE LOWER DETECTION LEVEL = 0
1977-JWE 1978
= 0.1-0.2 ug/L
.15 ug/L
Utilityb
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mine0
Beaver Falls
Wheeling
Hunting ton
Cincinnati
Louisville
Evansville
Total or Mean
West Viewd
o
T3
01
f.
O
01
CO
rH
CO
EC
CO
01
H
10
10
11
13
20
11
23
15
14
11
138
12
^
00
*U »3
c
3 i-H
0 •
i*-i O
•1 ^
cfl
CO
O)
H
0
0
0
0
0
0
0
0
0
0
0
0
i-4
00
•0 3
c
0 •
*l | r]
. v
'co
X
en
01
0
0
0
0
0
0
0
0
0
0
0
0
c
o
•H
4J
CO >J
4-1 00
C 3
01
O rH
§0
AV
rH
rH q
CO 01
CO
01
S3
00
3
O
•H
CO
1 — 1 \ *
CO C
EC 01
O
1 §
a u
K
1
O
•rH 1-J
4-» q — *
CO 0) Ol
>4-< T3 «
C 01 O
O 4-1 f\
0 O.
C/3 0> rH
4J Ct
en co EC
01
a
H
01
3
T3 -^
01 00
B 3
•H rH
c o
o
W rH
CO
CO EC
0)
a
H
.
O
4J q ^
CO 01 00
3 *§ =>
H rH
5 *-> v
o a.
W § rH
4-4 CO
CO cfl EC
1J
H
G
0)
3
T3 — -
0) OO
e 3
•H rH
q o
o v
o
C/j rH
CO
10 EC
01
a
H
a = Clear well effluent
b = see Figure 1
c = Western Pennsylvania Water Co., Hays Mine Plant
d = Ground water supply
CONCENTRATIONS NOT CORRECTED FOR EXTRACTION LOSSES,
205
-------
SURVEY FOR BASE-NEUTRAL EXTRACTABLE NON-HALOGENATED HYDROCARBONS
Analyses were conducted on raw and finished sample extracts by GC/flame
ionization detector (GC/FID) and by GC/MS for the non-halogenated extractable
hydrocarbons listed in Table 6. These compounds can be generally grouped as
phthalate esters and polyaromatic hydrocarbons (PAH). Approximate lower de-
tection levels by GC/FID varied for these compounds from 0.5 ug/L to 10 ug/L;
lower detection levels by GC/MS-SIM were 0.1 ug/L.
Implementation of a rigorous quality assurance program, as detailed in
Section 5, was necessary after interferences were noted in data produced for
these compounds from the first four months of sampling and analysis (July
through October 1977). The quality control program included a solvent group
concept whereby two solvent blanks were extracted, concentrated and analyzed
with each group of four project samples. Interferences were controlled and
all data from November 1977 through June 1978 were statistically corrected.
Data from the earlier period were discarded.
Phthalates (Quality assurance data: Tables F-l to F-3.)
GC/FID chromatograms of solvent blanks and sample extracts generally con-
tained responses presumptively identified as phthalate compounds at concentra-
tions at and below the approximate lower detection levels (routine lower quan-
tification levels of 0.5 ug/L to 5.0 ug/L depending on the compound). GC/MS-
SIM confirmed the presumptive identifications of these interferences in sol-
vent blanks as phthalates. Concentrations of these contaminants reported in
solvent blanks by GC/FID were statistically handled and used in the correction
of all sample data. A single compound, bis(2-ethylhexyl) phthalate, that co-
eluted with 1,2-benzanthracene and/or chrysene, was found in solvent blanks
and field extracts well in excess of the approximate lower detection level of
1 ug/L. Statistical corrections at a 95% confidence level of 1.4 ug/L to 4.4
ug/L were applied to sample data for this compound. A few sample extracts
contained bis(2-ethylhexyl) phthalate in excess of the statistical correction
but these reports were questioned because of the random nature of the contam-
ination. The other phthalate compounds were not detected in sample extracts
at concentrations exceeding statistical corrections.
Field extracts did not contain dimethyl phthalate above 5.0 ug/L, diethyl
phthalate above 2.0 ug/L, di-n-butyl phthalate above 0.5 ug/L, or butyl benzyl
phthalate above 2.0 ug/L. Extreme variability of extraction recovery data
prevented their application to field extracts to suggest concentrations above
which these phthalates were not likely present in field waters. Because of
the random nature of bis(2-ethylhexyl) phthalate contamination and the
extreme variability of its extraction recovery data, this phthalate could not
be evaluated in field waters.
Polyaromatic Hydrocarbons (Field data: Tables 110 to 114. Quality assurance
data: Tables F-l to F-3.)
PAH compounds were generally not found in samples collected from November
1977 through June 1978 at concentrations exceeding approximate GC/FID lower
detection levels (0.5 ug/L to 10 ug/L depending on the compound). However,
206
-------
numerous low level responses were apparent at PAH retention times in GC/FID
chromatograms from most utility locations, particularly in the winter months
of 1977-78. Initial GC/MS-SIM analyses of a few such selected raw and fin-
ished extracts confirmed the presence of some of the PAH compounds at 0.1
ug/L or greater. Further GC/MS-SIM evaluations were then undertaken to quali-
tatively define PAH compounds at levels >0.1 ug/L in extracts of raw and fin-
ished water samples from each utility. These evaluations were generally done
on a one-time basis for each utility. Extracts from several GAG influent and
effluent sequences were also evaluated.
The GC/MS-SIM qualitative results of those evaluations for PAH compounds
are presented in Tables 110 through 114. Positive confirmations of the com-
pounds were based on their presence at 0.1 ug/L or greater in sample extracts.
Solvent blanks were also analyzed by GC/MS-SIM and did not contain responses
for any of the PAH compounds, nor did chromatograms produced by GC/FID analy-
sis of solvent blanks.
Tables 110 and 111 present data for utilities located on the Ohio,
Allegheny, Monongahela and Beaver Rivers and for West View's ground water.
The importance of these data is that they indicate the confirmed presence of
some of the PAH compounds in raw and finished waters of the utilities at con-
centrations equal to and in excess of 0.1 ug/L. It is important to note that
the effect of treatment cannot be evaluated on the basis of a single sample
sequence, particularly for a single compound, because the data are qualita-
tive, quality assurance data suggest highly variable extraction recoveries,
and identifications are just above the lower detection level for these com-
pounds by GC/MS-SIM.
The data also indicate the absence of eight other PAH compounds in ex-
tracts from several utility finished waters. Additional GC/MS-SIM analysis of
these seven compounds was not undertaken because positive confirmations were
not indicated in initial attempts.
Two sample sequences from the Wheeling Water Department were GC/MS-SIM
analyzed, the first sequence collected in the winter season, the second col-
lected in early summer. GC/FID analyses of those sequences had produced
visually different chromatograms. Low level responses were apparent in the
chromatograms of February raw and finished extracts but were not observed in
the chromatograms from samples collected in June. A difference in the number
of PAH compounds present in winter and early summer was also supported by the
MS data as presented in Table 111.
A significant qualitative difference in raw and finished waters was
suggested by GC/MS-SIM analysis of extracts from utilities where treatment
included GAG filtration/adsorption. These data are presented in Tables 112 to
114. At the Western Pennsylvania Water Company, seven or eight PAH compounds
were present in raw water extracts at or above 0.1 ug/L in two sequences eval-
uated. With the exception of naphthalene, the compounds were not present at
0.1 ug/L in the associated finished water extracts. The finished water was
representative of treatment including GAC filtration/adsorption (Table 112.)
PAH compounds present in the extracts of raw waters and of GAC influent
waters at or above 0.1 ug/L appeared to be removed by GAC filtration/adsorp-
207
-------
tion at the Huntington Water Corporation and at the Beaver Falls Authority
(Tables 113 and 114, respectively). Removal appeared to be more effective
with some GACs than with others. In addition to the qualitative nature of the
MS data and the variability of extraction recoveries, the GAG type, age and
hydraulics should be considered in interpretation of the data.
ry -1
In research done by others in January 1977, raw and finished water
samples from the Western Pennsylvania Water Company Hays Mine Plant (WPW) and
the Huntington Water Corporation were analyzed for six PAH compounds. At WPW,
a total concentration of 0.6 ug/L for the compounds evaluated was reported for
the raw water, including the reported presence of 0.4 ug/L of fluoranthene.
The total concentration of PAH compounds reported in the finished water was
0.003 ug/L, a concentration well below the 0.1 ug/L for GC/MS lower detection
levels of PAHs reported in Table 112. At the Huntington Water Corporation,
however, the total concentration reported for the raw water was 0.06 ug/L;
that reported for the finished water was 0.007 ug/L. Both concentrations were
below the detection level at which PAH compounds were confirmed by project
data in 1978.
All qualitative data presented for project utilities are based on the
presence at >0.1 ug/L of some or all of a group of seven to eight PAH com-
pounds in the extract of a field sample. An extract containing six PAH com-
pounds at a concentration of >0.1 ug/L (Table 111) would contain a total con-
centration for those compounds of >0.6 ug/L. While the relationship of the
concentration in the extract to that present in the field sample is not
defined because of variable extraction recoveries (Table F-l), it is very
likely that concentrations were higher in the field samples. The World
Health Organization has recommended22 that the concentration of six represen-
tative PAH compounds be limited to 0.2 ug/L in treated surface waters. One of
the six representative compounds was fluoranthene, a PAH confirmed in project
extracts.
Because these GC/MS-SIM data are generally based on a single sequence at
each project utility, they should be considered as initial findings. It is
apparent, however, that some PAH compounds were present during the winter
months of 1977-78 in raw and finished waters. Some GAG filter/adsorbers
appeared to be effective in their removal. Research into the presence and
significance of polynuclear aromatic hydrocarbons in drinking water is
required. (Text continues on page 214.)
