:PA-560/6-77-026
ENVIRONMENTAL MONITORING
NEAR INDUSTRIAL SITES
VINYLIDENE CHLORIDE
•US-TECHNICAL INFORMATION CBNIER
\
AUGUST 1977
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF TOXIC SUBSTANCES
WASHINGTON, D.C. 20460
-------
EPA-560/6-77-026
SAMPLING AND ANALYSIS OF SELECTED TOXIC SUBSTANCES
Task I - Vinylidene Chloride
Contract No. 68-01-4115
Dr. Vincent J. DeCarlo
Supervisor, Special Actions Group
Office of Toxic Substances
U.S. Environmental Protection Agency
401 M Street, S.W.
Washington, D.C. 20460
Prepared for
Environmental Protection Agency
Office of Toxic Substances
Washington, D.C. 20460
October 1977
-------
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 con-
stitute endorsement or recommendation for use.
-------
CONTENTS .
Sections
I
II
III
IV
V
1
7
. 8
38
51
References 137
11
-------
FIGURES
No. Title
1 Chromatogram of vinylidene chloride and selected hydro-
carbons. 16
2 Calibration curve for vinylidene chloride 17
3 Chromatograms of unbrominated and brominated vinylidene
chloride and chloroethane on OV-101 column ...... 19
4 Chromatograms before and after bromination of 12 chlor-
inated hydrocarbons on OV-101 column ......... 20
5 VOA chromatogram of 12 chlorinated hydrocarbons on Car-
bowax 1500 column 23
6 Sampling Tjain 25
7 Recovery of vinylidene chloride versus volume of air .
sampled 28
8 Stability of vinylidene chloride in carbon disulfide . . 30
9 Bromination of vinylidene chloride: Effect of excess
Br 32
10 Bromination of vinylidene chloride: Effect of time. . . 32
11 Calibration curve for vinylid^ne chloride by VOA tech-
nique. 34
12 Release of vinylidene chloride from solids ....... 36
13 Geographical location of recommended sampling sites. . . 50
iii
-------
FIGURES (Continued)
No. Title Page
14 Sampling locations and concentrations of VDC in air
at Dow Chemical, Plaquemine, Louisiana 55
15 Wind patterns during sampling at Dow Chemical, Plaque-
mine, Louisiana 60
16 Extracted ion current plots - Dow air sample No. 13 .. 61
17 Sum plot of Dow air sample No. 13........... 63
18 Multiple ion scan of 3 ng VDC and 10 ml of Dow water
sample W-2 66
19 Sum plot of Dow water sample W-2. 69
20 Sampling locations and concentrations of VDC in air at
PPG Industries, Lake Charles, Louisiana 71
21 Wind patterns during sampling at PPG Industries, Lake
Charles, Louisiana 77
22 Extracted ion current plots for VDC in PPG air samples
Nos. 3 and 5 78
23 Sum plot of PPG air sample No. 3 81
24 Sum plot of PPG air sample No. 5 82
25 Sum plot of PPG water sample W-l. ........... 84
26 Mass spectrum of VDC in PPG water sample W-l 85
27 Sampling locations and concentrations of VDC in air at
Dixico, Dallas, Texas 88
28 Wind patterns during sampling at Dixico, Dallas, Texas. 93
29 Sampling locations and concentrations of VDC in air at
Dewey and Almy, Owensboro, Kentucky ......... 95
30 Wind patterns during sampling at Dewey and Almy .... 100
iv
-------
FIGURES (Concluded)
