PA-560/6-77-019
ENVIRONMENTAL MONITORING
NEAR INDUSTRIAL SITES:
POLYCHLORONAPHTHALENES
T>
JULY 1977
-------
ENVIRONMENTAL MONITORING NEAR INDUSTRIAL SITES: POLYCHLORONAPHTHALENES
by
M. D. Erickson, R. A. Zweidinger, L. C. Michael and E. D. Pellizzari
Research Triangle Institute
Post Office Box 12194
Research Triangle Park, North Carolina 27709
Contract No. 68-01-1978
Project Officer
Dr. Vincent J. DeCarlo
Office of Toxic Substances
Washington, D. C. 20460
U. S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF TOXIC SUBSTANCES
WASHINGTON, D. C. 20460
-------
DISCLAIMER
This report has been reviewed by the Office of Toxic Substances, U. S.
Environmental Protection Agency, and approved for publication. 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
-------
ABSTRACT
This research program was initiated to evaluate possible environmental
contamination by polychlorinated naphthalenes (PCNs). The program included
sampling and analytical method development and the collection and analysis
of field samples.
Air, water, soil, sediment and biota were collected from sites near a
PCN manufacturer, and six potential users. A glass fiber filter and two
precleaned polyurethane foam plugs in tandem were used for PCN collection.
Recovery of the PCNs from the foam and filter was accomplished by triple
extraction with toluene. The concentrated extract was chromatographed on a
silica gel column and the final volume reduced to 2 ml in a Kuderna-Danish
apparatus. Samples were analyzed by gas chromatograph/quadrupole mass
spectrometer/computer. The instrument was operated in the multiple ion
detection mode which permitted the detection of <50 pg of a PCN isomer
o
CvO.3 ng/m in air, ^0.2 yg/5, for water and M3.5 yg/kg for soil). The
presence of PCNs was confirmed from full scan mass spectra or by monitoring
the chlorine isotope ratio. PCNs were found at all sites sampled although
appreciable quantities were found at only three sites, near the manufacturer
and two possible users.
This report was submitted in fullfillment of Contract No. 68-01-
1978 by the Research Triangle Institute under the sponsorship of the U. S.
Environmental Protection Agency. This report covers the period June 11,
1976 to December 11, 1976 and work was completed as of February 9, 1977.
iii
-------
CONTENTS
Page
Abstract 11:L
-t -v
Figures
Tables xix
Acknowledgements xxn
List of Abbreviations xxiii
1.0 Summary and Conclusions •"•
2.0 Introduction ->
3.0 Method Development 6
3.1 Sampling 6
3.1.1 Apparatus °
3.1.1.1 Air Sampler Design 6
3.1.1.2 Clean-up Polyurethane Foam Plug 8
3.1.1.3 Collection of PCNs on Polyurethane Foam 1^
3.1.1.4 Water Samplers 22
3.1.2 Selection of Sampling Sites 23
3.1.3 Sampling Protocol 23
3.1.3.1 Air 24
3.1.3.2 Water 24
3.1.3.3 Soil 24
3.1.3.4 Miscellaneous 25
3.1.4 Meteorology 25
3.1.4.1 Measurements 2^
3.1.4.2 Macrometeorological Data
3.1.5 Sample Storage 26
3.2 Analytical Methodology 26
27
3.2.1 Apparatus, Instruments and Chemicals
27
3.2.1.1 Air Equipment
3.2.1.2 Soil Equipment 27
27
3.2.1.3 Water Equipment
IV
-------
CONTENTS (Cont'd)
Page
3.2.1.4 Meteorological 28
3.2.1.5 Chemicals 28
3.2.1.6 Analytical Instrumentation 2°
3.2.2 Sample Workup Procedures .... 28
3.2.2.1 Perchlorination Methods 28
3.2.2.2 Extraction of PCNs from Water 29
3.2.2.3 Extraction of PCNs from Soil 31
3.2.2.4 Extraction of PCNs from PUF and GFF 33
3.2.2.5 Column Chromatography 38
3.2.2.6 Volatile Organics from Aqueous Samples (VOA) 38
3.2.3 Instrumental Analysis and Data Reduction 42
3.2.3.1 GC/MS of PCNs 42
3.2.3.2 GC/ECD of PCNs 57
3.3 Quality Control 57
3.3.1 Method Validation 57
3.3.2 Controls and Blanks 57
3.3.3 Sample Log 58
3.3.4 Instrumentation Control 58
3.3.5 Data Evaluation and Quality Control 62
3.3.6 Confirmation of Results 62
3.3.7 Archival Storage 62
3.3.8 Assessment of Error 62
3.3.8.1 Collection 63
3.3.8.2 Extraction from Sampling Media 63
3.3.8.3 Column Chromatography 63
3.3.8.4 Addition of Standards 63
3.3.8.5 Final Volume of Sample 63
3.3.8.6 RMR 64
3.3.8.7 GC/MS Analysis 64
3.3.8.8 Total Error 64
4.0 Sampling and Analysis Near Koppers Chemical and Coatings
Plant, Bridgeville, PA 66
-------
Contents (Cont'd)
Page
4.1 Field Sampling 66
4.1.1 Air 66
4.1.2 Soil 66
4.1.3 Water 71
4.1.4 Miscellaneous 71
4.2 Analyses 71
4.3 Results and Discussion 71
4.3.1 Air 75
4.3.2 Soil 83
4.3.3 Water 86
4.3.4 Miscellaneous Samples 86
4.3.5 Isomeric Distribution of PCNs 90
5.0 Sampling and Analysis of Polychlorinated Naphthalenes in the
Vicinity of Manning Paper Company, Green Island, New York ... 94
5.1 Field Sampling 94
5.1.1 Air 97
5.1.2 Water 97
5.1.3 Soil and Sediment 100
5.1.4 Summary of Manning Sampling 100
5.2 Analyses 100
5.3 Results and Discussion 100
6.0 Sampling and Analysis of Polychlorinated Naphthalenes in the
Vicinity of Cornell Dubilier Electronics Corporation,
New Bedford, Massachusetts 106
6.1 Field Sampling 106
6.1.1 Air 106
6.1.2 Soil 106
6.1.3 Water 109
6.2 Analyses 109
6.3 Results and Discussion 109
6.3.1 Soil 109
6.3.2 Water 109
vi
-------
Contents (Cont'd)
Page
7.0 Sampling and Analysis of Polychlorinated Naphthalenes in the
Vicinity of Sprague Electric Company, North Adams,
I -I O
Massachusetts XJ-J
7.1 Field Sampling 113
7.1.1 Soil 113
7.1.2 Water 117
7.1.3 Summary of Sprague Sampling
7.2 Analyses 117
7.3 Results and Discussion
7.3.1 Soil 117
7.3.2 Water 117
8.0 Sampling and Analysis of Polychlorinated Naphthalenes in the
Vicinity of General Electric Company, Fort Edward,
New York 121
8.1 Field Sampling 121
8.1.1 Air 121
8.1.2 Soil 127
8.1.3 Miscellaneous 127
8.1.4 Water 127
8.1.5 GE-FE Discharge 128
8.1.6 Summary of GE-FE Sampling 128
8.2 Analyses 128
8.3 Results and Discussion
8.3.1 Air Samples Collected on Tenax
8.3.2 Volatile Organics in Water 129
9.0 Sampling and Analysis of Polychlorinated Naphthalenes in the
Vicinity of General Electric Company, Hudson Falls,
New York 138
9.1 Field Sampling 138
9.1.1 Air 138
9.1.2 Water 143
9.1.3 Soil 143
9.1.4 Summary of GE-HF Sampling 143
vii
-------
Contents (Cont'd)
Page
9.2 Analyses 143
9.3 Results and Discussion of Samples Collected Near General
Electric, Hudson Falls, New York (GE-HF) 143
10.0 Sampling and Analysis of Polychlorinated Naphthalenes in the
Vicinity of Cornell Dubilier Electronics Company, Sanford,
North Carolina 146
10.1 Field Sampling 146
10.1.1 Air
10.1.2 Water
10.1.3 Soil
10.1.4 Miscellaneous 149
10.1.5 Summary 149
10.2 Analyses 152
10.3 Results and Discussion 152
10.3.1 Air 152
10.3.2 Water 152
10.3.3 Soil 152
10.3.4 Amber-Colored Resinous Solid Sample 162
11.0 References 164
Appendix A: Analytical Methods for Polychlorinated
Naphthalenes 168
Appendix B: Summary of Perchlorination Reaction Development
Efforts 180
Appendix C: Gas Chromatographic-Mass Spectrometric Analysis
of Halowaxes and Aroclors 204
viii
-------
Figures
Number Page
1 Sampling apparatus configuration for collection of
polychlorinated naphthalenes 7
2 Air sampler 9
3 Gas liquid chromatography-electron capture detection
CO
( Ni) - foam clean-up; first toluene extraction .... H
4 Gas liquid chromatography-electron capture detection
( Ni) - foam clean-up; third toluene extraction .... 12
5 Gas liquid chromatography-electron capture detection
( Ni) - foam clean-up; sixth toluene extraction .... 13
6 GC-MS analysis of the fifth toluene extract of a
foam plug, multiple ion detection 15
7 Gas chromatography-electron capture detection analysis
of H-1014 recovered from polyurethane foam after air
sampling and H-1014 standard. (a) Before column
clean-up (1 yl injection of a 30 ml sample); (b) After
column clean-up (1 yl injection of a 15 ml sample);
(c) Halowax 1051 (5 ng injection) 16
8 Sampling apparatus configuration for collection
efficiency determination 20
9 Sectioning of soil plugs 32
10 Gas liquid chromatography-electron capture detection
£ -3 £rj)
( Ni) - first extraction of Halowax 1014 ^ from
foam 35
11 Gas liquid chromatography-electron capture detection
/: o M3\
( Ni) - second extraction of Halowax 1014 from
foam 36
12 Gas liquid chromatography-electron capture detection
63 (R)
( Ni) - third extraction of Halowax 1014 ^ from
foam 37
ix
-------
Figures (Cont'd)
Number
13 Apparatus for VOA purge ^3
14 GC/MS of a Halowax 1014™ Halowax 1051^ and
d1 ..-anthracene mixture illustrating temperature
programming ^4
15 GC/MS analysis of a pesticide mixture using multiple
ion detection with PCN ions 48
16 GC/MS analysis of a pesticide mixture using multiple
ion detection with PCN ions 49
17 GC/MS of Aldrin using multiple ion detection for
polychloronaphthalene detection 50
18 Gas chromatograph-mass spectrometric-multiple ion
detection response relative to d.--anthracene as a
function of concentration for selected ions 51
19 Field sampling protocol for ambient air 61
20 Map of Koppers with sampling locations for PI -
10/25/76 69
21 Map of Koppers with sampling locations for P2 -
10/26/76 70
22 The length of wind speed class proportional to its
frequency of occurrence 76
3
23 Total PCN concentrations (ng/m ) in air near Koppers
Chemical and Coatings Plant, Inc., Period 1 77
24 Map of Koppers with sampling locations (Period 1) .... 78
3
25 Total PCN concentrations (ng/m ) in air near Koppers
Chemical and Coatings Plant, Inc., Period 2 79
26 Map of Koppers with sampling locations (Period 2) .... 80
27 Average distribution of PCNs on sampling media 82
28 Total PCN concentrations (yg/kg) in soil near
Koppers Company (Period 2) 84
29 Distribution of PCNs on soil near Koppers Company .... 85
30 Total ion current chromatogram from gc/ms analysis of
organics collected near Koppers Chemical Company,
Bridgeville, Pennsylvania (P2/C7/L21) 87
-------
Figures (Cont'd)
Number Page
31 Total ion current chromatogram of gc/ms analysis
of organics collected near Koppers Chemical Company
Bridgeville, Pennsylvania (P2/C7/L20) 88
32 Gas chromatography/mass spectroscopic analysis of
ambient air near Koppers (P2/C7/L20) 89
33 Isomeric distribution of PCNs between soil and air
at Koppers 93
34 Map of Green Island and surrounding area with sampling
locations for Manning - PI - 11/10-11/76 95
35 Map of Green Island and surrounding area with sampling
locations for Manning - P2 - 11/11-12/76 96
36 Total ion current chromatogram from GC/MS analysis of
organics collected near Manning Paper Company, Green
Island, New York (P2/C5/L11) 105
37 Map of the area surrounding Cornell Dubilier, New
Bedford, MA with sampling locations for PI -
11/13/76 108
38 TIC chromatogram of GC/MS analysis of CDE composited
soil sample Ill
39 Map of the area surrounding Sprague Electric with
sampling locations for PI - 11/14/76 115
40 Map of the area to the west of Sprague Electric with
sampling locations for PI - 11/14/76 116
41 TIC chromatogram of GC/MS analysis of Sprague
composited soil sample (P1/C1/L1-6) 119
42 Map of the area surrounding General Electric -
Ft. Edwards with sampling locations for PI -
11/15-16/76 125
43 Map of the area surrounding General Electric -
Fort Edwards with sampling locations for P2 -
11/16-17/76 126
XI
-------
Figures (Cont'd)
Number Page
44 Gas chromatography/mass spectroscopic analysis of
air samples collected above water outfall of General
Electric Company, Ft. Edwards, New York
(P1/C5/L18) 133
45 Total ion current chromatogram from gas chromatography/
mass spectroscopic analysis of organics purged from
water samples collected in outfall stream near General
Electric Company, Ft. Edward, New York (P1/C4/L15) . . . 135
46 Total ion current chromatogram from gas chromatography/
mass spectroscopic analysis of organics purged from
water samples collected in outfall stream near
General Electric Company, Ft. Edward, New York
(P1/C4/L14) 136
47 Total ion current chromatogram from gas chromatography/
mass spectroscopic analysis of organics purged from
water samples collected from Hudson River below
outfall stream near General Electric Company, Ft.
Edwards, New York (P2/C2/L5) 137
48 Map of the area surrounding General Electric -
Hudson Falls with sampling locations for PI -
11/17-18/76 139
49 Map of the area surrounding General Electric -
Hudson Falls with sampling locations for P2 -
11/18-19/76 140
50 Map of the area surrounding Cornell Dubilier,
Sanford, NC showing sampling locations for PI -
12/7-8/76 147
51 Map of the area surrounding Cornell Dubilier,
Sanford, NC showing sampling locations for P2 -
12/8-9/76 148
52 Map of the area surrounding Cornell Dubilier,
Sanford, NC showing sampling locations for P3 -
1/6/77 151
xii
-------
Figures (Cont'd)
Number
3
53 Total PCN concentration (ng/m ) in air near
Cornell Dubilier Electronics Company, Sanford,
NC (Period 1) 156
2
54 Total PCN concentration (ng/m ) in air near
Cornell Dubilier Electronics Company, Sanford,
NC (Period 2) 157
55 Total PCN concentrations (yg/kg) in soil near
Cornell Dubilier Electronics Company, Sanford,
NC (Period 3) 158
56 Map of the area surrounding Cornell Dubilier,
Sanford, NC showing sampling locations for
P3 - 1/6/77 159
57 Annual Wind Rose, Raleigh-Durham Airport, North
Carolina, 1955-1964 160
58 Fall Wind Rose, Raleigh-Durham Airport, North
Carolina, 1955-1964 161
59 TIC chromatogram of GC/MS analysis of amber-colored
resinous solid collected near Cornell Dubilier
Electronics Company, Sanford, NC (P3/C3/L9) 163
A-l Thermal desorption/high resolution interface manifold
for gas chromatography 176
B-l Halowax 1051 ^ and interfering peak 182
B-2 Halowax 1051® on 2% OV-17 183
B-3 Perchlorinated Aroclor 1248 ^ 184
r f\
B-4 Gas chromatogram ( Ni electron capture detection)
of perchlorinated Halowax 1051 W 191
f o
B-5 Gas chromatogram ( Ni electron capture detection)
of perchlorinated Halowax 1014 ^ 192
B-6 Gas chromatogram ( Ni electron capture detection)
of perchlorinated Halowax 1099 ^ 193
f O
B-7 Gas chromatogram ( Ni electron capture detection)
of Halowax 1014 perchlorinated using antimony
pentachloride 194
Xlll
-------
Figures (Cont'd)
Number
CO
B-8 Gas chromatogram ( Ni electron capture detection)
of Halowax 1014^perchlorinated using antimony
pentachloride and sulfuryl chloride 195
B-9 Gas chromatogram ( Ni electron capture detection) of
Halowax 1014 ^ standard 196
B-10 Gas chromatogram ( Ni electron capture detection) of
Halowax 1014 perchlorination reaction. Reaction
quenched immediately after addition of antimony
pentachloride at -78° 197
/: o
B-ll Gas chromatogram ( Ni electron capture detection) of
Halowax 1014 perchlorination reaction. Reaction
quenched after 0.75 hour at room temperature 198
fi o
B-12 Gas chromatogram ( Ni electron capture detection) of
Halowax 1014 perchlorination reaction. Reaction
quenched after 0.75 hour at room temperature and 1.0
hour, at 40° 199
B-13 Gas chromatogram ( Ni electron capture detection)
of Halowax 1014 ^perchlorination reaction. Reaction
quenched after 0.75 hour at room temperature and 2.0
hour at 40" 200
B-14 Gas chromatogram ( Ni electron capture detection)
of Aroclor 1232 ^ standard 202
/- o
B-15 Gas chromatogram ( Ni electron capture detection)
of Aroclor 1232 VEy reacted under Halowax perchlori-
nation conditions 203
C-l Gas chromatographic-mass spectrometric.analysis of
Halowax-1000 ^; (a) Total ion current 204 ng); (b)
Multiple ion detection (4.08 ng); (c) Multiple ion
detection (4.08 ng) 205
C-2 Gas chromatographic-mass spectrometric analysis of
Halowax-1000 ; multiple ion detection of 4.08 ng ... 206
xiv
-------
Figures (Cont'd)
Number Page
C-3 Gas chromatographic-mass spectrometric analysis of
Halowax-1001 ^; (a) Total ion current (208 ng); (b)
Multiple ion detection (4.10 ng); (c) Multiple ion
detection (4.10 ng) 207
C-4 Gas chromatographic-mass spectrometric analysis of
Halowax-1001 ; multiple ion detection (4.10 ng) .... 208
C-5 Gas chromatographic-mass spectrometric analysis of
Halowax-1013 ^; (a) Total ion current (180 ng); (b)
Multiple ion detection (3.76 ng); (c) Multiple ion
detection (3.76 ng) 209
C-6 Gas chromatographic-mass spectrometric analysis of
Halowax-1013 ; multiple ion detection (3.76 ng) .... 210
C-7 Gas chromatographic-mass spectrometric analysis of
Halowax-1014 ^; total ion current (232 ng) 211
C-8 Gas chromatographic-mass spectrometric analysis of
Halowax-1014 , (a) Multiple ion detection (4.64 ng);
(b) Multiple ion detection (4.64 ng) 212
C-9 Gas chromatographic-mass spectrometric analysis of
Halowax-1031 ; (a) Total ion current (218 ng); (b)
Multiple ion detection (4.36 ng) 213
C-10 Gas chromatographic-mass spectrometric analysis of
Halowax-1031 ; multiple ion detection (4.36 ng) .... 214
C-ll Gas chromatographic-mass spectrometric analysis of
Halowax-1031^; multiple ion detection (4.36 ng) .... 215
C-12 Gas chromatographic-mass spectrometric analysis of
Halowax-1051 ; (a) Total ion current (230 ng); (b)
Multiple ion detection (4.60 ng); (c) Multiple ion
detection (4.60 ng) 216
C-13 Gas chromatographic-mass spectrometric analysis of
Halowax-1099 ®; (a) Total ion current (222 ng); (b)
Multiple ion detection (4.44 ng); (c) Multiple ion
detection (4.44 ng) .- 217
xv
-------
Figures (Cont'd)
Number Pag§_
C-14 Gas chromatographic-mass spectrometric analysis of
Halowax-1099 ^; multiple ion detection
(4.44 ng) 218
C-15 Gas chromatographic-mass spectrometric analysis
of Aroclor-1232 ^H (a) Total ion current (248 ng); (b)
Multiple ion detection (248 ng); (c) Multiple ion
91 n
detection (248 ng)
C-16 Gas chromatographic-mass spectrometric analysis of
Aroclor-1232 ; (a) Multiple ion detection (248 ng);
(b) Multiple ion detection (248 ng); (c) Multiple
ion detection (248 ng) 22°
C-17 Gas chromatographic-mass spectrometric analysis of
Aroclor-1232 ; (a) Multiple ion detection (248 ng);
(b) Multiple ion detection (248 ng); (c) Multiple
ion detection (248 ng) 221
C-18 Gas chromatographic-mass spectrometric analysis of
Aroclor-1242 ™ (a) Total ion current (192 ng); (b)
Multiple ion detection (192 ng); (c) Multiple ion
detection (192 ng) 222
C-19 Gas chromatographic-mass spectrometric analysis of
Aroclor-1242 ^; (a) Multiple ion detection (192 ng);
(b) Multiple ion detection (192 ng) 223
C-20 Gas chromatographic-mass spectrometric analysis of
Aroclor-1242 ^; (a) Multiple ion detection (192 ng);
(b) Multiple ion detection (192 ng) 224
C-21 Gas chromatographic-mass spectrometric analysis of
Aroclor-1248 ^H (a) Total ion current (192 ng); (b)
Multiple ion detection (192 ng); (c) Multiple ion
detection (192 ng) 225
C-22 Gas chromatograpic-mass spectrometric analysis of
Aroclor-1248 ^; multiple ion detection (192 ng) 226
xvi
-------
Figures (Cont'd)
Number
C-23 Gas chromatographic-mass spectrometric analysis of
Aroclor-1248 ; (a) Multiple ion detection (192 ng);
(b) Multiple ion detection (192 ng) 227
C-24 Gas chromatographic-mass spectrometric analysis of
Aroclor-1248 ^H (a) Multiple ion detection (192 ng);
(b) Multiple ion detection (192 ng) 228
C-25 Gas chromatographic-mass spectrometric analysis of
Aroclor-1248 i (a) Multiple ion detection (192 ng);
(b) Multiple ion detection (192 ng) 229
C-26 Gas chromatographic-mass spectrometric analysis of
Aroclor-1248 ^; (a) Multiple ion detection (192 ng);
(b) Multiple ion detection (192 ng) 23°
C-27 Gas chromatographic-mass spectrometric analysis of
Aroclor-1248 ™; multiple ion detection (192 ng) 231
C-28 Gas chromatographic-mass spectrometric analysis of
Aroclor-1248 ^; (a) Multiple ion detection (192 ng);
(b) Multiple ion detection (192 ng) 232
C-29 Gas chromatographic-mass spectrometric analysis of
Aroclor-1254 , total ion current (198 ng) 233
C-30 Gas chromatographic-mass spectrometric analysis of
Aroclor-1254 ; (a) Multiple ion detection (198 ng);
(b) Multiple ion detection (198 ng) 23^
C-31 Gas chromatographic-mass spectrometric analysis of
Aroclor-1254 ; (a) Multiple ion detection (198 ng);
(b) Multiple ion detection (198 ng) 235
C-32 Gas chromatographic-mass spectrometric analysis of
Aroclor-1268 ; (a) Multiple ion detection (198 ng);
(b) Multiple ion detection (198 ng) 235
C-33 Gas chromatographic-mass spectrometric analysis of
Aroclor-1268 ™ (a) Multiple ion detection (180 ng);
(b) Multiple ion detection (180 ng) 237
xvii
-------
Figures (Cont'd)
Number Page
C-34 Gas chromatographic-mass spectrometric analysis of
Aroclor-1268®; total ion current (180 ng) 238
C-35 Gas chromatographic-mass spectrometric analysis of
Aroclor-1268 ^; multiple ion detection (180 ng) .... 239
C-36 Gas chromatographic-mass spectrometric analysis of
Aroclor-1268 ^; (a) Multiple ion detection (180 ng);
(b) Multiple ion detection (180 ng) 240
C-37 Gas chromatographic-mass spectrometric analysis of
Aroclor-1268 ^; multiple ion detection (180 ng) .... 241
XVZll
-------
Tables
Number Page
1 Retention of Halowax 1014 ^ and 1051 on Olympic
dO
2135 ^ Polyurethane Foam During Air Sampling 18
2 Collection Efficiency of Polychloronaphthalenes .... 21
3 Effects of Interval Sampling on the Concentration of
PCNs in Water 22
4 Recoveries of Polychloronaphthalenes from Water .... 30
5 Extraction of Halowax 1014 from Aqueous Solution
with Toluene vs. pH 31
6 Extraction Efficiency of Chlorinaphthalenes from
Soil 34
7 Summary of Column Chromatography Clean-up
Experiments 39
8 Polychloronaphthalenes Parent Ion and MID Ion
(m/e) Values 46
9 Linear Regression Results of MID Linearity Study
for Polychloronaphthalenes 52
10 Relative Molar Responses for Polychloronaphthalenes
Isomers 54
11 GC/MS/COMP Numerical Data Printout 55
12 Recovery of Chloronaphthalenes from Control Samples . . 59
13 Gas Chromatography/Mass Spectrometry Analysis of
Blank Samples 60
14 Reproducibility of GC/MS Analysis of Samples for
PCNS. Replicate Injections of an Air Sample Extract -
Koppers, P2/C1/L1 - Filter 65
15 Total Estimated Analytical Error 64
16 Sampling Protocol for Koppers Company, Chemical and
Coatings Plant, Bridgeville, Pennsylvania 67
xix
-------
Tables (Cont'd)
Number Pa§e
17 Results of Analysis of Samples Collected Near
Koppers Chemicals and Coatings, Inc., Bridgeville,
T, 72
Pennsylvania
18 Compounds Identified in Ambient Air Near Koppers
Chemical Company (P2/C7/L20) 91
19 Sampling Protocol for Manning Paper Company, Green
Qft
Island, New York y°
20 Results of Analysis of Samples Collected Near Manning
Paper Company, Green Island, New York
21 Sampling Protocol for Cornell Dubilier, New Bedford,
Massachusetts
22 Results of Analysis of Samples Collected in the
Vicinity of Cornell Dubilier Electronics Corporation,
New Bedford, Massachusetts
23 Results of Full Scan GC/MS Analysis of the Composited
Soil Sample 112
24 Sampling Protocol for Sprague Electric Comapny,
North Adams, Massachusetts
25 Results of Analysis of Samples Collected in the
Vicinity of Sprague Electric Comapny, North Adams,
Massachusetts 1-^
26 Results of Full Scan GC/MS Analysis of Sprague
Composited Soil Sample (P1/C1/L1-6) 12°
27 Sampling Protocol for General Electric, Ft. Edwards,
New York 122
28 Results of Analysis of Samples Collected Near
General Electric Company, Fort Edwards, New York ....
29 Compounds Identified in Air Samples Above Outfall
at General Electric Company, Ft. Edwards, New York . . .
30 Sampling Protocol for General Electric, Hudson Falls,
New York 141
xx
-------
Tables (Cont'd)
Number Page
31 Results of Analysis of Samples Collected Near
General Electric Company, Hudson Falls,
New York 144
32 Sampling Protocol for Cornell Dubilier Electronics
Company, Sanford, North Carolina
33 Results of Analysis of Samples Collected Near
Cornell Dubilier Electronics Company, Sanford,
North Carolina 153
34 Results of Full Scan GC/MS Analysis of Amber-Colored
Resinous Solid Collected Near Cornell Dubilier,
Sanford, North Carolina (P3/C3/L9) 162
A-l Operating Parameters for GC/MS/COMP System 175
B-l Perchlorination Reaction Results 186
xxi
-------
ACKNOWLEDGEMENTS
The authors wish to thank Dr. W. D. Bach, Jr. of the Research Triangle
Institute (RTI), for assistance with the meteorology and Messrs S. D.
Cooper, R. B. Keefe, and B. J. Parker (RTI) for their assistance with the
field sampling. The mass spectral analysis by Dr. J. T. Bursey and Mr. L.
Kelner (RTI) is gratefully acknowledged.
Valuable assistance and discussions were obtained from Mr. J. Ciancia,
Environmental Protection Agency Region II; Dr. J. Margason, Environmental
Protection Agency, Research Triangle Park; Drs. V- J. DeCarlo and G. E.
Parris, Environmental Protection Agency, Office of Toxic Substances,
Washington, D.C. Dr. L. Ballard of Nutech, Incorporated, Durham, North
Carolina is thanked for the excellent manufacturing of the air samplers.
Lastly, we wish to thank the many people at each sampling site who permitted
use of their'yards, business establishments, etc. and electricity for our
sampling efforts.
xxxi
-------
LIST OF ABBREVIATIONS
CDE
CDS
BCD
FID
GC
GC/MS
GE-FE
GE-HF
GFF
K-D
Koppers
Manning
MID
PCB
PCN
PPT
PUF
RMR
RSD
SD
Sprague
TIC
VOA
Cornell Dubilier Electronics Corporation, New Bedford,
Massachusetts.
Cornell Dubilier Electronics Company, Sanford, North
Carolina.
Electron Capture Detector.
Flame lonization Detector.
Gas Chromatography.
Gas Chromatography/Mass Spectrometry.
General Electric Company, Fort Edwards, New York.
General Electric Company, Hudson Falls, New York.
Glass Fiber Filter.
Kuderna-Danish Evaporator
Koppers Company, Incorporated, Koppers Chemical and
Coatings Plant, Bridgeville, Pennsylvania.
Manning Paper Company, Green Island, New York.
Multiple Ion Detection.
Polychlorinated Biphenyl.
Polychlorinated Naphthalene.
Parts per trillion.
Polyurethane Foam.
Relative Molar Response.
Relative Standard Deviation.
Standard Deviation.
Sprague Electric Company, North Adams, Massachusetts,
Total Ion Current.
Volatile Organic Analysis.
xxiii
-------
1.0 SUMMARY AND CONCLUSIONS
Polychlorinated naphthalenes (PCNs) are used principally as dielec-
trics in the capacitor industry. Due to their toxicity and chemical
stability, their presence in the environment must be evaluated. Environ-
mental samples were collected near several representative PCN manufacturing
and use sites in the United States and analyzed by gas chromatography/mass
spectrometry to assess the levels of PCNs near these sites. The objective
of this research was to obtain a general assessment of PCN pollution,
rather than detailed monitoring data.
Samples were collected near suspected PCN manufacturer (Koppers
Chemical and Coatings Plant, Bridgeville, Pennyslvania), a suspected PCN-
containing paper manufacturer: (Manning Paper Company, Green Island, New
York), and five capacitor manufacturers: Cornell Dubilier Electronics
Corporation, New Bedford, Massachusetts; Sprague Electric Company, North
Adams, Massachusetts; General Electric Company, Fort Edward, New York;
General Electric Company, Hudson Falls, New York; and Cornell Dubilier
Electronics Company, Sanford, North Carolina). Two of the sites (Cornell
Dubilier Electronics Corporation, New Bedford, Massachusetts and Sprague
Electric Company, North Adams, Massachusetts) were designated as "second-
ary" sites and were sampled according to an abbreviated protocol. At the
other five geographical sites, soil, sediment, biota, 24 hour integrated
water and two consecutive 24 hour integrated air samples were collected,
and analyzed for PCNs. In addition, miscellaneous related water samples
were collected and analyzed where appropriate.
All sampling methods were developed, tested and validated prior to
field sampling. An air sampler was developed in which PCNs were collected
on a glass fiber filter (GFF) and two polyurethane foam (PUT) plugs in
tandem. Suitable cleanup of the PUF was achieved using multiple washings
o
with hot toluene. A flow rate of no more than about 60 J,/min (90 m /24 hr
-------
sample) collected all but monochloronaphthalene efficiently. Higher flow
rates substantially reduced the collection efficiency. Samples were col-
lected at each site using four air samplers located along the arms of
orthogonal transects of the plant site. The orientation of the transects
and distance of the samples from the site were determined by meteorological,
topographical, and other considerations.
Water samples were collected upstream and downstream using a peristal-
tic interval sampler or (where appropriate) by "grab" techniques. Three to
ten core soil samples were collected within a 10-50 m location and composited
for analysis. "Near" and "far" locations along each arm of the transects
were sampled. In addition, "grab" soil samples were collected at locations
suspected of contamination such as sanitary landfills, dumps, etc.
Evaluation of PCN solubility showed toluene to be a superior solvent
to hexane which is customarily used in PCB analysis. Consequently,
recovery of PCNs from PUF and GET was best accomplished by triple extrac-
tion with toluene at 25°. Water samples were extracted by partitioning
into toluene and soil samples were extracted using a mixture of acetone
and toluene after first treating the soil with diethyl ether to loosen the
PCNs from the matrix. The toluene extract volume was reduced without loss
of PCNs using a Snyder column on a flat-bottom boiling flask and further
reduced under a N« stream at 25°. The concentrated extract was chromato-
graphed with hexane on a silica gel column and finally reduced to 2 ml
using a Kuderna-Danish apparatus.
