United State* Office of EPA-660/&-79O03
Environmental Protection Toxic Subttancet February 1979
Agency Washington DC 20460
Toxic Substances
x°/EPA Environmental Monitoring
Near Industrial Sites
Acrylonitrile
EPA 560/6-79-003
-------
EPA-560/6-79-003
ENVIRONMENTAL MONITORING NEAR INDUSTRIAL SITES
ACRYLONITRILE
BY
John Going
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
Contract No. 68-01-4115
Elizabeth Bryan
Technical Director
Vincent J. DeCarlo
Project Officer
Office of Toxic Substances
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
NOTICE
This report has been reviewed by the Office of Toxic Substances,
Environmental Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the views and policies
of the Environmental Protection Agency. Mention of trade names or commercial
products is for purposes of clarity only and does not constitute endorsement
or recommendation for use.
-------
CONTENTS
Sections Page
I Summary 1
II Introduction 4
III Sampling and Analysis Protocol . 5
IV Method Development for Sampling and Analysis 12
V Selection of Sampling Sites 63
VI Discussion of Results 71
References 265
iii
-------
TABLES
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Data Summary for Program Task 3
Summary of Organic Compounds Found in Air and Water Samples.
Gas Chromatographic Conditions
Operating Parameters for GC/MS Analysis of Acrylonitrile
Operating Conditions for Purge and Trap Analysis of
Gas Chromatographic Conditions
Analysis of Acrylonitrile in Air
Analysis of Acrylonitrile in Water
Chromatographic Behavior of Potential Interfering Compounds.
Gas Chromatographic Conditions for Acrylonitrile Analysis. .
Survey of Potential Adsorbents Using Methanol
Recovery of AN From Spiked Adsorbents Using Methanol ....
Humidity~Temperature Observations
Stability of Acrylonitrile in Carbon Bisulfide
Recovery of Acrylonitrile on Charcoal
Recovery of AN in Water Following Adsorption
Results of Azeotropic Distillation Studies
Efficiency of Thermal Desorption of Acrylonitrile
iv
Page
2
3
8
9
11
13
16
17
23
24
26
26
33
34
36
37
41
43
50
-------
TABLES (Continued)
Number Page
20
21
22
23
24
25
26
27
28
29
Components of Purge and Trap System
Operating Conditions for Purge and Trap Analyses of
Acrylonitrile
Purge and Trap System for Acrylonitrile - Specific
Instructions
Retention Times
Sampling Sites
Recommended Sampling Sites
Sampling Schedule
Field Sampling Summary
Summary of Air Sampling Parameters
Air Sampling Data for American Cyanamid, New Orleans,
Louisiana
52
55
56
60
64
69
72
73
74
77
30 Water and Sediment Sampling Data for American Cyanamid, New
Orleans, Louisiana 78
31 Soil Sampling Data for American Cyanamid, New Orleans,
Louisiana 79
32 Meteorological Conditions During Sampling at American
Cyanamid, New Orleans, Louisiana: Source of Data, New
Orleans International Airport, New Orleans, Louisiana. . . 80
33 Acrylonitrile Concentrations in Air Samples From American
Cyanamid, New Orleans, Louisiana 81
34 Acrylonitrile Concentrations in Water Samples From American
Cyanamid, New Orleans, Louisiana 86
35 Acrylonitrile Concentrations in Soil and Sediment Samples
From American Cyanamid, New Orleans, Louisiana 87
v
-------
TABLES (Continued)
Number Page
36 Air Sampling Data for American Cyanamid, Linden, New Jersey
2/22/78-2/23/78 95
37 Air Sampling Data for American Cyanamid, Linden, New Jersey
2/22/78-2/23/78 96
38 Water and Soil Sampling Data for American Cyanamid, Linden,
New Jersey 98
39 Meteorological Conditions During Sampling at American
Cyanamid, Linden, New Jersey, February 21, 1978 to
February 22, 1978: Source of Data, Newark, New Jersey
Airport 100
40 Meteorological Conditions During Sampling at American
Cyanamid, Linden, New Jersey, February 22, 1978 to
February 23, 1978: Source of Data, Newark, New Jersey
Airport 101
41 Acrylonitrile Concentrations in Air Samples From American
Cyanamid, Linden New Jersey 2/21/78-2/22/78 102
42 Acrylonitrile Concentrations in Air Samples From American
Cyanamid, Linden, New Jersey 2/22/78-2/23/78 103
43 Acrylonitrile Concentration in Water Sample From American
Cyanamid, Linden, New Jersey: Azeotropic Distillation . . 107
44 Acrylonitrile Concentrations in Soil Samples From American
Cyanamid, Linden, New Jersey 107
45 Air Sampling Data for Monsanto, Texas City, Texas 112
46 Water and Soil Sampling Data for Monsanto, Texas City,
Texas. 113
47 Wind Conditions During Sampling at Monsanto, Texas City,
Texas: Data From Galveston Airport and Two Meteorological
Stations in the Field 114
48 Acrylonitrile Concentrations in Air Samples From Monsanto,
Texas City, Texas 115
vi
-------
TABLES (Continued)
Number ?age
49 Acrylonitrile Concentrations in Water Samples From Monsanto,
Texas City, Texas: Azeotropic Distillation 121
50 Acrylonitrile Concentrations in Soil Samples From Monsanto,
Texas City, Texas 122
51 Compounds Identified by GC/MS in Air Samples From Monsanto,
Texas City, Texas 124
52 Air Sampling Data for Monsanto, Decatur, Alabama (3/8-3/9) . 129
53 Air Sampling Data for Monsanto, Decatur, Alabama (3/10-3/11) 130
54 Water, Soil and Sediment Sampling Data for Monsanto Company,
Decatur, Alabama (3-8-78) 132
55 Meteorological Conditions During Sampling at Monsanto,
Decatur, Alabama 133
56 Acrylonitrile Concentrations in Air Samples From Monsanto,
Decatur, Alabama, for Two Sampling Periods 135
57 Acrylonitrile Concentrations in Water Samples From Monsanto,
Decatur, Alabama 142
58 Acrylonitrile Concentrations in Soil and Sediment Samples
From Monsanto, Decatur, Alabama 143
59 Compounds Identified by GC/MS in Air Samples From Monsanto,
Alabama 146
60 Air Sampling Data for DuPont, Lugoff, South Carolina .... 151
61 Soil, Water, and Sediment Sampling Data, DuPont, Lugoff,
South Carolina 153
62 Meteorological Conditions During Sampling at DuPont, Lugoff,
South Carolina 154
63 Acrylonitrile Concentrations in Air Sampling From DuPont-
May, Lugoff, South Carolina 156
vii
-------
TABLES (Continued)
Number Page
64 Acrylonitrile Concentrations in Water Samples From DuPont-
May, Lugoff, South Carolina: Azeotropic Distillation. . . 161
65 Acrylonitrile Concentrations in Soil and Sediment Samples
From DuPont-May, Lugoff, South Carolina 162
66 Compounds Identified by GC/MS in Air Sample From DuPont,
South Carolina 166
67 Air Sampling Data for DuPont, Waynesboro, Virginia 169
68 Soil, Water, and Sediment Sampling Data, DuPont, Waynesboro,
Virginia 170
69 Meteorological Conditions During Sampling at DuPont,
Waynesboro, Virginia 171
70 Acrylonitrile Concentrations in Air Sampling From DuPont,
Waynesboro, Virginia . . . 173
71 Styrene and Acrylonitrile Concentrations in Air Samples From
DuPont, Waynesboro, Virginia 176
72 Acrylonitrile Concentrations in Water Samples From DuPont,
Waynesboro, Virginia: Azeotropic Distillation 177
73 Acrylonitrile Concentrations in Soil and Sediment Samples
From DuPont, Waynesboro, Virginia 177
74 EICP Data for Compounds Identified by GC/MS in Air Sample
From DuPont, Waynesboro, Virginia 180
75 Air Sampling Data for Borg-Warner, Washington, West Virginia 183
76 Water and Soil Sampling Data for Borg-Warner, Washington,
West Virginia 185
77 Wind Conditions During Sampling at Borg-Warner, Washington,
West Virginia 186
78 Acrylonitrile Concentrations in Air Samples From Borg-Warner,
Washington, West Virginia 188
viii
-------
TABLES (Continued)
Number Page
79 Styrene and Acrylonitrile Concentrations in Air Samples From
Borg-Warner, Washington, West Virginia 191
80 Acrylonitrile Concentrations in Water Samples From Borg-
Warner, Washington, West Virginia: Azeotropic
Distillation 192
81 Acrylonitrile Concentrations in Soil Samples From Borg-
Warner, Washington, West Virginia 192
82 EICP Data For Compounds Identified by GC/MS in Air Samples
From Borg-Warner, Washington, West Virginia 196
83 Air Sampling Data for B. F. Goodrich, Louisville, Kentucky. . 203
84 Soil and Water Sampling Data, B. F. Goodrich, Louisville,
Kentucky 204
85 Meteorological Conditions During Sampling at B. F. Goodrich,
Louisville, Kentucky, Source of Data - Standiford Field,
Louisville, Kentucky 205
86 Acrylonitrile Concentrations in Air Samples From B. F.
Goodrich, Louisville, Kentucky 207
87 Comparison of Acrylonitrile and Styrene Levels in Air Samples
From B. F. Goodrich, Louisville, Kentucky 210
88 Acrylonitrile Concentrations in Water Samples From B. F.
Goodrich, Louisville, Kentucky: Azeotropic Distillation. . 210
89 Acrylonitrile Concentrations in Soil Samples From B. F.
Goodrich, Louisville, Kentucky 211
90 EICP Data For Compounds Identified by GC/MS in Air Sample
From Goodrich, Louisville, Kentucky 214
91 Air Sampling Data for Monsanto, Addyston, Ohio 218
92 Soil and Water Sampling Data, Monsanto, Addyston, Ohio. . . . 219
93 Meteorological Conditions During Sampling at Monsanto,
Addyston, Ohio, Source of Data - Greater Cincinnati Airport 221
ix
-------
TABLES (Continued)
Number Page
94 Acrylonitrile Concentrations in Air Samples From Monsanto,
Addyston, Ohio 222
95 Comparison of Acrylonitrile and Styrene Levels in Air Samples
From Monsanto, Addyston, Ohio 226
96 Acrylonitrile Concentrations in Water Samples From Monsanto,
Addyston, Ohio: Azeotropic Distillation 227
97 Acrylonitrile Concentrations in Soil Samples From Monsanto,
Addyston, Ohio 228
98 EICP Data for Compounds Identified by GC/MS in Air Sample
From Monsanto, Addyston, Ohio 232
99 Air Sampling Data for Uniroyal, Painesville, Ohio 235
100 Soil and Water Sampling Data, Uniroyal, Painesville, Ohio . . 237
101 Meteorological Conditions During Sampling at Uniroyal,
Painesville, Ohio 238
102 Acrylonitrile Concentrations in Air Samples From Uniroyal,
Painesville, Ohio 239
103 Comparison of Acrylonitrile and Styrene Levels in Air Samples
From Uniroyal, Painesville, Ohio 242
104 Acrylonitrile Concentrations in Water Samples From Uniroyal,
Painesville, Ohio: Azeotropic Distillation 243
105 Acrylonitrile Concentrations in Soil Samples From Uniroyal,
Painesville, Ohio 243
106 EICP Data for Compounds Identified by GC/MS in Air Sample
From Uniroyal, Painesville, Ohio 247
107 Air Sampling Data for Vistron, Lima, Ohio 252
108 Soil and Water Sampling Data, Vistron, Lima, Ohio 253
109 Wind Conditions During Sampling at Vistron, Lima, Ohio. ... 254
x
-------
TABLES (Concluded)
Number Page
110 Acrylonitrile Concentrations in Air Samples From Vistron,
Lima, Ohio 255
111 Acrylonitrile Concentrations in Water Samples From Vistron,
Lima, Ohio: Azeotropic Distillation 259
112 Acrylonitrile Concentrations in Soil Samples From Vistron,
Lima, Ohio 259
113 EICP Data for Compounds Identified by GC/MS in Air Sample
From Vistron, Lima, Ohio 262
xi
-------
FIGURES
Number Page
1 Chromatogram of 15 ng Acrylonitrile in Carbon Disulfide
on Durapak OPN 19
2 Calibration Curves of Acrylonitrile in Carbon Disulfide ... 20
3 Chromatogram of Acrylonitrile and 26 Potential Interferences. 22
4 Recovery of Acrylonitrile Versus Loading 27
5 Apparatus for Evaluating Air Sampling Train 29
6 Effect of Humidity on Recovery of Acrylonitrile 30
7 Relationship of Temperature and Relative Humidity at 12 mg/
liter Absolute Humidity 32
8 Recovery of Acrylonitrile From Calgon OL Charcoal at
Different Humidities 35
9 High and Low Resolution Mass Spectra of Acrylonitrile and
Background 38
10 Stability of Acrylonitrile in Water 40
11 Purging System of Volatile Organic Analysis 45
12 Percentage of Acrylonitrile Remaining During Purging With
Helium 46
13 Cumulative Percent of Acrylonitrile Breakthrough on
Chromosorb 104 48
14 Cumulative Percent of Acrylonitrile Breakthrough on Porapak N 49
15 Trapping and Desorbing System for Volatile Organic Analysis . 51
16 Effect of Temperature on Desorption of Acrylonitrile 53
17 Effect of Time on Desorption of Acrylonitrile 53
xii
-------
FIGURES (Continued)
Number Page
18 Calibration Curve 57
19 Chromatograph of 10 ml of 0.1 ppb AN 58
20 Chromatogram of Potential Interferences 59
21 Comparison of GC/MS Sensitivity for Acrylonitrile 61
22 Location of Sampling Sites 70
23 Sampling Locations at American Cyanamid, Fortier, Louisiana. . 75
24 Wind Patterns During Sampling at American Cyanamid, Fortier,
Louisiana 82
25 Annual Wind Rose for New Orleans, Louisiana 83
26 Gas Chromatogram of Sample A-4 85
27 SIM Plots of 6 ng AN and Air Samples A-4, A-4D and A-5 .... 89
28 Sampling Locatins at American Cyanamid, Linden, New Jersey . . 94
29 Sampling Sectors Defined by Region II, EPA 97
30 Wind Patterns During Sampling at American Cyanamid, Linden,
New Jersey, 2/21/78-2/23/78 104
31 Wind Patterns During Sampling at American Cyanamid, Linden,
New Jersey, 2/22/78-2/23/78 105
32 SIM Plots of 6 ng AN and Air Samples A-2(l), A-3(l), A-4(l)
and A-3(2) 108
33 Sampling Locations at Monsanto, Texas City, Texas Ill
34 Wind Patterns During Sampling at Monsanto, Texas City, Texas . 118
35 Annual Wind Rose for Galveston, Texas 119
36 Mass Chromatogram for Air Sample MTX A-l 125
xiii
-------
FIGURES (Continued)
Number Page
37 Mass Chromatogram for Air Sample MIX A-7 126
38 Sampling Locations at Monsanto, Decatur, Alabama on Day 1 . . 127
39 Sampling Locations at Monsanto, Decatur, Alabama on Day 2 . . 128
40 Wind Patterns During Sampling at Monsanto, Decatur, Alabama . 138
41 Wind Patterns During Sampling at Monsanto, Decatur, Alabama . 139
42 Wind Rose for Huntsville, Alabama (1/60-12/64) Using Star
Program Output 140
43 SIM Plots for Monsanto, Decatur, Alabama 144
44 Extracted Ion Current Profiles for Monsanto, Decatur,
Alabama 145
45 Mass Chromatogram Showing Additional Compounds Identified in
Monsanto, Decatur, Alabama Air Sample A-4 147
46 Extracted Ion Current Profiles of Propionitrile Specific Ions
for Monsanto, Decatur, Alabama Water Samples 149
47 Sampling Locations at DuPont, Lugoff, South Carolina 150
48 Wind Patterns During Sampling at DuPont-May, Lugoff, South
Carolina 158
49 Wind Rose for Columbia, South Carolina 159
50 SIM Plots for DuPont, Lugoff, South Carolina 163
51 Extracted Ion Current Profiles for DuPont, Lugoff, South
Carolina Water Sample W-3 (azeotropic distillate) Along
With a 5-ng AN Standard 164
52 Mass Chromatogram Showing Additonal Compound Identified in
DuPont, Lugoff, South Carolina Air Sample A-8 165
53 Sampling Locations at DuPont, Waynesboro, Virginia 168
xiv
-------
FIGURES (Continued)
Number . Page
54 Wind Patterns During Sampling at DuPont, Waynesboro,
Virginia 174
55 Wind Rose for Lynchburg, Virginia (1/69-12/73) Using Star
Program Output 175
56 SIM Plots for DuPont, Waynesboro, Virginia Air Sample A-2.
Along With a 4-ng AN Standard 178
57 Mass Chromatogram, Showing Additional Compounds Identified
in DuPont, Waynesboro, Virginia Air Sample A-6 181
58 Sampling Locations at Borg-Warner, Washington, West Virginia. 182
59 Wind Patterns During Sampling at Borg-Warner, Washington,
West Virginia 189
60 Extracted Ion Current Profiles for Borg-Warner, Washington,
West Virginia Air Samples A-2 and A-2D Along with a 40-ng
AN Standard 194
61 Extracted Ion Current Profiles for Borg-Warner, Washington,
West Virginia Air Samples A-3 and A-3D Along with a 40-ng
AN Standard 195
62 Mass Chromatogram for Borg-Warner Air Sample A-2D 197
63 Mass Chromatogram for Borg-Warner Air Sample A-2D 198
64 Mass Chromatogram for Borg-Warner Air Sample A-3 199
65 Mass Chromatogram for Borg-Warner Air Sample A-3D 200
66 Sampling Locations at B. F. Goodrich, Louisville, Kentucky. . 202
67 Wind Patterns During Sampling at B. F. Goodrich, Louisville,
Kentucky 208
68 Wind Rose for Louisville, Kentucky 209
69 SIM Plots for B. F. Goodrich, Louisville, Kentucky, Air
Samples A-l and a 2-ng AN Standard 212
xv
-------
FIGURES (Continued)
Number Page
70 SIM Plots for B. F. Goodrich, Louisville, Kentucky, Water
Sample Distillates W-l and W-2 and a 1-ng AN Standard. . . 213
71 Mass Chromatogram Showing Additional Compounds Identified
in B. F. Goodrich, Louisville, Kentucky, Air Sample A-l. . 215
72 Sampling Locations at Monsanto, Addyston, Ohio 217
73 Wind Patterns During Sampling at Monsanto, Addyston, Ohio. . 223
74 Wind Rose for Cincinnati, Ohio 224
75 SIM Plots for Monsanto, Addyston, Ohio, Air Samples A-3 and
A-4 and a 2-ng AN Standard 229
76 SIM Plots for Monsanto, Addyston, Ohio, Water Sample Distil-
lates W-l and W-2 and a 1-ng AN Standard 230
77 Mass Chromatogram Showing Additional Compounds Identified
in Monsanto, Addyston, Ohio, Air Sample A-4 233
78 Sampling Locations at Uniroyal, Painesville, Ohio 234
79 Wind Pattern During Sampling at Uniroyal, Painesville, Ohio. 240
80 SIM Plots for Uniroyal, Painesville, Ohio, Air Samples A-l
and A-2 and a 2-ng AN Standard 244
81 SIM Plots for Uniroyal, Painesville, Ohio, Water Sample
W-2, Water Sample Distillates W-2 and W-3 and a 1-ng AN
Standard 245
82 Mass Chromatogram Showing Additional Compounds Identified
in Uniroyal, Painesville, Ohio Air Sample A-2 248
83 Acetonitrile SIM Plots for Uniroyal, Painesville, Ohio
Water Sample W-2, Water Sample Distillates W-2 and W-3
and a 1-ng AN Standard 249
84 Sampling Locations at Vistron, Lima, Ohio 251
85 Wind Patterns During Sampling at Vistron, Lima, Ohio 256
xvi
-------
FIGURES (Concluded)
Number Page
86 Annual Wind Rose for Findlay, Ohio (1/50-12/54) Using Star
Program Output 257
87 SIM Plots for Vistron, Lima, Ohio, Air Samples A-2, A-3 and
A-5 and a 2-ng AN Standard 260
88 Extracted Ion Current Plots for Vistron, Lima, Ohio, Air
Sample A-4 and a 40-ng AN Standard 261
89 Mass Chromatogram for Vistron, Lima, Ohio, Air Sample A-4 . . 263
90 Mass Chromatogram for Vistron, Lima, Ohio, Air Sample A-5 . . 264
xvii
-------
SECTION I
SUMMARY
The objective of this research program was to provide the Environmental
Protection Agency (EPA) with sampling and analysis capabilities, and to deter-
mine on a task basis the levels of toxic substances in the air, water, soil,
and tissues at designated sources and ambient locations throughout the United
States. The first and second tasks which have been completed on this program
were vinylidene chloride and acrylamide. The third task was acrylonitrile (AN).
Methods for sampling and analyzing AN in air, water, soil and sediment
were developed and a protocol was prepared and approved. Air samples were col-
lected on charcoal and desorbed with carbon disulfide for gas chromatographic
(GC) analysis. Water samples were analyzed by a modified purge and trap tech-
nique or with prior azeotropic distillation. GC analysis with a nitrogen
specific Hall detector was used. Soils and sediments were extracted with water
and analyzed as water samples.
Eleven industrial sites were sampled for AN. A summary of the analytical
results from the eleven plants is given in Table 1. The high and low concen-
trations of AN in air, water, soil or sediments are given for each plant. AN
in air was found at all the plants at levels from < 0.1 to 325 ^g/m • The
highest levels were found at an ABS, SAN plant and an acrylonitrile, acryl-
amide plant. The presence of AN correlated with the observed wind patterns.
The highest levels were found downwind of the plant or at stations crosswind
but very close to the plant. The highest levels of AN found in water were
3,500 and 4,300 ^g/liter from an acrylic, modacrylic fiber plant and a nitrile
elastromer plant, respectively. There was no apparent correlation between air
levels and water levels. No AN was found in the soils and sediments.
The presence of AN in the air and water samples was confirmed by GC/MS.
High resolution selected ion monitoring techniques were employed to achieve
the required sensitivity and selectivity.
In the process of obtaining GC/MS confirmation of AN in air and water
samples, the identities of some other compounds in those samples were estab-
lished. Table 2 summarizes the compounds found in the air and water samples.
-------
TABLE 1. DATA SUMMARY FOR PROGRAM TASK 3
Site
American Cyanamid
New Orleans, LA
American Cyanamid
Linden, NJ
Monsanto
Texas City, TX
Monsanto
Decatur, AL
DuPont
Lugoff, SC
DuPont
Waynesboro , VA
Borg-Warner
Washington, W. VA
Goodrich
Louisville, KY
Monsanto
Addyston, OH
Uniroyal
Painesville, OH
Vistron
Lima , OH
Product
Acrylonitrile
Acrylamide
Acrylonitrile
Acrylic, modacrylic
fibers
Acrylic fibers
Acrylic, modacrylic
fibers
ABS, SAN resins
Nitrile elastomers
ABS, SAN resins
ABS, SAN resins
Nitrile elastomers
Acrylonitrile ,
Acrylamide
Air (/j,g/m3)
High Low
13.6 < 0.1
15.9 <0.1
8.9 <0.3
4.2 <0.1
1.1 <0.1
7.0 <0.2
325 <0.2
4.3 <0.2
1.1 <0.2
3.1 <0.1
148 <0.2
Water (/ig/,0) Soil
High Low (tig/kg)
<0.1 -0.5
0.8 — < 50
0.4 < 0.1 <100
3,500 <0.1 <50
19.7 <0.1 <50
<1.3 <50
1.9 1.4 <50
2.0 <1.4 <400
8.0 <1.4 <400
4,300 9.3 <400
<1.3 <100
Sediment
(Mg/kg)
< 0.5
—
—
<50
<50
<50
—
—
—
—
—
-------
TABLE 2. SUMMARY OF ORGANIC COMPOUNDS FOUND IN AIR AND WATER SAMPLES
Air
Acrylonitrile
Xylenes
Ethylbenzene
Dichlorobenzenes
Toluene
Trimethylbenzenes
Styrene
Water
Acrylonitrile
Acetonitrile
Propionitrile
•H
B <
Cfl hJ
e
Cfl "
!^i CD
0 S
cfl
C
4-> M
(3 3
CO 4-1
CO cfl
C o
O 0)
S 0
X
X
—
X
X
—
—
X
—
X
o
w
**
4-> M-l
C 4-1
O 0
PL, j>0
£3
X
X
—
X
X
—
—
X
—
—
"^
^
ft
O
^i
o
4-1 CO
c v
0 C
PM >-,
3 cfl
O &
X
X
X
X
X
—
—
—
—
—
^
.
^f
tj "
0) C
C 0
1-1 4-1
cfl 00
& C
1 -H
00 £.
-n CO
O cfl
X
X
X
—
X
X
X
X
—
—
s^
C
^
a)
JC* t~H
O -H
•H >
(J CO
13 -rl
0 3
O O
O >-l
X
X
X
X
X
X
—
X
X
—
ffi
O
^
O C
4-1 O
C 4J
Cfl CO
CO >-,
pj PO
X
X
X
X
X
X
—
X
X
—
33
o
„
Q)
rH
rH -H
cfl >
>-. CO
O Cl)
£.S
C cfl
X
X
X
X
X
X
—
X
X
—
M
C O
o
V-i "
4-1 CO
co E
•H -H
> >J
X
X
X
X
X
X
X
—
NA-/
NA
a/ GC/MS analysis not performed.
-------
SECTION II
INTRODUCTION
Acrylonitrile is a widely used chemical intermediate in the manufacture
of acrylic fibers, synthetic rubbers and plastics. Its use in the manufacture
of acrylic fibers and copolymer resins accounts for most of the 1.5 billion
pounds of acrylonitrile produced annually (1976 data) in the United States.
In 1977, interim reports on the toxicity of acrylonitrile were released.—
A study of the chronic toxicity of ingested acrylonitrile to rats showed the
development of various tumors including CNS and ear canal tumors among the
treated animals, but not among the controls. In a later report, OSHA was in-
formed of preliminary results of an epidemiology study demonstrating an excess
of cancer among workers exposed to acrylonitrile at a textile fibers plant.—'
On January 17, 1978, OSHA issued an Emergency Temporary Standard for acrylo-
nitrile, lowering the permissable exposure level from 20 ppm as an 8-hr TWA to
2 ppm with a ceiling level of 10 ppm for any 15-min period during the 8-hr day.—
On July 29, 1976, MRI Project No. 4280-C entitled "Sampling and Analysis
of Selected Toxic Substances" had been initiated. The objective of this
program was to provide the EPA with sampling and analysis capability to deter-
mine the level of toxic substances in air, water, soil, and sediment from
designated sources and ambient locations throughout the United States. The
third and final task on this program, issued August 17, 1977, was the sampling
and analysis for acrylonitrile.
This report describes Task III of the program as follows: Section III,
Sampling and Analysis Protocol; Section IV, Method Development for Sampling
and Analysis; Section V, Site Selection; and Section VI, Discussion of
Results.
-------
SECTION III
SAMPLING AND ANALYSIS PROTOCOL
The approved protocol that was applied to field sampling and analysis
is described in this section. The experimental method development work that
lead to this protocol is described in the following section.
SAMPLING PROTOCOL
Air Sampling
Aerial photographs were obtained by the EPA for most of the recommended
sites. These were inspected by MRI and EPA and used in the selection of
sampling stations around the sites. Approximately six stations were deployed
around a site, with care taken to insure that more than one station was down-
wind. If possible, at least one station was established at the plant fence
line in each of four quadrants surrounding the plant. More than one air
sampler was deployed at some of the stations as part of the quality assurance
program discussed below. The sampling probes were set at 1.5 m above ground
pointing downward and air was sampled continuously for 24 hr at 1 liter/min.
During each test, certain meteorological data were collected. This in-
cluded temperature, relative humidity and general wind behavior. A log of
weather observations was obtained if possible after the test period from the
nearest National Weather Station. A description of any source activity during
the test that could affect the reliability of the samples was noted.
The air sampling train consisted of the following components assembled
in the following order: (a) top charcoal tube; (b) backup charcoal tube;
(c) critical orifice; and (d) 24-v DC vacuum pump.
The charcoal tubes were constructed of glass and were 8 mm OD x 6 mm ID
x 18 cm in length. They were packed with approximately 1.5 g 20 to 50 mesh
Calgon OL charcoal that was activated at 250°C for 1 hr under a stream of
nitrogen. Minimum lengths of natural rubber tubing were used to connect the
sampling tubes to the pump. Flow was maintained at 1 liter/min by using a
hypodermic syringe needle as a critical orifice. Flow rates were checked in
-------
the field every 2 to 3 hr. Power was supplied to the DC pump by two 24-v
lantern batteries connected in parallel. The pump and batteries were encased
in a cardboard-polystyrene shipping box which was then anchored at the sampling
site. The reliability of this system has been demonstrated in several previous
OTS studies..±§7 After sampling was completed, the tubes were capped with rubber
slip-on septa, wrapped in aluminum foil and stored over dry ice for shipment
to MRI. At MRI, they were kept on dry ice until analyzed.
Quality assurance of the AN sampling program was maintained in three ways.
First, blanks were taken to the site, and treated as if they were real samples.
At least four blanks were used at each site. Second, duplicate air samples
were collected at two stations, with at least one of them being downwind of
the suspected source. Finally, spiked sampling tubes were used at two stations,
one upwind and one downwind, to establish the actual recovery of AN under real
conditions.
Water Sampling
If liquid wastes were being discharged into a waterway at an accessible
location, samples were collected at any observable outfall. If no outfall
was evident, e.g., submerged discharge pipe, the samples were collected from
any obvious upwelling. One or two samples were collected downstream of the
discharge point and one sample was collected upstream. For the readily acces-
sible sites, a series of grab samples were composited throughout the air
sampling period. When sites are difficult to reach and require extensive
hiking and/or boating, one or two grab samples were collected. The samples
were collected and composited in 1-gal brown glass bottles with Teflon-lined
caps. The samples were acidified at the site to a pH < 4 by the addition of
sulfuric acid. As the samples were collected, they were placed on ice and
then kept at ice temperature until analyzed.
Soil Sampling
Due to the volatility of AN, it was not expected that a significant
quantity would be present in the soil. As a precaution, however, soil samples
were collected, generally at or near selected air sampling stations. Soil
from the top 12 cm were placed in 16-oz bottles and stored over dry ice until
analyzed.
Sediment Sampling
If possible, sediment samples were collected at each point where water
samples were collected. An Eckman dredge was used for collecting sediment
samples. The samples were kept on dry ice until analyzed.
-------
ANALYSIS PROTOCOL
Air Sample Analysis
Exactly 5 ml of Mallinckrodt reagent grade C$2 was placed in a 7-ml vial
fitted with a Teflon^-faced septum. Each bottle of CS£ was first screened for
the presence of interfering peaks. The charcoal from the sampling tubes was
gently but swiftly added to the CS2 with gentle swirling. After all of the
charcoal was added, the cap was replaced and the vial shaken periodically.
The desorption of AN was allowed to proceed for at least 0.5 hr. If the
sample was not analyzed immediately, it was stored in a freezer until needed.
The AN was analyzed by GC with flame ionization detection using the conditions
listed in Table 3. Confirmation of samples thought to contain AN was made by
GC/MS using the conditions listed in Table 4. The limit of detection for
this protocol was estimated to be 0.3 fg/m although lower detection levels
were achieved. The detection limit is influenced by background interference
plus sample size and can be expected to vary from plant to plant.
The quality assurance samples from the field sampling were analyzed with
the air samples. All the samples were coded in a fashion that did not reveal
the source or nature of that sample. Quality assurance of the analysis was
maintained by (a) analyzing a standard after every fifth sample and (b) forti-
fying samples.
Water Sample Analysis
Two complementary techniques were used for analyses of the water samples:
azeotropic distillation and purge and trap.
Azeotropic Distillation—
The distillation apparatus consists of an insulated glass column, 35 x
2.0 cm, packed with 1/8 in. glass helices, a 1-liter single-necked distillation
flask and heating mantle and a mini-distillation head with a low volume holdup
(Ace Glass, 9322). A 500 ml water sample was added to the distillation flask
along with 25 ml methanol and 5-ml 18 N l^SO^ and then distilled at 1 ml/min.
The first 10 ml of distillate were collected and analyzed for AN using gas
chromatography with a Hall electrolytic conductivity nitrogen selective de-
tector. The chromatographic conditions are given in Table 3. All of the
field samples were treated and analyzed by this technique. The limit of
detection ranged from 0.1 to 1.3 //g/liter. This range was due mainly to a
variability in the response of the Hall detector to AN. Analytical quality
assurance was tested by spiking and analyzing field samples.
Purge and Trap—
Some water samples were also analyzed by the purge and trap technique. A
10-ml sample was transferred to the purging apparatus (Figure 11, pg 45) and
-------
TABLE 3. GAS CHROMATOGRAPHIC CONDITIONS
oo
Column
System (solvent)
Air Samples (CS2)
Water Samples
Azeotrope Distil-
lates (CH3OH)
Purge and Trap
(none)
Soil and Sediment
Samples
Purge and Trap
(None)
Direct Aqueous
Injection
Packing
80/100 mesh
Durapak OPN/
Porasil C
80/100 mesh
Chromosorb
101
80/100 mesh
Chromosorb
101
80/100 mesh
Chromosorb
101
80/100 mesh
Chromosorb
101
Dimensions
9 ft x 1/8 in.
Stainless Steel
6 ft x 1/8 in.
Stainless Steel
6 ft x 1/8 in.
Stainless Steel
6 ft x 1/8 in.
Stainless Steel
6 ft x 1/8 in.
Stainless Steel
Carrier
Helium at
40 ml/min
Helium at
35 ml/min
Helium at
45 ml/min
Helium at
45 ml/min
Helium at
45 ml/min
Temperature
Injector Column Detector
160°C 130°C for 2 min 270°C
15°C/min to 150°C
180°C 100°C for 1 min 850°C
10°C/min to 150°C
a/
160°C- 80°C for 5 min 850°C
10°C/min to 150°C
160°C- 80°C for 5 min 850°C
10°C/min to 150°C
180°C 100°C for 1 min 850°C
10°C/min to 150°C
Detector Type
Flame ionization
Hall electrolytic
conductivity
Hall electrolytic
conductivity
Hall electrolytic
conductivity
Hall electrolytic
conductivity
a/ Thermal desorption temperature.
-------
TABLE 4. OPERATING PARAMETERS FOR GC/MS ANALYSIS OF ACRYLONITRILK SAMPLES
Atr samples
High resolution, SIM mode
Full scan mode
Gas chromatograpli
Type:
Column:
Carrier:
Temperature:
Varian 2700
6 ft x 2 mm, 80/100 mesh,
Durapak OPN/Porastl C
Helium at 30 ml/mln
Injector, 150°C
Column, 135°C
Varian 2700
6 ft x 2 mm, 80/100 mesh,
Durapak OPN/Porasil C
Helium at 30 ml/mln
Injector, 150°C
Column temperature pro-
gram - 80°C for 1 mln,
then 8°C/mtn to 150°C
Water samp1es (Azeotroplc distillation)
Full scan mode
Varian 2700
6 ft x 2 mm, 80/100 mesh,
Cromasorb 101
Helium at 30 ml/mln
Injector, 150°C
Column temperature program -
100°C for 1 min, then 10°C/
mln to 150°C
Low resolution SIM Mode
Varian 2700
6 ft x 2 mm, 80/100 mesh,
Cromasorb 101
Heltum at 30 ml/m1n
Injector, 150°C
Column temperature program -
100°C for 1 roln, then 10°C/
mln to 150°C
Mass Spectrometer
Type:
Data system:
Separator:
Masses monitored:
Resolution:
Tonization
potential:
Multiplier gain:
Varian 311A
Varian Spectrosystem
100 MS
Watson-Blemann
52.0187, 53.0265
4,000
70 ev
It x 106 - 2 x 107
Vnrtan 311A
Varian Spectrosystem
100 MS
Watson-Biemann
20-250
1,000
70 ev
4 x 106 - 2
10'
Varian 311A
Varian Spectrosystem 100 MS
Watson-Blemann
20-250
1,000
70 ev
4 x 106 - 2 x 107
Varian 311A
Varian Spectrosystem 100 MS
Watson-Biemann
53, 52
1,000
70 ev
4 x 106 - 2 x 107
-------
analyzed according to the conditions given in Table 5. A limit of detection
of 0.1 fig/liter had been established using clean water.
Samples which appeared to contain AN by either azeotropic or purge and
trap analysis were confirmed by GC/MS using the azeotropic distillation
method. The GC/MS operation conditions are given in Table 4.
Soil Samples
The soil samples were analyzed using a modification of the purge and trap
technique, although no method development work was done. An untreated 10 g
soil sample was extracted with 50 ml of water using ultrasonic agitation. The
sample was centrifuged and the supernatant withdrawn and filtered. Some of
the early samples were analyzed by applying the purge and trap technique to
10 ml of the supernatant. Most of the samples were analyzed of direct
aqueous injection of the supernatant. The analysis conditions are given in
Table 3. The limit of detection using direct aqueous injection/Hall detection
ranged from 50 to 400 ^g/kg. The range was due both to a variation in the
Hall detector response and to variations in the background. When purge and
trap was applied to the extract, a limit of detection of ~0.5 fig/kg was
achieved.
Sediment Samples
The sediment analysis protocol was similar to the soil analysis protocol.
Each sediment sample was first centrifuged to separate water which was removed.
A 10-g portion of the sample was then extracted with 50 ml of water and the
extract analyzed by the conditions given in Table 3. For the limited sediment
samples processed, the limit of detection was 50 /^g/kg.
10
-------
TABLE 5.. OPERATING CONDITIONS FOR PURGE AND TRAP ANALYSES OF ACRYLONITRILE
Purge:
Gas: Helium at 20 ml/min
Purge Time: 30 min
Purge Temperature: 85 °C
Sample Size: 10 ml
Desorption:
Trap Material: 0.3 g 50/80 mesh Porapak N
Carrier 1: Helium at 45 ml/min
Desorption Temperature: RFL Controller at 85 = 194°C peak temperature
Desorption Time: 5 min
Gas Chromatograph:
Carrier 2: Helium at 45 ml/min
Column: 6 ft x 1/8 in. Stainless steel with 80/100 mesh Chromosorb 101
Column Temperature: 80°C for 3 min then 8°C/min to 150°C
Transfer Line: 200°C
*.
Detector:
Hydrogen Flow: 60 ml/min
Solvent: 15% n-propanol in water at 0.5 ml/min
Pyrolysis Oven: 850°C
Mode: Catalytic reduction using nickel wire
11
-------
SECTION IV
METHOD DEVELOPMENT FOR SAMPLING AND ANALYSIS
LITERATURE STUDY
The initial action taken on this task was a computerized literature search
of Chemical Abstracts 1970 to 1977 as well as NTIS. The search was keyed on
acrylonitrile and analysis or variants of analysis. The relevant literature
was obtained and is summarized below.