No, Title
3
31 Concentrations of VDC (^g/m ) in air at nine sampling
stations at Tennessee Eastman, Kingsport, Tennessee. . 102
32 Wind patterns during sampling at Tennessee Eastman . . . 107
33 Extracted ion current plot of Tennessee Eastman air
sample No. 9 for VDC 108
34 Sum plot of Tennessee Eastman air sample No. 9 110
35 Sampling locations and concentrations of VDC (lig/m ) in
air at Dow Chemical, Midland, Michigan 113
36 Extracted ion current plot of Dow Chemical air sample
No. 3 for VDC 118
37 Extracted ion current plot of Dow Chemical air sample No.
11 for VDC 118
38 Mass spectrum of VDC in Dow Chemical, Midland, Michigan
air sample No. 3 ........... 119
39 Mass spectrum of VDC in Dow Chemical, Michigan, air sam-
ple No. 11 120
40 Wind patterns during sampling at Dow Chemical. ..... 122
41 Sum plot of Dow Chemical, Michigan, air sample No. 3 . . 125
42 Sum plot of Dow Chemical, Michigan, air sample No. 11. . 126
43 Sum plot of Dow Chemical, Michigan, water sample W-3 . . 129
44 Multiple ion scan of VDC in drinking water 134
-------
TABLES
No. Title Page
1 Data Summary for Program Task 1 2
2 VDC Concentrations in Drinking Water from Four U.S.
Cities 4
3 Summary of Organic Compounds Found in Air Samples. ... 5
4 Summary of Organic Compounds Found in Water Samples. . . 6
5 Gas Chromatographic Conditions for VDC Analysis. .... 10
6 Retention Times on Durapak OPN Column. ......... 15
7 Retention Times on Carbowax 1500 Column 22
8 Laboratory Recovery Studies of VDC on Charcoal ..... 24
9 Field Sampling Breakthrough Studies 26
10 VDC Collected 27
11 Stability of VDC Adsorbed on Charcoal 29
12 Stability Study 35
13 Analysis of Solids . 37
14 Monomer and Polymer Synthesis and Polymer Processing
(1974) 39
15 Producers of Vinylidene Chloride Monomer
40
VI
-------
TABLES (Continued)
No. Title
16 Estimated Annual Emissions of VDC in Polymer Synthesis. . 41
17 Estimated Annual Emissions of VDC in Polymer Fabrica-
tion 42
18 Estimated Annual Emissions of VDC in Cellophane Coating . 43
19 Estimated Emissions of VDC in the Processing of PVDC
Polymers. ...... ........... 44
20 Major Manufacturers of Extruded PVDC Products 45
21 Location and Product Profile of Recommended Industrial
Sites 47
22 Properties and Previous Results of Finished Waters Se-
lected for Sampling 49
23 Sampling Schedule ....... 52
24 Field Sampling Summary. .... 53
25 Summary of Air Sampling Parameters. 53
26 Air Sampling Data for Dow Chemical, U.S.A., Plaquemine,
Louisiana 56
27 Weather Conditions During Sampling at Dow Chemical, USA,
Plaquemine, Louisiana ....... ..... 57
28 VDC in Air Samples from Dow Chemical, Plaquemine, Lou-
isiana. 59
29 Summary of GC-MS Analysis of Dow Air Sample No. 13. ... 64
30 VDC Concentrations in Water from Dow Chemical, Plaquemine,
Louisiana 67
31 Summary of GC-MS Analysis of Dow Water Sample W-2 .... 68
32 Air Sampling Data for PPG Industries, Lake Charles, Lou-
isiana. . ........... 72
vii
-------
TABLES (Continued)
No. Title Page
33 Weather Conditions During Sampling at PPG Industries,
Lake Charles, Louisiana 74
34 VDC Concentrations in Air Samples from PPG Industries,
Lake Charles, Louisiana. ................ 75
35 Summary of GC-MS Analysis of PPG Air Sample No. 3 79
36 Summary of GC-MS Analysis of PPG Air Sample No. 5 80
37 VDC Concentrations in Water from PPG Industries, Lake
Charles, Louisiana ...... 83
38 Summary of GC-MS Analysis of PPG Water Sample W-l 87
39 Air Sampling Data for Dixico, Dallas, Texas 89
40 Weather Conditions During Sampling at Dixico, Dallas,
Texas 91
41 VDC Concentrations in Air Samples from Dixico, Dallas,
Texas 92
42 Air Sampling Data for Dewey and Almy, Owensboro, Kentucky. 96
43 Weather Conditions During Sampling at Dewey and Almy,
Owensboro, Kentucky 97
44 VDC Concentration in Air Samples from Dewey and Almy,
Owensboro, Kentucky. ..... 98
45 Air Sampling Data for Tennessee Eastman Company, Kingsport,
Tennessee 103
46 Weather Conditions During Sampling at Tennessee Eastman,
Kingsport, Tennessee 104
47 VDC Concentrations in Air Samples from Tennessee Eastman,
Kingsport, Tennessee 106
48 Summary of GC/MS Analysis of Tennessee Eastman Air Sample
No. 9 109
viii
-------
TABLES (Concluded)