Sample analysis was accomplished by quadrupole gas chromatograph/mass
spectrometry/computer (GC/MS/COMP) using an OV-101 column. To improve
sensitivity and specificity for PCNs, the instrument was operated in the
multiple ion detection (MID) mode, permitting detection of <50 pg of an
3
individual PCN isomer or (for the volume of air sampled) about 0.3 ng/m .
The presence of PCNs was confirmed by either full scan GC/MS/COMP or by
monitoring the chlorine isotope ratio using M and M + 2 or M + 4 mass
ions.
Polychlorinated naphthalene levels in soil near Koppers Chemical and
Coatings Plant, Bridgeville, Pennsylvania were found to be an average of
2400 yg/kg with a range of 480 to 5800 yg/kg. Air samples were found to
-------
3 3
contain from 25-450 ng/m (avg. 150 ng/m ) on the first day and 120-2900
O Q
ng/m (avg. 1400 ng/m ) the second day. A small dead fish was found to
contain 39 yg/kg and locally obtained apples, 90 yg/kg. Only trace
amounts of PCNs were found in the water samples. All eight chloronaphtha-
lene isomers were found in the air and soil samples, with mono-, di-, and
trichloronaphthalene predominating in air and tri-, tetra-, and pentachloro-
naphthalene predominating in soil. The isomeric distribution is a signifi-
cant parameter in evaluating the health effects of PCN contamination due to
the varying toxicity of different isomers.
Samples collected near Manning Paper Comapny, Green Island, New York
contained low-to-undetectable PCN concentrations with an average soil
concentration of 10 yg/kg (range: undetectable to 34 yg/kg), an average
3 33
air concentration of 1.5 ng/m (range: 0.3-2.5 ng/m ) and 1.4 ng/m
3
(range: undetectable to 3.1 ng/m ) on two successive days, and generally
undetectable amounts of PCNs in water.
A composite of the soil samples collected near Cornell Dubilier
Electronics Corporation, New Bedford, Massachusetts was found to contain
a total of 500.yg/kg PCNs. Water samples did not contain detectable
amounts of PCNs.
Two composite soil samples collected near Sprague Electric Company,
North Adams, Massachusetts were found to contain totals of 44 and 52 yg/kg
PCNs. Water samples did not contain detectable amounts.
Air samples collected near General Electric Company, Ft. Edward, New
3
York were found to contain an average of 3.1 ng/m (range: undetectable to
33 3
7.3 ng/m ) and 1.2 ng/m (range: undetectable to 4.9 ng/m ) total PCNs on
consecutive days. Soil samples collected along the transects contained an
average of 2.4 yg/kg (range: undetectable to 7.3 yg/kg) total PCNs. Grab
soil samples contained an average of 7.0 yg/kg (range: undetectable to 21
yg/kg) total PCNs. Water collected by interval samplers did not contain
detectable amounts of PCNs. Total PCN levels of undetectable, 0.6 yg/£
and 5.5 yg/£ were found in water collected by "grab" sampling.
Air samples collected near General Electric Company, Hudson Falls,
3
New York were found to contain an average of 5.6 ng/m (range: 2.8-9.3
3
ng/m ) total PCNs. Soil samples were found to contain an average of 4.5
-------
yg/kg (range: 0.3-12 yg/kg) total PCNs. Water samples did not contain
detectable amounts.
Samples collected near Cornell Dubilier Electronics Company, Sanford,
3 3
North Carolina contained an average of 19 ng/m (range: 9.8-33 ng/m ) on
3 3
the first day and 17 ng/m (range: 9.8-33 ng/m ) on the second day in air,
240 yg/kg (range: undetectable to 470 yg/kg) in soil, and 0.6 yg/& in
water downstream of the plant (undetectable in upstream sample). An
amber-colored resinous solid collected near the plant was found to contain
920 Ug/kg PCNs, mostly as the di-, tri-, and tetrachloronaphthalene.
For the sake of perspective, comparison of PCN concentrations in soil
found in this study with those found in similar studies of PCNs and PCBs
may be helpful. The first PCN concentrations reported for environmental
samples in the United States were 1250-5000 yg/kg in sediment samples
collected from a Florida drainage ditch. A nearby aircraft overhaul
hanger was suspected as a possible source. The values found in our
study are lower than those reported for PCBs in soils near a PCB manu-
facturing site in Illinois where concentrations from 130-20,700 yg/kg with
(9)
an average concentration of 3900 yg/kg were detected (about four times
greater than our findings). These values are approximately ten times
those reported for PCBs in bottom sediments of major drainage basins of
the United States which ranged from 1.2-160 yg/kg with an overall aver-
age of 13.5 yg/kg. These results suggest that transport of PCN and PCBs
by water over any significant distance is slight; a reasonable conclusion
in light of their low solubility in water.
-------
2 . 0 INTRODUCTION
Polychlorinated naphthalenes (PCNs) have recently come under suspicion
as environmental contaminants because of their chemical and toxicological
similarities to polychlorinated biphenyls (PCBs) . Marketed in the
United States as Halowaxes w , PCNs are primarily used to impregnate
capacitor tissue paper and as engine oil additives.
Occupational exposure to PCNs has resulted in dermal reactions (chlor-
acne) , liver damage ' , and in a few cases, death. Penta- and hexa-
chloronaphthalene appear to be primarily responsible for the toxic effects
of PCNs.(1'4)
Polychlorinated naphthalenes have been detected in a variety of
environmental media including dead cormorants in the Netherlands , a
drainage ditch in South Florida near an airport overhaul hanger (1250-
5000 yg/kg) ' ' and in sediment from the Guadalupe River, California.
These findings suggest a potential environmental hazard. It was the purpose
of this research to develop analytical methods for collection and analysis
of PCNs in ambient air and to apply these methods to study the occurrence
of PCNs near five or more manufacture or user sites.
Polychlorinated naphthalenes and their environmental effects and
( 1 3 8^
toxicity have been recently reviewed. ' '
-------
3.0 METHOD DEVELOPMENT
3.1 SAMPLING
It was the objective of this research to collect samples of environ-
mental media at various suitable sites to obtain a reliable assessment of
environmental contamination by PCNs. To this end, sampling apparatus was
developed or acquired and thoroughly tested, sampling sites were selected,
and sampling protocols developed before embarking on any field sampling
trips.
3.1.1 Apparatus
3.1.1.1 Air Sampler Design
A number of problems have been encountered with the impinger method
for collecting pesticides and consequently it has been withdrawn as an EPA
standard method. No other method has been substituted for this purpose.
Substitute methods are currently being evaluated and the most promising of
these is based upon polyurethane foam (PUF) as a sorbent. The
primary limitation of this material is a relatively high background. In
order to compensate for this background, large samples at high flow rates
have been collected for analysis. Hi-Vol samplers were modified to accept
the sampling module (Figure 1) and a by-pass valve used to regulate the
flow rate to ^300 5,/min. The load on the pump motor plus the loss of
ventilation resulted in overheating and occasionally total destruction of
the pump motor. In addition to expense and inconvenience, such an occurrence
(14)
has lead to sample contamination. In view of this problem the sampler
was redesigned taking into account the particular constraints pertaining to
this system.
The pressure drop imposed by the glass fiber filter and sorbent
material is such that flow rates greater than 300 &/min are not practical.
-------
GLASS FIBER FILTER
WIRE MESH SUPPORT
POLYUR ETHANE
FOAM PLUGS
\
II
V
.s^\Vv,
\ ^
\
\-V\\A:
. i
J
STAINLESS STEEL
COUPLING
COUPLING BOLTS
WIRE MESH SUPPORT
SCALE Icm = 4cm
Figure 1. Sampling apparatus configuration for collection
of polychlorinated naphthalenes.
-------
For this reason pumps which could operate continuously at about 300 £/min
with an estimated pressure drop of 130 mm of Hg were investigated. Cast
rotary vane pumps satisfied this requirement.
Air samplers, custom designed by Nutech Corporation (Durham, NC),
were used for the collection of chloronaphthalenes in ambient air. The
sampler shown schematically in Figure 2 uses a Cast oilless rotary vane
vacuum pump model no. 1022. This pump is rated to deliver 280 &/min air
flow at no pressure drop and 250 &/min air flow at 130 mm Hg. The selection
of this model was based in part on the fact that it has a motor mounted
drive rather than belt drive which results in a more compact, lighter unit
for field sampling.
The flow rate is monitored at the pump exhaust to avoid the problems
associated with correcting for pressure changes during the sampling period.
The pressure at the pump is monitored with a vacuum gauge which serves to
signal possible malfunctions.
3.1.1.2 Clean-up Polyurethane Foam Plug
Previous workers using PUF to collect PCBs have used hexane,
acetone, and/or petroleum ether in a Soxhlet extractor to remove impurities
prior to sampling. Initial tests with hexane-extraction showed no decrease
in the level of impurities in the extracts after several successive extrac-
tions and it was further shown (vide infra) that in fact, PCNs were not
readily soluble in hexane. Considerable effort was expended to find a
method which would successfully clean up PUF for use in PCN sampling. The
procedure finally developed involves manual extraction with hot toluene
followed by drying in a vacuum oven. The complete standardized procedure
is detailed in Appendix A. Details of the development of the procedure are
discussed below.
When it was determined through initial extraction attempts with
acetone and petroleum ether in a Soxhlet apparatus for 12 hours that an
adequate removal of the electron capturing contaminants from samples of
Olympic 4214 polyether type polyurethane foam (Olympic Products Company,
Greensboro, NC) was not achieved, further development of PUF clean-up was
pursued. In order to better estimate the mass of contaminants being ex-
tracted 100 to 200-fold concentrates of petroleum ether extracts were
-------
VACUUM
GAUGE
COLLECTION MODULE
EXHAUST
FLOW METER
ROTARY VANE VACUUM
PUMP
Figure 2. Air sampler.
-------
analyzed by GC/Flame lonization Detection (FID) which showed that most of
these extracted materials were insensitive to FID. No improvement with
additional petroleum ether extractions was observed. Similar extraction
of a virgin piece of foam with toluene at 90° for 2 hr indicated a much
higher degree of contaminant removal although no significant decrease in
number or size of chromatograph peaks was observed with a second extrac-
tion. Subsequent extractions of this same piece of foam with hexane at
60° revealed smaller, but still substantial quantities of contaminants
remaining.
In light of the solubility results, a 1 g piece of foam was manually
extracted at 100° for 10 min with successive 50 ml volumes of toluene.
Analysis by GC/ECD indicated the background contamination could be reduced
to acceptable levels by five 10 min extractions. Chromatograms for the
first, third, and sixth extractions are presented in Figures 3, 4, and 5,
respectively.
Since it was necessary to prepare a large number of polyurethane foam
plugs for field air sampling a study was made of the conditions required
to remove electron-capturing interferents by extraction of several foam
plugs simultaneously. Greatest success was attained by squeezing the foam
in toluene at 100°C in the bottom of a four liter beaker with a one liter
Erlenmeyer flask. Four extractions with redistilled toluene at 100° re-
duced the background contaminants to a level acceptable to GC/ECD. This
procedure is faster and more efficient than Soxhlet extraction.
A higher-density and less-crosslinked polyurethane foam, Olympic
dD
2315 ^ (Olympic Products Co., Greensboro, NC) was selected as the adsorbent
material for collecting chloronaphthalenes from ambient air primarily on
the basis of lower background on GC/ECD. Foam plugs, 5 cm diameter x 13
cm long, were cut from sheets of polyurethane foam with an electric knife.
Following five successive extractions with toluene at 100° for approximately
10 min per extraction the plugs were placed in individual 9 oz glass jars
and dried in vacuo at 50°C for 12 hours. Upon removal from the vacuum
oven, the jars were capped and wrapped with aluminum foil to prevent
photodegradation of the foam. This cleanup procedure was followed for all
foam plugs used in air sampling.
10
-------
Pulse
Attenuation
Column
100 u sec
128 x 1Q-'1 afs
180 x 0.2 cm 2% OV-101 on
100/120 mesh Gas Chroro Q
170°
300°
250"
30 cc/min
3
cc
M- 64 x NT11 alt
10
12
14
Time (min)
16
Figure 3. Gas liquid chromatography-electron capture
detection (63Ni) - foam clean-up; first
toluene extraction.
11
-------
Pulse
Attenuation
Column
I
41
-------
Pulse - 100 u sec
Attenuation = 32 x 10'11 afs
Column ='180 x 0.2 cm 2% OV-101 on
100/120 mesh Gas Chrora Q
Tcol - 200°
Tdet - 300°
Tini = 250°
Ng rlow - 30 cc/min
Time (min)
Figure 5. Gas liquid chromatography-electron capture detection (
foam clean-up; sixth toluene extraction.
13
-------
Each of the five toluene extracts of the foam was analyzed by GC/MS,
using the MID ions for the PCNs. Since the foam plugs were extracted in
groups of four, simultaneously, the extracts contained contaminants from
four plugs. The first extract contained large peaks in all channels, many
with long retention times (greater than that for CinCl0). Each successive
1U o
extract contained fewer and less intense peaks. The fifth extract (Figure
6) contained three peaks in the 164 channel and three in the 196 channel.
This background would not interfere substantially with the analysis of
samples containing a total of ^200 ng or more of monochloronaphthalene.
Dichloronaphthalene has a different retention time and was free of inter-
ference.
3.1.1.3 Collection of PCNs on Polyurethane Foam
(12 13")
The previous work on collection of PCBs using PUF ' ' indicated
that it may be the most suitable medium for collection of PCNs. Investiga-
tions under this project determined that PUF is an acceptable collection
medium for PCNs provided that the plugs are properly cleaned prior to use,
the flow rate through the plugs is low enough to prevent breakthrough, and
that some loss of lower chlorinated (particularly monochloronaphthalene)
PCN isomers is acceptable.
The breakthrough volume and recovery of PCNs from PUF plugs was
evaluated by treating a standard precleaned plug with 200 yg of Halowax
(S)
1014^ in toluene applied to one end of the foam plug, the solvent removed
in vacuo, and the treated foam plug placed with the treated surface up in
the air sampling train over a clean foam plug. Air was drawn through the
3
two tandem foam plugs at 147 Jl/min until 53 m of air had passed through
the sample. The two foam plugs were then extracted with toluene as above.
Only 20% of the Halowax 1014 ^ was found on the treated foam plug and none
on the back-up foam plug. When this low recovery was encountered, an
additional extraction with 200 ml of toluene at 80°C was made to make
certain the problem was not incomplete extraction. No additional Halowax
©
was recovered in this extract. The profile of the Halowax 1014 ^extracted
from the foam was distorted from the profile of the standard with the later
eluting peaks being diminished. This is illustrated in Figure 7 which
14
-------
53.0-1
M.fl-
•*T
196M
30.0
23.0-Si-^ii
• ..202*1
a
FILE l\
S3.3-
3:6*2
300«1
TILE £1 NO.
ifl/t 1-5.2XOVI01. iro«
Figure 6. GC-MS analysis of the fifth toluene extract of
a foam plug, multiple ion detection.
15
-------
Lit
(O
13
*§
o: u
uj S
Q '
K ^o
o£-
oo
u =
Q:
Column - 180 x 0.2 cm glass
2% OV-101 on 100/120
mesh Gas Clirom Q
Attenuation - 64 x 10-H afs
- 100 \isec
- 200°
•= 250°
- 300°
- 30 ml/mln
Column - 180 x 0.2 cm
glass 2% OV-101
on 100/120 mesh
Gas Chrom Q
Attenuation - 64 x 10~H
afs
Pulse - 100 psec
Tcol - 200"
- 250°
- 300°
Flow = 30 ml/min
Column - 180 x 0.2 cm glass
2Z OV-101 on 100/120
mesh Gas Chrom Q
Attenuation •= 84 x 10~H afs
1'ulse <• 100 iibuc
- 200"
- 250°
Tdct - 300°
N2 Plow - 30 ml/mln
2 4
I (min.)
2 4
t (min.)
2 4
t (min.)
Figure 7. Gas chromatography-electron capture detection analysis of H-1014 recovered
from polyurethane foam after air sampling and H-1014 standard. (A) Before
column clean-up (1 pi injection of a 30 ml sample); (B) After column clean-
up (1 yl injection of a 15 ml sample); (C) Halowax 1014 (5 ng injection).
-------
contains chromatograms of the foam extract, before and after removal of
interferences on a silicic acid column and a standard of Halowax 1014
Vacuum stripping of the chloronaphthalenes from the foam was a possible
explanation for the poor retention observed. Photodecomposition may also
have been responsible for the significant loss of applied Halowax . In
fact a subsequent experiment in which two air sampling trains were run
simultaneously implicated photosensitivity as the cause of some of this
loss. In this experiment one sampling train contained two tandem plugs
with 200 ]4g of Halowax 1014 ^ applied on the top plug and the other con-
tained a similarly spiked single plug. Both sampling units, shielded from
sunlight by wrapping with aluminum foil, were run as previously described
although a large difference in vacuum registered on the two pumps (<25 mm
Hg for one plug; >180 mm Hg for two plugs) no major difference in retention
of Halowax 1014^ was observed. Seventy five percent recovery was observed
for the two plug series and 64% recovery for the single plug. These
results suggest that the chloronaphthalenes may be photosensitive and that
retention is not only a function of the vacuum applied, but also a function
of flow rate.
Flow rate through the air samplers was found to have a significant
effect on the amount of Halowax which could be recovered. Table 1
presents the results of several experiments at different flow rates.
Experiment 1, indicates possible photodegradation of chloronaphthalenes.
The only difference between Experiment 1 and the two plug series of Experi-
ment 2 is that the plugs in Experiment 2 were shielded from sunlight. A
study by Lewis et_ al_. ' indicated that PCBs, particularly in the presence
of diethylamine, are sensitive to sunlight. Lewis and co-workers actually
employed this photosensitivity to facilitate determination of Mirex in
the presence of PCBs. Experiments 2, 3 and 4 show improved retention of
chloronaphthalenes at lower flow rates even when the sampling period is
extended to 24 hr. A 24 hr sampling period at 62 £ per minute reflects a
sampling volume of 90 m . For an atmospheric PCN concentration of 1 part
per trillion (ppt), 62 &/min would result in the accumulation of VLOO Ug of
Halowax^on the foam plug at 100% efficiency of collection. This flow
rate was employed for field sampling.
17
-------
Table 1. RETENTION OF HALOWAX 1014® AND 1051^ON OLYMPIC 213f
POLYURETHANE FOAM DURING AIR SAMPLING.
00
Flow Rate
Experiment Plug (£/min)
1
2
3
4
Top ^ 150
Bottom J
Top (only) 140
Top ^ 140
Bottom J
Top (only) 71
Top "^ 71
Bottom \
Top "1 62
Bottom )
Top ^ 62
Bottom j
Sampling Halowax Applied3 Halowax Recoveredb % Recovered
Period (hr) 1014 1051 1014 1051 1014 1051 Notes
6
6
6
6
6
24
24
232 pg
0
232
232
0
200
200
0
232
0
Blank
^27 pg
0
128
143
6.6
153
170
0
222 220
0 0
0 0
0
VL2 Apparatus not
shielded from
sunlight
55
62
..3
76
85
0
215 95 97
000
000
000
Halowax applied by pipetting standard onto top surface of plug followed by vacuum drying.
Halowax recovered by three, 100 ml extractions with toluene at room temperature for 5 minutes.
-------
One of the most important performance criteria for a sampling method
is collection efficiency and knowledge of this parameter is essential to
the interpretation of results in terms of atmospheric concentrations. To
define this parameter, an experiment was conducted in which the apparatus
illustrated in Figure 8 was employed. Halowax 1014 (464 yg) and Halowax
1051® (444 yg) were added to a slurry of 320 g of 3 mm glass beads in
hexane. After evaporating the hexane the loaded beads were divided into 2
x 100 g and 2 x 60 g portions. Each of the 100 g portions was placed in
sampling trains ahead of the glass fiber filter. Two PUF (Olympic 2315^0
plugs were placed in the sampling module and the pump operated for 24 hr
at 62 £/min. For four hours during this period the section housing the
glass beads on one of the apparatus was slightly warmed with heating tape.
Following the sampling period each of the foam plugs and glass bead portions
was extracted with toluene and the extracts analyzed by GC/ECD. Each of
the 60 g portions of glass beads was similarly analyzed to determine the
actual loading of Halowax . The loading was observed to be non-uniform
hence no quantitative conclusions can be drawn for this experiment about
the collection efficiency of chloronaphthalenes on foam, however, some
interesting results concerning volatility were revealed. The components
of Halowax 1014 were observed to be much more volatile than the components
of Halowax 1051 . In general, volatility appeared to decrease with
increasing chlorination, and is greatly increased by a slight rise in
temperature. Semi-quantitatively speaking, chloronaphthalenes which were
desorbed from the glass beads were proportionately adsorbed on the first
foam plug. No chloronaphthalenes were detected on the second plug.
In a repeat experiment under more controlled conditions, 0.5 ml each
of mono-, di-, tri-, tetra- and octachloronaphthalene standards were
pipetted into a common vessel, the total volume reduced to 0.5 ml, and the
PCNs quantitatively transferred to glass beads contained in the sampling
apparatus displayed in Figure 8. This resulted in the application of 116
yg of 2-chloronaphthalene, 176 yg of 1,2-dichloronaphthalene, 208 yg of
1,2,3-trichloronaphthalene, 366 yg of 1,2,3,4-tetrachloronaphthalene and,
o
210 yg of octachloronaphthalene to the beads. After drawing ^90 m of
ambient laboratory air through the samplers at 62 £/min the glass beads,
19
-------
STAINLESS STEEL COUPLING
GLASS FIBER F'LTER
MESH SUPPORT
POLYuRETHANE
FOAM PLUGS
PYREX GLASS BEADS
WIRE MESH SUPPORT
COUPLING BOLTS
WIRE MESH SUPPORT
SCALE }>. = I i
in.
Figure 8. Sampling apparatus configuration for collection
efficiency determination.
20
-------
glass fiber filter and foam plugs were analyzed. The results presented in
Table 2 reveals several interesting phenomena. Firstly, it appears that
at the temperature (16°C) and flow rate (^62 A/min) of the sampled air
monochloronaphthalene passes through both foam plugs. This is substantiated
by the fact that 78-97% of dichloronaphthalene was found on the second
foam plug and approximately 100% of trichloronaphthalene was found on the
first foam plug. This is undoubtedly a result of decreasing volatility
with increasing degree of chlorination. Although the lower air temperatures
during field sampling would have reduced this volatility somewhat, concen-
trations of monochloronaphthalene observed in air samples must be regarded
as minima. Secondly, only about one-half of the octachloronaphthalene was
removed. The possibility of adsorption to glass surfaces prompted rinsing
of interior surfaces of the glassware used in the sampling apparatus with
toluene. However, no increase in recovery was observed. Photosensitivity
of the higher chlorinated PCN isomers, a phenomenon suspected from previous
experiments discussed above and reported as a destruction mechanism for
PCBs ' , may be responsible for this unexplained loss and prompted
further investigation.
Table 2. COLLECTION EFFICIENCY OF POLYCHLORONAPHTHALENES
Glass
Glass
PUF No
PUF No
TOTAL
Glass
Glass
PUF No
PUF No
TOTAL
Beads
Fiber Filter
. 1
. 2
Beads
Fiber Filter
. 1
. 2
m/e 164
.47
a
.89
1.5
2.86
.95
.044
1.18
1.57
3.75
% Recovered From Each Medium
m/e 196 m/e 230 m/e 266
1
6
77
85
2
10
96
110
.36
.06
.63
.8
.9
.91
.067
.9
.8
.7
6.
112.
.
120.
13.
.
95.
.
109.
88
11
3
73
0
7
14
2
286
3
26
88
115
45
-
59
-
104
.6
.17
.3
.16
.0
.6
-
.3
-
.9
m/e 404
55.2
—
—
55.2
68.2
—
—
—
68.2
detected
21
-------
To assess photosensitivity a mixture of chloronaphthalenes (2-
chloronaphthalene, octachloronaphthalene and Halowax 1014^— about 500
yg/ml) in toluene was irradiated for eight hours with ultra-violet light
from a germicidal lamp. The PCN mixture was analyzed by GC/MS before and
after irradiation. No significant differences were observed and, therefore
it does not appear that PCNs unaccounted for in collection efficiency and
recovery studies were lost through photodecomposition.
3.1.1.4 Water Samplers
Water samplers were used without modification (vide infra). The
efficiency of collection was tested to assure that PCNs were not adsorbed
to the walls of the tubing.
The water sampling procedure was tested by sampling an aqueous solu-
tion spiked with Halowax 1014 ^ and octachloronaphthalene (about 250
]Jg/&). The extracts of aliquots of water before and after sampling were
analyzed by GC/MS and the integrated areas compared. The average concen-
trations were actually found to be greater in the sampled water than in the
unsampled as shown in Table 3. Thus, loss of PCNs through adsorption to
the tubing or other losses during sampling were assumed to be negligible.
Table 3. EFFECTS OF INTERVAL SAMPLING ON THE
CONCENTRATION OF PCNs IN WATER
Ratio
Compound (Sampled/Unsampled)
C10H5C13 0.83
C1()H4C14 0.89
C1()H3C15 1.03
C10H2C16 1.34
C10HC17 2.27
C10C18 1.49
Mean 1.31
S.D. 0.54
22
-------
3.1.2 Selection of Sampling Sites
Sites were selected to represent industries which manufacture and use
PCNs. Use sites were selected within the electrical capacitor industry,
both the manufacture of capacitor tissue and the capacitors themselves.
To this end, a computer search (Predicast), Thomas Registry, Chem. Sources,
(8 1^
Chemical Buyers Guide, two EPA reports^ ' , Cramer Electronics Catalog,
Pioneer Standard Electronics Catalog and the various state industrial
(8 1 3^
directories were consulted. The sole domestic producer of PCNs ' ' is
Koppers Company, Inc. at the Koppers Chemical and Coatings Plant, Bridge-
ville, PA, so this represented the manufacturing site for sampling.
Manning Paper Co., Green Island, NY was selected as a representative
capacitor tissue manufacturer. It is listed ' as a producer of
electrical insulating papers. General Electric Company, Hudson Falls, NY;
General Electric Co., Ft. Edward, NY; Sprague Electric Co., North Adams,
MA, Cornell Dubilier Electronics Corp., New Bedford, MA; and Cornell
Dubilier Electronics Co., Sanford, NC were selected as representative
capacitor manufacturers. At these sites, a variety of capacitors ranging
from automobile capacitors to electric power station capacitors are manu-
factured. With the exception of Koppers Company, Inc., there was no
direct published evidence that any of the above listed manufacturers use
PCNs. A detailed discussion of the known activities and other relevant
information appears in the sampling and analysis section for each sampling
site.
3.1.3 Sampling Protocol
The sampling protocol was developed to collect, air, water, soil,
sediment, vegetation, and aquatic organism samples which would effectively
assess the environmental contamination by PCNs near the suspected manu-
facturer and use sites. For all media, the prevailing philosophy of the
sampling protocol was to collect samples which would be likely to yield
positive results and which would reflect the extent of environmental
contamination and its effect (either directly or indirectly) on humans.
23
-------
3.1.3.1 Air
Sampling locations were selected within four quadrants around the
sampling site. Selection of locations along transects of the suspected
PCN source was made after surveying the site for wind speed and direction,
air stability, terrain, population, and site accessibility. At each site
four samplers were deployed along the arms of two orthogonal transects
with two sets of air samples collected over consecutive 24 hr periods.
During the entire collection time, meteorology was monitored continuously
at a representative location for the entire site and intermittently at
each sampling location.
3.1.3.2 Water
Where appropriate, water samples were collected using the interval
samplers. Generally, samples were collected simultaneously upstream and
downstream of the suspected PCN source. Suspected sources were identified
by surveying the site for plant outfalls, local inquiries, inquiries at
the wastewater treatment plant serving the plant, and in some cases,
discussion with EPA Regional personnel. The samplers were deployed in the
stream or river such that the inlet of the sampling tubing was well-
removed from the river bank and in the main flow as much as possible.
Water was generally collected from 10-30 cm beneath the surface with the
samplers calibrated to pump 50% of a 15 min duty cycle for 24 hours. This
resulted in the collection of about 4 £ in 24 hours.
When interval sampling was inappropriate, 1 £ grab samples were
collected. These situations included mud puddles, stagnant ditches, ocean
bays and streams which were not amenable to interval sampling.
3.1.3.3 Soil
Using the transects set up for air sampling, soil samples were collec-
ted at "near" and "far" locations along each of the four arms of the
transects. Locations were selected according to meteorology, terrain,
population, and accessibility. At each location, several core samples
were taken within a 10-50 m radius which were representative of the location.
Generally, attempts were made to obtain samples from high and low ground,
24
-------
grassy and barren, under trees and in the open, etc. Every attempt was
made to collect samples from areas which did not appear to have been
recently disturbed. Each core sample was stored in a glass jar with a
foil-lined cap and returned to the lab where the samples from each location
were composited for analysis. During the first sampling trip (Koppers),
ten core samples were collected at each location. It was the opinion of
all personnel involved that this protocol was probably not necessary for
obtaining a representative sample of the location, so on subsequent sampling
trips only 3-5 core samples (at the discretion of the sampling personnel)
were collected.
3.1.3.4 Miscellaneous
Sediment, garden vegetation, and aquatic biota were collected according
to their availability. Sediment was usually collected in conjunction with
water samples. At the discretion of sampling personnel, other samples of
interest were collected — generally industrial solid waste. Details of
these samples and their collection are included in the sampling protocol
for each site.
3.1.4 Meteorology
3.1.4.1 Measurement s
The macrometeorology and micrometeorology were recognized as important
factors in the transport of PCNs. With respect to air sampling, the
specific meteorology of the site was important during the entire sampling
period. Continuous measurements of temperature, wind direction and wind
run were recorded on an MRI Mechanical Weather Station (MRI) at a central
location throughout the sampling period. The strip chart record allowed
subsequent reference to calm periods, wind direction shifts, etc., which
could affect PCN transport in air. Intermittent measurements were made
approximately six times daily at each sampling location. Humidity, tempera-
ture, wind speed, and wind direction were measured and noted along with
general conditions (rain, snow, cloud cover, odors, etc.). Measurements
at each location provided a cross-check with the MRI record.
25
-------
3.1.4.2 Macrometeorological Data
Pertinent meteorological data for the sampling period were obtained
from the nearest National Weather Service office. This data was used to
augment that obtained at the sampling location.
Average annual and seasonal data were obtained from the National
Climatic Center, Asheville, NC for construction of wind roses. The wind
rose was used to correlate the theoretical PCN deposition with that found
by sampling and analysis.
3.1.5 Sample Storage
The samples were preserved in the state in which they were collected
as nearly as possible. To prevent degradation by microbial action or
volatilization losses, all samples were kept cold. Solids (soils, foam
plugs, etc.) were frozen and aqueous samples were stored at 5°C.
3.2 ANALYTICAL METHODOLOGY
The analysis of PCNs in environmental samples requires special method-
ology due to a number of factors. The volatility of PCNs spans a wide
range and care must be exercised to prevent loss of lower chlorinated
species. Because of the potential presence of a large number of chlori-
nated pesticides and PCBs in samples, the analytical method must remove as
many interferents as possible and be sensitive to PCNs in the presence of
inseparable interferents (notably PCBs). These factors were considered in
the selection and development of all analytical procedures.
Samples were analyzed by GC/MS using multiple ion detection after
extraction and cleanup. The specific methods used are detailed in Appendix
A. All methods employed were tested and evaluated during the course of
this project. Initially, perchlorination of PCNs followed by GC/ECD
analysis was attempted. Perchlorination, a technique often used for PCB
analysis, would involve conversion of all PCN isomers to octachloronaphtha-
lene to yield a single GC peak with a long retention time, hopefully removed
from that of interferents. Perchlorination was found to be fraught with
difficulties including side reactions, low recoveries, and inconsistent
reaction yields, and hence was abandoned. The GC/MS method ultimately
26
-------
employed allowed positive identification of individual PCNs and permitted
PCN identification even in the presence of much higher concentrations of
PCBs.
The following discussion details the analytical method development and
presents the results of method validation tests.
3.2.1 Apparatus, Instruments and Chemicals
3.2.1.1 Air Equipment
Air samples were collected on 5 cm diameter x 13 cm long polyether
type polyurethane foam plugs cut from sheets of Olympic 2315 (Olympic
Products Co., Greensboro, NC) using an electric knife. The foam plugs were
preceded in the air stream by 11 cm diameter glass fiber filters (Gelman
Type A-E). The sampling media were contained in an aluminum foil-wrapped
glass module constructed of Kimax glass process pipe and fittings (Kimble
Products, Toledo, OH). Air was drawn through the sampling module by a
custom manufactured Nutech Corporation (Durham, NC) sampler which used a
Cast oilless rotary vane vacuum pump Model No. 1022.