208
-------
O
VO
TABLE 110. MASS SPECTROMETRY-SELECTED ION MONITORING (SIM) CONFIRMATION'
OF BASE-NEUTRAL EXTRACTED NON-HALOGENATED PRIORITY POLLUTANTS
N. Utility
N. Date
Compound N^ Water
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene and/or Anthracene
Fluoranthene
Pyrene
1 , 2-Benzanthracene
and/or Chrysene
3, 4-Benzof luoranthene and/or
11,12-Benzofluoranthene
Benzo(a)pyrene
Indeno(l,2:C,D)pyrene
1,2:5, 6-Dibenzanthracene
and/or 1,12-Benzoperylene
Fox Chapel
1-31-78
R
+
-
-
-
-
-
-
F
+
-
-
-
-
-
-
Wilkinsburg
2-15-78
R
-
-
-
-
-
-
-
F
-
-
-
-
-
-
-
Pittsburgh
1-23-78
R
TR
TR
-
+
+
+
+
F
+
-
-
+
+
+
+
-
-
-
-
-
Beaver Falls! West Viewb
3-28-78
R
-
-
+
+
+
+
-f
F
+
+
+
+
+
+
+
-
-
-
-
-
6-1-78
R
-
-
+
-
+
+
+
F
-
+
+
-
+
-
+
detection level approximately 0.1 ug/L
+ = present £0.1 ug/L in extracted concentrate of sample
- = not detected£0.1 ug/L in extracted concentrate of sample
TR = trace
Ground water supply
R = raw
F = finished
-------
TABLE 111. MASS SPECTROMETRY-SELECTED ION MONITORING (SIM) CONFIRMATION0
OF BASE NEUTRAL-EXTRACTED NON-HALOGENATED PRIORITY POLLUTANTS
to
M
O
N. Utility
NV Date
Compound >v Water
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene and/or Anthracene
Fluoranthene
Pyrene
1 , 2-Benzanthracene
and/or Chrysene
3,4-Benzofluoranthene and/or
11 ,12-Benzof luoranthene
Benzo(a)pyrene
Indeno(l,2:C,D)pyrene
1,2:5, 6-Dibenzanthracene
and/or 1,12-Benzoperylene
Wheeling
2-21-78
R
-
+
+
+
+
+
+
F
+
+
+
+
+
+
+
-
-
-
-
-
6-21-78
R
+
-
-
-
-
+
+
F
-
-
-
-
+
+
+
Cincinnati
2-13-78
R
+
+
+
+
+
+
+
F
+
-
+
+
+
+
+
-
-
-
-
-
Louisville
12-13-77
R
+
-
-
-
+
+
+
F
+
-
-
-
-
-
-
Evansville
2-15-78
R
_
-
-
-
+
+
+
F
+
-
-
+
+
+
+
-
-
-
-
-
detection level approximately 0.1 ug/L
+ = present ^0.1 ug/L in extracted concentrate of sample
- = not detected^0.1 ug/L in extracted concentrate of sample
TR = trace
Ground water supply
R = raw
F = finished
-------
TABLE 112. MASS SPECTROMETRY-SELECTED ION MONITORING (SIM) CONFIRMATION OF BASE-NEUTRAL
EXTRACTABLE NON-HALOGENATED PRIORITY POLLUTANTS, WESTERN PENNSYLVANIA WATER COMPANY
N. Date
Compound N^ Water
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene and/or Anthracene
Fluoranthene
Pyrene
February 14, 1978
Raw
+
+
+
+
+
+
+
Finishedb»c
+
-
-
-
-
-
-
April 12, 1978
Raw
+
+
+
+
+
+
+
Finished13 >d
+
-
-
-
-
-
-
MS-SIM detection level approximately 0.1 ug/L
+ = present ^0.1 ug/L in sample extract
,- = not detected ^0.1 ug/L in sample extract
Treatment includes filtration/adsorption
GAG = Filtrasorb 400
Approximate loading rate =2.3 m/hr (1.0 gpm/ft^)
Approximate EBCT = 18 minutes
Depth = 76 cm (30 inch) GAG
,GAC age =26 months
GAG age =28 months
-------
ro
M
to
TABLE 113. MASS SPECTROMETRY-SELECTED ION MONITORING (SIM) CONFIRMATION OF BASE-NEUTRAL
EXTRACTABLE NON-HALOGENATED PRIORITY POLLUTANTS, HUNTINGTON WATER CORPORATION
N. Date
Compound ^\ Water
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene and/or Anthracene
Fluor an thene
Pyrene
February 14, 1978
R
+
+
+ .
+
+
+
+
GACb
Inf
+
+
-
+
+
+
+
Effc
-
-
-
-
-
-
-
Effd
+
-
-
-
-
-
-
Effe
-f
-
-
-
-
-
-
F
+
-
-
-
+
+
+
March 14, 1978
R
+
'. -
-
-
+
+
+
GACb
Inf
+
-
-
-
+
+
+
Efff
-
-
-
-
_
_
-
F
-
-
-
-
+
+
+
detection level approximately 0.1 ug/L
+ = present ^0.1 ug/L in sample extract
- = not detected ^.0.1 ug/L in sample extract
GAG = WVW 14x40
Depth = 76 cm (30 inch) GAC
Approximate loading rate = 6.1 m/hr (2.6 gpm/ft^)
Approximate EBCT = 7.2 minutes
,GAC age = 8 months
GAC age =16 months
fGAC age = 34 months
GAC age = 9 months
GAC = granular activated carbon
R = raw
Inf = influent
Eff = effluent
F = finished
-------
TABLE 114. MASS SPECTROMETRY-SELECTED ION MONITORING (SIM) CONFIRMATION OF BASE-NEUTRAL
EXTRACTABLE NON-HALOGENATED PRIORITY POLLUTANTS, BEAVER FALLS AUTHORITY
to
I-1
\ Date
\
Compound \Water
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene and /or Anthracene
Fluoranthene
Pyrene
January 2,, 1978
GACb>c
Inf
+
+
-
+
+
+
+
F400
Eff
-
-
-
-
-
-
-
FC
Eff
+
-
-
-
-
+
-
ICI
Eff
+
+
-
+
+
+
+
January 18, 1978
GACb»d
Inf
+
-
-
+
+
+
-
F400
Eff
+
-
-
-
-
-
-
FC
Eff
+
-
-
+
-
-
-
r ici
Eff
+
+
-
+
+
-
-
pIS-SIM detection level approximately 0.1 ug/L
+ = present ^0.1 ug/L in sample extract
- = not detected ^0.1 ug/L in sample extract
Depth = 61 cm (24 inch) GAG
Approximate loading rate = 3.1-3.5 m/hr (1.3-1.5 gpm/ft2)
Approximate EBCT = 10.1-11.4 minutes
,GAC age = 3% months
GAG age = 4 months
GAG = granular activated carbon
Inf = influent
Eff = effluent
F400 = Filtrasorb 400
FC = Filtrasorb C
ICI = Hydrodarco 8x16
-------
ORGANIC COMPOUNDS NOT DESIGNATED AS PRIORITY POLLUTANTS
GC/MS identification of recurring unknowns was attempted when the GC/
Hall, GC/FI or GC/alkali detector responses indicated sufficient concentra-
tion (1 ug/L) for GC/MS analysis. Some recurring unknowns were identified,
others were not.
trans-1,2-Dichloroethylene
This compound was confirmed by GC/MS-SIM at concentrations at or above
0.1 ug/L once in finished water at Wheeling, once in raw, Filtrasorb 400 GAG
effluent and finished water at Beaver Falls, and once in finished ground water
at West View. GC/Hall analyses of these utilities' waters presumptively indi-
cate the occasional presence of this compound.
Squalene
Squalene was identified by GG/MS in an untreated surface water at
Wheeling at a concentration exceeding 1 ug/L. The compound had a retention
time of 1.65 relative to hexachlorobenzene when using the procedure detailed
in Appendix D.
1,2,3,4-Tetrahydronaphthalene (Tetralin)
Tetralin was identified by GC/MS in an untreated surface water at
Louisville at a concentration exceeding 1 ug/L. The compound had a retention
time of 0.41 relative to hexachlorobenzene when using the procedure detailed
in Appendix D.
6-Tertiary butyl meta cresol and 2,6-Tertiary dibutyl meta cresol
These cresols were identified once by GC/MS in untreated surface water at
Wilkinsburg and in untreated ground water at West View. The 6-tertiary butyl
meta cresol was identified by GC/MS in an untreated surface water at Fox
Chapel. Concentrations were at or above 1 ug/L in each sample. Retention
times relative to hexachlorobenzene were 0.67 for the butyl cresol and 0.93
for the dibutyl cresol when using the procedure detailed in Appendix D.
Unidentified Compounds Resulting from Chlorination
At several utilities, compounds were detected in chlorinated waters that
were rarely detected in raw waters. These compounds may be products of chlor-
ination or may be contaminants in chlorine used for disinfection. When de-
tected, concentrations in in-plant waters were typically lower than concen-
trations in finished waters possibly because chlorine contact time in in-plant
waters was less than in finished waters or because finished waters had been
chlorinated twice. Concentrations of these compounds were insufficient for
GC/MS identification.
Raw and finished water data for three unidentified base-neutral extract-
able halocarbons are presented in Tables 116 through 118. These data demon-
strate the presence of these unidentified halocarbons in finished waters at
greater frequency and at higher concentrations than in raw water. Data pre-
214
-------
sented In Tables 84 and 85 demonstrate the same for a compound which was pre-
sumptively identified as 2-chloronaphthalene but which could not be GC/MS
confirmed as 2-chloronaphthalene and could not be identified. It may have
been a halocarbon resulting from the application of chlorine.
These unidentified halocarbons were detected less frequently and at lower
concentration at utilities (West View, Fox Chapel, Wilkinsburg, Western
Pennsylvania Water Company) that demonstrated lower formation of trihalo-
methanes than other utilities (see Table 47), suggesting that these halocar-
bons, like the trihalomethanes, may be chlorination products.
At the Western Pennsylvania Water Company in July, a purgeable halocarbon
was detected in chlorinated waters that was not detected in raw water. The
compound could not be GC/MS identified. These data are presented in Table
115. This compound was not detected at other times at the utility. A
purgeable halocarbon with a similar relative retention time was frequently
found in Beaver Falls' finished water but rarely in its raw water. It could
not be GC/MS identified.
TABLE 115. UNIDENTIFIED PURGEABLE HALOCARBON3 DATA
WESTERN PENNSYLVANIA WATER COMPANY, JULY 5-14, 1978
GC/HALL DETECTOR, (MEAN VALUES)
Water Concentration," ug/L"
raw
chlorinated raw
clarified
settled
GAG filtered
finished
ND
0.3
0.6
0.7
0.6
0.3
aUsing procedure described in Appendix B,
compound has retention time of approxi-
mately 0.70 relative to 1,4-dichlorobutane.
bQuantification based on 1,4-dichlorobutane.
215
-------
TABLE 116. UNIDENTIFIED BASE-NEUTRAL EXTRACTABLE HALOCARBONa DATAb'C
JULY 19 7 7- JUNE 1978, GC/HALL DETECTOR
Utility
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mined
West View6
Beaver Falls
Wheeling
Hun ting ton
Cincinnati
Louisville
Evansville
Total or Mean
West Viewf
Raw Water
o
-o
o>
u
CO
01
C/}
U
O
CO
01
B
H
12 1
9
11
12
11
18
12
21
11
11
11
139
11
60
r-t
0
A\
c
o
1*4
u
o
CO
-------
TABLE 117. UNIDENTIFIED BASE-NEUTRAL EXTRACTABLE HALOCARBON DATA
JULY 1977-JUNE 1978, GC/HALL DETECTOR
1-*
'
Utility
Fox Chapel
Wilkinsburg
Pittsburgh
WPW/Hays Mined
West Viewe
Beaver Falls
Wheeling
Hun ting ton
Cincinnati
Louisville
Evansville
Total or Mean
West Viewf
Raw Water
tj
o
TJ
0)
CJ
CO
cu
CJ
o
CO
cu
0
H
12
9
11
12
11
18
12
21
11
11
11
139
11
i^|
60
3
,— |
O
A\
*U
£«
3
O
CJ
CO
cu
0
>r4
H
0
0
0
0
0
0
0
0
0
0
0
0
0
^^
60
3
r- )
O
V
*o
c
o
fa
CJ
CJ
CO
cu
,1
H
0
0
0
1
0
1
3
0
0
0
1
6
0
^
c *•>»
CU 60
,E 3
5s
I— (
c •
0 0
r, A\
ns T)
j,j C
4J 3
C 0
CU fa
CJ
C
O
o
rj
nJ
s
C
o
•H
4J
CO
(-1
4-1
C
01
o
c
o
CJ
CJ
o
I
a
•H
^
cfl
S3
Finished Water
M
o
fa
•o
cu
CJ
M
01
C«
CJ
O
to
CU
e
•H
11
10
11
13
—
20
12
24
16
15
11
143
12
j
-«^1.
60
3
i— 1
.
0
A\
TD
c
3
O
fa
CJ
0
CO
cu
£•
•l_l
H
0
0
2
0
—
2
0
4
3
4
2
17
0
_,
""«•«*.
60
3
i-H
•
0
V
T3
C
3
O
fa
U
CJ3
CO
CU
B
•rH
H
1
1
0
3
—
5
7
8
3
6
3
37
0
^
C "**^
01 60
^
ts
r-t
c •
0 0
'£ A\
CO TD
t-i C
C 0
01 fa
a
c
o
a
CO
CU
0.1
0.2
0.2
0.2
0.3
0.4
0.2
c
o
•H
CO
1*4
4-J
C
01
a
c
o
CJ
CJ
O
§
e
•H
^
CO
S
0.2
0.2
0.3
0.3
0.4
0.5
0.5
a = Using procedure described in Appendix D, compound has retention
time of approximately 0.77 relative to hexachlorobenzene.
b = Quantification based on hexachlorobenzene.
c = NOT CORRECTED FOR EXTRACTION LOSSES.
d = Western Pennsylvania Water Co., Hays Mine Plant.
e = Ohio River at West View.
f = Ground water supply.