No. Title
49 VDC Concentrations in Water from Tennessee Eastman,
Kingsport, Tennessee 112
50 Air Sampling Data for Dow Chemical, Midland, Michigan . 114
51 Weather Conditions During Sampling at Dow Chemical, USA,
Midland, Michigan 115
52 VDC Concentrations in Air at Dow Chemical, Midland, Mich-
igan ... 116
53 Summary of GC/MS Analysis of Dow Air Sample No. 3 ... 123
54 Summary of GC/MS Analysis of Dow Air Sample No. 11. . . 124
55 VDC Concentrations in Water from Dow Chemical, Midland,
Michigan 127
56 Summary of GC/MS Analysis of Dow Water Sample W-3 . . . 130
57 Drinking Water Samples from Five U.S. Cities 131
58 VDC Concentrations in Drinking Water from Four U.S.
Cities 133
59 Solid Final Product Samples 135
60 VDC Concentrations in Solid Products 136
IX
-------
SECTION I
SUMMARY
The purpose of this program was to provide sampling and analysis
capabilities to EPA's Office of Toxic Substances, so that the levels of
suspected toxic substances in air, water, soil, and sediment at desig-
nated locations throughout the United States could be determined. The
first task was the sampling and analysis for vinylidene chloride (VDC).
Methods for sampling and analyzing VDC in air, water, and solids
were evaluated. A protocol was developed and approved.
Six industrial plants were selected for sampling. The plants repre-
sented manufacturers of VDC monomers, manufacturers of VDC polymers, and
fabricators of VDC polymers. Drinking water samples from four major U.S.
cities were collected. Several types of solid final products made from
VDC polymers or co-polymers were obtained from producers or purchased.
A summary of the results from the industrial plant samples is shown
in Table 1. The high and low concentration of VDC in air and water samples
for each plant are listed along with a description of the type of usage
of VDC. In general, VDC in air was found only at monomer and polymer pro-
duction facilities. The highest level detected, 51.8 u.g/m3 (0.014 ppm)
was found downwind of a monomer production plant at the property line.
At sites where VDC was found, its1 distribution correlated with the wind
behavior that existed during the sampling period. The greatest distance
at which VDC was detected away from plant property was 0.6 miles. VDC
was detected, however, at a station which was at a property line but known
to be 1.5 miles from the VDC production facilities.
The presence of VDC in the samples was confirmed by both (a) bro-
mination of VDC followed by gas chromatographic analysis of the result-
ing l,2-dibromo-l,l-dichloroethane, and (b) gas chromatography/mass spec-
trometry (GC/MS) analysis of the original VDC sample.
-------
Table 1. DATA SUMMARY FOR PROGRAM TASK 1
Site
Dow Chemical
Plaquemine, Louisiana
PPG Industries
Lake Charles, Louisiana
Dixico
Dallas, Texas
Dewey and Almy
Owensboro, Kentucky
Tennessee Eastman
Kingsport, Tennessee
Dow Chemical
Midland, Michigan
Process
Producer of VDC
Producer of VDC
User of VDC to produce
methyl chloroform
User of PVDC barrier
coating latex
Producer of barrier
coating latex
Producer of modacrylic
fibers
Use of VDC to produce
polymers
Fabricator of extrusion
resins
Air (ug/m3) Water (ug/l)
High Low High Low
9.6
51.8
ND
ND
< 1
25
ND
ND
ND
ND
ND
0.2
550
b/
ND
< 1
< 1
ND
ND
b/
ND
ND
ND
a/ ND = Not detected.
b/ No sample available,
-------
The highest level of VDC in water was 550 ppb, found in an industrial
wastewater canal. Generally, the level was less than 1 ppb and was detect-
able by GC/MS only.
Confirmation of VDC in the water samples was made by GC/MS analysis.
The results of analysis of drinking waters from four major U.S. cities
are summarized in Table 2. VDC was detected only in the Miami, Florida
sample. The results are confirmed since the samples were analyzed by GC/MS.
VDC was found in two samples of Saran Wrap at 4.9 and 58 ppm. No VDC
was detected in PVDC mono filaments, modacrylic fibers or meat packaging
film.
In the process of obtaining GC/MS confirmation of VDC in air and
water samples, the identities of other compounds in those samples were
established. Tables 3 and 4 summarize the compounds found in the air and
water samples, respectively.