Measurement of Acrylonitrile
The measurement techniques reported for the analysis of acrylonitrile
were reviewed first without regard to the source of the sample, e.g., air,
water, etc. The useful techniques were divided into instrumental and chemical
categories. The instrumental techniques have utilized gas chromatography,
polarography and infrared spectroscopy. The chemical techniques have been
based upon the chemical reactivities of the two functional groups. The reac-
tions which have proved useful are (a) hydrolysis of the nitrile group and
(b) additions to the double bond. All these procedures are reviewed in
greater detail below.
Gas Chromatography—
The reported procedures using gas chromatography have, at least in recent
years, relied on flame ionization detection. An attempt was made by Barrett—
to use electron capture detection; however, the sensitivity was reported to be
one-fifth of that obtained with flame ionization detection. No absolute de-
tection limits were reported. The column packings have tended to be of the
porous polymer variety. The advantage of these packings is in their ability
to resolve acrylonitrile from methanol and to be used for direct aqueous injec-
tion. The reported systems are summarized in Table 6. All of the procedures
utilized flame ionization detection.
Polarography—
The polarographic determination of acrylonitrile was first reported by
Bird and Hale.—' The method was later coupled with an azeotropic distillation
to separate acrylonitrile from other components.—i-i' The analysis was done at
-1.95 volts versus SCE using tetramethylammonium iodide supporting electrolyte.
12
-------
TABLE 6. GAS CHROMATOGRAPHIC CONDITIONS
Packing
Conditions
Comments
Reference
Chromosorb 104, 60/80
mesh
Tenax
0.4% Carbowax 1500 on
Carbopak A
Porapak !Q, 50/80 mesh
Porapak Q, 50/80 mesh
Porapak N, 50/80 mesh
Chromosorb 101, 50/60
Porapak Q, 50/80 mesh
a/
120 C, 84 cc/min N2r
6.5 min,
80°C, 15 cc/min N2, -,
2 ft x 1/8 in. Teflon
100 C, 30 cc/min He, -,,
6 ft x 1/8 in. stain-
less steel
155°C, 50 cc/min N2, -,
4 ft x 1/4 in. stain-
less steel
160°C, 30 cc/min N2,
3.2 min, 5 ft x 1/8 in.
stainless steel
170°C, 40 cc/min N2,
10.5 min, 9 ft x 1/8
in. stainless steel
1106C to 200° C at 18eC/
min He,-, 8 ft x 1/8 in.
stainless steel
156°C, 50 cc/min He,
11.8 min, 12 ft x 1/8
in. stainless steel
Used by American Cyanamid
for air sample analysis
Used by American Cyanamid
for water analysis
Head space analysis of
residual monomer
NIOSH method for acrylo-
nitrile in air
Poor resolution from
methanol
Resolved from methanol
ASTM tentative method for
nitriles in water
Used with a trapping col-
umn for combustion efflu-
ents
(9)
(9)
Steicher (10)
(11)
Barrett (2)
Barrett (2)
ASTM D337K12)
Bellar (13)
-------
TABLE 6. (concluded)
Packing Conditions Comments Reference
Carbowax 20M on 60/80 80°C, 30 cc/min N2, Early method for (6)
Chromosorb W 4 min, 6 ft x 1/8 in. acrylonitrile in air
stainless steel
207. Carbowax 400 in 70°C, 100 cc/min He, Coelutes with meth- Lysyj (14)
30/60 Chromosorb W 8 min, 12 ft x 1/4 acrylonitrile
in. aluminum
a] Column temperature,,carrier gas and flow rate, retention time, column parameters.
-------
Although a detection limit of 0.1 ppm was reported,— no mention was made of
the accuracy, precision or recovery at this level.
Infrared Spectroscopy—
The determination of acrylonitrile by IR has been reported using a 10-m
gas cell at 20*C and 760 mm Hg.A' At a wavelength of 10.49 f/m, the limit of
detection is apparently about 0.5 ppm (v/v).
Hydrolysis of the Nitrile Group—
In strong base, acrylonitrile will be hydrolyzed to ammonia and the
acrylate anion. The resulting ammonia has been determined by the Nessler
method. ' r The procedure has a detection limit of about 20 ppm.
Reactions of the Alkene Group—
Both colorimetric and titrimetric procedures have been developed based
upon a reaction of the alkene group. The colorimetric procedures utilize
photochemical bromination which results in the formation of cyanogen bromide.
The cyanogen bromide reacts with a benzidene-pyridine solution to produce
species absorbing at 535 nm.— The limit of detection is about 0.25 ppm.
The well-known titrimetric procedures follow the reaction of acryloni-
15/
trile with excess dodecanethiol.—' The excess thiol is then measured by
iodometric titration with 12 or amperometric titration with silver. The pro-
cedure works best for 2 to 200 mg acrylonitrile.
In '.summary, the most promising measurement technique, worthy of further
investigation, was judged to be gas chromatography. The use of nitrogen
specific detectors was not found in the literature and was selected for
investigation.
Analysis of Air Samples
Only a limited number of methods for sampling (and analysis) of acrylo-
nitrile in air have been reported. These are summarized in Table 7 indicating
the sampling or trapping technique, the measurement technique and the reported
limit of detection. The measurement techniques were discussed previously.
None of the procedures had the necessary sensitivity as they existed. The
most promising trapping medium appeared to be charcoal with methanol as the
eluate. This approach was selected as the starting point for additional
studies.
Analysis of Water Samples
A summary of methods for the anlaysis of acrylonitrile in water is given
in Table 8. None of the procedures had the necessary sensitivity. The purge
and trap technique could have the required sensitivity; however, the efficiency
15
-------
TABLE 7. ANALYSIS OF ACRYLONITRILE IN AIR
CT-
Trapping
medium
Charcoal
Water
Charcoal
Carbo sieve B
(150, 50 mg)
Water (3 ml)
1% H2S04
-
2 ml H^SO,
Sampling
condition
Unreported
Unreported
0.2 (pm, 20 8.
maximum
1 j?Pm, 10-20 j?
Unreported
Unreported
1 ml air sample
0,4 0pm, volume
Measurement
technique
Gas Chromatography
Hydrolysis, Color-
imetry
Gas Chromatography
Gas Chromatography
Bromination,
Colo rime try
Hydrolysis, Color-
imetry
Gas Chromatography
Hydrolysis, color-
Limit of
detection Comments
Not reported Used for OSHA
compliance
500 ^g/m
Range 5-135 mg/m
3
Range 40-750 mg/m
400 ii>g/m
Unreported
0.5 ppm
Unreported
Reference
(9)
Rogaczewska (0)
OSHA Method
5156 (11)
Barrett (2)
Russkikh (16)
Aarato (7)
(6)
(6)
5 jC aix sample
imetry
Infrared Spectro- ; 0.5 ppm
scopy
(6)
-------
TABLE 8. ' ANALYSIS OF ACRYLONITRILE IN WATER
Sample
treatment
Measurement
technique
Limit of
detection
Comments
References
Direct aqueous injection Gas Chromatography
0.1 mg/^
Purge and trap
Extraction to remove
cations
Gas Chromatography/
Mass Spectrometry
Polarography
3.5
Direct aqueous injection Gas Chromatography/FID 1
Titrimetric 0.5 mg/|
Azeotropic distillation Hydrolysis, Color- 2
with methane 1 imetry
Azeotropic distillation
with methanol
Polarography
0.1 mg/j
Amer Cyan (9)
Purging and trapping effi-
ciencies unknown
Industrial waste water
(17)
Ponomarev (18)
General nitrile procedure ASTM D3371 (12)
Surface water Wronski (19)
Correction for aceto-
nitrile necessary
Recovery not reported
at low levels
Stefanescu (20)
- Daues (4)
-------
of the procedure had not been established nor had the optimum conditions been
well-defined. This approach was believed to warrent further investigation.
Another possible approach selected for study was combination of preconcentra-
tion by azeotropic distillation with quantitation by gas chromatography using
a nitrogen selective detector.
GAS CHROMATOGRAPHIC STUDIES
The initial experimental work was directed toward developing gas chromato-
graphic procedures that could be used to evaluate sampling methods and hope-
fully could be used for real sample analysis. The selection of systems for
study had to consider the probable solvent, the compatibility of detectors
with that solvent and the column that would resolve AN from the solvent as
well as compatibility with that solvent. The probable solvents and their roles
were thought to be (a) carbon disulfide (charcoal desorbent), (b) methanol
(charcoal desorbent, azeotropic distillate), and (c) water (direct aqueous in-
jection). The available detectors were flame ionization (FID) and Hall elec-
trolytic conductivity (Hall). Carbon disulfide was compatible with only the
FID while methanol and water was compatible with either detector. The following
studies were directed toward selecting a column and evaluating the total system.
The purge and trap studies are included here although that procedure did not
involve a solvent.
Carbon Disulfide/FID
Carbon disulfide has frequently been used as a desorption solvent for
air samples collected on charcoal, etc. Columns packed with Chromosorb 104
and Carbowax 1500/Carbopak C were tested with a 10-ppm solution of AN in CS2•
It was not possible to find conditions at which the AN would be sufficiently
separated from the solvent for accurate analysis at concentrations below 10
ppm. In an earlier task for the OTS, a Durapak OPN/Porasil C column had been
used for the analysis of vinylidene chloride in carbon disulfide. This column
was able to adequately separate AN from CS2 using the conditions listed in
Table 10.
A chromatogram of AN in CS- was found to contain an impurity, noted in
Figure 1, which eluted with a relative retention time of 0.96 when compared
to AN. This impurity does not seriously interfere as relative standard
deviations (n = 6) of 1.9 and 2.2% at 1 and 100 ppm, respectively, were
achieved. Also a calibration curve from 1.0 to 460 ng AN, shown in Figure 2,
was quite linear.
The chromatographic behavior of 26 compounds which could be considered
as potential interferences in the analysis of AN in air was established. The
compounds chosen were ones that (a) have been found in previous industrial air
monitoring studies, (b) have been reported in urban or rural air at concentra-
tions greater than 1 ppb (v/v), or (c) are related to acrylonitrile production
18
-------
I I
4 6
Minutes
10
Figure 1. Chromatogram of 15 ng Acrylonitrile in Carbon Bisulfide
on Durapak OPN
19
-------
lOOOi-
800
600
JO)
"53
400
200
w
I
I
I
I
I
25 50
Monograms Acrylonitrile
I
I
J
0 100 200 300 400 500
Na nog rams Acrylonitrile
Figure 2. Calibration Curves of Acrylonitrile in Carbon Bisulfide
20
-------
or use. A chromatogratn of 24 of the 27 compounds in CS~ is shown in Figure 3.
The GC conditions were modified from Table 3 and are noted in Table 9 along
with the retention times of the 27 compounds. Also noted in Table 9 are re-
ported air concentrations for some of the compounds.
AN was well-resolved from the commonly occurring hydrocarbons such as
benzene, toluene, xylenes, etc., as well as the related compounds such as
acetonitrile, acrolein and styrene. The results did not suggest that any
serious interference problems are likely.
Methanol/Hall
Methanol was used as a desorbent solvent for AN trapped on charcoal in
the NIOSH procedure (11). It has also been used in the isolation of AN from
water by azeotropic distillation. The literature study revealed that a Porapak
N column provided reasonable resolution of AN from methanol. This column was
used at the conditions listed in Table 10 in conjunction with the Hall electro-
lytic conductivity detector operated in the nitrogen mode. This system
allowed a highly selective measurement of < 0.25 ng AN.
When the Porapak N column was used with a FID, the strong methanol re-
sponse limited the sensitivity to about 10 ppm.
Water/FID and Hall
Acrylonitrile in water can be analyzed by direct aqueous injection gas
chromatography (DAIGC) using well-established conditions (12). This technique,
however, is limited in sensitivity to about 1 ppm with FID. A DAIGC system
was established for the purpose of studying the stability of AN under a
variety of conditions in various types of water. The chromatographic condi-
tions are given in Table 10. The DAIGC system also is compatible with the
Hall detector. This combination was used for analysis of aqueous extracts
of soils and sediments.
Purge and Trap/FID and Hall
Initially a 0.2% Carbowax 1500/Carbopak'C column was tested for its
chromatographic properties. Acrylonitrile eluted quite rapidly even at low
column temperatures and would not be expected to be adequately resolved from
early eluting impurities. Chromosorb 102 was briefly studied and eliminated
for the same reasons. An established GC system using Chromosorb 101 was found
to work satisfactorily for the purge and trap experiment and is described in
Table 10.
21
-------
U
\
41
•C
2
2
4
3
10 12 14 16 18 20 22 24 26 28 30
Minutes
Figure 3. Chromatogram of acrylonitrile and 26 potential interferences
22
-------
TABLE 9. CHROMATOGRAPHIC BEHAVIOR OF POTENTIAL INTERFERING COMPOUNDS
Compound
Hexane
Methylene chloride
Carbon tetrachloride
Chloroform
Benzene
Tetrachloroethylene
1 , 2-Dichloroethane
Acetaldehyde
Octane
1 , 2-Dichloropropane
Toluene
1,1, 2-Trichloroethane
Acrolein
Acrylonitrile
Decane
Xylene isomer
Ethyl benzene
Acetonitrile
Xylene isomer
Styrene
m-Dichlorobenzene
Xylene isomer
1,1,2, 2-Tetrachloroethane
Propionitrile
Benzonitrile
Dibutyl Phthalate
Acrylic acid
R ^ '
11.0
11.1
12.3
13.5
15.8
16.1
16.8
16.9
17.9
18.8
19.9
20.4
20.4
21.8
23.5
23.5
23.5
24.5
24.5
24.9
25.6
25.6
26.8
> 30
> 30
> 30
> 30
fcR
tR, AN
0.50
0.51
0.56
0.62
0.72
0.74
0.77
0.78
0.80
0.86
0.91
0.94
0.94
1.00
1.08
1.08
1.08
1.12
1.12
1.14
1.17
1.17
1.23
> 1.4
> 1.4
> 1.4
> 1.4
Typical Air Concentrations
ppb (v/v)
0.11
0.02
1.3-54
0.6-10 , 0.04
0.7-3.4
0.3-39
0.14
0.2-2.7
1.9-8.7
0.7-40, all isomers
a/Analyzed using the following chromatographic conditions:oven temperature, 60°C for 30 min; 6°C/min to
150°C; hold 12 min; all other conditions as before. See Table 3 for normal conditions.
-------
TABLE 10. GAS CHROMATOGRAPHIC CONDITIONS FOR ACRYLONITRILE ANALYSIS
GC system
Column
Conditions
Carrier
Retention time
Carbon disulfide/
flame ionization
detection
Methanol/Hall
conductivity
Water/flame ion-
ization detection
and Hall
Purge and trap/
flame ionization de-
tection and Hall
9 ft x 1/8 in. stainless
steel packed with 80/100
mesh Durapak OPN/Porasil C
6 ft x 2 mm glass packed
with 50/80 mesh Porapak N
6 ft x 1/8 in. stainless
steel packed with 80/100
mesh Chromosorb 101
6 ft x 1/8 in. stainless
steel packed with 80/100
mesh Chromosorb 101
130°C for 2 min
15°C/min to 150°C
and hold 5 min
150°C
160°C
Helium at 40
ml/min
Helium at 35
ml/min
Nitrogen at
40 ml/min
120°C for 2 min Nitrogen at
10°C/min to 150°C 45 ml/min
7.0 min
5.5 min
2 min
6.0 min
-------
AIR SAMPLING
Survey of Potential Adsorbents
The development of an air sampling protocol was directed toward the use
of a solid adsorbent trap. Various adsorbent-desorbent solvent combinations
were screened in the following manner. Various adsorbents (—0.5 g) were added
directly to a 10-ppm solution of AN in methanol. The level of AN was period-
ically monitored to establish if any of the materials would adsorb AN from the
solvent. If an adsorbent would actually remove AN from a particular solvent,
then that solvent could not be expected to be useful for desorption and was
eliminated. The results are summarized in Table 11. The loss of AN was negli-
gible with the porous polymers Porapak N and Chromosorb 104, but was of concern
with the charcoals. Next, 20 ^g of AN in 1 //I hexane was added directly to
—0.5 g of the adsorbents. Two milliliters of methanol were added and the
vial was sonicated for 5 min. The recoveries are given in Table 12. The
porous polymers gave acceptable recoveries; the charcoals did not. While
these results favor the polymers, it was felt that they would not be likely
to have a high enough breakthrough volume for 24-hr samplings. In fact, a
breakthrough volume of 35 liters has been reported for Porapak N in a 4 in. x
1/4 in. tube.^i-' Consequently, additional studies with the charcoal adsorbents
were conducted. A study was made with Fisher charcoal to determine if the
percent recovery was related to the amount of AN added.
Vials containing 0.5 g charcoal were spiked with 2, 20 and 200 Mg AN and
extracted with methanol or CS2 as before. Figure 4, Curve 4, shows the depend-
ency of recovery by methanol extraction on loading. This suggested that the
charcoal has some active sites and that the methanol was unable to cause desorp-
tion from these sites. Since as little as 20 Mg AN may be all that is collected
in a 24-hr sample, this system was unacceptable.
The recoveries with CS2 were much less dependent on loading level and
appeared to be best for Calgon OL charcoal (Figure 4, Curve 3). Consequently,
Calgon OL was selected for further testing as an air sampling adsorbent.
Air Sampling Studies with Calgon OL
The air sampling studies were made using 8 mm OD x 18 cm length glass
sampling tubes. These were packed with 1.5 g charcoal, plugged with glass
wool, and capped with slip-on septa when not in use. A dimpled inlet tube was
attached to the charcoal tube which allowed the addition of known quantities
of AN into the air stream. It was initially established that at a flow rate
of 1 liter/min, all of the AN evaporated and moved into the sampling tube
within 30 min. No AN was ever found in the inlet tube. The desorption of AN
from the charcoal was accomplished by transferring the inlet glass wool and
charcoal to a suitable vial, adding 5 ml CS2, capping with a Teflon®-lined
septum and sonicating for 5 min.
25
-------
TABLE 11. SURVEY OF POTENTIAL ADSORBENTS USING METHANOL
Adsorbent
Calgon charcoal PCB 20X50
Calgon charcoal OL
Fisher coconut charcoal
Porapak N 50/80 mesh
Chromosorb 104 50/80 mesh
0.25 hr
91
87
84
94
96
% recovery
0.75 hr
83
84
83
91
93
3.0 hr
80
76
81
91
94
TABLE 12. ' RECOVERY OF AN FROM SPIKED ADSORBENTS USING METHANOL
Adsorbent
% recovery
Calgon PCB
Calgon OL
Fisher coconut
Porapak N
Chromosorb 104
52
48
60
93
83
26
-------
100
80
60
o
o
&
+*
9}
g 40
20
O Fisher Coconut,
• Fisher Coconut, C$2
A Calgon OL, C$2
• Calgon PCB, C$2
i i i i i i i I
i i i i i i
I I I I I I I I I
10 100
Acrylonitrile Added, Micrograms
1000
Figure 4. Recovery of Acrylonitrile Versus Loading
27
-------
Preliminary studies indicated that when large volumes of air were pulled
through the trap (M..5 nr), the recovery was only-^60%. The air being pulled
through the traps was laboratory air, pre-filtered through charcoal with a
relative humidity of about 50% at 72°F. The humidity of the air was suspected
as being important in affecting the recoveries. The air sampling system was
modified to add H20 to a purified stream of dry tank air. The humidity was
monitored on stream by a Hygrodynamics humidity sensing element and could be
changed from 0% to >95%. The apparatus is shown in Figure 5. The experiments
were scaled down from a 17 x 0.8 cm OD tube containing 1.5 g charcoal to a
6.5 x 0.6 cm OD tube containing 0.2 g charcoal. The maximum volume of air was
reduced to 150 liters although the sampling rate was maintained at 1 liter/min.
A 150 liter air sample in the reduced system corresponds to a normalized sample
volume of 750 liters/g. This is comparable to the 640 liters/g sample volume
obtained for a 16 hr sample on the normal system.
The effect of humidity on total recovery was tested on tubes that were
spiked with 25 /xg AN and then 10 liter or 150 liters of air at various R.H.'s
was drawn through them. The results are shown in Figure 6. The recovery for
a 10 liter air sample is reasonable and does not appear to be unduly affected
by humidity. When the air volume was increased to 150 liters, however, the
recovery decreased markedly with increasing humidity. This was shown to be a
displacement phenomenon by segmenting the charcoal into four sections during
the 150 liter, 96% R.H. run. The average level found in each of the sections
for duplicate runs was 7, 8, 9 and 17% starting from the front of the tube.
In other words, the AN had migrated from the front section into the next
three sections and probably beyond. The low recovery was then shown not to be
a problem of short term stability by the following study. A series of tubes
were prepared at 50% and 96% R.H. using 10 liters of air and either analyzed
immediately or stored 24 hr and then analyzed. The recoveries for samples
prepared at both humidities were completely independent of storage time and
all above 90%.
For the tests conducted with 150 liters of air, the recovery at 50% R.H.
was from 61 to 65% which was marginally acceptable. Two approaches were used
in attempting to increase the recovery. The first approach was to simply add
a backup tube of the same dimensions to be analyzed separately. A recovery
of 88% was achieved with about 30% of the total AN being on the second tube.
The other approach was to dry the gas air stream with a dessicant before it
reached the charcoal. A drying tube, 15 cm x 1.2 cm ID, filled with 7.2 g
Drierite that had been activated at 200°C, was placed ahead of the charcoal
tube. Air at 50% R.H. was sampled through the tubes at 1 liter/min and the
R.H. of the effluent was continuously monitored. At the end of 1 and 2 hr,
the humidity had increased to only 11% and 43%, respectively. By changing
the drying tube every 2 hr, it should be possible to sample in a high humidity
environment.
28
-------
R.H. Sensor
Control Valves
V
Flowmeter •
•o
Gas Dispersion
Bottle
Air
(Charcoal Filtered)
Conditioned Air ~ 5 Ipm
t
Charcoal
Acrylonitrile in Hexane
Vacuum Pump
APPARATUS FOR EVALUATING AIR SAMPLING TRAIN
Figure 5.
-------
100
80
b
51 60
8
40
20
I I I 1
I I
I I
10 Liters
20 40 60
Relative Humidity
80
100
Figure 6. Effect of Humidity on Recovery of Acrylonitrile
30
-------
The concept was first tested at 57% humidity using the above drying tube
and one charcoal tube. The AN in hexane was added to the inlet end of the
Drierite drying tube. After sampling air at 1 liter/min for 75 min, the
charcoal tube was analyzed and found to contain 76% of the added AN. Thus
the AN appears to be unretained by the Drierite.
It appears that an acceptable level of recovery can be obtained at a 50%
R.H. level and that the recovery could be improved by use of a drying tube.
It also seemed likely that at high R.H., e.g., 96%, recovery will be poor.
This may not actually be the case if the absolute humidity rather than the
relative humidity is considered. The absolute humidities at 22.2*C (728F)
are 2.0, 11.8 and 18.8 mg/liter for 11, 56 and 90% relative humidities, re-
spectively. In terms of absolute humidity, a level of -^12 mg/liter appears
to be the highest tolerable level.
The 12 mg/liter level of humidity would correspond to higher relative
humidities with decreasing temperatures. Figure 7 shows the 12 mg/liter line
as a function of temperature versus relative humidity and is generally divided
into acceptable and unacceptable regions. By considering the long term
weather observations at or around some of the sites, it was shown how often
the unacceptable conditions will be encountered. Table 13 was prepared from
10-year weather data observations for Lake Charles, Louisiana and Evansville,
Indiana, for the month of January. The data are arranged to show the fre-
quency of observations as a function of temperature ranges and relative
humidity ranges. Shown parenthetically is the maximum absolute humidity that
could occur within each category, i.e., highest temperature and highest R.H.
The only cases where absolute humidity will be equal to or exceed ^ 12 mg/liter
for the situations defined by Table 13 are when the temperature is above 80°F
and the relative humidity is above 50%. At a location such as Evansville,
Indiana, those conditions only occurred during 9.5% of the observations. In
other words, 90% of the time, the absolute humidity level will be below ~12
mg/liter. This was expected to be the case for the non-coastal sites. At
Lake Charles, Louisiana, however, the absolute humidity will exceed -^12 mg/
liter about 50% of the time.
Since most of the humidity experiments were conducted with tubes con-
taining 0.2 g charcoal, additional recovery experiments were run at 70%
R.H., 72°F (14 mg/liter) with the 1.5 g tubes that will be used in the field.
The possibility of lowering the humidity level by use of a Drierite drying tube
was also part of the experiment. Consequently, the experiment was made using
four sampling trains consisting of a front and a back tube (17 cm x 0.8 cm OD)
with 1.5 g charcoal per tube. One train was spiked at the front tube inlet
with 50 //g AN. Two trains were modified by placement of a 1.2 cm OD x 1.0 cm
ID x 15 cm length glass drying tube containing 7.2 g Drierite ahead of the
31
-------
LL.
O
£
o
0)
Q.
J
IUU
80
60
40
20
n
12mg/e ^^•>v^
Absolute Humidty
—
1 1 1 1 1
1 1 1 1
Unacceptable
^~ •
—
Acceptable
1 1 1 1
100
Relative Humidity
Figure 7. Relationship of temperature and relative humidity at 12 mg/liter
absolute humidity
32
-------
TABLE 13. HUMIDITY-TEMPERATURE'OBSERVATIONS
Relative Humidity
Temperature °F 0-50% 50-80% 80-90% 90-100%
< 50°F 3.5 (4.7)-/ 15.6 (7.5) 8.8 (8.5) 9.0 (9.4)
Lake Charles, Louisiana
(4.7)^ 15.6 (7.5)
50-80°F 7.4 (12.6) 16.9 (20.2) 9.4 (22.8) 26 (25.3)
Evansville, Indiana
< 50°F
50-80°F
90.5% (12.
J i s (?n
6)^
2) 2.8 (22.8)
1
3.2 (25.3)
a/ Percent of occurance (Maximum mg/liter of H20).
b_/ The 90.5% observation refers to all conditions within the box. This was
not subdivided since these are all the conditions where the humidity
level is acceptable.
33
-------
charcoal tubes. These trains were spiked at the inlet of the drying tube with
50 /ig AN. The fourth train was used as a blank and was not spiked. Air was
conditioned at 70% R.H., 72°F (14 rag/liter H20) and pulled through the train
at 0.6 liter/min for 12 hr. The Drierite tubes were changed every 2 hr through-
out the sampling period. The tubes were then desorbed with CS2 and analyzed
for AN. The results are summarized in Table 14. Surprisingly, there was no
large difference in the recovery level for the trains with or without drying
tubes. Unfortunately, a smaller than desired volume of air was pulled through
the tubes. The duration of experiment, 12 hr, was limited by the need for peri-
odic changing of the drying tubes. The study was then repeated in quadrupli-
cate without drying tubes but for a 24-hr period. The sampling rate was
1 liter/min, and the R.H. was set at 70% (72*F). The results are also reported
in Table 14. The total recovery dropped to about 40% with most of the AN on
the backup tube.
TABLE 14. RECOVERY OF AN FROM CALGON OL CHARCOAL
Volume of
air
432
432
432
1,200
1,200
1,200
1,200
Weight of
charcoal
1.4
1.4
1.4
1.4 .
1.4
1.4
1.4
Liters/
gram
308
308
308
860
860
860
860
Drying Relative humidity %
tube at 72°F
Yes
Yes
No
No
No
No
No
70%
70%
70%
70%
70%
70%
70%
Front
51
69
63
1
5
ND
4
Recovery
Back
ND
ND
ND
101
36
40
43
Total
51
69
63
102JI/
41
40
47
aj Rejected as anomolous.
As the results in Table 14 show, the Drierite drying tubes did not
significantly improve recovery at relatively high humidities (>14 mg/liter).
Consequently, they were not used in the field sampling. Also, the results
again showed that recovery drops considerably when the sampling volume -is
large and the humidity is high.
The results of most of the air sampling experiments have been summarized
graphically in Figure 8. The purpose of the figure is to show the effect
on recovery of AN of increasing sample volume at various humidity levels.
The sample volume scale has been normalized to liters of air per gram of
charcoal to permit the use of data collected with both" 0.2 and 1.4 g traps.
34
-------
100
80
60
o
-------
A second scale shows the volume of air sampled using the 1.4 g traps. The
results show quite clearly the deleterious effect of a humidity level of
>12 mg/liter (60% R.H. at 72°F). The results also show, however, that the
recovery is at an acceptable level at humidities up to about 12 mg/liter.
The average recovery of AN for the seven experiments made with air volumes
of 750 liters/g or greater and at relative humidities up to 60% was 75%. This
data is represented by the top two curves in Figure 8. The standard deviation
of these seven measurements was 9.4%.
Stability of Acrylonitrile Solutions in Carbon Bisulfide
A series of six solutions of AN in CS2 were prepared at 10 ppm and placed
in glass vials. Three were stored in a refrigerator at 3°C and three were
stored in a freezer at -17*C. The samples were analyzed by GC/FID at 0, 4, 6
and 12 days after preparation. These results of the analysis are summarized
in Table 15 in terms of recovery versus a freshly prepared external standard.
The results obviously showed that (a) storage at 3°C is identical to storage
at -17 *C and that (b) the solutions are stable for at least 12 days at those
temperatures.
TABLE 15. STABILITY OF ACRYLONITRILE IN CARBON BISULFIDE
% Recovery
Sample Storage Temperature 0 Days 4 Days 6 Days 12 Days
1
2
3
4
5
6
3°C
3°C
3°C
-17°C
-17°C
-17°C
105
104
104
103
101
102
102
90
99
99
99
99
103
108
109
112
108
113
102
100
100
102
93
103
Stability of AN on Charcoal
A series of charcoal tubes was spiked by injecting 50 /^g of AN in hexane
into the middle of the tubes. The tubes were 8 mm OD x 6 mm ID x 18 cm in
length and contained 1.5 g charcoal. The 24 tubes were capped with silicon
rubber septa caps, wrapped in foil and divided into two equal groups. Half
of the tubes were stored in a freezer at -17*C and half were placed on dry
ice at -78*C. Three tubes were randomly selected and analyzed immediately.
Thereafter, three tubes each from each set were analyzed on days 1, 4 and 24.
36
-------
The tubes were desorbed and analyzed by the prescribed protocol (Section III)
and the results are given in Table 16. Stability was observed for up to 8 days
for tubes stored at -17°C or -78°C. The two remaining tubes which were being
stored at -78*C were desorbed and analyzed on day 24 and recoveries of 71 and
100% were found. It appears that the tubes can be considered to be stable for
up to 24 days when stored at -78CC.
TABLE 16. RECOVERY OF ACRYLONITRILE ON CHARCOAL
% Recovery^./
Storage time Storage temperature
(days) (-17°C) (-78°C)
0 104, 104, 118^
1 83, 91, 94
4 114, 97, 93
8 105, 105, 105
24
87,
100,
107,
71,
87
94,
103
100
110
, 105
a/ Spiked with 50 yg.
b_/ Analyzed immediately without storage.
Gas Chromatography/Mass Spectrometry Confirmation
The confirmation of AN in CS£ proved to be an extremely difficult prob-
lem, due to the intense background at the two most intense ions monitored in
the low resolution mode. A full scan GC/MS approach was tried initially but
found not to have the necessary sensitivity. The sensitivity of the tech-
nique was limited to -^15 ng by the background contribution from the column.
By operating the instrument in a high resolution, selected ion monitoring mode
(SIM), it was possible to monitor AN in the presence of the background, thus,
lowering the detection limit to < 1 ng. The C3H3N+ and C3H2N"1" ions with m/e
values of 53.0265 and 52.0187, respectively, were monitored.
The background, presumably from the column, was due to C^E^+ and C4H,+
ions with m/e values of 53.0391 and 52.0313, respectively. Figure 9 shows
the actual high and low resolution mass spectra at m/e values of 52 and 53.
The final operating conditions are given in Table 4.
37
-------
C3H2N+
52.0187
M
AM
C4H7+
52.0301
= 4,000
C3H3N+
53.0265
53.0391
M
AM
= 1,000
Figure 9. High and low resolution mass spectra of acrylonitrile and
background
38
-------
Limit of Detection in Air Samples
In order to demonstrate the general limit of detection of AN in air, a
series of spiked blanks were prepared. The C$2 extract from a blank charcoal
tube used at the Uniroyal site was used. Known quantities of AN (1.4, 2.8 and
5.3 jtg) were added to 1 ml of the extract. After each addition of AN, the ex-
tract was analyzed for AN by injecting ~5 /il of the solution. The limit of
detection of 0.5 ng per injection was estimated from the curve of detector
response versus ng of AN.
With an injection volume of 5 fj.1 and an extract volume of 5 ml, the total
quantity of AN is 0.5 ^g. This represents the quantity of AN which must be
collected from the air to given the minimum acceptable signal. Assuming an
air volume.of 1.5 m^, the air level of AN must be 0.5 fig/1.5 m^ or 0.3
to produce the minimum acceptable signal.
WATER SAMPLES
In developing a protocol for water, it was difficult to separate sampling,
sample treatment and sample analysis. The results will be discussed more in
terms of the various overall schemes. The major sampling concern was the
stability of the sample. This will be discussed first followed by the results
of four analysis schemes.
Stability of Acrylonitrile Solutions
The two variables that were tested in studying the aqueous stability of
AN were water type and pH. A series of solutions was prepared in duplicate
at 10 ppm AN from distilled water and from Mississippi River water. Prior to
spiking, the pH was adjusted to 4 or 10 or left unadjusted. The samples were
all stored at room temperatures in vials with Teflon®-lined caps and had a
headspace equal to 25% of the sample volume. The samples were analyzed at 0,
1, 6 and 23 days after preparation. The results are shown in Figure 10 as the
average of duplicate samples. As can be seen, there were day-to-day variations
in the analysis, but there was no indication of sample decomposition occurring
in the distilled water samples that can be related to pH (hydrolysis).
Furthermore, these results indicate that AN was not lost to the atmosphere,
at least in stoppered vials, and thus it was not necessary to collect samples
with zero headspace. All samples in the river water showed decomposition at
23 days, although pH did appear to have stabilizing effect in that the pH 4
sample contained 67% of the original concentration while the pH 10 sample
was totally decomposed. The sample with pH unaltered had decomposed totally
at 6 days.
39
-------
4 5
Days
O PH4
a pH Unadjusted
A PH 10
•ff-
23
Figure 10. Stability of Acrylonitrile in Water
40
-------
Direct Aqueous Injection Gas Chromatography (DAIGC)
The technique of DAIGC Is an established method for AN samples of 1 ppm
or greater (12). It was used routinely in the studies of the stability of AN
in water. The limited sensitivity (~-l ppm), however, made it inadequate for
the analysis of real samples.
Adsorption from Water
No reference was found in the literature of any attempts to use solid
adsorbents to remove AN from water. Preliminary experiments were performed
to evaluate its potential. Five adsorbents (-^l g) were packed in chromatog-
raphy columns arid 500 ml of a 20-ppb aqueous AN solution were passed through
them at 4 ml/min. After all the water had passed, the columns were dried by
briefly blowing air through them. The adsorbents were then eluted with 25 ml
methanol at a rate of 5 ml/min. The eluates were analyzed for AN by GC/Hall
and the results are listed in Table 17. The poor recoveries may be due to low
adsorption from water or to low desorption by methanol. Nevertheless, the
results were not encouraging and no further studies were made.
TABLE 17. RECOVERY OF AN IN WATER FOLLOWING ADSORPTION
Adsorbent % recovery
Calgon PCB 30
Fisher coconut 35
Porapak N 0
Chromosorb 101 0
Chromosorb 104 0
Azeotropic Distillation
AN forms an azeotrope with methanol that distills at 61.4°C and is 38.7%
acrylonitrile. By adding a small amount of methanol to water, AN has been
isolated by azeotropic distillation.2J Since methanol does not form an azeo-
trope with water, the distillate is largely methanol.
41
-------
The distillation apparatus consisted of an insulated glass column, 35 x
2.0 cm, packed with 1/8 in. glass helices, a 1-liter single-necked distilla-
tion flask and heating mantle and a minidistillation head with a low volume
holdup (Ace Glass, 9322). The nominal operating conditions used as a starting
point were: volume of water, 500 ml; methanol, 25 ml; acid, 5 ml l^SO^ (18 N);
distillation rate, 1 ml/min; fraction collected, first 10 ml.
None of the published studies had reported the recovery of the technique
or the effect of variables. Specific variables were considered to be acidity,
level of methanol, distillation rate and water type. Various experiments were
conducted in order to establish the effect of these variables and determine
optimum conditions for recovery of AN.
The experimental studies of these variables were also designed to de-
termine the distillation curve of AN. Initially, two or three 10-ml aliquots
of the distillate were collected for analysis. Using four samples at 50 ppb,
the average recovery was 87% and 7.8% in the first and second 10-ml cuts,
respectively. Column flooding frequently prevented the collection of a third
cut. Refinement of the apparatus by decreasing the size of the distillation
head increased the fraction of AN in the first 10 ml of distillate to 92%.
The second fraction was not analyzed.
A summary of the study of the effects of acidity, methanol level and
water types is given in Table 18. When no methanol was added, severe column
flooding occurred and no effective distillation could be accomplished. The
recovery was acceptable at 25 ml methanol but decreased again with 50 ml.
No effect was observed when the pH of the solution was unadjusted; however,
the recovery was very poor when the pH was raised to 10. This probably is
due to loss of AN by hydrolysis. The need for acid was to insure that the
solution was not basic. The distillation rate, as governed by the heat ap-
plied to the distillation pot, also had no effect. Similarly, the water type
(distilled, tap or Mississippi River) did not significantly affect the
recovery.
The recovery was good from 10 to 500 ppb. The technique can serve to
"clean-up" and concentrate environmental samples for GC/MS analysis.
Purge and Trap
The low boiling point of AN suggested that it should be amenable to the
purge and trap technique. However, studies of the purging and/or trapping
conditions could not be found in the literature, despite its inclusion in the
"Priority Pollutant" list. Preliminary studies were.made using fairly conven-
tional conditions, e.g., purging of 10 ml of water with helium at 30 ml/min
for 10 min at room temperature; trapping on Tenax and flashing at 180*C;
42
-------
TABLE 18. RESULTS OF AZEOTROPIC DISTILLATION STUDIES
Parameter varied^/ % recovery]?/
Methanol added
0 ml flooded
10 ml 59
25 ml 89, 94
50 ml 64, 77
Acidity
5 ml H2SOA (18 N) 89, 94
0 ml H2S04 101
pH 10 20
Distillation rate^-'
61 89, 94
66 100
Water type
Distilled 89, 94
Tap 99
Mississippi River 80
AN level
5 yg (10 ppb) 82
12.5 yg (25 ppb) 98
25 yg (50 ppb) 89, 94
50 yg (100 ppb) 88
100 yg (200 ppb) 85
250 yg (500 ppb) 88
a/ The conditions before optimization were: volume, 500 ml; methanol, 25 ml;
acid, 5 ml H2SO^ (18 N); distillation rate, 61; water type, distilled;
AN, 25'yg (50 ppb).
b_/ The percent recovery is the sum of the first and second 10-ml aliquots.