217
-------
N)
I-1
oo
o o z:
i-i p* re
o H- co
§o rt
re
O^ 5tJ i*(
H- 3
« <
PJ re
rt
re
CO
e
TD
-O
(O
3
i-l PJ 3
rt to
re <
CO PJ
rt 3
< PJ'
H-
re s!
S PJ
• rt
re
^
n
o
pa
BJ
co
S
H-
re
PJ
s
rt
a = Using proced
time of ap
b = Quantificati
c = NOT CORRECTE
a o -a c
3 H n
TI o re
O cr x
pa PJ H. a.
co 3 re
n re PJ co
X a. rt n
H re >-»
so o M H-
> 3 s- cr
n re
H ET O D,
M re •
O X 00 H-
n
t™* I"^ ^ ^^
o (-1 re T3
W> O M "O
M O rt 3
OT CT H- CU
• re < H-
3 re x
N
re rt o
3 O •
re
• 3" 0
PJ "O
n o
£§
o o.
i-(
cr PJ
re co
N It
re re
3 rt
re re
• 3
rt
§
re
cr.
rt
H-
•p
o
o
G
O
O
H
0
rt
PJ
O
&
§
Co
VO
-
NJ
o
O
I-1
co
NJ
CO
O
*•
O
VO
Evansville
p
o
o
p
-
-
o
<3N
O
VO
Louisville
H1
O
O
Ul
-
o
o
CO
o
Cn
Cincinnati
P
o
o
s;
Cn
CO
O
**
O
Cn
Hun ting ton
ro
o
h-1
ho
Cn
N3
0
NJ
O
CO
Wheeling
NJ
o
-
N>
N>
O
O
NJ
O
N5
Beaver Falls
M
oo
o
o
N>
0
M
l->
O
VO
O
VO
re
CO
rt
£
re
P
o
o
1
1
1
1
1
1
1 WPW/Hays Mined
N3
-
O
o
o
M
CO
O
M
1 Pittsburgh
i— i
o
o
H
O
o
Wilkinsburg
VO
o
o
o
o
o
o
o
y
PJ
T)
re
NJ
0
O
tr
o
o
H-
rt
Times GC Searched For
Times GC Found £0.1 ug/L
Times GC Found <0.1 ug/L
Mean Concentration When
Found £0.1 ug/L
Maximum GC Concentration
Times GC Searched For
Times GC Found ^0.1 Ug/L
Times GC Found <0.1 ug/L
Mean Concentration When
Found £0.1 ug/L
Maximum GC Concentration
PJ
BJ
pi-
re
Tl
3
re
o.
re
H
M
00
•
M
0
t"^ H^
I-1 M
VO M
1 (33
cj oo
3 w
t4 1
30 S
O
n M
CTABLE
L DETEC
o
I
!>
£j4
V
o
-------
REFERENCES
1. Rook, J. J., Formation of Haloforms During Chlorination of Natural
1 Waters. Water Treatment Exam. 23:234 (1974).
2. Bellar, T. A., J.. J. Lichtenberg, and R. C. Kroner. The Occurrence of
Organohalides in Chlorinated Drinking Water. J. Am. Water Works Assoc.
66:703 (1974).
3. Symons, J. M., e£ ad. National Organics Reconnaissance Survey for
Halogenated Organics. J. Am. Water Works Assoc. 67:634 (1975).
4. Symons, J. M., e£ al. Interim Treatment Guide for the Control of
Chloroform and Other Trihalomethanes. WSRD, MERL, U. S. Environmental
Protection Agency, Cincinnati, Ohio, (June 1976).
5. Stevens, A. A., et al. Chlorination of Organics in Drinking Water. J.
Am. Water Works Assoc. 68(11):615-620 (November 1976).
6. Love, 0. T., _et al. Treatment for the Prevention or Removal of Trihalo-
methanes in Drinking Water. In: Interim Treatment Guide for the Control
of Chloroform and Other Trihalomethanes, J. M. Symons, et al. WSRD,
MERL, U. S. Environmental Protection Agency, Cincinnati, Ohio, (June
1976). Appendix 3.
7. Stevens, A. A., and J. M. Symons. Trihalomethane and Precursor
Concentration Changes Occurring During Water Treatment and Distribution.
J. Am. Water Works Assoc. 69(10):546 (October 1977).
8. U. S. Environmental Protection Agency. Sampling and Analysis Procedures
for Screening of Industrial Effluents for Priority Pollutants. EMSL,
Cincinnati, Ohio, (March 1977, Revised April 1977).
9. The Radian Corporation. Methods for Gas Chromatographic Monitoring of
EPA's Consent Decree Priority Pollutants. Austin, Texas 78766. Pre-
pared for: ASTM Symposium on Measurements of Organic Pollutants in Water,
(June 1978).
10. Hewlett Packard 3380A Integrator Instrument Manual. Avondale,
Pennsylvania. (March 1975, revised June 1976).
11. APHS, AWWA, WPCF. Standard Methods for the Examination of Water and
Wastewater. 14th ed., (1976).
219
-------
REFERENCES (continued)
12. Palin, A. T. Analytical Control of Water Disinfection with Special
Reference to Differential DPD Methods for Chlorine, Chlorine Dioxide,
Bromine, Iodine and Ozone. J. Inst. Water Engr. 28(3):139 (May 1974).
13. Geldreich, E. E., H. D. Nash, and D. Spino. Characterizing Bacterial
Populations in Treated Water Supplies: A Progress Report. Microbiologi-
cal Treatment Branch, WSRD, MERL, U. S. Environmental Protection Agency
Cincinnati, Ohio, (1978).
14. Engineering-Science, Inc. An Investigation of the Effect of Open
Storage of Treated Drinking Water on Quality Parameters.
EPA-600/1-77-027, HERL, U. S. Environmental Protection Agency,
Cincinnati, Ohio, (May 1977).
15. United States Environmental Protection Agency. National Interim Primary
Drinking Water Regulations. Federal Register. 40(248):59566-59588
(December 24, 1975).
16. United States Environmental Protection Agency. Interim Primary Drinking
Water Regulations. Federal Register. 43(28):5756 (February 9, 1978).
17. Ingols, R. S., and G. M. Ridenour. Chemical Properties of Chlorine
Dioxide in Water Treatment. J. Am. Water Works Assoc. 40:1270 (1948).
18. Symons, J. M. Interim Treatment Guide for Controlling Organic Contamin-
ants in Drinking Water Using Granular Activated Carbon. WSRD, MERL, ORD,
U. S. Environmental Protection Agency. (January 1978).
19. Ohio River Valley Water Sanitation Commission, unpublished data.
Cincinnati, Ohio.
20. The National Research Council. Drinking Water and Health. National
Academy of Sciences. Washington, D. C., (1977).
21. Basu, O.K., and J. Saxena. Polynuclear Aromatic Hydrocarbons in
Selected U. S. Drinking Waters and Their Raw Water Sources. En. Sci.
and Tech. 12(7):795-798 (July 1978).
22. World Health Organization. International Standard for Drinking-Water.
3rd ed. Geneva, Switzerland, (1971). p. 37.
220
-------
APPENDIX A
GENERAL ORGANIC LABORATORY PROCEDURES
GLASSWARE CLEANING AND HANDLING
Sample Bottles
Three sizes of sample containers were used for project organic sampling.
Forty mL Flint glass vials with Teflon-lined screw caps were used for collec-
tion of purgeable samples. Two hundred and seventy mL standard laboratory
Pyrex glass bottles with Teflon-lined screw caps were used for collection and
storage of terminal level purgeable samples. Gallon Pyrex glass bottles with
Teflon-lined screw caps were used for collection of extractable samples. In
the laboratory at the time of analysis, 12 mL Flint glass vials with Teflon-
lined screw caps were used to contain a transferred portion of the 40 mL
samples.
Forty mL and 12 mL vials were cleaned with detergent and tap water,
rinsed with deionized tap water and oven treated at 250-300°C for two hours.
After cooling, sodium thiosulfate powder was added to each 40 mL vial to eli-
minate residual chlorine at the sample site; these vials were tightly capped
and stored or packed for shipment to the sample site.
Two hundred and seventy mL bottles were washed in the same manner as the
vials. After rinsing, they were kiln heated for two hours at 250°C. Sodium
thiosulfate was not added. Thirty mL of concentrated buffer solution was
added in order to maintain the utility's finished water pH during storage.
The bottles were tightly capped and stored or packed for shipment.
Gallon bottles were washed with detergent and tap water, rinsed with
deionized tap water, rinsed with acetone, and given a final rinse with methy-
lene chloride. The gallon bottles were drained and air dried. After approx-
imately one gram of sodium thiosulfate was added, each bottle was tightly
capped and stored or packed for shipment.
The Teflon caps were washed with detergent and tap water, rinsed with
deionized tap water and air dried.
Laboratory Glassware
All laboratory glassware used in handling project samples was cleaned by
washing with detergent and tap water, rinsing with deionized tap water and
air drying. This included such extraction glassware as Kuderna-Danish (K-D)
221
-------
evaporation apparatus, funnels, separatory funnels, graduated cylinders, one-
liter amber bottles for storage of extracts prior to concentration, and 2 mL
ampules for storage of concentrated extracts. Additionally, separatory fun-
nels were chromic acid washed. K-D apparatus was methylene chloride rinsed,
washed with detergent, rinsed with deionized tap water and oven dried at 110°C
for 30 minutes. To minimize interference from phthalate esters, these proce-
dures were revised for all extraction glassware to include distilled water
rinsing, acetone rinsing and kiln firing at 400°C for 30 minutes.
Materials
Detergent used in washing was RSB-35, a surface active agent from the
Pierce Chemical Company. Austin, Texas, tap water was used for washing and
deionized Austin tap water for rinsing. Solvents for rinsing were Burdick and
Jackson distilled-in-glass quality. Anhydrous sodium thiosulfate (Baker
Analyzed Reagent) was used in the designated sample containers for residual
chlorine reduction.
Buffers used during storage of terminal level purgeable samples were pre-
pared with halide-free (Baker Analyzed Reagent) chemicals and low organic dis-
tilled water.
PREPARATION OF LOW ORGANIC WATER
Water used for purgeable blank analyses, preparation of purgeable stan-
dards and rinsing of purging apparatus was prepared from deionized tap water.
The water was sparged for 30 minutes with zero grade nitrogen.at 100-200 cc/
minute and then sparged continuously at a reduced rate until used.
Distilled water used for recovery tests for extractable compounds, for
rinsing laboratory glassware and for preparation of buffers was prepared in
the following manner. Deionized tap water was distilled over a solution of
potassium permanganate and sodium hydroxide. During the distillation, a
stream of zero grade nitrogen was passed through the aqueous solution at. 50-
100 cc/minute. The distilled water was used from the receiver on the still or
stored in a 20 liter glass bottle with a Teflon-lined screw cap. (The storage
bottle was cleaned with chromic acid, washed with detergent and tap water and
rinsed sequentially with deionized tap water, acetone, methylene chloride and
low organic distilled water.)
OTHER CONTROLS
General
Only high purity laboratory products were employed in the analytical pro-
cedures. Solvents used were Burdick and Jackson distilled-in-glass quality.
Standard solutions of the Priority Pollutants of interest were prepared from
99+% pure reference standard compounds. Gases were zero grade purity and were
cleaned using a 5A molecular sieve placed after the regulators. Further
cleaning of purge and carrier gases for purgeable analyses was. achieved with
the use of a 6.4 mm (%-inch) OD by -28 cm stainless steel trap packed with
Tenax and Chromosorb 102 placed in the gas line after the molecular sieve.
222
-------
These traps were cleaned periodically by disconnecting them and heating at
200°C. System transfer lines were stainless steel. For purgeable analyses,
short transfer lines from the desorption unit to the GC columns were used to
eliminate "memory" problems in the system. Teflon parts were eliminated from
the system where temperatures were in excess of 150°C.