-------
Table 2. VDC CONCENTRATIONS IN DRINKING WATER FROM FOUR U.S. CITIES
Sample VDC
Cincinnati, Ohio
Pre-chlorination ND
Post-chlorination ND
Lawrence, Massachusetts
Pre-chlorination ND
Post-chlorination ND
Miami, Florida
Pre-chlorination 0.059
Post-chlorination 0.045
Waterford, New York
Pre-chlorination a/
Post chlorination a/
Philadelphia, Pennsylvania
Pre-chlorination ND
Post-chlorination ND
ND = Not Detected
a/ Sample not received in time for scheduled analysis.
-------
Table 3. SUMMARY OF ORGANIC COMPOUNDS FOUND IN AIR SAMPLF.S
Dow Chemical
Compound Plaquemine, Louisiana
Vinylidene chloride X
trans-Dichloroethylene -
Hexanes
Cyclohexane or hexene
Methylene chloride
Carbon tetrachlorlde X
cis-Dichloroethylene
Chloroform ' ~ X
Trichloroethylene X
C-6 Methyl ketones
Methyl chloroform
Benzene X
Tetrachloroethylene X
Toluene X
1,1,2 -Trlch loroethane
1,2-Dichloropropane X
PPG Industries Dixico Dewey and Almy Tennessee Eastman Dow Chemical
Lake Charles, Louisiana Dallas, Texas Owensboro, Kentucky Kingsport, Tennessee Midland, Michigan
X XX
X
X X
X
X
X X
X
X
X
X
X X
X XX
X
X X
X
-------
Table 4. SUMMARY OF ORGANIC COMPOUNDS FOUND IN WATER SAMPLES
Compound
Methylenc chloride
Vlnylidene chloride
Dime thoxyme thane
Ch lorobromomethane
trans-Dich loroethylene
cis -Dichloroethy lene
Chloroform
Ethylene dichloride
Dihromome thane
Methyl chloroform
Carbon tetrachloride
1,2-Dichloropropane
Benzene
Trichloroethylene
l-Chloro-3-bromopropane
1 , 1 , 2 -Trich loroethane
Methyl cyclohexane
Bromoform
Tetrachloroethylene
Toluene
Dov Chemical PPG Industries
Plaquemine, Louisiana Lake Charles, Louisiana
X
X X
X X
X X
X X
X X
X X
X
X
X
X X
x x
X
X
X X
X
Dixico Dewey and Almy Tennessee Eastman Dow Chemical
Dallas, Texas Owensboro, Kentucky Kingsport, Tennessee Midland, Michigan Drinking waters
X
X XX
X
X
X
X
X
X
X
X
X
X
X
-------
SECTION II
INTRODUCTION
A concern about the environmental exposure to vinylidene chloride
(VDC) has developed recently, stimulated at least partially by its struc-
tural similarly to vinyl chloride and trichloroethylene, both carcino-
gens. Studies of the carcinogenicity of VDC with test animals, however,
have been contradictory and controversial. No clear understanding of its
threat to the environment is available at this time. It is also not known
to what extent the general population is being exposed to VDC.
On July 29, 1976, MRI Project No. 4280-C(1) entitled "Sampling and
Analysis of Selected Toxic Substances" was initiated. The objective of
this program was to provide the EPA with sampling and analysis capability
to determine the level of toxic substances in air, water, soil, and sediment
from designated sources and ambient locations throughout the United States.
The first task on this program was the sampling and analysis for VDC.
This report describes Task I of the program as follows: Section III,
Experimental Methods; Section IV, Site Selection; and Section V, Discussion
of Results.
-------
SECTION III
EXPERIMENTAL METHODS
SAMPLING PROTOCOL
Air Sampling
Air was sampled at 7 to 13 stations deployed around the industrial
sites. Meteorological data obtained from the National Climatic Center
was used to establish the most probable ind patterns for the time of
sampling. A small array of sites (three to five) was established at the
property line at an upwind, a downwind and two crosswind points. The re-
maining stations were positioned in an array upwind and downwind, with
most of the stations being downwind. If possible, the downwind sampling
stations were positioned such that samples were collected at the property
line and at a minimum of two different distances from the property line.
The sampling probes were set at 4 to 6 ft above ground at all stations
and air was sampled continuously for 24 hr.
The meteorological conditions that existed during each test obtained
from the nearest National Weather Station. Wind speed and direction, tem-
perature and precipation data were included in the data log.