Duplicate results are also given.
c/ Refers to Variac setting.
43
-------
Chromosorb 101 GC column with FID. The recoveries at these conditions were
virtually nil. It was then decided to isolate the steps in the procedure and
determine their efficiencies individually.
Purge Studies—
The purging system, shown in Figure 11, was designed so the sample was
introduced into the purge tube through 1/16 in. Teflon® tubing from a syringe
fitted with Teflon®valves. This system also permitted the periodic withdrawal
of the test solution through the same tubing. A solution of 10 ml of 10-ppm
AN was purged with helium and samples were analyzed periodically by DAIGC in
order to monitor the percent of AN remaining. Purge temperature and flow rate
(purge volume) were varied with the objective of reducing the volume of purge
gas required to remove >90% of the AN. The effect of purge solution tempera-
ture and purge gas flow rate on the purging efficiency is shown in Figure 12.
At 20°C, only 80% of the AN had been purged by ~2 liters of gas. Raising
the temperature to 67 to 68°C greatly increased the purging efficiency. Over
90% was purged by about 0.8 liter of helium. The flow rate did not appear
to have any significant effect. The temperature effect was further studied
by purging at 85°C and 20 ml/min. The volume required to purge 99% was re-
duced to about 0.5 liters. At a flow rate of 20 ml/min and a temperature of
85°C, a purge time of 30 min was judged to be adequate for quantitative
purging.
Trap Studies—
During the purge studies, a preliminary test of the trapping efficiency
of Tenax®-GC was made. Ten milliliters of a 50-ppb AN solution at 68°C was
purged for 30 min with 30 ml/min of helium. Two Bendix Flasher Tenax tubes
were attached in tandem to trap the purged AN. The trapped AN was thermally
desorbed at 180°C onto the GC column (Table 3) for quantitation. The recovery
of AN on the first and second trap was 57 and 12%, respectively, for a net
recovery of 67%. The presence of AN in the second tube indicates that the
trapping step was inefficient; i.e., the elution volume was exceeded.
Since the study had shown the Tenax had not trapped the AN when 0.9 liter
of purge gas was used, a study was made to find a better trapping material.
The breakthrough or elution volumes were measured using the same purge system.
The trapping materials were packed in the 3 in. x 1/4 in. stainless steel
tubes. A second trap, containing Tenax®, was added after the first trap con-
taining the test material. The test material was spiked with 1.0 g AN, and
helium was passed through the two traps at 20 ml/min. Every 10 min (0.2 liter
of helium), the Tenax "^ tube was removed for analysis and a second Tenax® tube
replaced the first tube. The Tenax® tubes were analyzed by thermal desorption-
gas chromatography. The Tenax® tubes were interchanged and analyzed until 80
to 90% of the AN had eluted. The tests were made with 0.25 g Chromosorb 104
using dry helium and with 0.33 g Porapak N using wet and dry helium. The
44
-------
Teflon Two-way Valve
with Luer-Locki
I Ocm Syringe
Teflon Two-way Valve
with Uwr-Lack & 1/16"
x 3" Stolnleu Steel Tubing
Swagelock Reducing
Union, Bored Through
3' x 1/4" Stalnleu
Steel Tube (Tenox)
He I turn Purge Gai
Figure 11. Purging System of Volatile Organic Analysis
45
-------
O 20°C, 30m//min
68° C, 30m//min
• 67° C, 15m//min
A 85° C, 20m//min
I I I I I I I
Purge Volume, Liters
Figure 12. Percentage of Acrylonitrtle Remaining During Purging with Helium
46
-------
results are shown in Figures 13 and 14 as the cumulative percent breakthroughs
of AN on Chromosorb 104 and Porapak N, respectively. There was little differ-
ence in the breakthrough volumes of Chromosorb 104 and Porapak N. Since
Chromosorb 104 is a copolymer of styrene and acrylonitrile, it was thought to
be wiser to use Porapak N as the trapping material. The experiment using
Porapak N was repeated with 10 ml of tap water in the purge tube in order to
simulate the actual purging conditions. This had no measurable effect on the
elution volume. Since it was just established that a purge volume of only
0.5 liter would be required, it was concluded that Porapak N would be an ade-
quate trap for AN.
Thermal Desorption—
The next step in the process that was evaluated was the desorption effi-
ciency. Fifty nanograms of AN in 10 £il water were injected directly onto the
adsorption tube. The tube was then inserted into the Bendix flasher at 100°C
or 180°C and a normal chromatographic analysis was conducted. The same amount
of AN was injected through a heated Swagelock T, fitted with a septum and
located just downstream of the flasher. A comparison of areas was made to
determine the desorption efficiencies listed in Table 19.
The system was transferred to the gas chromatograph equipped with a Hall
detector operating in the nitrogen selective mode. The trap and desorption
system became an integral part of the gas chromatograph. A schematic arrange-
ment of the new system, without the purge equipment, is shown in Figure 15.
The mass spectrometer is not in the system. The trap is a straight piece of
1/8 x 10 in. stainless steel tubing packed with 0.3 g of 50/80 mesh Porapak N.
The trap is wrapped permanently with a heating tape which is controlled by an
RFL Controller. A list of components is given in Table 20.
The thermal desorption operating .parameters were reestablished since
both a new trap configuration and a different heating technique were being used.
The two variables that were evaluated were the thermal desorption temperature
and the desorption time. A known quantity of AN in H20 was injected onto the
head of the trap via a septum. The desorption was then made and a normal
chromatographic analysis was performed. The first variable studied was the
desorption temperature. A constant desorption time of 5 min was maintained
by switching valve 2 (see Figure 15) to take the trap out of the GC system.
The desorption temperature was determined by the dial setting on the RFL con-
troller. The actual maximum temperature achieved was measured by a thermo-
couple placed between the heating tape and the trap. The results are shown
in Figure 16. A temperature of 155 to 194°C is required for complete desorp-
tion. A desorption temperature of 172°C (80 on the dial) will be used since
the temperature limit of Porapak N is ~190°C. The time required for complete
desorption was investigated at both 172°C and 194°C. As before, desorption
time was controlled by valve 2. The results are graphed in Figure 17.
47
-------
00
1001
I 80
o
60
o>
o
o
t>
CO
V
40
>~ 20
•0°
1.0
2.0
Purge Volume, Liters
3.0
4.0
Figure 13. Cumulative Percent of Acrylonitrile Breakthrough on Chromosorb 104
-------
100
8 80
I 60
u
"I
o
-
40
CO
v
c
_o
_
20
O Purged with Dry Helium
A Purged with Water Saturated Helium
I I I
I I I
I I I I I I I I
0.4 0.8
1.2 1.6 2.0 2.4
Purge Volume, Liters
2.8 3.2 3.6 4.0
Figure 14. Cumulative Percent of Acrylonitrile Breakthrough on Porapak N
-------
TABLE 19. EFFICIENCY OF THERMAL DESORPTION OF ACRYLONITRILE
% Desorbed
Trapping Material 100°C 180°C
Tenax® 96 91, 93
Chromosorb 102 95, 98, 98 91, 86, 97, 86
Chromosorb 104 106, 99, 100, 98
Porapak N 102, 105, 98, 103, 102
Calgon OL Charcoal — 13
50
-------
Helium (GC)
Helium (GC)
Heated
Lines
Gas Chromatograph
Hall
Detector
Mass
Spectrometer
Figure 15. Trapping and desorbtng system for volatile organic analysis
Materials Belong Te:
0
WnsWngU.V^C 20460
51
-------
TABLE 20. COMPONENTS OF PURGE AND TRAP SYSTEM
Purge System
Glenco Valve, No. 3070
4 - 8 y sintered glass frit
Glenco jacketed column, No. 3065-13
Hamilton Valve, No. FT1 (1/16 in.)
Hamilton Valve, No. FL1
Hamilton Syringe, No. 610CH,LL
Glenco Coupling No. 3020-2X4
Hanke E-52 Constant Temperature Circulator
Trap System
Carle Valve No. 5521
Carle Valve No. 2011
1/8 x 10 in. Stainless steel trap
Briskeat heat tape, No. B1N-2-1/2
RFL Temperature Controller No. 872
Fenwal 100 K ohm Glass Bead Thermistor
Gas Chromatograph
Varian Model 3700
Detector
Tracer Hall Electrolytic Conductivity, Model 700
52
-------
400
300
o
I
200
100
•I
(172°C) (194°C)
5 Minute Desorption
(138°C)
I
65
70 75 80
Temperature Control
85
90
Figure 16. Effect of temperature on desorption of acrylonitrile
400
300
o
0)
~S 200
Q_
100
T
• Temperature Control at 80
— — • Temperature Control at 85
I
I
0246
Desorption Time (Min)
Figure 17. Effect of time on desorption of acrylonitrile
8
53
-------
Desorption was negligible in the first 1.5 rain but was complete by 3 min. No
effect of desorption temperature was noted. A desorption time of 5 min will
be used.
Operation of the Final System—
The inclusion of the purging operation into the thermal desorption, chro-
matographic analysis required optimizing a number of variables. It was found
that one of the more critical steps was the removal of water vapor from the
trap and tubing following the purge step. When the trap was desorbed immedi-
ately, a significant quantity of water was transferred to the GC. This caused
an erratic response in the Hall detector. However, by switching valve 2 at the
end of the purge, dry carrier could be passed through the trap and line and
vented. It was subsequently established that a 5 min drying would allow the
detector to operate properly but would not result in the loss of any AN. The
final operating conditions are listed in Table 21. The overall analytical
process on a time scale is listed in Table 22.
Using the optimized procedure, a calibration curve from 0.1 to 10 ppb
AN in water was generated and is shown in Figure 18. Triplicate analyses were
done at the 0.1 ppb level. A recovery of 96% was attained with a relative
standard deviation of 10%. A typical chromatogram for 10 ml of 0.1 ppb AN in
tap water is shown in Figure 19.
A limited study of several potential interferences was made. The com-
pounds selected for study were all nitrogen-containing since the Hall detector
is selective for nitrogen. The list was also limited to structurally similar
nitriles which might be expected to co-exist with AN. The chromatogram of
the mixture, given in Figure 20, shows that none of the compounds will inter-
fere. The retention times and relative retention times are given in Table 23.
GC/MS Confirmation
The GC/MS procedure was developed as a confirmation for either the azeo-
tropic distillation technique or for direct aqueous injection technique. The
chromatographic conditions optimized for these methods were used with minor
variations. Initially, AN was detected by collecting full scan mass spectra
and presenting the data as extracted ion current plots. Later SIM was used to
improve the sensitivity. Figure 21 compares the EICP for 5 ng of AN with the
SIM plot for 1 ng AN. The final operating conditions are given in Table 4.
Limit of Detection in Water Samples
The limit of detection for the azeotropic distillation procedure varied
due to changes in response of the Hall detector. The variable factor was the
ng of AN detectable in an injection volume of 5 to 8 /zl. The minimum quantity
of AN detectable on a given day was estimated from injections of standards
containing 0.5 to 1.0 ng AN.
54
-------
TABLE 21. OPERATING CONDITIONS FOR PURGE AND TRAP ANALYSES OF ACRYLONITRILE
Purge:
Gas: Helium at 20 ml/min
Purge Time: 30 min
Purge Temperature: 85°C
Sample Size: 5 to 10 ml
Desorption:
Trap Material: 0.3 g 50/80 mesh Porapak N
Carrier 1: Helium at 45 ml/min
Desorption Temperature: RFL Controller at 80 = 172*C peak temperature
Desorption Time: 5 min
Gas Chromatograph:
Carrier 2: Helium at 45 ml/min
Column: 6 ft x 1/8 in. Stainless steel with 80/100 mesh Chromosorb 101
Column Temperature: 80°C for 3 min then 8°C/min to 150°C
Transfer Line: 200°C
Detector:
Hydrogen Flow: 60 ml/min
Solvent: 15% n-propanol in water at 0.5 ml/min
Pyrolysis Oven: 850°C
Mode: Catalytic reduction using nickel wire
55
-------
TABLE 22. PURGE AND TRAP SYSTEM FOR ACRYLONITRILE - SPECIFIC INSTRUCTIONS^/
Time,
Min
-
0
30
35
40
43
47
Valve Position Thermal
No. 1 No. 2 Purge Gas Desorption G.C. Program^/
2 1 off off off - (80°C)
1* 1 on* off off - (80°C)
2* 1 off* off off - (80°C)
2 2* off on* on - (80°C)*
2 1* off on on - (80°C)
2 1 off off* on (110°C)
2 1 off off off (150°C)*
Vent at
Hall Detector
Vented
Vented
Vented
Vented
Vented
Unvented*
Vented*
Operation
Ready Condition
Purge and Trap Occuring
Purge completed, trap switch out
and being dried.
Trap switch to G.C. and thermally
desorbed.
Trap switched off G.C., but heated
to clean.
Trap heat off, Vent closed and
detector receiving column effluei
Acrylonitrile has eluted, G.C. has
reached 150CC and will return to
80@C. Column effluent vented.
a./ An asterick appears where a specific action is required. When more than one action is required at the same
time, proceed from left to right.
b_/ 80°C for 5 min then 10°C/min to 150°C and then return to 80°C.
-------
100
10
<
LLJ
<
1.0
0.1
0.1 1.0
ACRYLONITRILE, ppb
Figure 18. Calibration curve
10
-------
Thermal Desorption
Column at 80° C
Column Programmed
at 10°C/Minute to
200° C
00
1 Nanogram
Acrylontrile
Column
Vent Closed
15
Minutes
Figure 19. Chromatograph of 10 ml of 0.1 ppb AN
-------
Thermal Desorption
Column at 80° C
Column Programmed
at 8°C/Min to 200
Ul
Column Vent
Closed
10
15
Minutes
Figure 20. Chromatogram of potential interferences
-------
TABLE 23. RETENTION TIMES
Compound tr> (min) tRfAN')
Acetonitrile 8.7 0.84
Acrylonitrile 10.4 1.00
Propionitrile 12.5 1.20
Isobutyronitrile 14.0 1.35
Methoxyacetonitrile 14.5 1.39
n-Butyronitrile 15.2 1.46
60
-------
4000
Five Nanograms Acrylonitrile
~ ¥ ' ~ P • I * I * I 1 'fl'l'f'l
27870 27880 87880 87800 87810 37880 87880 87840 87890 2VQOU
a.
t Nonograr*
(x;o)
M/'« • 52
I ' ' 1 ' ' 1 ' ' I ' ' I ' ' I ' ' I ' ' I ' ' I ' ' | ' ' I ' ' I !
!23<.3870310M12
MINUTE. 5
b.
Figure 21. Comparison of GC/MS sensitivity for acrylonitrile. a. Full-
scan mode; b. Selected ion monitoring.
61
-------
The detection limit then becomes:
ng AN required per injection x 10 ml distillate = Mg/liter detection limit
volume injected (5-8 /il) 500 ml sample
The limit of detection for the purge and trap procedure was not rigorously
established. However, a 0.1 /^g/liter clean water sample was analyzed and
yielded a chromatographic signal that was easily measured.
SOIL AND SEDIMENT SAMPLES
No method development was completed on the analysis of AN in soil and
sediment. It was felt that water extraction of these media should be highly
effective due to the good solubility of AN in water. Furthermore, adequate
methods existed for the analysis of AN in water.
The limit of detection for soil and sediment analysis procedure was
influenced by the Hall detector response as discussed above. Using the
minimum detectable quantity of AN on the analysis day, limit of detection
becomes
ng AN required per injection x 50 ml extract volume _ /xg_ , . , .
volume injected (5-8 jul) 10 g soil (sediment) kg
62
-------
SECTION V
SELECTION OF SAMPLING SITES
A list of suggested plant sites for monitoring acrylonitrile in ambient
air was prepared by Bill Lamason, Strategies and Air Standards Division,
Environmental Protection Agency. The text of this report is reproduced below.
"Date: August 30, 1977
Subject: Suggested Plant Sites for Monitoring of Acrylonitrile in the
Ambient Air
From: Bill Lamason,
Strategies and Air Standards Division
To: Vince DeCarlo
Office of Toxic Substances
Based on production research I have done on acrylonitrile producers and
users, as well as information and guidance provided by John Bachmann, Joe
Cirvello, and Don Lokey of PSB, here is our list of recommended facilities
for monitoring:
(See Table 24)
Larger facilities were chosen to maximize the chance of finding ambient
air concentrations of acrylonitrile. A variety of acrylonitrile users was
selected due to lack of adequate emissions information on these facilities.
Also chosen were plants in metropolitan areas or other areas of dense
population, where possible, to determine maximum exposure threat. Age of a
facility was also considered, if known, as a measure of emission control.
Utilizing the assumption that a new facility would have much better control
equipment, a range of new and old facilities was selected.
Geographic location of facilities within the country was noted as the
states of Illinois, Louisiana, and Texas were considered to have stronger
air pollution standards. Thus, plants in these states were assumed to be
better controlled.
63
-------
TABLE 24. SAMPLING SITES
Facility
Product produced
Reasons for recommendation
Comments
Monsanto, Texas City, TX
Acrylonitrile
Vistron Corp. (SOHIO)
Lima, OH
Acrylonitrile
Acrylamide
American Cyanamid
New Orleans, LA
DuPont, Camden, SC
Acrylonitrile
Acrylic fibers
Monsanto, Decatur, AL
Acrylic and
Modacrylic fibers
Large, new facility; Texas Acceptable
air standards are stricter,
therefore, likely to have
well controlled emissions.
Large facility; outside of
Texas, Louisiana, or
Illinois, therefore,
probably less controlled.
Smallest of major producers;
located in an area of large,
dense population.
Site of DuPont epidemiology
study; second largest use
facility in country; located
outside of Texas, Louisiana,
and Illinois, therefore,
probably has less atmospheric
emission controls.
Largest use facility in the Acceptable
nation; outside of Texas,
Louisiana, and Illinois,
therefore, probably has less
atmospheric emission controls.
Reported to be no longer
producing acrylamide.
The effect on acrylo-
nitrile production must
be determined.
Acceptable
Acceptable
(continued)
-------
Facility
TABLE 24. (continued)
Product produced Reasons for recommendation
Comments
Ln
Monsanto, Addyston, OH
ABS and SAN
resins
Borg-Warner, Washington, W ABS and SAN
resins
Goodrich, Louisville, KY
American Cyanamid,
Linden, NJ
Nitrile
Elastomers;
ABS resins
Acrylamide
DuPont, Montague, MI
Nitrile Barrier
resins
Largest acrylonitrile user
among ABS-SAN resin produ-
cers; upwind of metropol-
itan Cincinnati.
Second largest acrylonitrile
user among ABS-SAN resin
producers.
Largest acrylonitrile user
among nitrile elastomer
producers; in the metropol-
itan Louisville area.
Largest acrylonitrile user
among acrylamide producers;
located in a large, dense
population in the New York
metropolitan area.
Largest user of acrylonitrile
among Nitrile Barrier resin
producers.
Acceptable
Acceptable
Acceptable
Not recommended. In-
accessible for air
sampling.
Acceptable
-------
If you have any questions or comments on this recommended list for moni-
toring of acrylonitrile facilities, please feel free to contact me or, after
September 12, Joe Cirvello or John Bachmann.
cc: Pat Hilgard
Joe Cirvello
John Bachmann"
The list of recommended sites is given as Table 24 with a few comments that
were added at the time the list was received. Several changes were made to
the list of recommended samping sites before a final list was developed. At
one point, the Borg Warner, Washington, West Virginia, site was replaced with
Borg-Warner, Ottawa, Illinois. This was then reversed when it became apparent
that the potential for population exposure was low at the Illinois site. The
American Cyanamid plant at Linden, New Jersey, was not recommended for sampling
due to its inaccessibility. This plant, however, was returned to the recommended
list in response to the request of Region II, EPA. The American Cyanamid,
Milton, Florida, plant was considered as a replacement or an addition but was
eventually eliminated due to a low potential for population exposure. DuPont,
Montague, Michigan, was removed when it was established that AN was not a
plant product. DuPont, Waynesboro, West Virginia, was inserted as the re-
placement. A Uniroyal plant at Scottsbluff, Louisiana, was considered as a
substitution for the American Cyanamid, Milton, Florida, plant but was poorly
situated for air sampling and was eliminated. The Uniroyal plant at Paines-
ville, Ohio, was the final addition to the list.
A meeting was held at the Office of the Toxic Substances on January 27,
1978, for the purpose of discussing Task 3, Acrylonitrile. The following were
present:
John Going MRI 816-753-7600
Roger Caiazza MRI 816-753-7600
Vince DeCarlo EPA-OTS 202-755-0300
Liz Bryan EPA-OTS 202-426-9819
Cindy McMillin EPA-OTS 202-755-0300
Joe Breen EPA-OTS 202-426-9819
Joe Cirvello EPA-DAQPS, 919-629-5355
Durham, NC
Phil Youngblood EPA-DAQPS, 919-541-5391
Durham, NC
66
-------
Morris Trichon EPA-Region II 212-264-9539
Gordon Howard EPA-EPIC 555-3110
The principal topics of discussion were sampling strategy, the ordering of
sites and the special task for Region II.
Mr. Youngblood recommended a sampling program that would include two 24-
hr samples and eight sampling stations. At least four of the stations were
to be downwind of the source at differing angles and distances. If possible,
the distance for maximum ground concentration would be calculated and a
sampler should be placed at that distance. To do this would require knowing
at least the exact location of source and the local wind conditions. OTS
would attempt to pinpoint the emission sources by sending photographs of the
plants to Regional and State officials. The best way to determine the on-site
wind conditions would be with a portable weather station. MRI, however, does
not have this type of equipment. Phil Youngblood said that he was aware of
at least three portable units at RTP which we might be able to use. Mr. Cirvello
would make the inquiries on this. The question of whether to collect samples
for 1 or 2 days was not resolved except for the Region II task. The primary
reason for collecting two samples seems to be an insurance mechanism for
getting one good sample. OTS's response was that if the first sample was
taken properly, the extra expense to collect the second sample is unwarranted.
The criteria for a good sample include having good wind conditions and ob-
taining on-site wind data.
Before discussing the sampling schedule, Dr. DeCarlo explained that
Region II was anxious to know whether the American Cyanamid plant at Linden,
New Jersey, was emitting acrylonitrile. Therefore, it was listed as the
second plant to be visited. Otherwise, it was agreed that sampling would
start at a less critical site such as New Orleans, skip a week and resume
with the New Jersey site, the Gulf Coast sites, the Mid-Atlantic and then the
Midwest sites. The probable order would be:
American Cyanamid, LA
American Cyanamid, NJ
Monsanto, TX
Monsanto, AL
DuPont, SC
DuPont, VA
Vistron, OH
Monsanto, OH
Goodrich, KY
67
-------
The sampling strategy to be used at Linden, NJ, would be altered to satisfy
the needs of Region II. Mr. Trichon would be responsible for the selection
of the sampling locations.
The list of plants sampled is given in Table 25 in the actual order of
sampling. The locations of the plants are shown in Figure 22.
68
-------
TABLE 25. RECOMMENDED SAMPLING SITES
Plant (Location)
American Cyanamid
New Orleans, LA
American Cyanamid
Linden, NJ
Monsanto
Texas City, TX
Monsanto
Decatur, AL
DuPont
Canden, SC
DuPont
Waynesboro , VA
Borg-Warner
Washington, W. VA
Goodrich
Louisville, KY
Monsanto
Addyston, .OH
Uniroyal
Painesville, OH
Vistron (SOHIO)
Lima, OH
Longitude and Latitude
Coordinates
90"16'08"
29*57-30"
74*12' 15"
40*35' 00"
94*53'48"
29*22' 30"
87*01' 18"
39*38'01"
80°39'15"
34*14'14"
78*53*13"
38°03'35"
81*40'49"
39° 15 '36"
85°49'52"
38*12' 50"
84*42'44"
39*08' 07"
81*14'08"
41°45'23"
84*07' 53"
40*42'49"
Longitude
Latitude
Longitude
Latitude
Longitude
Latitude
Longitude
Latitude
Longitude
Latitude
Longitude
Latitude
Longitude
Latitude
Longitude
Latitude
Longitude
Latitude
Longitude
Latitude
Longitude
Latitude
EPA
Region Product
VI Acrylonitrile
II Acrylamide
VI Acrylonitrile
IV Acrylic and Modacrylic
Fibers
IV Acrylic Fibers
III Acrylic and Modacrylic
Fibers
III ABS and SAN Resins
TV Nltrlle elastomers,
ABS and SAN Resins
V ABS and SAN Resins
V Nltrlle elastomers
V Acrylonitrile
Acrylamide
Original Basis of Recommendation
Smallest of major producers; located in an
area of large, dense population.
Largest acrylonitrile user among acrylamlde
producers; located in a large, dense popula-
tion in the New York metropolitan area.
Large, new facility; Texas air standards are
stricter, therefore, likely to have well con'
trolled emissions.
Largest use facility in the nation; outside
of Texas, Louisiana, and Illinois, therefore
probably has less atmospheric emission con-
trols.
Site of DuPont epidemiology study; second
largest use facility in country; located
outside of Texas, Louisiana, and Illinois,
therefore, probably has less atmospheric
emission controls.
Substituted Site
Second largest acrylonitrile user among
ABS-SAN resin producers.
Largest acrylonitrile user among nitrile
elastomer producers; in the metropolitan
Louisville area.
Largest acrylonitrile user among ABS-SAN
resin producers; upwind of metropolitan
Cincinnati.
Added site.
Large facility; outside of Texas,
Louisiana, or Illinois, therefore,
69
-------
O Acrylonitrile Producer
O Acrylamide Producer
A Acrylonitrile, Acrylamide Producer
• Acrylic, Modacrylic Fibers
• ABS, SAN Resins and/or Nitrile Elastomers
Figure 22. Location of Sampling Sites by EPA Regions
-------
SECTION VI
DISCUSSION OF RESULTS
The eleven selected industrial sites were sampled between February 7,
1978, and June 13, 1978. The actual sampling schedule is shown in Table 26;
presampling site visits were conducted on the day prior to sampling. A sum-
mary of the air and water samples collected at the eleven industrial sites is
given in Table 27 and a summary of the air sampling parameters follows in
Table 28. The results from the eleven plants are discussed below.
AMERICAN CYANAMID COMPANY, NEW ORLEANS, LOUISIANA
Field Sampling
Presampling Survey—
A presampling survey of the American Cyanamid Company plant, New Orleans,
Louisiana, was conducted on February 6, 1978. The plant is located in
Jefferson Parish, Louisiana, in South Kenner along State Highway 18 about
6 miles east of Luling, Louisiana. The plant lies in a semirural area with
the Mississippi River to the north. The heaviest population area around the
plant is directly north across the Mississippi River. There are smaller resi-
dential areas to the east and the west. The area south of the plant is un-
populated swampland. The nearest highway to the south is over 4,000 m away.
The NPDES permit shows two outfalls which discharge wastewater into the
Mississippi River. Both are readily visible where they pass over the highway.
The locations of the outfalls are shown in Figure 23. This plant produces
acrylonitrile, acrylamide, polyacrylamide, melamine, urea and various inor-
ganics.
Air Sampling—
Air sampling was conducted using eight battery powered air samplers at
five sampling stations located upwind, downwind, and laterally with respect
to the plant. Samplers were located from 0 to 1.2 km from the plant perimeter
on a levee and rights-of-way and were positioned at a height of 1.5 m.
71
-------
TABLE 26. SAMPLING SCHEDULE
Date
Company
Location
Product
2/7 - 2/8
2/21 - 2/23
2/28 - 3/1
3/8 - 3/11
American Cyanamid
American Cyanamid
Monsanto
Monsanto
3/22 - 3/24 DuPont
3/29 - 3/30 DuPont
4/4 - 4/5 Borg-Warner
4/24 - 4/25 Goodrich
4/26 - 9/27 Monsanto
5/23 - 5/25 Uniroyal
6/13 - 6/14 Vistron
New Orleans, LA
Linden, NJ
Texas City, TX
Decatur, AL
Lugoff, SC
Waynesboro, VA
Acrylonitrile
Acrylamide
Acrylonitrile
Acrylic, Modacrylic
fibers
Acrylic fibers
Acrylic, Modacrylic
fibers
Washington, W. VA ABS, SAN resins
Louisville, KY
Nitrile elastomers, ABS,
SAN resins
Addyston, OH ABS, SAN resins
Painesville, OH Nitrile elastomers
Lima, OH Acrylonitrile, Acrylamide
72
-------
TABLE 27. FIELD SAMPLING SUMMARY
Site
American Cyanamid
New Orleans, LA
American Cyanamid
Linden, NJ
Monsanto
Texas City, TX
Monsanto
Decatur, AL
DuPont
Lugoff, SC
DuPont
Waynesboro, VA
Borg-Warner
Washington, W. VA
Goodrich
Louisville, KY
Monsanto
Addyston, OH
Uniroyal
Painesville, OH
Vistron
Lima, OH
Total
Air Samples
10
10
18
11
9
Water Soil and/or
Samples Sediment Samples
104
36
37
73
-------
TABLE 28. SUMMARY OF AIR SAMPLING PARAMETERS
American Cyanamid
New Orleans, LA
American Cyanamid
Linden, NJ
Monsanto
Texas City, TX
Monsanto
Decatur, AL
DuPont
Lugoff, SC
DuPont
Waynesboro , VA
Borg-Warner
Washington, W. VA
Goodrich
Louisville, KY
Monsanto
Addyston, OH
Uniroyal
Painesville, OH
Vistron
Lima, OH
Average Sampling Volume,^
(No. of Samplers)
1,170 (8)
1,160 (18)
1,310 (10)
1,320 (18)
2,660 (11)
1,540 (9)
1,390 (8)
1,390 (7)
1,460 (8)
1,470 (8)
1,210 (7)
Average Sampling
Time (hr)
23.7
20.1
22.7
23.0
46.2
26.1
25.2
22.6
28.3
26.7
21.6
Average
Rate (//min)
0.82
0.96
0.96
0.96
0.96
0.98
0.92
1.03
0.86
0.92
0.93
74
-------
Air Station
- Water & Sediment Station
Figure 23. Sampling Locations at American Cyanaraid, Fortier, Louisiana.
Numbers in parenthesis represent levels of AN in air (/^g/m ),
75
-------
Sampling was carried out during the period 0900 (CST) February 7, 1978,
to 1100 (CST) February 8, 1978. The five sampling locations are shown in
Figure 23 and described in Table 29.
A duplicate sample was collected at station A-4. This station was selected
as the most probable point of maximum impact. Additional samplers were also
installed at stations A-l and A-4 to collect air samples using spiked sample
tubes. These stations were selected as the upwind and downwind stations, re-
spectively.
Water and Sediment Sampling—
Water and sediment samples were taken at two locations, one upstream and
one downstream from the discharge. The water samples were acidified to pH < 3
at the time of collection. The sampling locations are shown in Figure 23 and
described in Table 30.
Soil Samples—
Soil samples were taken February 8, 1978, at two of the air sampling
stations by removing the top approximately 12 mm of topsoil from the area
near the samplers. Sampling data are given in Table 31.
Meteorological Conditions
Meteorological conditions that existed during the sampling period were
obtained from the National Weather Service, New Orleans International Airport.
Additional humidity measurements were made on-site. The data are summarized
in Table 32.
Sample Analysis
Air Samples—
The results of the analysis of the charcoal tubes are given in Table 33.
Acrylonitrile was found at only two of the five sampling stations. The levels
of AN in micrograms per cubic meter are shown on the site map given in Fig-
ure 23. Figure 24 shows the wind rose for the sampling period. For comparison,
the annual wind rose from hourly observations (New Orleans, LA) follows as
Figure 25. As expected, the sites where AN was found were the downwind sites,
A-4 and A-5. Without knowing the location of the AN production or storage
facilities, it is not possible to comment on the differences in AN levels at
the two stations.
The quality of the results was monitored in several ways. First, it is
significant that no AN was found in any of the backup tubes, including the
spikes. This indicates that breakthrough did not occur and that no AN
was lost in this fashion.
76
-------
TABLE 29. AIR SAMPLING DATA FOR AMERICAN CYANAMID. NEW ORLEANS, LOUISIANA
Location of Sampling Station
Sample U.S. Geological
No. survey coordinates Description
A-l
A-1S£/
A- 2
A- 3
A- 4
A-4DS/
A-4S
A- 5
3317850 mN
763550 mE
3317540 mN
764200 mE
3317650 mN
764380 mE
3316770 mN
762630 mE
3317260 mN
762630 mE
On levee 100 m north of
fence line
On Kenner Road 300 m
north of Texas and Pacific R.R.
South end of Moss Lane
South end of Portiere Heights
Road
On Portiere Heights Road
300 m south of Highway 18
Total
sampling
(hr)
24.1
24.1
23.6
23.9
23.5
23.5
23.5
23.7
Sampling
rate
(jtf/min)
0.73
0.67
1.01
0.96
1.04
0.63
0.64
0.93
Total
volume
(m3)
1.05
0.96
1.43
1.37
1.46
0.89
0.90
1.32
Sampling
height
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
aj S = Spiked sample; D = Duplicate sample.
-------
TABLE 30. WATER AND SEDIMENT SAMPLING DATA FOR AMERICAN CYANAMID, NEW ORLEANS, LOUISIANA
Sample U.S. Geological
No. survey coordinates
Location of sampling station
Description
Sample Volume % Moisture
type collected by weight
W-l
3317550 mN
762190 mE
1,845 m upstream of discharge 001
Grab 3.79 £
water
W-2
3317960 mN
765280 mE
3,858 m downstream of discharge 001 Grab 3.79
water
Se-1
3317550 mN
762190 mE
1,845 m upstream of discharge 001 Sediment 100
Se-2
3317960 mN
3,858 m downstream of discharge 001 Sediment 100
oo
-------
TABLE 31. SOIL SAMPLING DATA FOR AMERICAN CYANAMID. NEW ORLEANS. LOUISIANA
Location of Sampling Station
Sample
No.
So-1
So-1
U.S. Geological
Survey coordinates
3317859 mN
763550 mE
3316770 mN
762630 mE
Description Sample Size
On levee 100 m north of fence line 100 cm3
South end of Fortiere Heights Road 100 cm3
% Moisture
by weight
-
-
VO
-------
TABLE 32. METEOROLOGICAL CONDITIONS DURING SAMPLING AT AMERICAN CYANAMID, NEW ORLEANS, LOUISIANA:
SOURCE OF DATA, NEW ORLEANS INTERNATIONAL AIRPORT. NEW ORLEANS, LOUISIANA
co
o
Time (LST)
2/07/78
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2/08/78
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
Wind
Direction
(00-36)
07
05
06
06
07
06
07
08
07
07
06
05
05
06
06
05
05
06
05
06
05
06
04
06
03
Speed
(kts)
15
15
14
12
10
11
10
9
11
9
11
11
13
10
13
14
11
13
12
8
9
9
9
9
10
Temperature
1.1
1.7
2.8
2.8
3.3
3.9
3.9
3.9
3.9
3.9
3.9
3.9
3.3
3.3
2.8
2.2
1.7
1.7
1.1
1.7
1.1
1.1
0.6
1.1
1.1
Humidity
Relative, % Absolute (mg/£)
55 (42l/) 2.9 (2.2i/)
55
50
50
50
50
50 3.2
55
55 (54l/) (3.4I/)
45
45
45
45 2.7
40
45
45
45
45
50 2.6
50
50
50
50
55
60 (72-') 3.1 (3. 7-i/)
Pasquill
Stability
Class
Precipi-
tation
4 (Neutral) None
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
Rain
Humidity measured on site
-------
TABLE 33. ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM
AMERICAN CYANAMID, NEW ORLEANS, LOUISIANA
GO
Sample
No.
A-l
A-1S
A-2
A-3
A-4
A-4D
A-4S
A-5
A-B1
A-B2
Sampling
2/7/78
2/8/78
2/7/78
2/8/78
2/7/78
2/8/78
2/7/78
2/8/78
2/7/78
2/8/78
—
—
Period
0920-
0925
1010-
0944
0945-
0936
1031-
1002
1020-
0959
—
—
Air Volume
(m3)
1.046
0.963
1.430
1.374
1.464
0.885
0.897
1.324
—
Sample GC/MS .
Type— Comments ug Found y,g/m Confirmation-
Front — <0.1 <0.1-' Yes
Front Spiked with 15 ug AN 6.8 45% Recovery
at Station A-l
Front — <0.1 <0.1
Front — <0.1 <0.1
Front — 20.2, 19.8- 13. (T Yes
Front Duplicate at Station 9.6, 9.8^-' 11.0 Yes
A-4
Front Spiked with 15 /ig AN 25.4, 25.3d-'
at Station A-4
Front — 0.4, 0.6- 0.4 No
Field — <0.1
Blanks — <0.1
aj All back-up tubes were analyzed. AN was < 0.1 //g.
b/ Yes = Confirmation of presence or absence of AN; No
cj Based on sample size and instrument detection unit.
d_/ Results of duplicate injections.
e/ Not corrected for recovery.
= Unconfirmed; sample below detection unit.
-------
NNW
NW
WNW
WSW
SW
I
SSW
Winds Greater Than 10 Knots
NNE
NE
ENE
ESE
SE
SSE
Winds Less Than or Equal to 10 Knots
Figure 24. Wind patterns during sampling at American Cyanamid,
Fortier, Louisiana
82
-------
NW
WNW
WSW
SW
NNW
SSW
I Winds Greater Than 12mph
I Winds Less Than 12mph
NNE
SSE
NE
ENE
ESE
SE
Figure 25. Annual wind rose for New Orleans, Louisiana (1/51 - 12/60)
83
-------
The absolute humidity is known to be an influencing factor in controlling
breakthrough. The measured humidity during the test was 2 to 4 mg/liter, well
below the critical level of ~12 mg/liter.
Sampling quality assurance was checked by the use of spiked sampling
tubes, upwind and downwind samples of the plant, spiked with 15 |ag AN. The
upwind results would indicate the level of recovery. The downwind results
would indicate whether AN was being destroyed by other plant emissions. The
upwind recovery was 6.8 jxg or 45%. A calculation of the recovery of the
downwind spike must take into account the amount of AN collected from the air
and desorbed into the charcoal. The calculation of the amount of AN collected
from air and subsequently desorbed relies on knowing the collection and de-
sorption efficiencies, i.e., the recovery. In other words, calculation of
the recovery in the downwind sample requires knowing the recovery factor!
Although the recovery of AN at the downwind spike could not be readily deter-
mined, its presence indicates that there is nothing in the downwind air that
destroyed or reacted with the trapped AN.
Analytical quality assurance was demonstrated by duplicate injections of
all positive samples plus fortification of one positive sample. The agreement
of duplicate injections was better than 2% for all samples except A-5 where
the observed level was fairly low, i.e., 0.3 and 0.5 jug/m^. An aliquot of the
sample A-4 was fortified with an equivalent quantity of AN and reanalyzed.