Air samples for collection of volatile and semi-volatile species were
collected on Tenax GC cartridges using a Nutech Model 221A sampling pump
(Nutech Corporation, Durham, NC) as described by Pellizzari. ~
3.2.1.2 Soil Equipment
Soil sample cores (5 cm diameter x 15-20 cm long) were cut using a
common garden bulb planter. The samples were placed in labeled one-quart
wide-mouthed jars with foil lined caps (Fisher Scientific).
3.2.1.3 Water Equipment
Unfiltered water samples were collected in amber four-liter bottles
with teflon-lined caps using a S7576 AC Interval Sampler (Horizon Ecology
Co., Chicago, Illinois) operated on a duty cycle of 7 1/2 minutes every 15
minutes (50%) at a flow rate of about 5 ml/min. Teflon tubing (3 mm id)
was used, except for a short section of silicone tubing necessary for
proper operation of the peristaltic pump. The pump and collection bottle
were housed in a custom-built wooden box for security. A 24 hr sampling
period was sufficient to nearly fill the sampling container.
27
-------
In cases where interval sampling was inappropriate (e.g., mud puddles,
lakes, etc.) grab water samples were collected in one-liter wide-mouthed
bottles with foil-lined caps (Fisher Scientific).
3.2.1.4 Meteorological
Continuous meteorological measurements (wind run, wind direction, and
temperature) were recorded using a MRI Mechanical Weather Station (Meteoro-
logical Research, Inc., Altadena, CA). Intermittent humidity and temperature
readings were taken using a sling psychrometer (Taylor Instrument Co.,
Rochester, NY). Wind direction and speed were measured using an anemometer
(Edmund Scientific Co., Barrington, NJ) and compass.
3.2.1.5 Chemicals
All solvents used were distilled in glass (Burdick and Jackson,
Muskegon, MI) and used without further purification. Where purity was
especially critical, solvents were redistilled in glass. Halowax mixtures,
Aroclor ^ mixtures, and pesticides were obtained from the Quality Assurance
Section, Environmental Toxicology Division, EPA, HERL, Research Triangle
Park, NC; individual PCN isomers were obtained from RFR Corp., Hope, RI;
silica gel was obtained from Davison Chemical Division, W. R. Grace, Balti-
more, MD and d---anthracene from Merck and Company, Rahway, NJ.
3.2.1.6 Analytical Instrumentation
Analysis of all samples for PCNs was accomplished using a Finnigan
3300 quadrupole GC/MS with a PDP/12 computer. Volatile and semi-volatile
organics were analyzed using a Varian CH-7 GC/MS with a Varian 620/L computer.
Gas chromatography-electron capture detection analyses were performed on a
Fisher Victoreen Series 4400 gas chromatograph.
3.2.2 Sample Workup Procedures
3.2.2.1 Perchlorination Methods
Perchlorination has been successfully applied to the analysis of
PCBs ~ , where all PCB isomers are converted to decachlorobiphenyl for
(25-27)
GC/EC detection. The procedure is also reported to work for PCNs.
(27)
In our laboratory, however, the methodology recommended by Analabs
28
-------
( 28)
using the procedure of Safe et al. did not prove useful in perchlori-
(S) (25)
nation of Halowax mixtures. Using the procedure of Armour , the
(E)
reaction was observed to convert Halowax mixtures to octachloronaphthalene
(C1AC1Q). However, this conversion was neither consistent nor complete
10 8
with recoveries generally from 10-50%. After expending considerable effort
on this technique with no foreseeable positive results, and with approval
of the project officer, the perchlorination procedure in combination with
GC/ECD was abandoned in favor of GC/MS. Details of the perchlorination
efforts and results are included in Appendix B.
3.2.2.2 Extraction of PCNs from Water
Toluene was chosen as the extraction solvent based on PCN solubility
studies conducted in this laboratory. The extraction procedure is as
follows: In a separatory funnel, shake a 200 ml aliquot of the water
sample with 25 ml toluene for five min. Repeat the extraction twice more
(total toluene volume = 75 ml) and dry over Na-SO,. Evaporate the solvent
just to dryness, reconstitute the sample in 1.0 ml hexane and proceed with
column chromatography clean-up. A detailed procedure is listed in Appendix
A.
Several extractions of water samples spiked with PCNs were conducted
to check recovery (Table 4). Polychlorinated naphthalenes were pipetted
into 200 ml tap water in hexane or toluene solution (1.0 ml). After thorough
mixing, the extraction was conducted as described previously. The mean
extraction efficiency is 0.90 H- 0.19 (see Table 4).
Since it is often found that extraction of organic compounds from
water samples is highly pH-dependent, the efficiency of extraction of PCNs
from water at different pH values was tested. The results of this experiment
are summarized in Table 5. The aqueous phase (2.0 ml) and 0.5 ml toluene
containing C^Clg at 140 pg/yl were shaken together in a vial and equili-
brated for four hours at 60°C with occasional shaking. The organic layer
was then analyzed directly by GC/ECD. No significant dependence between
recovery and pH was observed, indicating the extraction of water samples
may be carried out without adjusting the pH. Consistent recovery of greater
than 100% is attributed to solvent evaporation during the experiment.
29
-------
Table 4. RECOVERIES OF POLYCHLORONAPHTIIALENES FROM WATER
co
o
Sample
No.
K-I
K-II
K-III
K-IV
K-V
K-VI
Spiked
Substances
C10C18
C10C18
C10C18
H-1014
H-1051
H-1000
Amount
Spiked
(Pg)
0.28
14
14
11.6
11.1
4.1
Extraction
Solvent
Hexane
Toluene
Toluene
Toluene
Toluene
Toluene
Percent Extracted
First Second Third Fourth Total
80
60 9 2 2 72
67.5 6 1 75
117 <5 <2 - 'VL2°
94 11 1 - 106
71 12 6 - 89
Mean 90.5
SD 18 . 8
RSD 20.8%
-------
Table 5. EXTRACTION OF HALOWAX 1014 ^ FROM AQUEOUS
SOLUTION WITH TOLUENE vs. pH
Aqueous Phase % Recovery
0.5M NaOH 114
pH = 10.00 buffer 105
pH = 7.00 buffer 106
Distilled water (pH =6.5) 119
pH = 5.08 buffer 104
3M HC1 123
To determine the efficiency of extracting chloronaphthalenes from
natural aqueous media, one gallon of very turbid water was taken from a
stream in Durham County, NC about one mile below a municipal waste water
treatment plant outfall. To one 200 ml aliquot was added 400 yg of Halowax
1014 , to another was added 4.64 yg of Halowax 1014 and 4.44 yg of
Halowax 1051 . After storage for 36 hours at 4°C, 86% recovery was observed
at the high concentration and 83 and 118% recovery, respectively, at the
low concentrations.
3.2.2.3 Extraction of PCNs from Soil
Prior to analysis, individual soil plugs were composited for each soil
sampling location by combining one-half of the top 2.5 cm of each plug as
illustrated in Figure 9. These semi-circular soil portions were subsequently
broken into smaller pieces and pulverized by vigorous shaking. Fifty gram
aliquots were removed from these composites for analysis. The soil extrac-
tion procedure was adapted from a method reported for pesticides, PCBs and
(29)
PCNs , and tested for its applicability to this research project. Local
soil samples were spiked with 20.4, 23.2 and 22.2 yg of Halowax 1000, 1014
and 1051 , respectively. The extraction of chloronaphthalenes from soil
was accomplished using the procedure described in Appendix A with one
modification: two parallel extractions were conducted, one using a hexane-
acetone mixture and the other a toluene-acetone mixture. Following the
31
-------
15-20 cm <
2.5 cm
5cm
Figure 9. Sectioning of soil plugs.
32
-------
extraction, quantitation was achieved using GC/ECD by comparison with
standards. The results, presented in Table 6, indicate no appreciable
difference in extraction efficiency between the two procedures. The toluene-
acetone method was chosen for its consistency with methods employed for
water and polyurethane foam.
3.2.2.4 Extraction of PCNs from PUF and GFF
Previous research on collection of airborne pesticides and PCBs on PUF
(28 29) (14)
used either petroleum ether ' or 5% ether in hexane to extract the
chlorinated hydrocarbons from the foam. Initial efforts using hexane to
clean-up the PUF (vide supra) were totally unsatisfactory, prompting
further investigation. A test of the solubility of Halowax and Aroclor
mixtures in hexane, acetone and toluene revealed by far the greatest solu-
bility in toluene. Halowax and Aroclor mixtures (10 mg) were instantly
soluble in 100 yl of toluene whereas solvation by hexane and acetone was
very slow and in some cases incomplete at room temperature.
Previous researchers have used Soxhlet extractors to extract
PCBs from PUF. Early experiments in this research indicated that Soxhlet
extraction was too time-consuming for the large number of PUF sampling
plugs needed and a more efficient manual extraction was developed.
To this end, a pilot study of the extraction of Halowax 1014 ^ from
(S)
a 1 g piece of Olympic 4214 VEy PUF was conducted by adsorbing 10 yg (10 ml
of a 1 yg/ml solution) of Halowax 1014 ^ onto a 1 g piece of foam which was
pre-cleaned as described above. Solution not retained by the foam was
drawn into the pores by compression and expansion of the foam. All solvent
was removed by placing the foam, contained in a loosely covered beaker, in
a vacuum oven at 40° for one hour. Extraction of the adsorbed Halowax^
was accomplished with successive 30 ml portions of toluene at room tempera-
ture for 10 min. After concentration of the extracts to 10 ml, 1 yl
aliquots were analyzed at 200°C on the 2% OV-101 column using GC/ECD.
Quantitation was attained by peak height comparison with a peak at 2.6 min
for the Halowax 1014^ mixture (0.2 mg/yl). The first, second and third
extracts contained 6.07, 1.48 and 0.4 yg, respectively, indicating a total
recovery of 79.5%. Chromatograms of these extracts appear in Figures 10,
11, and 12, respectively. No losses have been attributed to concentration
techniques.
33
-------
Table 6. EXTRACTION EFFICIENCY OF CHLORINAPHTHALENES FROM SOIL
Percent Recovery
Hexane-Acetone
To luene-Ac e tone
C1()H7C1
64
84
C10H6C12 C1QH5C13 C1()H4C14
77 - 160
46 - 115
C10H3C15
66
79
C10H2C16
86
122
C1()HC17
83
105
C10C18
98
119
Average
90.6
95.7
co
-F-
-------
\
Pulse
Attenuation
Column
Tool
Tdet
ini
j fl
ow
100 v sec
32 x 10-1 1 afs
180 x 0.2 cm 2" OV-101 on
100/120 mesh Gas Chrom Q
170°
300°
250"
30 cc/min
_L
4 6
Time (min)
10
Figure 10. Gas liquid chromatography-electron capture detection ( Ni)
first extraction of Halowax 1014 from foam.
35
-------
Pulse = 100 u sec
Attenuation = 64 x 10-11 afs
Column = 180 x 0.2 cm 2% OV-101 on
100/120 mesh Gas Chrom Q
« 200°
= 300°
= 250'
ow = 30 cc/min
Tdet
Tin]
N2 fl
246
Time (min)
Figure 11. Gas liquid chromatography-electron capture detection (
second extraction of Halowax 1014 from foam.
63
36
-------
5*
O
5;
-------
Extending this evaluation to standard size plugs, the recovery of a
1.0 yg sample of Halowax 1014^applied to Olympic 2315 ^"was found to be
100% when extracted four times for five min with 200 ml toluene. The
sample was applied to the foam in toluene and the solvent removed in vacuo.
(R)
It should be noted that all of the Halowax 1014 ^was removed in the first
two extractions.
Another factor which may influence PCN recovery is the penetration of
PCNs into the foam. The latter was evaluated by injecting Halowax 1014
into the center of a plug with a glass syringe. After drying in vacuo, 225
yg (97%) of the Halowax 1014 ^was recovered.
As used for the extraction of field samples, the PUF extraction proce-
dure not only quantitatively recovered PCNs, but was quick, and required a
minimum of glassware. To assure complete recovery of PCNs collected in
the field, the GFF was analyzed in a manner similar to that for PUFs.
3.2.2.5 Column Chromatography
The column chromatography procedure used for cleaning up PCN extracts
from foam plugs, water samples, and soil samples was adapted from that
de
A.
(29)
described previously. The procedure is described in detail in Appendix
The results of several trials, summarized in Table 7, show that the
column clean-up procedure easily removed most pesticides and polyurethane
foam background without loss of PCNs. In addition to analysis of indi-
vidual fractions by GC/ECD, the total hexane fractions were combined and
analyzed to provide additional recovery data. The wide variance in results
obtained from GC/ECD is due mostly to errors in the sample injection volumes.
This column clean-up procedure removes all recognized potential interferents
tested (see footnote "c", Table 7) except aldrin and PCBs.
3.2.2.6 Volatile Organics from Aqueous Samples (VOA)
A few water samples suspected of containing appreciable quantities of
volatile organic materials were analyzed according to the modified VOA
(30—32)
procedure outlined below. The volatiles were purged from the
sample and adsorbed on Tenax GC. The Tenax was then analyzed by GC/MS/COMP
using a Varian CH-7 interfaced with a high resolution glass capillary GC
column.
38
-------
Table 7. SUMMARY OF COLUMN CHROMATOGRAPHY CLEAN-UP EXPERIMENTS
Sample
Number
KVII
KVIII
KIX
Substance Amount Hexane
Chromatographed (yg) Fraction
C10C18 28 1
3
4
Total
Combined
H-1014 23.2 1
2
3
4
5
6
Total
Combined
Combined
H-1014 23.2 1
Concentrated 2
Foam Extract 3
4
Total
Combined
Combined
Volume
(ml)
17
22
10
5
54
54
5
13
10
10
10
10
58
20
11
12
12
55
55
55
Analysis
(yg)
22.3
3.2
0.1
0.06
26.2
36.5
0
7.9
7.9
1.2
0.1
0
17.2
32.7
21.9
15.0
4.7
0.6
0
20.2
24.8
17.3
Percent Toluene Volume Analysis
Recovery Fraction (ml) (yg)
94
130
74
141
95
87
107
75
1
2
10
10
a
b
-------
Table 7. (cont'd)
•t-
o
Sample
Number
KX
KXI
Substance Amount Hexane
Chromatographed (ug) Fraction
H-1014 23.2 1
2
3
4
5
Total
Combined
Combined
Pesticide0 0.1-0.2 1
Mixture of each 2
3
4
5
Volume
(ml)
14
10.5
11
10.5
12
58
57
57
10
10
10
10
10
Analysis
(mg)
11.5
9.2
0.9
-0.3
0
21.9
21.0
22.6
0
0
54
42
6
Percent Toluene Volume Analysis
Recovery Fraction (ml) (|jg)
94
91
98
Total
50
102
103
1
2
3
4
KXII
H-1014
Pesticide
Mixture
4.6
1-2
1
2
3
4
5
10.0
9.8
9.9
10.6
10.7
*•«£
0.2J
i
Total
51.0
4.2
91
1
2
3
6.2
11.9
10.5
5.6
10
10
10
e
f
g
o
j
k
1
-------
Table 7. (cont'd)
NOTES
aPeak at RRT =0.95 from foam extract is eluting
In addition to peak at RRT = 0.95, peak at 0.80 is observed. These two
compounds are the most noticeable features of the foam extract.
CPesticide Mixture Lindane (0.07 yg) , heptachlor (0.11 yg), aldrin (0.10 yg),
p,p'-DDD (0.20 yg), and p,p'-DDT (0.25 yg) .
Aldrin is only pesticide eluted with hexane.
Q
Small amount of heptachlor.
All pesticides except aldrin.
o
Some dieldrin detected.
Aldrin off scale, halowax present as measured.
Aldrin on scale, small amount of heptachlor.
All pesticides, except aldrin. Dieldrin concentration low.
Dieldrin predominates. Some traces of other pesticides.
Very small amount of dieldrin.
41
-------
In order to avoid contamination, all glassware used in the purging
apparatus was washed and heated to 400°C in a glassware oven, with the
exception of the fritted glass bubblers and the thermometers. Both were
soaked in dilute HNO,,, then the thermometer rinsed well with distilled
water and distilled water pulled through the bubblers. Clean Tenax GC
cartridges (10 cm x 1.5 cm i.d.) were desorbed at 270°C for 20-30 min under
a stream of helium to remove contaminants, then cooled to room temperature
in a capped glass tube.
The sample (250 ml) was placed in a 250 ml round bottom, three-necked
flask. The necks were occupied by a thermometer, a fritted glass bubbler
and a condenser containing a small plug of oven-treated glass wool (see
Figure 13). With the flow rate adjusted to 25 ml/min, the samples were
purged at 40-45°C for 1.75 hour, including the time spent coming up to
temperature. Loaded cartridges were stored in a freezer awaiting analysis
by GC/MS.
This method has been validated by radioisotope recovery methods for a
number of organic compounds including acetone, acetonitrile, benzene,
(32)
toluene, phenol and dimethylbenzanthracene.
3.2.3 Instrumental Analysis and Data Reduction
3.2.3.1 GC/MS of PCNs
Analysis of all samples for PCNs was accomplished using a Finnigan
3300 quadrupole GC/MS with a PDP/12 computer. The 180 cm x 2 mm i.d. glass
column, packed with 2% OV-101 on Chromosorb W was held at a temperature of
150° for three minutes, programmed to 230° at 8°/min and held isothermally
until all peaks had eluted. The individual chloronaphthalenes were well
resolved and the last peak (C,0Clg) eluted in approximately 18 min as
illustrated in Figure 14. The flow rate was 30 cc/min, helium. The ioniza-
tion voltage was nominally 70 eV and detector voltages were between 1.8 and
2.2 kV- Full scan spectra were obtained from m/e 110-500 and MID ions were
set at the nominal masses discussed below. Exact mass settings were made
using a standard PCN mixture which also served as an instrument check and
as a cross-check on retention times.
42
-------
Tenax GC Cartridge
Teflon Fitting
Condenser
Thermometer
He Input
Bubbler
Figure 13. Apparatus for VGA purge.
43
-------
lea.a-.
43.
.e-
£r*. 196*1
' 188*1
«B-a-]^^.^,e_.j;;L/%^T3^^^^ 266*1
' '• r-^i i-i'iv-. /^"S- r. I - V.— • •* -.< -r- <'.' f • .•-yVrv^ '^'•"''-fiVi.;;r.Vi . 'i' .Vi'i ", n 9^n*l
230*1
isa.a-
88.5-
sa.a-
43.3-
z 5 19
FILE El NO. 5
f1tX.2;:OViaui7Q»5M.4*^n.-9.2.2>CV,4NGxrCL.OCT26
IS
404*5
368*3
336*2
300*1
Figure 14.. GC/MS of a Halowax 1014 , Halowax 105r5^
and d^Q-anthracene mixture illustrating
temperature programming.
44
-------
Analysis of PCNs by GC/MS provides positive identification of PCN
isomers and allows analysis even in samples where interferents (such as
PCBs) are present in much greater concentration. Gas chromatography/mass
spectrometry represents state-of-the-art analytical methodology and is
clearly the method of choice in terms of specificity and information
content.
Samples were analyzed using the multiple ion detection mode to provide
maximum sensitivity. The results were, in many cases, confirmed by measuring
the chlorine isotope ratios by MID using different ions or (if the concen-
tration was sufficient) by full scan mass spectrometric measurement. The
details of the analytical method are discussed below and in Appendix A.
Multiple Ion Detection—Multiple ion detection is an operational mode
for a quadrupole gas chromatograph/mass spectrometer where up to nine m/e
values are monitored through the chromatogram. Preset voltages for each
m/e value are step-jumped at short time intervals. Since this technique
allows integration of ion intensity for a longer time period for the desired
ions than in the customary full-scan mode, the sensitivity of the instrument
is increased by approximately two orders of magnitude. By judicious selec-
tion of m/e values to be monitored, interference by unwanted compounds can
usually be minimized.
Eight ions were selected for monitoring: one from the parent cluster
for each of the eight chlorinated naphthalenes (C1QH7C1-C Clg). Although
the parent ions were not necessarily the most intense, the probability of
interference by PCN fragment ion or other contaminants was reduced. Ions
were chosen from the M (parent), M + 2, or M + 4 m/e values (Table 8)
according to an optimum combination of greatest intensity and least inter-
ference from PCBs and other PCNs. The ions selected are listed in Table 8.
As expected, the MID scans of the Halowax mixtures (ca. 4 ng) in Figures
C-l-C-14 of Appendix C closely matched the total ion current (TIC) plots
from normal GC/MS runs. Gas chromatography/mass spectrometry using MID is
clearly sensitive to 4 ng of Halowax^mixture. This sensitivity is com-
pared with the 200 ng amounts needed for good quality full scan spectra.
Interferences—MID scans of Aroclorwmixtures (ca. 20.0 ng) were ob-
tained using the m/e values selected for PCN quantitation (Figures C-15-C-37
45
-------
Table 8. POLYCHLORONAPHTHALENES PARENT ION AND MID ION (m/e) VALUES.
C1()H7C1
C10H6C12
C1()H5C13
C10H4C14
CinH_Cl,
10 3 5
C10H2C16
C H Cl
ID 1 /
C10C18
M
162
196
230
264
298
332
366
400
M + 2
164
198
232
266
300
334
368
402
M + 4
166
200
234
268
302
336
370
404
MID Ion
164
196
230
266
300
336
368
404
"Double" MID
162, 164
196, 198
230, 232
264, 266
298, 302
332, 336
366, 370
400, 404
Theoretical
Response Ratio
100/33
100/65
100/98
100/131
100/106
100/161
100/224
100/298
-------
of Appendix C) to test the analytical specificity. For the most part, the
PCBs are not detected, even at a 50-fold greater concentration than the
PCNs, although several PCB peaks were observed in the 164 and 196 channels
which represent fragments of higher molecular weight molecules. This
represents a potential interferent in PCN detection, however comparison of
retention times indicates the peaks are not at retention times for all
isomers of mono- and dichloronaphthalenes.
Using the MID ions for PCNs, a mixture of chlorinated pesticides
(lindane, heptachlor, aldrin, heptachlor epoxide, dieldrin, p,p'-DDD, and
p,p'-DDT; 6-25 ng each) was analyzed and gave several early-eluting peaks
which might interfere with PCN analysis (Figures 15-17). Since all of
these compounds, except aldrin are removed by the silica gel column cleanup,
the potential for interference is eliminated. Aldrin may present diffi-
culties as it elutes early in the gas chromatogram and gives a sharp peak
in most MID channels (Figure 17). This potential interferent in the analysis
of field samples was noted, but not detected.
External Standard—An external standard, dlf.-anthracene, was added to
standard mixtures and samples for retention time correlation and quantitation.
The d.. --anthracene was monitored in the MID mode at m/e 188, its parent
ion. This compound eluted relatively early in the chromatogram and presented
no potential interference. The results of an analysis of a mixture of
Halowax-1014 , Halowax-1051 ^ and d.--anthracene (Figure 14) illustrates
this utility.
MID Response Linearity and Detection Limit—A mixture of 2-chloronaphtha-
lene, Halowax 1014^, octachloronaphthalene, and d. n-anthracene was prepared
at several different concentrations and the MID responses measured. The
response of each PCN isomer was calculated relative to that of the d._-
anthracene to normalize injection volume and instrumental variations. The
results for the selected isomers C^E^I (m/e = 164), C10H5C13 (m/e = 230),
C10H2C16 (m/e = 300), and C^Clg (We = 404) are plotted vs. concentration
in Figure 18. Linear regression analysis (Table 9) indicates that, within
experimental error, instrumental response is linear from <1 ng to at least
500 ng, which approaches the solubility limit of PCNs in hexane. The
detection limit has not been specifically evaluated, but <50 pg of an
47
-------
ICO.O-i
00
.0-
ico.a
"""''"' ^^::^^-^!^i>:?r>^^^ 196 •
T™*"^
10
15
RU EI MO. 1
?tir.fllX.2::OViei. t99»C.30CC/M.-3. l.aKV.9-2
Figure 15. GC/MS analysis of a pesticide mixture using multiple ion
detection with PCN ions.
-------
100.0-
ea.e-
60.0-
40.0-
20.0
nJjX'r'xV.Vt''l..'V ,.,'rf, ..'','
'2£^£££££%£^^ * 3(! iU J
FILH El HO. I
PEt,r.l1lX.2XOViai.l9e*C.30CCxM.-9,
Figure 16. GC/MS analysis of a pesticide mixture using multiple ion detection
with PCN ions.
-------
80.0-
60.0-
40.0-
20.0-
.6-
lao.a-
m
o eo.0-
60.0-
-J.' i£>. Q t* -40.0-
£/> ^~* *•:"* •-*•
-»- *-j ^-' ^
^ kv_ *"
0 ^ ^^
3 -p £J ;3
" c^ ~1 O
C" t<: 5" .0-
C2^ 3
K'I _
iJ^^rt^,
. Ti-rrT..rp-nmr»T|
,T-n.r.1rpi.rT-..r,T1
LI
100. fl-
ee. 0-
5 10
.0-
100.6-
aa.e-
. 60.0-
1 40.0-
iw j •^«<^5V1.^
> l-fn-fT.i[«i?im-«-jr»rTT-ti»nwm>i-rininr»riTr|in"mi|-"<-^'^^|«'...J...... J...T..-|-.. |. ,...,. .J|rtr\,,jvr
5 10
.tl"
j
—-
V —
5 10
>^J
^
^••••W-/^''^*^...^^^^^^
v»t
-------
1000
8
164
X 230
O 300
D 404
Figure 18. Gas chromatograph-mass spectrometric-multiple ion detection
response relative to dig-anthracene as a function of
concentration for selected ions.
51
-------
individual PCN has been reproducibly detected. This translates into an
3 3
ultimate sample concentration of about 0.3 ng/m for air (based on 90 m
sample), 0.2 yg/£ for water (based on 200 ml aliquot) and 0.5 Ug/kg for
soil (based on 50 g aliquot).
Table 9. LINEAR REGRESSION RESULTS OF MID LINEARITY STUDY
FOR POLYCHLORONAPHTHALENES
Isomer
m/e
Linearity
Slope
Y-intercept
n
n = number of
C10
164
1
2
0
5
points
Determination of
H7C1
.00
.18
.365
C10H
230
0.
0.
~0.
17
5C13 C10H
300
982 0.
10 0.
18 "8.
17
2C16
982
78
85
C10
404
0
0
~4
17
C18
.977
.23
.29
in calculation
Relative
Molar
Response — The
calculation of relati'
molar response (RMR) for the quantitation of sample components precludes
the need for a calibration curve. The RMR is calculated as the integrated
peak area of a known amount of the compound, A° , , with respect to the
integrated peak area of a known amount of standard, A° , (in this case d1(.-
anthracene) , according to the equation
o o
A , /moles . (A , ) (mw . ) (g .,)
m _unk - unk = l_unk^ - unk *std (Equation 1}
From this calculated value, the concentration of an identified compound in
a sample is calculated by rearranging Equation 1 to give
(Aunk> (mWunk} (Sstd) 2^
2)
(Astd) (mwstd) (RMR)
The use of RMR for quantitation in GC/MS has proven successful in
(21)
repeated application to similar research problems.
52
-------
The RMRs for the available PCN isomers were calculated from the
numerical integrations of peaks observed in the appropriate MID channel.
The RMRs listed in Table 10 are mean values of four injections of each of
two concentrations (VL ng and ^60 ng). The similarity of the values
obtained at both concentrations is further indication of linear response
and implies that the RMR values are valid at least over this range of
concentrations. Since standards for isomers of C10H3C15, C^E^Cl^, C^HCl^
were not available, the RMR for each of these compounds was arbitrarily set
at 0.75, the mean value for the five PCN isomers determined.
The randomness of the RMR with extent of chlorination is surprising,
considering the change in ionization cross-section expected by systemati-
cally replacing hydrogens with chlorines. In addition, the small differ-
ences among the values (differing by no more than a factor of 3) is in
marked contrast to ECD or FID detectors. For example, the sensitivity of
phen1
(28)'
ECD to decachlorobiphenyl is about 10 greater than the sensitivity to
monochlorobiphenyl.
"Double" MID Confirmation—The identification of PCNs in samples was
confirmed by "double" MID GC/MS analysis, wherein two ions of the parent
cluster for each isomer were monitored (Table 8). The intensity ratio
should agree with the isotopic abundance (Table 8). If the intensity ratio
was incorrect (outside a 10-20% error margin), the results were assumed to
be spurious and were entered in the final data tabulations as "not detected".
Full Scan Confirmation—In many cases, the PCN concentration found by
MID analysis was sufficient to enable confirmation using the full scan mode
of the mass spectrometer. This not only provided the needed confirmation,
but also permitted identification of other components in the sample.
Volatile and Semi-volatile Organics Collected on Tenax—Tenax GC
cartridges containing organic compounds were analyzed by GC/MS/COMP using a
Varian CH-7 GC/MS with a Varian 620/L computer. Chromatographic separations
were accomplished on a 0.35 mm i.d. x 100 m glass SCOT capillary column
coated with OV-101 stationary phase and prepared in this laboratory. Using
(18—21^
a custom-designed inlet-manifold the cartridges were thermally
desorbed with the volatime components trapped in a liquid nitrogen-cooled
nickel capillary and subsequently revolatilized and injected into the
53
-------
Table 10. RELATIVE MOLAR RESPONSES FOR POLYCHLORONAPHTHALENE
ISOMERS
a
Low Concentration
Isomers
2-Chloronaphthalene
1, 2-Dichloronaphthalene
1,2, 3-Trichloronaphthalene
1,2, 3, 4-Tetrachloronaphthalene
Octachloronaphthalene
a
High Concentration
Isomers
2-Chloronaphthalene
1, 2-Dichloronaphthalene
1,2, 3-Trichloronaphthalene
1,2,3, 4-Tetrachloronaphthalene
Octachloronaphthalene
Total6
2-Chloronaphthalene
1 , 2-Dichloronaphthalene
1,2, 3-Trichloronaphthalene
1,2,3, 4-Tetrachloronaphthalene
Octachloronaphthalene
MEAN
Amount
(ng)
1.39
1.06
1.25
1.10
1.26
Amount
(ng)
69.6
52.8
62.4
54.9
60.0
Amount
(ng)
RMRb
0.502
0.937
0.658
1.024
0.321
RMRb
0.439
1.172
0.805
1.328
0.330
RMR
0.470
1.054
0.732
1.176
0.326
0.752
SDC
0.063
0.068
0.087
0.088
0.240
. SD
0.046
0.076
0.037
0.042
0.026
SD
0.06
0.14
0.10
0.17
0.14
0.36
RSDd
0.122
0.072
0.132
0.086
0.771
RSD
0.106
0.064
0.046
0.032
0.077
RSD
0.130
0.135
0.137
0.147
0.442
0.484
Four injections
Relative Molar Response Standard = d.. --anthracene - 0.828 ng (0.276
ng/yl)
^Standard Deviation
Relative Standard Deviation = SD/RMR
2Eight Injections
54
-------
Varian 1700 gas chromatograph. Complete details of the system and its
operation are included in Appendix A.
Data Handling and Interpretation—The data output of the GC/MS was
screened at several levels and logged. After calculation, the results were
recorded on summary sheets for each site. At that point, the results were
evaluated and a decision made regarding the need for repeat analysis,
confirmation by "double" MID, or confirmation by full scan analysis. The
details of the data handling and interpretation are discussed below.
Data Output Format—The GC/MS/COMP system produced numerical data
printout as illustrated in Table 11 for a representative sample. This is
the raw data for a soil sample collected near Koppers Chemical and Coatings
Plant, P2/C4/L11. The identification line contains the sample identifica-
tion, GC oven temperature, programming conditions, attenuation, detector
voltage, injection volume, and date of analysis. The nominal mass is
printed in the first column; the precise mass of each channel (to five
significant figures) is set daily to compensate for instrumental fluctua-
tions. The peak intensity, retention time (MIN), integrated AREA, number
of data points'in the integration (PNTS), and background (BKGND) are printed
in successive columns.
Table 11. GC/MS/COMP NUMERICAL DATA PRINTOUT
R23GABVLKOP-1-10, 150* 3M8*/M, -9, 1.8 kV, 1 MCL, J28
MASS
164
196
230
188
266
300
336
368
404
PEAK
40.98
52.80
47.62
02.68
26.10
16.42
02.00
00.38
00.06
MIN
0.916
1.882
3.799
4.599
7.116
9.249
10.816
12.832
15.183
AREA
258.86
646.10
916.46
54.20
1115.76
448.82
53.00
07.96
01.78
PNTS
42
84
103
65
159
149
120
49
51
BKNGD
28.84
11.72
05.54
13.64
03.46
02.14
01.66
01.60
01.60
55
-------
Calculation of Concentrations — Using the integrated peak area (numeri-
cal printout) for the RMR for each isomer (Table 10) , concentrations of
PCNs were calculated. The amount of trichloronaphthalene (m/e = 230) in a
sample, for example, was calculated using the data in Tables 10 and 11 and
Equation 2 .
g , (Aunk) (mwunk) (gstd} fv . ,.