Interference from Laboratory Air
Possible sources of laboratory air contamination include laboratory sol-
vents, cleaning compounds, refrigerants and building materials. Contamination
from the air cannot easily be eliminated. Therefore, the laboratory insured
that system parts which came into contact with the project samples, carrier
gasses or purge gasses were not exposed to laboratory air. A Luer-Lok Valve
was used on the purging vessel to introduce the sample and then close out
laboratory air. Project samples were rapidly introduced to the purging vessel
after uncapping in order to minimize exposure to laboratory air.
SAMPLE STORAGE
Upon receipt at the laboratory, samples were numbered and recorded. Both
purgeable and extractable samples were refrigerated at 2-10°C.
At the time of analysis, a portion of the 40 mL purgeable sample was
transferred headspace free to a 12 mL vial sealed with a Teflon-lined screw
cap. The 12 mL vials were stored at 2-10°C for reanalysis, if desired.
When possible, purgeable samples were analyzed within two weeks of
receipt at the laboratory. During a long period, however, when instrumenta-
tion was revised, these samples were held refrigerated for four to six months
before analysis.
Extractable samples were extracted as soon as laboratory time permitted.
The extract was either concentrated the same day or was stored in one liter
amber glass bottles sealed with Teflon-lined screw caps at 2-10°C overnight
for concentration the next day. All concentrates were stored at 2-10°C in 2
mL ampules sealed with Teflon-lined septa.
Extractable samples were typically extracted and concentrated within
three days of receipt at the laboratory. During one period, however, when
procedures were revised to minimize interferences, these samples were held
refrigerated for three to six weeks before extraction and concentration.
223
-------
APPENDIX B
EQUIPMENT AND ANALYTICAL PROCEDURES
FOR PURGEABLE HALOCARBON PRIORITY POLLUTANTS
STANDARDS
Primary standard solutions at one part per thousand were prepared as a
group from 99+% pure halocarbon standard compounds in Burdick and Jackson dis-
tilled- in-glass quality methanol in a volumetric flask as follows. The flask
was partially filled with methanol. Because the halocarbons are volatile,
these liquids were weighed in a tared microsyringe to prevent evaporation dur-
ing measurement. A 10 uL syringe was rinsed twice with a standard compound
and then brought to a predetermined volume of the standard by weight. This
volume was introduced into the methanol along with several methanol rinsings
of the syringe. The process was repeated for each purgeable standard compound
and the final solution was brought to volume in the flask with methanol. This
stock solution was transferred to vials sealed with Teflon-lined septa for
freezer storage for up to six months.
A secondary standard in methanol at twenty parts per million was prepared
from the primary standard and similarly sealed in vials for freezer storage
for up to six months. The secondary standard solution was used for daily pre-
paration of calibration standards at ten parts per billion (ug/L) by dilution
in low organic water. A single vial of secondary standard was used daily for
up to three weeks, with the Teflon septum being replaced with each use.
Primary and secondary standard solutions of internal standard 1,4-dichlo-
robutane were prepared in the same manner.
EQUIPMENT
A purge, trap and desorption device was interfaced to a Tracor model 560
gas chromatograph equipped with a digital temperature programmer. The GC was
interfaced to a Tracor model 700 Hall electrolytic conductivity detector.
Output from the system was integrated and recorded by a Hewlett Packard model
3380A integrator.
Initially, purge, trap and desorption was performed by a Tekmar model
LSC-1. This unit was replaced by purge, trap and desorption units made by
Radian Corporation.
PROCEDURE
Forty mL sample vials were opened and a portion of the sample was rapidly
224
-------
transferred to a 5 mL syringe for introduction to a purging vessel. The
remaining portion was transferred headspace free into a 12 mL storage vial
and sealed with a Teflon-lined screw cap.
The internal standard, 1,4-dichlorobutane, was introduced by syringe to
the purging vessel. The sample was purged with nitrogen at 40 cc/minute for
twelve minutes. The volatile compounds were trapped on a resin bed of 10 cm
of Tenax GC followed by 5 cm of Chromosorb 102 in a glass-lined 3.5 mm OD
stainless steel trap.
When the purging was complete, the trapped compounds were desorbed for
three minutes with the Radian Corporation made unit. A desorption temperature
of 180°C was reached in approximately 40 seconds. Desorption was onto the
head of a GC column at room temperature.
The GC was equipped with a 3.7- m by 0.35 cm glass column packed with
0.2% Carbowax 1500 on 60/80 mesh Carbopack C. The 0.3 m pre-column contained
3% Carbowax 1500 on 60/80 mesh Chromosorb W-HP. The GC column oven was
rapidly heated to 60°C, held at 60°C for four minutes, then programmed to
170°C at 8°C/minute. When only the trihalomethane compounds were being ana-=
lyzed (terminal level samples), the column oven was rapidly heated to 60°C
after desorption, the initial four minute hold was deleted, and the tempera-
ture was programmed from 60° to 170°C at 10°C/minute. The carrier gas was
nitrogen at 40 cc/minute.
The electrolytic conductivity detector was operated in the halogen speci-
fic mode. The HP 3380A integrator was operated in the internal standard mode.
Quantification by internal standard was based on the formula:
Cy = (Ay x Ry x Ci)/(Ai x Ri)
where
Cy = concentration, ug/L, of compound y in sample
Ay = chromatographed area of compound y in sample
Ry = response factor for compound y in calibration
Ai = chromatographed area for internal standard in sample
Ri = response factor for internal standard in calibration
Ci = concentration, ug/L, of internal standard in sample
concentration, ug/L, in calibration
response factor - chromatographed area ln calibration
Between sample analyses, the sample syringe and the purging apparatus
were rinsed three times with low organic blank water* At the end of each
day's operation or after a sample analysis with high organic concentrations,
the syringe and purging apparatus were rinsed with acetone and blank water.
Between sample analyses, the trap was baked out at 180°C for three minutes and
cooled to room temperature and the GC column was cooled to room temperature.
This procedure applied to the handling of calibration standards, USEPA
reference samples, system blanks and project samples.
225
-------
APPENDIX C
QUALITY ASSURANCE DATA FOR
PURGEABLE HALOCARBONS
The data presented here were generated as part of the quality assurance
program discussed in Section 5. The analytical procedure employed for purge-
able halocarbons is detailed in Appendix B. Interpretation of project purge-
able halocarbon data presented in Sections 6 and 7 was, in part, based on this
quality assurance data.
226
-------
TABLE C-l. SIGNIFICANCE OF CHLOROFORM DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of USEPA Standards
True value, ug/L
Blank corrected mean of standard
run as unknown, ug/L
Mean blank, ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
9.13
10.0
0.15
+ 10
± 14
8
10.1
10.9
+ 7
± 1
2
68.5
70.9
+ 4
± 14
83
74.6
81.7
+ 10
± 1
2
Reproducibility of Laboratory Standards
True value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.1
0.11
0.04
+ 10
± 55
5
0.25
0.21
0.04
- 16
± 10
5
0.5
0.42
- 16
± 8
8
0.1 - 0.5
- 9
± 22
18
1.0
0.94
- 6
± 9
8
10
9.4
- 6
± 20
57
100
102
<0.1
+ 2
± 3
3
200
196
<0.1
- 2
± 8
3
227
-------
I4O
120
-» IOO
2
0
h
Qf.
\-
z
111
u
z
o
u
80 .
CO
O
I
u
III
40 .
20 .
10 .
5.0-
± \°\
A '
* o° *•
, • •
• - O
ib • w •
b •'
Ji.. o^o
• *o
.°Vv •
s..
••
o*
*o o
OF DATA SETS
• FIELD REPUICATE SET
o REPLICATE ANALYSIS OF
SINGLE FIELD SAMPLE
10 20 30 50
DEVIATION ABOUT MEAN ( °/o
Figure C-l. Precision of instantaneous chloroform data.
228
-------
5.0j
4.0.
3.0_
?!
- 2.0
2
O
I- I.O
z
111
U
Z
0
U
I.O
U
I
o
3
lu
O.I -
± 23 %
OF DATA SETS
. FIELD REPLICATE SET
O REPLICATE ANALYSIS OF
SINGLE FIELD SAMPLE
IO 2O 3O
50
} 100
•
o
o
•
•
o
IO 2O 3O 5O ^ IOO
DEVIATION ABOUT MEAN , °/o
Figure C-2. Precision of instantaneous chloroform data.
229
-------
325-
3OO-
-
250 -
J
to
X
"^
- 2OO -
2
g
P
Q/
^ 150 -
Ul
U
2
0
O
IOO -
o
I
o
< 50"
111
±20%
O
*o
0
o
o
ft* *
o
o ,
o
o
°^ 0
0
0 0
o
o
o
^\
0 °
• o
% • *
0 °*
•I .• °
V»°» ° 0
J* 0 0« o°
• * •/ °
v>
-•^ ^5% OF DATA SETS
>
j
o
o
• FIELD REPLICATE SET
0 REPLICATE ANALYSIS OF
SINGLE. FIELD SAMPLE
10 20 30
5O ^ 100
DEVIATION ABOUT MEAN , °/o
Figure C-3. Precision of terminal chloroform data.
230
-------
TABLE C-2. SIGNIFICANCE OF BROMODICHLOROMETHANE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of USEPA Standards
True value, ug/L
Blank corrected mean of standard
run as unknown, ug/L
Mean blank, ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.8
0.64
- 20
± 2 .
2
1.19
1.97
ND
+ 65
± 11
8
9.2
9.3
+ 1
± 0
2
11.6
16.2
+ 36
± 19
83
Reproducibility of Laboratory Standards
True value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.1
0.08
ND
- 20
± 13
5
0.25
0.21
ND
- 16
± 10
5
0.5
0.42
ND
- 16
± 7
8
0.1 - 0.5
- 13
± 10
18
1.0
0.98
ND
- 2
± 6
8
10
9.4
- 6
± 19
57
50
53.5
ND
+ 7
± 8
6
ND = not detected
231
-------
25
20
15
-J
", 100
DEVIATION! ABOUT ME AM , «/o
Figure C-4. Precision of instantaneous bromodichloromethane data.
232
-------
70 -
60 -
2~ 50 -
O
5
Ql
h
Z 40
UJ
u
7.
O
(J
_c\l 30 -
u
CO
I
o
z 2O "**
UJ
JO -
o ot>
o
o o
|o
± 15%
00
o
• FIELD REPLICATE SET
o REPLICATE ANALYSIS OP
SINGLE FIELD SAMPLE
95% OF DATA SETS
I ' \
10 2O 3O 5O J IOO
DEVIATION ABOUT MEAN , %
Figure C-5. Precision of terminal bromodichloromethane data.
233
-------
TABLE C-3. SIGNIFICANCE OF DATA FOR DIBROMOCHLOROMETHANE AND/OR
CIS-1,3-DICHLOROPROPENE AND/OR 1,1,2-TRICHLOROETHANE
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of USEPA Standards3
True value, ug/L
Blank corrected mean of standard
run as unknown, ug/L
Mean blank, ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
1.0
0.80
- 20
± 1
2
2.74
1.87
ND
- 32
± 9
8
7.1
6.7
- 6
± 1
2
17.2
14.4
- 16
± 25
83
Reproducibility of Laboratory Standards13
True value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.38
0.30
ND
- 21
± 7
5
0.96
0.84
ND
- 13
± 8
5
1.5
1.47
ND
- 2
± 2
3
1.92
1.74
ND
- 9
± 7
5
3.0
3.23
ND
+ 8
± 4
3
3.85
3.52
ND
- 9
± 5
5
2
-------
15 .
—
^ l°-
>.