The recovery of added AN was 100%.
A gas chromatogram of sample A-4 was obtained at conditions chosen to maxi-
mize resolution. This is shown in Figure 26 with the AN peak noted. Tentative
identifications of some of the ~25 peaks, based on extracted ion current
plots (see following section), are also noted on the chromatogram.
Water Samples—
The results of the analyses to the two water samples are given in
Table 34. Both the azeotropic distillation technique and the purge and trap
technique were used for the analyses. No AN was detectable by either proce-
dure. The lowest limit of detection was 0.10 ppb, using the purge and trap
procedure. Sample W-2 was spiked at 0.1 ppb and analyzed by purge and trap.
The recovery of added AN was 102%.
Soil and Sediment Samples—
The soil and sediment samples were extracted with water and the extracts
were analyzed by both direct aqueous injection and purge and trap. The re-
sults are summarized in Table 35. Soil sample So-2 was found to contain a very
low level of AN, i.e., 0.5 /iig/kg. This sample was downwind of the plant.
Sediment sample Se-1 was thought to contain AN based on the analysis of the
azeotropic distillate. No AN, however, was found by the purge and trap
analysis. It was concluded that no detectable AN was present in the sediment.
84
-------
00
10
12
14
16
18
20
22
MINUTES
Figure 26. Gas chromatogram of sample A-4
-------
TABLE 34. ACRYLONITRILE CONCENTRATIONS IN WATER SAMPLES FROM
AMERICAN CYANAMID, NEW ORLEANS, LOUISIANA
Azeotropic distillation Purge and Trap
Sample Initial Volume
No. (ml)
W-l 500
W-2 500
W-5^/
Final Volume /ig Found Volume Analyzed .
(ml) (original sample) /Ig/j?— (ml) ng Found fj.R/&—
10 < 0.4 <0.8 10 < 1 < 0.1
10 <0.4 <0.8 10 <1 <0.1
10 0.051 (102%) 0.1
aj Based on instrument detection limit and sample size.
W Spiked with 0.050 ptg AN.
00
-------
TABLE 35. ACRYLONITRILE CONCENTRATIONS IN SOIL AND SEDIMENT SAMPLES FROM
AMERICAN CYANAMID. NEW ORLEANS. LOUISIANA
Sample
No.
So-1
So- 2
Se-1
Se-2
Sample
size
10 g
10 g
10 g
10 g
Extraction
volume
50 ml
50 ml
50 ml
50 ml
yg Found
(original sample)
DAIGCaV
< 2
< 2
-v 3
< 2
P&TW
< 0.005C./
< 0.005
< 0.005
< 0.005
yg/kg
DAIGCJ!/ P&Tb/
< 200 < 0.5£/
< 200 ^0.5
30Qd/ < 0.5
< 200 < 0.5
oo aj Direct aqueous injection gas chromatography.
• b_/ Purge and trap.
cl Based on instrument detection limit and sample size.
d/ Not confirmed by P and T analysis.
-------
Gas Chromatography/Mass Spectrometry Confirmations—
The confirmation of AN in some of the samples was made by GC/MS. A full
scan GC/MS approach was tried initially but found not to have the necessary
sensitivity. The sensitivity of the technique was limited to -«-15 ng (injected)
by the background contribution from the column. By operating the instrument
in a high resolution, selected ion monitoring mode (SIM), it was possible to
monitor AN in the presence of the background, thus, lowering the detection
limit to ^1 ng. The C3H3N+ and C3H2N+ ions with m/e values of 53.0265 and
52.0187, respectively, were monitored. The background peaks were due to C^E^+
at 53.0391 and C4H4+ at 52.0313.
The presence of AN in air samples A-4 and A-4D was confirmed. Figure 27
gives the SIM plot for these samples as well as for a 6-ng standard of AN.
Sample A-5 was found to contain AN by GC analyses. Its presence could not be
confirmed by GC/MS as it was below the limit of detection. The upwind sample
A-l was confirmed as not containing AN.
AMERICAN CYANAMID, LINDEN, NEW JERSEY
Field Sampling
Presampling Survey—
A presampling survey of the American Cyanamid plant, Linden, New Jersey,
was conducted on February 21, 1978. Prior to inspecting the site, a meeting
was held with Bill Librizzi, Morris Trichon, John Toomey, Eric Chasanoff and
one unidentified person, all from Region II, EPA. Mr. Chasanoff had prepared
the following document to assist in the selection of sampling stations.
Mr. Trichon provided some specific information about the AN process. The
AN production facilities are located in the northwest section of the plant
property. The company recently shut down an older facility and started up
the new operation with double the original capacity. The same type of wet
scrubber emission control is reported to be in use on the new facility. Mr.
Trichon also reported on February 24 that there was some disruption in the AN
production from 2300 on February 22, 1978, to 0700 on February 23, 1978.
Apparently the "drier" was shut down. The extent of the effect on AN produc-
tion is not known. The affected period occurred during the last 8 hr of the
second day of sampling.
The site was then inspected by Mr. Thomas of MRI and the Region II repre-
sentatives. Six sampling stations were selected and mutually agreed upon.
88
-------
6 Nanogrdim Acrylonitrile
X200
' I ' * I
49
MINI/TE.S
I
6
t
V*
£
Air Sample A- 4
X 100
MINUTES
I
9
£
v>
Air Sample A-4D
X200
I • • I • • I
3*5
MI MUTE: s
Air Sample A- 5
X200
I
2
A 9
MINUTES
I
8
Figure 27. SIM plots of 6 ng AN and air
Samples A-4, A-4D and A-5
89
-------
DA *£:
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Preliminary Determination on Monitoring Site_ Selection for the
Ambient Toxic Study _
FROM: Eric Chasanoff, Advisor
Impact Assessment Section, AB
TO? Morris Truchan
Air Facilities Branch
Having surveyed the area surrounding the plant John Toomey of Edison
and myself have selected twenty-two possible sites in Staten Island and
New Jersey for the placement of the portable monitors. The sites range
from 300 meters to 3 kilometers from the plant and most of the sites are
located near residential housing. The area around the plant was divided
into eight sectors (A-H) and each sector is forty-five degrees in width.
With the exception of sector A (unplowed roads made site surveying im-
possible) at least two possible monitoring sites were selected in each
sector.
Methodology
The "PT series" of models were used to determine the source's
impact area. After careful evaluation of the results we have decided
that the source's impact is insignificant at Distances in excess of
3 kilometers. Therefore, all the sites were located within 3 kilo-
meters of the plant.
The following site locations are listed below:
SECTOR A: No sites selected.
SECTOR B: East of Meridth Avenue in Chelsa (vacant lot).
Between Meridth Avenue and Spencer Road, Northwest of
South Avenue in Chelsa (open area).
North of Travis Road and South Avenue intersection
(open area).
SECTOR C: Abandoned storage tanks west of West Shore Expressway,
southwest of Meridth Avenue.
Swamp east of West Shore Expressway, northeast of Meridth
Road.
90
EPA FORM 1320-6 (REV. 3-76)
-------
- 2 -
Victory 31vd and Cannon Avenue (vacant lot)
North of Cannon Street in Travis (vacant lot).
SECTOR D: Glen Street at Patterson Cemetery east of West Shore
Expressway; Travis.
Glen Street at Sylvan Cemetery east of Hest Shore
Expressway; Travis.
Wild Avenue southeast of Victory Blvd near, the Arthur Kill
powere plant; Travis (vacant lot).
SECTOR E: Northeast of Roosevelt Avenue and east of John Street at
the abandoned RR tracks Carteret.
On Pershing Avenue next to Boro Hall or at the sports
field.
SECTOR F: Off Tremly Point Road at intersection to BP storage tanks
Tremely Point (unused area).
Off Roosevelt Avenue leading northeast to the New Jersey
Central RR tracks about 400 meters east of Exit 12 on New
Jersey Turnpike, Carteret. (open area).
South of Roosevelt Avenue and east of New Jersey Turnpike
at an apartment house development, Carteret.
SECTOR G: Off Tremley point road at sharp turn in road next to storage
tanks, (open area south of road).
Grasselli Road east of Turnpike, (site next to RR tracks).
South of intersection between S. Woods Avenue and Tremley
point road just west of Turnpike and across from the American
Cyamaide offices; Tremley (vacant lot).
In Tremley park off S. Woods Avenue.
SECTOR H: North Grasselli road ar fork (RR tracks near landfill).
North of Grasselli road opposite the GAP building.
Each sector will be influenced by a certain wind direction. Below is a
table listing the wind direction, downwind sector and upwind sector.
91
-------
Wind Direction (L)_ Downwind Sector Upwind Sector
360° - 3150 (NNW) D H
315° - 270° (WNW) C G
270° - 225° (wsw) B F
225° - 180° (SSW) A E
180° - 135° (SSE) H D
135° - 90° (ESE) G C
90° - 45° (ENE) F B
45° _ 60° (NNE) E A
The upwind monitors will be required so as to establish a background con-
centration that is independent of the plant.
The downwind monitors will be needed to record the impact due to the plant.
All sites that are located near residential areas will require portable
monitors if that sector is downwind of source. The monitors will record
representative concentrations that people are exposed to.
All sites that are not located near residential areas can be used if the
sector is upwind of the source. The monitors will record the background
concentration independent of the plant contribution.
In summary, with the exception of Sector A, an adequate number of sites
have been selected that will best represent the impact of the plant.
92
-------
The plant is located on Tremley Point in Union County, New Jersey.
Tremley Point lies in a fork created by the Rahway River and the Arthur Kill.
Linden, New Jersey, is 2,000 m to the northwest; Carteret, New Jersey, is
1,500 m to the south-southwest, and Travis, New York is 1,200 m to the east-
southeast.
No NPDES information was requested for this plant as it was not originally
scheduled for sampling. The plant has been known to transport its wastes by
barge for ocean dumping.
The plant is a major producer of acrylamide and polyacrylamide products.
Acrylonitrile is produced for internal consumption in the acrylamide produc-
tion. The plant also produces a variety of pesticides, pesticide products
and surface active agents.
Air Sampling—
Air sampling was conducted using nine battery powered air samplers at six
stations located downwind, upwind, and laterally with respect to the plant.
Samplers were located from about 500 m to 1,500 m from the acrylonitrile
facility and were positioned at a height of 1.5 m.
Sampling was carried out in two segments during the time period from
1500 (EST) February 21, 1978, to 1000 (EST) February 23, 1978. The six
sampling stations are shown in Figure 28 and described in Tables 36 and 37.
The following Figure, No. 29, shows the sectors and potential sampling loca-
tions proposed by E. Chasanoff of Region II. The selected sampling stations
are also noted in this figure.
A duplicate sampler was operated at station A-l. This station was pre-
dicted by Region II as being the point of maximum impact. Air samples were
collected using spiked sampling tubes at this station and at station A-6.
Water Sampling—
A single 3-liter grab water sample was collected February 23, 1978, from
the Region II helicopter. The sample was collected at a point about 30 m
from shore in the Tremley Point Reach of the Arthur Kill about midway in rela-
tion to the large building on the plant property. The sampling location is
shown in Figure 28 and described in Table 38.
Our special thanks goes to Mr. Bill Librizzi who arranged for the heli-
copter used in collecting the water sample.
93
-------
/• *^ifcl3:S-'J
«<'—L^.X'0 \"
01
M T3
%
• Cfl
^ -O
(11
co C
M O
rn
fc-S
13 60
OJ M
13 -H
C «
SM-4
o
CO
>-, (0
U rH
CU
C >
03 CU
O rH
•H
H 4-J
Q) C
| ,
O CO rH
•H 0) CU
•U (0 >
rt 0) -H
O JS 4J
O 4J O
rH (3 CU
CU (X
00 M CO
13 nj CU
•rl P- ri
6 -H
cfl
en
CX3
es
60
•rl
-------
TABLE 36.: AIR SAMPLING DATA FOR AMERICAN CYANAMID, LINDEN, NEW JERSEY 2/21/78-2/22/78
Sample
No.
A-l
A- 2
A- 3
ui A-4
A-5
A- 6
A-6S
Location of
U.S. Geological
survey coordinates
4494000 mN
568850 mE
4493280 mN
568170 mE
4493520 mN
568270 mE
4494710 mN
568020 mE
4493590 mN
565440 mE
4494560 mN
566730 mE
sampling station
Description
Cannon Street and Victory
Avenue (Travis , New York)
West side of Victory Avenue
Glen Street and Melven Street
(Travis-, New York)
North end of Meredith Road
Road through oil storage area
Tremley Point Road and Sea
Explorer Road
Total
sampling
(hr)
21.2
21.2
21.2
21.1
21.1
21.2
20.1
21.0
21.0
Sampling
rate
0.780
1.028
0.758
0.778
0.983
0.885
1.122
1.100
0.973
Total
volume
(m3)
0.992
1.306
0.965
0.983
1.246
1.123
1.357
1.389
1.228
Sampling
height
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
a/ D = Duplicate sample; S = Spiked sample.
-------
TABLE 37. AIR SAMPLING DATA FOR AMERICAN CYANAMID, LINDEN, NEW JERSEY 2/22/78-2/23/78
Sample
No.
A-l
A- ID3/
A-lSa/
A-2
A- 3
A-4
A-5
A-6
A-6S
Location of
U.S. Geological
survey coordinates
4494000 mN
568850 mE
4493280 mN
568170 mE
4493520 mN
568020 mE
4494710 mN
568020 mE
4493590 mN
565440 mE
4494560 mN
566730 mE
sampling station
Description
Cannon Street and Victory
Avenue (Travis, New York)
West side of Victory Avenue
Bridge over 440
Glen Street and Melven Street
(Travis, New York)
North end of Meredith Road
Road through oil storage area
Tremley Point Road and Sea
Explorer Road
Total
sampling
(hr)
18.2
18.2
18.2
19.9
19.9
20.9
18.8
19.4
19.4
Sampling
rate
(£/min)
0.937
1.21
0.769
0.883
1.01
0.797
1.067
1.120
1.170
Total
volume
(m3)
1.022
1.324
0.841
1.055
1.202
0.999
1.205
1.300
1.358
Sampling
height
(m)
1.5
1.5
1-5
1.5
1.5
1.5
1.5
1.5
1.5
a/ D = Duplicate sample; S = Spiked sample.
-------
•••"" ,.j**Z
\ ) ^/- *•*.*• ."
t ;*. (' / ""v • • • • •
»,§ \ ' Ni.y • • • •
V-1-...ixr '.••>• . ..*
Mb ••.•X., '"
• \ InmK*. • • V ' •'.-. 1
•\ • • z. *r-
\ •«•/*? • • -^n
.•• . './• • • . .' \ '
i ^-=^««*' :' ->'•
^ X* •... •-•-;.^\... y .
r/ ^^" 1. / ' V- I : :
t J* x '•^jJi'wr 'j- / • .'i*
I>A X ' * &-/f^
/ -*• /\A *:'£^:Mjr/<'
if
0
1
f
N
500
1
1000
1
Scale - Meters
Figure 29. Sampling sectors defined by Region II, EPA
97
-------
TABLE 38. WATER AND SOIL SAMPLING DATA FOR AMERICAN CYANAMID. LINDEN, NEW JERSEY 2/23/78
Location of sampling stations
Sample
No.
W-l
So-1
So- 2
So- 3
U.S. Geological
survey coordinates
4494050 mN
567450 mE
4493520 mN
568270 mE
4494710 mN
568020 mE
4494560 mN
566730 mE
Description
30 m offshore from large
building
Glen Street and Melven
Street Travis
North end of Meredith Road
Tremley Point Road and Sea
Explorer Road
Sample
type
Grab
water
Soil
Soil
Soil
Sample
size
3 H
200 g
200 g
200 g
% Moisture
by weight
-
-
-
-
oo
-------
Soil Sampling—
Soil samples were collected February 23, 1978, from the area near air
sampling stations A-3, A-4 and A-6 by removing the top approximately 12 mm of
soil. The sampling locations are shown in Figure 28 and described in Table 38.
Meteorological Conditions
The meteorological conditions that existed during the two sampling periods
were obtained from the Newark Airport, Newark, New Jersey. Humidity measure-
ments were made on-site. The data are summarized in Tables 39 and 40 for the
first and second sampling period, respectively.
Sample Analysis
Air Samples—
Air samples were collected on two consecutive days at six sampling sta-
tions. The results of the analysis of the charcoal tubes are given in Tables
41 and 42 for the first and second day of sampling, respectively. The levels
in micrograms per cubic meter are also shown on the map in Figure 28.
Figures 30 and 31 show the wind rose patterns for the two sampling periods.
On the first day, the winds were predominately from the northwest and north-
northwest. Consequently, stations A-2 and A-3 were directly downwind of the
AN source about 85% of the time and A-l was downwind 15% of the time. The
air levels at these sites generally reflect the wind patterns. Low levels
were also found at A-5 and A-6. Station A-6 was upwind of the reported AN
facility but still close to the plant itself. Station A-5 was largely cross-
wind of the AN facility.
On the second day of sampling, the winds shifted slightly to the west.
They were from the northwest and north-northwest 70% of the time and from
the west-northwest 30% of the time. The maximum air levels of AN occurred at
station A-3 while no AN was detected at stations A-l, A-2 or A-5. A low level
of AN, 0.4 jUg/nH was again found at station A-6. The levels found on the
second day were notably lower than those on the first day. This could be
attributed to a disruption in the AN process known to have occurred during the
second sampling period. No AN was found on any of the backup tubes, indicating
that no breakthrough had occurred. The absolute humidity ranged from 1.5 mg/
liter to 3.7 mg/liter during the sampling period.
Sampling quality assurance was checked for each sampling period by the
use of spiked sampling tubes. The spike samplers were deployed at stations
A-l and A-6. Station A-l was the downwind station predicted to be near the
point of maximum impact. Station A-6 was upwind or crosswind of the plant.
99
-------
o
o
TABLE 39. METEOROLOGICAL CONDITIONS DURING SAMPLING AT AMERICAN CYANAMID,
LINDEN, NEW JERSEY, FEBRUARY 21, 1978 TO FEBRUARY 22, 1978:
SOURCE OF DATA, NEWARK, NEW JERSEY AIRPORT
Wind
Time (LST)
2/21/78
1700
1800
1900
2000
2100
2200
2300
2400
2/22/78
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
Direction
(00-36)
31
31
31
32
31
32
33
34
33
34
34
32
33
34
32
29
30
30
31
33
Speed
(knots)
10
11
10
8
11
10
8
7
10
7
9
7
9
5
3
5
10
7
12
13
Temperature Absolute Humidity^/
(°C) (mg/£)
- 1 2.0
- 1
- 2
- 3
- 4
- 4 1.2
-5
- 6
- 6
- 8
- 9
- 9
- .9
- 9
— 9 —
- 9
- 8
- 7 3.7
- 6
- 5
Precipitation
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
-------
TABLE 40. METEOROLOGICAL CONDITIONS DURING SAMPLING AT AMERICAN CYANAMID,
LINDEN, NEW JERSEY, FEBRUARY 22, 1978 TO FEBRUARY 23, 1978:
SOURCE OF DATA, NEWARK, NEW JERSEY AIRPORT
Time (LST)
2/22/78
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2/23/78
0100
0200
0300
0400
0500
0600
0700
0800
0900
Direction
(00-36)
300
310
330
330
310
310
310
310
300
300
310
310
310
310
310
310
300
300
310
300
300
310
310
310
Wind
Speed
(knots)
7
12
15
15
14
12
18
15
16
17
14
17
14
17
13
12
14
11
12
9
12
7
11
12
Temperature Absolute Humidity^/
(°C) (nig/4)
- 7 3.7
- 6
- 5
- 4
- 3
- 3 1.5
- 3
- 4
- 4
- 4
- 4
- 4
- 5
- 5
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4 3.4
- 4
Precipitation
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
][7 Humidity measured on site.
-------
TABLF. 41. ACRYLON1TR1LE CONCENTRATIONS IN AIR SAMPLES FROM
AMERICAN CYANAMTD, LINDEN NEW JERSEY 2/21/78-2/22/78
O
NJ
Sample
No. Sampling period
A-l 2/21/78 1653-
A-1D 2/22/78 1101
A- IS
A-2 2/22/78 1500-
1055
A-3 2/22/78 1508-
1100
A-4 2/22/78 1422
1116
A-5 2/22/78 1615
1204
A-G 2/22/78 1620-
A-6S 1142
A-BS
A-BS
/
KC-S£'
KC-S
A-B
A-B
Air volume
(m3)
1.022
1.324
0.841
1.055
1.202
0.999
1.205
1.300
1.358
_
-
-
-
-
Sample
type5' Comments
Front
Front Duplicate
Front Spiked with 28 pg AN
Front
Front
Front
Front
Front
Front Spiked with 28 |ig AN
Front Spiked with 28 ug AN
Front Spiked with 28 |jg AN
Front Spiked with 28 pg AN
Front Spiked with 28 Mg AN
Front Field Blank
Front Field Blank
Mg Found
0.6
0.6
17.9
16.8
1.9
< 0.1
0.8
0.6
19.0
21.6
21.2
1.8.2
19.5
< 0.1
< 0.1
Mg/m3
0.6^
0.5
Recovery not
calculated
15.9
1.6
< 0.1*'
0.7
0.5
Recovery not
calculated
77% Recovery
76% Recovery
65% Recovery
70% Recovery
< 0.1
< 0.1
GC/MS
Conf Irmationb/
-
-
Yes
Yes
Yes
No
_
-
.
-
-
-
•
"
jJ/ All back-up Lubes were analyzed. No AN was found.
b/ Yes =f Confirmed presence or absence of AN; No = Unconfirmed; Sample
c/ Level not corrected for
recovery.
d/ Based on sample size and instrument
e/ Sample tube spiked and
detection limit.
below detection
limit.
stored at HRI.
-------
TABLE 42. ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM
AMERICAN CYANAHID, LINDEN. NEW JERSEY 2/22/78-2/23/78
O
U>
Sample
No.
A-l
A-1U
A- IS
A- 2
A- 3
A-4
A-5
A-6
A-6S
A-BS
A-BS
KC-S*/
KC-S
A-B1
A-B2
Sampling period
2/22/78 1104-
2/23/78 0815
2/23/78 1055-
0800
2/23/78 1100-
0808
2/23/78 1116-
0825
2/23/78 1206-
0915
2/23/78 1143-
0845
_ _
""" ~
-
-
Air volume Sample
0.992 Front
1 . 306 Front
0.965 Front
0.983 Front
1.246 Front
1.123 Front
1.357 Front
1.389 Front
1.228 Front
Front
Front
Front
Front
Front
Front
Comments pg Found
< 0.1
Duplicate < 0.1
Spiked with 28 pg AN 11.6
(11.6 pg spike
0.0 pg air)
< 0.1
6.5
< 0.1
< 0.1
0.5
Spiked with 28 pg AN 13.0
Spiked with 28 pg AN 17.3
Spiked with 28 pg AN 18.2
Spiked with 28 pg AN 18.2
Spiked with 28 pg AN 19.5
Field blank < 0.1
Field blank < 0.1
CC/MS
pg/nH Conf irmationjl'
< O.L£/
< 0.1
42% Recovery
< 0.1
5.2*'
< 0.1
< 0.1
0.4
Recovery not
calculated
62% Recovery
65% Recovery
65% Recovery
70% Recovery
< 0.1
< 0.1
-
-
_
Yes
_
-
_
-
_
_
-
-
a/ All back-up tubes were analyzed. No AN was found.
b/ Yes = Confirmed presence or absence of AN; No = Unconfirmed; Sample below detection limit.
cj Based on sample size and instrument detection limit.
dj Level not corrected for recovery.
e_/ Sample tube spiked and stored at MRI.
-------
NNW
NW
WNW
WSW
SW
SSW
I
I
Winds Greater Than 10 Knots
Winds Less Than or Equal to 10 Knots
NNE
NE
ENE
SE
SSE
Figure 30. Wind patterns during sampling at American Cyanamid,
Linden, New Jersey, 2/21/78-2/23/78
104
-------
NNW
NNE
NW
NE
WNW
ENE
WSW
SW
SE
SSW
SSE
I Winds Greater Than-10 Knots
I Winds Less Than or Equal to 10 Knots
Figure 31. Wind patterns during sampling at American Cyanamid,
Linden, New Jersey, 2/22/78-2/23/78
105
-------
On each sampling day, two blank tubes were spiked at the site and immediately
stored on dry ice. Two identical charcoal tubes that had been left at Kansas
City were also spiked and stored on dry ice. On analysis of the tubes, it
was found that AN was present in the ambient air at low levels at both spike
sites A-l and A-6 on the first day and at A-6 on the second day. For the rea-
sons previously discussed, recovery of the spikes was not calculated where AN
was found in the air. The recovery of spike A-l on the second day was 42%.
The recoveries for the blanks spiked on site were 77 and 76% for the first day
and 62 and 65% for the second day. The recoveries for the tubes spiked in
Kansas City were 65 and 70%. The levels of AN at the duplicate samples, 0.5
and 0.6 /ig/m , on day 1 were too low to be considered as a measure of precision.
Water Samples—
The one water sample was analyzed by the azeotropic distillation proce-
dure and a level of 0.8 ppb was found. The water sample was also spiked at
MRI at 10 ppb and reanalyzed by the same technique. An 84% recovery of the
added AN was achieved. The results are given in Table 43.
Soil Samples—
The soil samples were extracted with water and the extracts were analyzed
by direct aqueous injection. The results are listed in Table 44. No AN was
found above the 50 M8/kg detection limit.
Gas Chromatography/Mass Spectrometry Confirmation
The confirmation of AN in some of the samples was done by GC/MS using
high resolution SIM mode.
The presence of AN in air samples A-2 and A-3 collected in the first
sampling period and in air sample A-3 collected on the second day was con-
firmed. The AN in air sample A-5 (day 1) was too low to be confirmed. The
absence of AN in upwind air sample A-4 was confirmed. The SIM plots are
shown in.Figure 32.
MONSANTO, TEXAS CITY, TEXAS
Field Sampling
Presampling Survey—
A presampling survey of the Monsanto plant, Texas City, Texas, was con-
ducted on February 27, 1978. The plant is located within the city limits of
Texas City, Texas. The plant is bounded on the north by an urban area, on the
west by an urban and industrial area, on the south by an industrial area, and
on the east by the Texas City Ship Channel. Other chemical production facili-
ties in the area are owned by Amoco and Union Carbide. The NPDES permit lists
six discharge points; one to Moses Lake and five to the Texas City Ship Channel.
106
-------
TABLE 43. ACRYLONITRILE CONCENTRATION IN WATER SAMPLE FROM AMERICAN
CYANAMID, LINDEN, NEW JERSEY; AZEOTROPIC DISTILLATION
Sample Initial Final jLtg -Found GC/MS
No. volume volume (original sample) /xg/l Confirmation
W-l 500 10 0.4 0.8
a/
W-1S- 500 10 4.2 8.4 (84% recovery)
ja/ W-l spiked with 5 fig AN.
TABLE 44. ACRYLONITRILE CONCENTRATIONS IN SOIL SAMPLES FROM
AMERICAN CYANAMID, LINDEN, NEW JERSEY
Sample
No.
So-1
So-2
So-3
Sample
size
10 g
10 g
10 g
Extraction
volume
25 ml
25 ml
25 ml
ptg Found
(original sample)
DAIGCSy
< 0.5
<0.5
<0.5
Ug/kg
DAIGC3-'
< 50
<50
<50
a/ Direct aqueous injection gas chromatography.
107
-------
6 Monograms Acrylonitrile
X200
V»
r
2
I
3
A 5
MINUTES
I • .' I • • I
818
Air Sample A-2, 1st Day
X200
I ' ' I ' ' \ ' ' I
2345
MINUTES
I
8
Air Sample A-3, Isr Day
X200
I
3
3 i 2 a A
f-
56789
MINUTES
Figure 32. SIM plots of ~6 ng AN and air samples A-2(l), A-3(l),
and A-3(2)
108
-------
Air Sample A-4, 1st Day
X400
<> 5
MINUTES
I
1
I
9
Air Sample A-3, 2nd Day
X200
4567
HINUTCS
Figure 32.(Concluded).
109
-------
The locations are shown in Figure 33. This plant produces acrylonitrile,
plastics, plasticizers, resins, styrene, elastomers, ABS resins, butadiene-
acrylonitrile latexes, polybutadiene-acrylonitrile/PVC latexes, and PVC com-
pounds .
Air Sampling—
Air sampling was conducted using 10 battery powered air samplers at 7
sampling stations located upwind, downwind, and laterally with respect to the
plant. Samplers were located from 0 km to 2 km from the plant perimeter along
right-of-ways and shore lines and were positioned at a height of 1.5 m.
Sampling was carried out during the period 1400 (CST), February 28, 1978,
to 1500 (CST), March 1, 1978. The sampling locations are shown in Figure 33
and described in Table 45.
Water Sampling—
Grab water samples were taken at three locations in the Texas City Ship
Channel. The first was taken at a point about 400 m north of outfall 002 and
about 25 m offshore. The second was taken near outfall 002 about 25 m off-
shore. The third was taken near the west end of the seatrain north slip.
These locations are shown in Figure 33 and described in Table 46.
Soil and Sediment Sampling—
Two soil samples were taken at air sampling stations A-2 and A-4 by re-
moving approximately 12 mm of topsoil near the samplers. Sampling locations
are shown in Figure 33 and described in Table 46. Due to the rocky nature of
the bottom of the ship channel we were unable to obtain a sediment sample.
Meteorological Conditions
The meteorological conditions that existed during the sampling period
were obtained on-site using two portable meteorological stations to continu-
ously monitor wind data and a sling psychrometer to measure humidity. Addi-
tionally, a record of weather observations from the National Weather Service
at Galveston Airport was obtained. The portable meteorological stations were
established at air sampling stations A-l and A-5. The meteorological data are
summarized in Table 47.
Sample Analysis
Air Samples—
Air samples were collected at seven sampling stations from 1400 hr,
February 28, 1978, to 1400 hr, March 1, 1978. The results of the analysis of
the charcoal tubes are given in Table 48. The AN levels in micrograms per
cubic meter are also given for the respective stations shown in Figure 33.
110
-------
Figure 33. Sampling locations at Monsanto, Texas City, Texas. Numbers in
parenthesis represent levels of AN in air (yg/m^).
Ill
-------
TABLE 45. AIR SAMPLING DATA FOR MONSANTO, TEXAS CITY, TEXAS
Sample
No.
A-l
A- 2
A- 3
A-4
A- 5
A-5D^-'
A-5S*/
A- 6
A-6D*/
A- 7
Location of
U.S. Geological
survey coordinates
3252860 mN
316450 mE
3249430 mN
314460 mE
3249430 mN
316350 mE
3251730 mN
318950 mE
3250850 mN
316980 mE
3251000 mN
314330 mE
3251380 mN
315730 mE
sampling stations
Description
Beach area east of Holiday
Inn.
Western portion of barge
canal.
Near junction of Snake
Island and Mainland.
2.75 km east of Bay Street
on Texas City Dike.
Northwest portion of Snake
Island.
Near corner of 8th Avenue
and 12th Street.
Corner of 4th Avenue and
3rd Street.
Total
sampling
(hrs)
22.2
22.2
22.5
22.5
23.0
23.0
23.0
22.9
22.9
22.5
Sampling
rate
(£/min)
0.861
0.885
0.880
0.916
0.955
1.055
0.945
1.000
1.060
1.030
Total
volume
1.15
1.18
1.19
1.24
1.32
1.46
1.30
1.36
1.46
1.39
Sampling
height
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
a/ D - duplicate sample
S - spiked sample
-------
TABLE 46. WATER AND SOIL SAMPLING DATA FOR MONSANTO. TEXAS CITY. TEXAS
Sample
No.
W-l
W-2
W-3
Location of sampling stations
U.S. Geological
survey coordinates Description
3251880 mN North of outfall 001.
316790 mE
3251480 mN Near outfall 002
316680 mE
3251110 mN Southwest of outfall 005
316230 mE
Sample
type
Grab water
Grab water
Grab water
Sample
size
1 gal.
1 gal.
1 gal.
% Moisture
by weight
_
—
-
So-1 (A-2)-?/ 3249430 mN
314460 mE
.
So-2 (A-4)-' 3251730 mN
318950 mE
Western portion of barge Soil
2.75 km east of Bay Street Soil
on Texas City Dike.
200 g
200 g
a_l Sampling site designation on map (Figure 33).
-------
TABLE 47. WIND CONDITIONS DURING SAMPLING AT MONSANTO, TEXAS CITY, TEXAS:
DATA FROM GALVESTON AIRPORT AND TWO METEOROLOGICAL STATIONS IN THE FIELD
Time
(LST)
Moteia/
a/
Snake Island-
Direction
(00-36)
Speed
(lets)
Direction
(00-36)
Speed
(kts)
Galveston AirportW
Direction Speed
(00-36) (kts)
2/28/78
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
25
20
17
18
18
18
19
19
20
22
27
2
3
7
7
6
6
7
7
4
3
1
19
19
21
20
20
21
22
22
21
21
20
20
21
21
20
21
22
21
10
7
10
9
10
9
8
8
3/1/78
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
29
35
35
34
03
03
04
03
02
01
36
1
6
7
2
8
10
9
9
35
2
2
4
4
4
5
4
4
4
4
4
3
6
5
4
4
6
6
6
6
8
13
11
14
12
12
16
10
10
13
a/
Climatronics wind direction and wind speed stations continuous sampling
recorded on strip chart. Data are average of four 15-min observations.
Galveston Airport hourly observations. Airport is closed during late
evening, early morning.
£/ Variable winds no average possible.
b/
114
-------
A-7
A-B
TABLE 48. ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM
MONSANTO, TEXAS CITY, TEXAS
FEBRUARY 28, 1978 TO MARCH 1. 1978
Sample
No.
A-l
A-2
A- 3
A-4
A- 5
A-5D
A-5S
A-6
A-6S
Sampling
2/28/78
3/1/78
2/28/78
3/1/78
2/28/78
3/1/78
2/28/78
3/1/78
2/28/78
3/1/78
2/28/78
3/1/78
period
1435-
1245
1506-
1318
1500-
1327
1555-
1428
1435-
1335
1510-
1405
Air
volume
(m3)
1.145
1.179
1.192
1.240
1.318
1.456
1.304
1.357
1.458
Sample
type3./
Front
Front
Front
Front
Front
Front
Front
Back
Front
Front
Back
GC/MS
Comments yg Found yg/m3— Confirmation^/
10.1
6.2
2.5
< 0.3
3.4
Duplicate 3.2
Spiked with 12 yg 3.5
1.0
1.2
Spiked with 12 yg 1.8
6.3
8.9
5.2
2.1
< 0.3
2.6
2.2
Recovery not
calculated
0.9
Recovery not
calculated
No
No
_
_
_
-
-
_
-
2/28/78 1540-
3/1/78 1413
1.391
Front
Front
Field blank
3.2
< 0.3
2.3
No
(continued)
-------
TABLE 48. (continued)
Sample
No.
A-B
A-BS
A-BS
KC-S
KC-S
KC-S
Air
volume Sample
Sampling period (m^) typea/
- - Front
- Front
- Front
- Front
- Front
- Front
Comments
Field blank
Spiked with 12 yg
Spiked with 12 yg
Spiked with 12 yg
Spiked with 12 yg
Spiked with 12 yg
yg Found
< 0.3
3.5
3.8
8.0
9.2
8.4
yg/m^/
-
29% recovery
32% recovery
67% recovery
77% recovery
70% recovery
GC/MS
Confirmation^/
-
a/ All backup tubes were analyzed. Those not shown contained < 0.3 yg.
b_/ Not corrected for recovery. Detection limit based on sample size and instrument response.
_c/ Yes = confirmed presence or absence of AN; No = unconfirmed, sample below detection limit.
-------
During the sampling period, the predominant wind directions were: south
and south by southwest - 44%, and north and north by northeast - 33%. The
wind rose for this period is shown in Figure 34. Wind speeds measured on-site
averaged 4 knots during the sampling period. As a result, stations A-l and
A-7 were downwind 44% of the time, stations A-2 and A-3 were downwind 33% of
the time, and station A-5 was generally crosswind. Stations A-4 and A-6 were
predominantly upwind or crosswind of the AN facility. For comparison, the
annual wind rose for hourly observations (Galveston, Texas) follows as Fig-
ure 35.
For several hours prior to sampling, the absolute humidity averaged 18
rag/liter. As determined earlier, favorable sampling conditions require abso-
lute humidities less than 12 mg/liter. Thus, sampling was postponed until
the absolute humidity began to drop. During the sampling period, the absolute
humidity dropped from an initial 15.5 mg/liter to a final of 10.5 mg/liter.
Station A-4, located 2 miles east northeast of the AN facility, was the
only station where AN was not detected (<0.3 /Ltg/m^) . As noted above, this
station was predominantly crosswind during the sampling period. Station A-6,
primarily upwind, showed an AN concentration of 0.9 ug/m-^. The major downwind
station, A-l, showed an AN concentration of 8.9 /ag/m^. Other downwind stations,
A-2, A-3 and A-5, had AN concentrations of 5.2 jug/m , 2.1 /^g/m and 2.4 jug/m^,
respectively. The closest station, A-7, was located at the NW plant perimeter
and was primarily downwind of the facility. The AN concentration at this site
was 2.3 Mg/rn-^. The results suggest that AN is generally present in the vicinity
of the AN facility and concentration is effected slightly by wind conditions.
This general presence of AN is consistent with information supplied by Region
VI, EPA, stating that Monsanto has five storage tanks for AN (not pinpointed)
with floating roofs which emit approximately 2.5 tons/year of AN.—'
Sampling quality assurance was checked by the use of spiked sampling
tubes. The spiked samplers were deployed at stations A-5 and A-6. Station
A-6 was predominantly upwind whereas station A-5 was predominantly downwind
or crosswind. In addition, two blank tubes were spiked at the site and im-
mediately stored on dry ice. Three blank tubes were returned and spiked at
Kansas City. Duplicate sampling at station A-5 was performed as a check of
precision. The results of the analyses of these tubes are also given in
Table 48. Spiked samplers A-5 and A-6 gave poor recoveries and were the only
stations where AN was found in the backup tubes. These samplers were spiked
when the absolute humidity was 15.5 mg/liter. It is known that high humidity
causes migration of AN; therefore, the low recoveries and presence of AN in
the backup tubes at these stations is most likely due to the high humidity
during sampling. Additionally, station A-5 was located on the shore of Snake
Island and was subjected to higher humidities than station A-6 which was
located inland. As would be expected, station A-5 showed a larger migration
of AN through the charcoal tubes. As mentioned above, gross recoveries of
117
-------
NNW
NNE
NW
WNW
WSW
SW
I
NE
ENE
SE
SSW ^ , | . -^ SSE
S
Winds Greater Than 10 Knots
Winds Less Than or Equal to 10 Knots
Figure 34. Wind Patterns during Sampling at Monsanto, Texas City, Texas.