= (RMR) (Equat10n2)
In this example,
Aunk - 916'46
A . . = 54.20
std
mw , = 230
unk
mw ^j = 188
std
gstd = °'276 ng/yl
and, BMR = 0.732.
By substitution,
(916. 46) (230) (0.276 x 10"9) = 7.80 ng/yl.
Sunk ~ (54. 20) (188) (0.732)
The concentration in the actual sample was calculated from the volume
of the extract, the recovery efficiency and from the volume of the sample
or aliquot. For air samples, the volume of air sampled is listed for each
sample in the sampling protocol for each location. For soil samples, a 50
g aliquot of the composited soil was analyzed; while for water samples a
200 ml aliquot was analyzed. For the sample cited above, the volume of the
extract was 2.0 ml, so the total amount of trichloronaphthalene is 15,600
ng; dividing by 50 g soil extracted gives an original soil concentration of
310 ng/g or 310 yg/kg.
Assessment of Results — The calculated results were entered onto the
appropriate summary sheet for that site. The results were also assessed
for the need for confirmation by "double" MID or full scan GC/MS. Results
which did not fit the prevailing isomeric distribution pattern, were unusu-
ally high or low, or were otherwise anomolous or interesting were subjected
to confirmation.
56
-------
3.2.3.2 GC/ECD of PCNs
Analysis using GC/ECD, limited to method development aspects of the
project, involved a Fisher Victoreen Series 4400 gas chromatograph using a
180 cm x 2 mm i.d. glass column packed with 2% OV-101 on 100/120 mesh Gas
Chrom Q at either 170° or 200° with a nitrogen flow rate of 30 ml/min. The
3.3 QUALITY CONTROL
To assure the validity of the results, careful quality control proce-
dures were maintained throughout the project. These included validation
and testing of all methods prior to use; logging of sampling protocol
sheets, sample analysis progress, and GC/MS analytical results in multiple
locations; analysis of blanks and controls; instrumentation control and
error propagation. These procedures assured the quality (accuracy, preci-
sion, completeness, and representativeness) of the data with two exceptions:
(1) poor recoveries were noted for PUF control samples stored with field
samples. The percent recovery found was used as a correction factor to the
air samples; (2) the collection efficiency of monochloronaphthalene (and to
a lesser extent dichloronaphthalene) on PUF was found to be poor, so all
values for monochloronaphthalene should be regarded as minima.
3.3.1 Method Validation
As previously discussed the analytical methods including GC/MS analysis,
extraction of polyurethane foam plugs, soil, and water, column chromatography,
and volatile organic purges were all validated and tested before use.
The sampling methodology, as discussed in a preceeding section was
validated through a series of tests to check collection efficiency of the
air and water samples. Under the sampling conditions used, the air samplers
collected PCNs efficiently, except for mono- and dichloronaphthalenes which
were only partially collected. The water sampling procedure was found to
collect PCNs with no losses.
3.3.2 Controls and Blanks
Through the use of controls and blanks, contamination or loss of
sample during storage was monitored. The recovery of PCNs from spiked
57
-------
polyurethane foam plugs was found to be about 40%, so the analytical results
were corrected by this factor. Blank polyurethane foam plugs were not
contaminated during storage.
Control samples were prepared as part of the quality control program
by spiking polyurethane foam plugs with known amounts of PCN mixtures and
were stored with field samples. The plugs were extracted and analyzed
after being stored for about as long as an average field sample from each
of the three sampling trips. Table 12 summarizes the results of these
experiments.
Parallel to the spiked polyurethane foam plugs, blank plugs were
stored with field samples and analyzed after being stored for about as long
as an average field sample. The results for these blanks are listed in
Table 13. No PCNs were found in the blank samples.
3.3.3 Sample Log
At the time of collection of each sample, a sample protocol sheet
(Figure 19) was filled out and stored in a loose leaf binder. Upon return
to the lab, copies of these protocol sheets were stored in two separate
locations to prevent accidental loss. In addition, a bound sample log book
was kept with sample identification codes for a running record of analysis
progress. Copies of the GC/MS analytical data were entered into the sample
log book for archival purposes.
3.3.4 Instrumentation Control
To insure that the total operating system was calibrated and in
proper working order, the Finnigan 3300 GC/MS was evaluated daily using a
standard reference mixture of testosterone and cholestane under a set of
reference criteria. In addition, a PCN reference mixture was subjected to
GC/MS analysis under the identical operating parameters as those to be used
for the analysis of field samples at the beginning of each working day. In
this manner, the performance of the GC column, the sensitivity of the mass
spectrometer, the calibration of the mass spectrometer and the performance
of the computer system were monitored by evaluating the results of the
reference mixture.
58
-------
Table 12. RECOVERY OF CHLORONAPHTHALENES FROM CONTROL SAMPLES
Sample Code
PA-2C
NY-ld
NY-2b
NC-lb
Mean
SD
Storage Time
Days* C1()H
89 55
72
37 36
37 13
35
20
7ci
.3
.6
.6
.1
.9
C10H6C12
55.2
26.5
37.0
68.9
46.9
18.9
% Recovery
C10H5C13 C10H4C14 C10H3C15
43.4
23.0
34.5
59.1
40.0
15.3
39.3
24.2
33.0
59.7
39.1
15.1
43.4
27.0
34.5
64.3
42.3
16.1
C10H2C16
30.9
22.3
33.4
63.2
37.5
17.8
C10HC17
33.2
22.0
31.8
70.0
39.3
21.1
C10C18
44.1
18.5
31.5
58.0
38.0
16.8
Mean
43.1
23.4
34.0
57.1
39.8
17.3
SD
8.9
2.9
2.0
18.1
3.5
Including storage on foam and storage awaiting analysis
116 yg Halowax 1014_; 111 yg Halowax 105lw applied to foam
222 yg Halowax 1051^ applied to foam
"232 yg Halowax 1014®;
11.6 yg Halowax
; 11.1 yg Halowax
applied to foam
-------
Table 13. GAS CHROMATOGRAPHY/MASS SPECTROMETRY ANALYSIS OF BLANK SAMPLES
Storage Time
Sample Code Days
C^Cl
C^Cl,.
C^HCl, C^Cl- Average
PA-2
PA-3
NY-3
NY-4
NC-2
60
60
72
72
38
Including storage of extract prior to analysis
-------
Dace:
Project No.
Municipal!ty_
Location
Operator
_
Seq.
Sampler/Rationale
State
Site
Sanple Code_
No.
Dimensions
(en)
Sorbent Date/Analytical Procedure
( H)
M)
M)
M)
DC Amps_
Sampling Rate (L?M)_
Vacuum ("Hg)
End: Tine
Rationale: Qual. Anal (F)
Quant. Anal. (E) Calibration (C) Start: Tine
Experimental: Lab (L)_
Remarks
Field (X)
Total: fnin).
_ftJ
~ i
ft"
"ft3
Volur.e Air/Cartridge:
0.0233
No. 'Salit
MAP:
Ti=e Tanip. Wet. Dry
Rel. Hur.id " Wind Dir./Speed /_
Cloud Odor
Reaarka
Tine
e=p. Wet.
Dry
Rel. Huaid_
Cloud
Reaarks
_" Wind Dir./Speed /_
Odor
Tenp. Wet.
Drv
Rel. Ku=id
Cloud ~
Resarks
Wind Dir./Speed /_
Odor
Tine
Tscp. Wet.
Rel. Hu=id
Cloud ~
Reaarks
Dry_
" Wind Dir./Speed /_
Odor
Figure 19. FIELD SAMPLING PROTOCOL FOR AMBIENT AIR
61
-------
3.3.5 Data Evaluation and Quality Control
Gas chromatography/mass spectrometry data was screened by the instru-
ment operator before printout and only peaks clusters near the appropriate
retention time were included in the integration. The operator also screened
the data for anomalous results. If the sum of the PEAK and BACKGROUND was
XLOO, the detector was saturated and the sample had to be re-run using a
lower sensitivity or smaller injection volume.
In addition to preliminary screening by the operator, the data was
also checked by the instrument supervisor to assure that the sample had
been run as per the request sheet and that the output was of appropriate
quality. The results were then released for interpretation.
Upon receipt of the GC/MS output, it was again checked for saturated
peaks and for correlation in retention times with those of the corresponding
known PCN isomers. The raw data was also added to the sample log book,
which kept a running account of the status of each sample.
3.3.6 Confirmation of Results
Confirmation of representative samples were obtained by (a) full-scan
GC/MS or (b) "double ion MID", where two ions of a chlorine isotope cluster
are measured simultaneously as discussed in a preceeding section.
3.3.7 Archival Storage
All samples (e.g., remaining portions of soil samples) sampling
protocol sheets, sample log books, notebooks, instrumental log books,
spectra, GC/MS output, magnetic tapes, and other records were retained for
archival purposes. After acceptance of the final report, the samples and
magnetic tapes will be discarded but all hard copy records will be perma-
nently archived.
3.3.8 Assessment of Error
The error of the entire analytical procedure is a function of the
errors at each step. The major sources of error and their estimated
magnitude are listed below.
62
-------
3.3.8.1 Collection
The error involved in collection of soil and water sample resides
primarily in the selection of a representative sample, however, the error
in air collection involves the additional factor of collection efficiency
of the GFF and PUT plugs. Studies of this efficiency indicate that a major
error for monochloronaphthalene may result from collection as discussed in
a preceeding section.
The error in measurement of the volume of air sampled is estimated at
+10%, due to the sensitivity and stability of the flow rates.
Any error in volume for soil and water samples is derived from gravi-
metric and volumetric errors, respectively. An error of 0.2% for soils and
0.5% for water is estimated at this step. There is no dependable method
for estimating the error or variability introduced by the selection of
representative samples without an extensive sampling and analytical effort.
3.3.8.2 Extraction from Sampling Media
Based on the recovery of PUF control samples discussed above (Table
12), the average SD for the individual PCNs was +17.3% which corresponded to
+43.3% (average RSD).
3.3.8.3 Column Chromatography
The average recovery of PCNs from the column clean-up step, as dis-
cussed above was 98.5%. Thus it may be assumed that the maximum loss at
this step is 5%.
3.3.8.4 Addition of Standards
An error in addition of standard would affect the calculation of the
amount of PCN in the sample. This error is that inherent in the volumetric
and gravimetric manipulation of the sample. A balance error of +0.7%,
volumetric error of +0.5%, dilution errors of +1% for pipetting and +0.5%
for volumetric error, and a pipetting error of 1% upon addition of standard
are estimated.
3.3.8.5 Final Volume of Sample
The sample is generally made up to a final volume of 2.0 ml. The
error in this step is +5%.
63
-------
3.3.8.6 RMR
The error in calculation of RMR can be estimated from the mean RSD in
Table 10. The average RSD for the five RMR values is +19.9%.
3.3.8.7 GC/MS Analysis
The reproducibility of the GC/MS was checked using an actual sample
and is summarized in Table 14. The average RSD of the five PCN isomers
detected was 12%.
3.3.8.8 Total Error
The total error, S , associated with the reported PCN concentrations
may be estimated by
S =Z S (Equation 1)
I « 1 -^ I
VJ.-1 J
where S. is the error associated with the i source of error. This
equation assumes that all sources of error are independent, which is a
reasonable assumption in this case. The total error is listed for air,
water, and soil samples in Table 15. Sediment, and other solid samples may
be assumed to have errors similar to those listed for soil.
Table 15. TOTAL ESTIMATED ANALYTICAL ERROR
Medium Propogated Error
Air + 50%
- 51%
Water + 24%
- 27%
Soil + 24%
- 26%
64
-------
U1
Table 14. REPRODUCIB1LITY OF GC/MS ANALYSIS OF SAMPLES FOR PCNs.
REPLICATE INJECTIONS OF AN AIR SAMPLE EXTRACT - KOPPERS, P2/C1/L1 - FILTER
Injection
1
2
3
Mean
SD
RSD
C10H7C1 C10H6C12
3.0
3.2
2.5
2.9
0.36
0.12
C1QH5C13
20.7
25.3
21.2
22.4
2.52
0.11
C10H4C14
10.0
12.1
9.9
10.7
1.24
0.11
C1()H3C15
9.3
11.5
10.5
10.4
1.10
0.11
C10H2C16 C10HC17 C10C18
0.6
0.7
0.8
0.7
0.10
0.14
Total
43.6 ng/m
52.9 ng/m3
45.0 ng/m
47.2
5.01
0.11
Average RSD =0.12
-------
4.0 SAMPLING AND ANALYSIS NEAR KOPPERS CHEMICAL AND COATINGS PLANT,
BRIDGEVILLE, PA
Koppers Company, Inc., Pittsburgh, PA (Koppers) was selected for
(1 8^
sampling since it is the sole U.S. producer of PCNs. ' ' The manufacture
of PCNs takes place at Koppers Chemical and Coatings Plant, Bridgeville,
PA and they are distributed under the trade name Halowax ^.
4.1 FIELD SAMPLING
Koppers is located in a southern suburb of Pittsburgh, PA in a deep
(60-100 m) valley. The plant site is relatively large and old (roughly,
600 x 100 m), containing several main buildings and numerous smaller build-
ings. A high level of activity was inferred from the 30-50 plumes of steam
and/or smoke observed at any one time. Several people indicated that a
white "snow" occurred which was usually deposited in the night.
Based on the above discussions and on general sampling protocol,
sampling locations were chosen both in the valley and on the hills surround-
ing the plant. The sampling protocol is summarized in Table 16.
4.1.1 Mr
Two 24 hr air samples were collected at each of four points along the
transects as shown in Figures 20 and 21. The MRI weather station was
located at L3 (Figure 20). No adverse weather was noted, although wind
direction and speed were highly variable.
4.1.2 Soil
Soil samples were collected along the transects according to the
general protocol. One soil sampling site (L13, Figure 21) was in an area
66
-------
Table 16.
SAMPLING PROTOCOL FOR KOPPERS COMPANY, CHEMICAL AND COATINGS PLANT
BRIDGEVILLE, PENNSYLVANIA.
o\
Meteorological Conditions
Period Cycle Location
10/25/76 Cla LI
PI
L2
L3
L4
C2a L5
L6
C3b L7
10/26/76 Cla LI
P2 L2
L3
L4
C2 L5
L6
L7
C3C LB
C4C L9
L10
Lll
L12
L13
L14
L15
L16
Sampling
Time
1358-14J2
1403-1435
1350-1339
1445-1404
1112-1125
1055-1125
2120-2150
1446-1551
1510-1530
1355-1428
1420-1412
1320-13203
1330-13093
1050C
1425
1800
1830
1600
1630
1700
1730
0900
0930
Sampling
Volume
91.8 m3
93.0 ra3
89.0 ni3
89.2 ra3
2.7 ft
2.6 ft
246 ft
90.0 m3
91.0 m3
91.8 m3
89.2 m3
2.4 ft
1.3 ft
0.8 ft
12 coresS
10 cores
10 cores
6 cores
10 cores
10 cores
10 cores
10 cores
Type of
Sample
APCN
APCN
APCN
Al'CH
WPCNd
WPCNe
AIIC
APCN
APCN
APCN
APCN
WPCN11
WPCN1
WPCNJ
apples
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
T CC) Z RH
8-4 93
8-4 93
8-4 93
8-4 93
llf
9*
7
4-2 55
4-2 55
4-2 55
4-2 55
gC
9^
6
6
7
7
7
6
0
1
Wind Dir./
Speed (kmph)
SW/5-NW/13
SW/5-NW/13
SW/5-NU/13
SW/5-NW/13
SW/5-NW/19
SW/5-NW/19
Other
slight rain, odor
benzophenone.
naphthalene
slight rain
slight rain
slight rain
slight rain
slight rain
mixed organic
of
NW/13-H/4 odor of naphthalene
NW/13-W/4
NU/13-W/4
NW/13-U/4
HW/15-W/4
NW/15-//4
NW/11
NU/11
NW/13
NW/13
NW/11
NU/11
Calm
Calm
odor of naphthalene
(continued)
-------
Table 16 (cont'd)
o\
00
Meteorological Conditions
Period Cycle Location
10/26/16 C6c«k L17
C7b»k L18
L19
L20
L21
Sampling
Time
0900
1048-1114
1048-1118
1126-1202
1217-1247
Sampling
Volume
15 cm
229 4
170 i
153 I
189 I
Type of
Sample
fish
AHC
A11C
A1IC
AHC
Ulnd Dlr./
T ("O Z RH Speed (kniph)
HW/6
3 NU/6
2 59 NW/6
NU/6
cloudy
cloudy.
cloudy,
cloudy
Other
naphthalene
mixed organic
e,
24 hr. composite samples
30 nun survey samples
grab samples
MO m downstrean from waste water out fall
MO m upstream from waste water out fall
water temperature
P
"5 cm diameter x 13 cm depth
MOO in upstream from Koppers Chemical property
same as Location 6, period 1
^drainage ditch
W10/29/76
Key to Sample Type: APCN - polychloronaphthalenes, air
AHC - hydrocarbons, air
WVCN - polychloronaphthalenes, water
SPCN - polycliloronaphthalenes, soil
-------
100 m
Figure 20. Map of Koppers with sampling locations for
PI - 10/25/76.
69
-------
Figure 21.
H
100 m
Map of Koppers with sampling locations for
P2 - 10/26/76.
70
-------
which local sources said used to be a dumping ground for Koppers' solid
waste.
4.1.3 Water
Interval samples were collected above and below the South Fayette
Township Sewage Treatment Plant (PI, L5 and L6). The plant manager stated
that Koppers did not use their system, but the proximity and possibility of
leakage into the system prompted the decision to sample. Other composite
samples were collected upstream (P2, L5) on Millers Run and downstream (P2
L6) on Chartiers Creek. A grab water sample (P2 L7) was collected from a
small ditch which contained the runoff and/or leachate from the old dumping
ground discussed in Section 6.1.2.
4.1.4 Miscellaneous
Several apples were collected from the apple trees on the farm at L4.
A small dead fish (carp, 40 g, M.5 cm) was collected in Millers Run. No
cause of death could be determined.
Five air samples were collected on Tenax. Both upwind and downwind
samples were taken and attempts were made to be directly in the plume of the
emissions from the plant for the downwind samples.
During the sampling period, a white flocculant crystalline material was
observed as airborne particulates and adhering to low vegetation near
sampling locations 9 and 10. This is presumably the "snow" noted by local
residents. A sample was collected.
4.2 ANALYSES
Air, soil and water samples were analyzed according to the procedure
detailed in Appendix A. The whole fish sample was homogenized in a blender
and then extracted using the soil extraction procedure. The apples were
likewise homogenized and also extracted accorded to the soil extraction
procedure. The white flocculant crystalline material was analyzed by direct
probe mass spectrometry (using the Finnigan 3300 instrument).
4.3 RESULTS AND DISCUSSION
The results of the analysis of samples collected near Koppers Company,
Bridgeville, Pennsylvania are summarized in Table 17.
71
-------
Table 17. RESULTS OF ANALYSIS OF SAMPLES COLLECTED NEAR KOPPERS CHEMICALS AND
COATINGS, INC., BRIDGEVILLE, PENNSYLVANIA
to
Period8 Cycle3 Location8
PI Clb Ll Fc
1
2
Total
L2 F
1
2
Total
L3 F
1
2
Total
L4 T
1
2
Total
C2f L5
L6
P2 Clb Ll F
1
2
Total
C1()H7C1
__d
96
49
140
__
1.8
1.8
0.2
—
1.8
0.9
__
__
—
—
0.4
—
230
130
360
C10H6C12
160
1.3
160
__
15
15
—
25
—
25
«
7.3
7.3
__
—
2.9
610
—
610
C10H5C13
120
5.2
130
Te
29
29
59
—
59
— .
18
_ —
18
0.5
~
22
750
—
770
Degree of Chlorlnatlon
C10H4C14 C10H3C15 C10H2C16 C10HC17
_
15 1.3
__ __
15 1.3
— — — — _—
3.0
3.0
— — — — __
5.0 —
-_ __ -_ __
5.0 — ~
« — — «-—
__ — — . __
__
0.5 0.4 T T
0.2
11 10 0.7
99 30
— — —
110 40 0.7
C10C18 Total
_-
390
55
450 ng/m
T
49
49 ng/m
0.2
89
1.8 -
91 ng/m
_ _
25
25 ng/m3
T 1.4 pg/fc
0.6 pg/Jl
47
— 1700
130
1900 ng/m
(continued)
-------
Table 17. (cont'd)
CO
Period Cycle Location
P2 Cl L2
L3
L4
C2f L5
L6
L7
JjiA^'l^
C41 L9
L10
Lll
F
1
2
Total
F
1
2
Total
F
1
2
Total
C10H7C1
140
42
180
970
45
1000
5.5
5.5
—
--
—
62
3.4
—
97
C10H6C12
1.8
250
250
1.0
860
860
32
32
—
—
—
8.9
15
3.2
81
Degree
C10H5C13 C10H4C14
22
330
350
32
920
950
1.
62
63
—
—
—
13
280
60
310
7.2
32
40
14
83
100
1 0.8
11
12
—
—
—
7.3
170
39
270
(continued)
of Chlorlnatlon
C10H3C15 C10H2C16 C10HC17
3.2 0.2
3.2 0.2
5.2 0.4 0.2
1.0
6.2 0.4 0.2
0.8
1.0
1.8
0.2
__
__
h^5'/vV a -h h
98 20 9.3
20 4.6 1.0
190 25 4.2
C1()C18 Total
35
750
42
830 ng/m3
0.2 53
2800
45
0.2 2900 ng/m3
2.6
110
110 ng/m3
0.2 pg/Jt
__
__
— h 90 ug/kg
4.0 600 yg/kg
0.6 130 ug/kg
2.3 990 ug/kg
-------
Table 17. (cont'd)
Period Cycle Location
P2 C4 L12
L13
U3A*
L14
US
U6
C61 L17
C10H7C1 C10H6C12
6.0 34
9.6 75
5300 >16000k
1.6 15
57
8.8
1.7
C10H5C13
370
880
>25000k
86
260
180
22
Degree
C10H4C14
710
760
18000
49
230
130
—
of Chlorination
C10H3C15 C10H2C16
900
310
15000
26
200
79
9.5
250
55
11000
6.3
30
18
3.6
C1()HC17
74
16
5200
1.5
6.8
5.6
2.2
C10C18
7.5
5.2
760
0.7
3.1
—
—
Total
2300 pg/kg
2100 pg/kg
>96000 pg/kg
190 pg/kg
780 pg/kg
440 pg/kg
39 pg/kg
See Table 16 for period, cycle and location designations in the sampling protocol
24 hour air samples
CF." Glass fiber filter; 1 - Top PUF Plug; 2 - Bottom PUF Plug
No PCN peak detected. Detection limit for air is about 0.3 ng/m , for soil about 0.5 pg/kg, and for water about 0.2 pg/Jt..
eTrace
Water samples
' interferences prevented accurate quantitation - <5 pg/kg
Soil Samples
•'Black crystalline component of sample L13
K
^GC/HS detector saturated
(_J)ish (whole, homogenized)
-------
4.3.1 Air
The frequency of occurrence of wind speed classes having lower limits
of 0, 4, 7, 11, 17 and 21 knots (1 kt = 0.5148 m/s) as a function of wind
direction (of a 16-point compass) at the Greater Pittsburgh Airport is shown
graphically in Figure 22. This distribution, called a wind rose is based on
observations every three hours from January 1, 1970 to December 31, 1974.
Westerly winds (WSW-WNW) occur most frequently with greater average speed
and with infrequent low (<4 kt) wind speeds. Low wind speeds are more
predominant in the southwest quadrant. Winds from the northwest quadrant
are infrequent and are seldom over 10 kt.
Although wind rose data is useful in assessing long term trends in the
macrometeorology, it does not always give insight to the ground level dis-
persion of an emission. This is especially true in the case of irregular
terrain such as found near Koppers. In this case, the micrometeorology must
be considered. These factors are included in the discussion below.
Total PCN concentrations for the air samples collected during Period 1
are presented in Figure 23, with detailed maps presented in Figure 24.
Comparable data for Period 2 is given in Figures 25 and 26. The average
3
concentration detected during the first 24 hr period was 150 ng/m . The
average concentration detected during the second 24 hr period was 1600
ng/m .
Weather records from the Greater Pittsburgh Airport show that during
Period 1 (10/25/76) light rain and fog was present with northerly winds at
11-15 km/hr until about 1830 EST. Visibility increased, the precipitation
ended and wind speeds increased to 18.5-30 km/hr, gusting to 40 km/hr in the
early evening. Skies remained cloudy overnight, keeping temperatures from
falling more than 3°C. By sunrise of the following day, the skies cleared
and northerly winds persisted at 20-26 km/hr throughout the day. An over-
cast of shallow convective clouds formed during the late morning and lasted
until late afternoon (Period 2). By evening, skies were clear with tempera-
tures decreasing to 6°C by midnight and wind speeds decreasing to approxi-
mately 10 km/hr for the night. At midnight, low clouds returned and remained
throughout the second sampling period.
75
-------
is]
7.0%
(8.1)
— 0
Wind Speed
Class
13.4%
(10.6)
(8.9)
8.8%
( ) Average speed for
given direction
Average speed (all
directions) 8.2 kt
ANNUAL WIND ROSE
PITTSBURGH, PENNSYLVANIA
1970-1974
Figure 22. The length of wind speed class proportional to its frequency
of occurrence.
76
-------
91
© — -
25 "~"~~"
(2)
49
450
Figure 23.
Total PCN concentrations (ng/m3) in air near Koppers
Chemical and Coatings Plant, Inc., Period 1.
77
-------
Figure 24. Map of Koppers with sampling locations (Period 1)
78
-------
1
I
2900 1
C3J
I
no
830
\
\ 1900
Figure 25. Total PCN Concentrations (ng/m ) in air near
Koppers Chemical and Coatings Plant, Inc.,
Period 2.
79
-------
Figure 26. Map of Koppers with sampling locations (Period 2)
80
-------
During the first sampling period, the brisk winds quickly diluted plant
emissions as the air moved southward toward Location 1. Atop the hills to
either side of the plant, at Locations 2 and 4, the winds were generally
unfavorable for measuring emissions transported directly from the plant.
Concentrations measured at these sites probably arose from horizontal and
vertical diffusion of emissions during the period.
In Period 2, especially during the night of October 26-27, generalized
meteorological conditions were extremely conducive to a localized wind
circulation pattern which would transport emissions toward the wastewater
treatment plant location. Clear skies and relatively dry air permitted the
ground, hillside and high ground to cool by radiation. The cold air flowed
toward lower elevation, accumulating as a cold air mass in the narrow val-
leys. As this mass of cool air deepened, warmer air was lifted aloft,
creating a stable interface. Apparently, the mass of cool air deepened to
such a height that the emissions had insufficient buoyancy to be lifted
above the interface. The generally northerly wind flow was not strong
enough to mix and dissipate this stable layer, so the emissions were trapped
within the cold air mass. This cool air drifted very slowly down the Sygan
Run and Millers Run Valleys, approximately in proportion to the slope of the
land and contrary to the flow of air further aloft, toward the wastewater
treatment plant sampling location. Anemometer records at that location
indicated little movement of air during the night. With the vertical dis-
persion limited by the stable air aloft, horizontal dispersion confined by
the valley walls, and very little movement of air past the emission loca-
tion, the entire valley floor became an emission reservoir, giving a high
concentration at sampling Location 3.
Although difficult to quantify, the depth of the cool air may have
increased during the night to an altitude of 70 m above the valley floor,
thereby affecting the hilltop monitoring locations. The PCN concentrations
there are lower because of greater dilution as the depth of the cool air
increased and because the samplers were probably above the stable interface
at least for part of the sampling period.
The average distribution of the PCNs on the air sampling media is
depicted in Figure 27. The bulk of the PCNs are collected on the first foam
81
-------
2.4% GLASS FIBER FILTER
92.8% FIRST POLYURETHANE
FOAM PLUG
4.8% SECOND POLYURETHANE
FOAM PLUG
SCALE Icm = 4cm
Figure 27. Average distribution of PCNs on sampling media.
82
-------
plug; however a portion of the heavier isomers are collected by the glass
fiber filter, indicating that they may be present in the air as particulates
or aerosols. Only monochlorobiphenyl eluted, to any great extent, through
the first foam plug to the second.
4.3.2 Soil
The soil analysis results are presented in Figure 28. The average
concentration found was 950 yg/kg.
The concentration of PCNs detected generally decreased with increasing
distance from the plant. The very high concentrations (2100 Ug/kg) at
Location 13 may not be due totally to air deposition. The samples compris-
ing this composite were collected in an area which, according to local
residents, was a place where Koppers has dumped solid waste. In fact,
analysis of a black crystalline residue, collected at this site revealed
>90,000 mg/kg total PCNs (where the exact value may be much higher since
only the surface of this material was extracted).
The distribution of PCNs in the soil samples is consistent with air-
borne transport and subsequent deposition by climatological wind distribu-
tion and local circulation patterns, such as the drainage flow discussed
previously. Prevailing southwesterly winds may be channeled to a more
southerly flow by the Millers Run Valley. The northerly wind flow is
likewise channeled up the valley, which would fail to reduce those wind
speeds.
For an effective emission height of about 25 m the maximum relative
concentration (concentration per unit emission per unit wind speed) should
( 331
occur from 500 to 1000 m downwind of the emitter. As the emission
height increases, the downwind distance for the maximum increases. It is
therefore feasible that the highest concentration would more likely be found
farther away from the source (jL.£., beyond the wastewater treatment plant)
rather than nearer the source.
The average degree of chlorination of PCNs detected in samples collec-
ted nearer the plant is slightly lower than that of samples collected farther
along the transect. This is illustrated in Figure 29. The mean degree of
chlorination found for "near" samples was 3.75 while 4.33 was found for
83
-------
990
39
190
{14)—
780
16
'600
Figure 28. Total PCN concentrations (ug/kg) in soil near
Koppers Company (Period 2).
84
-------
500i
400-
CO
60
a.
c
o
* 3001
u
o
C03
n)
200-
9 NEAR SAMPLES
OFAR SAMPLES
8
Degree of Chlorination
Figure 29. Distribution of PCNs on soil near Koppers Company.
85
-------
"far" samples. No clear explanation is apparent for this slight difference
other than differences in the physical properties of the various isomers,
emission history and meteorological conditions.
4.3.3 Water
The water samples taken during Period 1 were from above (L6) and below
(L5) the South Fayette Township Wastewater Treatment Plant outfall. This
plant does not directly serve Koppers and hence only a very slight increase
in PCNs is observed in water collected below the treatment plant outfall.
The values are near the detection limit, but were confirmed by the "double
MID" technique. The values therefore do not appear to be artifacts, but are
nevertheless low.
Analysis of composited water samples collected during Period 2 above
(L5) and below (L6) the Koppers plant reveals only a trace of CnnH0Cl, in
1U Z D
the upstream sample while no PCNs were detected downstream.
4.3.4 Miscellaneous Samples
The small dead fish (carp, 40 g, 'VLS cm) retrieved from Millers Run was
found to contain appreciable levels of PCNs (39.4 yg/kg total) with approxi-
mately 57% accounted for by trichloronaphthalene. No other dead or living
organisms were observed either upstream or downstream from the Koppers
plant, consequently no cause of death can be assigned. In addition, it is
not known whether the PCNs detected had been ingested by the fish or were
adsorbed either before or after death.
Approximately one dozen red apples (variety unknown), obtained from the
ground at Location 14, were found to contain a total PCN concentration of
about 90 yg/kg.
The white flocculant crystalline material that was observed as airborne
particulates was found to be phthalic anhydride. This conclusion was sup-
ported by a melting point of M.28°C (lit. 131.6°C).
Two of the five air samples collected on Tenax GC cartridges were
selected for GC/MS analysis as the two most likely to produce positive
results. The TIC chromatograms (Figures 30 and 31) were examined and the
more interesting (Figure 32, P2/C7/L20) was selected for data output and
86
-------
300DCI-,
00
•-J
EOOOO-
10000-
0500 05SO 0606 0650 0700 07SO 0800 08SO
SPECS 10500 - 10990 LH I 277 1 BBK3N 1 XHCPHBULKOP>'T1FI»X2FEB77''1 0 OHO STEP SPECS'! IHT- 1000
0900
OSSO
Figure 30. Total ion current chromatogram from gc/ms analysis of organics collected near
Koppers Chemical Company, Bridgeville, Pennsylvania (P2/C7/L21).
-------
30t)00-i
eoooo-
00
co
10000-
r-i"T|i"li""
r
""!
-'i ......... I1"1""!
0000 0050 0100 01SO 0200
SPECS 10000 - 10152 Lrt 12771xLCn3Ml^HCPflBVLROTT3F3/'a FEB'77xJOO
lf'T'l'n''"n'^"T"n'l"ni'|H'iT"riTl ...... ri"T"i""i ..... TT-
0250 0300 0350
STEP SPEC9-J IMT- JOOO
OH 00
0450
0500
Figure 31.