—
2
g
t 4<0"
M
h
2
LJ
u
2
O
u
I.O
• Z <£<
0
o
• 0
* 0
•* CL °
. ° ff*» . •
!**
°* •
• • ^
fc ^ 0
2 o o
• *o » •
* °QD •
o o •
* V °, •
b vo
-• 15 % OF DATA SETS
. FIELD REPLICATE SET
• o REPLICATE ANALYSIS OF
SINGLE FIELD SAMPLE
0
i
< j , , , 1 1 1 —
U
M
L
CO
I
O
IO 2O 3O
50
0-5
S
°
\
IO
2O 3O 5O
DEVIATION! ABOUT MEAN ,
* too
Figure C-6. Precision of instantaneous dibromochloromethane data.
235
-------
25
^ 20
^
z
0 15 •*
h
S
h
z
tu
o
2
0
0 |O -
o
c\l
L
cO
I
O
z
111
^r
-C
— »•
± 2
o
•
0
0 0
•
0
o* o
0 0
0 O o
0
£
v> •
cP •
O (
• rt
w
o ^
* 0*
• •
., °.*0
• * o
0
o o
o« o
.*
L2%
• FIELD REPLICATE SET
o RFPLICATE ANALYSIS OF
SINGLE FIELD SAMPLE
o
0
o
0
-« 15% OF DATA SETS
10 20 30 50 }IOO
DEVIATION ABOUT MEAN , %
Figure C-7. Precision of terminal dibromochloromethane data.
236
-------
TABLE C-4. SIGNIFICANCE OF BROMOFORM DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL = 0.1 ug/L
Reproducibility of USEPA Standards
True value, ug/L
Blank corrected mean of standard
run as unknown, ug/L
Mean blank, ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
2.85
2.35
ND
- 18
± 11
8
4.8
4.76
- 1
± 1
2
9.2
10.2
+ 11
± 2
2
14.2
14.8
+ 4
± 20
83
Reproducibility of Laboratory Standards
True value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.1
<0.1
ND
5
0.25
0.17
ND
- 32
± 12
5
0.5
0.33
ND
- 34
± 36
8
1.0
0.77
ND
- 23
± 7
8
5.0
4.73
ND
- 5
± 5
6
10
9.8
ND
- 2
± 13
57
ND = not detected
237
-------
TABLE C-5. SIGNIFICANCE OF CARBON TETRACHLORIDE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of USEPA Standards
True value, ug/L
Blank corrected mean of standard
run as unknown, ug/L
Mean blank, ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
1.68
1.32
ND
- 21
± 35
8
1.9
1.83
- 4
± 1
2
3.9
3.85
- 1
± 1
2
12.6
11.8
- 6
± 33
83
Reproducibility of Laboratory Standards
True value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.1
0.08
ND
- 20
± 50
5
0.25
0.20
ND
- 20
± 10
5
0.5
0.38
ND
- 24
± 6
8
0.1 - 0.5
+ 22
± 19
18
1.0
0.87
ND
- 13
± 14
8
10
10.1
+ 1
± 23
56
ND = not detected
238
-------
z
0 1.5.
O & _J I.O.
O K\
Z
-------
4.0
^ 3.0 .
2
0 1.5
H
tf
h
tu
U I.O .
2
0
U
CO
c
? 0-5-
U
2 ;
UJ
^
o
o
>
• FIELD REPLICATE SET
• o REPLICATE ANALYSIS OF
Q SINGLE FIELD SAMPLE
o
o
*
%
* ° 0 0 0
° • §
, , r— —————— _^^_ *
10 20 30 50
DEVIATION ABOUT MEAN °/o
Figure C-10. Precision of terminal bromoform data.
240
-------
TABLE C-6. SIGNIFICANCE OF DICHLOROIODOMETHANE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Precision Indicated by Field Replicate Data Sets
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
<0.1
44
± 81
0.1 - 0.2
12
± 40
i.o
1
± 10
Precision Indicated by Replicate Analyses of
Single Field Sample
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
<0.1
13
± 101
0.15
1
± 100
>0.15
0
241
-------
TABLE C-7. SIGNIFICANCE OF CHLOROBENZENE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of Laboratory Standards
True Value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.1
<0.1
ND
5
0.25
0.20
ND
- 20
± 10
5
0.5
0.44
ND
- 12
± 11
5
1.0
0.86
ND
- 14
± 5
5
10
9.7
- 3
± 37
57
ND = not detected
Precision Indicated by Field Replicate Data Sets
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
<0.1
7
± 100
0.1 - 0.8
6
± 59
1.4 - 2.9
6
± 29
Precision Indicated by Replicate Analyses of
Single Field Sample
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
0.1
2
± 100
>0.1
0
242
-------
TABLE C-8. SIGNIFICANCE OF 1,1-DICHLOROETHANE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of Laboratory Standards
True Value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.1
0.10
ND
0
± 20
4
0.25
0.22
ND
- 12
± 23
5
0.5
0.51
ND
+ 2
± 12
5
0.1 - 0.5
- 4
± 18
14
1.0
0.99
ND
- 1
± 26
5
10
10.1
ND
+ 1
± 20
55
ND = not detected
Precision Indicated by Field Replicate Data Sets
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
<0.1
11
± 181
0.1 - 0.4
11
± 81
>0.4
0
Precision Indicated by Replicate Analyses of
Single Field Sample
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
^0.1
5
± 60
>0.1
0
243
-------
TABLE C-9. SIGNIFICANCE OF 1,2-DICHLOROETEANE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of USEPA Standards
True value, ug/L
Blank corrected mean of standard
run as unknown, ug/L
Mean blank, ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
1.0
0.87
- 13
±1
2
1.39
1.80
ND
+ 29
± 16
8
3.1
3.2
+ 3
± 2
2
27.2
34.1
+ 25
± 16
83
Reproducibility of Laboratory Standards
True value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.1
0.15
ND
+ 50
± 7
5
0.25
0.32
ND
+ 28
± 9
5
0.5
0.41
ND
- 18
± 45
8
0.1 - 0.5
+ 14
± 24
18
1.0
0.97
ND
- 3
± 5
8
10
9.7
_ o
± 14
56
ND = not detected
244
-------
TABLE C-10. SIGNIFICANCE OF 1,2-DICHLOROETHANE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Precision Indicated by Field Replicate Data Sets
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
<0.1
20
± 105
0.1 - 0.3
5
± 53
>0.3
0
Precision Indicated by Replicate Analyses of
Single Field Sample
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
^0.1
7
± 100
>0.1
0
245
-------
TABLE C-ll. SIGNIFICANCE OF 1,2-DICHLOROPROPANE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of Laboratory Standards
True Value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.1
0.07
ND
- 30
± 29
5
0.25
0.21
ND
- 16
± 10
5
0.5
0.44
ND
- 12
± 7
5
0.1 - 0.5
- 19
± 15
15
1.0
0.89
ND
- 11
± 7
5
10
9.3
- 7
± 19
56
ND = not detected
Precision Indicated by Field Replicate Data Sets
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
<0.2
12
± 89
>0.2
0
Precision Indicated by Replicate Analyses of
Single Field Sample
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
^0.25
2
± 100
>0.25
0
246
-------
TABLE C-12. SIGNIFICANCE OF TRANS-1,3-DICHLOROPROPENE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibillty of Laboratory Standards
True Value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.1
<0.1
ND
5
0.25
0.19
ND
- 24
± 11
5
0.5
0.40
ND
- 20
+ 10
5
1.0
0.83
ND
- 17
± 8
5
10
9.4
- 6
± 16
44
ND = not detected
Precision Indicated by Field Replicate Data Sets
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
<0.1
2
± 100
£0.1
0
Precision Indicated by Replicate Analyses of
Single Field Sample
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
<0.1
0
>0.1
0
247
-------
TABLE C-13. SIGNIFICANCE OF 1,1,1-TRICHLOROETHANE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of USEPA Standards
True value, ug/L
Blank corrected mean of standard
run as unknown, ug/L
Mean blank, ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
11.2
11.4
+ 2
± 29
83
Reproducibility of Laboratory Standards
True value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.1
0.60
0.04
+ 500
± 25
5
0.25
0.65
0.04
+ 160
± 34-
5
0.5
0.73
+ 46
± 10
8
0.1 - 0.5
+ 200
± 21
18
1.0
1.08
+ 8
± 12
8
10
10.1
+ 1
± 23
56
248
-------
TABLE C-14. SIGNIFICANCE OF TRICHLOROETHYLENE DATA
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of USEPA Standards
True value, ug/L
Blank corrected mean of standard
run as unknown, ug/L
Mean blank, ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
19.0
19.9
+ 5
± 30
83
Reproducibility of Laboratory Standards
True value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
0.17
0.29
0.14
+ 71
± 38
5
0.43
0.52
0.14
+ 21
± 15
5
0.82
0.65
0.59
- 21
± 18
3
0.86
1.18
0.14
+ 37
± 13
5
0.17 - 0.86
+ 32
± 21
18
1.64
1.44
0.59
- 12
± 4
3
1.74
2.16
0.14
+ 24
± 5
5
17.4
16.5
- 5
± 24
43
249
-------
TABLE C-15. SIGNIFICANCE OF DATA FOR
1,1,2,2-TETRACHLOROETHANE AND/OR TETRACHLOROETHYLENE
PURGEABLE HALOCARBONS, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Reproducibility of USEPA Standards3
True value, ug/L
Blank corrected mean of standard
run as unknown, ug/L
Mean blank, ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
8.8
12.0
+ 36
± 32
83
Reproducibility of Laboratory Standards13
True value, ug/L
Blank corrected mean
of standard run as
unknown, ug/L
Mean blank, ug/L
Relative error from
true value, %
(accuracy)
Standard deviation
about mean, %
(precision)
Number of tests
1— 3
0.14
0.13
0.12
- 7
± 31
5
0.35
0.27
0.12
- 23
± 15
5
0.42
0.19
0.16
- 55
± 5
3
0.70
0.61
0.12
- 13
± 11
5
0.84
0.54
0.16
- 36
± 13
3
0.14 - 0.84
- 23
± I?
21
1.41
1.23
0.12
- 13
± 11
5
14.1
13.8
- 2
± 25
53
bfor tetrachloroethylene but based on co-eluting standards
both compounds
250
-------
I4O -
12O -
J 1OO -
2
0 80 4
h
QL
h
•2
uJ
O
Z
O
o
I
h
I-
2
UJ
4O -
20
±2O %
0. 0
0*
• o
'• O O •
-------
4OO
1
%
35O
i^
1
0
•
o
3
0 *
300^8 o
2~
Z 0
2 • ° o
— • °
h
4
ff 250 -
Z
111
u
2
O
0 2OO -
^
I
h
h-
2 ISO -
^
UJ
i.
IOO -
60 J-
• • o
° • ° o
8
o
o o
0
)
3o . °
o!
0° ° °
o
D* ?
o
0
o
• o
0
J°* *
o
r ° .
• «•"*
• * o
• °o^
r? !' o
JW ^ %^
o
. • o
15% OF DATA SETS
0
• FIELD REPLICATE SET
§ o REPLICATE ANALYSIS OF
SINGLE FIELD SAMPLE
o
o
o
20 30 50
DEVIATION ABOUT MEAN °/o
>/|0o
Figure C-12. Precision of terminal total trihalomethane data.
252
-------
APPENDIX D
EQUIPMENT AND ANALYTICAL PROCEDURES FOR
BASE-NEUTRAL EXTRACTABLE HYDROCARBONS
STANDARDS
Calibration standards were prepared gravimetrically according to the
nature of the particular compound. Volatile liquids were weighed in a tared
microsyringe to prevent evaporation during measurement. Solids were weighed
in a tared beaker. Standard compounds of 99+% purity were used. Primary
standard solutions at one part per thousand were made up in Burdick and
Jackson distilled-in-glass quality solvents. Methylene chloride was used to
solubilize the halogenated compounds. Methylene chloride was then exchanged
for hexane in a Kuderna-Danish apparatus. The solvent exchange was carried
out in three steps to insure that all methylene chloride was removed. Primary
standard solutions of non-halogenated compounds were prepared in hexane with
benzene occasionally being used to aid solubility. A secondary dilution from
the primary stock was made in hexane to a ug/L working level and was stored in
hypovials sealed with Teflon-lined septa for up to six months in a freezer.