118
-------
NW
WNW
WSW
SW
NNW
SSW
I Winds Greater Than 12mph
I Winds Less Than 12mph
NNE
SSE
NE
SE
ENE
Figure 35. Annual wind rose for Galveston, Texas (1/51 - 12/60)
119
-------
the spikes were not readily calculated but apparently low. This was also
true of the blank spikes which averaged 27% recovery. However, the Kansas
City blank spikes had an average of 75%. Although recoveries were lower than
desired, the results appear to be reproducible. This is evidenced by the
duplicate sampling at station A-5 which had only a 17% range in the results.
During the GC analysis of the air samples, relatively high levels of ad-
ditional components were observed. Xylene and ethylbenzene were tentatively
identified by GC retention time. The nearest sampling site, A-7, contained
approximately 50 /ng xylene/iP and 10 jug ethylbenzene/m-^. For samples A-l,
A-2, A-3 and A-5, xylene and ethylbenzene were roughly quantitated at 20 /ig/
m^ and 4 jiig/m , respectively. For the crosswind sites, A-6 and A-4, xylene
was approximated at 5 fJLg/ra^ and 1 jug/m^, respectively. Ethylbenzene levels
at A-6 and A-4 were less than 1 /xg/m^. The higher levels of xylene and ethyl-
benzene were generally found at sites where elevated levels of AN were found.
This suggests that the plant is contributing the xylene and ethylbenzene air
levels. The presence of xylenes and ethylbenzene was confirmed by GC/MS using
full scans and extracted ion current profiles (see below). Also identified
in the air samples were dichlorobenzenes and toluene.
Water Samples—
The results of the analysis of the three water samples are given in
Table 49. Both the azeotropic distillation technique and the purge and trap
technique were used for the analyses. AN was found in sample W-l at a level
of 0.4 ppb. This water sample was taken approximately 400 m north of outfall
002 (see Figure 33). Samples W-2 and W-3 did not contain AN (<0.1 ppb).
Sample W-3 was spiked at MRI at 9.5 ppb and reanalyzed by the azeotropic dis-
tillation technique. A recovery of 76% of the added AN was achieved.
Soil Samples—
The two soil samples were extracted with water and analyzed by the direct
aqueous injection technique employing the Hall detector. The soil sample lo-
cations are shown in Figure 33 and the analytical results are shown in Table 50.
No detectable AN was found in the soil samples.
Gas Chromatography/Mass Spectrometry Confirmations—
The presence of AN in air samples A-l and A-2 was not confirmed by high
resolution GC/MS-SIM, although they had been found to contain 8.9 fJig/ra? and
5.2/ig/m^, respectively, by GC/FID. The inability to confirm AN in these
samples might have been due to the fact that GC/MS analysis was performed 23
days after GC/FID analysis. Although lab samples have exhibited acceptable
storage stabilities for up to 24 days, the stability of AN in samples A-l and
A-2 must be considered unknown due to the possible presence of other compounds.
Confirmation of AN in water sample W-l via GC/MS-SIM was also not successful.
120
-------
TABLE 49. ACRYLONITRILE CONCENTRATIONS IN WATER SAMPLES FROM
MONSANTO, TEXAS CITY. TEXAS: AZEOTROPIC DISTILLATION
Sample
No.
W-l
W-2
W-3
W-35^-7
Initial
volume
500 ml
500 ml
500 ml
500 ml
Final
volume
10 ml
10 ml
10 ml
10 ml
ug Found
(original sample) ug/£
0.2 0.4
< 0.06 < O.I-3-/
< 0.06 < 0.1—
3.6
(76% recovery)
GC/MS
confirmation-
No
No
No
a/ Based on instrument detection limit and sample size.
b/ Yes = confirmed presence or absence of AN; No = unconfirmed, below detection limit.
c/ Spiked with 4.75 ug AN
-------
TABLE 50. ACRYLONITRILE CONCENTRATIONS IN SOIL SAMPLES FROM MONSANTO. TEXAS CITY, TEXAS
yg Found
Sample
No.
So-1
So-2
Sample
size
10 g
10 g
Extraction
volume
25 ml
25 ml
(original
sample) /
DAIGC^
< l^
< 1
r
pg/kg/
a/
< 100
< 100
a^f Direct aqueous injection gas chromatography.
b/ Based on instrument detection limit and sample size.
ho
NJ
-------
The presence of other selected organics in the air and water samples was
tested by the use of extracted ion current profiles. The compounds found in
the air samples A-l and A-7 are listed in Table 51 along with the EICP data
used to confirm their presence. The following compounds were shown not to be
present: propionitrile, isopropylnitrile, vinylacetonitrile, methacrylonitrile,
ri-butyronitrile, pyruvonitrile, lactonitrile, 2-methylbutylnitrile, 2-methyl-
propylnitrile, acrylamide, acrolein, styrene, chlorobenzene, trimethyl benzenes,
1,1,2,2-tetrachloroethane, 1,1,2-trichloroethane and benzene. The mass
chromatograms for air samples A-l and A-7 are shown in Figures 36 and 37,
respectively, with the identified compounds noted.
MONSANTO CHEMICAL COMPANY, DECATUR, ALABAMA
Field Sampling
Presampling Survey—
A presampling survey of the Monsanto plant, Decatur, Alabama, was con-
ducted on March 6, 1978. The plant is located approximately 1 km east of
Decatur city limits on a small peninsula bordered by Wheeler Lake on the east,
the Tennessee River on the north and Bakers Creek on the west in Morgan
County, Alabama. Other chemical production facilities and industries in the
immediate area are: Amalgamize Corporation, located due east of the Monsanto
plant on Bakers Creek, 3M Company, located northeast of the Monsanto plant on
the Tennessee River and Alabama Brick and Ceramic Tile Manufacturing located
due west of the Monsanto facility in Decatur city limits.
Air Sampling—
Air sampling was conducted using eight battery powered air samplers at
five sampling stations located upwind, downwind, and laterally with respect
to the plant.
Sampling was conducted during the time periods of 1200 (CST), March 8,
1978, to 1200 (CST), March 9, 1978, and 1230 (CST), March 10, 1978 to 1330
(CST), March 11, 1978. The sampling locations are shown in Figures 38 and 39
and described in Tables 52 and 53.
Winds during the first sampling period were primarily from the west and
northwest. Intermittent heavy rainfall was encountered which wet the charcoal
traps before the samplers could be shut off. These samples were not considered
valid due to the wetting of the charcoal traps. Thus, sampling was repeated
following the termination of rainfall.
123
-------
TABLE 51. COMPOUNDS IDENTIFIED BY GC/MS IN AIR SAMPLES
FROM MONSANTO. TEXAS CITY, TEXAS
Sample
MTX
A-l
MTX
A- 7
Xylenes
m/e = 91,106,105
I = 100,46,17
100, 39, 11^
100,43,12
Ethylbenzene
91,106,51
100,32,13
100,30,13
100,43,8
Dichlorobenzenes
146,148,113
100,62,11
100, 63, -b/
100, 60, -b/
Toluene
91,92
100,65
100,60
100,58
aj The table contains the observed relative intensities of the three m/e values listed and being
used as the criteria for identification. The expected intensities are listed in the heading
with the m/e values.
b/ m/e 113 not measurable due to low signal-to-noise level.
-------
NJ
Ln
5000-
f 000-
3000-
£000-
1000-
a)
C
>
J3
o
U
a) o > -H >,
X! Q X!
i i i i ['i i i i p I i i I r i i i i i i IT i i ri i j i i i i i i i I'l i i i i i i i i i i i i i i i i i i ri i i i r i i i i i i j I'T i i i i i i i i i i ri [ r i'. i i i i i r| ! i i i • ,
8670 8680 8690 8700 8710 8720 8730 8740
SPECS 28670 - 28780 JM MTX fl-1 3UL 0.GOING 3-31-78
Q 7 6 Q
r-r^r^r-r-!
Q7? v 57S:
STEP SPEC#=1 TIC= 100
Figure 36. Mass Chromatogram for Air Sample MTX A-l.
-------
K)
Q) e
C
c
0) J-i 0)
WD
i-H 4
^ N
u C
>-. co a)
5000-
_
~
-
*
f 000-]
-
-
3000-^
-
eooo-
.
^
1 000-^
•*•
_
"1""
4J
\
1
1
\
\
\
\
^ ^-<
O
[V] ^J ^j ^
n c o c
1 . -a en i-i «
^j I C »C i~l
1 tO Q) 0 >,
\ C 0) 0) Q
\ (!) C r
\3 0)
' I | K
r-l i~n p
O >^
H
\
V^ \j
1 | I T l"r rTI'I'l*^"! ! I ! 'I T 1 !"J J i i I-T-I i |n | i i i-i j-r~i i rj i m ~i r~r T T j i r i i i i r i i-j i i i i i i i |
H
>>
s
\
\
v/
\
"^--_
SPEC* S8790 - 2S900 JM MTX fl-7
8830 8840
3UL J.GOING
8850 8860
3-31 -78
8870 8380 8890 8900
STEP SPECtt't TIC= .100
Figure 37. Mass Chromatogram for Air Sample MTX A-7.
-------
'i I
•*, i
3.
- -IS/-
\ 'r\
...o.
, 'OOP'" ' . \
_ W H EELER LAKE
u -
£ .
6 ELEV 556 i , J
S
a
'\
..
(/
5 ".. «;
1
^ ^f».
""^1^
; **&*, 7
. -12/^. ° \
, 7 \ ^~.
•~>,'r-v DKCAq-UH xv
k-4)-1 — '•-^••>p . H ...„ \ n
I
/
pccalur Junction/
8 /
i x
i?/
"'/
I I
i .J*
r 's^^-l ^^;;;sr, VMI. ^f
*^5/\'\f \ '^jysip&-., ^ ^8
i /•?' :.( ^-A-= ! V^^^fe^'^V.^ N ..-'X*0
-' f4. ;'M^>^ >f '^S^N^^^x \ /1"SHR""
!V /A •-•;-• .•• iiV? , V '/S<\>4^N>VS."*- N/
!» •- —A - .....—^^t^ey^WXrs^^fe^v:.^: X _ . .. .
v£vv
^7SV*>
OK^tu- H«t»,^ ^'\^^
\. ^
Figure 38. Sampling locations at Monsanto, Decatur, Alabama on Day 1 (3/8 - 3/9). Numbers in
parenthesis represent levels of AN in air
-------
NJ
00
_
_ ••~ifc«~l - \
j>f~* • J. ' V- . I .^ \
-l Air Sampling Site
l) Water Sampling Site
" 3' '.'. i-'-' *•; - - "vAf^
*-^~—..-:.<* I!a:;-§ ,
WHEELER LAKE
6 ELEV 556 i
\\ q> ,„
:V WHEELER/^fO - l!
: ""1
f ili^-H ; •
- ' r- I' V
^ ^ .J\
Figure 39. Sampling locations at Monsanto, Decatur, Alabama on Day 2 (3/10 - 3/11). Numbers in
parenthesis represent levels of AN in air (pg/m3).
-------
TABLE 52.. AIR SAMPLING DATA FOR MONSANTO. DECATUR. ALABAMA (3/8-3/9)
Site
No.
A-l
A-1S£/
A-2
A-2D3-/
A-2S&7
A-3
A-3D*/
A-4
A- 5
Location
U.S. Geological
Survey coordinates
3833430 mN
503120 mE
3830600 mN
497620 mE
3830720 mN
496820 mE
3831250 mN
500000 mE
3832700 mN
496800 mE
of sampling stations
Description
Swan Creek Waterfowl Manage-
ment Area
500 m east of Beltline Road
on Highway 20
Valving station
Decatur wastewater treatment
plant
Alabama state dock west of
plant
Total
sampling
(hr)
21.6
21.6
23.3
23.3
23.3
22.4
22.4
22.7
23.7
23.7
Sampling
rate
(jC/min)
0.903
0.630
0.741
0.951
0.962
0.886
0.741
0.941
1.000
0.979
Total
volume
(m3)
1.17
0.82
1.03
1.33
1.34
1.19
0.997
1.28
1.42
1.39
Sample
height
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
ay D = ouplicate sample; S = spiked sample.
-------
TABLE 53.. AIR SAMPLING DATA FOR MONSANTO, DECATUR, ALABAMA (3/10-3/11)
LO
o
Site
No.
A-l
A- 2
A- 3
A-4
A-4Da-/
A-4S£/
A-5
A-5Sa/
Location of
U.S. Geological
Survey coordinates
3833430 mN
503120 mE
3830600 mN
497620 mE
3830950 mN
500750 mE
3831250 mN
500000 mE
3832700 mN
496800 mE
sampling site
Description
Swan Creek
Waterfowl management area
500 m east of Beltline Road on
Highway 20
S.W. corner of Neher Street N.W.
and Market Street N.W.
Decatur wastewater treatment
plant
Alabama state dock
west of plant
Total
sampling
Chr)
20.9
23.2
22.6
23.3
23.3
23.3
24.3
24.3
Sampling
rate
(2-/min)
1.305
0.835
1.052
0.963
0.899
1.182
1.293
0.903
Total
volume
(m3)
1.636
1.163
1.426
1.347
1.257
1.652
1.885
1.317
Sample
height
-------
During the second sampling period no rain was encountered, and winds
blew primarily from the south and south-southeast. As a result of the rain-
terminated sampling mentioned above, no back-up charcoal tubes were available
for the second day of sampling.
Water Sampling—
Grab water samples were taken at four locations in the Tennessee River
along the northern border of the Monsanto plant on March 10, 1978. These
locations are shown in Figure 39 and described in Table 54.
The NPDES permit for this plant indicates there are four discharge points
into the Tennessee River. An aerial photograph shows two obvious discharge
points, one of which is located at water sampling site W-3 in Figure 38.
Another discharge point visible in the aerial photograph is located approxi-
mately 150 m upstream (east) of this point. W-l was taken upstream of the
plant. W-2 was taken at the loading dock upstream of the two discharges noted
in the photo. W-3 was taken about 20 m offshore of the farthest west discharge
visible in the photograph. W-4 was taken downstream of the plant. Due to
insufficient information, it was not possible to correlate the NPDES outfall
designations to the aerial photograph or water sampling sites.
Soil and Sediment Sampling—
Two soil samples were obtained on March 11, 1978. Sampling was accom-
plished by removing approximately 12 mm of topsoil from air sampling locations
A-4 and A-5; each sample totaled approximately 500 cm . Soil sample locations
are shown in Figure 38 and described in Table 54.
Sediment samples were obtained on March 10, 1978. Sampling was accom-
plished using an Eckman Dredge at water sampling locations W-2, W-3 and W-4,
and were procured at approximately the same time water samples were taken.
Each sample totaled approximately 500 cm . Sediment sample locations are
shown in Figure 38 and described in Table 54.
Meteorological Conditions
The meteorological conditions that existed during the two sampling periods
were obtained on-site using a portable meteorological station to continuously
monitor wind data and a sling psychrometer for humidity data. The meteorologi-
cal station was established at A-5. The data are summarized in Table 55.
Air Samples—
Ten air samples were collected on the first day of sampling at five
sampling sites. Eight air samples were collected on the second day of sampling
at five sampling sites, four of which were unchanged from the first sampling.
The results of the analysis of all the charcoal tubes are given in Table 56.
The levels in micrograms per cubic meter (/Jg/m3) are also shown in Figure 38
for day 1 and in Figure 39 for day 2.
131
-------
TABLE 54., WATER, SOIL AND SEDIMENT SAMPLING DATA FOR MONSANTO COMPANY. DECATUR, ALABAMA (3-8-78)
Location of sampling site
Sample
No.
W-l
W-2
W-3
W-4
So-1 (A-4)^/
So-2 (A-5)^/
Se-1
Se-2
Se-3
U.S. Geological
Survey Coordinates
3832350 mN
499000 mE
3832350 mN
498500 mE
3832680 mN
497990 mE
3833210 mN
497330 mE
3831250 mN
500000 mE
3832700 mN
496800 mE
3832350 mN
498500 mE
3832680 mN
497990 mE
3833210 mN
497610 mE
Description
Monsanto boat dock located
near sewage disposal plant
Monsanto dock near pumping
station
Effluent discharge point
near Bakers Creek
Downstream near pipelines
crossing
Decatur wastewater treat-
ment plant
Alabama state dock west of
plant
Monsanto dock near pumping
station
Effluent discharge point
near Bakers Creek
Downstream near pipelines
crossing
Sample
type
Grab Water
Grab Water
Grab Water
Grab Water
Soil
Soil
Sediment
Sediment
Sediment
Volume % Moisture
collected by weight
3 £
3 £
3 £
30
X/
500 cm3
500 cm3
500 cm3
500 cm3
500 cm3
a/ Sampling site designation on map (Figure 38).
-------
TABLE 55. METEOROLOGICAL CONDITIONS DURING SAMPLING AT MONSANTO. DECATUR. ALABAMA
10
Co
Wind*/
Date
Day 1
3-08-78
3-09-78
Time
(LST)
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
Direction
00-36
05
06
05
05
09
24
25
27
27
27
27
27
27
27
28
31
31
35
28
30
31
28
32
28
27
28
Speed Temperature^/ Relative
(kts) (°C) (%)
07 18 80
07
06
04 15 89
10
07
10 8 86
08
07
08
06 6 92
06
06
05
05
04 5 92
04
05 -
03
03
05
04
03
03
05
05
Absolute
(rag/*)
12.5
11.4
7.3
6.5
6.3
Ccontinued)
-------
TABLE 55.(continued)
u>
WindaV
Date
Day 2
3-10-78
3-11-78
Time
(LST)
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
Direction
00-36
29
31
27
27
27
21
21
21
20
21
21
27
21
20
18
18
16
16
16
17
17
18
16
-
-
-
Speed
(kts)
07
05
06
06
05
03
03
04
03
03
04
00
02
03
06
07
08
08
08
07
06
07
13
-
-
-
Temperature^/ Relative
(°C) (%)
_ _
7 78
-
-
-
-
8
-
8 72
-
-
-
8 72
-
-
—
8 79
-
^
-
-
-
-
-
13 70
Absolute
(mg/A)
_
5.9
5.9
5.9
6,4
8.9
aj On-site data from meteorological station.
b/ Data taken on-site.
-------
TABLE 56. ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM
MONSANTO, DECATUR, ALABAMA. FOR TWO SAMPLING PERIODS
Sample
No.
Sampling period
Air
volume
(m3)
Sample
type^' Comments
b/ GC/MS
yg Found yg/mj Confirmation—
Day 1
(3/8-3/9)
1-A-l
1-A-1S
l-A-2
1-A-2D
1-A-2S
l-A-3
l-A-4
1-A-4D
l-A-5
1-A-5D
3/8/78 1500-
3/9/78 1235
3/8/78 1240-
3/9/78 1155
3/8/78 1320-
3/9/78 1200
3/8/78 1200
3/9/78 1140
3/8/78 1300-
3/9/78 1125
1.170
0.816
1.03
1.33
1.34
1.28
1.42
1.39
1.19
0.997
Front
Front Spiked with 16 yg
AN
Front
Front Duplicate
Front Spiked with 16 yg
Back
Front
Front
Front Duplicate
Front
Front Duplicate
< 0.1
3.5
4.1
3.3
9.7
1.0
5.4
2.3
1.8
< 0.1
< 0.1
< 0.1
22% recovery
4.0
2.5
Recovery not
calculated
4.2
1.6
1.3
< 0.1
< 0.1
—
Yes
-
Yes
-
Day 2
(3/10-3/11)
2-A-l
2-A-2
3/10/78 1515-
3/11/78 1210
3/10/78 1400-
1.636
1.163
Front^-7
Front^
< 0.3
< 0.3
< 0.2
< 0.3
-
3/11/78 1315
(continued)
-------
TABLE 56. (concluded)
10
Air
Sample volume
No. Sampling period (nH)
2-A-3 3/10/78 1405- 1.426
3/11/78 1240
2-A-4 3/10/78 1325- 1.347
2-A-4D 3/11/78 1245 1.257
2-A-4S 1.652
2-A-5 3/10/78 1245- 1.885
2-A-5S 3/11/78 1300 1.317
A-B - - -
A-B - - -
A-BS - - -
A-BS - - -
A-BS - - -
KC-S - - -
KC-S - - -
Sample
type^/
Front^-7
Front !/
Front!/
Front^-/
Front^-7
Front4-7
Front!/
Front!/
Front!/
Front^-7
Front!/
d/
Front—
j /
ft /
Front!7
Comments
Spiked with 16 pg
Spiked with 16 pg
Field blank
Field blank
Spiked with 16 pg
in field
Spiked with 16 pg
in field
Spiked with 16 pg
in field
Spiked with 16 pg
in Kansas City
Spiked with 16 pg
in Kansas City
b/ GC/MS
pg Found pg/m^ — Conf irmation~
< 0.3
< 0.3
< 0.3
9.5
2.1
12.2
< 0.2
< 0.2
11.2
14.1
12.5
10.1
8.6
< 0.2
< 0.2
< 0.2
59% recovery -
1.1
Recovery not -
calculated
70% recovery
88% recovery -
78% recovery
63% recovery -
54% recovery -
a/ All backup tubes were analyzed. Those not shown contained < 0.1 pg AN.
t>/ Not corrected for recovery. Detection limit based on sample size and instrument response.
c/ Yes = confirmed presence or absence of AN; No = unconfirmed, sample below detection limit.
d/ Backup tubes were not available for sampling on Day 2.
-------
Figures 40 and 41 show the wind rose patterns for the two sampling periods.
For comparison, the annual wind rose from the Star Program (Huntsville, AL)
follows as Figure 42. On the first day, winds were predominantly from the
west and northwest. Consequently, stations A-2 and A-3 were downwind
approximately 21% of the time and station A-4 was downwind, but at a greater
distance, approximately 35% of the time. The air levels at these sites gen-
erally reflect the wind patterns. The highest concentration of AN, 4.2 /zg/m ,
was found at station A-3. AN levels at station A-4 averaged 1.5 jug/m3. The
air levels at the duplicate stations, A-2 and A-2D, were relatively high at
4.0 jug/m and 2.5 fig/m3, respectively. No AN was detected at the upwind sta-
tions A-5 and A-5D (<0.1 /jg/m3). Station A-l, 5,000 m east-northeast from
the AN facility, was the outermost sampling site. Although it was generally
crosswind or downwind about 30% of the time, no AN was detected at this site.
As mentioned earlier, rainfall terminated this sampling period, but not before
the charcoal tubes had become wet.
On the second day of sampling, all sampling sites remained where they
were during day 1 sampling, except station A-3. This station was moved to a
site southeast of the AN facility. During this sampling period, the pre-
dominant wind directions were: south - 34%, south-southeast - 17%, southwest
and south-southwest - 25%, and west - 17%. As a result, stations A-4 and A-5
were near crosswind sites, station A-l was a far crosswind site, and stations
A-2 and A-3 were upwind sites. During this sampling period, only station A-5
contained AN at 1.1 g/m3. The absolute humidity ranged from 5.9 mg/liter to
12.5 mg/liter during the sampling period.
Sampling quality assurance was checked for each sampling period by the
use of spiked sampling tubes, blank field spikes, blank lab spikes, and blank
tubes. The spiked samplers were deployed at stations A-l and A-2 for day 1
sampling. Station A-l was located 5,000 m east-northeast of the AN facility
and was predominantly crosswind and downwind. The recovery of spike A-l was
22%. Station A-2 was located 1,000 m west of the AN facility and was primarily
crosswind. The recovery of this spike was not calculated due to the occurrence
of ambient AN. For sampling conducted on the second day, spiked samplers were
deployed at stations A-4 and A-5. Station A-4 was 1,200 m southeast of the
facility and was generally a crosswind site. Station A-5 was also upwind of
the AN facility, being 800 m west-northwest. The recovery of the A-4 spike
was 59%. The recovery of the A-5 spike was not calculated due to the presence
of ambient AN. Three blank tubes were spiked at Monsanto and immediately
stored on dry ice. The recoveries for the blank spikes were 70%, 88% and 78%.
Two identical tubes were spiked at MRI and stored on dry ice. These recover-
ies were 63% and 54%. No AN or interference was found on the blank tubes.
137
-------
NNW
NNE
NW
WNW
WSW
SW
SSW
SSE
NE
ENE
ESE
SE
I Winds Greater Than 10 Knots
I Winds Less Than or Equal to 10 Knots
Figure 40. Wind Patterns during Sampling at Monsanto, Decatur, Alabama (3/8-3/9),
138
-------
NNW
NNE
NW
NE
WNW
ENE
WSW
SW
SE
SSW
SSE
I
I
Winds Greater Than 10 Knots
Winds Less Than or Equal to 10 Knots
Figure 41. Wind Patterns during Sampling at Monsanto, Decatur, Alabama (3/10-3/11),
139
-------
NNW
NNE
NW
NE
WNW
ENE
WSW
ESE
SW
SE
ssw
• Winds Greater Than 12mph
I Winds Less Than 12mph
SSE
Figure 42. Wind rose.for Huntsville, Alabama (1/60 - 12/64) using Star
program output
140
-------
Water Samples—
The four water samples were first analyzed by the azeotropic distillation
(AD) technique. The three water samples which were positive for AN were then
reanalyzed by direct aqueous injection. The results are given in Table 57.
Sample W-l, collected upriver from the plant, contained no detectable AN by
AD. Sample W-2, collected near the dock, was found to contain 180 ppb AN by
AD and 220 ppb by direct injection. The sample collected at the outfall, W-3,
was found to contain 3,500 and 3,800 ppb AN by AD and direct aqueous injection,
respectively. Water sample W-4, collected about 800 m offshore and downstream
of W-3, was found to contain 19 and 48 ppb AN by AD and direct aqueous injec-
tion, respectively. Agreement between the two techniques was good for the
samples W-2 and W-3 which contained the higher levels of AN. The agreement
for sample W-4 was not as good, however. The level of AN was close to the
detection limit for direct aqueous injection.
Sample W-4 was spiked at 9.5 ppb AN by the addition of 4.75 /ng AN fol-
lowing the first distillation. Additional methanol was added and the process
repeated. The analysis of the distillate showed a recovery of 78%.
Soil and Sediment Samples—
Two soil samples and three sediment samples were extracted with water and
the extracts analyzed by the direct aqueous injection technique. No detect-
able AN was present in any of the samples. These results are given in Table 58.
Gas Chromatography/Mass Spectrometry Confirmation—
The presence of AN in day 1 air samples A-4 and A-2D was confirmed.
Figure 43 gives the high resolution SIM plot for these samples as well as for
a 4-ng standard of AN.
The presence of AN in water samples W-2 and W-3 was confirmed by GC/MS
using extracted ion profiles. Sample W-4 was found to contain AN by GC
analyses but its presence could not be confirmed by GC/MS as it was below the
limit of detection. The extracted ion current profiles for these samples and
a 5-ng AN standard are shown in Figure 44.
Day 2 air sample A-4 was analyzed by GC/MS using the full scan mode and
EICP's for 21 possible compounds. Day 2 air sample A-4 did not contain AN but
was selected for additional analysis because it contained higher levels of
"other" compounds. By this method, air sample A-4 was found to contain isomers
of xylene, dichlorobenzene, and toluene. The EICP data are listed in Table 59.
Not found was propionitrile, isopropylnitrile, vinylacetonitrile, methacrylo-
nitrile, ii-butyronitrile, pyruvonitrile, lactonitrile, 2-methylbutylnitrile,
2-methylpropylnitrile, acrylamide, acrolein, ethylbenzene, styrene, chloro-
benzene, trimethylbenzenes, 1,1,2,2-tetrachloroethane, 1,1,2-trichloroethane
and benzene. The mass chromatogram and identified compounds are shown in
Figure 45. Water samples W-2, W-3 and W-4 were also analyzed.
141
-------
TABLE 57. ACRYLONITRILE CONCENTRATIONS IN WATER SAMPLES FROM MONSANTO. DECATUR. ALABAMA
Azeotropic distillation
Sample
No.
W-l
W-2
W-3
W-4
W-4S-/
Initial volume
(ml)
500
500
500
500
500
Final volume
(ml)
10
10
10
10
10
yg Found
(original sample)
0.06
89
1,750 3
9.5
3.7 (78% recovery)
DAIGCb/ GC/MS
yg/2. yg/£ confirmation0-'
< 0.1
180
,500
19
-
fc _.
218 Yes
3,800 Yes
48 No
-
a/ Based on instrument detection limit and sample size.
b/ Direct aqueous injection chromatography
cj Yes = confirmed presence of AN; No = unconfirmed, sample below detection limit.
d_/ W-4 spiked with 4.75 yg AN following first distillation.
-------
TABLE 58.: ACRYLONITRILE CONCENTRATIONS IN SOIL AND SEDIMENT SAMPLES FROM MONSANTO,
DECATUR, ALABAMA
Sample
No.
SO-1
So-2
Se-1
Se-2
Se-3
Sample
size
10 g
10 g
10 g
10 g
10 g
Extraction
volume
25 ml
25 ml
25 ml
25 ml
25 ml
yg Found
(original sample)
DAIGC3-/
< 0.5
< 0.5
< 0.5
< 0.5
< 0.5
DAIGC-/
- -,u-
< 50
< 50
< 50
< 50
a/ Direct aqueous injection gas chromatography.
b/ Based on instrument detection limit and sample size.
-------
Four Nanogrami Acrylonitrile
x 2000
m/e -53.0265-
4 5
Minutes
Air Sample A-4
x 2000
I I
_A
^TJllv-/^M
T I I
i HI ft
1 IT
4 5
Minutes
Air Sample A-2D
x 2000
^u^^^
LrL*L-"r1' J\
^«^-m^/X^
•n HUTIlT
4
Minutes
Figure 43. SIM Plots for Monsanto, Decatur, Alabama
Air Samples A-2D and A-4 Along with a 4-ng Acrylonitrile Standard.
144
-------
ii
Figure 44. Extracted Ion Current Profiles for Monsanto, Decatur, Alabama
Water Samples W-2, W-3 and W-4 along with a 5-ng AN Standard.
145
-------
TABLE 59. COMPOUNDS IDENTIFIED BY GC/MS IN AIR SAMPLES
FROM MONSANTO, ALABAMA
Sample
MAL
A-4
Xylenes
m/e = 91,106,105
I = 100,46,47
100,50,17-
Dichlorobenzenes
146,148,113
100,62,11
100, 65, -b/
Toluene
91,92
100,65
100,69
a,/ The table contains the observed relative intensities of the three
m/e values listed and being usee as the criteria for identifi-
cation. The expected intensities are listed in the heading with
the m/e values.
b/ m/e 113 was measureable due to low signal-to-noise level.
146
-------
5000-,
tooo-
£000-
1 000-
• i ri i P'l 1111 f i f 11»i n 111 111 ri 111 ri 11111111111111 11 [ i !•< i'| 11111 r 11 r| 11 i'i [ i i
8550 8560 8570 8580 8590 8600 8610 8620
SPECS £8550 - £8660 JM MRL fl-4 3UL J.GOING 3-31-78
'"i""i""i I' I
8630 86tO 8650 8660
STEP SPE.C«=l TIC' 100
Figure 45. Mass Chromatogram Showing Additional Compounds Identified in
Monsanto, Decatur, Alabama Air Sample A-A.
-------
The samples did not contain any of the other compounds considered. Propio-
nitrile was found in W-2, W-3 and the azeotropic distillate of W-2. The EICP's
of these samples are shown in Figure 46. The observed intensity ratios for
the ions 54, 26 and 52 were 100:12:29 and 100:17:28 for samples W-2 (azeotropic
distillate) and W-3, respectively. This compares reasonably well with the re-
ported ratios of 100:34:18 found in the Eight Peak Index of Mass Spectra. The
identification was also supported by the agreement of the retention times
relative to AN. The relative retention times for the samples were 1.24 and
the expected relative retention time was 1.20.
DUPONT, LUGOFF, SOUTH CAROLINA
Field Sampling
Presampling Survey—
A presampling survey of the DuPont-May plant, Kershaw County, South
Carolina, was conducted on March 22, 1978. The plant is located directly
southeast of the small community of Lugoff, South Carolina, and approximately
3 km west of Camden, South Carolina. The facility is bordered by the Jefferson
Davis Highway (Route 1) on its northern boundary, the Wateree River to the
east, the Seaboard Air Line Railroad to the south, and Kershaw County Road,
S 28.38, to the west. Only one other chemical production facility was noted
in the immediate area, Union Carbide, located at the intersection of County
Road S 28.133 and the Seaboard Air Line Railroad right-of-way on the northeast
corner.
According to the NPDES permit, wastewater is discharged into the Wateree
River at two points, 001B and 002. The maximum flow rate from 001B is 8 mgd
with 633 kg/day daily average suspended solids and 1,784 kg/day daily average
kjeldahl nitrogen. The plant produces acrylic, modacrylic, polyamide and
polyester fibers.
Air Sampling—
Air sampling was conducted using 11 battery powered air samplers at eight
sampling locations located in a circular array with respect to the plant.
Sampling was conducted during the period 1045 hr (EST), March 22, 1978,
to 1015 hr (EST), March 24, 1978. The sampling locations are shown in Figure
47 and described in Table 60.
Wind conditions were light and quite variable during the first 24 hr;
therefore, sampling was continued an additional 24 hr to assure an adequate
sample.
148
-------
30.000-,
20,000-
*• z
VO —
I 10.000-
Woter Sample W-2
10 15
I
20
Water Sample W-2
(Azeoliopic Oislillale)
10
15 20
Watei Sample W-3
\
10
T
VJL,
26
15 20
MINUTES
Figure 46. Extracted Ion Current Profiles of Propionitrile Specific Ions
For Monsanto, Decatur, Alabama Water Samples.
-------
• THE ARMY
G1NEERS
XtONorowN i4 Mr
RAUO* r.HOSSHQAOS 6 f^l
•- &
|A-11 Air Sampling
Station
Water Sampling
Station
Figure 47. Sampling locations at DuPont, Lugoff, South Carolina. Numbers
in parenthesis represent levels of AN in air (yg/m ).
150
-------
TABLE 60. AIR SAMPLING DATA FOR PUPONT. LUGOFF, SOUTH CAROLINA
Location of sampling stations
Sample
No.
A-l
A- 2
A-2D£/
A- 3
A- 4
A- 5
A- 6
A-6S£/
A- 7
A- 8
A-8S*/
U.S. Geological
survey coordinates
3789040 mN
531770 mE
3786250 mN
532550 mE
3786450 mN
531380 mE
3787100 mN
530850 mE
3788830 mN
533880 mE
3787350 mN
533080 mE
3787840 mN
530470 mE
3788400 mN
531000 mE
Description
West end of Howard
F. Speaks Bridge
1,250 m east of
S28.133 and S28.38
S28.133 and S28.38
On Ward Drive near
railroad
Wateree Blvd. and
Mitchell Drive
1,200 m NNW of A- 2
Corner of Burdell
Road and US-1
Near junction of
US-1 and Co. Rd. 5
Total
sampling
(hr)
46.9
45.7
45.4
45.7
46.1
44.5
46.5
46.2
47.3
47.3
Sampling
rate
(H/rain)
1.05
1.05
0.890
1.05
0.966
1.02
0.857
0.951
0.734
1.04
Total
volume
(m3)
2.95
2.87
2.43
2.88
2.67
2.73
2.39
2.63
2.09
2.96
Sampling
height
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
a] D = duplicate sample; S = spiked sample.
-------
Water Sampling—
Grab water samples were taken at four locations along the eastern DuPont
plant border and the Wateree River on March 23 and 24, 1978. These locations
are shown in Figure 47 and described in Table 61.
Soil and Sediment Sampling—
Four soil samples were obtained on March 24, 1978. Sampling was accom-
plished by removing approximately 12 mm of topsoil for air sampling locations
A-l, A-3, A-5 and A-8; each sample totaled approximately 500 cm-*. Soil sample
locations were essentially taken at four compass points in relation to the
plants. The exact locations are indicated in Figure 47 and described in
Table 61.
Sediment samples were obtained on March 23, 1978. Sampling was accom-
plished using an Eckman Dredge at water sampling locations W-l, W-3 and W-4.
Each sample totaled approximately 500 cm . Sediment sample locations are
shown in Figure 47 and described in Table 61.
Meteorological Conditions
The meteorological conditions that existed during the sampling period
were obtained on-site using a portable meteorological station to continuously
record wind data and a sling psychrometer to record humidity data. The port-
able meteorological station was located at air sampling station A-8. The
data are summarized in Table 62.
Sample Analysis
Air Samples—
Eleven air samples were collected at eight sampling stations around the
DuPont-May acrylonitrile facility. Table 63 contains the results of the analy-
sis of the charcoal tubes. In addition, the levels in micrograms per cubic
meter are given in Figure 47. During the sampling period, winds were very
calm, averaging less than 1 knot, as shown by the wind rose in Figure 48
and the data in Table 62. For comparison, the annual wind rose from hourly
observations (Camden, South Carolina) follows as Figure 49. The predominant
winds were from the north (26% of the time). Due to these wind conditions,
none of the sampling stations could be designated as upwind or downwind. AN
was found at all sampling stations, except A-6 at levels between 0.1 and 1.1
/j,g/m . AN was not detected at station A-6.
Sampling quality assurance was checked by the use of spiked samplers at
stations A-6 and A-8 and by the use of spiked blank tubes. The recovery for
the spiked samples was inexplicably low while the recovery for spiked blanks
was abnormally high (100-161%). This is in sharp contrast to the results of
the four previous sites. No explanation is apparent although it is felt that
the problem more likely resides with the spiking solution than the sampling
tubes.
152
-------
TABLE 61. SOIL, WATER, AND SEDIMENT SAMPLING DATA, DUPONT, U1GOFF, SOUTH CAROLINA
Sample
No.
So-1 (A-l)a/
So-2 (A-3)a/
So-3 (A-5)!/
So-4 (A-8)^
W-l
W-2
W-3
W-4
Se-1 (W-l)£/
Se-2 (W-3)5/
Se-3 (W-4)£/
Location
U.S. Geological
survey coordinates
3789040 mN
531770 mE
3786450 mN
531380 mE
3788830 mN
533880 mE
3788830 mN
531000 mE
3789170 mN
531860 mE
3788730 mN
532240 mE
3788540 mN
532530 mE
3788240 mN
532790 mE
3798170 mN
531860 mE
3788540 mN
532530 mE
3788240 mN
532790 mE
of sampling stations
Sample
Description type
West end of Howard F. Speaks Bridge Soil
S28.133 and S28.31 Soil
Wateree Blvd. and Mitchell Drive Soil
Near Junction of US-1 and Co. Rd. 5 Soil
West end of Howard F. Speaks Bridge Grab water
About 500 m. south nf Howard F. Speaks Grab water
Bridge
About 900 m. south of Howard F. Speaks Grab water
Bridge
Near Old Ferry Road Grab water
West end of Howard F. Speaks Bridge Sediment
About 900 m. south of Howard F. Speaks Sediment
Bridge
Near Old Ferry Road Sediment
Sample % Moisture
size by weight
500 cc
500 cc
500 cc
500 cc
1.5 i.