The sampling train for the collection of VDC from ambient air con-
sisted of the following components assembled in the order given: (a)
probe (directed into the wind); (b) top charcoal tube; (c) back-up char-
coal tube; (d) critical orifice; and (e) 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 1.8 g 6/14 mesh
Fisher charcoal that was activated at 400°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 0.5 or 1 liter/min
by using a hypodermic syringe needle as a critical orifice and was checked
in the field every 3 to 4 hr. Power was supplied to the DC pump by two
24-v lantern batteries connected in parallel. The pump and batteries
8
-------
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 previous OTS studies.!.' 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, the tubes were kept on dry ice until
analyzed.
Water Sampling
Water samples were collected in 25 ml glass screw-top vials fitted
with caps lined with Teflon®-rubber laminate discs (Teflon® side down)
obtained from Pierce Chemical Company. All water samples were grab sam-
ples. The vials were filled with a water sample to overflowing and then
immediately sealed with the septa cap with no headspace. As the samples
were taken, they were placed on ice and then kept at ice temperature (4
to 6 C) until they were analyzed.
Consumer Product Sampling
The consumer products that were analyzed for residual monomer were
obtained from the manufacturer or purchased directly at a retail outlet.
ANALYSIS PROTOCOL
Air Samples
Exactly 4 ml of reagent grade C&2 was placed in a 7-ml vial filled
with a Teflon -faced septum and cooled on dry ice. The charcoal from the
adsorption tubes was slowly added to the CS£ with gentle swirling. After
all the charcoal has been added, the cap was replaced and the vial shaken
periodically. The desorption of the VDC from the charcoal was allowed to
proceed for at least 45 min. The VDC was analyzed by GC with flame ioniza-
tion detection using the conditions listed in Table 5 as AIR-PRIMARY.
For GC confirmation of identity, 1.0 ml of the CS2 solution was placed
in a 3.5 ml vial. An approximately 500-fold molar excess of B^ was added
based upon the analytical results obtained in the primary analysis. After
30 min, the excess B^ was removed by two extractions with an equal volume
of 0.5 M NaOH. The C&2 was dried by passage through a microcolumn of Na2SO^
and diluted to 1.0 ml if necessary. Analysis for brominated VDC was performed
by GC using the conditions listed in Table 5 as AIR-CONFIRMATION.
Additional confirmation was obtained by GC/MS operated under the fol-
lowing conditions using the AIR-PRIMARY column.
-------
Table 5. GAS CHROMATOGRAPHIC CONDITIONS FOR VDC ANALYSIS
Sample
AIR-PRIMARY
AIR-CONFIRMATION
Column
6 ft x 4 mm glass packed
with Durapak OPN on
80/100 Porasil C
6 ft x 1/8 in. stainless
steel packed with 1.5%
OV-101 on 80/100
Chromosorb W
Column
temperature
60°C for 2.0 min,
' 20°/min to 120°C.
Hold 10 min
90°C
Carrier flow
Nitrogen at 38
ml/min
(Helium at 30 ml/
min for GC/MS)
Nitrogen at 38
ml/min
Species
analyzed
VDC
BR VDC
WATER-PRIMARY
6 ft x 1/8 in. stainless
steel packed with 0.2%
Carbowax 1500 on
Carbosieve C
60°C for 3.0 min,
then 10°/min to
140°C. Hold 5
min
Nitrogen at 40
ml/min
VDC
WATER-CONFIRMATION
(GC/MS)
6 ft x 1/8 in. stainless
steel packed with 0.2%
Carbowax 1500 on
Carbosieve C
60°C for 3.0 min,
then 10°/min to
140°C. Hold 5
Helium at 30 ml/
min
VDC
SOLIDS-PRIMARY
6 ft x 4 mm glass packed
with Durapak OPN on 80/
100 Porasil C
90°C
Nitrogen at 38
ml/min
VDC
-------
Instrument:
Data System:
Separator:
lonization Potential:
Emission Current:
Multiplier Gain:
Mass Range:
Scan Rate:
.:• Resolution:
Sampling Rate:
Varian MAT 311A Mass Spectrometer/Varian 2700
Gas Chromatograph
Varian Spectro System 100 MS
Watson/Bieman
70 ev
1 milliamp
107
45-200
5.5 sec/decade
1,000
6 m7
Water Samples
The water samples were analyzed by the purge and trap technique. The
purging apparatus consisted ofa35mmx8mmID glass column (Glenco, Inc.)