Internal standard hexachlorobenzene for the calibration standard was prepared
in the same manner.
Prepared working level calibration standards of the base-neutral extract-
able compounds were examined by GC/MS. The presence and elution order of the
project priority pollutants listed in Tables 5 and 6 were confirmed.
EQUIPMENT
O
The USEPA Priority Pollutant Protocol for analysis of base-neutral
extractable compounds by gas chromatography/mass spectrometry (GC/MS) was
revised as necessary by the laboratory to enable routine analysis of concen-
trated sample extracts by GC/Hall detector (GC/Hall) and GC/flame ionization
detector (GC/FID).9
A Tracor model 560 gas chromatograph equipped with a digital temperature
programmer was interfaced to a Tracor model 700 Hall electrolytic conductivity
detector and to a Tracor FI detector. Output from the system was integrated
and recorded by a Hewlett Packard 3380A integrator.
PROCEDURE
The basic extraction and analysis procedures that were used are described
in the USEPA's Protocol.8 Several modifications were made by the laboratory
as listed below:
253
-------
1. Three liters of samples were extracted.
2. After adjusting the pH to greater than eleven, a methanol
solution of hexachlorobenzene was added as an internal
standard to each sample and solvent blank to be extracted.
3. The sample was serially extracted with one 250 mL portion
and two 150 mL portions of distilled-in-glass methylene
chloride.
4. After concentrating the volume of the combined methylene
chloride extracts to one milliliter, 10 mL of distilled-in-
glass hexane was added and the volume was again concentrated
to 1.0 mL + 0.05 mL.
Modifications made in the analysis of the halogenated base-neutral
extractable Priority Pollutants were:
1. A Hall electrolytic conductivity detector operated in the
halogen specific mode was used for detection of all halogen
compounds in this fraction including the pesticides.
2. Nitrogen was the carrier gas at 40 cc/minute.
3. The GC column temperature was programmed, after an initial
four minute hold at 50°C, from 50°C to 280°C at 8°C/minute,
with a final fifteen minute hold.
Quantification by the HP 3380A integrator for both halogenated and non-
halogeanted compounds was calculated as follows:
Cy = Ay x Ry
where
Cy = concentration, ug/L, of compound y in sample
Ay = chromatographed area for compound y in sample
Ry = response factor for compound y in calibration
Response factor = concentration. ug/L. in calibration
chromatographed area in calibration
It should be noted that C is the concentration of the compound in the
sample assuming 100% extraction efficiency.
254
-------
APPENDIX E
QUALITY ASSURANCE DATA FOR
EXTRACTABLE HALOCARBONS
The data presented here were generated as part of the quality assurance
program discussed in Section 5. The analytical procedure employed for extrac-
table halocarbons is detailed in Appendix D. Interpretation of project
extractable halocarbon data presented in Sections 6 and 7 was based, in part,
on this quality assurance data.
255
-------
TABLE E-l. SIGNIFICANCE OF 1,4-DICHLOROBENZENE DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION -LEVEL = 0.1 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17.
85
± 8
2
1.67
62
± 4
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.36
- 13
± 29
3
1.67
1.72
+ 3
± 10
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
17
± 75 -
0.1-0.4
. 7
± 78
1.7
1
± 14
Replicate Analysis
of Single Field Sample
<0.1
18
± 68
0.1-0.9
11
± 19
1.3
1
± 100
a = 3000 concentration factor
b = Each test performed in triplicate
256
-------
TABLE E-2. SIGNIFICANCE OF 1,3-DICHLOROBENZENE DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTORS
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
63
± 11
2
1.67
55
± 4
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.35
- 17
+ 28
.3
1.67
1.72
+ 3
± 9
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
4
± 107
0.1-0.3
3
± 100
1.3-3.3
3
± 72
Replicate Analysis
of Single Field Sample
<0.1
7
± 111
0.5
2
± 58
>0.5
0
a = 3000 concentration factor
b = Each test performed in triplicate
257
-------
TABLE E-3. SIGNIFICANCE OF 1,2-DICHLOROBENZENE AND/OR HEXACHLOROETHANE DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Both Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.33
57
± 6
2
3.33
71
± 1
1
Reproducibility of Both Standards by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.83
0.88
+ 6
± 23
3
3.33
3.42
+ 3
± 8
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
9
± 82
0.1-0.6
4
± 93
1.1
1
± 5
Replicate Analysis
of Single Field Sample
<0.1
13
± 53
0.1-0.5
7
± 3
>0.5
0
a = 3000 concentration factor
b = Each test performed in triplicate
258
-------
TABLE E-4. SIGNIFICANCE OF 1,2,4-TRICHLOROBENZENE AND/OR
HEXACHLOROBUTADIENE DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Both Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.33
61
± 15
2
3.33
31
± 5
1
Reproducibility of Both Standards by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.83
0.84
+ 1
± 16
3
3.33
3.52
+ 6
± 10
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
7
± 77
0.1-0.3
4
± 54
>0.3
0
Replicate Analysis
of Single Field Sample
<0.1
7
± 68
0.1-0.6
5
± 13
>0.6
0
a = 3000 concentration factor
b = Each test performed in triplicate
259
-------
TABLE E-5. SIGNIFICANCE OF BIS(2-CHLOROISOPROPYL) ETHER AND/OR
BIS(2-CHLOROETHYL) ETHER DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTORS
APPROXIMATE LOWER DETECTION LEVEL =0.2 ug/L
Extraction of Both Standards from Distilled Waterb
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests0
0.33
56
± 24
2
3.33
84
± 8
1
Reproducibility of Both Standards by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value,
% (accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.83
0.28
- 66
± 26
3
3.33
3.59
+ 8
± 10
37
a = 3000 concentration factor
b = Blank corrected. See Appendix G.
c = Each test performed in triplicate
There were no field replicate data sets or replicate
analyses data sets in which these compounds were detected.
260
-------
TABLE E-6. SIGNIFICANCE OF BIS(2-CHLOROETHOXY) METHANE DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION LEVEL = 0.1-0.2 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
49
± 51
2
1.67
63
± 6
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.39
- 6
± 8
3
1.67
1.80
+ 8
± 10
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
5
± 71
£0.1
0
Replicate Analysis
of Single Field Sample
<0.1
9
± 98
0.1-0.2
3
± 20
>0.2
0
a = 3000 concentration factor
b = Each test performed in triplicate
261
-------
TABLE E-7. SIGNIFICANCE OF HEXACHLOROCYCLOPENTADIENE DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION LEVEL = 0.1-0.2 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
56
± 36
2
1.67
26
± 5
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.33
- 21
± 16
3
1.67
1.74
+ 5
± 13
37
a = 3000 concentration factor
b = Each test performed in triplicate
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
1
± 150
^0.1
0
Replicate; Analysis
of Single Field Sample
^0.1
6
± 100
>().!
0
262
-------
TABLE E-8. SIGNIFICANCE OF 2-CHLORONAPHTHALENE DATAg
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
50
± 22
2
1.67
53
± 3
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.37
- 12
± 3
3
1.67
1.73
+ 4
± 16
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
0.1-0.4
5
± 21
>0.4
0
Replicate Analysis
of Single Field Sample
<0.1
2
± 100
0.1-1.3
12
± 27
a = 3000 concentration factor
b = Each test performed in triplicate
263
-------
TABLE E-9. SIGNIFICANCE OF 4-CHLOROPHENYL PHENYL ETHER DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
55
± 17
2
1.67
63
± 3
I
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.39
- 6
± 11
3
1.67
1.72.
+ 3
± 14
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
2
± 67
£0.1
0
Replicate Analysis
of Single Field Sample
<0.1
4
± 100
0.2
1
J: 100
>0.2
0
a = 3000 concentration factor
b = Each test performed in triplicate
264
-------
TABLE E-10. SIGNIFICANCE OF 4-BROMOPHENYL PHENYL ETHER AND/OR^-BHC DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Both Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.33
83
± 8
2
3.33
68
± 2
1
Reproducibility of Both Standards by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.83
0.63
- 24
± 10
3
3.33
3.50
+ 5
± 11
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
1
± 0
^0.1
0
Replicate Analysis
of Single Field Sample
<0.1
1
± 100
^0.1
0
a = 3000 concentration factor
b = Each test performed in triplicate
265
-------
TABLE E-ll. SIGNIFICANCE OF tf-BHC (LINDANE) AND/OR 5-BHC DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.33
55
± 6
2
3.33
61
± 4
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.83
0.71
- 14
± 10
3
3.33
3.49
+ 5
± 10
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
4
± 60
£0.1
0
Replicate Analysis
of Single Field Sample
<0.1
5
± 40
£0.1
0
a = 3000 concentration factor
b = Each test performed in triplicate
266
-------
TABLE E-12. SIGNIFICANCE OF HEPTACHLOR AND/OR p-BHC DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Both Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.33
57
± 4
2
3.33
61
± 4
1
Reproducibility of Both Standards by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.83
0.74
- 11
± 3
3
3.33
3.46
+ 4
± 11
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
8
± 61
0.1-0.4
5
± 45
>0.4
0
Replicate Analysis
of Single Field Sample
<0.1
13
± 73
0.1-0.4
2
± 57
>0.4
0
a = 3000 concentration factor
b = Each test performed in triplicate
267
-------
TABLE E-13. SIGNIFICANCE OF ALDRIN DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTORa
APPROXIMATE LOWER DETECTION LEVEL = 0.1 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
55
± 17
2
1.67
63
± 3
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.35
- 17
± 6
3
1.67
1.77
+ 6
± 11
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
10
± 60
0.1-0.9
6
± 21
>0.9
0
Replicate Analysis
of Single Field Sample
<0.1
5
± 79
0.1-0.3
4
± 18
>0.3
0
a = 3000 concentration factor
b = Each test performed in triplicate
268
-------
TABLE E-14. SIGNIFICANCE OF HEPTACHLOR EPOXIDE DATAg
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Standards from Distilled
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
75
± 16
2
Water
1.67
57
± 2
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.39
- 7
± 9
3
1.67
1.73
+ 4
± 9
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
1
± 61
£0.1
0
Replicate Analysis
of Single Field Sample
<0.1
2
± 155
>0.1
0
a = 3000 concentration factor
b = Each test performed in triplicate
269
-------
TABLE E-15. SIGNIFICANCE OF a-ENDOSULFAN DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
7
± 11
2
1.67
10
± 4
I
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.1.7
0.11
- 35
± 1
3
0.42
0.42
+ 1
± 21
3
1.67
1.73
+ 4
± 10
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
7
± 82
0.2
1
± 35
>0.2
0
Replica1:e Analysis
of Single Field Sample
<0.1
18
± 80
0.1
1
± 0
>0.1
0
a = 3000 concentration factor
b = Each test performed in triplicate
270
-------
TABLE E-16. SIGNIFICANCE OF DDT DATA g
BASE-NEUTRAL EXTEACTABLE HALOCARBON, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Standards from Distilled
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
49
± 11
2
Water
1.67
52
± 13
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value, %
(accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.33
- 21
± 12
3
1.67
1.73
+ 4
± 20
37
Precision of Field Data
Range, ug/L
Number of sets where
mean lies in range
Standard deviation
about mean, %
Field Replicate
Data Sets
<0.1
1
± 100
0.1
1
± 100
>0.1
0
Replicate Analysis
of Single Field Sample
<0.1
1
± 170
^0.1
0
a = 3000 concentration factor
b = Each test performed in triplicate
271
-------
TABLE E-17. SIGNIFICANCE OF DIELDRIN AND DDE DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Both Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests ^
0.17
62
± 12
2
1.67
58
± 4
1
a = 3000 concentration factor
b = Each test performed in triplicate
Reproducibility of Both Standards by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value,
% (accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.83
0.77
- 7
± 8
3
3.33
3.45
+ 4
± 9
37
There were no field replicate data sets or replicate
analyses data sets in which these compounds were detected,
272
-------
TABLE E-18. SIGNIFICANCE OF ENDRIN DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR5
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
67
± 18
2
1.67
70
± 10
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value,
% (accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.34
- 19
± 6
3
1.67
1.81
+ 8
± 15
37
a = 3000 concentration factor
b = Each test performed in triplicate
There were no field replicate data sets or replicate
analyses data sets in which this compound was detected.