1.5 t
1.5 i
1.5 I
500 cc
500 cc
500 cc
a/ Sampling site designation on map (Figure 47).
-------
TABLE 62. METEOROLOGICAL CONDITIONS DURING SAMPLING AT DUPONT. LUGOFF, SOUTH CAROLINA
Time
Date (LST)
3/22/78 1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
3/23/78 0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
Wind
Direction
(00-36)
04
36
045
14
04
06
02
06
04
34
31
34
35
35
02
02
34
36
14
18
32
12
05
27
14
32
32
21
17
. 24
Speed
(kts)
05
04
03
03
03
02
01
< 01
< 01
< 01
< 01
< 01
< 01
< 01
< 01
< 01
< 01
< 01
01
< 01
< 01
< 01
01
02
01
02
01
01
02
02
Humidity
Temperature Relative Absolute
(°C) (%) (rag/ JO
- - -
- - -
24 30 6.5
_
_
_
17 46 6.7
_
_
_
_ _ _
15 50 6.4
_
_ _ _
_
- - -
9 80 7.3
_
_ _ _
_
_
15 78 10.0
_
_ _ _
21 63 11.3
_
_ _ _
_
_
(continued)
-------
TABLE 62.(continued)
Wind
Time
Date (LSI)
1700
1800
1900
2000
2100
2200
2300
2400
3/24/78 0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
Direction
(00-36)
18
21
22
35
36
32
34
04
33
30
33
31
13
35
34
36
03
15
Speed Temperature
(kts) (°C)
01
01 19
01
01
< 01
01
03 17
01
< 01
< 01 13
< 01
< 01
< 01
< 01
01
01
01
01 19
Humidity
Relative Absolute
(%) Gng/ A)
53 8.9
-
-
-
-
70 9.9
- -
_ -
76 8.5
-
-
-
-
-
-
-
71 11.9
-------
TABLE 63. ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM DUPONT-MAY. LUGOFF, SOUTH CAROLINA
Sample
No.
A-l
A- 2
A-2D
A- 3
A- 4
A- 5
A- 6
A-6S
A- 7
A- 8
A-8S
A-BS
A-BS
KC-S
KC-S
Air
Sampling volume
period (m^)
3/22/78
3/24/78
3/22/78
3/24/78
3/22/78
3/24/78
3/22/78
3/24/78
3/22/78
3/24/78
3/22/78
3/24/78
3/22/78
3/24/78
3/22/78
3/24/78
-
-
-
-
1220- 2.951
1015
1145- 2.875
0930 2.45
1135- 2.880
0920
1115- 2.673
0920
1230- 2.732
0900
1155- 2.390
0935 3.021
1105- 2.635
09.5
1045- 2.086
1000 2.959
-
-
-
-
Sample
typejl/
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Mg
Comments Found
1.0
0.7
Duplicate 0.7
0.6
0.3
0.3
< 0.3
Spiked with 13 yg AN 0.3
0.8
2.3
Spiked with 13 yg AN 4.6
Spiked with 13 yg AN 21.0
Spiked with 13 yg AN 7.9
Spiked with 13 yg AN 16.2
Spiked with 13 yg AN 13.0
b/ GC/MS
yg/m^ Confirmation^-'
0.3
0.2
0.3
0.2
0.1
0.1
< 0.1
Recovery not -
calculated
0.3
1.1 Yes
Recovery not
calculated
161% Recovery
61% Recovery
125% Recovery
100% Recovery
(continued)
-------
TABLE 63.(continued)
Sample
No.
A-B
A-B
Air
Sampling volume Sample
period (m3) type Comments
- - - Front Field Blank
- Front Field Blank
yg GC/MS
Found yg/m3 Conf irmation£/
< 0.3
< 0.3
a/ All backup tubes were analyzed. Those not shown contained < 0.3 yg.
b_/ Not corrected for recovery. Detection limit based on sample size and instrument response.
c/ Yes = confirmed presence or absence of AN; unconfirmed, sample below detection limit.
-------
NNW
NW
WNW
WSW
SW
SSW
I
I
Winds Greater Than 10 Knots
Winds Less Than or Equal to 10 Knots
NNE
NE
ENE
ESE
SE
SSE
Figure 48. Wind patterns during sampling at DuPont-May, Lugoff, South Carolina.
158
-------
NNW
NNE
NW
NE
WNW
ENE
WSW
ESE
SW
SE
SSW
SSE
• Winds Greater Than 12mph
I Winds Less Than 12mph
Figure 49. Wind rose for Columbia, South Carolina (1/51 - 12/60)
159
-------
Water Samples—
The results for the analysis of the four water samples by azeotropic dis
tillation are given in Table 64. The upriver sample W-l contained no detect-
able AN. Samples W-2 and W-3, taken from plant outfalls, contained 1.1 fj.g/
liter and 19.7 /ug/liter, respectively. The downriver sample contained 0.4 /ng
liter.
Soil and Sediment Samples—
Four soil samples and three sediment samples were extracted with water
and analyzed by direct aqueous injection. The results are listed in Table 65.
No AN was found above the 50 Mg/kg detection limit.
Gas Chromatography/Mass Spectrometry Confirmation—
Air sample A-8 was found to contain 1.1 jug/m-' AN by GC analysis. Con-
firmation was accomplished by high resolution GC/MS-SIM. Figure 50 gives the
SIM plot for air sample A-8 and a 4-ng standard of AN.
The presence of AN in water sample W-3 was confirmed by GC/MS using full
scan with extracted ion current profiles. The EICP's are shown in Figure 51
along with 5-ng AN standard.
Selected samples were analyzed for additional compounds using the full
scan mode and EICP's for 21 possible compounds.
Air sample A-8 contained toluene, isomers of xylene, and diclorobenzene.
Not found were propionitrile, isopropylnitrile, vinylacetonitrile, methacrylo
nitrile, n-butyronitrile, pyruvonitrile, lactonitrile, 2-methylbutylnitrile,
2-methylpropylnitrile, acrylamide, acrolein, chlorobenzene, ethylbenzene,
styrene, trimethylbenzenes, 1,1,2,2-tetrachloroethane, 1,1,2-trichloroethane
and benzene. The mass chromatogram of air sample A-8 is shown in Figure 52
with the identified compounds noted above the chromatogram. The EICP data
are given in Table 66. None of the tested compounds were found in the water
sample.
DUPONT, WAYNESBORO, VIRGINIA
Field Sampling
Presampling Survey—
A presampling survey of the DuPont plant, Waynesboro, Virginia, was con-
ducted on March 27, 1978. The plant is located in Waynesboro, Virginia, and
is bordered on the north and east by a residential area, on the south by a
Thiokol facility manufacturing carpet backings, and on the west by the South,
River.
160
-------
TABLE 64. ACRYLONITRILE CONCENTRATIONS IN WATER SAMPLES FROM DUPONT-MAY, LUGOFF,
SOUTH CAROLINA: AZEOTROPIC DISTILLATION
Sample
No.
W-l
W-2
Initial
volume
500 ml
500 ml
Final
volume
10 ml
10 ml
yg Found
(original sample)
< 0.06
0.54
GC/MS
yg/£ Confirmation^./
< O.ia/
1.1
W-3
500 ml
10 ml
9.87
19.7
Yes
W-4
500 ml
10 ml
0.21
0.4
_§_/ Based on instrument detection limit and sample size.
b/ Yes = confirmed presence or absence of AN; No = unconfirmed, sample below detection limit.
-------
N)
TABLE 65. ACRYLONITRILE CONCENTRATIONS IN SOIL AND SEDIMENT SAMPLES FROM DUPONT-MAY, LUGOFF,
SOUTH CAROLINA
Sample
No.
So-1
So-2
So-3
So-4
Se-1
Se-2
Se-3
Sample
size
10 g
10 g
10 g
10 g
10 g
10 g
10 g
Extraction
volume
25 ml
25 ml
25 ml
25 ml
25 ml
25 ml
25 ml
yg Found
(original sample)
DAIGC£/
< 0.5
< 0.5
< 0.5
< 0.5
< 0.5
< 0.5
< 0.5
yg/kg
DAIGCa/
< 5 Ok/
< 50
< 50
< 50
< 50
< 50
< 50
a/ Direct aqueous injection gas chromatography.
b_/ Based on instrument detection limit and sample size.
-------
Four Nanograms Aerylonitrile
x2000
M/e = 53.0265-
M/e = 52.0187-
Air Sample A-8
x 2000
mnlWVHr
jwur\y»nrtfrN1^ JJL
i
ill, i I
0 1
4 5
Minutes
Figure 50. SIM Plots for DuPont, Lugoff, South Carolina
Air Sample A-8 along with a 4-ng AN Standard.
163
-------
4000
J Five Nanograms Acrylonifrile
^T i - - » i f i L • * • i • i _* i _* i
'Jf/97O 27000 £7890 27900 37910 27920 27990 27940 279SO 27900
Azeotropic Distillate of
Water Sample W-3
20S4O
2B27O
2S
55
28910 29320
\
Figure 51. Extracted Ion Current Profiles for DuPont, Lugoff, South Carolina
Water Sample W-3 (azeotropic distillate) along with a 5-ng AN Standard.
164
-------
; o o o -i
4000-
3000-
<* £000-
1000*-
\
\
\
0- ri ri i rr r i I f i i rr i i 'i T i 1 i i i i'i i i i
c
0)
o
H
0)
0)
N
C
0)
O
M
Q) O 0)
0 .-I C
0) X
-------
TABLE 66. COMPOUNDS IDENTIFIED BY GC/MS IN AIR SAMPLE
FROM DUPONT, SOUTH CAROLINA
Sample
DSC
A-8
Xylenes
m/e = 91,106,105
I = 100,46,47
100, 46,19^
Dichlorobenzenes
146,148,113
100,62,11
100, 71, -b/
Toluene
91,92
100,65
100,62
a_l The table contains the observed relative intensities of the three
m/e values listed and being used as the criteria for identifica-
tion. The expected intensities are listed in the heading with
the m/e values.
b/ m/e 113 not measurable due to low signal-to-noise level.
166
-------
Existing discharges are characterized as follows: (a) Discharge 001:
Combined discharge from waste treatment plant and interception sewer system
carrying diluted waste streams. Average flow is 5.3 mgd. Organic nitrogen
permitted level is 18 ppm. (b) Discharge 007: Once-through noncontact cool-
ing water. Average flow is A mgd. (c) Discharge Oil: Storm sewer draining
area of new nylon plant. Discharge intermediate. The DuPont plant manu-
factures cellulose acetate, cellulosic fibers, and the noncellulosic acrylic,
modacrylic and polyurethane fibers.
Air Sampling—
Air sampling was conducted using nine battery-powered air samplers at
six sampling stations located upwind, downwind, and laterally with respect
to the plant. Samplers were located from 0 to 300 m from the plant perimeter
on public land and in one case on private property with permission of the
resident. Samplers were positioned at a height of 1.5 m.
Sampling was carried out during the period 1100 (EST) March 29, 1978, to
1500 (EST) March 30, 1978. The sampling stations are shown in Figure 53 and
described in Table 67.
Water Sampling—
Grab water samples were taken at three locations; one upstream (W-l),
one downstream (W-3), and one near discharge point Oil (W-2), as described
in the NPDES permit. The downstream sample (W-3) was taken on the west side
of the river and W-2 was taken on the east side. A difference in the color
of the South River on the east and west sides was noted. The water on the
west side of the river was clear and dark brown in color. The water on the
east side by the plant was turbid and greenish white in appearance. A defi-
nite line of demarcation was observed about midstream. The sampling locations
are shown in Figure 53 and described in Table 68.
Soil and Sediment Sampling—
Three soil samples were taken at air sampling stations A-2, A-4 and A-6
by removing approximately 12 mm of topsoil from the area near the samplers.
Sediment samples were taken at each of the water sampling stations using an
Eckman dredge.
Locations of soil and sediment sampling sites are shown in Figure 53 and
described in Table 68.
Meteorological Conditions
The meteorological conditions that existed during the sampling period
were obtained on-site using a portable meteorological station to continuously
record wind data and a sling psychrometer to record humidity data. The port-
able meteorological station was located at air sampling station A-6. The
meteorological data are summarized in Table 69.
167
-------
0V
00
;^mrv^5
—-^-v * \X •:• r-.Y \\
—^ : ,<~N v. .'-.- r . w ?-. A
y\\\\tf&£--
Figure 53. Sampling locations at DuPont, Waynesboro, Virginia. Numbers in
parenthesis represent levels of AN in air
-------
TABLE 67.S AIR SAMPLING DATA FOR DUPONT, WAYNESBORO. VIRGINIA
Site
No.
A-l
A- 2
A- 3
A-3D3./
A-3S£/
A-4
§ A-5
A-6
A"6SS/
Location
U.S. Geological
Survey coordinates
4215220 mN
685750 mE
4214200 mN
685520 mE
4214360 mN
625710 mE
4214560 mN
686100 mE
4214650 mN
685900 mE
4214710 mN
684970 mE
of sampling stations
Description
S.W. corner, parking lot
Wenonah Elementary School
East side Dauphine St. 1 km
south of Main St.
East side Dauphine St 0.75
km south of Main St.
N.W. corner of Elkin Circle
and Eleventh St.
S.E. corner Eleventh St. and
Dauphine St.
464 Arch Ave.
Total
sampling
(hr)
25.42
25.50
26.67
26.67
26,67
26.5
26.5
25.25
25.25
Sampling
rate
0.970
1.008
0.888
0.824
1.106
1.100
1.018
1.131
1.121
Total
volume
(m3)
1.48
1.54
1.42
1.32
1.32
1.75
1.62
1.71
1.70
Sampling
height
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
a/ D = duplicate sample; S = spiked sample.
-------
TABLE 68. SOIL, WATER, AND SEDIMENT SAMPLING DATA, DUPONT. WAYNESBORO, VIRGINIA
Sample
No.
So-1 (A-2)£/
So- 2 (A-4)£/
So- 3 (A-6)£/
W-l
W-2
W-3
Se-1 (W-l)-3-/
Se-2 (W-2)3-/
Se-3 (W-3)£/
Location
U.S. Geological
Survey coordinates
4214200 mN
685520 mE
4214560 mN
686100 mE
4214710 mN
684970 mE
4214450 mN
684550 mE
4214830 mN
685450 mE
4215250 mN
685420 mE
4214450 mN
634550 mE
4214830 mN
685450 mE
4215250 mN
685420 mE
of sampling stations
Description
East side Dauphine St. 1 km
south of Main St.
N.W. corner of Elkin Circle
and Eleventh St.
464 Arch Ave.
50 m downstream of Wayne
Ave. bridge
West side of Dupont by
credit union
300 m upstream of Main St.
bridge, west side of river
50 m downstream of Wayne
Ave. bridge
West side of Dupont by
credit union
300 m upstream of Main St.
bridge, west side of river
Sample Sample % Moisture
type size by weight
Soil 500 cm3
Soil 500 cm3
Soil 500 cm3
Grab water 1.5 H
Grab water 1.5 I
Grab water 1.5 H
Sediment 500 cm3
Sediment 500 cm3
Sediment 500 cm3
a/ Sampling site designation on map (Figure 53).
-------
TABLE 69. METEOROLOGICAL CONDITIONS DURING SAMPLING AT DUPONT, WAYNESBORO, VIRGINIA
Time
Date (LSI)
3/29/78 1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
3/30/78 0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
Wind
Direction
00-36
—
'
-
24
32
31
34
36
04
36
32
31
31
11
09
34
36
30
22
27
09
05
02
08
13
13
16
17
Speed Temperature
(kts) (°C)
- —
18
-
06
06
08
04
04
02
02
03
01
< 01
03
02
01
< 01
01
< 01
01
01
03
04 7
04
04
04
04
03 12
Humidity
Relative Absolute
(%) (ing/ 4)
— —
39 6.1
-
-
-
-
-
-
-
-
-
-
-
-
__ _
-
-
-
-
-
-
-
56 4.2
-
-
-
-
47 4.9
-------
Sample Analysis
Air Samples —
Nine air samples were collected at six sampling stations deployed around
the DuPont acrylonitrile facility. The results of the analysis of the char-
coal tubes are presented in Table 70. The levels in micrograms per cubic
meter are also shown in Figure 53. Figure 54 shows the wind rose pattern for
the sampling period. For comparison, the annual wind rose from the Star
Program (Lynchburg, VA) follows as Figure 55. As can be seen from the wind
rose, no upwind or downwind stations could be designated for the sampling
period. Additionally, 47% of the sampling period was calm and, when there
was wind, it averaged less than 3 knots. The only site where AN was found
was station A-2 at 7.0 (ig/m3. Levels at all other sites were less than 0.2
For sampling quality assurance, spiked samplers were deployed at stations
A-3 and A-6. Recoveries of these spikes were 64% and 53%, respectively. Two
blank tubes were spiked on-site and stored on dry ice. These recoveries were
53% and 48%. Blank spikes prepared at MRI, using identical charcoal tubes,
had recoveries of 43%, 49%, 52% and 49%. Blank charcoal tubes that were taken
to DuPont, returned to MRI on dry ice, and analyzed showed no AN or interfer-
ing compounds.
Styrene was suggested by GC analyses and the levels at each sampling site
appear in Table 71. The highest styrene level, 23.0 /ig/m , occurred at sta-
tion A-l. The second highest level was at station A-6. Duplicates at this
station were in close agreement with levels of 9.9 Mg/m and 8.8 /ig/m . All
other levels were in the range of 1.9 /jg/m to 4.3 jtig/m . Triplicate sampling
at station A-3 agreed very closely with results of 2.1, 2.3 and 2.3 /ng/m3.
Styrene, however, was not confirmed when GC/MS analysis were made (see below).
Water Samples —
The results for the- analysis of the three water samples are listed in
Table 72. These samples were analyzed by the azeotropic distillation proce-
dure. No AN was found above the 1.7 /jg/liter detection limit.
Soil and Sediment Samples —
Three soil samples and three sediment samples were extracted with water
and the extracts analyzed by direct aqueous injection. No detectable AN ( < 50
|Ug/kg) was found. These results are given in Table 73.
Gas Chromatography/Mass Spectrometry Confirmation —
The presence of AN in air sample A-2 was confirmed by GC/MS using high
resolution SIM. Figure 56 shows the SIM plot for air sample A-2 as well as a
4-ng AN standard.
172
-------
TABLE 70., ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM DUPONT. WAYNESBORO, VIRGINIA
Sample
No.
A-l
A- 2
A- 3
A- 3D
A-3S
A- 4
A- 5
A- 6
A-6S
A-BS
A-BS
KC-S '
KC-S
KC-S
KC-S
A-B
A-B
Air
Sampling volume
period (m^)
3/29/78
3/30/78
3/29/78
3/30/78
3/29/78
3/30/78
3/29/78
3/30/78
3/29/78
3/30/78
3/29/78
3/30/78
_
-
-
-
-
-
-
1200- 1.479
1345
1230- 1.452
1400
1130- 1.421
1510 1.318
1.770
1150- 1.776
1420
1200- 1.619
1430
1325- 1.714
1400 1.698
_ _
-
-
-
-
-
-
Sample
typefi/
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Front
Comments
Duplicate
Spiked with 21 yg AN
Spiked with 21 yg AN
Spiked with 21 yg AN
Spiked with 21 yg AN
Spiked with 21 yg AN
Spiked with 21 yg AN
Spiked with 21 yg AN
Spiked with 21 yg AN
Field blank
Field blank
ug
Found
< 0.3
10.3
< 0.3
< 0.3
13.5
< 0.3
< 0.3
< 0.3
11.1
11.1
10.1
9.1
10.3
10.9
10.2
< 0.3
< 0.3
GC/HS
yg/m3P_' confirmation^./
< 0.2
7.0
< 0.2
< 0.2
64% Recovery
< 0.2
< 0.2
< 0.2
53% Recovery
53% Recovery
48% Recovery
43% Recovery
49% Recovery
52% Recovery
49% Recovery
-
_
Yes
-
'
—
-
-
-
—
-
-
-
-
-
-
-
a/ All backup tubes were analyzed. AN was < 0.2
b_/ Not corrected for recovery. Limit of detection based on sample size and instrument response.
c/ Yes = Confirmed presence or absence of AN; No = Unconfirmed, sample below detection limit.
-------
NNW
NNE
NW
WNW
WSW
SW
NE
SE
SSW
SSE
ENE
ESE
46.7% Calm
I Winds Greater Than 10 Knots
I Winds Less Than or Equal to 10 Knots
Figure 54. Wind Patterns during Sampling at DuPont, Waynesboro, Virginia.
174
-------
NNW
NNE
NW
NE
WNW
ENE
WSW
ESE
SW
SE
SSW
SSE
I Winds Greater Than 12mph
I Winds Less Than 12mph
Figure 55. Wind rose for Lynchburg, Virginia (1/69 - 12/73) using Star
program output
175
-------
TABLE 71. STYRENE AND ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM
DUPONT, WAYNESBORO, VIRGINIA
Sample No. Styrene (yg/m3) Acrylonitrile (yg/m3)
A-l 23.0 < 0.2
A-2 1.9 7.0
A-3 2.1, 2.3, 2.3 < 0.2, < 0.2
A-4 2.8 < 0.2
A-5 4.3 < 0.2
A-6 9.9, 8.8 < 0.2
176
-------
TABLE 72. ACRYLONITRILE CONCENTRATIONS IN WATER SAMPLES FROM DUPONT, WAYNESBORO, VIRGINIA:
AZEOTROPIC DISTILLATION
Sample
No.
W-l
W-2
W-3
Initial
volume
500
500
500
Final
volume
10
13
10
yg Found
(original sample)
< 0.7
< 0.9
< 0.7
yg/Jt
a/
< 1.3-
< 1.7
< 1.3
GC/MS
confirmation^-'
~~
—
"
aj Based on instrument detection limit and sample size.
b/ Yes = confirmed presence or absence of AN; No = Unconfirmed, below detection limit.
TABLE 73. ACRYLONITRILE CONCENTRATIONS IN SOIL AND SEDIMENT SAMPLES FROM DUPONT,
WAYNESBORO, VIRGINIA
Sample
No.
So-1
So-2
So-3
Se-1
Se-2
Se-3
Sample
size
10 g
10 g
10 g
10 g
10 g
10 g
Extraction
volume
25 ml
25 ml
25 ml
25 ml
25 ml
25 ml
yg Found
(original sample)
DAIGC3/
< 0.5
< 0.5
< 0.5
< 0.5
< 0.5
< 0.5
yg/kg
DAIGC-
< 50^
< 50
< 50
< 50
< 50
< 50
a/ Direct aqueous injection gas chromatography.
b_/ Based on instrument detection limit and sample size.
-------
00
Four Nanograms Acrylonitrile
x 400
M/e =53.0265
M/e =52.0187
l I 1 l I I I I I I I i 1 1 1 L
4 5
Minutes
Air Sample A-2
x 200
i i I
J l l I i i I l J
4 5
Minutes
Figure 56. SIM Plots for DuPont, Waynesboro, Virginia Air Sample A-2 Along with a 4-ng AN Standard.
-------
The presence of other selected organics in the air samples was tested by
the use of extracted ion current profiles.* The compounds found in air Sample
No. 6 are listed in Table 74 along with the EICP data used to confirm their
presence. The following compounds were shown not to be present: propio-
nitrile, isopropylnitrile, vinylacetonitrile, methacrylonitrile, iv-butyro-
nitrile, pyruvonitrile, lactonitrile, 2-methylbutylnitrile, 2-methylpropyl-
nitrile, acrylamide, acrolein, chlorobenzene, 1,1,2,2-tetrachloroethane,
1,1,2-trichloroethane and benzene. The mass chromatogram for air sample A-6
is shown in Figure 57 with the identified compounds noted.
BORG-WARNER, WASHINGTON, WEST VIRGINIA
Field Sampling
Presampling Survey—
A presampling survey of the Borg-Warner plant, Washington, West Virginia,
was conducted on April 3, 1978. The plant is located in the community of
Washington, West Virginia, a rural area approximately 3 miles west of Parkers-
burg, West Virginia. The plant is bounded on the south by farmland, on the
west by the Ohio River, on the north by a small area containing Junior League
baseball fields next to a DuPont facility, and on the east by DuPont Road.
The NPDES permit indicates one discharge point into the Ohio River at mile
point 191.5. The DuPont plant to the north produces nitric acid, nitroglycerine,
dynamite, water gels, fibers (synthetic, polyamide, and noncellulosic) and
resins (acrylic, acetal, fluorocarbon, polyamide, and polyvinyl butyral). The
Borg-Warner plant produces ABS and SAN resins, other acrylonitrile copolymer
resins, and plating chemicals.
Air Sampling—
Air sampling was conducted using eight battery-powered air samplers at
four sampling stations located upwind, downwind, and laterally with respect
to the plant. Samplers were located from 0 to 1,100 km from the plant perim-
eter on public land and rights-of-way and were positioned at a height of 1.5 m.
Sampling was carried out during the period from 1000 (EST) April 4, 1978,
to 1600 (EST) April 5, 1978. A duplicate sampler was set up at A-3 at 0435
(EST) April 5, 1978, after the wind shifted and was operated until the end of
the sampling period. The sampling locations are shown in Figure 58 and described
in Table 75. Intermittent rains were encountered during the period from 1500
(EST) April 4, 1978 to 1800 (EST) April 5, 1978; however, no water directly
entered the tubes.
Air sample A-6 was selected for GC/MS analysis/full scan mode because it
contained several compounds at appreciable levels.
179
-------
TABLE 74." EICP DATA§/ FOR COMPOUNDS IDENTIFIED BY GC/MS IN AIR SAMPLE
FROM DUPONT, WAYNESBORO. VIRGINIA
Sample
No.
DVA
A- 6
Xylenes
m/e = 91,106,105
I = 100,46,17
100,42,14^
Ethylbenzene
91,106,51
100,32,13
100,23,11
Trimethylbenzenes
105,120,119
100,59,15
100,25,13
Toluene
91,92
100,65
100,51
a./ The table contains the observed relative intensities of the three m/e
values listed and being used as the criteria for identification.
The expected intensities are listed in the heading with the m/e
values.
180
-------
5000-
00
4000-
3000-n
£000-
1000-
I
Qi
Q)
N
C
01
•H
H
I
QJ
C
a)
N
a)
a>
B
eooo
SPECif 31980 - 32110 Jfl DVfi H6P 3UL P.K. 4-13-78
• 11111111 i, 11111 j i /11 j 11 ii j i, 111! m-j 1111 j -i 11 lyrrrrprrryrn-rp-n-rj
£050 £ 1 0 0
STEP S PE C S=1 TIC = 100
Figure 57. Mass Chromatogram, Showing Additional Compounds Identified in
DuPont, Waynesboro, Virginia Air Sample A-6.
-------
C»
NJ
r.-^v*^ .^.i
x - "V\ v;/-' T ^
-^U^-"^ //>>:, M'1
^..WJ ;V • .-
-v>> ...••/•.'. ._ ^y-.:
L^'fJ^-^ '.r\•• ^-^~~~.).-r-:l'Jiii-'/^^. "•'"•'•/. ' ' ' |001| >• Aqueous Discharge
.,«5...-'f-/k^-ffityf'fi'-yv^^' c- "^, • '"- ' ... .^...
|[A-T) Air Sampling Station
(W- J) Water Sampling Station
|001| + Aqueous Discharge
(LuatCKI Ml
A647 " MW
1 330^'0 FEF T ' :; VA )
•<4
Figure 58. Sampling locations at Borg-Warner, Washington, West Virginia. Number in parenthesis
represent levels of AN in air
-------
TABLE 75. AIR SAMPLING DATA FOR BORG-WARNER, WASHINGTON. WEST VIRGINIA
Location
Sample U.S. Geological
No. Survey coordinates
A-l 4346650 mN
440560 mE
A- 2 4346120 mN
A-2Da-/ 441790 mE
A-2S*/
A- 3 4345210 mN
A-3D^ 441580 mE
00
OJ
A-4 4345160 mN
A-4S^/ 440300 mE
of sampling stations
Description
On Ohio side, Sellars
Road and Baltimore &
Ohio Railroad
Junior League baseball
field-Washington Bottom
900 meters south of
West Virginia
34 on Dupont Road
At old lock station
near Lewis Cemetary
Total
sampling
(hr)
22.65
27.58
27.58
27.58
28.17
10.58
28.58
28.58
Sampling
rate
(£/min)
0.820
0.984
0.917
1.042
0.824
1.085
0.905
0.830
Total
volume
(m3)
1.115
1.628
1.517
1.724
1.393
0.689
1.553
1.424
Sampling
height
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
a_/ D - duplicate sample, S - spiked sample.
-------
Water Sampling—
Grab water samples were taken at two locations; one upstream and one
downstream of the Borg-Warner facility. The upstream sample was taken at a
point near the municipal water works. The downstream sample was taken at a
point approximately 1,100 m downstream of the discharge point. Samples were
taken from shore on the West Virginia side. Water sampling sites, W-l and
W-2, are shown in Figure 58 and described in Table 76.
Soil and Sediment Sampling—
Two soil samples were taken; one at air sampling station A-2, and the
other at air sampling station A-4. Samples were collected by removing approxi-
mately 12 mm of topsoil from the area near the samplers. Sampling locations
are shown in Figure 58 and described in Table 76. Sediment samples were not
taken.
Meteorological Conditions
The meteorological conditions that existed during the sampling period
were obtained from the National Weather Service at Wood County Airport and
from an on-site meteorological station. The wind conditions from the National
Weather Service and the temperature and humidity readings measured in the
field are listed in Table 77. The portable meteorological station was es-
tablished at air sampling station A-4 but the recorder failed to function
properly.
Sample Analysis
Air Samples—
The results of the analysis of the charcoal tubes are given in Table 78.
Acrylonitrile was detected at three of the four sampling stations. According
to the wind patterns during the sampling period, shown in Figure 59, stations
A-l and A-4 were predominantly upwind, and stations A-2 and A-3 were down-
wind. Station A-4 had no detectable AN and station A-l had a level of only
o
0.3 /Ltg/m . The levels of stations A-2 and A-3, on the other hand, averaged
about 86 and 250 ^g/m-% respectively. The levels of AN in micrograms per cubic
meter are shown on the site map given in Figure 58.
Sampling quality assurance was checked by the use of spiked sampling
tubes. The spiked samplers were deployed at stations A-2 and A-4. Station
A-4 was predominantly upwind and the recovery of added AN at this site was
32%. Due to the high level of AN in the ambient air at station A-2, the re-
covery could not be calculated. Additionally, two blank tubes were spiked at
the site and immediately stored on dry ice. These blank spike recoveries
were 49% and 53%. Three blank tubes were returned and spiked at Kansas City.
Recoveries for these blank spikes were 40%, 42% and 43%. Analysis of blank
charcoal tubes revealed no interfering compounds nor acrylonitrile above the
detectable limit of 0.3 /ng per tube.
184
-------
TABLE 76. WATER AND SOIL SAMPLING DATA FOR BORG-WARNER, WASHINGTON, WEST VIRGINIA
CO
Sample
No.
W-l
W-2
So-1 (A-2)£/
So-2 (A-4)a7
Location
U.S. Geological
Survey coordinates
4348560 mN
451570 mE
4345160 mN
440290 mE
4345160 mN
440300 mE
4346120 mN
441790 mE
of sampling
Description
Near municipal water
works
At old lock station
At old lock station
Junior League baseball
field-Washington bottom
Sample Sample
type size
.Grab Water 1 £
Grab Water \S.
Soil 200 cm3
Soil 200 cm3
% Moisture
by weight
.^
_
__
a/ Sampling site designation on map (Figure 58).
-------
TABLE 77. WIND CONDITIONS DURING SAMPLING AT BORG-WARNER, WASHINGTON, WEST VIRGINIA
oo
Time
Date (LSI)
4/4/78 1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
4/5/78 0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
Winda/
Direction
(00-36)
22
22
23
22
24
23
25
24
24
26
27
25
28
28
27
31
30
28
30
34
34
34
34
36
34
33
4
Speed TemperatureW
(kts) (°C)
12
14
10
15
15 17
13
10
08
08
13
13
09
07 15
10
12
12
10
13
11 12
09
10
08
08
08
05 10
06
04
Humidityb./
Relative Absolute
(%) (ing/l)
.
- -
- -
-
100 14.2
-
-
-
-
-
-
-
78 10.3
- -
-
_ _
- -
- -
76 8.2
- -
-
- -
- -
-
87 8.2
-
_ _
(continued)
-------
TABLE 77.(continued)
WindJL/
Time
Date (LST)
1300
1400
1500
1600
Direction
(00-36)
17
24
24
35
Speed Temperature^/
(kts) (°C)
06
05
08 16
08
Humidityb/
Relative Absolute
(%) (mg/A)
-
73 9.7
a./ From National Weather Service, Wood County Airport.
b/ Taken on-site.
00
—i
-------
TABLE 78. ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM BORG-WARNER. WASHINGTON, WEST VIRGINIA
00
00
Sample Sampling
No. period
A-l 4-4-78 1504-
4-5-78 1343
A-2 4-4-78 1115-
4-5-78 1450
A-2D
A-2S
A- 3 4-4-78 1100-
4-5-78 1510
A-3D 4-5-78 0435-
4-5-78 1510
A-4 4-4-78 1030-
4-5-78 1505
A-4S
A-BS
A-BS
KC-S
KC-S
KC-S
A-B
A-B
Air volume Sample
(m3) type
1.115 Front
Back
1.628 Front
Back
1.517 Front
Back
1.724 Front
Back
1.393 Front
Back
0.689 Front
Back
1.553 Front
Back
1.424 Front
Back
Front
Front
Front
Front
Front
Front
Front
Comments
Duplicate
Spiked with 24 yg AN
Duplicate
Spiked with 24 yg AN
Spiked with 24 yg AN
Spiked. with 24 yg AN
Spiked with 24 yg AN
Spiked with 24 yg AN
Spiked with 24 yg AN
Field blank
Field blank
yg Found
0.3
Interference
150
12.1
110
< 0.4
160
0.4
242
Interference
224
< 0.4
< 0.3
Interference
6.8
0.8
11.6
12.6
10.0
9.4
10.3
< 0.3
< 0.3
GC/MS
yg/m3- Confirmation^/
0.3
99.6
72.5
78. 9-93. G£'
173.7
325.1
< 0.2
32% Recovery
49%
53%
42%
40%
43%
-
_
-
Yes
-
Yes
:
—
Yes
-
Yes
—
Yes
-
-
—
-
-
-
-
-
-
a/ Not corrected for recovery. Detection limit based on sample size and instrument response.
b/ Yes = Confirmation of presence or absence of AN; No = Unconfirmed; sample below detection limit.
cj With and without subtraction of spiked AN.
-------
NNW
NNE
NW
NE
WNW
ENE
WSW
ESE
SW
SE
I
SSW
Winds Greater Than 10 Knots
SSE
I Winds Less Than or Equal to 10 Knots
Figure 59. Wind Patterns during Sampling at Borg-Warner, Washington, West Virginia.
189
-------
Duplicate sampling at station A-2 was performed as a check of precision.
These duplicates showed good agreement with results of 99.6 /^g/m3 and 72.5 ^g/
m3. Due to the high level of ambient AN at station A-2, the 24 ug AN added
to the spike sampler can almost be considered negligible. The air level of AN
is thus 93.0 /Ltg/m3 without regard for the spike or 78.9 /Ltg/m3 if the spike is
accounted for by subtracting 24 ug from the total found. In either case, the
results agree well with the two unspiked samples. Two samplers were positioned
at station A-3; however, they were not duplicates. Air sample A-3 was col-
lected from 1100 hr, April 4, 1978, to 1510 hr, April 5, 1978, for a total of
approximately 28 hr sampling. Air sample A-3D was collected from 0445 hr,
April 5, 1978, to 1510 hr, April 5, 1978, for a total of approximately 11 hr.
Thus, the results from station A-3D reflect the ambient AN levels for the
last 11 hr of the sampling period, whereas station A-3 results reflect ambient
AN levels for the total 28 hr sampling period. From the results it appears
that the highest AN level, 325 /Lig/m3, occurred during 0435 to 1510 hr on
April 5, 1978.
The presence of styrene was indicated at the two downwind stations, A-2
and A-3, and averaged 19.0 /Lig/m3 and 43.7 jJig/m , respectively. Triplicate
sampling at station A-2 was in good agreement with individual results of
19.7, 16.2 and 21.1^g/m3. Styrene levels were generally consistent with
acrylonitrile levels, i.e., high at downwind stations and not detected at
upwind stations as shown in Table 79. It is also notable that the styrene
level in A-3D is significantly higher than in A-3. This trend agrees with
the AN results for these samples and again indicates that the higher level
of emissions occurred during the last 11 hr of sampling. Styrene was con-
firmed by GC/MS (see below).
Water Samples—
Water samples were analyzed by the azeotropic distillation procedure.
The results for the analysis of the two water samples are given in Table 80.
Sample W-l, collected upriver of the plant, contained 1.4 (Ug/liter. The down-
river sample, W-2, contained 1.9 //g/liter. Sample W-2 was spiked at 9.5 ppb
by the addition of 4.75 ^g AN following the first distillation. Additional
methanol (10 ml) was added and the process repeated. The analysis of the
distillate showed a recovery of 65%.
Soil Samples—
Two soil samples were extracted with water and the extracts analyzed by
direct aqueous injection. No AN was found above the 50 ^6g/kg detection limit.
These results are listed in Table 81.
Gas Chromatography/Mass Spectrometry Confirmation—
The confirmation of AN in some of the samples was attempted by GC/MS
using the high resolution, selected ion monitoring (SIM) mode. By this pro-
cedure, the upwind air sample A-4 was confirmed as not containing AN.
190
-------
TABLE 79. STYRENE AND ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES
FROM BORG-WARNER, WASHINGTON. WEST VIRGINIA
Sample No. Styrene (yg/m3) Acrylonitrile (yg/m3)
A-l < 0.5 0.3
A-2 19.7 99.6
A-2D 16.2 72.5
A-2S 21.2 79-93
A-3 33.0 174
A-3D 54.4 325
A-4 < 0.6 < 0.2
191
-------
Sample
No.
TABLE 80. ACRYLONITRILE CONCENTRATIONS IN WATER SAMPLES FROM BORG-WARNER,
WASHINGTON, WEST VIRGINIA: AZEOTROPIC DISTILLATION
Initial
volume
(ml)
Final
volume
(ml)
ug Found
(original sample)
GC/MS
Confirmation]!'
W-l
W-2
W-2S£/
500
500
500
10
10
10
0.7
0.95
3.1
(65% recovery)
1.4
1.9
a/ Based on sample size and instrument detection limit.
,_, b_/ Yes = Confirmed presence or absence of AN; No = Unconfirmed; below detection limit,
VD
c/ Spiked with 4.75 pg AN.
Sample
No.
TABLE 81. ACRYLONITRILE CONCENTRATIONS IN SOIL SAMPLES FROM
BORG-WARNER, WASHINGTON, WEST VIRGINIA
Sample
size
Extraction
volume
yg Found
(original sample)
DAIGC£/
ug/kgV
So-1
So-2
10 g
10 g
25 ml
25 ml
< 0.5
< 0.5
< 50
< 50
a/ Direct aqueous injection gas chromatography.
b_/ Based on sample size and instrument detection limit.