Total ion current chromatogram of gc/ms analysis of, organics collected near
Koppers Chemical Company, Bridgeville, Pennsylvania (P2/C7/L20).
-------
30000-,
zoooo-
10000-
00
\O
v—.A
22
23
0000 0050 0100 0150 0300 0250 0300 0350
SPECS 10000 - 104SE LM 1 277 1 'LCM3N 1 /HCPfiB VLROTT3F3'2 FEB'77/100 STEP SPECtt-1 IHT- 1000
0400
_J
0450
0500
Figure 32. Gas chromatography/mass spectroscopic analysis of ambient air near
Koppers (P2/C7/L20).
-------
interpretation which involved plotting of 412 mass spectra and identifying
the compounds from these spectra. The results of this identification are
presented in Table 18. Although the styrene peak is very high relative to
that normally seen in ambient air, the compounds identified were not directly
relevant to this project, so further analysis and data interpretation of the
Tenax cartridges was not pursued.
4.3.5 Isomeric Distribution of PCNs
All eight chloronaphthalenes were found to varying degrees in air and
soil samples in the vicinity of the Koppers plant. As illustrated in Figure
33, the air samples contained predominantly mono, di-, and trichloronaphtha-
lenes, whereas the soil samples contained most tri-, tetra-, and penta-
chloronaphthalene isomers. This difference could be due to weathering and
aging of the soils, or an average historical deposition of more highly
chlorinated PCNs than was observed in the air during our sampling periods.
Due to the poor collection efficiency (vide supra), the amount of mono-
chloronaphthalene (and to a much lesser extent, dichloronaphthalene) may, in
fact, be greater than that represented in Figure 33.
The knowledge of isomeric distribution and accompanying health effects
information may be important to the assessment of the toxicological impli-
cations of this data. The toxicity of the chloronaphthalenes, similar to
that of PCBs, is highly dependent on the organism exposed, the route of
(1 8^
exposure, and the extent of chlorination. ' Humans are susceptible to
physiological damage from PCNs and several occupationally related deaths
have been recorded. The clinical manifestations of PCN poisoning are
liver necrosis and chloracne. The effects of low-level exposure have not
been reported. Individual PCN homologs exhibit markedly different toxicity
levels. Mono- and dichloronaphthalenes are generally considered non-toxic;
tri- and tetrachloronaphthalenes exhibit some toxicity, but the penta- and
hexachloro analogs are responsible for the known cases of severe poisoning
and are highly toxic. Contrary to this trend, octachloronaphthalene is
considered relatively innocuous. No information is available on the rela-
tive toxicities of the various positional isomers.
90
-------
Table 18. COMPOUNDS IDENTIFIED IN AMBIENT AIR NEAR
KOPPERS CHEMICAL COMPANY (P2/C7/L20)
Chromatographic
Peak No.
1
2
4
5
7
8
9
10
11
12
12a
12b
13
14
15
15a
15b
16
17
18
19
20
21
22
23
24
25
26
27
28
Compound
CO,
CF2C12
n-propane
CC13F
CH2C12
C2C13F3 (Freon 113, BKG)
acetone
C,H..
6 14
3-methyl pentane
perfluorobenzene (standard)
n-hexane
chloroform
perfluorotoluene (standard)
methyl chloroform
benzene
propyl acetate isomer
2-methylhexane
2,3-dimethyl pentane
trichloroethylene
n-heptane
C7H14
C7H14
toluene
C8H18
C8H16
n-octane
tetrachloroethylene
C8H16
ethylbenzene
meta- and /or p_ara-xylene
(continued)
91
-------
Table 18 (cont'd)
Chromatographic
Peak No. Compound
29 styrene
29a ortho-xylene
30 isopropylbenzene
32 C,-alkyl benzene
33 CU-alkyl benzene
34 C,-alkyl benzene
35 a-methylstyrene
35a 1,2,4-trimethylbenzene
36 1,2,3-trimethylbenzene
38 diethylbenzene
39 C,-alkyl benzene (tentative)
40 C,-alkyl benzene
41 ii-undecane
42 C,-alkyl benzene
43 C,-alkyl benzene
44 Ce-alkyl benzene
44a C10H12
45 tetrahydronaphthalene
46 naphthalene
92
-------
234 567
DEGREE OF CHLORINATION
Figure 33. Isomeric distribution of PCNs between soil and air
at Koppers. o - Air, x - Water.
93
-------
5.0 SAMPLING AND ANALYSIS OF POLYCHLORINATED NAPHTHALENES IN THE VICINITY
OF MANNING PAPER COMPANY, GREEN ISLAND, NEW YORK
Samples were collected near Manning Paper Company, Green Island, New
York on November 10-12, 1976. The samples collected were found to contain
low-to-undetectable amounts of polychlorinated napthalenes (PCNs). The
sampling and analysis are discussed in detail below.
5.1 FIELD SAMPLING
(34) (34)
Manning has 325 employees and manufacturers rope, kraft papers ,
various grades and colors of paper for electrical insulation, filter paper,
saturating paper and abrasive backing paper. Manning has been reported
as a manufacturer of capacitor tissue paper and therefore a potential user
of PCNs.
Manning began operation in 1914 and is presently housed in a compact
two story facility in downtown Green Island, New York. The facility is
bordered on three sides by a crowded mixed residential and commercial
district and on the fourth by a lumber yard, railroad, small stream and
Interstate highway immediately to the west. The plant appeared to be in
operation 24 hours a day during the sampling period.
Green Island is a small (33,000) town in the Troy, New York metro-
politan area. Most of the houses and businesses appear to be much as they
were 30-50 years ago. Besides Manning, which is in the center of town,
Bendix Friction Materials Division, Ford (radiators), and Zak, Inc. are
large industries clustered about six blocks north of Manning. Troy, to the
east; Watervliet to the Southwest; and Cohoes, to the northwest; are larger
cities, but similar in their appearances. The attitude of the indigenous
population was such that the procurement of secure sampling locations was
difficult. Both home owners and businessmen warned of the danger of theft
and vandalism.
94
-------
Figure 34. Map of Green Island and surrounding area with sampling
locations for Manning - PI - 11/10-11/76.
- Manning Plant Site
95
-------
Figure 35. Map of Green Island and surrounding area with sampling
locations for Manning - P2 - 11/11-12/76.
- Manning Plant Site.
96
-------
5.1.1 Air
Despite initial difficulties, two 24 hour air samples was obtained
(35)
along transects of the site as shown in Figures 34 and 35. The air
samplers were placed relatively far from the plant in order to optimize
collection of emitted materials from a 50-70 m smokestack which occasionally
belched forth black smoke.
The terrain of the area is generally flat, except for the Hudson, which
flows in a 10 m deep channel and Interstate 787 which is 5-10 m above grade.
After a 10 cm snowfall the first day (before samplers were set out), the
weather was humid, overcast and calm. The temperature generally hovered
slightly above freezing.
The sampling protocol data are summarized in Table 19.
5.1.2 Water
There was no evidence of sewage treatment facilities for Manning. In
1967 , the plant had no treatment facilities and was expected to tie into
a municipal (North Albany) system when it was built. An employee of the
Rensselaer County Wastewater Treatment Plant stated that Green Island was
not hooked up to their plant and that he thought they dumped sewage directly
into the Hudson. However, no outfalls were observed and no sign of water
emission from Manning was detected.
Interval samplers were located upstream and downstream of Green Island.
Upstream was near Location 1 and downstream at Rensselaer County Wastewater
Treatment Plant about 3 km from Manning. One 24 hour sample was collected
at each site. The only problem encountered was that the tide (1 m) came in
after sampling had begun and almost inundated the samplers.
A grab water sample was taken from the west bank of the Hudson at about
the middle of Green Island (downstream from Manning). Two grab water sam-
ples were obtained from the small (3 m wide) stream which marks the western
boundary of Green Island. One was upstream from Manning and the other
downstream. Again, no outfalls or other signs of pollution from Manning
were noted, except that the stream flora appeared much denser and more
varied.
97
-------
Table 19. SAMPLING PROTOCOL FOR MANNING PAPER COMPANY,
GREEN ISLAND, NEW YORK.
VO
00
Meteorological Conditions
Period Cycle Location
11/10,11/76 Cla LI
PI L2
L3
LA
C2b L5
L6
L7
L8
L9
L10
Lll
L12
C3 L13
L14
11/11,12/76 Cla LI
P2 L2
L3
L4
C2e L*
L6
C3h L7
C4J L8
L9
L10
Sampling
Time
1630-1630
1530-1530
1657-1555
1700-1510
1545
1615
1515
1445.
1100
1200
1230
1130
1200-2030
1250-2230
1645-1540
1550-1641
1610-1200
1525-1728
1100
1115
1150
1100
1115
1115
Sampling
Volume (t)
91,400
85,900
85,800
82,800
3 coresc
5 cores
5 cores
5 cores
5 cores
5 cores
3 cores
3 cores
2.7
2.9
85,600
92,800
93,000
97,300
1 J,
i X.
5 cores
1 fc
1 ft,
0.95 H
Type of
Sample
APCN
APCH
APCN
APCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
HPCN
WPCN
APCN
APCN
APCN
APCN
SPCNf
SPCNg
SPCN1
WPCNf
WPCNg
WPCNk
(continued)
T (°C) * RH
1-4 56-71
1-4 52-91
1-4 52-91
1-4 50-81
1-4
1-4
-2-3 50-100
-2-4 59-80
-3-4 51-88
-2-3 60-100
Wind Dir./
Speed (kmpli) Other
NW/0-1 Cloudy
W/0-1 slight rain in evening
W/0-1 snow in morning
C snow in morning
NH/0-7 Cloudy
W/0-3 Cloudy
W/0-3 Cloudy
SW/0-3 Cloudy
-------
Table 19. (cont'd)
Period Cycle Location
C5 Lll
L12
L13
Sampling
Time
1244-1309
1157-1223
1157-1223
Meteorological Conditions
Sampling
Volume (Jt)
190
246
214
Type of
Sample
AHC1
AHCm
Alien
Wind Dir./
T (°C) % RH Speed (krapli)
1.5 NE/0-3
1.5 N/0-5
1.5 N/0-5
Other
strong odor of burning
brake linings
24 hour air samples on PDF
Soil sampling along transects
cAbout 5 cm diameter, 13 cm depth
Water samples taken at 100% sampling rate
Crab sediment samples
Small stream 100 m west of Manning, downstream
gSmall stream 100 m west of Manning, upstream
'Grab soil samples
Representative sites around Colonle Landfill
Crab water samples
%lest bank of Hudson river, downstream from Manning
Downwind of Manning, hopefully in plume from smokestack
""llpwind of Manning, upwind of Bendix
Upwind of Manning, downwind of Bendix
Key to Sample Type: APCN - polychloronaphthalenes, air
AHC - hydrocarbons, air
UPCN - polychloronaphthalenea, water
SPCN - polychloronauhthalenea, soil
-------
5.1.3 Soil and Sediment
Three to five core samples were obtained at each of eight sites along
the transects as shown in Figure 19. As nearly as possible, representative
soils of each site were obtained. Multiple samples for each site were
composited after return to the lab to yield eight samples for extraction and
analysis.
Two sediment samples were obtained at the sites where grab water
samples were collected in the stream west of Manning.
Six grab soil samples were collected at the Colonie Dump a few miles
away. While this was cited by some natives as the only place where Manning
would dump solid wastes, nothing at the dump was indicative of industrial
paper wastes. These samples were composited in the lab to yield one sample
for analysis.
5.1.4 Summary of Manning Sampling
The samples collected at Manning are summarized in Table 19. There are
no subjective indications that Manning was producing PCN-impregnated paper.
In fact, the plant was rather innocuous during our visit, with the exception
of intermittant black smoke from the smokestack. During our upwind sampling
near Bendix, there was a strong odor of burning clutch plates which is
probably due to the phenolic resin being manufactured there.
5.2 ANALYSES
Air, soil and water were analyzed according to the procedure detailed
in Appendix A.
5.3 RESULTS AND DISCUSSION
The results of sample analysis for samples collected near Manning Paper
Company are summarized in Table 20.
The air samples collected contained low-to-undetectable PCN levels.
Some of the values were confirmed by "double MID" GC/MS and do not appear to
be artifacts. No pattern of geographic or isomeric distribution was ob-
served. Because of the low levels found in the filter and first plug, the
backup plug was not analyzed. This decision was made by the Project Officer.
100
-------
Table 20. RESULTS OF ANALYSIS OF SAMPLES COLLECTED NEAR MANNING PAPER COMPANY,
GREEN ISLAND, NEW YORK.
Period3 Cycle3 Location3 CIAHTCI cmHi:cli CinHqC1o
1U / 10 o 2. 1U j 3
PI Clb LI FC ~d
1
2
Total — —
L2 F 0.3
2
Total 0.3
L3 F — — 0.6
2
Total -- — 0.6
L3 F — — 0.6
1 0.5 0.8
2 0.5
Total 1.0 0.8 0.6
C2e L5
L6 — — 4.6
L7 — 2.2 9.5
Degree of Chlorlnation
C10H4C14 C10H3C15 C10H2C16 C10HC17 C10C18 Total
1.0 0.5 1.0 — — 2.5 ng/m3
1.0 0.5 1.0 — — 2.5 ng/m3
0.3
0.3 ng/m3
0.2 — — — — 0.8
0.2 — — — — 0.8 ng/m3
0.6
1.3
0.5
2.4 ng/m
4.9 — -- — 4.9 Mg/kg
4.6 Vig/kg
11 11 — — — 34 ug/kg
(continued)
-------
Table 20. (cont'd)
Period Cycle Location
PI C2 L8
L9
L10
Lll
L12
C3f L13
L14
P2 Clb LI
o
ro
L2
L3
L4
<10«7"
—
—
~
—
—
T8
—
F
1
2
Total
F
1
2
Total —
F ~~
i
2
Total
F 0.2
1 — "
2
Total 0.2
Degree of Chlorlnatlon
C10H6C12 C10H5C13 C10H4C14 C10H3C15 C10H2C16 C10HC17
4.0
8.1 2.0 5.2 1.8 0.9
3.6 1.6 1.3 0.4 3.2
_
—
__
—
M ••_ •»« • « •»_ — —
1.0 — — 1.3
1.0 — — 1.3
— _ _ — __ — —
1.8 1.3
— 1.8 1.3
— — — — — — __
—
—
T8
—
Tg
(continued)
clflci8 Total
4.0 Vig/kg
2.3 20 Mg/kg
1.4 12 pg/kg
—
—
T8
__ _ —
2.3
2.3 ng/m3
3.1
3.1 ng/m3
__
—
—
0.2
—
0.2 ng/m3
-------
Table 20. (cont'd)
Period
11/11/76
P2
Cycle Location
C2e L5
L6
C36 L7
CA L8
L9
L10
Degree of Chlorinatlon
C10H;C1 C10H6C12 C10H5C13 C10H4C14 C10H3C15 C10H2C16 C10HC17 C10C18 Total
0.6 — -- -- 0.4 1.1 — 2.6 ug/kg
1.1 0.7 — — 1.8 ug/kg
6.2- — — -- 0.9 0.7 0.3 — 8.1 ug/kg
—
0.2 0.1 — — 0.3 ug/fc
—
See Table 19 for period, cycle, and location designations in the sampling protocol.
Air samples
F - Glass fiber filter; 1 - Top PUF plug; 2 • bottom PUF plug (not analyzed at direction of Project Officer)
No PCN peak detected. Detection limt for air was about 0.3 ng/m3, for soil about 0.5 Ug/kg, and for water
about 0.2 ug/i.
Soil samples
Water samples
-------
One of the three air samples collected on a Tenax GC cartridge was
analyzed by GC/MS/COMP. Examination of the TIC chromatogram (Figure 36) and
lack of relevant findings in the samples analyzed from other locations
prompted the decision not to plot and interpret the mass spectra.
104
-------
30000-.
aoooo-
ioooo-
0-
1000 1050 1100 1 ISO 1200
SPECS 11000 - lllSt Ln 1277inDE2M3/CMMYTRnM5T3F3^af'Ea77/1 OOnOVl
iaso 1300
STEP SPECO-1 INT- 1000
~ '-I-•>--['<•••,
1350
1100
I H 5 0
1SOO
Figure 36. Total ion current chromatogram from GC/MS analysis of organics
collected near Manning Paper Company, Green Island, New York
(P2/C5/L11).
-------
6.0 SAMPLING AND ANALYSIS OF POLYCHLORINATED NAPHTHALENES IN THE VICINITY
OF CORNELL DUBILIER ELECTRONICS CORPORATION, NEW BEDFORD, MASSACHUSETTS
(37)
Cornell Dubilier (CDE) has about 750 employees and manufactures
/OQ\
electronic components and capacitors. They are reported to utilize PCN-
impregnated paper in their processes.
The site was sampled on Saturday, November 13, 1976. A large number of
cars in a small parking lot and on adjacent streets indicated the plant was
operating. The plant is an old (1900-1920 vintage) three-story brick build-
ing. There are no major "industrial" features such as major plumbing or
(39)
cooling towers. The plant is located on a peninsula (Figure 37) called
the "Ft. Area" at the south end of New Bedford. The area surrounding CDE on
three sides is older residential and on one side is the mouth of the Acush-
net River (at least 1 km wide, brackish water).
6.1 FIELD SAMPLING
During the sampling, the temperature was about 5°C, the wind generally
from the northwest at 0-12 kmph, and the skies mostly clear. Details of
samples collected are presented in Table 21.
6.1.1 Air
A faint white emission was observed from the main smokestack. An air
sample was collected with a Tenax cartridge in a position on the sea wall
such that the stack emission was coming down directly toward the sampler.
An upwind air sample was taken at the corner of Cleveland and David Streets.
6.1.2 Soil
Six soil samples were collected from representative sites 0-5 blocks
from the plant site (Figure 37). Samples were collected in a ditch, a
vacant lot, the playground occupying the last half of the same block as CDE
and some residential yards.
106
-------
Table 21. SAMPLING PROTOCOL FOR CORNELL DUBILIER, NEW BEDFORD,
MASSACHUSETTS
Period Cycle Location
11/13/77 Cla Ll
PI L2
L3
L4
L5
L6
C2d L7
t3
C3 L9
L10
Sampling
Time
0940
0945
0950
0955
1000
1010
0940
1000
1008-1039
1047-1113
Sampling
Volume (St.)
1 core
1 core
1 core
1 core
1 core
1 core
1
1
238
229
Type of
Sample
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
WPCN
WPCN
AHCe
AHCf
Meteorological Conditions
Wind Dir./
T (°C) 7. RH Speed (kmph) Other
5C 54 NW/0-16 Clear, gusty wind
4.5 52 NW/0-19
4.5 52 NW/0-16
Grab soil samples
About 5 cm diameter x 13 cm depth
Meteorological conditions did not change noticeably from site to site during course of sampling, except for the wind velocity, which was gusting
Crab water samples
Downwind, in plume from smokestack, on seawall
Upwind
Key to Sample Type: SPCN - polychloronaphthalenes, soil
WPCH - polychloronaphthalenes, water
AHC - hydrocarbons, air
-------
CORNELL - DUBIUER
SMOKE STACK
Figure 37.
CORNELL-DUBILIER
NEW BEDFORD, MA.
Map of the area surrounding Cornell Dubilier, New Bedford,
MA with sampling locations for PI - 11/13/76.
108
-------
6.1.3 Water
Two grab water samples were collected, one from standing water in a
ditch about 300 m north of CDE and one from the brackish water at the edge
of the bay. No streams or ponds were available for sampling.
6.2 ANALYSES
Soil and water samples were analyzed according to the procedures de-
scribed in Appendix A. The two air samples were not analyzed.
6.3 RESULTS AND DISCUSSION
The results of the analysis of samples collected near Cornell Dubilier
Electronics Corporation in New Bedford, Massachusetts are summarized in
Table 25.
6.3.1 Soil
Due to the fact that CDE was a "secondary" sampling site, the six soil
samples were composited into a composite sample. The PCN levels found
(Table 22) were sufficient to warrent full scan GC/MS confirmation (Figure 38,
Table 23). Tetra- and pentachloronaphthalene were confirmed and di- through
octachlorobiphenyl were identified.
6.3.2 Water
No PCNs were found in the two water samples.
109
-------
Table 22. RESULTS OF ANALYSIS OF SAMPLES COLLECTED IN THE VICINITY OF
CORNELL DUBILIER ELECTRONICS CORPORATION, NEW BEDFORD, MASSACHUSETTS
Degree of Chlorination
Perioda Cyclea Location3 CinH7Cl C1ftH,Cl0 C.-H.Cl, C.-H.Cl. CinH,Cl, CinH«Cl, ClnHCl_. CinCl0 Total
lu / 1U o 2. 1U 3 j 10 i\ 4 ID j 3 11) 2 o 1U / ID o
PI
cib
C2C
L1-6C — d 11 180 230
Li 7 — — ~~~ •""• -~~
L8
76 5.0 3.3 -- 500 pg/kg
a
See Table 21 for Period, Cycle and Location designations in the sampling protocol
Soil Samples
Q
Locations 1-6 composited prior to sample workup
No PCN peak detected. Detection limit for soil is about 0.5 yg/kg,
for water about 0.2
water Samples
-------
100.0-
88.0-
60.0-
40.8-
§ 20.0-
CJ
M
(U
.0-
r
5
10
Time (mln)
TIC*2
I I /'I'd i ii i 11*1111 j-| 11 i i ill rivrrr
ri3GSNBCDl-b,2%QViei,150*3HQ*/M,-8,1.6,3-17-77
Figure 38. TIC chromatogram of GC/MS Analysis of CDE composited
soil sample.
-------
Table 23. RESULTS OF FULL SCAN GC/MS ANALYSIS OF CDE
COMPOSITED SOIL SAMPLE
Retention Time Compound
3.34 C12HgCl2
5.74 - 5.89 C12H7C13
6'24 C10H4C14
6.79 - 8.54 C12H6C14
7.29 C12H7C13 4-
8.19 - 10.59 C12H5C15
8.34 - 9.49 C10H3C15
9.29 - 12.14 C12H4C16
11.54 - 13.00 C12H3C17
13.24 CHC1
112
-------
7.0 SAMPLING AND ANALYSIS OF POLYCHLORINATED NAPHTHALENES IN THE VICINITY
OF SPRAGUE ELECTRIC COMPANY, NORTH ADAMS, MASSACHUSETTS
Sprague Electric Company (Sprague), a 50 year-old company with over
1,000 employees, makes a wide variety of electrical components , in-
cluding capacitors and foil.^ ' Sprague is a sprawling complex of old
multi-story buildings in the center of town. On the main site are 5-10
buildings of varying size, the largest being three stories tall and over 100
m long. A second site ("Brown St. Plant"), two or three blocks west, con-
sists of a single, large three story building, and a third site, about 1.5
km east is a large brick structure with five stories. All buildings appear
to be at least 50 years old.
North Adams is a small city (20,000) in the Northwestern Massachusetts
mountains. The city is in a deep, narrow valley running east-west, with a
branch valley running south.
We sampled the site on Sunday, November 14, 1976 (see Table 24). There
were less than five cars at each site, indicating the plant was not opera-
ting. No fumes, or other signs of activity were noted.
7.1 FIELD SAMPLING
7.1.1 Soil
Eleven soil samples were collected at various sites around Sprague as
(41)
shown in Figures 39 and 40. A series of five samples were obtained in
an array around the two plant sites downtown. Four soils were taken along
transects at least 1 km from the plant. Two soil samples were taken near
the plant on the east edge of town. Since Sprague was reported to dispose
of their solid waste at the city landfill^ ', one soil sample was taken at
the landfill. No particular evidence of electrical manufacturing waste was
observed.
113
-------
Table 24. SAMPLING PROTOCOL FOR SPRAGUE ELECTRIC COMPANY,
NORTH ADAMS, MASSACHUSETTS.
P.-rioi Cycle Location
11/14/76 Cla LI
PI L2
L3
L4
L5
L6
L7
L8
L9
-,d L10
" Lll
C3e L12
C4 L13
L14
LI 5
LL6
Sampling
Time
1420
1430
1440
1450
1455
1545
1630
1640
1705
1500
1505
1610
1500
1510
1700
1715
Sampling
Volume (8.)
1 core
1 core
1 COTB
1 core
1 core
1 core
1 core
1 core
1 core
1 core
1 core
lf
0.90
0.80
0.75
0.95
Mefcorolo;,ic.-U Cjr.ditloni-
Type of l/ind Elr./
S:i;..yA<= 1' (°C) % B.H Sptal (ta>ph) Otlier
SPCN 6C 43 W/0-16 Clear
SPCN
SPCH
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
WPCNB
UPCNh
WPCNl
WPCNJ
Key Co Sample Type:
Samples were taken along a transect of the main Sprague plant, as shown on the map.
About 5 cm diameter x 13 era depth
°Durlng the sampling period, meteorological conditions were constant, except variable wind speed.
Samples were taken near Sprague plant on east edge of North Adams
eLandEtll, 5 km south of city
Representative soil collected
^Upstream of all Sprague activity
Downstream of East plant, upstream of two downtown sites
Downstream of all Sprague plants, upstream of sewage plant outfalls
^Downstream of Sprague plants and sewage plant outfall
SPCN - polychloronanhthalenes, soil
WPCN - pojychloronnphthalenes, water
-------
Figure 39. Map of the area surrounding Sprague Electric with Sampling Locations for PI -
11/14/76.
-------
Figure 40. Map of the area to the west of Sprague Electric with sampling
locations for PI - 11/14/76.
116
-------
7.1.2 Water
Four grab water samples were collected. One sample was obtained
upstream of all Sprague sites (near the east plant location). A second
sample, taken downstream of the eastern plant, served as the upstream
sample for the two downstream sites. A third sample was obtained downstream
of the downtown plants, but upstream of the sewage plant. A fourth sample
was obtained downstream of the sewage plant.
No major effluent sources were noted, although two small (<10 cm) pipes
drain from the main plant directly into an adjacent concrete flume (Hoosic
River). It should be noted that Sprague's discharge situation is uncertain.
(43)
Although reportedly they discharge aluminum sulfates, their discharge
permit application is being processed. In addition, the sewage system was
about one or two weeks away from changeover from the old plant at North
(43)
Adams to a newly constructed plant in Williamstown.
7.1.3 Summary of Sprague Sampling
The three large smokestacks at the main plant site were virtually idle
(one stack was faintly emitting during the presampling site visit on October 26,
1976). No fugitive emissions were visible during either visit. Three local
passersby said there was no smell or pollution from Sprague.
7.2 ANALYSES
Soil and water samples were analyzed according to the procedures de-
tailed in Appendix A.
7.3 RESULTS AND DISCUSSION
The results of the analysis of samples collected near Sprague Electric
Company in North Adams, Massachusetts are summarized in Table 25.
7.3.1 Soil
Due to the fact that Sprague was a "secondary" sampling site the 12
soil samples were composited into two composite samples representing six
locations each (L1-L6 and L7-L12). The PCN levels found (Table 25) were
judged sufficient to attempt full scan data on one of the samples, as shown
in Figure 41 and Table 26. High background and low PCN levels prevented
identification, although C^E^Cl^ was tentatively identified at 7.09 minutes.
7.3.2 Water
No PCNs were found in the water samples.
117
-------
Table 25. RESULTS OF ANALYSIS OF SAMPLES COLLECTED IN THE VICINITY
OF SPRAGUE ELECTRIC COMPANY, NORTH ADAMS, MASSACHUSETTS
oo
Period3
PI
o a
Cycle Location C1 nH7Cl
Clb L1-6C ~d
L7-12d
C4e L13
L14
L15
L6
Degree of
C10H6C12 C10H5C13 C10H4C14 C10H
9.4 18 8.
13 25 12
— — — —
—
—
— —
Chlorination
3C15 C10H2C16 C10HC17
7 4.5 3.8
1.2
— —
—
—
—
C10C18 Total
44 ng/kg
— 52 ng/kg
— —
—
—
—
See Table 24 for Period, Cycle and Location designations in the sampling protocol
Soil samples
£
Locations 1-6 and 7-12 composited into two samples prior to workup
No PCN peak detected. Detection limit for soil is about 0.5 pg/kg, for water
about 0.2 pg/5,
water samples
-------
.A.
TIC*10
-•rrv, . ITt-rryl-. rr. .-. .: , r
r.-rrvp-rmJ.i.,..
1
IB
15
20
I14GSNASE1-6,2XOV101,150*3M3*/M,-8, 1.6,03-17-77
Figure 41. TIC chromatogram of GC/MS analysis of Sprague
composited soil sample (P1/C1/L1-6).
-------
Table 26. RESULTS OF FULL SCAN GC/MS ANALYSIS OF SPRAGUE
COMPOSITED SOIL SAMPLE (P1/C1/L1-6)
Retention Time Compound
7.09 - 8.39 C12H6C14
7.09 C10H4C14
9.04 - 10.24 C12H5C15
10.24 - 11.19 C12H,C16
120
-------
8.0 SAMPLING AND ANALYSIS OF POLYCHLORINATED NAPHTHALENES IN THE VICINITY
OF GENERAL ELECTRIC COMPANY, FORT EDWARD, NEW YORK
General Electric (GE) operates two large capacitor plants in Hudson
Falls, New York and Ft. Edward, New York. The plants, although in dif-
ferent villages, are only about 1 km apart. General Electric Company,
Electronic Components, Capacitor Department employs 1,883 people distribu-
ted between the two plants in New York and a third in Lansing, North
Carolina. Both plants appeared to operate two shifts daily, with a third
shift for cleanup and maintenance according to local sources.
The Ft. Edwards plant (GE-FE), a relatively new facility (probably
post-World War II), spreads over a site about 500 x 800 m and manufactures
/oc\
small capacitors, primarily for use in automobiles. Several additions
are apparent and, in fact, an addition at the southwest corner of the
site was being started during our visit. This is reportedly a treatment
plant to handle PCBs. '
The vicinity of GE-FE was sampled on November 15-17, 1976. Nine
air, five water, 18 soil, and two miscellaneous samples were collected.
The sampling protocol is summarized in Table 27.
8.1 FIELD SAMPLING
8.1.1 Air
Four air sampling sites were selected on a transect and two 24 hour
samples collected at each site (Figures 42 and 43).^ ' Since there was
no large smokestack, any emissions would most likely be fugitive. There-
fore, sampling sites were selected fairly close to the plant.
121
-------
Table 27. SAMPLING PROTOCOL FOR GENERAL ELECTRIC, FT. EDWARDS, NY.
to
N>
Period Cycle Location
11/15/76 Cla LI
L2
L3
L4
C2° L5
L6
L7
L8
L9
L10
Lll
L12
C3C L13
L14
C4h L15
L16
L17
C5 L18
Sampling
Time
0850-0914
0855-0930
0907-0852
0850-0947
1430
1240
1400
1200
1500
1000
1530
1215
1425-1630
1512-1620
1130
1240
1500
1600-1628
Sampling
Volume (I)
90,000
91,800
89,200
93,200
3 cores
4 cores
3 cores
3 cores
3 cores
3 cores
3 cores
4 cores
4.0
3.1
1
1
1
225
Meteorological Conditions
Type of Wind Dir./
Sample T (*C) 7. RU Speed (kmph) Other
APCN -7-7 50-100 Variable, 0-3 Hanging from clear to
overcast
APCN
APCN
APCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
WPCNf
WPCN8
WPCN* strong odor of kerosene
WPCNj
WPCNk
AIIC1
(continued)
-------
Table 27. (cont'd)
Period Cycle Location
11/J6, 17/76 Cla LI
P2
L2
L3
C2m L4
L5
L6
L8
C39 L9
L10
Lll
C4h L12
L14
L15
Sampling
Time
0926-1049
0943-0912
0909-0853
1245
1247
1530
1535
1600
1540
1530
1550
1245
1530
1605
1600
Sampling
Volume (ft)
94,800
87,700
88,700
•vO.5
5 cores
2 cores
3 rolls
1 ft
1 ft
1 ft
1 ft
1 ft
Meteorological Conditions
Type of Wind Dir./
Sample T (°C) Z RH Speed (kmph) Other
APCN -5-6b 51-89 NW/0-10 Variable, clear to
overcast
AFCN
APCN
SPCNn
SPCN°
SPCNP
SPCMq
SPCNr
MPCN1
MPCN"
MPCNV
WPCN1
WPCNV
Key to Sample Type: APCN - polychloronaphthalenes, air
AHC - hydrocarbons, air
WPCN - polychloronaphthalenes, water
SPCN - polychloronaphthalenes, soil
MPCN - polychloronaphthalenes, miscellaneous
-------
Table 27. (cont'd)
3.
24 hour air samples on polyurethane foam
Meteorological conditions monitored at centrally located MRI weather
station. There were no perceptible local variations.