273
-------
TABLE E-19. SIGNIFICANCE OF DDD AND B-ENDOSULFAN DATA
BASE-NEUTRAL EXTEACTABLE HALOCARBON, GC/HALL DETECTOR3
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Both Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests^
0.33
30
± 12
2
3.33
27
± 7
1
Reproducibility of Both Standards by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value,
% (accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.33
0.22
- 33
± 22
3
0.83
0.73
- 12
± 11
3
3.33
3.44
+ 3
± 10
37
a = 3000 concentration factor
b = Each test performed in triplicate
There were no field replicate data sets or replicate
analyses data sets in which these compounds were detected.
274
-------
TABLE E-20. SIGNIFICANCE OF METHOXYCHLOR DATA
BASE-NEUTRAL EXTRACTABLE HALOCARBON, GC/HALL DETECTOR
APPROXIMATE LOWER DETECTION LEVEL =0.1 ug/L
Extraction of Standards from Distilled Water
Concentration, ug/L
Mean recovery, %
Standard deviation about mean, %
Number of tests
0.17
62
± 12
2
1.67
56
± 19
1
Standard Reproducibility by Direct Injection
True value, ug/L
Mean of standard run as unknown,
ug/L
Relative error from true value,
% (accuracy)
Standard deviation about mean, %
(precision)
Number of tests
0.42
0.23
- 45
± 51
3
1.67
1.84
+ 11
± 42
37
a = 3000 concentration factor
b = Each test performed in triplicate
There were no field replicate data sets or replicate
analyses data sets in which this compound was detected,
275
-------
APPENDIX F
QUALITY ASSURANCE DATA FOR
NON-HALOGENATED EXTRACTABLE HYDROCARBONS
The data presented here were generated as part of the quality assurance
program discussed in Section 5. The analytical procedure employed for extrac-
table halocarbons is detailed in Appendix D. Interpretation of project
extractable halocarbon data presented in Section 7 was based, in part, on this
quality assurance data.
276
-------
TABLE F-l. EXTRACTION3 OF STANDARDS FROM DISTILLED WATER, BASE-NEUTRAL EXTRACTABLE NON-HALOGENATED
PRIORITY POLLUTANTS, GC/FLAME IQNIZATION DETECTOR13
Compound
Naphthalene
Acenaphthylene
Acenaphthene
Dimethyl phthalate
Fluorene
Diethyl phthalate
Phenanthrene and Anthracene
Di-n-butyl phthalate
Fluoranthene
Pyrene
Butyl benzyl phthalate
bis(2-Ethylhexyl) phthalate
and 1, 2-Benzanthracene
and Chrysene
Benzo(a)pyrene
Indeno ( 1 , 2 : c , d) pyr ene
1,2:5, 6-Dibenzanthracene
and 1,12-Benzoperylene
Approximate
Lower Detection
Level (ug/L)
0.5
0.5
1
5
0.5
2
1
0.5
1
0.5
2
1
5
10
10
Mean Recovery ± Standard Deviation, %
Concentration, 1.5 ug/L
Test 1
6±5
19±2
21±2
ND
15 ±6
17±11
19±2
27±3
19±2
19±1
7±7
20±3
ND
ND
ND
Test 2
51±21
51±34
49±27
J23±10
NFB
57±23
55156
25±27
45±38
NFB
NFB
29±18
39±16
43±15
Test 3
87±13
104±15
80±7
136±163
62±27
24+4
81±5
58±39
80±12
73±4
33±20
47+18
58±8
NFB
NFB
Test 4
62±4
53±1
71±15
ND
98±8
ND
102±4
118±10
109±10
109±10
96±28
54±5
ND
ND
ND
Concentration, 10 ug/L
Test 1
91±18
65±24
79±1
32±12
81±2
48±21
79±2
68±6
81±3
83±2
51±15
71+10
60±3
73±20
63±10
Test 2
43±14
32±23
47±13
47±17
47±17
25±6
58±14
30±16
64±22
57±12
NFB
NFB
18±7
21±6
37±11
Test 3
71±9
76±10
69±10
63±12
70±18
23±323
82±8
79±11
87±6
85+6
31±8
51+5
46±8
53±10
45±15
N)
fEach test performed in triplicate
3000 Concentration factor
ND = Not detected
NFB = Not found after blank correction
-------
TABLE F-2. STANDARD REPRODUCIBILITY BY DIRECT INJECTION, BASE-NEUTRAL EXTRACTABLE NON-HALOGENATED
Compound
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Dimethyl phthalate
Diethyl phthalate
Phenanthrene and Anthracene
Di-n-butyl phthalate
Fluoranthene
Pyrene
Butyl benzyl phthalate
bis(2-Ethylhexyl) phthalate
and 1,2-Benzanthracene
and Chrysene
3, 4-Benzof luoranthene and
11, 12-Benzof luoranthene
Benzo(a)pyrene
Indeno ( 1 , 2 : C , D) pyr ene
1,2:5, 6-Dibenzanthracene
and 1, 12-Benzoperylene
~s rr~~
Approximate
Lower Detection
Level
(UK A.)
0.5
0.5
1
0.5
5
2
1
0.5
1
0.5
2
1
5
5
10
10
True
Value
(ug/L)
10
10
10
10
10
10
20
10
10
10
10
30
20
10
10
20
Mean Concentration
of Standard
Run as Unknown
n = 15
(ue/L)
10.5
10.2
10.2
10. Oa
10. la
10.5
20.4
10.4
10.2
10.1
10.6
30.7
19.7
9.7
10.1
20.4
Standard
Deviation
About Mean
7
/o
(precision)
± 9.5
± 3.9
± 4.9
± 5.0
± 8.9
± 12
± 4.4
± 5.8
± 4.9
± 5.0
± 5.7
± 4.6
± 9.6
± 20
± 5.9
± 9.8
Relative
Error From
True Value
%
(accuracy)
+ 5.0
•+2.0
+ 2.0
0
+ 1.0
+ 5.0
+ 2.0
+ 4.0
+ 2.0
+ 1.0
+ 6.0
+ 2.3
- 1.5
- 3.0
+ 1.0
+ 2.0
N3
vj
00
-------
TABLE F-3. STANDARD REPRODUCIBILITY BY DIRECT INJECTION, BASE-NEUTRAL EXTRACTABLE NON-HALOGENATED
Approximate
Lower Detection
Level
Compound (ug/L)
Naphthalene
Acenaphthylene
Acenaphthene
Dimethyl phthalate
Fluorene
Diethyl phthalate
Phenanthrene and Anthracene
Di-n-butyl phthalate
Fluoranthene
Pyr ene
Butyl benzyl phthalate
bis(2-Ethylhexyl) phthalate
and 1, 2-Benzanthracene
and Chrysene
Benzo(a)pyrene
Indeno ( 1 , 2 : C , D) pyr ene
1, 2: 5,6-Dibenzanthracene
and 1, 12-Benzoperylene
0.5
0.5
1
5
0.5
2
1
0.5
1
0.5
2
1
5
10
10
True
Value
(ug/L)
5
5
5
}io
5
10
5
5
5
5
15
5
5
10
Mean Concentration
of Standard
Run as Unknown
n = 15
(ug/L)
5.2
5.1
5.1
J10.6
5.6
10.2
4.9
4.9
5.0
5.1
15
5.0
5.2
9.9
Standard
Deviation
About Mean
(precision)
± 13
± 5.9
± 12
|± 21
± 18
± 5.9
± 18
±18
± 6.0
± 5.9
± 15
± 12
± 15
± 27
Relative
Error From
True Value
(accuracy)
+ 4.0
+ 2.0
+ 2.0
1+ 6.0
+ 12
+ 2.0
- 2.0
- 2.0
0
+ 2.0
0
+ 4.0
- 1.0
-J
VD
-------
APPENDIX G
SOLVENT IMPURITIES AND HALOGENATED BY-PRODUCTS
OF SOLVENT IMPURITIES
Burdick and Jackson distilled-in-glass methylene chloride contains a
small amount of cyclohexene as a preservative. In the extraction laboratory
this compound reacts with any free chlorine present in project field samples
to produce dichlorocyclohexane as a reaction product. Dichlorocyclohexane
has the same retention time under the procedures described in Appendix D as
bis(2-chloroethyl) ether and bis(2-chloroisopropyl) ether. It was necessary,
then, to add thiosulfate to the sample bottle to quench free chlorine at the
sample site.
This phenomenon was demonstrated in the laboratory when free chlorine
spiked distilled water was extracted under the procedures described in
Appendix D to produce 50 ug/L false positive reports of bis-chloro ethers.
Even with thiosulfate present in all sample bottles, a 0.04 to 0.3 ug/L
false positive bis-chloro ether peak was present in all field samples chroma-
tograms. The peak was also present in all solvent blank chromatograms. It
was hypothesized that prior to extraction, a small amount of free chlorine
resulted from methylene chloride degradation and reacted with the preservative
to produce dichlorocyclohexane.
280
-------
APPENDIX H
ATTEMPTED ANALYSIS OF BASE-NEUTRAL EXTRACTABLE
ORGANO-NITROGEN COMPOUNDS
The compounds listed in Table H-l are the nitrogen containing base-
neutral extractable Priority Pollutants. Analysis for these compounds in pro-
ject concentrated sample extracts was attempted. A Tracor model 702 nitrogen-
phosphorous alkali flame ionization detector (sensitized to nitrogen) was
interfaced to a Tracor model 560 gas chromatograph. The detector output was
integrated and recorded by a Hewlett Packard 3380A programmable integrator.
The GC/alkali detector lower levels of detection are also listed in Table H-l.
A typical chromatogram resulting from direct injection of calibration stand-
ards at 6.66 ug/L is shown in Figure H-l. Extraction recoveries for calibra-
tion standards in distilled water were evaluated at three concentrations: 1.66
ug/L, 3.33 ug/L and 6.66 ug/L. These data are included in Table H-l. System
blank evaluations (including extraction solvents) indicated occasional inter-
ference in areas of the chromatogram unrelated to Priority Pollutant retention
times.
TABLE H-l. EXTRACTION RECOVERIES AND DETECTION LEVELS OF
Compound
Nitrobenzene
2 , 6-Dinitrotoluene
2 , 4-Dinitrotoluene
N-nitrosodiphenylamine
Benzidine
Lower Detection3
Level (ug/L)
4.0
0.4
0.1
0.4
4.0
0.5
Average
1.66 ug/L
Standard
61
50
72
89
3.33 ug/L
Standard
58b
80
73
84
63
79
6.66 ug/L
Standard
( °/\
\/o )
87
87
91
103
87
aWith
^Only one determination.
Sample chromatograms produced under a thorough quality control program
contained numerous peaks, some being presumptively identified as Priority
Pollutants. See Figure H-2. GC/MS confirmation of the identifications, how-
ever, was not possible. For example, benzidine was frequently reported in
project samples at concentrations ranging from 1.0 to 15 ug/L. For confirma-
tion to occur by GC/MS, samples would have had to contain 20 to 50 ug/L of
benzidine in order to elicit a sufficient scanning mode response. A compar-
able concentration was needed for scanning mode confirmation of the other
nitrogen compounds. Problems were also involved in GC/MS confirmation by
281
-------
selected ion monitoring. According to the USEPA Protocol,8 GC column condi-
tioning with benzidine is necessary to chromatograph adequately the nitrogen-
containing Priority Pollutants. Benzidine used in column conditioning
resulted in an interference in confirmation attempts by selected ion monitor-
ing. Other analytical methods likely available for characterization of this
group of compounds were beyond the scope of the project.