-------
The downwind samples were analyzed by GC/MS using the full scan mode
and extracted ion current profiles (EICP). EICP's were obtained for the AN
specific ions with m/e values of 53, 52 and 26. By this method, the presence
of AN was confirmed in downwind air samples A-2, A-2D, A-3 and A-3D. Figures
60 and 61 give the EICP's for these samples along with a 40-ng AN standard.
The presence of other selected organics in the air samples was tested by
the use of extracted ion current profiles. The compounds found in the air
samples A-2, A-2D, A-3 and A-3D are listed in Table 82 along with the EICP
data used to confirm their presence. The following compounds were shown not
to be present: propionitrile, isopropylnitrile, vinylacetonitrile, methacrylo-
nitrile, ii-butyronitrile, pyruvonitrile, lactonitrile, 2-methylbutylnitrile,
2-methylpropylnitrile, acrylamide, acrolein, styrene, chlorobenzene, trimethyl
benzenes, 1,1,2,2-tetrachloroethane, 1,1,2-trichloroethane and benzene.
Toluene was present in air samples A-2, A-2D, A-3 and A-3D. The presence of
styrene was confirmed in samples A-2D and A-3. Ethylbenzene and isomers of
xylene were found in Samples A-3 and A-3D. Trimethylbenzene was present in
sample A-3D. The mass chromatograms for the samples are given in Figures 62
through 65; the identified compounds are noted.
B. F. GOODRICH, LOUISVILLE, KENTUCKY
Field Sampling
Presampling Survey—
A presampling survey of the B. F. Goodrich chemical plant in Louisville,
Kentucky, was conducted on April 24, 1978. The plant is located immediately
west of Shively, Kentucky (a suburb of Louisville). It is bounded on the
north by Bells Lane, on the east by Campground Road and 1-264, on the south
by ALCOA Aluminum, and on the west by AIRCO Carbide. Other chemical produc-
tion facilities in the immediate area are DuPont, Rohm and Haas, and American
Synthetic Rubber Corporation.
The B. F. Goodrich plant produces elastomers, plastics and resins, and
latexes including butadiene-acrylonitrile latexes, polybutadiene-acrylonitrile-
polyvinyl chloride latexes, and styrene-butadiene latexes. According to the
NPDES permit, outfalls 001, 002 and 003 discharge into Paddys Run which flows
into the Ohio River.
193
-------
40-ng ACRYLONITRILE STANDARD
B9
iM03iS7oataaoat8KKM40D3i4iosi 4209149091440
JOOOO
Z7OOO
aooo
ZSDOO
MOOD
Zl ODD
10000
1
ISOOO
14000
11000
I
4000-
aooo
1000
0
31
AIR SAMPLE A-2
•^X-N
81
Figure 60. Extracted Ion Current Profiles for Borg-Warner, Washington, West Virginia
Air Samples A-2 and A-2D along with a 40-ng AN Standard.
194
-------
40-ng ACRYLONITRILE STANDARD
zauuu
MOOD
83000
28000
21000
ITOOD
21 ODD
19000
II
II
15000
14000
laopo
12000
11000
10000
TBIIU
•ooo
4000
1000
o
at
AIR SAMPLE A-3
I7OOO
MOOD
WOOD
Z1C
I
I
ITOOD
laggo
isooo
14000
laooo
laooo
MOOD-
10000
4000
1000-
0
AIR SAMPLE A-3D
ss
IW081 •! Oil UOtl H031040»l 0BO91 OK»1 «TO
si e4Wi«o»i «osi troii
Figure 61. Extracted Ion Current Profiles for Borg-Warner, Washington, West Virginia
Air Samples A-3 and A-3D along with a 40-ng AN Standard.
195
-------
TABLE 82. EICP DATA- FOR COMPOUNDS IDENTIFIED BY GC/MS IN AIR SAMPLES
FROM BORG-WARNER, WASHINGTON, WEST VIRGINIA
Xylenes
Sample m/e 91,106,105
No. I 100,46,17
BWV
A- 2
BWV
A-2D
BWV 100,37,13
A- 3
BWV 100,40,8
A- 3D
Ethylbenzene Trimethylbenzenes Toluene
91,106,51 105,120,119 91,92
100,32,13 100,59,15 100,65
100,44
100,58
100,17,11 — 100,63
100,17,8 100,29,11 100,55
Styrene
104,103,78
100,40,30
__
100,45,10
100,55,12
100,58,29
a/ The table contains the observed relative intensities of the three m/e values listed and being
used as the criteria for identification. The expected intensities are listed in the heading
with the m/e values.
-------
SOOO-i
H 0 0 0 -
3000-
£000-
1000-
1600
SPEC« 31SSO - 31690 Jtl
1650
STEP SPECS'1 TIC* 100
Figure 62. Mass Chromatogram for Borg-Warner Air Sample A-2D.
-------
VO
00
5000-1
HOOO-
3000-
8000-
1000-
1111111 |ri Ml 111 f| 11 r'l p 1111111111111 [ n r 1111 ri'| i n ri 1111 |Ti i m-rrrfTTi i n i n ITTTI 1111 n : 1111
1600 1650
SP£C« 315SO > 31690 JM
STEP S P E C « =1 TIC = 100
Figure 63. Mass Chromatogram for Borg-Warner Air Sample A-2D.
-------
£ 0 0 0 —
t 0 0 0 -
3000-
01
c
01
N
Cl
Ol
H-1
^ £000-
1 000-
Of .,
^
1 S
\ ( I
\ '^ o
\ \, H
""l""1""1""1""1""1""1""1""' 1'"1'"
C - % Q) 0) >-< O)
t-i rH t-H >v-H
vs^_ .. — —^\ 4-
1700 1750
BPECit 31700 - 31830 JM BWV A3P 1UL P . K . 4-13-78'
1800
STEP S P E C « =1 TIC = 100
Figure 64. Mass Chromatogram for Borg-Warner Air Sample A-3.
-------
5000-1
4000-
NJ
O
o
30
£0
10
00-
00-
.
.
*
-
00-
0,
1 c
01
N
1 0) C
1 r-l 0)
1 II 5
1 iJ f>-,
\ 0 ^ -
01 M Ol OJ
1 3 0 C C
\ 1-1 •< — -^.. -,
\
V
"
Tf M 1 1 1 1 1'| 1 1 1 1 |'i n 1 1 1 1 1 1 1 1 m 1 1 1 1) 1 1 1 11 1 1 M i j 1 1 1 1 j ri M 1 1 r n | ri 1 1] 1 1 1 1 1 1 1 1 1 |'i 1 1 1 j i r
01
0)
N
C
0)
r^
0) -U
c R3PD UJL
1950
P.K. 4-13-78
STEP SPEC«=1 TIC
1 0 0
Figure 65. Mass Chromatogram for Borg-Warner Air Sample A-3D.
-------
The DuPont plant in the area is southwest of the Goodrich plant and pro-
duces elastomers, vinyl fluoride, and certain other chlorinated and fluorinated
compounds. The Rohm and Haas facility is also to the southwest and produces
methyl methacrylate monomer, acrylic resins, acrylic process aids, impact
improvers for PVC, and polymethyl methacrylate sheet. The American Synthetic
Rubber Corporation produces polybutadiene, polybutadiene-styrene, carboxylated
styrene-butadiene, styrene-butadiene copolymer resins, high solids latexes,
liquid rubbers, and special application liquid polymers.
Air Sampling—
Air sampling was conducted using seven battery-powered air samplers at
four sampling stations located upwind, downwind, and laterally with respect
to the plant.
Sampling was conducted during the period 1000 (EST) April 24, 1978, to
1400 (EST) April 25, 1978. The sampling locations are shown in Figure 66 and
described in Table 83. During the sampling period, winds blew primarily from
the north and occasional rain occurred. No water entered the tubes directly.
Water Sampling—
Two grab water samples were taken on April 25, 1978, one upstream and
one downstream of the plant. Water sampling locations are shown in Figure 66
and described in Table 84.
The NPDES permit for this plant indicates that it discharges into Paddy
Run which then enters the Ohio River near Mile Point 613. The upstream sample
was taken from a boat launching ramp on the Indiana site just north of the
Shermon Benton Bridge approximately 6.4 km upstream of Paddy Run. The down-
stream sample was taken from the riverbank on the Kentucky side approximately
2.0 km downstream of Paddy Run.
Soil Sampling—
Soil samples were taken on April 25 at two locations. Sampling was ac-
complished by removing approximately 12 mm of topsoil from the area near
Sampling Stations A-2 and A-3. Sampling locations are shown in Figure 66 and
described in Table 84.
Meteorological Conditions
The meteorological conditions that existed during the sampling period
were obtained from the National Weather Service at Standiford Field, Louisville,
Kentucky. Some humidity measurements were made on-site. The data are summar-
ized in Table 85.
201
-------
.......j.;Vp«k
te:>v
••/.,,
'lentil
Site •:•'•
JiVJ
. -. « ~«" • <£_ii^A-l-,' . ...J^.
f^^/:j;^.
$b:nrfe:;/l,rV''
£~.^._. 0,.;X'>] li|
L'y.. •..?»./ •...:-/• I | .. •.:• >^-r^
brriia:, ,.,^rr-• i>.'^Tr% ,C i ;
-—^ -•]•—^ ,.-^u -V^*; •
^ftilffifl
IN
A **// Subsutiy *""/
xp^^.^
-< .< \ /• /{ •r'-'P'SffiPif** - •"•
'••••>• j^^ ^M^^-"
<***-"_ ^J0 ^r-^ ,^#^-..--->^x:'';^
••• .x /^ X^^C- '•.'"-•... - -^-
^ x"/ MII. fr%? • / ^^? " • ^^«. . 'n '<^*r
, ' /:- ««« **£•/. ^ ^^J-/^M\
' // s&S •-!-,.'/ .•'iv-^'-i' >V« >«? Ml
' ^' /^$( <0A1<°-2^: H->/*V^0?->
r^ - --^-iW/EK, : \{2i'tsS^ j
' M^M^;^: u^Zz^l1
>*V • V •:y^^^J /-:>..• -., 17=—... . : 1
[A-1 JAir Sampling
^__^ Station
|W-1 [Water Sampling
Station
^ Discharge Point
1000 m
• —. '
Riverside
. .•'.;;;..Garden3
' '
M«--
' '?
l%rfr/''gt(H:
Figure 66. Sampling locations at B. F. Goodrich, Louisville, Kentucky.
Numbers in parenthesis represent levels of AN in air (yg/m )
202
-------
TABLE 83. AIR SAMPLING DATA FOR B. F. GOODRICH. LOUISVILLE. KENTUCKY
o
u>
Sample
No.
A-l
A-2
A-2DJ7
A-3
A-4
Location
U.S. Geological
Survey coordinates
4230720 mN
602580 mE
4229470 mN
600000 mE
4230520 mN
603020 mE
4231200 mN
600450 mE
•>
of sampling stations
Description
On Bells Lane approximately
400 m west of 1-264
On Senn Road approximately
700 m west off Campground
Road
Access raod from Bells Lane
to 1-264 East
Gate on road to boat ramp
approximately 1.5 km north
of Five Mile Lane - Indiana
side
Total
sampling
(hr)
25.9
26.0
25.7
25.7
25.7
25.9
25.8
Sampling
rate
(i/min)
1.13
0.611
0.829
0.910
0.876
0.850
1.08
Total
volume
(m3)
1.75
0.954
1.28
1.40
1.35
1.32
1.68
Sampling
height
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
a/ D = Duplicate sample
S = Spiked sample
-------
TABLE 84. SOIL AND WATER SAMPLING DATA, B. F. GOODRICH, LOUISVILLE. KENTUCKY
Location
Sample U.S. Geological
No. Survey coordinates
So-1 (A-2)-/ 4229470 mN
600000 mE
So-2 (A-3)-/ 4230520 mN
603020 mE
W-l
—
W-2 4729270 mN
g 599200 mE
-fs
of sampling stations
Sample Sample
Description type size
On Senn Road approximately Soil 250 cm3
700 m west of Campground
Road
Access road from Bells Lane Soil 250 cm3
to 1-264 East
Boat ramp under Shemon Benton Grab water 1.5 S.
Bridge
From riverbank approximately Grab water 1.5 ft
2 km downstream of Paddy Run
% Moisture
by weight
.
_
_
_
a/ Sampling site designations on map (Figure 66).
-------
TABLE 85. METEOROLOGICAL CONDITIONS DURING SAMPLING AT B. F. GOODRICH,
LOUISVILLE, KENTUCKY, SOURCE OF DATA - STANDIFORD FIELD,
LOUISVILLE, KENTUCKY
Time
Date (LST)
April 24, 1978 1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
April 25, 1978 0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
Wind
Direction
00-36
16
00
00
00
06
12
02
32
02
02
36
01
36
04
01
35
36
01
00
36
00
00
01
35
29
31
31
34
35
Speed Temperature—
(kts) (°C)
04
00
00
00
03
08
10
09
10
06
06
11
10
05
04 12.2
06
04
04 10.0
00
05
00
00
05
03 10.0
06
04
09
10 11.2
14
Humidity^
Relative Absolute
(%) (mg/j?)
76 8.21
93 8.74
93 8.74
94 10.53
a/ Measurements made on-site.
205
-------
Sample Analysis
Air Samples—
Seven air samples were collected at four sampling sites deployed around
the B. F. Goodrich plant. The results of the analysis of the charcoal tubes
are presented in Table 86. The levels in micrograms per cubic meter are also
shown in Figure 66. The wind rose pattern for the sampling period is given
in Figure 67. For comparison, the annual wind rose from hourly observations
for Louisville, Kentucky, follows as Figure 68. Winds were predominately from
the north making Station A-l upwind and Stations A-2, A-3 and A-4 crosswind.
The only site where AN was found was station A-l at 4.3 /Lig/m^. Levels at all
other sites were less than 0.2 /zg/m^. Although station A-l was an upwind site,
the fact that it was positioned immediately north of the plant probably accounts
for the detection of AN at this site.
Sampling quality assurance was checked by spiked samplers deployed at
sampling stations A-l and A-2. The recovery of the A-l spike was not calcu-
lated due to the presence of ambient AN. A recovery of 19% was calculated
for the A-2 spike. Two blank tubes were spiked at B. F. Goodrich and immedi-
ately stored on dry ice. The recoveries for the blank spikes were 91 and 87%.
Two identical tubes were spiked at MRI and stored on dry ice. These recoveries
were 82 and 75%. No AN or interference was found on the blank tubes.
During the GC/FID analysis of air samples, styrene was identified by GC
retention time and quantitated. These results are given in Table 87. For
comparison, the AN levels are repeated in this table. The sampling site with
the highest styrene level also happens to be the site where the highest AN
levels were found. This suggests that the plant is contributing to the styrene
air levels. The only site not containing styrene was sampling site A-3, a
near, crosswind site. Styrene, however, was not confirmed by GC/MS (see below).
Water Samples—
Two water samples were analyzed by the azeotropic distillation method.
These results are shown in Table 88. The upstream water sample W-l did not
contain AN above the detection limit of 1.4 jLtg/liter. The downstream water
sample W-2 contained 2.0 ^g/liter. This sample was taken from the Ohio River
bank (Kentucky side) approximately 2 km downstream of Paddy Run.
During the GC analysis of the water samples, acetonitrile was detected
and quantitated. The levels for both water samples, upstream and downstream,
were 2 jug/liter. These identical results imply that the acetonitrile source
is upstream of the B. F. Goodrich facility.
206
-------
TABLE 86. ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM B. F. GOODRICH. LOUISVILLE. KENTUCKY
ho
O
Sample
No.
A-l
A- IS
A- 2
A-2D
A-2S
A- 3
A- 4
A-BS
A-BS
KC-S
KC-S
A-B
A-B
Sampling
period
4/24/78 1050-
4/25/78 1245
4/24/78 1120-
4/24/78 1302
4/24/78 1102-
4/24/78 1253
4/24/78 1150-
4/24/78 1338
-
-
-
-
-
Air
volume Sample
(m3) type3/
1.75 Front
0.95 Front
Back
1.28 Front
1.40 Front
1.35 Front
1.32 Front
1.68 Front
Front
Front
Front
Front
Front
Front
Comments
Spiked with 20 ug AN
Duplicate
Spiked with 20 pg AN
Spiked with 20 ug AN
Spiked with 20 ug AN
Spiked with 20 ug AN
Spiked with 20 ug AN
Field blank
Field blank
Mg
Found
7.6
13.6
2.1
< 0.3
< 0.3
3.9
< 0.2
< 0.2
18.2
17.4
16.3
15.0
< 0.2
< 0.2
1
Ug/m3b/
4.3
Recovery not calculated
< 0.2
< 0.2
19% Recovery
< 0.2
< 0.2
91% Recovery
87% Recovery
82% Recovery
75% Recovery
-
GC/MS
confirmation0./
Yes
-
_
-
-
_
_
_
-
-
-
-
al All backup tubes were analyzed. Those not shown contained < 0.3 ug.
b_/ Not corrected for recovery. Detection limit based on sample size and instrument response.
cj Yes = Confirmed presence or absence of AN; No = Unconfirmed, sample below detection limit.
-------
NNW
NNE
NW
NE
WNW
ENE
WSW
SW
SE
I
I
SSW ^^_ _ -^ SSE
Winds Greater Than 10 Knots
Winds Less Than or Equal to 10 Knots
Figure 67. Wind patterns during sampling at B. F. Goodrich,
Louisville, Kentucky.
208
-------
NW
WNW
WSW
NNW
NNE
NE
ENE
SW
SSW ^^_ ^^^ SSE
I Winds Greater Than 12mph
I Winds Less Than 12mph
Figure 68. Wind rose for Louisville, Kentucky (1/51 - 12/60)
SE
209
-------
TABLE 87. COMPARISON OF ACRYLONITRILE AND STYRENE LEVELS
Sample
No.
A-l
A-2
A-2D
A-3
A-4
AN
concentration
(ug/m3)
4.3
< 0.2
< 0.2
< 0.2
< 0.2
Styrene
concentration
(UK/in3)
5.7
1.3
1.4
< 0.1
1.9
TABLE 88. ACRYLONITRILE CONCENTRATIONS IN WATER SAMPLES FROM
B. F. GOODRICH, LOUISVILLE, KENTUCKY: AZEOTROPIC DISTILLATION
Initial Final
Sample volume volume ug Found GC/MS
No. (ml) (ml) (original sample) ug/l confirmation-
W-l
W-2
500
500
10 < 0.7 < 1.4^
10 1.0 2.0
Yes
Yes
a/ Yes = Confirmed presence or absence of AN
No = Unconfirmed, sample below detection limit.
b_/ Based on instrument detection limit and sample size.
210
-------
Soil Samples—
Two soil samples were extracted with water and the extracts analyzed by
the direct aqueous injection technique. No detectable AN was present in
either sample. These results are given in Table 89.
TABLE 89. ACRYLONITRILE CONCENTRATIONS IN SOIL SAMPLES FROM
B. F. GOODRICH. LOUISVILLE. KENTUCKY
Sample
No.
So-1
So-2
Sample
size
(g)
10.8
10.0
Extraction
vo lume
(ml)
25
25
ug Found
(original sample)
DAIGCS/
< 3.7
< 4.0
ug/kg
DAIGC-'
< 37(£/
< 400
a/ Direct aqueous injection gas chromatography.
b/ Based on instrument response and sample size.
Gas Chromatography/Mass Spectrometry Confirmation—
The confirmation of AN in the air sample A-l was made by GC/MS operated
in high resolution, selected ion monitoring (SIM) mode. Figure 69 gives the
SIM plot for this sample as well as for a 2-ng standard of AN.
The azeotropic distillates of water samples W-l and W-2 were analyzed by
GC/MS operated in the SIM mode. The ions at m/e 53 and 52 were monitored to
detect acrylonitrile and the ions at m/e 41 and 39 were monitored to detect
acetonitrile. Figure 70 shows the AN SIM plots for W-l, W-2 and a 1-ng AN
standard. The presence of AN was difficult to confirm due to a high back-
ground, especially for the m/e 53 ion. It appears, however, that AN is con-
firmed in the downstream sample W-2 and is at a much lower level, if at all
present, in upstream sample W-l.
Air sample A-l was analyzed by GC/MS using the full scan mode and EICP's
for eight other possible compounds. By this method, air sample A-l was found
to contain isomers of xylene, isomers of trimethylbenzene, ethylbenzene,
toluene, and dichlorobenzene. The EICP data are listed in Table 90. Not
found were styrene, benzene, and acrylonitrile. Styrene was detected by GC
analysis in several air samples. The inability to confirm by GC/MS has been
noted previously and is suspected to be due to instability of styrene. The
mass chromatogram with the identified peaks noted is shown in Figure 71.
211
-------
CO
2 Monograms Acrylonitnle
(X 1000)
__JI_M_1U . -i J ill* i tt.l J, I,* M ,
^ ' ^4 ' " judA+f90*
t II I i 1 < KB! 1 ft J A JB J i A /
3 A
MINUTES
Goodrich A-l
(X 1000)
jir^^1
*i MH MM a i a i a •g'
-" --
8
I
2
I • • I r
78
MINUTES
Figure 69. SIM plots for B. F. Goodrich, Louisville, Kentucky, air samples
A-l and a 2-ne AN standard.
212
-------
1 Nonogram Acrylonirril*
(X50)
"53
M/e -52
I ' ' I ' ' I ' ' I ' I ' I
349678
MINUTES
I ' ' I ' I ' ' I
9 10 11 12
Goodrich W-l Diitillofo
(X5)
'« =52(X20)
1 ' I ' ' I ' I I r I ' I I I I I I I
0 123*36 789 1O 11 12
MINUTES
Goodrich W-2 Oiitillate
53 (X 1)
M/, =52 (X 10)
i • i • • i • ' i • T i • • i • ' i ' ' r • • i r • i • • i • • i
12343B783101112
MINUTES
Figure 70. SIM plots for B. F. Goodrich, Louisville, Kentucky, water sample
distillates W-l and W-2 and a 1-ng AN standard.
213
-------
a/
TABLE 90. EICP DATA- FOR COMPOUNDS IDENTIFIED BY GC/MS IN AIR SAMPLE
FROM GOODRICH, LOUISVILLE, KENTUCKY
Sample m/e
No. I
GKY
A-l
Xylenes
91,106,105
100,46,17
100,43,17
Ethylbenzene
91,106,51
100,32,13
100,43,10
Trimethylbenzene
105,120,119
100,59,15
100,38,15
Toluene
91,92
100,65
b/
Dichlorobenzene
146,148,113
100,62,11
100, 50, -
a./ The table contains the observed relative intensities of the three m/e values listed and being
used as the criteria for identification. The expected intensities are listed in the heading
with the m/e values.
b/ Ion current saturated at the retention time.
c/ Peak not seen due to low signal/noise ratio.
KJ
-------
II
N
I
-C
u
30000-1
20000-
10000-
0 -~
SPECS
i-T-T-ri i i | i i i i f i i i i [ i i i i i i T-Ti-pn-
'980 6990 7000 ,"'
- .'100 JI1 GOODRICH ft-
J.G
8--23x/a
TTT i iT~rT-r-p i I I I i T I i | i-r-rrp-i I I | I n T-rrT-rr p I i I p~i
' i) 'I 0 ? (" 5 n 7060 70 J' 0 ? 0 8 0
STEP SPECS=1 HIT- 1000
rr-r-ri I I \'t ^TT-I
0 V 0
Figure 71. Mass chromatogram showing additional compounds identified in
B. F. Goodrich, Louisville, Kentucky, air sample A-l.
-------
Acetonitrile was detected in the azeotropic distillate of water samples
W-l and W-2 by GC analysis. It could not be confirmed by GC/MS-SIM due to
an extremely high background at m/e 41 and 39.
MONSANTO, ADDYSTON, OHIO
Field Sampling
Presampling Survey—
A presampling survey of the Monsanto plant, Addyston, Ohio, was conducted
on April 26, 1978. The plant lies within the corporate city limits of
Addyston, Ohio, immediately west of Cincinnati. The plant is bounded on the
north by US-50 highway, on the east by Muddy Creek, on the south by the Ohio
River, and on the west by an uninhabited area. Residential areas are located
north of the plant and across the Ohio River in Kentucky. No other indus-
tries are located in the immediate area. Thfc Monsanto plant produces formal-
dehyde and plastics and resins including, acrylonitrile-butadiene-styrene
resins and styrene-acrylonitrile resins. The NPDES permit lists two dis-
charges into the Ohio River offshore and north of Muddy Creek.
Air Sampling—
Air sampling was conducted using eight battery powered air samplers at
five sampling stations located upwind, downwind, and laterally with respect
to the plant.
Sampling was conducted during the time period 0900 (EST) April 26, 1978,
to 1500 (EST) April 27, 1978. The sampling locations are shown in Figure 72
and described in Table 91. During the sampling period winds were primarily
from the north-northeast.
Water Sampling—
Two grab water samples were taken on April 27, 1978, one upstream and one
downstream of the plant.
The NPDES permit for this plant indicates two discharge points into the
Ohio River near Mile Point 484. The upstream sample was taken from River Park
approximately 1.1 km upstream of the discharge points. The downstream sample
was taken from shore on the Ohio side approximately 1.0 km downstream of the
discharge points. These locations are shown in Figure 72 and described in
Table 92.
216
-------
V«
XK
n'f • - - <- -" -"- • . V- -•" J ; S '•ifttN
.*,-/' Co ",.'^Vi/,yi \ ^ %
^^p\^-;^54^. \ ^"
; vr*fC#- %Sfe;^^^ \ **
\^>':.^ .$T-:$tf:-'^, )
- .- \ ' •• ••':(,_ , I • 'i -. • -. •> ^--^/j
(l.M.O)
Figure 12. Sampling locations at Monsanto, Addyston, Ohio. Numbers in parenthesis represent levels
of AN in air
-------
TABLE 91. AIR SAMPLING DATA FOR MONSANTO. ADDYSTON, OHIO
Sample
No.
A-l
A-2
A-2S3/
A-3
A~3Da7
A-3S-
t-o
00 A-4
A-5
Location
U.S. Geological
Survey coordinates
4334580 mN
698100 mE
4334120 mN
697940 mE
4333050 mN
697320 mE
4333450 mN
698230 mE
4334540 mN
697150 mE
of sampling stations
Description
Second street between Grigsby
and Baker
On U.S. 50 north of plant
On Kentucky 8 southwest of
plant
20 m north of U.S. 50 on Kirby
On U.S. 50 approximately 400
m west of plant
Total
sampling
(hr)
28.5
29.0
29.0
27.3
27.3
27.3
28.8
28.9
Sampling
rate
(tf/min)
0.824
0.778
0.813
0.968
0.464
1.04
0.975
1.04
Total
volume
(m3)
1.41
1.35
1.41
1.58
0.76
1.70
1.68
1.81
Sampling
height
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
a/ D = Duplicate sample
S = Spiked sample
-------
TABLE 92. SOIL AND WATER SAMPLING DATA, MONSANTO, ADDYSTON. OHIO
^0
Location of sampling stations
Sample
No.
So-1 (A-3)-/
So-2 (A-4)5/
W-l
W-2
U.S. Geological
Survey coordinates
4333050 mN
697320 mE
4333450 mN
698230 mE
4332950 mN
698150 mE
4334570 mN
696820 mE
Description
On Kentucky 8 southwest of
plant
20 m north of U.S. 50 on
Kirby
River park approximately 650
m southeast of plant
Approximately 750 m north-
west of plant
Sample
type
Soil
Soil
Grab water
Grab water
Sample
size
250 cm3
250 cm3
1.5 i
1.5 H
% Moisture
by weight
-
-
-
-
a/ Sampling site designations on map (Figure 72). !
-------
Soil Sampling—
Soil samples were taken on April 27, 1978, at two locations. Sampling
was accomplished by removing approximately 12 mm of topsoil from the area near
air sampling stations A-3 and A-4. Sampling locations are shown in Figure 72
and described in Table 92.
Meteorological Conditions
The meteorological conditions that existed during the sampling period
were obtained from the Greater Cincinnati Airport. Some humidity measure-
ments were made on-site. The data are summarized in Table 93.
Sample Analysis
Air Samples—
Eight air samples were collected at five sampling stations around the .
Monsanto acrylonitrile facility. Table 94 contains the results of the analy-
sis of the charcoal tubes. In addition, Figure 72 gives the levels in micro-
grams per cubic meter at each sampling station. As can be seen from the wind
rose shown in Figure 73, the winds were predominately from the north-northeast.
For comparison, the annual wind rose from hourly observations for Cincinnati,
Ohio, follows as Figure 74. Wind speeds were greater than 10 knots (39%) of
the sampling period. As a result, station A-3 was downwind during most of the
sampling period. Stations A-4 and A-5 were crosswind while station A-l was
far upwind and station A-2 was near upwind. These designations are consistent
with analytical results in that both upwind stations, stations A-l and A-2,
had levels < 0.2 /ng/m while Station A-3, the downwind station, had a level
of 1.1 jUg/m3. Crosswind stations A-4 and A-5 had levels of 0.9 and < 0.2 /^g/
m , respectively.
Sampling quality assurance was checked by the use of spiked samplers at
stations A-2 and A-3 and by the use of spiked blank tubes. Spiked sampler
A-2 had a recovery of 46%, but the A-3 spike recovery was not calculated due
to the presence of AN in the ambient air. Spiked blank tubes had recoveries
of 79 and 72%. Identical spiked blanks prepared at MRI had recoveries of 60
and 58%. The integrity of the charcoal tubes was checked by the extraction
and analysis of blank tubes and found to be free of AN or interference. Dup-
licate sampling at station A-3 was in good agreement with levels of 1.1 and
1.0
220
-------
TABLE 93. METEOROLOGICAL CONDITIONS DURING SAMPLING AT MONSANTO,
ADDYSTON. OHIO, SOURCE OF DATA - GREATER CINCINNATI AIRPORT
Time
Date (LST)
April 26, 1978 0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
April 27, 1978 0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
Wind
Direction
00-36
04
04
05
04
05
05
05
04
04
02
02
02
02
01
03
02
03
05
36
02
02
36
35
36
03
02
01
01
03
01
36
Humidity^/
Speed Temperature**/ Relative Absolute
(kts) (°C) (%) (wx/t.)
14 - -
12 14.4 51 6.32
14 -
15
16
15
16
16
14
15
14
10
11
08
09
08
08
10
08
07
06
07
08
10
10
10
13
11
11
14
08
a/ Measurements made on-site.
221
-------
TABLE .94. ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM MONSANTO. ADDYSTON. OHIO
N3
K>
ro
Sample
No.
A-l
A- 2
A-2S
A- 3
A-3D
A-3S
A-4
A- 5
A-BS
A-BS
KC-S
KC-S
A-B
A-B
Air
Sampling volume Sample
period (m^) typeS'
4/26/78 1020- 1.41 Front
4/27/78 1448
4/26/78 0952- 1.35 Front
4/27/78 1445 1.41 Front
Back
4/26/78 1145- 1.58 Front
4/27/78 1400 0.76 Front
1.70 Front
Back
it/26/78 1005- 1.68 Front
4/27/78 1453
4/26/78 0945- 1.81 Front
1440
- - - Front
- - - Front
- - - Front
- - - Front
- - - Front
- - - Front
Comments
Spiked with 24 ug AN
Duplicate
Spiked with 24 ug AN
Spiked with 24 ug AN
Spiked with 24 ug AN
Spiked with 24 ug AN
Spiked with 24 ug AN
Spiked wiht 24 ug AN
Spiked with 24 ug AN
Ug
Found
< 0.2
< 0.2
3.0
8.0
1.8
0.8
9.2
2.9
1.6
< 0.2
19.0
17.2
14.3
14.0
< 0.2
< 0.2
GC/MS
Ug/m3— confirmation^./
< 0.2
< 0.2
46% Recovery
1.1 Yes
1.0
Recovery not calculated
0.9 Yes
< 0.2
79% Recovery
72% Recovery
60% Recovery
58% Recovery
-
a/ All backup tubes were analyzed. Those not shown contained < 0.3 ug-
i
W Not corrected for recovery. Detection limit based on sample size and instrument response.
c^l Yes = Confirmed presence or absence of AN; No = Unconfirmed, sample below detection limit.
-------
NNW
NNE
NW
NE
WNW
ENE
WSW
ESE
SW
SE
SSW
SSE
I Winds Greater Than 10 Knots
I Winds Less Than or Equal to 10 Knots
Figure 73. Wind patterns during sampling at Monsanto,
Addyston, Ohio.
223
-------
NNW
NNE
NW
NE
WNW
ENE
WSW
SW
SE
SSW
• Winds Greater Than ]2mph
I Winds Less Than 12mph
SSE
Figure 74. Wind rose for Cincinnati, Ohio (1/51 - 12/60)
224
-------
Styrene in the air samples was detected and quantitated. These results
are presented in Table 95. The AN levels at corresponding sites are also
given for comparison. The highest styrene levels were found at A-2, a near
upwind site, and A-5, a predominately crosswind site. AN was not described at
these sites. The sampling station containing the highest AN levels, station
A-3, was the site where the lowest styrene levels were found. No conclusions
can be made from the difference since nothing is known about the storage or
use patterns of these compounds. Styrene was not confirmed by GC/MS analysis
(see below).
Water Samples—
The results for the analysis of the two water samples by azeotropic dis-
tillation are given in Table 96. The upriver water sample W-l contained
<1.4 /ig/liter while the downriver sample W-2 contained 8.0 /ng/liter. The
downriver sample was taken approximately 1 km downstream of the Monsanto dis-
charge points. After distillation, sample W-l was spiked with 25 /Lig AN,
10 ml of methanol was added, and the sample was redistilled. Recovery was
calculated to be 93%.
Acetonitrile was also detected in the water samples and quantitated.
Both upstream and downstream samples contained 4 /xg/liter. Equal levels for
both samples implies an acetonitrile source upstream of the Monsanto plant.
It is interesting to note that acetonitrile was also detected at Goodrich,
Louisville (129 river miles downstream of Addyston, Ohio) at a level of 2 fj,g/
liter.
Soil Samples—
Two soil samples were extracted with water and the extracts analyzed by
direct aqueous injection. These results are shown in Table 97. No AN was
detected in either sample.
Gas Chromatography/Mass Spectrometry Confirmation—
The presence of AN in air samples A-3 and A-4 was confirmed by GC/MS-SIM.
Figure 75 gives the SIM plot for these samples as well as for a 2-ng standard
of AN.
The azeotropic distillates of water samples W-l and W-2 were analyzed by
GC/MS operated in the SIM mode. The ions at m/e 53 and 52 were monitored to
detect acrylonitrile and the ions at m/e 41 and 39 were monitored to detect
acetonitrile. Figure 76 shows the AN SIM plots for W-l, W-2 and a 1-ng AN
standard. The presence of AN was difficult to confirm due to a high back-
ground, especially from the m/e 53 ion. It appears, however, that AN is con-
firmed in the downstream sample W-2 and is at a much lower level, if at all
present, in upstream sample W-l.
225
-------
TABLE 95. COMPARISON OF ACRYLONITRILE AND STYRENE LEVELS
_ IN AIR SAMPLES FROM MONSANTO. ADDYSTON, OHIO _
AN Styrene
Sample concentration concentration
No. _ (Pg/m3) _ (ng/rn3)
A-l <- 0.2 2.2
A-2 < 0.2 5.6
A-3 1.1 0.7
A-3D 1.0 1.8
A-4 0.9 3.2
A-5 < 0.2 4.6
226
-------
TABLE 96. ACRYLONITRILE CONCENTRATIONS IN WATER SAMPLES FROM
MONSANTO. ADDYSTON, OHIO: AZEOTROPIC DISTILLATION
Initial Final
Sample volume volume yg Found GC/MS
No. (ml) (ml) (original sample) Mg/£ confirmationjL/
W-l 500 10 < 0.7 < 1.4b/ Yes
W-1S£/ 500 10 23 -
(92% Recovery)
W-2 500 10 4.0 8.0 Yes
aj Yes = Confirmed presence or absence of AN; No = Unconfirmed, sample
below detection limit.
b/ Based on instrument response and sample size.
c/ Spiked with 25 yg AN.
227
-------
TABLE 97. ACRYLONITRILE CONCENTRATIONS IN SOIL SAMPLES FROM
MONSANTO. ADDYSTON. OHIO
Sample
No.
So-1
So-2
Sample
size
(8)
10
10
Extraction
volume
(ml)
25
25
ug Found
(original sample)
DAIGCS/
< 4
< 4
Hg/kg
DAIGC3/
< 40(£/
< 400
a/ Direct aqueous injection gas chromatography.
b_/ Based on instrument response and sample size.
228
-------
2 Nanograms Acrylonitrile
(X 1000)
Monwnto A-3
(X 1000)
I
2
Monjanfo A-4
(X 1000)
I
3
I
6
MINUTES
M/.
8
34
MINUTES
I i T ,
7 8
Figure 75. SIM plots for Monsanto, Addyston, Ohio, air samples A-3 and A-4
and a 2-ng AN standard.
229
-------
1 Nanogram AcrylonitriU
(X50)
M/f S3
52
T I ' ' I ' ' I ' ' I ' I ' ' I * I ' I I I t ' I
0 12 3496 7 09 10 11 12
MINUTE 3
MomantoW-l DUHIIat*
' ' I ' ' I ' ' I ' ' I ' ' I ' ' I
' '
(X5)
52 (X 20)
MINUTES
MonwntoW-2 Distillate
M/, -52(X20)
0123^3 6 78 9 10 11 12
MINUTES
Figure 76. SIM plots for Monsanto, Addyston, Ohio, water sample distillates
W-l and W-2 and a 1-ng AN standard.
230
-------
Air sample A-4 was analyzed by GC/MS using the full scan mode and EICP's
for eight other possible compounds. By this method, air sample A-4 was found
to contain isomers of xylene, isomers of trimethylbenzene, ethylbenzene,
toluene and dichlorobenzene. The EICP data are listed in Table 98. Not found
were styrene, benzene and acrylonitrile. Styrene was detected by GC analysis
in several air samples. The inability to confirm by GC/MS has been noted
previously and is suspected to be due to instability of styrene. The mass
chromatogram with the identified peaks noted is shown in Figure 77.
Acetonitrile was detected in the azeotropic distillate of water samples
W-l and W-2 by GC analysis. It could not be confirmed by GC/MS-SIM due to an
extremely high background at m/e 41 and 39.
UNIROYAL, PAINESVILLE, OHIO
Field Sampling
Presampling Survey—
A presampling survey of the Uniroyal plant in Painesville, Ohio, was con-
ducted on May 22, 1978. The plant is located north of the city of Painesville,
Ohio. It is bounded on the north by vacant land, on the east by a Glyco chemi-
cal plant, on the south by Highway 535, and on the west by a Diamond Shamrock
facility.