Soil samples collected along transects
About 5 cm diameter x 13 cm depth
G
24 hour water samples using interval samplers
Upstream of both GE-FE and GE-HF plants. Collection tubing froze
up overnight, probably several hours of collection lost.
Downstream of both GE-FE and GE-HF plants. Collection tubing froze
up overnight, probably several hours of collection lost.
Grab water samples
Collected from" GE-FE outfall into Hudson River
Collected in slough about 20 m from southwest corner of GE-FE
fence
t)ownstream from GE-FE and GE-HF near downstream interval sampler.
Green floating solid in the river was collected.
Sample collected on bluff above GE-FE outfall (see footnote "i")
Grab soil samples
Sediment from edge of Hudson River about 50 m downstream from GE-FE
outfall
Beige slime in GE-FE outfall. Sample scraped from rocks, leaves, etc.
Collected in valley about 10 m below old Ft. Edwards dump
^Collected from capacitor-dumping area on top of old Ft. Edwards
dump
Collected near capacitor dumping areas in new Ft. Edwards dump
g
Miscellaneous samples
Some water, foil, and paper chopped out of an open large capacitor
"Unburned rolls of paper - probably capacitor paper
•y
Representative small capacitor
WCollected from 50 cm wide, stream directly below (20 m) old Ft. Edwards
dump
XCollected from stream about 1 km west of old Ft. Edwards dump just
above influence into Hudson River. Probably contains runoff from
old Ft. Edwards dump
^Collected from pond at new Ft. Edward landfill
124
-------
HUDSON FALLS
SEWAGE TREATMENT
PLANT
Figure 42. Map of the area surrounding General Electric - Ft. Edwards
with sampling locations for PI - 11/15-16/76.
125
-------
HUDSON FALLS
SEWAGE TREATMENT
PLANT
^J^l HUDSON
FALLS
LANDFILL
Figure 43. Map of the area surrounding General Electric - Fort Edwards
with sampling locations for P2 - 11/16-17/76.
126
-------
8.1.2 Soil
Three to five samples were collected at each of eight sites (a near
and far site along each transect) as shown in Figure 42. An attempt was
made to obtain soil samples representative of each site. The samples
were composited upon return to the lab to yield eight samples for analysis.
8.1.3 Miscellaneous
Several grab soil samples were collected at three dump sites in the
area (these could be considered to be either the Ft. Edwards or Hudson
Falls site).
A visit to the old Ft. Edwards dump about 8 km south of town was
very interesting. This dump was reportedly used from 1953-65 by GE for
( ^fi"}
disposal of their solid wastes. It is not currently used for any
type of dumping. Hundreds of capacitors (assumedly) slightly smaller
than a pack of cigarettes were strewn in several spots around the site.
In one spot a dozen or more large (60 x 40 x 10 cm) capacitors were seen.
A composite soil was taken downhill of the dump and another from the top
near where many small capacitors lay.
A grab water sample was collected from a small (50 cm wide) stream
flowing below the dump. Another water sample was taken from a stream
(presumed to be the same stream) where it flowed into the Hudson.
The dump had been burned over, leaving small unburned rolls of
paper. Since these were unburnable, it was supposed that they may be
PCN-impregnated. A representative sample was collected. A sample of
water, paper, and foil were collected from an open large capacitor. A
small capacitor was also collected.
The new Ft. Edwards landfill was reportedly used by GE to dispose of
solid waste from 1954 to 1975. ' Again, both large and small capacitors
were observed. Two representative samples of soil were obtained.
The Hudson Falls landfill was reportedly used by GE to dispose of
/ o f.\
solid waste from 1969-1975. ' No samples were taken from this site.
8.1.4 Water
Because of the extremely steep bluffs overlooking the Hudson and the
lack of electrical power near the river, a suitable sampling site between
127
-------
the GE-FE and GE-HF plant could not be located. Therefore, we collected
water samples from one site upstream and one downstream of both plants as
shown in Figures 42 and 43. An upstream sample was collected near the
Hudson Falls Sewage Treatment Plant, about 500 m upstream (north) of
GE-HF. The river bank was about 10 m high and steep enough that a rope
was a necessary aid in climbing up and down the bank. The downstream
sample was collected under a highway bridge in Ft. Edwards. This location
was the first available sampling site downstream. It was about 3 km
south of GE-FE. During our first attempt at obtaining an integrated 24
hour sample, the Teflon lines froze on both samplers. Since subsequent
nights were also well below freezing, samples were collected for 12.5
hours during above-freezing daylight hours.
8.1.5 GE-FE Discharge
Two grab water samples were collected on separate days from the out-
fall of the GE-FE discharge into the Hudson. This outfall was an 80-100
cm pipe which gushed forth a milky effluent with the distinct and over-
powering odor of kerosene or turpentine. A sample of a beige slime which
covered the rocks in the outfall, was collected, along with attached
rocks, leaves, etc. A sample of the river sediment was collected about
30 m downstream from the discharge.
8.1.6 Summary of GE-FE Sampling
Subjectively, this site does not appear to be a "dirty" industry.
Nearby residents complained of a glowing red emission from the stacks
late at night, but we never observed this. One major exception to our
impression of the site was the very foul discharge into the Hudson. The
vapors were so dense at the outfall that eye glasses of the person collec-
ting the samples steamed over, he became faint, and developed a headache.
The only difficulty encountered was that of obtaining integrated water
samples.
8.2 ANALYSES
Air, soil and water samples were analyzed by the procedures detailed
in Appendix A. The grab water samples (P1/C4/L15, P1/C4/L16 and P1/C4/L17)
were analyzed by the VOA procedure detailed in Section III.
128
-------
8.3 RESULTS AND DISCUSSION
The analysis of samples collected at GE-FE are summarized in Table
28. Only low-to-undetectable amounts of PCN were found in these samples.
One exception to this conclusion is a sample collected down an embankment
below the old Ft. Edward dump. This sample (P2/C2/L7) contained appreci-
able amounts of most PCN isomers, indicating that these substances may be
leaching from the materials in the dump.
8.3.1 Air Samples Collected on Tenax
Sample P1/C5/L18 from GE-FE, collected on a bluff above the outfall
from the plant, was found to contain a large number of compounds as
evidenced by the TIC chromatogram as shown in Figure 44. Mass spectra
numbers 2240-2420 were plotted and interpreted (Table 29). Although the
mass spectra were not interpreted in detail, the sample contains primarily
hydrocarbons and alkyl aromatics, presumably kerosene components. No
PCNs were detected.
8.3.2 Volatile Organics in Water
Three grab water samples collected at the outfall were purged for
volatile organic analysis (VOA) and analyzed by GC/MS/COMP using the
Varian CH-7. The TIC chromatograms are presented in Figures 45-47.
Since the composition of these samples appeared similar to that found in
the air sample taken on the bluff above the outfall [(P1/C5/L18), Figure
44], these samples were not interpreted.
129
-------
Table 28. RESULTS OF ANALYSIS OF SAMPLES COLLECTED NEAR
GENERAL ELECTRIC COMPANY, FORT EDWARDS, NEW YORK
Period3 Cycle3 Location* C,nH,Cl C,nH,Cl, C,_H,C1,
1U / 1U o e. 10 5 3
PI Clb LI FC 0.3 — d
1 — — «
2
Total 0.3
L2 F —
2
Total
L3 F — — 1.8
1 -- -- 2.0
2
Total — — 3.8
L4 F —
1
2
Total — —
C2e /L5;r— ib'i.y f, *!':••- .^
L6 — —
L7 1.6 — —
Degree of Chlorlnation
C10H4C14 C10H3C15 C10H2C16 C10HC17 C10C18 Total
0.3
— 0.3 ng/m3
__
— — — — — —
1.4 0.6 — -^ — 3.8
1.5 -- -- — — 3.5
2.9 0.6 — — — 7.3 ng/m3
0.3 — — 0.3
1.8 1.3 1.5 — 4.6
1.8 1.6 1.5 — 4.9 ng/ra3
2.5 4.8 — — — 7.3 yg/kg
—
5.1 — — — — 6.7 yg/kg
L9
0.3
0.1
0.6
1.4 2.4 yg/kg
(continued)
-------
Table 28. (cont'd)
Degree of Chlorination
C10H7C1 C10H6C12 C10H5C13 C10H4C1A C10H3C15 C10H2C16 C10HC17 C10C18 T°tal
Period Cycle Location
PI
P2
C2 L10
Lll
L12
C4
Cl
L1A
L16
L17
LI
L2
L3
F
1
2
Total
F
1
2
Total
F
1
2
Total
3.6 1.2
1.0
1.0
0.6
0.7
1.0
5.5 Ug/1
0.6 vg/i
1.0
1.0 ng/m
2.9 3.9
1.0 2.9 3.9 ng/m
(continued)
-------
Table 28. (cont'd)
ro
Degree of Chlorination
Period Cycle Location C^Cl C^Cl,, C^Cl^ C^C^ C^C^ C10H2C16 C^HCl,
P2 Clb L4 F -- — — — — —
1
2
Total
C2e ^L7y~<*yi
-------
««•« '«• «!«• tl*» ««•• t«t 131.
•'ici !!••» • lift! in if//i/no(SMf/cnnvrcfltriP't rct'77'itnovi fTir t»tci*i znr> I»M
Figure 44. Gas chromatography/mass spectroscopic analysis of air samples
collected above water outfall at General Electric Company,
Ft. Edwards, New York (P1/C5/L18).
-------
Table 29. COMPOUNDS IDENTIFIED IN AIR ABOVE OUTFALL
AT GENERAL ELECTRIC COMPANY, FT. EDWARDS, NEW YORK
Chromatographic
Peak No.
1
2
3
4
7
8
9
9a
10
11
14
15
16
16a
22
24
23-29
Compound
C -alkyl benzene
C^-alkyl benzene
C_ -alkyl benzene
C_-alkyl benzene
C -alkyl benzene
C -alkyl benzene
G10H22
C,-alkyl benzene
C11H24
C11H16
°10H16
C11H24
C -alkyl benzene
C12H26
C12H26
C12H26
C- 0-alkane isomers
134
-------
30000-,
01
eoooo-
i oooo
76SO 7700 7750 7SOO 7850 7900
8043 LM 12771XHDES6U1XCMMYFEOEX1 FEB•77/1 0000V10 STEP SPECOM INT- 1000
60SO
Figure 45. Total ion current chromatogram from gas chromatography/mass spectroscopic
analysis of organics purged from water samples collected in outfall stream
near General Electric Company, Ft. Edward, New York (P1/C4/L15).
-------
30000-1
co
Os
soooo-
10000-
81 00
SPCCD ei
•I •) i-T-pl""!1-'"!""'1"1!1'"1"'!1 11"
8150 8200 8ZSO 6300 6350 8400 8450 8SOO
863H Ltl 12771./RBKS8M3/CMHVFEGE/1 r EB' 7T \ 0 OHOV1 0 STEP SPEC8-1 IMT • 1000
Figure 46. Total Ion current chromatogram from gas chromatography/mass spectroscopic
analysis of organics purged from water samples collected in outfall stream
near General Electric Company, Ft. Edward, New York (P1/C4/L14).
-------
30000-1
10000-
6700
SPECB 8700
8750 6800 8850 8900
LM 1 2771 XRBK4GMI ^CMNYFEBE/ 1 FEB • 77/ 1 0 OMOV I 0
89SO 9000
STEP SPECW'l IHT- 1000
90SO
9100
9150
9EOO
Figure 47. Total ion current chromatogram from gas chromatography/mass spectroscopic
analysis of organics purged from water samples collected from Hudson River
below outfall stream near General Electric Company, Ft. Edwards, New York
(P2/C2/L5).
-------
9.0 SAMPLING AND ANALYSIS OF POLYCHLORINATED NAPHTHALENES IN THE VICINITY OF
GENERAL ELECTRIC COMPANY, HUDSON FALLS, NEW YORK
Hudson Falls (Figure 48) is a village of 8,000 people located on the
Hudson River in upstate New York. Ft. Edwards, located directly south of
Hudson Falls has 3,700 residents. The terrain of the area is generally
flat, except for the Hudson which flows in a 20-40 m deep channel with very
steep banks.
The Hudson Falls plant (GE-HF) is considerably older than the Ft.
Edwards plant. It consists of 8-15 buildings (depending on how one differ-
entiates interconnecting structures) tightly packed into a small site on
the edge of the Hudson River. The major building along Sumpter Street is
30 x 150 m and three stories high. Judging from the materials in the
storage yard, this plant manufactures large capacitors for use in electrical
substations.
To the north of the plant is the Hudson Falls Sewage Treatment Plant.
To the east is a residential neighborhood. To the south is an older manufac-
turer, Sandy Hill, which appears to make steel shafts, etc. To the west is
the Hudson River.
The site was sampled on November 16-19, 1976. Eight air, three water,
and seven soil samples were collected. The sampling protocol is summarized
in Table 30. Because of the proximity of this plant site to that of General
Electric Company, Ft. Edwards, New York and certain sampling difficulties,
some of the samples were taken from common locations.
9.1 FIELD SAMPLING
9.1.1 Air
Air samples were collected on November 17-19, 1976. A transect was
jp and sites selected fairly far away (Figures
HF because of the large (at least 50 m) smokestack.
(43)
set up and sites selected fairly far away (Figures 48 and 49) from GE-
138
-------
HUDSON FALLS
SEWAGE TREATMENT
PLANT —
*^T, HUDSON
FALLS
LANDFILL
Figure 48. Map of the area surrounding General Electric - Hudson Falls
with sampling locations for PI - 11/17-18/76.
139
-------
HUDSON FALLS
SEWAGE TREATMENT
PLANT _
Figure 49. Map of the area surrounding General Electric - Hudson Falls
with sampling locations for P2 - 11/18-19/76.
140
-------
Table 30. SAMPLING PROTOCOL FOR GENERAL ELECTRIC, HUDSON FALLS, NEW YORK
Period Cycle Location
11/17.18/76 Cla LI
PI L2
L3
L4
C2C L5
L6
L7
L8
C3
11/18.19/75 Cl3 LI
P2 L2
L3
L4
C2f L5
L6
Sampling
Time
1100-0931
1115-1023
1035-0957
1135-1045
1225
1120
1210
1135
1250
0951-1024
1035-0941
1010-1006
1058-0920
0900
0827-2109
Sampling
Volume («.)
84,100
86,400
87,300
86,500
3 cores
3 cores
3 cores
3 cores
0.8 .
91,700
86,300
89,400
83,600
3.4
4.0
Meteorological Conditions
Type of Wind Dir./
Snmple T (°C) 7. RH Speed (kraph) Other
APCN -3-8b 60-81 variable 0-10 Overcast
APCN
APCN
APCN
SPCN
SPCN
SPCN
SPCN
WPCN6
APCN 0-7b 56-75 variable 0-10 variable cloudiness
APCN
Al'CN
APCN
WPCN8
WPCNh
Key to Sample Type: APCN-polychloronaphthalenes, air
WPCN-polychloronaphthalenes, water
SPCN-polychloronapl\thalenes, soil
-------
Table 30. (continued)
a
24 hour air samples on polyurethane foam.
Meteorological data monitored on MRI weather station. No perceptible
local variations.
Q
Soil samples collected along transect. Note: Several soil samples from
GE-FE sampling protocol are on GE-HF transect and thus were not duplicated.
Samples collected on 11/16/76.
About 5 cm diameter x 13 cm depth.
Q
Grab water samples collected from puddle in Sumpter Street, about 3 m
from edge of GE-HF main manufacturing building. Collected on 11/15/76.
On 11/18/76, the area was asphalted.
24 hour water samples using interval samplers. Sampler located upstream
of GE-HF and GE-FE .as in GE-FE. protocol.
%ater lines had clogged with no sample- collected. Grab sample was
collected.
Samples downstream of both GE-HF and GE-FE as in GE-FE protocol. Sampler
was run at 100% sampling rates to keep lines from freezing.
142
-------
9.1.2 Water
An integrated water sample was collected at L6, about 3 km downstream
of GE-HF (also downstream of General Electric Co., Ft. Edward, New York).
The corresponding upstream sampler became clogged during sampling, so a
grab sample was collected at the end of the sample period. A grab water
sample was collected from a puddle in Sumpter Street about 5 km from the
main manufacturing building. This water may have been seeping up from some
underground source. Three days later the area was asphalted over. A large
outfall from the plant into the Hudson River was observed but not sampled
because of the necessity of scaling a cliff on GE property.
9.1.3 Soil
Soil samples were collected in four locations along the GE-HF transect.
The other four locations were also on the GE-FE transect and samples from
this protocol will be used to complete the set.
9.1.4 Summary of GE-HF Sampling
No particular difficulties were encountered during the sampling of GE-
HF, except a clogged water collection line and those imposed by the presence
of the Hudson River.
9.2 ANALYSES
Air, soil, and water samples were analyzed according to procedures
detailed in Appendix A.
9.3 RESULTS AND DISCUSSION OF SAMPLES COLLECTED NEAR GENERAL ELECTRIC,
HUDSON FALLS, NEW YORK (GE-HF)
The analyses of samples collected near GE-HF are summarized in Table
31. As with GE-FE, the results are almost all near or below the detection
limit, although two soil samples (P1/C2/L7 and P1/C2/L8) contained PCNs in
concentrations about an order of magnitude higher than at other locations.
On advice of the Project Officer, the air samples from P2 were not analyzed,
since the values obtained from PI were low.
143
-------
Table 31. RESULTS OF ANALYSIS OF SAMPLES COLLECTED NEAR
GENERAL ELECTRIC COMPANY, HUDSON FALLS, NEW YORK
Period Cycle Location
PI Clb LI
L2
L3
L4
C2e L5
L6
L7
L8
C3£ L9
C1()H7C1 C10H6C12
1 0.8 2.0
2
Total 0.8 2.0
1 — 3.5
2
Total — 3.5
1 — 0.3
2
Total — 0.3
P — — —
1
2
Total
1.0
0.3
__
1.9
—
Degree of Chlorlnatlon
C10H5C13 C10H4C14 C10H3C15 C10H2C16 C10HC17 C10C18 Total
2.8
2.8 ng/m3
0.5 — — 0.5
__ __ 3.5
0.5 — — 4.0 ng/m3
3.4 — 3.4
2.5 — — — — -- 2.8
2.5 — — — 3.4 — 6.2 ng/m3
0.8 — — -- 1.5 2.3
4.8 — — 1.3 1.0 — 7.0
4.8 0.8 — 1.3 1.0 1.5 9.3 ng/m3
1.0 VJg/kg
0.3 yg/kg
6.8 3.5 1.6 — — — 12 pg/kg
1.5 1.3 4.7 Mg/kg
__
(continued)
-------
Table 31. (cont'd)
Period Cycle8 Location8
Degree of Chlorination
C10H7C1 C10H6C12 C10H5C13 C10H4C14 C10H3C15 C10H2C16 C10HC17 C10C18 Total
P2
C2f L5
L6
01
See Table 30 for Period, Cycle and Location designations in the sampling protocol
24 hour air samples
"F - Glass fiber filters; 1 - Top PUF plug; 2 - Bottom PUF plug (not analyzed aa per directions for Project Officer)
No PCN peak detected
eSoil samples
Water samples
-------
10.0 SAMPLING AND ANALYSIS OF POLYCHLORINATED NAPHTHALENES IN THE VICINITY
OF CORNELL DUBLIER ELECTRONICS COMPANY, SANFORD, NORTH CAROLINA
Cornell Dubilier Electronics, Sanford, North Carolina (CDS) employs
(44)
between 500-1000 people and manufactures "fixed capacitors". Since
(44)
this plant purchases capacitor tissue, foil, and chemicals , it is a
potential user of PCNs.
Cornell Dubilier began operations at this site in 1953 and currently
occupies a one story 50 x 200 m building in an industrial park at the
southern outskirts of Sanford, North Carolina. To the northwest is a small
housing development. To the northeast and southeast are widely spaced
mixed commercial and residential buildings. To the west and southwest is
an industrial-commercial area with small textile companies, a shopping
center, and miscellaneous small businesses.
Sanford is a town of 12,000 located in the central Piedmont region of
(44)
North Carolina where prominent industrial products are textiles and bricks.
10.1 FIELD SAMPLING
10.1.1 Air
Two 24-hour air samples were collected at each of four points along
(45)
a transect as shown in Figures 50 and 51. The plant is low, there are
no large smokestacks, and the terrain is fairly flat, so the samplers were
placed relatively close (275-900 m) to the plant. A water tower at the
south corner of the plant served as a landmark. Weather conditions were
constantly monitored by the MRI weather station which was set up in a
secure area at Location 3 (Figure 50).
During the collection of the air samples, two short puffs of black
smoke were observed emitting from a smokestack at the plant. A light
146
-------
ATLANTIC
WESTERN
RAILROAD
Figure 50. Map of the area surrounding Cornell Dubilier, Sanford, NC
showing sampling locations for PI - 12/7-8/76.
147
-------
ATLANTIC
WESTERN
RAILROAD
Figure 51. Map of the area surrounding Cornell Dubilier, Sanford, NC
showing sampling locations for P2 - 12/8-9/76.
148
-------
steady snow/rain during the first sampling period resulted in a total of
1.6 cm of precipitation. The sampling protocol is summarized in Table 32.
10.1.2 Water
Upstream and downstream water samples were obtained from a small
stream which runs along the southern edge of the CDS plant site (about 10
m from the plant itself). The stream was about 0.5-1.6 m wide above the
plant and widens out into a swampy lagoon beside the plant. The downstream
sample was collected near a culvert which channels the stream under a dirt
lane. The flow of water appeared to be significantly greater (about 3-4
fold) below the plant than above. The sampling protocol is summarized in
Table 32.
10.1.3 Soil
Three soil samples were collected at each of eight locations sur-
rounding CDS as shown in Figure 52 (Ll-8). The samples were combined after
returning to the lab to yield eight composite samples for analysis. Four
samples were collected within 50 m of the CDS plant (although not on company
property) along each side of the plant. The other four samples were collec-
ted 800-1000 m from the plant on each of the transects. The sampling
protocol data is summarized in Table 32.
10.1.4 Miscellaneous
Several chunks of an amber-colored resinous material were observed
near the CDS plant. A sample (about 20 cm diameter, 5 cm thick) was
collected.
10.1.5 Summary
Due in part to the cooperation of the people at the sampling loca-
tions, the sampling of air and water went very smoothly. The weather
refused to be cooperative, and the bulk of the sampling was accomplished
under a light, intermittent rain or snow. Due to the weather and other
factors, the soil sampling was delayed. When the soil samples were collec-
ted, the sampling was accomplished with ease. The plant appeared relatively
innocuous, although occasional puffs of black smoke were observed. Several
residents stated that they had never smelled any emissions from the plant.
149
-------
Table 32. SAMPLING PROTOCOL FOR CORNELL DUBILIER ELECTRONICS COMPANY,
SANFORD, NORTH CAROLINA
Meterological Conditions
Period Cycle Location
12/7-8/76 Cl3 LI
PI
L2
L3
L4
C2C L5
L6
tn 12/8-9/76 Cla LI
0 P2 L2
L3
L4
1/6/77 Cle LI
P3 L2
L3
L4
L5
L6
L7
L8
C2 L9
Sampling
Time
1431-1418
1446-1434
1413-1350
1509-1451
1325-1508d
1450-1515d
1450-1400
1445-1416
1407-1445
1502-1432
1230
1245
1300
1315
1330
1400
1415
1430
1445
Sampling
Volume (£)
88,000
88,900
88,200
89,000
4.0
4.0
87,800
87,800
9,200
87,800
3 cores
3 cores
3 cores
3 cores
3 cores
3 cores
3 cores
3 cores
Type of
Sample T (°C) Z RH
APCN 10-lb 90
APCN
APCN
APCN
WPCN
WPCN
APCN 8-lb 40
APCN
APCN
APCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
SPCN
amber colored resinous solid
Wind Dlr./
Speed (kmph) Other
SW/3-6 1.65 cm precipitation
during 24 hour period
N/3-11 Intermittent Snow
24 hour sample on PUF
bTemperature and wind continuously monitored by MRI at I<3. Conditions at other locations were not
significantly different
CInterval water samples collected at 50% sampling rate
Sampling period was 25.5 hours
CSoil sample collection delayed due to weather and other problems
Key to Sample type:
SPCN - polychloronaphthalenes, air
WPCN - polychloronaphthalenes, water
SPCN - polychloronaphthalenes, soil
-------
ATLANTIC
WESTERN
RAILROAD
Figure 52. Map of the area surrounding Cornell Dubilier, Sanford, NC
showing sampling locations for P3 - 1/6/77.
151
-------
10.2 ANALYSES
Air, soil, and water samples were analyzed by the procedures detailed
in Appendix A. The amber-colored resinous soil sample (P3/C3/L9) was
analyzed by refluxing a 5 g piece for 15 min with toluene. The toluene
extract was then concentrated and chromatographed as all other samples.
10.3 RESULTS AND DISCUSSION
The results of analysis of samples collected near Cornell Dubilier
Electronics Company in Sanford, North Carolina are summarized in Table 33.
10.3.1 Air
The air samples all contained PCNs in amounts significantly greater
3
than the background. The average concentration found was 19 ng/m with a
3 3
range of 9.8-31 ng/m for the first sampling period and 17 ng/m (from 9.8-
33 ng/m ) the second day. No significant pattern is discernible, neither
with respect to isomeric distribution nor distribution along the transects
as illustrated in Figures 53 and 54.
10.3.2 Water
The water samples contained low-to-undetectable PCN concentrations.
10.3.3 Soil
Total PCN concentrations for the soil samples collected during P3 are
presented in Figure 55 with a detailed sampling map in Figure 56. The
average PCN concentration found in soil was about 240 yg/kg. Two samples,
L2 and L5 are roughly an order of magnitude higher than the other samples.
Air-borne deposition does not appear to be the cause of these high concen-
trations, since, as illustrated in Figures 57 and 58, the annual and fall
wind roses for the area indicate predominant northeasterly and southwesterly
winds. These wind patterns are not concurrent with the soil PCN concen-
trations found. The area at L2 had been graded within the past several
years and is near where the amber resinous solid (P3/C2/L9) was found.
Since this sample contained higher than average concentrations of PCNs, it
is possible that other solid wastes which may have been mixed into the soil
during grading to account for this high concentration. The sample at L5 was
obtained along the bank of a small (M. m wide) stream which flowed along
152
-------
Table 33. RESULTS OF ANALYSIS OF SAMPLES COLLECTED NEAR
CORNELL DUBILIER ELECTRONICS COMPANY, SANFORD, NC
(Jl
CO
Period" Cycle3 Location3
h r
PI Cl LI F
1
2
Total
L2 F
1
2
Total
L3 F
1
2
Total
L4 F
1
2
Total
C2f L5
L6
C10H7C1
0.3
0.3
—
—
Te
T
—
0.3
0.3
—
C10H6C12
—
0.8
0.8
0.5
0.5
1.8
1.8
—
0.5
0.5
--
Degree of Chlorlnation
C10«5C13 C10H4C14 C10H3C15 C10H2C16
—
2.5
2.5
2.8
2.8
6.0
6.0
—
4.0
4.0
—
0.5
— —
13 3.3
13 3.3
1.1
4.3
5.4
2.0
2.0
0.5
2.0
2.5
__
0.1
—
—
—
1.9
3.0
4.9
__
—
—
0.4
1.0
1.4
—
C10HC17
—
5.8
5.8
1.6
3.0
4.6
—
—
0.6
—
0.6
—
C10C18 Total
— —
25
25 ng/m3
6.3 11
6.3 20
13 31 ng/n>3
9.8
9.8 ng/m3
1.4 2.9
7.8
1.4 11 ng/m3
—
0.6 iig/J.
(continued)
-------
Table 33. (cont'd)
Period Cycle Location
P2 Clb LI
L2
L3
L4
P3 Cl8 LI
L2
L3
L4
F
1
2
Total
F
1
2
Total
F
1
2
Total
F
1
2
Total
ft ti p-l /» tl rj-1
C10H?C1 C1()H6C12
-
—
—
0.8
0.8
1.6
— —
0.8 0.8
0.8 0.8
— —
3.8
3.8
0.4
4.8
__
0.7
C10H5C13
„
7.5
7.5
—
5.0
5.0
«.
1.0
1.0
16
16
10
210
—
2.0
Degree of Chlorination
C10H4C14 C10H3C15 C10H2C16 C10HC17
__
1.5 0.8
1.5 0.8
1.0 1.7 1.5
—
1.0 1.7 1.5
— — — . —
2.3 — — 1.5
2.3 — — 1.5
6.3
6.3
15 6.8 1.3
170 39 3.0 25
2.1 2.3
—
C10Clg Total
— —
9.8
9.8 ng/m3
5.0
5.8
11 ng/m3
15
— 15 ng/m
6.5 33
6.5 33 ng/ro
34 tig/kg
470 yg/kg
7.1 12 yg/kg
2.7 yg/kg
(continued)
-------
Table 33. (cont'd)
Degree of Chlorination
Period Cycle Location C^Cl C^Clj C^C^ C^Cl^ C^Clj C^Clg C^HCl, C1()C18
L5 — 1.3 52 130 55 5.3 3.3 5.8
L6 — — — — — 1.0 1.3 2.6
L7 11 — 2.5 2.3 2.4 1.3 1.7 8.7
L8 —
C2h L9 15 71 660 160 18
U1
Ui
a. See Table 32 for period, cycle and location designations in the sampling protocol.
b. Air sample
c. F = glass fiber filter; 1 » top PUF plug; 2 - bottom PUF plug; (not analyzed at direction of Project Officer).
d. No PCN peak detected. Detection limit for air is about 0.3 ng/m , soil about 0.5 yg/kg, and water about 0.2 yg/J..
e. Trace
f. Water samples
g. Soil samples
h. Amber -colored resinous solid.
Total
250 yg/kg
4.9 vig/kg
30 pg/kg
—
920 yg/kg
-------
9.8
25
2
Figure 53. Total PCN concentration (ng/m ) in air near Cornell Dubilier
Electronics Company, Sanford, NC (Period 1).
156
-------
/ 15
9.8
"•"N
I
33
/ I!
Figure 54.
Total PCN concentration (ng/m ) in air near Cornell Dubilier
Electronics Company, Sanford, NC (Period 2).
157
-------
30
250
12
4.9
Figure 55. Total PCN concentration (yg/Kg) in soil near Cornell Dubilier
Electronics Company, Sanford, NC (Period 3).
158
-------
ATLANTIC
WESTERN
RAILROAD
Figure 56. Map of the area surrounding Cornell Dubilier, Sanford, NC
showing sampling locations for P3 - 1/6/77.
159
-------
% FREQUENCY SCALE
0123456
CD
21 16 10 6 3 0 kt
WIND SPEED CATEGORY
8.0
AVERAGE WIND SPEED
ALL DIRECTIONS: 6.9 kt
7.6
ANNUAL WIND ROSE
Figure 57. RALEIGH-DURHAM AIRPORT, NORTH CAROLINA
1955-1964
160
-------
% FREQUENCY SCALE
0123456
21 16 10 6 30 kt
WIND SPEED CATEGORY
8.4
6.7
8.4
6.8
AVERAGE WIND SPEED
ALL DIRECTIONS: 6.6 kt
FALL WIND ROSE
Figure 58. RALEIGH-DURHAM AIRPORT, NORTH CAROLINA
1955-1964
161
-------
the edge of an area used to store old production equipment and other materi-
als. It may be that the PCNs found at this site were either deposited or
leached from the storage area.
10.3.4 Amber-Colored Resinous Solid Sample
This sample (P3/C3/L9) was found to contain 920 yg/kg PCNs, which
was confirmed by "double MID" analysis and also by full scan GC/MS as
shown in Figure 59 and Table 34. The large diffuse peak at about 15
minutes is not identifiable, but is probably due to constituents of the
"plastic". The presence of trichlorobenzene and tetrachlorobenzene, while
not relevant to this project may be of interest.
Table 34. RESULTS OF FULL SCAN GC/MS ANALYSIS OF AMBER-COLORED
RESINOUS SOLID COLLECTED NEAR CORNELL DUBILIER, SANFORD,
NORTH CAROLINA (P3/C3/L9)
Retention Time Compound
0.44 trichlorobenzene
0.89 tetrachlorobenzene
1.84 dichloronaphthalene
4.69 trichloronaphthalene
162
-------
100.fl-
.0-
60.0-
40.0-
20.0-
.0-
-1
*•".•>•>.•••.•»i'»'s>"*j
«
5 10
B3GS15NFCD,2%OV-101,150°3m8°/m-8, 1.8 kV, 03-17-77
15
Figure 59. TIC chromatogram of GC/MS analysis of amber-colored resinous solid
collected near Cornell Dubilier Electronics Company, Sanford, NC
(P3/C3/L9).