An evaluation of the largest GC/alkali detector response presumptively
identified as benzidine in a sample at 15 ug/L was attempted by GC/MS. A
likely identification of the compound eliciting the response was squaline, a
naturally occurring nitrogen compound ubiquitous in the environment. Because
of the lack of GC/MS support for presumptive GC/alkali detector data, this
analytical task was abandoned.
282
-------
indole (internal standard)
-2,6-dinitrotoluene
N-nitrosodiphenylatnine
-2,4-dinitrotoluene
-benzidine
-3,3-dichlorobenzidine
Figure H-l. Typical gas chromatogram of base-neutral extractable Priority
Pollutants calibration standard using alkali flame ionization detector.
283
-------
unknown
indole (internal standard)
unknown
2,6-din±trotoluene
N-nitrosodiphenylamine
* -benzid ine
3,3-dichlorobenzidine
note: other peaks are unknowns
Figure H-2. Typical gas chromatogram of base-neutral
extractable sample using alkali flame ionization detector.
284
-------
APPENDIX I
MASS SPECTROMETRY EQUIPMENT AND ANALYTICAL PROCEDURES
The USEPA Protocol for analysis of Priority Pollutants by gas chromato-
graphy/mass spectrometry (GC/MS)8 was closely followed by the GC/MS labora-
tory. Hewlett-Packard 5982A and 5985 combined gas chromatographs/mass spec-
trometers (GC/MS) and a Hewlett-Packard 5944A dedicated data system were used.
The MS systems utilized jet separators for the GC effluents. The system per-
formance was optimized daily for the analysis of 20 nanograms of decafluoro-
triphenylphosphine.
For analysis of purgeable halocarbons, a Tekmar model LSC-1 Liquid Sample
Concentrator was interfaced to the GC/MS system. While a sample was purged,
the GC oven was cooled to a subambient temperature of -50°C. Desorption from
the Tekmar was achieved in 8 minutes at 180°C onto the head of the GC column.
At the end of the 8 minute period, the GC oven temperature hadBreached appro-
ximately -20°C. The temperature was then rapidly raised to 60°C and program-
med according to protocol. MS scanning was started immediately.
285
-------
APPENDIX J
ORGANIC SAMPLING PROCEDURES
INSTANTANEOUS LEVEL PURGEABLE SAMPLING PROCEDURE
The 40 mL bottles for the sampling of purgeable compounds contain powder-
ed sodium thiosulfate. This substance must not be lost during sampling.
Therefore, it is extremely important that the sample water gently flow into
the bottle such that the bottle will be filled with little or no spillover.
If the water to be sampled is not tapped, use a beaker to introduce the
sample water to the 40 mL bottle. This beaker should have been thoroughly
washed, rinsed with distilled water and air dried. At the sample site, rinse
the beaker several times with the sample water prior to collection.
Remove the cap from the bottle to be filled, being careful not to spill
any of the thiosulfate out of the bottle. Avoid fingering the lip of the
bottle. Fill the bottle carefully with gently running water from the tap or
from the beaker until a convex meniscus forms above the lip of the bottle.
Carefully place the cap on the bottle and screw it securely in place. The
displaced meniscus will run down the sides of the bottle. Invert the bottle
several times. There should be no air space in the bottle larger than this
letter 0 . Dry the bottle off, label it properly and secure it with trans-
parent tape. Refrigerate it in the dark until sample shipping time.
TERMINAL LEVEL PURGEABLE SAMPLING PROCEDURE
Two bottles are required for this procedure. A 270 mL bottle is used for
sample storage during which time available trihalomethane precursor will react
with chlorine to form trihalomethanes. The sample will be collected in this
bottle. A 40 mL bottle contains powdered thiosulfate to stop the trihalo-
methane reaction and is used to ship the sample for analysis. The 270 mL
bottle will be shipped back empty to the laboratory for cleaning.
The 270 mL bottles contain a buffer with a pH at or near the utility's
finished water pH. This buffer must not be lost during sampling. Therefore,
it is extremely important that the sample water gently flow into the bottle
such that the bottle will be filled with little or no spillover.
To ensure the reaction reaching its formation potential, the sample is
usually chlorinated. Therefore, prior to sample collection, a stock chlorine
solution must be prepared.
A chlorine stock solution bottle and a 10 mL pipette should be readied
286
-------
prior to preparing the solution. Wash them and thoroughly rinse them with
distilled water. Allow them to dry. Weigh out 800 mg of reagent grade
Ca(OCl)2 and add it to 1.0 liter of distilled water. This should give a stock
strength of approximately 400 mg/L free chlorine. This solution should be
stored in a dark or aluminum foil wrapped glass stoppered bottle in a refri-
gerator that is free of organic chemicals, glues, solvents, etc. If is has
been stored for longer than a week prior to use, discard it and prepare a new
solution.
After chlorinating this sample, storing it for the designated time and
transferring to the 40 mL bottle containing thiosulfate, it will be necessary
to determine the free chlorine residual of the sample remaining in the 270 mL
bottle. The buffer in that bottle, however, may interfere with the chlorine
measurement. It will be necessary, therefore, to prepare an acid solution so
that the PH can be adjusted prior to making the chlorine measurement. For
this purpose dilute one part reagent grade H2SOz, into 40 parts distilled
water.
Immediately before sampling, pipette 10 mL of the stock chlorine solution
into the 270 mL bottle, being careful not to lose any of the buffer. Cap the
bottle. Go to the sample location.
Remove the cap from the 270 mL bottle being filled, being careful not to
spill any of the chlorine and buffer solutions in the bottle. Avoid fingering
the lip of the bottle. Fill the bottle carefully with gently running water
from the tap or from the beaker until a convex meniscus forms above the lip of
the bottle. Carefully place the cap on the bottle and screw it securely in
place. The displaced liquid will run down the sides of the bottle. Gently
invert the bottle several times to mix the sample and buffer and chlorine
solutions. There should be no airspace in the bottle larger than this letter
"0." Dry the bottle off, label it properly, and secure it with transparent
tape. Store it in the dark at a temperature approximating that of the
finished water until it is time to transfer it to the 40 mL bottle.
At the specified transfer time, remove the cap from the 40 mL bottle to
be filled, being careful not to spill any of the thiosulfate. Avoid fingering
the lip of the bottle. Remove the cap from the 270 mL bottle. Pour the
sample carefully from the 270 mL storage bottle into the 40 mL bottle until a
convex meniscus forms above the lip of the bottle. Carefully place the cap
on the bottle and screw it securely in place. The displaced liquid will run
down the sides of the bottle. Invert the bottle gently several times and
check for air bubbles. Dry the bottle off, label it properly, and secure the
label with transparent tape. Refrigerate the 40 mL bottle in the dark until
sample shipping time.
There should be approximately 230 mL of sample remaining in the 270 mL
bottle. Use this 230 mL to determine the remaining free chlorine residual by
whatever means you normally use for determination of free chlorine residual.
Measure out the volume required. Add the acid solution drop by drop until the
solution is very near pH 7. Then continue with the routine procedure for the
utility's free chlorine residual determination. Record this residual.
287
-------
EXTRACTABLE SAMPLING PROCEDURE
The gallon bottles for the sampling of extractable compounds will arrive
at the utility containing granular thiosulfate.
Remove the cap from the bottle. Fill the gallon bottle carefully with
gently running water from a tap or from a beaker. Fill the bottle to very
near the top, being careful not to lose any of the thiosulfate. This bottle
does not have to be filled airspace free. Fill it to very near the top. Cap
the bottle. If the outside of the bottle was wetted, dry it off. Label it
properly and secure it with transparent tape. Refrigerate it in the dark
until sample shipping time.
288
-------
APPENDIX K
PROCEDURE AND MEDIUM FORMULA FOR A
MEMBRANE FILTER - STANDARD PLATE COUNT
The laboratory apparatus needed is basically identical to that required
for the total coliform procedure as written under 909A, pages 928 to 931, a
through k, Standard Methods for the Examination of Water and Wastewater, 14th
Edition, 1976 (SM). The exception is that the medium is to be used as an agar
only; therefore, the description of absorbent pads is not applicable.
Medium and Preparation
Peptone 2 grams
Gelatin 2.5 grams
Glycerol 1.0 mL
Agar 1.5 grams
Distilled Water 100. mL
Adjust to pH 7.1 with NaOH (N) and autoclave for five minutes at 121°C.
Sterile medium is dispensed in 4-6 mL volumes into 60 by 15 mm petri dishes.
If possible medium should be prepared daily; however, prepared plates of
sterile medium can be stored at 4°C for one week.
The procedure for sample filtration is identical to sample filtration for
determination of total and fecal coliforms by the membrane filter technique.
The same precautions should be taken when rolling the membrane onto the agar
surface to avoid air bubble entrapment.
The selection of sample size should be determined as if the standard
pour plate procedure were to be utilized, particularly if raw water is
examined. When finished, potable water is examined, it is suggested that 100,
50, 25, 10 or 1-mL volumes be filtered.
The exact volume must instantly be determined by the analyst. It is
recommended that three different volumes for each sample be routinely filtered
due to normal variations in total bacterial density regardless of the source
of the sample.
Culture plates are incubated for 48 hours in an inverted position in an
incubator which maintains a 35° ± 0.5 °C temperature. All colonies regardless
of size and color are counted.
Report the total bacterial density in terms of total bacteria/1 mL.
289
-------
Compute the count by the following equation:
Total bacteria colonies/1 mL = ^ oflamplffiltered = ^nsity/1 mL
Membrane filters showing confluent growth, over 200 colonies, or colon-
ies which cannot be individually discerned should not be used for calculating
total bacterial density.
290
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/2-80-028
3. REC
3IEt>
TITLE ANDSUBTITLE
Water Treatment Process Modifications for Trihalo-
methane Control and Organic Substances in the Ohio
River
REPORT DATE
March 1980 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
Ohio River Valley Water Sanitation Commission
8. PERFORMING O
PERFORMING ORGANIZATION NAME AND ADDRESS
Ohio River Valley Water Sanitation Commission
414 Walnut St.
Cincinnati, Ohio 45202
10. PROGRAM ELEMENT NO.
C6irir-SOSl. Task 44
11. CONTRACT/GRANT NO.
R804615
2. SPONSORING AGENCY NAME AND ADDRESS .
Municipal Environmental Research Laboratory-Cinti, Ohio
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final (Oct. 1976-Aug. 1979)
14. SPONSORING AGENCY CODE
EPA/600/14
5. SUPPLEMENTARY NOTES
Project Officers:
Walter A. Feige
Jack DeMarco
513/684-7496
513/684-7282
6. ABSTRACT
Plant-scale studies at seven water utilities using the Ohio, Allegheny, Beaver,
and Monongahela Rivers as their source of supply evaluated various water treatment
process modifications for both the control of trihalomethane levels and the modifica-
tions' impact on bacteriological quality of the finished water. Process modifications
studied, based on comprehensive organic analysis, included relocation of the chlorine
application point, chlorination/ammoniation, partial or complete substitution of
chlorine dioxide for chlorine and placement of four different types of virgin
granular activated carbons in filter beds. Supplemental studies included organic
analysis of monthly raw and finished water samples collected for a one-year period at
each of 11 participating water utilities. In addition to providing plant facilities
and personnel, the 11 utilities joined USEPA in funding this project, which was con-
ducted by the Ohio River Valley Water Sanitation Commission.
This report was prepared in fulfillment of USEPA Grant R-804615 for project
activities for the period October 1976 to August 1979.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Water supply, water treatment, activated
carbon treatment, halogen organic com-
pounds, chloroform, chloromethanes,
chlorination, microbiology, quality
assurance
trihalomethane control,
chlorine dioxide, water
utilities, specific
organic compounds
13 B
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
307
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
291
U.S. GOVERNMENT PRINTING OFFICE: 1980 -657-146/5648
------- |