ff?)
fitted with a fine glass frit at the bottom and a Teflon compression union
JS> ~
with Teflon rings at the top that held the Tenax-GC-packed tube. A stream
of prepurified nitrogen further purified by a charcoal trap was employed
at the rate of 38 ml/min to purge the water samples. A 3 in. x 1/4 in. stain-
less steel tube packed with 60/80 mesh Tenax-GC® (Applied Science Laboratories,
Inc.) was used to trap VDC purged from water. The water sample was poured
into the back of a 10 ml glass syringe (American Hospital Supply Corporation)
to overflowing, the plunger as then replaced and pushed to the 5 or 10 ml
level. This technique was used to avoid a headspace in the syringe. The
sample was introduced from the syringe into the top of the VOA column,
minus the Tenax® tube, and the plunger forced down. The Tenax tube was
replaced immediately and left in the system for 10 min. The packed tube
was removed from the VOA apparatus and put into the Bendix flasher oven
where it was immediately thermally desorbed onto the GC column. VDC was
analyzed by a flame ionization detector using the column and conditions
listed in Table 5 as WATER-PRIMARY.
Samples were confirmed by GC/MS using the same column described under
WATER-CONFIRMATION. The operation conditions when full mass spectra were
collected were as follows:
Instrument:
Data System:
Separator:
lonization Potential:
Emission Current:
Multiplier Gain:
Mass Range:
Scan Rate:
Re so lution:
Sampling Rate:
Varian MAT 311A Mass Spectrometer/Varian 2700
Gas Chromatograph
Varian Spectro System 100 MS
Watson/B
70 ev
1 milliamp
107
45-200
5.5 sec/decade
1,000
6 KHZ
11
-------
Some samples were analyzed by the same instrumentation but operated
in the Multiple Ion Scan (MIS) mode. The four most intense peaks in the
mass spectrum of VDC, 61, 96, 98, and 63, were monitored.
Solid Samples
A known weight (1 to 5 g) of the solid was placed in a crimp-seal
vial and sealed. The vial was heated to 100°C and held there for 90 min.
A 100-^1 sample of headspace was analyzed for VDG by GC using the condi-
tion listed in Table 5 as SOLIDS-PRIMARY. The samples were analyzed in
triplicate if sufficient material was available.
METHOD DEVELOPMENT FOR SAMPLING AND ANALYSIS
A meeting was held on August 31, 1976, at Cincinnati EPA laborator-
ies for the purpose of discussing the proposed sampling and analysis pro-
gram for VDC. The following were in attendance.
Andrew E. O'Keeffe Environmental Monitoring Support Laboratory,
Research Triangle Park (EMSL-RTP)
Hal Richter Office of Air Quality and Pollution Standards,
Research Triangle Park (OAQPS-RTP)
John Margeson Quality Assurance Branch, Environmental Mon-
itoring Support Laboratory, Research Triangle
Park (QAB-EMSL-RTP)
John Going Midwest Research Institute (MRI)
Perry Brunner Office of Toxic Substances (OTS)
Thomas A. Bellar Environmental Monitoring Support Laboratory,
Cincinnati (EMSL-CIN)
Jim Lichtenberg Environmental Monitoring Support Laboratory,
Cincinnati (EMSL-CIN)
John G. Cobler Dow Chemical, Midland, Michigan
William A. Coniglio Office of Toxic Substances (OTS)
After extensive discussion of alternate sampling and analysis procedures
the following items were a consensus of the group:
* The optimum air sampling material was expected to be activated
charcoal with carbon disulfide as a desorbing solvent.
* An electron capture detector was reported by Mr. Cobler, of Dow
Chemical, as being more sensitive than a flame ionization detec-
tor and was used by Dow Chemical. In a subsequent phone conversa-
tion, Mr. Cobler stated that in fact the electron capture detec-
tor was incompatible with carbon disulfide and that a FID was
actually used.
12
-------
* Mr. Cobler stated that VDC could be separated from carbon disul-
fide by GC using a Durapak OPN column. No significant interfer-
ences were reported.
* The problem of maintaining quality control was discussed. The
possibility of obtaining charcoal tubes loaded with a known amount
of VDC to be used as quality control checks was suggested.