The Glyco chemical plant produces fatty acids, hydrogenated tallow, hy-
drogenated vegetable oil, and distilled oleic acid. The Diamond Shamrock
facility produces metal coatings and electrodes. The Uniroyal plant produces
polybutadiene-acrylonitrile and butadiene-acrylonitrile rubber-PVC resin
blends. The discharge information indicated that the plant had two discharges
into the Grand River.
Air Sampling—
Air sampling was conducted using eight battery powered air samplers at
five sampling stations located upwind, downwind and laterally with respect to
the plant.
Sampling was conducted during two successive time periods. The first
sampling period was from 1000 May 23, 1978, to 0130 May 24, 1978. The second
sampling period was from 1030 May 24, 1978, to 1400 May 25, 1978. At some
point during the evening of May 23, 1978, all three samplers were taken from
the downwind sampling station A-4. The first set of samples were discarded
and a second sampling was conducted. The sampling station locations are shown
in Figure 78 and described in Table 99. Winds blew primarily from the north
and west during the second sampling period.
231
-------
TABLE 98. EICP DATA- FOR COMPOUNDS IDENTIFIED BY GC/MS IN AIR SAMPLE
FROM MONSANTO, ADDYSTON, OHIO
Sample m/e
No. I
MOH
A- 4
Xylenes
91,106,105
100,46,17
100,45,14
Ethylbenzene
91,106,51
100,32,13
100,45,7
Trimethylbenzene
105,120,119
100,59,15
100, 43, b/
Toluene
91,92
100,65
100,61
Dichlorobenzene
146,148,113
100,62,11
100,60,7
a./ The table contains the observed relative intensities of the three m/e values listed and being
used as the criteria for identification. The expected intensities are listed in the heading
with the m/e values.
b_/ Peak not seen due to low signal/noise ratio.
NJ
u>
-------
30000-1
NJ
U)
to
£0000-
10000-
7130 71MO 7150 7160
7240 dtl MOMSflflTO H-4 J.G.
jviTrr i i i i pr
0 0 7 £ I 0 ? df.ii
STEP SPEC«=1 INI- 1000
rpi~!Tt-| T i I i']~ri' i t I 3
7 £ 3 0 7 a M 0
7 c 5 0
Figure 77. Mass chromatogram showing additional compounds identified in Monsanto,
Addyston, Ohio, air sample A-4.
-------
81"
|Ljs!
if'
f^
out 1t*#f>
n L'fit't
i, _,jp^^PSs®''1KS"^
'' :• hfairPt ifalhV]^"/^ - H\ $ x . •• \<(3.1)
\r ••• !l-lpf;>.w^,rt;1'
-------
TABLE 99. AIR SAMPLING DATA FOR UNIROYAL, PAINESVILLE. OHIO
Location of sampling stations Total Sampling
Sample U.S. Geological
No. Survey coordinates
A-1 a/
A- IS-'
A- 2
A- 3
A- 4
A-5 .
a/
A-5D-
A-5S^
4623480 mN
480760 mE
4622370 mN
480300 mE
4622210 mN
480760 mE
4622580 mN
480850 mE
4622670 mN
479980 mE
sampling rate
Description (hr) (£/min)
On Hardy Road approximately 26.7 0.961
1200 m north of 535 26.7 0.921
On 535 approximately 500 m 26.8 0.814
west of Hardy Road
Behind Gee Gee's Restaurant- 26.7 0.969
Hardy Road and 535
On Hardy Road approximately 26.5 0.798
400 m north of 535
On 535 approximately 1200 m 26.7 0.955
west of Hardy Road 26.7 0.911
26.7 1.00
Total Sampling
volume height
(m3) (m)
1.54 1.5
1.47 1.5
1.31 1.5
1.56 1.5
1.27 1.5
1.53 1.5
1.46 1.5
1.61 1.5
a^l D = duplicate sample; S = spiked sample.
-------
Water Sampling—
Three grab water samples were taken on May 25, 1978. Contact was made
with Mr. Walter M. Iliff, the plant engineer, who informed us that the plant
effluent goes to a water treatment plant operated by Uniroyal and is then
pumped to the Lake County wastewater treatment facility in Mentor, Ohio. With
Mr. Iliff's help we obtained a sample at the influent to the Uniroyal plant
treatment facility and at the effluent from the Uniroyal plant treatment
facility. Some problem with a surfactant causing foaming was being experienced
at the Uniroyal treatment facility at the time of sampling. A sample of the
final effluent from the Lake County treatment facility was also obtained. The
sampling locations are described in Table 100.
Soil Sampling—
Soil samples were taken on May 25 at two locations. Sampling was accom-
plished by removing approximately 12 mm of topsoil from the area near sampling
stations A-4 and A-5. The sampling locations are shown in Figure 78 and de-
scribed in Table 100.
Meteorological Conditions
The meteorological conditions that existed during the sampling period
were taken from Lost Nation Airport, Willoughby, Ohio, and on-site. Wind data
were taken from Lost Nation Airport and temperature and humidity data were
taken on-site. The meteorological conditions are summarized in Table 101.
Sample Analysis
Air Sampling—
Eight air samples were collected at five sampling sites deployed near the
Uniroyal facility. The results of the analysis of the charcoal tubes are
given in Table 102. The AN levels in micrograms per cubic meter are also given
for the respective stations shown in Figure 78.
Figure 79 shows the wind rose pattern for the second sampling period.
Stations A-2, A-3 and A-5 were downwind of the plant, A-4 was crosswind and
A-l was upwind. The winds, however, were light when blowing and were calm 50%
O
of the time. The highest level of AN, 3.1 jJig/m , was found at the downwind
station A-2. AN was also found at the other two downwind stations A-3 and A-5.
The observation of AN at the upwind site A-l probably results from a differ-
ence in wind patterns from the site and from the Lost Nation Airport.
Sampling quality assurance was checked by the use of spiked sampling tubes.
The spiked samplers were deployed at stations A-l and A-5. The recoveries
were not calculated for these spiked samplers due to the presence of ambient
AN. Four blank tubes were spiked at the site with two levels of AN, 24 jug and
12 jug. These spiked blanks were capped and stored on dry ice until analyzed.
236
-------
TABLE 100. SOIL AND WATER SAMPLING DATA, UNIROYAL, PAINESVILLE, OHIO
Sample
No.
S-l (A- 4) a/
S-2 (A-5)-^/
W-l
W-2
Location of sampling
U.S. Geological
Survey coordinates
4622580 mN
480850 mE
4622670 mN
479980 mE
4622540 mN
479540 mE
4622500 mN
479450 mE
stations
Sample
Description type
On Hardy Road Soil
approximately
400 m north of
535.
On 535 approx- Soil
imately 1,200 m
west of Hardy Road.
Influent to Uni- Grab water
royal plant water
treatment facility.
Effluent from Uni- Grab water
royal plant water
Sample % Moisture
size by weight
250 cm3
250 cm3
1.5 H
1.5 X,
W-3
treatment facility.
Effluent from
Lake County waste-
water treatment
facility.
Grab water
1.5
a/ Sampling site designation on map (Figure 78).
-------
TABLE 101. METEOROLOGICAL CONDITIONS DURING SAMPLING AT UNIROYAL, PAINESVILLE. OHIO
N3
CO
oo
Time
Date (LST)
5/24/78 0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
5/25/78 0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
Wind-''
Direction
00-36
_
33
32
33
-
32
-
-
-
-
-
-
-
-
-
-
-
-
33
-
33
-
03
Speed Temperature^-'
(kts) (°C)
00
05
05 13
10
00
05 18
00
00
00
00
00
00
00
00
00
00
00 20
00
05
00
07
-
05
Humidity-/
Relative Absolute
(%) (mg/£)
_
-
100 11.2
-
-
75 11.9
-
-
-
-
-
-
_ _
-
-
-
62 11.8
-
-
-
-
-
— —
a./ Data from Lost Nation Airport, Willoughby, Ohio.
b/ Data taken on-site.
-------
TABLE 102. ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM UNIROYAL, PAINESVILLE, OHIO
Sample
No.
A-l
A-1S
A- 2
A- 3
A- 4
A- 5
A-5D
A-5S
A-BS
A-BS
A-BS
A-BS
A-B
A-B
Sampling
period
5/24/78 1104-
5/25/78 1343
5/24/78 1043-
5/24/78 1330
5/24/78 1050-
5/25/78 1335
5/24/78 1125-
5/25/78 1355
5/24/78 1035-
5/25/78 1320
- -
-
-
-
-
Air
volume Sample
(m3 ) typej!./
1.54 Front
1.47 Front
Back
1.31 Front
1.56 Front
1.27 Front
1.53 Front
1.46 Front
1.61 Front
Back
Front
Front
Front
Front
Front
Front
ygb/
Comments Found
1.4
Spiked with 24 yg AN 5.2
3.0
4.1
0.7
< 0.1
1.1
Duplicate 1.0
Spiked with 24 yg AN 7.3
2.0
Spiked with 24 yg AN 15.5
Spiked with 24 yg AN 14.0
Spiked with 12 yg AN 6.4
Spiked with 12 yg AN 6.4
Field blank < 0.1
Field blank < 0.1
GC/MS
yg/m confirmation^/
0.9
Recovery not
calculated
3.1
0.4
< 0.1
0.7
0.7
Recovery not
calculated
64% Recovery
58% Recovery
53% Recovery
53% Recovery
-
Yes
-
—
Yes
_
_
_
-
-
-
-
-
-
a./ All backup tubes were analyzed. Those not shown contained < 0.1 yg.
b/ Not corrected for recovery. Detection limit based on sample size and detector response.
c/ Yes = Confirmed presence or absence of AN; No = Unconfirmed, sample below detection limit.
-------
NNW
NNE
NW
WNW
WSW
SW
I
SSW
SSE
Winds Greater Than 10 Knots
Winds Less Than or Equal to 10 Knots
NE
SE
(Calm 50%)
ENE
ESE
Figure 79. Wind pattern during sampling at Uniroyal, Painesville, Ohio.
240
-------
These recoveries were: 24 g level - 64% and 58%; 12 g level - 53% and 53%.
No AN or interference was found on the blank tubes.
During GC analysis, styrene was detected and quantitated. These results
appear in Table 103 along with the corresponding AN levels for comparison. The
styrene levels were virtually identical with a range of 6.5 jug/m^ to 8.1 jug/m^.
The highest levels, 8.1 jJig/m and 7.6 /xg/m^ were found at station A-5. Styrene,
however, was not confirmed by GC/MS (see below).
Water Samples—
One of the water samples, W-l, was broken during shipment. Sample W-2
which represents the effluent from the Uniroyal water treatment plant was
analyzed after azeotropic distillation and found to contain 4,320 ppb AN.
This sample was reanalyzed by direct aqueous injection gas chromatography
(DAIGC) and found to contain 3,360 ppb AN. Sample W-3 represents the effluent
from the Lake County wastewater treatment plant which receives the effluent
(W-2) from the Uniroyal treatment plant. This sample was analyzed after azeo-
tropic distillation and found to contain 9.3 ppb AN. The results are listed
in Table 104.
Sample W-3 was spiked at 10 ppb by the addition of 5 £ig AN following the
first distillation. Additional methanol was added and the distillation re-
peated. Analysis of this distillate showed a recovery of 74%.
Soil Samples—
Two soil samples were extracted with water and the extracts analyzed by
direct aqueous injection. No AN was found above the 400 (Ug/kg detection limit.
These results are listed in Table 105.
Gas Chromatography/Mass Spectrometry Confirmation—
The confirmation of AN in two of the air samples was made by GC/MS oper-
ated in a high resolution, selection ion monitoring (SIM) mode. The presence
of AN in air samples A-l and A-2 was confirmed. Figure 80 gives the SIM plot
for these samples as well as for a 2-ng standard of AN.
The raw water sample W-2 as well as the azeotropic distillates of W-2
and W-3 were analyzed by GC/MS operated in the SIM mode. The ions at m/e 53
and 52 were monitored to detect acrylonitrile and the ions at m/e 41 and 39
were monitored to detect acetonitrile. Figure 81 shows the AN SIM for the three
samples and for a 1-ng AN standard. The presence of AN is unquestionably con-
firmed in all samples.
Air sample A-2 was analyzed by GC/MS using the full scan mode and EICP's
for eight other possible compounds.
241
-------
TABLE 103. COMPARISON OF ACRYLONITRILE AND STYRENE LEVELS IN AIR SAMPLES
FROM UNIROYAL, PAINESVILLE. OHIO
Sample
No.
AN Cone.
(yg/m3)
Styrene Cone.
(yg/m3)
A-l
A-2
A-3
A-4
A-5
A-5D
0.9
3.1
0.4
< 0.1
0.7
0.7
6.7
6.6
6.5
6.9
8.1
7.6
242
-------
.o
UJ
TABLE 104. ACRYLONITRILE CONCENTRATIONS IN WATER SAMPLES FROM UNIROYAL,
PAINESVILLE. OHIO: AZEOTROPIC DISTILLATION
Sample
No.
W-l
W-2
W-3
W-3Sd/
Initial
volume
(ml)
500
500
500
500
Final
volume
(ml)
10
10
10
pg
Found DAIGCb/
(original sample) yg/&.§/ yg/&
Broken in shipment
2,160 4,320 3,360
4.65 9.3
3.7 (74% Recovery)
GC/MS
conf irmation£/
_
Yes
Yes
aj Detection limit based on sample size and instrument response.
b/ Direct aqueous injection gas chromatography.
_c/ Yes = Confirmed presence or absence of AN; No = Unconfirmed, sample below detection limit.
d/ Spiked with 5 yg AN after first distillation.
TABLE 105. ACRYLONITRILE CONCENTRATIONS IN SOIL SAMPLES FROM UNIROYAL, PAINESVILLE OHIO
yg Found
Sample Sample Extraction (original sample) yg/kg
No. size volume DAIGC£/ DAIGC3./
So-1 10 g 25 ml < 4 <
So-2 10 g 25 ml < 4 < 400
a/ Direct aqueous injection gas chromatography.
b/ Detection limit based on sample size and instrument response.
-------
>-
•—•
(/I
2 Monograms Acrylonitrile
(X 1000)
»_•**" J*f*"* • "f J. *'J* M I U
"•"•^ ' /X ' J> ^.JAH?**™*
t 11 A >> A ^« * n * a at i i . A /
y-
»-^
VI
I T — i r
3 A
MINUTE.S
. | • , |
7 8
Uniroyal A-l
(X 1000)
M/e =53.0265
i i—j • r j-
0 1 2
j • i I i
a A
MINUTES
i
5
I
6
Uniroyal A-2
(X 1000)
I
2
I
3
MINUTELS
8
i j • T^
•7 8
Figure 80. SIM plots for Uniroyal, Painesville, Ohio, air samples A-l and A-2
and a 2-ng AN standard.
244
-------
I Nonogram Auylo
(X50)
M/. -52
I ' • I • ' I ' ' I ' ' I ' ' I ' ' I ' ' I ' ' I ' ' I ' ' I ' ' I
123436 78 9 10 11 12
MINUTE. 9
Uniroyo! W-2
(X 10)
M/t "52
1 ' I ' ' 1 ' ' 1 ' I ' ' I ' 1 ' I ' I ' ' 1 ' I ' ' 1 ' 1
0 12 3436189 10 11 12
MINUTES
'Jniroyol W-2 Oi,nllol»
(X I)
M/. -52
' ' I I ' I I I ' I I ' ' I ' I ' ' [ ' ' I ' ' I
0129498 18S1O1112
MINUTES
LMroyn! W-] OlirilldK
(X25)
M/« -53
• 52
''] ' ' I ' ' I ' ' I ' ' I ' ' I ' ' I ' ' I ' ' I ' ' I ' ' .
0 123498 T89 10 11 12
MINUTES
Figure 81. SIM plots for Uniroyal, Painesville, Ohio, water sample W-2.
water sample distillates W-2 and W-3 and a 1-ng AN standard.
245
-------
Air sample A-2 was found to contain isomers of xylene, isomers of tri-
raethylbenzene, ethylbenzene, toluene, and dichlorobenzene. The EICP data are
listed in Table 106. Not found were styrene, benzene, and acrylonitrile.
Styrene was detected by GC analysis in several air samples. The inability to
confirm by GC/MS has been noted previously and is suspected to be due to in-
stability of styrene. The mass chromatogram with the identified peaks noted
is shown in Figure 82.
Acetonitrile had not been detected in the GC analysis of the water
samples. The samples were nevertheless screened for acetonitrile by the GC/MS
protocol and the SIM plots are shown in Figure 83. Acetonitrile was definitely
found in the W-2 distillate. Its presence in W-2 raw or W-3 distillate could
not be confirmed.
VISTRON, LIMA, OHIO
Field Sampling
Presampling Survey—
A presampling survey of the Vistron plant, Lima, Ohio, was conducted on
June 12, 1978. This plant is located within Shawnee Township south-southwest
of Lima, Ohio. The plant is bounded on the north and east by a Sohio oil
refinery, on the south by a Shell Oil tank farm and Buckeye Road, and on the
west by Amanda Road. The Sohio plant at this location produces propylene and
elemental sulfur. The Shell Oil facility is primarily a propane storage
facility. The Vistron plant produces acetonitrile, acrylamide, acrylonitrile,
ammonia, ammonium nitrate, carbon dioxide, hydrogen cyanide, nitric acid, urea,
acrylic acid catalysts, acrylonitrile catalysts, and nitrile resin. The NPDES
permit indicates a discharge point into the Ottawa River approximately 10 m
north of the bridge at Adgate Road and Amanda Road. The discharge rate is
2.1 to 3.7 mgd.
Contact was made with Jay Killen, Hal Knieriemen, and Mr. Munko from
Sohio. Mr. Knieriemen and Mr. Munko conducted concurrent sampling and were
very helpful in providing access to some of the sampling points as well as
pointing out the AN facility and aqueous discharge points.
Air Sampling—
Air sampling was conducted using seven battery powered air samplers at
five sampling locations located upwind, downwind, and laterally with respect
to the plant. Samplers were located from 0 to 500 m from the plant perimeter
on refinery property, tank farm areas, and road rights-of-way and were posi-
tioned at a height of 1.5 m.
246
-------
TABLE 106. EICP DATA- FOR COMPOUNDS IDENTIFIED BY GC/MS IN AIR SAMPLE
FROM UNIROYAL, PAINESVILLE, OHIO
Sample m/e
No. I
URO
A- 2
Xylenes
91,106,105
100,46,17
100,42,17
Ethylbenzene
91,106,51
100,32,13
100,42,6
Trimethylbenzene
105,120,119
100,59,15
100, 44, b/
Toluene
91,92
100,65
100,100
Dichlorobenzene
146,148,113
100,62,11
100,67,22
aV The table contains the observed relative intensities of the three m/e values listed and being
used as the criteria for identification. The expected intensities are listed in the heading
with the m/e values.
b_/ Peak not seen due to low signal/noise ratio.
NJ
-P-
-------
30000-1
20000-
NJ
•O
C»
10000-
6630 63MO 69SO 6360 6370 6880
SPECS 6830 - 6-560 JM UNIRQYAL fl-2 J . G
6890
6900
6910 6 V 2 0 6930 6'? 4 0
STEP S P E C tf =1 IM T = 1000
9 7 0
Figure 82. Mass chromatogram showing additional compounds identified in Uniroyal, Painesville, Ohio
air sample A-2.
-------
I Monogram Ac«>onilril
(X20)
I ' ' I ' ' I
a a 10 ia 12
Uniroyol W-2
(X 10)
I • • I ' ' I ' • I
1 2 3 4 9 B 7 8 3 10 II 12
I • • I • • I • • I • • I
0123498 189 1O 11 12
Figure 83. Acetonitrile SIM plots for Uniroyal, Painesville, Ohio
water sample W-2, water sample distillates W-2 and W-3 and
a 1-ng acetonitrile standard.
249
-------
Sampling was carried out during the period from 1900 (EDT) June 13, 1978,
to 1800 (EDT) June 14, 1978. The sampling locations are shown in Figure 84
and described in Table 107. During sampling winds blew predominantly from the
north.
Water Sampling—
Grab water samples were taken at four locations; one upstream of the dis-
charge, one at the discharge from the treatment ponds, one at the discharge
into the river, and one approximately 700 m downstream of the discharge. Water
sampling sites are shown in Figure 81 and described in Table 108.
Soil Sampling—
Two soil samples were taken; one at air sampling station A-2 and one at
air sampling station A-5. Samples were collected by using a coring device
supplied by Vistron to remove a sample approximately 7.5 cm in diameter and
10.0 cm deep. Sampling locations are shown in Figure 84 and described in
Table 108.
Meteorological Conditions
The wind conditions that existed during the sampling period were obtained
from the Vistron plant meteorological station. Humidity readings were made
on-site using a sling psychrometer. The conditions are summarized in Table 109.
Sample Analysis
Air Samples—
Seven air samples were collected at five sampling stations deployed
around the Vistron facility. The results of the analysis of the charcoal
tubes are presented in Table 110. The levels in micrograms per cubic meter
are also shown in Figure 84. Figure 85 shows the wind rose pattern for the
sampling period. For comparison, the annual wind rose from the Star Program
for Findlay, Ohio, follows as Figure 86. As can be seen from the wind rose
pattern the wind blew predominantly from the north and north-northwest. As a
result, stations A-2 and A-3 were predominantly downwind with A-l and A-5
being upwind. A-4 was located in very close proximity (50 m) to the acrylo-
nitrile manufacturing site. Stations A-l and A-5 had no detectable AN. Sta-
tions A-2 and A-3 averaged about 14.5 and 16.1 ^g/m^, respectively. A-4 had
the highest level of AN, 141 /j,»/m^ (average of two stations) , probably due to
its close proximity to the AN manufacturing site.
Sampling quality assurance was checked by the use of spiked sampling
tubes. The spiked sampler was deployed at station A-5 which was upwind. Re-
covery of added AN at this site was 40%. Additionally, a blank was spiked at
the site and immediately stored on dry ice. The recovery on this tube was 71%.
Analysis of blank charcoal tubes revealed no interfering compounds nor acrylo-
nitrile above the detectable limit of 0.3
250
-------
7.5 MINUTE SERIES (TOPOGRAPHIC) /
ro
Ul
"t';'« - ' ?/'l_H ;;' *" |
^'-'i-4?^zj3.
.-. i. >^T- .'.I,-; I /!. .}[
I Cntni
riHwhScf/ :.
' *^ ;
j-=pQsgg?far4^_ ^^v
©
(w^J)
^-' • '^«J.-^'<^.--^n^ii^i>^-^Mii
^f—=^:^^fe^^^-17^44
u {T;. ., -is;."-• • /_£•££-''\j^ 3s Vw*ar-^vS^
;^-!--M
V^P^-::. \.J ^..^ [UJ
•><^^:--ii ::; W^b-Vr|jy
v,i...^iL-;-^tLvyfe^r^
^"•::;"^""^e^^
_£I^^^gJ|^a^^
f^MW^^W^^.
•{•:•:•}• ^k \-:'^:}; C(s~<.. ^J/fi?,"^'-^^-*
ira^-'r='"-:^t(i34
' Plant Site ^r^^1
•-.\i>'\^< _
'(16.1)
•'. --I /I / (.1.1" I ^ I~O- t * '•-••-. • I' * I J i
:-r ,i/fcyfK iVsgLJC ^^J^ tii
:V".""3^0
-------
TABLE 107. AIR SAMPLING DATA FOR VISTRON, LIMA, OHIO
.
Site
No.
A-l
A-2
A- 3
A-4
A-4D^
A-5
Location
U.S. Geological
Survey coordinates
4511170
742400
4510160
742730
4510450
742320
4510550
742510
4510850
741680
mN
mE
mN
mE
mN
mE
mN
mE
mN
mE
of sampling stations
Description
Tank farm area north of plant
perimeter
On Buckeye Road 400 meters east
of railroad tracks
Shell tank farm near tank L-18
Immediately east of acrylonitrile
facility
On Amanda Road south of Adgate
Road
Total
sampling
(hr)
21.4
21.7
21.7
21.5
21.5
21.8
21.8
Sampling
rate
(£/min)
1.10
0.99
0.82
0.95
0.75
1.07
0.85
Total
volume
(m3)
1.41
1.28
1.07
1.23
0.97
1.40
1.11
Sampling
height
(m)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
a/ D = duplicate sample; S = spiked sample.
-------
TABLE 108. SOIL AND WATER SAMPLING DATA, VISTRON, LIMA. OHIO
Location of sampling stations
Sample U.S. Geological
No. Survey coordinates ' Description
S-l
(A-2)5/
S-2
(A-5)^
W-l
W-2
W-3
W-4
4510160 mN
742730 mE
4510850 mN
741680 mE
4511540 mN
741750 mE
4511200 mN
741690 mE
4510980 mN
741670 mE
4510270 mN
741480 mE
On Buckeye Road 400 m east of rail-
road tracks
On Amanda Road south of Adgate Road
Ottawa River approximately 650 m
north of W-3
Discharge from biodegradation pond
Main discharge, 20 m north of Adgate
Road bridge
Approximately 700 m downstream from
W-3
Sample Sample % Moisture
type size by weight
Soil 250 cm3
Soil 250 cm3
Grab 1.5 £
water
Grab 1.5 £
water
Grab 1.5 H
water
Grab 1.5 I
water
a/ Sampling site designation on map (Figure 84).
-------
TABLE 109. WIND CONDITIONS DURING SAMPLING AT VISTRON, LIMA, OHIO
Date
6/13/78
6/14/78
._
Time
(LSI)
1900
2000
2100
2200
2300
2400
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
Wind^./
Direction
(00-36)
35
35
35
36
36
33
34
35
35
35
32
33
34
34
36
35
35
03
30
36
18
11
16
33
Speed
(kts)
09
04
04
03
04
02
04
04
04
04
02
02
02
03
03
00
02
04
02
02
02
02
02
02
Humidity^-/
Temperature^-/ Relative Absolute
(°C) (%) (mgM)
20 42 7.27
-
-
_ _ _
16.1 63 8.65
— _ _
— — _
— _ _
— _ _
_ _ _
— _ —
— _ —
12.7 70 7.84
_ _ ..
_ _ _
_ _ _
— _ _
17.8
— — _
_ _ _
— — —
— _ _
27.2 32 8.35
_
a/ From meteorological station at Vistron Plant, Lima, Ohio.
b/ Taken on-site.
-------
TABLE 110. ACRYLONITRILE CONCENTRATIONS IN AIR SAMPLES FROM VISTRON, LIMA. OHIO
Sample
No.
A-l
A- 2
A- 3
A-4
A-4D
A- 5
A-5S
A-BS
A-B
Air
Sampling volume Sample yg GC/MS
period (m^) type3.' Comments Found yg/m^— Confirmation^-
6/13/78
6/14/78
6/13/78
6/14/78
6/13/78
6/14/78
6/13/78
6/14/78
6/13/78
6/14/78
-
—
2025- 1.409 Front < 0.2 < 0.2
1750
1950- 1.280 Front 18.5 14.5
1730
1930- 1.073 Front 17.2 16.1
1720
2015- 1.229 Front 165 134
1745 0.975 Front Duplicate 145 149
1925- 1.400 Front < 0.3 < 0.2
1710 1.112 Front Spiked with 24.5 yg AN 9.8 40% Recovery
Front Spiked with 24.5 yg AN 17.5 71% Recovery
Front Field blank < 0.3
-
Yes
Yes
Yes
Yes
—
aj All backup tubes were analyzed. AN was < 0.2 yg/m^.
b/ Not corrected for recovery. Limit of detection based on sample size and instrument response.
c_/ Yes = confirmed presence or absence of AN; No = Unconfirmed, sample below detection limit.
-------
WNW
WSW
NW
I
I
SW
NNW
SSW
Winds Greater Than 10 Knots
Winds Less Than or Equal to 10 Knots
NNE
SSE
NE
SE
ENE
Figure 85. Wind patterns during sampling at Vistron, Lima, Ohio.
256
-------
NW
WNW
WSW
SW
NNW
SSW
• Winds Greater Than 12mph
I Winds Less Than 12mph
NNE
SSE
NE
SE
Figure 86. Annual wind rose for Findlay, Ohio (1/50 - 12/54) using Star
program output
ENE
ESE
257
-------
Water Samples—
The results for the analysis of the four water samples are listed in
Table 111. These samples were analyzed by the azeotropic distillation proce-
dure. No AN was detected in any of the water samples. Sample W-l was spiked
with 10.0 pg of AN following the first distillation. An additional 10 ml of
methanol was added and distillation repeated. The analysis of the distillate
showed a recovery of 90%.
Soil Samples—
Two soil samples were extracted with water and the extracts analyzed by
direct aqueous injection. No AN was found above the 100 Ug/kg detection limit.
These results are listed in Table 112.
Gas Chromatography/Mass Spectrometry Confirmation—
The confirmation of AN in the air samples was done by GC/MS using both a
high resolution SIM mode and a low resolution full scan mode. The presence of
AN in air samples A-2 and A-3 and the absence of AN in upwind sample A-5 was
confirmed in the high resolution SIM mode. The SIM plots for these samples
plus a 2-ng AN standard are shown in Figure 87.
Air samples A-4 and A-5 were analyzed by full scan GC/MS. AN was
readily confirmed in A-4 as illustrated by the EICP's shown in Figure 88 for
A-4 and a 4-ng AN standard.
No GC/MS analyses of water samples was necessary since no AN had been
found in the GC analyses.
Air samples A-4 and A-5 were analyzed by GC/MS using the full scan mode
and EICP's for eight other possible compounds. By this method, both air
samples were found to contain isomers of xylene, isomers of triethylbenzene,
ethylbenzene, toluene, and dichlorobenzene, and A-4 was found to also contain
benzene and acrylonitrile.
The EICP data are listed in Table 113 and the mass chromatograms with the
identified peaks noted are shown in Figures 89 and 90 for A-4 and A-5, re-
spectively.
Styrene was detected by GC analysis in several air samples. The inability
to confirm by GC/MS has been noted previously and is suspected to be due to
instability of styrene.
258
-------
TABLE 111. ACRYLONITRILE CONCENTRATIONS IN WATER SAMPLES FROM VISTRON. LIMA OHIO: AZEOTROPIC DISTILLATION
Sample Initial . Final ug Found
No. volume (ml) volume (ml) (original samples) Mg/£
W-l 500 .10 < 0.7 < 1.3§-/
W-1S£/ 500 10 9.0
(90% Recovery)
W-2 500 10 < 0.7 < 1.3
W-3 500 10 < 0.7 < 1.3
W-4 500 10 < 0.7 < 1.3
a/ Based on instrument detection limit and sample size.
b/ Yes = Confirmed presence or absence of AN; No = Unconfirmed, below detection limit.
N>
o _c/ Spiked with 10 ug AN.
TABLE 112. ACRYLONITRILE CONCENTRATIONS IN SOIL SAMPLES FROM VISTRON, LIMA,
yg Found
Sample Sample Extraction . (original sample)
No. size volume DAIGCJL'
S-l 10 g 50 ml < 1.0
S-2 10 g 50 ml < 1.0
GC/MS
confirmation^.'
-
-
-
OHIO
Kg/Kg
DAIGCa/
< look/
< 100
aj Direct aqueous injection gas chromatography.
_b/ Based on instrument detection limit and sample size.
-------
2 Nanogrami Acrylanifrile
(X 1000)
3 A
MINUTES
Villron A-2
(X 1000)
I I I I I I
123496
MINUTES
Vi,;.3B A-J
I A '000)
Mum.^ ^u ^^^^^u.
L^ i i \_fju* ' ' ' 'if i MI'HU i uy«_miuijrM_«>/
1 ' 1 ' ' 1 ' ' 1 ' ' 1 ' ' 1 ' ' I '
3 1 2 3 * 3 8
MINUTES
r+mf* ^^
' 1 ' ' 1 ' '
7 8
Vimo" 1-5
i.x 1000)
u. ,,,,,• ..... "^..•'^-•"-ni..
3 1
i j i i | i i | i
234
MINUTES
, , . . | i i | i
9 B 7
1 ' '
8
Figure 87. SIM plots for Vistron, Lima, Ohio, air samples A-2, A-3 and A-5 and
a 2-ns AN standard.
260
-------
30000-,
eoooo-
ioooo-
40 ng Acrylonitrile
11 n 11 n 1111 n p 111111111111111111111111111111111 [i i
6690 6700 6710 6730 6770 6740
SPECS 6690 - 7580 JO 4OH9 flCM J.6. 8/2:
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
6750 6760
6770 6780 6790 6500
STEP SPECS"! INT* 1000
68)0 6320
Vistron Air Sample
A-4
JUJWA.
52
7390 7400 7410 7480 7430 7440 7450 7460 7470 7480 7490 7500 7510
Figure 88. Extracted ion current plots for Vistron, Lima, Ohio, air sample A-4 and a 40-ng AN standard.
-------
TABLE 113. EICP DATA^/ FOR COMPOUNDS IDENTIFIED BY GC/MS IN AIR SAMPLE
FROM VISTRON. LIMA, OHIO
Sample m/e
No. I
VOH
A-4
VOH
A- 5
Xylenes
91,106,105
100,46,17
b/
100,45,15
Ethylbenzene
91,106,51
100,32,13
b/
100,45,8
Trimethylbenzene
105,120,119
100,59,15
0 /
100,35,7-'
100,64,14
Toluene
91,92
100,65
b/
b/
Acrylonitrile
53,52,26
100,94,36
100,57,29
—
Dicblorobenzene
146,148,113
100,62,11
„ /
100,43,--'
100,62,8
Benzene
78,52,77,51
100,25,23,16
100,63,69,67
—
a/ The table contains the observed relative intensities of the three m/e values listed and being used as the
criteria for identification. The expected intensities are listed in the heading with the m/e values.
b_/ Ion current saturated at the retention time.
cf Peak not seen due to low signal/noise ratio.
NJ
0>
N3
-------
30000-1
80000-
loeoo-
7390 7400 7410 7480 7430 7440 7450
SPECS ?390 - 7520 JM VISTROM fi-4 J.O. 6'23-'78
7460
7470 7430 7490 7500
STEP SPECS=1 IMT= 1000
Figure 89. Mass chromatogram for Vistron, Lima, Ohio, air sample A-4.
-------
a
i
J
30000-
20000-
NJ
CTi
.[s
JOOOO-
™
™
T
n i-r-i S
7£50 7260 7270 72S!) 7290 7300 7310
SPECO 7250 - 7380 JH VISTRON fl-5 J.G. 8.'23-'7S
7320 7330 7340 7350 7360
STEP SPEC«=l IHT= 1000
Figure 90. Mass chromatogram for Vistron, Lima, Ohio, air sample A-5.
-------
REFERENCES
1. Federal Register, 4510-26, Vol. 43,!(No. 11), pp. 2586-2625, January 17, 1978.
la. "Sampling and Analysis of Selected Toxic Substances," Final Report No.
EPA-560/6-76-021, Ethylene Dibromide, EPA Contract No. 68-01-2646.
"Sampling and Analysis of Selected Toxic Substances," Final Report No.
EPA-560/6-77-026, Vinylidene Chloride, EPA Contract No. 68-01-4115.
"Sampling and Analysis of Selected Toxic Substances," Final Report No.
EPA-560/6-78-001, Acrylamide, EPA Contract No. 68-01-4115.
2. Barrett, W. J., "Sampling and Analysis of Four Organic Compounds Using
Solid Sorbents," Southern Research Institute Report SORI-EAS-74-301,
U.S. Department of Health, Education and Welfare.
3. Bird, W. L., and C. H. Hale, Anal. Chem., 24, 586 (1952).
4. Daues, G. W., and W. F. Hammer, Anal. Chem., 29, 1035 (1957).
5. Petrova, L. J., Chem. Abst., 76, 127898y (1972).
6. "Acrylonitrile," Am. Ind. Hyg. Assoc. J.. 31, 529 (1970).
7. Aarato, S., Chem. Abst., 79, 57280b (1973).
8. Rogaczewska, T., Chem. Abst., 80, 21137r (1974).
9. Personal Communication, N. Jones, American Cyanamid, Fortier, Louisiana.
10. Steicher, R. J., Anal. Chem., 48, 1398 (1976).
11. "Acrylonitrile, Method No. 5156," National Institute for Occupational
Safety and Health, Analytical Methods, Set K.
12. Method D3371, Annual Book of ASTM Standards, Vol. 31, 1974.
13. Bellar, T. A., and J. E. Sigsby, Jr., Environ. Sci. Tech., 4, 150 (1970).
14. Lysyj, I., Anal. Chem., 32, 771 (1960).
265
-------
15. Bessing, D. W., Anal. Chem., 21, 1074 (1949).
16. Russkikh, A. A., Chem. Abst.. 78, 101566e (1973).
17. "Sampling and Analysis Procedures for Screening of Industrial Effluents
for Priority Pollutants," U.S. EPA, EMSL, Cincinnati, Ohio, March 1977,
18. Ponomarev, Y. P., Chem. Abst., 82, 34803 (1975).
19. Wronski, M., Chem. Abst., 81, 175988c (1974).
20. Stefanescu, T., Chem. Abst.. 80, 40770J (1974).
21. Russell, J., Environ. Sci. Tech., 13, 1175 (1975).
22. Personal communication, February 27, 1978, from Oscar Ramirez, Acting
Director, Surveillance and Analysis, Region VI, to Vincent DeCarlo
Supervisor, Special Actions Group, OTS (WH-557).
266
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
2.
3. RECIPIENT'S ACCESSIOf*NO.
4. TITLE AND SUBTITLE
ENVIRONMENTAL MONITORING NEAR INDUSTRIAL SITES:
ACRYLONITRILE
5. REPORT DATE
December 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
John E. Going, Phil Kuykendahl, Sue Long
Jon Onstot, Ken Thomas
8. PERFORMING ORGANIZATION REPORT NO.
^PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-01-4115
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
Office of Toxic Substances
Washington, DC 20460
13. TYPE OF REPORT AND PERIOD COVERED
Task Final 8/77-12/78
14. SPONSORING AGENCY CODE
EPA-OTS
15. SUPPLEMENTARY NOTES
16. ABSTRACT
A sampling and analysis program was conducted to determine the levels of acrylo-
nitrile in air, water, soil and sediment around 11 industrial sites. The industrial
sites included acrylonitrile and acrylamide producers, acrylic and modacrylic fiber
producers, ABS and SAN resin producers and nitrile elastromer producers.
Air was collected at the perimeters of the industrial sites as 24-composites using
charcoal adsorption tubes. Soil samples were collected at the air sampling stations.
Grab water samples were collected upstream, downstream and if possible at plant
discharges.
Air levels ranged from 0.1 to 325 /^g/m ; water levels ranged from nondetectable
to 4,300 ^g/liter. No acrylanitrile was found in the soils or sediments. The sample
extracts were analyzed by gas chromatography and confirmed by gas chromatography/mass
spectrometry.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Acrylonitrile
Air
Water
Soil
Sediment
Sampling
Analysis
Environmental Monitoring
Industrial Plants
Organic
Chemistry
13. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
318
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
267
------- |