-------
11.0 REFERENCES
1. Kover, F. D. Environmental Hazard Assessment Report. Chlorinated
Naphthalenes. EPA 560/8-75-001 (December, 1975).
(§)
2. Halowax^, Chlorinated Naphthalene Oils and Waxlike Solids, Technical
Bulletin, Koppers Co., Pittsburgh, PA.
3. Brinkman, U. A. Th., and H. G. M. Reymes. Polychlorinated Naphthalenes,
J. Chromatogr., 127, 203-243 (1976).
4. Chloronaphthalenes, Amer. Ind. Hyg. Assoc., Hygienic Guide Series,
Jan-Feb, 1966.
5. Koeman, J. H., H. C. W. Van Velzen-Blad, R. DeVries, and J. G. Vos.
Effects of PCB and DDE in Cormorants and Evaluation of PCB Residues
from an Experimental Study. J. Reprod. Fert., Suppl. 19, 353-364
(1973).
6. Crump-Wiesner, H. J., H. R. Feltz and M. L. Yates. A Study of the
Distribution of Polychlorinated Biphenyls in the Aquatic Environment.
U. S. Geol. Surv. J. of Res., _!, 603-7 (1973).
7. Law, L. M., and D. F. Goerlitz. Selected Chlorinated Hydrocarbons in
Bottom Materials from Streams Tributary to San Francisco Bay. Pest.
Mon. Mon. J. , .8(1), 33-34 (1964).
8. Howard, R. H., and P. R. Durkin. Preliminary Environmental Hazard
Assessment of Chlorinated Naphthalenes, Silicones, Fluorocarbons,
Benzenepolycarboxylates and Chlorophenols. EPA-560/2-74-001 (November,
1973).
9. Stratton, C. L., and J. B. Sosebee, Jr. PCB and PCT Contamination of
the Environment Near Sites of Manufacture and Use. Environ. Sci.
Tech., 10, 1229-33 (1976).
10. Dennis, D. S. Polychlorinated Biphenyls in the Surface Waters and
Bottom Sediments of the Major Drainage Basins of the United States.
In National Conference on Polychlorinated Biphenyls, Nov. 19-21,
1975, Chicago, 111., EPA 560/6-75-004, (PB253-248).
164
-------
11. Bidleman, T. F., and C. E. Olney, Science, 183, 516 (1973).
12. Bidleman, T. F. , and C. E. Olney, Bull. Environ. Contamin. Toxicol. ,
11, 442 (1974).
13. Evaluation of a Method for the Analysis of Airborne Polychlorinated
Biphenyls, Environmental Science and Engineering, Gainesville, Florida,
Contract No. 68-01-2978, OTS, EPA, Washington, D. C. 20450, 1976.
14. Lewis, R. G., A. R. Brown, and M. D. Jackson, Paper No. 78, 173rd
National ACS Meeting, New Orleans, LA, March 25, 1977.
15. Lewis, R. G., R. C. Hanisch, K. E. MacLeod and G. W. Sovocool, J.
Agr. Food Chem., _24, 1030 (1976).
16. Registry, Thomas
17. Ciancia, J., Hudson-Champlain and Metropolitan Coastal Comprehensive
Project Industrial Waste Survey: John A. Manning Paper Co., Green
Island, N. Y., 1967 (USEPA, Region II).
18. Pellizzari, E. D., Development of Method for Carcinogenic Vapor
Analysis in Ambient Atmospheres, Research Triangle Institute, Report
Prepared for Office of Research and Development, U. S. Environmental
Protection Agency, EPA-650/2-74-121, Contract No. 68-02-1228, ROAP 21
BEG, Task 05, Program Element No. 1AA010, 148 pp., July, 1974.
19. Pellizzari, E. D., Development of Analytical Techniques for Measuring
Ambient Atmospheric Carcinogenic Vapors, Contract No. 68-02-1228, EPA
600/2-75-076, 187 pp., Nov., 1975.
20. Pellizzari, E. D., Development of Analytical Techniques for Measuring
Ambient Atmospheric Carcinogenic Vapors, Contract No. 68-02-1228,
1976 Annual Report, accepted.
21. Pellizzari, E. D., Identification and Analysis of Ambient Air Pollu-
tants Using the Combined Techniques of Gas Chromatography and Mass
Spectrometry, EPA Contract No. 68-02-2262, 1976 Annual Report,
accepted.
22. Pellizzari, E. D., J. E. Bunch, R. E. Berkley and J. McRae, Anal.
Lett., j), 45 (1976).
23. Pellizzari, E. D., J. E. Bunch, R. E. Berkley and J. McRae, Anal.
Chem., 48, 803 (1976).
165
-------
24. Huckins, J. N. , J. E. Swanson, and D. L. Stalling, J. Assn. Offic.
Anal. Chem., _57, 416, (1974).
25. Armour, J. A., J. Assn. Offic. Anal. Chem., 56, 988 (1973).
26. Hutzinger, 0., S. Safe, and V. Zitko, Intern. J. Environ. Anal.
Chem., 2^, 95 (1972).
27. Perchlorination Procedures, Analabs, Inc. North Haven, Conn., 06472.
28. Safe, S., in National Conference on Polychlorinated Biphenyls,
November 19-21, 1975, Chicago, II., F. A. Ayer, ed., EPA-560/6-75-004
(PB-253 248) 1976, p. 94.
29. Goerlitz, D. F., and L. M. Law, J. Assn. Offic. Anal. Chem., 57, 176
(1974).
30. Bellar, T. A., and J. J. Lichtenberg, The Determination of Volatile
Organic Compounds at the yg/& Level in Water by Gas Chromatography,
EPA-670/4-74-009, 1974.
31. Bellar, T. A., and J. J. Lichtenberg, J. Amer. Water Works Assn., 66,
739 (1974).
32. Pellizzari, E. D., USEPA Contract No. 68-03-2368, unpublished results,
1977.
33. Turner Jr., D. B., Workbook of Atmospheric Dispersion Estimates,
USDHEW, Public Health Service Publication No. 999-AP-26. Cincinnati,
Ohio, p. 29.
34. 1971 New York State Industrial Directory, New York State Industrial
Directory, 2 Penn 'Plaza, NY, 10001.
35. Map Source: Capital District including Albany, Troy, and Schenectady,
Arrow Maps. Dist. by Troy News Co., 176 3rd St., Troy, NY, 12181.
36. Personal Communication, John Ciancia, EPA Region II, Edison, NJ.
37. Personal Communication, New Bedford, Massachusetts Chamber of Com-
merce, November, 1976.
38. Personal Communication, E. E. Green, Gulf South Research Institute,
New Iberia, LA., November, 1976.
39. Map Source: The Prince and Lee Co., Publishers, 270 Orange St., New
Hanover, CN, 06509, Dist. by Service News Co., Popes Is., New Bedford,
MA.
40. Directory of Manufacturers, Massachusetts Department of Commerce,
1971.
166
-------
41. Map Source: Map of North Adams, Berkshire Ct., MA, obtained from
City Hall, Revised June, 1976.
42. Personal Communication, Richard McKnight, Hoosic Water Quality Dist.,
City Hall, Williamstown, MA.
43. Map Source: USGS 7 1/2' Hudson Falls, NY, 1966.
44. North Carolina Directory for Manufacturing Firms, 1974-1975, North
Carolina Department of Labor, Raleigh, North Carolina.
45. Map Source: Sanford City Limit and One Mile Jurisdiction, June 30,
1976, distributed by Sanford, NC City Offices.
167
-------
APPENDIX A
ANALYTICAL METHODS FOR POLYCHLORINATED NAPHTHALENES
168
-------
PROCEDURE FOR CLEANUP OF POLYURETHANE FOAM PLUGS
1. Cut 5 cm diameter x 13 cm long plugs from sheet of Olympic 2315
polyurethane foam.
2. Mark each plug with an identification number in the top using a hot
wire.
3. Place four plugs in bottom of clean four liter beaker, add 500 ml hot
toluene (100°C).
4. Compress the plugs 10 times using a one liter Erlenmeyer flask.
5. Let sit five minutes on steam bath.
6. Repeat Steps 4 and 5.
7. Compress the plugs and decant the toluene.
8. Add 250 ml fresh, hot toluene and repeat steps 4 through 7.
9. Repeat Step 8 three times (total of five extractions).
10. Using clean tweezers, transfer each plug into a foil-wrapped wide-
mouthed jar and cover loosely with a foil-lined cap.
11. Dry in vacuo at 50° for 12 hours.
12. Remove from oven, tighten cap and store away from potential contami-
nants.
PROCEDURE FOR EXTRACTION OF CHLORONAPHTHALENES FROM POLYURETHANE FOAM
PLUGS AND GLASS FIBER FILTERS
1. Using cleaned tongs, remove foam plugs and filters from storage jars
and place them in 400 ml beakers.
2. Add 150 ml of toluene to beakers containing foam plugs and 50 ml
toluene to beakers containing filters.
3. Compress the foam plug 10 times to the bottom of the beakers with a
125 ml Erlenmeyer flask, soak for five minutes and compress an addi-
tional 10 times.
4. Squeeze the toluene out of the plug and decant into a flat bottom
boiling flask. Similarly decant the toluene from the glass fiber
filter into a separate flask.
5. Repeat Steps 2 through 4 two more times.
6. Concentrate in a flat bottom boiling flask topped with a Snyder column
to approximately 15 ml.
169
-------
7. Transfer concentrate to 1.5 x 120 mm culture tube, assuring quanti-
tative transfer with small portions of petroleum ether. Blow down
under N- at <25°C just to dryness.
8. Dilute to approximately 1 ml with hexane and proceed with column
cleanup.
9. Concentrate column eluant with a Kuderna-Danish (K-D) apparatus to 2.0
ml.
PROCEDURE FOR EXTRACTION OF WATER SAMPLES
1. Record total water volume.
2. Take 200 ml aliquot.
3. In separatory funnel, shake 5 min with 25 ml toluene.
4. Repeat Step 3 two times (total extract volume is ca. 75 ml).
5. Dry the combined toluene layers with ^2 g of Na_SO .
6. Reduce volume of combined extracts in flat-bottom boiling flask
topped with a Snyder column to approximately 15 ml. Transfer concen-
trate to 15 x 120 mm culture tube, assuring quantitative transfer with
small portions of petroleum ether.
7. Blow-down under N_ just to dryness.
8. Reconstitute the sample in a small amount of hexane and proceed with
column chromatography cleanup procedure.
PROCEDURE FOR EXTRACTION OF POLYCHLORONAPHTHALENES FROM SOIL SAMPLES
1. Weigh 50 g of soil into a 1 quart screw cap jar.
2. Add 50 ml of diethyl ether, shake and allow to stand overnight.
3. Remove diethyl ether in vacuo at room temperature.
4. Add 5 ml of distilled-deionized water to dampen soil.
5. Add 40 ml of acetone and shake for 20 minutes.
6. Add 80 ml of toluene and shake an additional 10 minutes.
7. Decant acetone-toluene extract through glass wool into a one liter
separatory funnel.
8. Repeat Steps 5 through 7 two more times.
9. Extract combined organic fractions with 500 ml water.
*
Procedure adapted from D. F. Goerlitz and L. M. Law, J. Assoc. Offie.
Anal. Chem., 57, 176-181 (1974).
170
-------
10. Back extract water three times with 25 ml portions of toluene.
11. Dry combined organic extracts with sodium sulfate.
12. Concentrate to 15 ml in a flat bottom boiling flask topped with a
Snyder column. Transfer concentrate to 15 x 120 mm culture tube,
assuring quantitative transfer with small portions of petroleum
ether.
13. Blow-down under N0 at <25°C just to dryness.
/
14. Dilute to approximately 1 ml with hexane and proceed with column
cleanup.
COLUMN CLEANUP PROCEDURE
1. Silica gel (Davison Chemical Division, W. H. Grace, Baltimore, MD),
Grade 923, 100-200 mesh is washed with toluene, followed by hexane,
dried at 130° for 16 hr and stored in a sealed amber bottle.
2. Using a 1.0 x 30 cm glass column, pack with a plug of glass wool,
silica gel in a hexane slurry to 10 cm height, and 1.0 cm Na^SO,.
3. Wash column with 50 ml hexane to settle the bed and clean any residual
contaminants.
4. Transfer sample to column in 1.0 ml or less solvent (preferably
hexane) with washing.
5. Elute the PCNs with 50 ml hexane.
6. The foam background and pesticides are eluted with toluene.
7- Concentrate hexane eluate in K-D apparatus, followed by nitrogen blow
down if necessary to achieve a detectable concentration.
8. Analyze by GC/ECD or GC/MS as described elsewhere.
GAS CHROMATOGRAPH/MASS SPECTROMETER ANALYTICAL CONDITIONS
Instrument: Finnigan 3300 Quadrupole gas chromatograph/mass spectrometer
with PDP/12 computer.
Column: 180 cm x 2 mm i.d. glass.
Column Packing: 2% OV-101 on Chromosorb W HP.
Oven Temperature: 150°, 3 min, 8%nin to 230°, Hold.
Flow Rate: 30 cc/min, helium.
MID Ions: 164, 188, 196, 230, 266, 300, 336, 368, 404 (nominal).
171
-------
Full Scan: 110-500 m/e.
lonization Voltage: 70 eV (nominal).
Detector Voltage: 1.8 - 2.2 kV.
DESCRIPTION OF GAS CHROMATOGRAPH/MASS SPECTROMETRY/COMPUTER INSTRU-
MENTATION
Varian MAT CH-7
The Varian MAT CH-7 is a low resolution magnetic sector single focusing
instrument. A resolution of 1,000 (10% valley definition) is attainable
by the instrument using low resolution slits; a resolution of 3,000 (10%
valley definition) may be attained by use of high resolution slits. Scan
speed is variable from approximately one sec/mass decade to 18 hr/mass
decade. Calibration of the system is routinely performed with perfluoro-
kerosene.
The Varian 620/L computer is on-line with the CH-7 system. Long-term
storage of data is on line-track, IBM-compatible magnetic tape. The com-
puter system subsequently treats the stored data in several different ways
to facilitate interpretation: (a) a reconstructed gas chromatogram (the
abscissa is spectrum number) is routinely made in order to correlate the
GC peaks (observed in the analog mode) with scan number; (b) any given
mass spectrum or the entire series of scans is corrected for background
signal (column bleed, other contaminants); (c) plots of intensities of
specific ions (mass fragmentography) are made from the scan data. This
information is often useful, when correlated with retention time data,
in simplifying the identification of particular compounds; (d) normalized
mass spectra are plotted using different types of normalization or ampli-
fication factors in order to facilitate identification; (e) hard copy out-
put of normalized data in digital form, with various forms of background
correction, is also available.
The GC system on the CH-7 mass spectrometer is a Varian Aerograph
1700 gas chromatograph. Glass capillary columns CVLOO m, prepared in
house) are used on this system. The capillary column is interfaced to
the ion source through a single-stage glass jet separator. The system is
equipped with a specially designed thermal desorption injection system ~ '
172
-------
to allow injection of volatiles desorbed from a Tenax cartridge onto the
capillary as a discrete-fine band.
The Varian 620/L computer interfaced with a MAT CH-7 mass spectrometer
consists of a 12K central process with teletype, high speed paper tape
reader/punch, and Statos-31 electrostatic printer/plotter. Its principal
bulk storage device is a DEC IBM-compatible magnetic tape unit. The com-
puter is also fitted with a hardware multiply/divide card and a dual disc
system to speed up mass spectral data processing.
The mass spectrometer interface to the computer consists of a hardware
multiplexer which allows the sampling of the multiplier output as well
as the total ion current monitor and a Hall effect generator which is
fitted in the CH-7 to sample the magnetic field. These signals, appro-
priately coded, are entered into the computer during data acquisition time.
Peak time and intensities are computed on-line and stored, along with Hall
probe and TIC information, on magnetic tape. In the continuous scanning
mode, spectra are acquired every 6-7 seconds for the entire duration of the
GC run. Using 600 ft tape reels, there is no difficulty in acquiring sets
of over 2,000 scans for any run, if needed.
After acquisition of the entire GC run, a complete TIC profile can be
generated and plotted on the Statos-31 recorder directly from the raw data.
This plot serves to indicate the quality of the GC run, and may also be
used to identify the scans of interest in the run. Generally, the mass
spectral data must be converted from time information to mass information
at this time. This is normally done by obtaining calibration constants
from a standard perfluorokerosene run produced before or after a given
series of unknown runs. Standard calibration curves are stored in the
computer's core memory.
Finnigan 3300 GC/MS with PDF/12 Computer
The Finnigan 3300 mass spectrometer has a mass range of 1000, with
unit resolution over the entire range. Calibration of the system is
routinely performed with FC-43 for lower mass ranges and tris(perfluoro-
heptyl)-a-triazine in the higher ranges.
The PDP/12 computer is on-line with the Finnigan system. Long term
storage of data is on LING tapes or removable disc packs. The computer can
173
-------
subsequently treat stored data in several different ways to facilitate
interpretation: (a) a reconstructed gas chromatogram is routinely made to
obtain retention times; scan number for a given gas chromatographic peak is
obtained by operator interaction with a CRT display; (b) any given mass
spectrum or an entire series of scans are corrected for background signal
(column bleed, septum bleed, etc.); (c) plots of intensities of specific
ions (mass fragmentography) are made from the scan data. This type of
information is often useful, when correlated with retention time data, for
simplifying the identification of particular compounds. Peak areas are
also readily obtainable from these mass chromatograms and can be used to
provide quantitative information; (d) normalized mass spectra are plotted,
using different types of normalization or amplification factors in order to
facilitate identification; (e) hard copy output of normalized data in
digital form, with various forms of background correction, is also avail-
able.
The GC system in use on the Finnigan mass spectrometer is a Finnigan
9500.
The basic hardware of the PDP/12 consists of an 8K central processor
fitted with a teletype, random access disc, CRT display and electrostatic
printer/plotter. The interface to the mass spectrometer was custom-de-
signed and built and consists of both analog to digital as well as digital
to analog interfaces. The latter involves several unique concepts in
interface design, since by using this system it is possible to put the
entire mass spectral scanning operation under computer control. Since
the data acquisition phase of the spectrometer operation is controlled
entirely by the computer, a large number of different types of acquisition
protocols have been implemented. For example, in the multiple ion detec-
tion mode, up to nine individual peaks can be selected within the entire
mass spectral range, and acquired for varying time intervals as selected
by the operator. In the repetative scanning mode, scan intervals down
to one scan per second are possible with entire scans recorded either on
LING tapes or disc.
All data processing operations are carried out interactively by
means of programs stored on the small computer.
174
-------
GC/MS ANALYSIS OF AIR SAMPLES COLLECTED ON TENAX
Volatile compounds were thermally desorbed from the Tenax GC collec-
tion cartridge, trapped in a cooled capillary and then injected onto the
gas chromatograph. The inlet-manifold used for thermally recovering
(1-5)
vapors trapped on cartridges is shown in Figure A-l. For analysis by
GC/MS/COMP, a Varian 1700 gas chromatograph was used to house the glass
capillary column which was connected to the inlet-manifold (Figure A-l).
A Varian MAT CH-7 mass spectrometer with a resolution of 2,000 equipped
with single ion monitoring capabilities was used in conjunction with the
GC. The mass spectrometer was interfaced to a Varian 620/L computer.
A 0.35 mm i.d. x 100 m glass SCOT capillary column coated with OV-101
stationary phase and benzyl triphenylphosphonium chloride (surfactant) was
used for effecting the resolution. The capillary column was conditioned
for 48 hours at 230°C and 1.5-2.0 ml/min of He flow. The operating param-
eters selected for the inlet-manifold GC/MS/COMP system are given in Table
A-l.
Table A-l.' OPERATING PARAMETERS FOR GC/MS/COMP SYSTEM
Parameter Setting
Inlet-manifold
desorption chamber 350°C
valve 220°C
capillary trap - minimum -195°C
maximum +180°C
thermal desorption time 4 min
GLC 100 m glass SCOT-OV-101
50 m glass SCOT-Carbowax 20M or DECS 20-240°C, 4°C/min
80-240'C
carrier (He) flow 0,3 ml/min
transfer line to MS 240°C
MS
scan range m/e 20 -*• 300
scan rate, automatic-cyclic 1 sec/decade
filament current 300 uA
multiplier 6.0
ion source vacuum ^4 x 10~6 torr
175
-------
COMPRESSION SPRING
PURGE
CAS
HEATING CARTRIDGE
CARRIER GAS
TO GLC CAPILLARY
IEATINS AND COOLING BATH
hi CAPILLARY TRAP
VALVE POSITION A
(SAMPLE OESORPTION)
CARRIER
GAS >
PURGE
GAS
TO
GLC
VENT
VALVE POSITION B
(SAMPLE INJECTION )
CARRIER
GAS
PURGE ^
GAS*
Figure A-l.
Thermal desorption/high resolution interface
manifold for gas chromatography.
176
-------
At the beginning and throughout the chromatographic run the mass
spectrometer was set to operate in the repetitive scanning mode. In this
mode, the magnet was automatically scanned exponentially upward from a
preset low mass to a high mass value (m/e 25-300). The scan is completed
in approximately three seconds. At this time the instrument automatically
resets to the low mass position in preparation for the next scan, and the
information accumulated by an on-line 620/L computer and stored on magnetic
tapes or the dual disk system. The reset period require approximately
three seconds. Thus, a continuous scan cycle at six seconds/scan maintained
and repetitively executed throughout the chromatographic run. The result
is the accumulation of a continuous series of mass spectra throughout the
chromatographic run.
Prior to running unknown samples, the system was calibrated with a
standard substance, perfluorokerosene, to determine the time of the ap-
pearance of the known standard peaks in relation to the scanning magnetic
field. The calibration table which was thus generated was stored in the
620/L computer memory. This procedure served to calibrate the masses over
the scanning range.
With the magnet continuously scanning, the sample was injected and
automatic data acquisition initiated. As each spectrum was acquired by
the computer, each peak which exceeded a preset threshold was recognized
and reduced to centroid time and peak intensity. This information was
stored in the computer core while the scan was in progress. In addition,
approximately 30 total ion current values and an equal number of Hall
probe signals were stored in the core of the computer as they were acquired.
During the three second period between scans this spectral information,
along with the spectrum number, was written sequentially on disks or mag-
netic tape, and the computer is reset for the acquisition of the next
spectrum.
Upon completion of the entire GC run, 300-1,000 spectra will have been
recorded. Depending on the information required, the data may either be
processed immediately or additional samples may be run, stored on magnetic
tape and the results examined at a later time.
177
-------
The mass spectral data were processed in the following manner.
First, the original spectral data were read and the total ion current (TIC)
information extracted. Then the TIC information was plotted against the
spectrum number on the Statos-31 recorder. The next stage of the processing
involved the conversion of spectral peak times to peak masses which was
done directly via the dual disk system. The mass conversion was accomplished
by use of the calibration table obtained previously using perfluorokerosene.
Normally one set of the calibration data was sufficient for an entire day's
data processing since the characteristics of the Hall probe are such that
the variation in calibration is less than 0.2 atomic mass units/day. A
typical time required for this conversion process for 1,000 spectra was 20
minutes.
Interpretation of Data: Compound Identification
After the spectra were obtained in mass-converted form, the full
spectral scans from the GC run was recorded on the Statos-31 plotter. The
TIC information available at this time was most useful for deciding which
spectra were to be analyzed. At the beginning of the runs where peaks were
very sharp, nearly every spectrum was be inspected individually to deter-
mine the identity of the component. Later in the chromatographic run when
the peaks were broader, only selected scans were analyzed.
Identification of resolved components was achieved by comparing
the mass cracking patterns of the unknown mass spectra to an eight major
peak index of mass spectra. Since the OV-101 SCOT capillary separates
primarily on the basis of boiling point, particular note was made of the
boiling point of the identified compound for comparison with the elution
temperature of the standard.
178
-------
REFERENCES
Pellizzari, E. D., Identification and Analysis of Ambient Air Pollu-
tants Using the Combined Techniques of Gas Chromatography and Mass
Spectrometry, EPA Contract No. 68-02-2262, 1976 Annual Report,
accepted.
Pellizzari, E. D., Development of Method for Carcinogenic Vapor
Analysis in Ambient Atmospheres, Research Triangle Institute,
Report Prepared for Office of Research and Development, U. S.
Environmental Protection Agency, EPA-650/2-74-121, Contract No.
68-02-1228, ROAP 21 BEG, Task 05, Program Element No. 1AA010, 148
pp., July, 1974.
Pellizzari, E. D., Development of Analytical Techniques for Measuring
Ambient Atmospheric Carcinogenic Vapors, Contract No. 68-02-1228,
EPA 600/2-75-076, 187 pp., Nov., 1975.
Pellizzari, E. D., Development of Analytical Techniques for Measuring
Ambient Atmospheric Carcinogen Vapors, Contract No. 68-02-1228,
1976 Annual Report, in preparation.
Pellizzari, E. D., B. Carpenter, J. Bunch and E. Sawicki, Environ.
Sci. Tech., 9, 556 (1975).
179
-------
APPENDIX B
SUMMARY OF PERCHLORINATION REACTION DEVELOPMENT EFFORTS
180
-------
Perchlorination was attempted under a variety of conditions. In a
typical experiment approximately 1 Ug each of Halowax lOSr5, Halowax
1014 ^H Aroclor 1248 ^ and a CHC13 blank were pipetted into separate reaction
vessels and the solvent evaporated under a nitrogen stream. About 1.0 ml
SbCl- (Cerac-Pure) was pipetted into a cold reaction vessel and immediately
capped. After overnight heating at 170° in an aluminum heating block, the
reaction tubes were cooled in ice. The reaction was quenched with 2 ml 25%
HC1, and the SbCl5 hydrolyzed at 60° for 20 min. If the yellow color
persisted, additional HC1 was added and the hydrolysis carried out at 100°.
Following the hydrolysis, the aqueous phase was extracted with 4 x 1 ml
benzene, the extracts eluted through a column of Na.SO,, combined, blown
down to dryness under a N~ stream and reconstituted to a known volume using
benzene. The sample was then analyzed by GC/ECD.
In lieu of a pure standard for Cn.Cl0, Halowax 1051 ^was initially
J-U o
used to calibrate the retention time (RT) for CinCl0. The gas chromatogram
(§)
of Halowax 1051^on 2% OV-101 (Figure B-l, top) contains 2 major peaks
with RT = 5 min and 9 min. The early eluting peak (RT = 5 min) was identi-
fied by GC/MS/COMP (Finnigan 3200) as C1()HC17 and the latter peak as C-^Clg.
There was an interfering peak in the chromatogram (Figure B-l, bottom)
which caused problems because its RT is the same as that of C1nHCl7. The
interfering peak was traced to a septum problem and was solved by changing
from a 2% OV-101 (200°) column packing to 2% OV-17 (230°). Figure B-2
illustrates the shift of this peak relative to C..-HC1-,.
(R)
The reaction consistently perchlorinates the Aroclor ^ polychloro-
biphenyls (PCB) mixture with little or no by products (Figure B-3), although
recovery is low 0X1.0-20% — no quantitation was performed) . The Halowaxes ^
did not consistently perchlorinate. It had been thought that CinHCl7 was
being formed in many of the trials, but discovery of the interference
mentioned above (Figure B-2) negated these results. With a change of GC
columns, it was found that perchlorination was not being achieved using the
above conditions and that in fact no detectable compounds were present.
Several possible causes of this loss were investigated. The possible loss
during nitrogen blow-down was shown to be negligible. No perceptible
(ID
change was noted in the Halowax 1014 ^ fingerprint pattern in a reconstituted
181
-------
c
o
•a
w
o
Pulse = 100 u sec
Attenuation - 32 x lO"11 afs
Column = 180 x 0.2 cm glass 2", OV-10!
on Chromosorb W (K?)
TC01 • 200°
Tdet = 300°
Tini • 200°
N? flow = 35 ml/min
I
Figure B-l.
3468
Time (mini
©
Halowax 1051 and interfering peak.
10
12
182
-------
Pulse
Attenuation
Column
100 u sec
32 x ID"'1 afs
180 x 0.2 cm glass 2", OV-17
on Chromosort) 750
230°
300°
260°
36 ml/min
220 pg h-1051
unknown interferent
= C10C18
i
4 6
Time (min)
10
Figure B-2. Halowax 1051 T,, on 2%.OV-17.
183
-------
o
Q.
V
ee
V
e
PI use
Attenuation
Column
Tcol
'det
T1nj
N2 How
100 u sec
32 x 10-11 afs
180 x 0.2 cm glass 22 OV-101
on Chronosorb W (HP)
202°
300°
220"
36 ml/min
0246 8
Time (min)
Figure B-3. Perchlorinated Aroclor 1248
10
184
-------
solution after 30 min of N_ stream blowing through the sample tube at 35°.
The possibility of loss due to adsorption on the Na2SO^ column was ruled
out after an experiment showed that the Halowax ^ concentrate was the same
before and after passage through the column. It is apparently important
that a tightly sealed, inert reaction vessel be employed to prevent loss or
contamination of the Halowax® at 170°. Hydrolysis vials (Regis), 5 ml
reaction vials (Kontes), 17 x 150 mm screw cap culture tubes (Kimble), and
vacuum hydrolysis vials (Pierce) were used as reaction vessels in these
experiments. None was totally satisfactory.
The purity of SbCl- from various sources was investigated by running
solvent blanks through the above procedure. All samples of SbCl^ tried
were found to contain some impurities, which elute early in the chromatogram.
Antimony pentachloride from Cerac-Pure was the purest, followed by that
from Ventron and then PCR. A blank sample prepared using Cerac-Pure SbCl5
was analyzed by GC/MS. Very small amounts of CgCl,, C-Clj-Br and C,Cl,Br2
(on the order of 0.1 ppm) were identified. These amounts were not great
enough to interfere significantly with PCN analysis.
The use of both benzene and hexane as solvents for extraction of
C10Clg from the hydrolyzed reaction mixture was investigated, and benzene
was found to be more efficient.
In order to extend the range of conditions evaluated for the per-
chlorination of PCNs, milder conditions, j^.j*. lower temperatures and shorter
times, were investigated. These conditions proved more satisfactory,
although two problems with the reaction have not been solved to date: (1)
apparent reaction by-products were observed even under the mildest of
reaction conditions; (2) recovery of the products was poor, generally less
than 50%. Several experiments were conducted in unsuccessful efforts to
correct these problems.
Table B-l summarizes the perchlorination trials conducted to demon-
strate effects of time and temperature. At lower reaction temperatures,
the vaporization of PCNs did not appear to be a problem nor did leakage
from reaction vessels due to high internal pressures.
Accordingly, the requirement of carefully sealing the reaction vessel
when two at higher temperatures was no longer so rigid. Screw cap culture
185
-------
Table B-l. PERCHLORINATION REACTION RESULTS
00
Reaction
Number
I
II
III
IV
V
VI
VII
VIII
IX
Reaction
Vessel Reagent Temp .
Ha Hexane 210°
(1 ml)
H CHC1, 90°
(1 ml)
H SbCl, 23°
H SbCl5 90°
H SbCl, 90°
H SbCl5 98°
H SbCU 23°
H SbCl5/S02Cl2 23°
H SbCl5/S02Cl2 23°
Time
(hr)
15
15
0
0.75
3.0
73
42
42
42
Starting
Material
H-1051
H-1051
H-1051
H-1051
H-1051
H-1051
H-1051
H-1051
H-1014
Amount
Reacted %
(gg) Analysis Recovered Comments
1.2
1,2
1.2
1.2
1.2
1.2
1.2
1.2
118
No reaction
No reaction
No reaction
C10C18 ' °
C10C18 ' °
C10C17 - 12%
Nothing observed in
Several samll peaks
C1QC18 is only
PCN observed
C1f.Cl0 is about
1U H
166% Hexane Solvent may have
evaporated to yield high
recovery
101% Rxn worked up by evaporating
CHC13, reconstituted in
hexane
66%
Very large peak
0.36
Peak at RRT - 0.
creased markedly
of C10C17
chromatogram
, RRT J PCNs
19% First extraction
52% Small amount of
a1itt"fno nonlro
at RRTC =
36 has in-
at expense
only
early-
13 times
SbCl5 23° 42
H-1014 118 C10Clg, C1()C17H
approx. equal
Many early eluting
peaks
(continued)
-------
Table B-l. (cont'd)
Reaction Reaction
Number Vessel
XI H
XII C
XIII C
XIV C
XV C
XVI C
XVII C
XVIII C
XIX C
Reagent
CHC13 (1ml)
SbCl5
SbCl,
SbClj
SbCl5
SbCl5
SbCl5
Sbci5f
SbCl,
Temp.
23°
70°
70°
70°
70°
70°
70°
170°
55"
55°
Time
. (hr)
42
od
0.25
0.50
1.0
2.0
2.0,
9
od
0.5
Starting
Material
H-1014
H-1014
H-1014
H-1014
H-1014
H-1014
H-1014
H-1014
H-1014
Amount
Reacted
(UK) -
118
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
Analysis Recovered Comments
No reaction. 71%
Fingerprint simi-
lar to standard
C10C18 + C10C17H
already forming
C,,.C1Q 60% No other PCN peaks
-LU o -v _
C10C17H
C10Clg is only 48%
peak observed
C, Cl 21% Small "Char" peak at
10 8 RRT = 0.59
ClnCl. 4% "Char" peak at RRT = 0.59
JJJ o .
Increasing
C1()C18 - "Char" peak at RRT =-0.59
much larger
C10C18 + C10C17H a"
already quite prominent
C.-Cl. predominates, 21%
1U o
c10ci7H : 1/7 CjLOci8
(continued)
-------
Table B-l. (cont'd)
co
00
Reaction
Number
XX
XXI
XXII
XXIII
XXIV
XXV
XXVI
Reaction
Vessel
C
C
C
C
C
C
C
Reagent
Sbci5/so2ci2
SbClj
Sbci5/so2ci2
SbCl5
SbCl,
SbClj
SbCl,
Temp.
55°
55°
55'
50°8
50°
50°
50«
Time
(hr)
0.5
1.0
1.0
1.0
1.0
1.0
1.0
Starting
Material
H-1014
H-1014
H-1014
C10C18
H-1051
H-1014
H-1099
Amount
Reacted
(ug)
1.2
1.2
1.2
0.28
1.2
1.2
0.8
Analysis
C10C18
C10C18 is only
product
C10Clfl is only
product
Char peak at
RRT - 0.73 about
0.25 that of C1()C18
C- Clg only product
C10CVC10C17H - 9
Traces of ClnCl-H
Recovered Commenca
29%
9%
17% Unknown peak at RRT -0.52
47%
38%
37%
10%
XXVII
XXVIII
and lower PCN's
SbCl,
SbClc
50° 1.0 A-1232 1.2
C12C110 peak
increased over standard.
Very little reaction.
C.nClQ and C,-C1,H are
1U o lu /
-78" 0 H-1014 1.2
already quite prominent
(continued)
1"Char" peak at RRT - 0.81
visible as small bump
-------
Table B-l. (cont'd)
00
Reaction
Number
Reaction
Vessel
Reagent
Temp.
Time
(hr)
Starting
Material
Amount
Reacted
(l.B)
Analysis
X
Recovered
Comments
XXIX
SbCI,.
23" 0.75 H-101A
1.2
C. Clg is largest 19% 1"Char" peak at RRT - 0.81
peak Visible as small bump
C1()C17H is quite 14%
prominent other PCN's
still present
XXX C SbClj
XXXI C SbClc
XXXII C SbCI,
XXXIII C SbCl,
XXXIV C SbCI,
23°
40°
23°
40°
23°
40°
23°
40°
23°
40°
0.75 H-1014
1.0
0.75 H-10H
2.6
0.75 H-1014
2
0.75 A-1232
2.6
°;765 Cioci8
1.2 C.-.d- predominates 15%
J.U o
C. gCl^t not detected
Some early-eluting
peaks present
1.2 Similar to XXX 9%
1.2 Very little early 25%
elutes observed
1.3 Very little reaction
observed
0.28 No Charing 8%
"Char" peak is -10% of C1()C18
"Char" peak is -27% of C]0Clg
"Char" peak is -27% of C Ci&
H - Pierce Chemical Co. - Vacuum Hydrolysis Tube
C •> Screw Cap Culture Tube with double Teflon liners
''SbCI added in 0.1 ml allquots: SbCl5/S02Cl2 added in 0.1/0.4 ml ratio.
KltT
Relative Retention Time.
RRT of C1QC18 - 1.00
f=o means reaction quenched Immediately after addition of SbCI,.
e()uant. of Samples XII-XVI is for first extract only.
For samples XVII-XXII, SBCI, added to tubes at 0° in ice bath.
n
Reagent added at room temperature for
reactions XXIII-XXV11.
Reagent added to tubes in dry Ice/acetone
bath for reactions XXVIII-XXXIV.
Recoveries for XXVIII-XXXIV are for one
extraction only.
*Valucs are for Clods found as percent
of theoretical calculated from amount
reacted. Reactions l-III.
H-1014
A-1232
Halowax-1014
Aroclor-1232
-------
tubes or centrifuge tubes with double Teflon liners proved the most advanta-
geous, both in terms of cost and ease of handling. Samples I, II and XI
illustrate the good recoveries obtained from tightly sealed samples under
typical reaction conditions.
Several results are worth special note from Table B-l. It appears
that the presence of S02C12 tempers the action of SbCl^ and gives better
yield and fewer unwanted by-products (compare reactions VII and VIII, IX
and X, XIX and XX, and XXI and XXII). Futher experiments would be needed
in this area before definite conclusions could be drawn. The yield of the
reactions at higher concentrations (IX-XI) is greater than normal, possibly
indicating that the use of smaller amounts of SbCl- would yield better
recoveries.
It appears that even with the addition of SbCl,- to the reaction vessel
in a dry ice-acetone bath and slow heating that the conditions may be too
severe. Possibly dilution of the SbCl- with SCLC1- (mentioned above), CgF.
or other suitable solvent may be a solution.
Figures B-4-B-6 are examples of the more successful perchlorination
attempts for three Halowax mixtures. Note that all of the chromatograms
contain early-eluting peaks which are attributed to unreacted material for
Figures B-5 and B-6. There are no peaks attributable to Halowax 1051
with relative retention time (RRT) = 0.12 (Cn_Cl0 = 100), so the large,
-LU o
early-eluting spike is unidentified.
The effects of the use of SCLCl- are illustrated in Figures B-7 and
B-8. While the percent conversion to C-0Clg is greater with S02C12 (Figure
B-8), the presence of a large peak at RRT =0.5 (RRT ^ C1QHC17) indicates
either partial reaction or by-products.
The effects of time and temperature on the reaction are illustrated in
(R)
Figures B-9-B-13. Figure B-9 is a chromatogram of Halowax 1015 w standard
for retention time comparison. The two large peaks at 4.5 min and 8.5 min
in Figure B-10 (XXVII) show that the concentrations of C1QHC17 and C-^Clg,
respectively, have grown markedly even at dry ice temperatures. After 45
min at 23° (Figure B-ll) , the C..-HC1- and C., _Clg peaks are even more pro-
nounced, and the by-product peak at 6.9 min (RRT = 0.82) is clearly evident.
In Figure B-12 the by-product peak has increased in prominence, while the
190
-------
n
'COL
Pulse Interval = TOOusec
Attenuation = 32 x 10~11 afs
Column = 180 x 0.2 cm, 1% OV-17 on
100/120 mesh Gas Chrom Q
= 215'
= 300°
= 270°
= 57 ml/min
1{il/1.0 ml Benzene Extract
of XXIV
I
o
u
41
c
_L
_L
JL
2 4 6 8 10
Time (min)
f\ "\
Gas chromatogram ( Ni electron capture detection)
.of perchlqrinated Halowax 1051 ®.
Figure B-4.
191
-------
Pulse Interval = lOOusec
Attenuation = 32 x 10"
180 x 0.2 cm, 1% OV-17 on
100/120 mesh Gas Chrom Q
215°
3005
270°
57 ml/min
1X/1.0 ml Benzene Extract
of XXV
10
12
Time (min)
c. o
Figure B-5. Gas chromatogram ( Ni electron capture detection) of
perchlorinated Halowax 1014 ®.
192
-------
•3
s
Pulse Interval
Anenuation = 32 x 10" ' afs
Column
'COL
rDET
rINJ
= 180 x 0.2 cm, 1% OV-17 on
100/120 mesh Gas Chrom Q
= 215°
«300°
= 270°
=« 57 ml/min
1X/1.0 ml Benzene Extract
of XXVI
10
Time (min)
to
Figure B-6. Gas chromatogram ( Ni electron capture
detection of perchlorinatecT'Halowax 1099
193
-------
a
c.
CC
w
a
"E
ac
Pulse Interval
Attenuation = 32 x 10~11 afs
Column
TCOL
TDET
TINJ
N2 flow
Sample:
<180x0.2 cm. i%OV-17on
100/120 mesh Gas Chrom Q
= 215'
300°
= 270'
- 57 ml/min
1X/2.0 ml Benzene Extract
of XXI
I
0 2 4 6 3 10
Time (min)
Figure B-7. Gas chromatogram ( Ni electron capture detection)
of Halowax 1014 ® -gerchlorinated using antimony
pentachloride.
194
-------
Pulse Interval = lOOMsec^
Attenuation
Column
TCOL
TDET
TINJ
N2 flow
Sample:
= 32x 10'" afs
= 180 x 0.2 cm. 1% OV-17 on
100/120 mesh Gas Chrome Q
= 215°
= 300°
= 270°
= 57 ml/min
1X/2.0 ml Benzene Extract
of XXII
Time (min)
go
Figure B-8. Gas chromatogram ( Ni electron capture detection) of
Halowax 1014. ^ perchlorinated using antimony penta-
chloride and sulfuryl. chloride.
195
-------
Pulse Interval =< 100/isec
Attenuation = 64 x TO"11 afs
Column
COL
'DET
'180x0.2cm.2%OV.17on
100/120 mesh Chromosorb 750
«213°
'300°
= 270°
• .44 ml/mm
400 pg H-1014 Standard
I i t I I I I III III
10
12
i
Figure B-9. Gas chroraatogram ( -Ni electron capture detection) of
-Halowax 1014 ® standard.
196
-------
I
£
a
Pulse Interval
Attenuation = 64 x 10"'
Column
afs
180x0.2cm,2%OV-17on
100/120 mash Chromosorb 750
= 213°
= 300'
= 270"
= .44 ml/min
1X/1.0 ml Benzene Extract
of XXVIII (t = 0)
6
10
12
Time (min)
Figure B-10.
Gas chromatogram_( Ni electron capture detection)
of Halowax 1014™ perchlorination reaction. Reaction
quenched immediately after addition of antimony penta-
chloride at -78°.
197
-------
Pulse Interval
Attenuation •
Column :
COL
64 x TO'11 afs
180 x 0.2 cm,
2%OV-17on
100/120 mesh
Chromosorb 750
218°
300°
270°
.44 ml/min
1X/1.0m!
Benzene Extract
of XXIX
(t = 45 min at 23"
Figure B-ll.
Time (mini
f\ T
Gas chromatogram ( Ni electron capture detection) of
Halqwax 1014 ® perchlorination reaction. Reaction
quenched after 0.75 hour at room temperature.
198
-------
Pulse Interval
Attenuation
Column
TCOL
TDET
TINJ
N2 flow
Sample:
« 100MS6C
= 64x 10'11 afs
= 180 x 0.2 cm, 2% OV-17 on
100/120 mesh Chromosorb 750
= 218°
= 300°
= 270°
.44 ml/min
1X/1.0 ml Benzene Extract of
45minat23M.O
hrat40')
10
12
Time (min)
Figure B-12.
f\ ^
Gas chromatogram ( Ni electron capture detection) of
Halowax 1014 © perchlorination reaction. Reaction
quenched after 0.75 hour at room temperature and 1.0
hour at 40°.
199
-------
Pulse Interval :
Attenuation
Column
COL
1 100;isec
'64x 10"'1 afs
: 180 x 0.2 cm, 2% OV-17 on
100/120 mesh Chromosorb 750
1X/1.0 ml Benzene Extract of
XXXI (t - 45 min at 23° 2 hr
at 40°)
10
12
Figure B-13.
Time (min)
f'j
Gas chromatogram ( Ni electron capture detection) of
Halowax 1014 ® perchlorination reaction.- Reaction
quenched after 0.75 hour at room temperature and 2.0
hour at..40°.
200
-------
C1()HC17 peak has virtually disappeared. After 2.0 hr (Figure B-13), the
by-product peak continues to increase, and the C10Clg peak is diminished.
At these mild perchlorination conditions the PCB mixtures examined
were not perchlorinated significantly as illustrated in Figures B-14 and
B-15. It does not appear from these results that suitable conditions will
be found for perchlorination of PCNs and PCBs in the same sample.
201
-------
Pulse Interval•
Attenuation :
Column •
TOOL
TDET
TINJ
N2 flow
Sample:
TOO^sec
32 x tO"1^ afs
180 x 0.2 cm, 1% OV-17 on
100/120 mesh Gas- Chrom Q
215°
300°
270°
57 ml/min
300 pgA-1232 Standard
Figure B-14.
Time (min)
63
Gas chromatogram ( Ni electron capture
detection) of Aroclor 1232 ® standard.
202
-------
Pulse Interval •
Attenuation •
Column
'COL
10Ojisec
32 x 10'1 ' afs
180 x 0.2 cm, 1% OV-17 on
100/120 mesh Gas Chrom Q
215°
300°
270°
57 ml/min
3X/2.0 ml Benzene Extract of
XXVII (A-1232)
Time (min)
Figure B-15.
,63,
Gas chromatogram ( ""Ni electron capture detection) of
Aroclor 1232 ® reacted under Halowax perchlorination
conditions.
203
-------
APPENDIX C
GAS CHROMATOGRAPHIC-MASS SPECTROMETRIC ANALYSIS OF
HALOWAXES^AND AROCLORS®
204
-------
28. CH
. i.sxv
130.Si
,164*5
-196*3
L. 230*5
IUU.3-]
ea.a-
368*15
Figure C-l.
a.
b.
c.
chromatographic-mass spectrometric analysis of Halowax
Total ion current (204 ng) .
Multiple ion detection (4.08 ng)
Multiple ion detection (4.08 ng)
205
-------
to
o
aa.a-
bD.tl-
O.B-
20.0-
.0-
V;^^
-^ '-i.
368*.' 5
2u6i 1C1-
iui-i i-10
111
FILE El HO. f,
II- ItiUU. 2XOVIO 1. 17fuC . SUCC.'M. -8.1. 3KV
Figure C-2. Gas chromatographic-mass spectrometrlc analysis of Halowax-1000 ;
multiple ion detection of 4.08 ng.
-------
•0.-—.
' \ •''•- r \ A
/. • •--,' v v \.
TIC*5
7
"-lO'-Jl
i>4
•J
i "e
SO.CH „-
aB.C-i
\
40.0-;*
fK
"1
- .
C-4
u
«3
o
I
""!''•"
3'
,J. ,- 3JC -0cr
+
es fn
o o
_*o _vn <->
O O CJ
— -^ U1
1U =0
r
/L,
!C
b
,... ... . 193*
230*
33.0-
S3.S-
20.0-
'•.^V'V-'rfV.trj.v.-r'.yjW-'.'^^v.y.'.;..^-^...",...',)/:-. f..V. .•."•'.•.••..V..:-.>.'^.-».-.'-,V. 363*30
..-.-..-A. .- A
',-•-. 300*20
Figure C-3. Gas chrotnatographic-mass spectronetric analysis of Halowax-1001 ;
a. Total ion current (208 ng)
b. Multiple ion detection (4.10 ng)
c. Multiple ion detection (4.10 ng)
207
-------
lea.a-
53.3J.- aC>L£'Z^--:^.-vV:^--^.^ 368*30
IT^~~
.3-1-
a 5 10
FILE EI NO. 7
Figure C-4. Gas chromatographic-mass spectrometric analysis of
Halowax 1001©; multiple ion detection (A. 10 ng).
208
-------
IOG.J-,
so.e-
26.0-
•»«%.•>;•
13'
IIC
. IS
23.3-!
• .C *4 - -
u
- SO -*-
. i.si
aS.a-1, ff.
A3.3-
20. a-.1
.a-
—r
18
.164
15
iC.O-l
:o.i-
MQ.
15
Figure C-5. Gas chromatographic-mass spectrometric analysis of Halowax-1013^;
a. Total ion current (180 ng)
b. Multiple ion detection (3.76 ng)
c. Multiple ion detection (3.76 ng)
209
-------
KJ
H->
o
ice.o-
ao.o-
eo.a-
40.0-
2fl.O-
.6-
u
m"
o
I .
i :
'^j//ffi;:?^^:^^^ 368*20
*20
10
IS
•.r.i...te.y.-i«iL 300*10
FILE E! NO. 8
H-IOl3.2*OVIOI.ird*C.3oCC/n.-S.I.S'KV
I ^^
Figure C-6. Gas chromatographic-ma'ss spectrometric analysis of Halowax-1013®-
multiple ion detection (3.76 ng) . '
-------
ICO.O-
a-j.e-
-ia.il-
20.li-
•'•V •
10
2.1
H- 1C I •». 2XOVI51 l?a*. JOCC/n. 1.6KV
Figure C-7. Gas chromatographic-mass spectrometric analysis of" Halowax-1014 ;
total ion current (232 ng)
-------
UlU.lV
£0.0
60. iV
•ia.0-
U
**
X
•-• rH #H r-t
o V <-> °
1 I i
"'T
10
to
I—1
to
100.0-
ea. b-
fcO.O-
•tu.&-
-i
r
i
.0- •
(I
•f~t *-.v -iT^-'-'-r* r_ .-•!-•* •• '-.-T.' '«'/"""~»i? ;"•'•'• -^'.7'.'"•'•-;'-.:.•-'"••" UV;'-"r V* ""p'.I'^"'..'"*« . J'ilf-*":.\v' * 'i*. " l •' «'•• ' 'r''. -.'"•'"«."" ""'; '*-"'••."•
^*A>Jj».viV*^'l^'^f"^'v^y!'y^;''iV;^^t^y^^:yv
v ... .. . 360
"•A. v. ••• - inn*in
^A^^X%".'..'y''' V.fa.^rf.',.^ ,y J.^.x.,.' ^..
^^••••T>[ir^»*^». p^w.. *ir|>y*iT>V*r|*wnw>< |«Trwr.»f ».t«T*r. rpyn*^.
•5 16
FILE £1
HO.
C'l1.-6. I.9KV
Figure C-8. Gas chromatographic-mass spectrometric analysis of Halowax-1014 ;
a. Multiple ion detection (A.64 ng)
b. Multiple ion detection (4.64 ng)
-------
83.8-
60.0-
40.0-
.28.0-
TIC*5
H-ia3i.2.-:oviai.3accxrt. i
133.OT
se.o-
€0.0-1
43. a-!
L/J
PILE s: NO.
H-133: .2.?CV
.164*1
Figure C-9. Gas chromato^raphic-mass spectrometric analysis of
Halowax-1031®;
a. Total ion current (218 ng)
b. Multiple ion detection (4.36 ng)
213
-------
80.0-
£0.0-
40.Q-
20.0
.0-
C10H6C12
c10..5ci3
•' "I"
*-i-r i-f-^t-^v-m . , j-T
' 1(3
196*5
230*5
ieo.a-i
so. fl-
ea, e-
.0-
FILE El MO. I
H-l03l.ixOVI0l,I7u'C.iULC/-n.-a.l.9KV
b i
tt\V;.^Vvi'«v;->'>-i-^C«»*vv^"V-"ii".;;j^v.v»v.-v.".V'* 196*5
230*5
Figure C-10.,, Gas chromatographic-mass .spectroraetric analysis of Halowax-103rH
a. Multiple ion detection (4.36 ng)
b. Multiple ion detection (A.36 ng)
-------
KKJ.U-,
Cn
tJ.fc
40.0-
20.0- i-
. >.'• '•*..••' i^':.;.v«.rr~y*.,^^.t.t^^y^^^.;^^-^/^v..
-------
.o-i
-J
TIC*10
H- i cs;.:; • :.>A-^.'v.>>,..;i'i...x..^
32.
o
=0
Figure C-12.
2-3
NO. 2
n.--?. l.3
-------
130.0-1
33.0-
SB.O-
•43.0-
29.3-
'» V, ,-. •? '*«' '•' \
.0-f
H-1099,2SOV101,170*C,20CC/M,1,SCT
ia
isa.c-
10
100.0-
30.3-
60.3-1
40.0-j
... TIC
164
NO. 3
foi
Figure C-13. Gas chromatographic-mass spectrometric analysis of Halowax-1099 ;
a. Total ion current (222 ng)
b. Multiple ion detection (4.44 ng)
c. Multiple ion detection (4.44 ng)
217
-------
103.0-1
33. OH
25 33
= !!_£ E! NO. 3
H1C5S, 2r:CV131. 193*C. 3aCC^r,. -3. 1.9KV
Figure C-14. Gas chromatographic-mass spectrometric analysis of
Halowax-1099 ^ multiple ion detection (4.44 ng).
218
-------
iee.a-i
so.a-
S0.0H
43.3-
3iJ
TIC*5
_i»__; .;_„.. i_i ._.
fl-1232.2/.-QV 131.173-'C.33CC/M. l.S
108.8-|.
j ,. ....- ...:
sa.a-j --- - ~
13
j
i
20.0-
\
.a-
1C3.3-.
;>':'-V'-;..:;s V/^,T 196*10
Figure C-15. Gas chromatographic-mass spectrometric analysis of
Aroclor-123^;
a. Total ion current (248 ng)
b. Multiple ion detection (248 ng)
c. Multiple ion detection (248 ng)
219
-------
ISO. 3-,
iS.l-
-It**
. 230*10
S3.3-
,.J
JC.3-: .
20.0-
i
'.''•,"/'••'.' .''.v**;'.>">'•";.*'.•'.'•/'•• 265*10
130.3-
1363*10
Figure C-16. Gas chroraatographic-mass spectrometrie analysis of
a. Multiple ion detection (248 ng)
b. Multiple ion detection (248 ng)
c. Multiple ion detection (248 ng)
220
-------
I
SO. 3-1
ea.3-
.3-
3
PILE El
HO. 1
120.3-.
30.;-;
-V«rr 300*10
330*10
MC. I
i. !.'0-C.3CC:/-M.-3, 1. 3"V
Figure C-17. Gas chromatographic-mass spectronetric analysis of
Aroclor-1232;
a. Multiple ion detection (248 ng)
b. Multiple ion detection (240 ng)
c. Multiple ion detection (248 ng)
221
-------
lea.a-.
so. fl-
, a-
40. fl-
29 0*
• •" -.' -*\ :"•.*..
'/- V *•' v v ^>_x Vr<*«s -ir*1!
v^»-*
a ' ' ' ' 4 ' ' 10'
aa.e-
ss.a-i
j. a-!
23.0-
ryi 154*10
53.3-
"•^r^*
i
Figure C-18. Gas chromatographic-mass spectrometric analysis of Aroclor
a. Total ion current (192 ng)
b. Multiple ion detection (192 ng)
c. Multiple ion detection (192 ng)
-124$
222
-------
JO.3-
.3-
230*10
iaa.3-1
sa.a-
sa.:
•"/*'••:>?; 256*10
MO.
Figure C-19. Gas chromato^raphic-mass spectrometric analysis of
Aroclor-1242®;
a. Multiple ion detection (192 ng)
b. Multiple ion detection (192 ng)
223
-------
so.e
£0.3-
-a.o-
20.0-
100.0-
V»-..-.-..-v/:- 336*10
SB. 3-
ao.:-
jV
.C-r-
a
PILE HI
HO.
368*10
Figure C-20. Gas chromatographic-mass spectrometric analysis of Aroclor-124Z^i
a. Multiple ion detection (192 ng)
b. Multiple ion detection (192 ng)
224
-------
so.an
•w.aH
23.6-:
•IIC*10
2S.5-"
12S.J-
33.0-1
f «
10
Figure C-21. Gas chromatographic-mass spectrometric analysis of
Aroclor-1248 ®;
a. Total ion current (192 ng)
b. Multiple ion detection (192 ng)
c. Multiple ion detection (192 ng)
225
-------
133.3-
sa.z-i
^^rr'':•,•*. ;.'.'.•;.;•.- /;_' '_•;_«• j; 196*10
53.C-
13
FI'-S £1 NO. 2
H- 12-1:5. iSCV.31, l75-»C.3£CCA1.
-3. l.JKV
Figure C-22. Gas chromatographic-mass spectrometric analysis of Aroclor-124^;
multiple ion detection (192 ng)
226
-------
8U.O-
£0.0-
.0-
100.a-
80.i)-
eo.B
Y
20.0-
.0-
12
FILE £1
"•'[ ,-r.-~^~T.-r-.
10
15
HO. 3
. l70«C.3CiCC/ri.-8.1.9KV
20
Figure C-23. Gas chromatographic-mass spectrometric analysis of Aroclor
a. Multiple ion detection (192 ng)
b. Multiple ion detection (192 ng)
-mft
-------
ioo.a-
30.0
Ed.il-
48.0-
•I T*
•i nr
hi
00
lua.o-
60.0-
40. G-
~T
15
FILE El MO. 3
H- 12
-------
60.0-
60.0-
JO.O-
iO.O-
y*~~***^..,..-^,
to
aa.e-
6B.O-
40.0-
26.6-
.6-
FILE El
15
HO. 3
10
it)
Figure C-25. Gas chromatographic-mass spectrometric analysis of Aroclor-1245^
a. Multiple ion detection (192 ng)
b. Multiple ion detection (192 ng)
-------
£0.0-
£0.0
JQ.O-
r*w_
^—--.^..^.^.^^ ,-.',•-
10
100.0-
80.0-
eo.o-
-10.0-
I'O.O-
.0-
FILE El MO. 3
M-l248.2::OVlQl. iru»C,3nCCxn.-8.1.9KV
Figure C-26. Gas chromatographic-mass spe'ctrometric analysis of Aroclor-1248®
a. Multiple ion detection (192 ng) '
b. Multiple ion detection (192 ng)
-------
to
100.On
60. fl-
.o-
-JO.O-
2U.C-
0-
FILE El 110. 3
fl-1243.2~OViai.1704C.30CC-t1.-8.1.9KV
Figure C-2'7. Gas chrotnatographic-mass spectrometric analysis of Aroclor-124Sfi;
multiple ion detection (192 ng)
-------
to
N>
loo.a-
80,0-
60.0-
40.ll-
20.0-
.u- -~
12
i £>a. fl-
a.o-
€0.0-
•M.O-
2U.O
.b-
I
25
FILE El
NO. 3
I'll
"JO
. 404'I•'
J.. ."..!••
--.C-li 1C
2t'j»lo
Figure C-28. Gas chromatographic-mass spectrometric analysis of Aroclor-
a. Multiple ion detection (192 ng)
b. Multiple ion detection (192 ng)
-------
to
to
o.e
.0.6-
tiJ.O-
•a.a
.0
10
TT
20
Figure C-29, Gas chromatoeraphlc-mass spectrometric analysis of
Aroclor-1254^; total ion current (198 ng).
-------
OJ
100.0-
eo.o-
c.U.0-
43.0-
20.0-
i^ 164*10
230*15
196*5
266*5
10
iBO.fl-
, o-
£6.0-
40.0-
**:?
-------
Bd.O-
4U.C-
20.0
i:. .V;.,-.'.y-V;.'.•;;'•,'>/\.'~'<'v»» 464*10
^/'•"'.^foM?^^ 230*15
.-. -^ ..v..v;-.V-...-..--•-"._,,•.• ..-.••..- -..',• --- -;.:. -- :..• — ." "-.'.'>'• -.--.-•-•*" 196*5
266*5
cn
86.0
61). 0-
10.0
.0-
-^—^- 336*15
_^-~ 368*10
-"^^^^ 40',* 10
FILE El HO. -)
H-i:'SJ.2;:OVIOI. l,'0.iC.30i:c -I1.-3.1.9KV
Figure C-31. Gas chromatographic-mass s-pectrometric analysis of Aroclor-1254®-
a. Multiple ion detection (198 ng) '
b. Multiple ion detection (198 ng)
-------
10
100.On
6a.fl-
ee.o-
.....:....,.„.:...... ••-•^..,,,.,r,.c..-w^v^Jl.^...^.,.'.''....'. __• ..". LZZV/ZT^
n
100.0-
eo.o-
60.0-
15
•ua.o
20.0-
.6-
l;'.1,V<'j^*-^;^^-T'' 300*20
JUy.V.W.-V/r^ 336*15
368*10
FILE El
HO. -4
Figure C-32. Gas chromatographic-mass spectrometric analysis of
a. Multiple ion detection (198 ng)
b. Multiple ion detection (198 ng)
330*20
336*15
368*15
404*10
-------
N>
aa.a-
SO.Ci-
-10.u-
20.0
.6-
^'•'•V.V-'V i'.ya'J- •.y.'-y;." ^>.':^i,;^.;'.r;::.' --"-. 1S6* 10
^^V^t;\-'i'l'X;'«V^'.'^;j'»^^a.'i*i8^'^>i'^'a'^'^CriW^^ 164*10
..^ . . . 230*10
12
IBO.O-
15
eo.o
eo.o
.e-
;^i^^xV'^v''^A^^^A^>^:^^^
- •• • •• • -/• • • ••••^SL.- •'•• " - • •'• .'-.-• --.- ••••*• '/••.-.•-: ..--..
! —.;-. •' .'. .- . -.• .- - . •' -
j±,:-y!±^'^te^^^>,\^^^^.:^:^.^,.;r ;.v.•:.,-,. .-, -.;. 196* 10
yi,-v^ii^-3i^-.*.»-.vv.^i^.«.^K/-^i.-.»A-vt'^ r,*>.^X^W-^1^r>':^^i'1T7'-:**-^ 230* 10
FILE Et HO. 5
Figure C-33. Gas chromatographlc-mass spectrometric analysis of ^roclor-126t
a. Multiple ion detection (180 ng)
b. Multiple ion detection (180 ng)
-------
10
w
oo
BO.O
60.0 _
40.0
20.0
in
""""ST
"~ir
n- lii.:;. il-ii'-j 101. l^fnC . -'.XC-II. -3-
Figure C-34. Gas chromatographic-mass spectrometrie analysis of Aroclot
ion current (180 ng)
total
-------
iaa.0-
ea.fl-
ea.u-
•iQ.e-
20.B-
iij^-V-.-.x.. .•',.-... • v.,- iX^iVL-'' .'•.v-'-"-.--. .......
If. H I.
10
FILE El HO- 5
f,-i:6a.2xovioi. iro«c.3«i:.-n.-s. LSI v
Figure C-35. Gas chromatographic-mass spectroraetric analysis of Aroclor-1268^;
multiple ion detection (180 ng)
-------
10B.il-
ea.o-
60.0-
40.0-
20.0
109.8-
iy'.:^.^,-^;.-^-^^^.^^,.^^ Z&l&^^^.&^^^if^^
FILE El 110. 5
fl-1268,2;:OVIO 1,1 ~('>:>*C.3iJCi>n,-8.1. 9KV
300*20
336*15
V;^-i^^v»SCA./-J.>v^^yf^?^
Figure C-36.
Gas chromatographic-mass spectrometric analysis of Aroclor-126{
a. Multiple ion detection (180 ng)
b. Multiple ion detection (180 ng)
-------
ae.e-
40.0-
336*15
368*10
FILE El
110. 5
M.-8. I.9KV
Figure:C-37. Gas chroraatographic-mass spectrometric analysis of Aroclor-126ff^ multiple
ion detection (180 ng)
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Polychlorinaphthalenes in Air, Water and Soil
5. REPORT DATE
June, 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
M. D. Erickson, R. A. Zweidinger, L. C. Michael,
E. D. Pellizzari
8. PERFORMING ORGANIZATION REPORT NO.
Task I Final Report
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute
Post Office Box 12194
Research Triangle Park, North Carolina 27709
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
EPA 68-01-1978
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Toxic Substances
U. S. Environmental Protection Agency
Washington, D. C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
FINAL June 11, 1976-Dec. 11. 1!
76
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This research program was initiated to evaluate possible environmental contami-
nation by polychlorinated naphthalenes (PCNs). The program included sampling and
analytical method development and the collection and analysis of field samples.
A glass fiber filter and two precleaned polyurethane foam plugs in tandem were
used for PCN collection. Recovery of the PCNs from the foam and filter was accom-
plished by triple extraction with toluene. The concentrated extract was chromato-
graphed on a silica gel column and the final volume reduced to 2 ml in a Kuderna-
Danish apparatus. Samples were analyzed by gas chromatograph/quadrupole mass spec-
trometer/computer. The instrument was operated in the multiple ion detection mode
which permitted the detection of <50 pg of a PCN isomer (^0.3 ng/m3 in air). The
presence of PCNs was confirmed from full scan mass spectra or by monitoring the
chlorine isotope ratio.
Air, water, soil, sediment and biota were collected from sites near a PCN manu-
facturer, and six potential users. PCNs were found at all sites sampled although
appreciable quantities were found at only three sites, near the manufacturer and two
possible users.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Polychlorinated naphthalenes (PCNs)
Air
Water
Soil
Fish
Biota
13. DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (This Report)
unclassified
21. NO. OF PAGES
266
20. SECURITY CLASS (This page)
unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA report number as it appears on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE ., . .
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, it used, in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
approval, date of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.). List author's affiliation if it differs from the performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Leave blank.
15. SUPPLEMENTARY NOTES
Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of, Presented at conference of,
To be published in, Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
the primary posting(s).
18. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
the public, with address and price.
19. &20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)
------- |