* A protocol for the sampling and analysis ofVVDC in water was pre-
sented by Mr. Tom Bellar, EMSL-CIN. The procedure uses a generally
accepted approach for the analysis of volatile organics in water
samples. The preference to conduct continuous sampling rather than
discrete grab sampling, as in the VOA technique, was stated by
Dr. Coniglio. The possibility of using charcoal for continuous
water sampling will be investigated.
* The importance of verification of compound identification was
emphasized. Identification by two columns and/or two detectors
was considered as the minimum verification required. However,
confirmation by gas chromatography/mass spectrometry (GC/MS) was
the preferred technique.
Gas Chromatographic Conditions
Carbon Disulfide Extracts of VDC - As a result of the project meeting
on August 31, 1976, in Cincinnati, it was assumed that VDC collected in
air samples would most likely be analyzed in a carbon disulfide solution.
Thus, the initial and major analytical problem would be the gas chromato-
graphic separation of VDC and the solvent. A series of columns were tested
using flame ionization detection for their ability to retard either VDC
or carbon disulfide. A 12-ft column packed with 10% SP-2100 on 100/120
Supelcoport showed marginal success. At a column temperature of 30°C, VDC
eluted approximately 0.1 to 0.2 min ahead of the solvent. It was observed
that the actual retention time of VDC decreased as the volume of sample
injected was increased. Furthermore, the peak shape was very dependent
upon the injection technique and varied with operators. Ultimately, this
column was judged to be unacceptable.
A 6 ft x 4 mm glass column packed with 80/100 mesh Durapak OPN on
Porasil C was then found to be satisfactory. The following conditions were
found to be optimum for separation of VDC from the solvent and other chlor-
inated hydrocarbons.
13
-------
Columns: D.ual 6 ft x 4 mm glass packed with 80/100 mesh Durapak OPN
on Porasil C
Detectors: Dual hydrogen flame ionization operated in differential
mode
Carrier: Nitrogen at 30 ml/min
Column Temperature: 60°C for 2.0 min then programmed at 10°C/min
to 100°C and held at 100°C for 5 min
Injector Temperature: 200°C
Detector Temperature: 275°C
The chromatographic behavior of 14 chlorinated hydrocarbons, 2-propa-
none and diethyl ether was determined. The absolute retention times and
retention times relative to VDC are listed in Table 6.
A chromatogram showing the elution peaks of most the compounds is
shown in Figure 1. Methyl chloride, vinyl chloride, 2-propanone, and di-
ethyl ether were omitted from the mixed standard since they all co-elute
with carbon disulfide. The only compound which would seriously interfere
is chloroethane. This was not expected to cause any serious problems ex-
cept at the PPG monomer production facility at Lake Charles, Louisiana.
PPG produces chloroethane at this plant at the rate of 120 million pounds
per year.
A stock solution of VDC was prepared by injecting 199 u.1 of VDC into
10 ml of CS2 at 0°C. The resulting 5.0 ppt solution was stored in a freezer
held at -13°C. Working standards were prepared daily by injecting a known
volume of the stock standard into a small volume of CS2 in a septum sealed
vial. A calibration curve of VDC in CS2 was prepared using the conditions
previously described. A plot of peak area versus nanograms VDC, shown in
Figure 2 was linear from 13 to 550 ng. The detection limit was about 10 ng.
Brominated Carbon Disulfide Extracts of VDC - Samples of VDC in carbon di-
sulfide were brominated as a means of confirming VDC identity. The prin-
cipal purpose of the bromination procedure was to establish whether a GC
peak on the Durapak OPN column was due to the presence of VDC or chloro-
ethane. Chloroethane had been shown to coelute with VDC on the Durapak OPN
column and could interfere with the analysis. Thus, the behavior of chlor-
oethane during the bromination was studied. The behavior of the chlor-
inated alkenes, cis-dichloroethene, trans-dichloroethenet and trichloro-
ethene was also established. It was determined that a 6 ft x 1/8 in. stain-
less steel column of 1.5% OV-101 on Chromosorb W at 90°C could be used
to separate brominated VDC from CS.
14
-------
Table 6. RETENTION TIMES ON DURAPAK OPN COLUMN
a/
Compound t^—' (min) tR, VDC
Chloroethane 2.28 0.53
Chloroethene 2.50 0.58
2-Propanone 3.10 0.72
Diethyl ether 3.70 0.86
1,1-Dichloroethene 4.33
Chloroethane 4.33 1.00
trans-1,2-Dichloroethene 5.13 1.18
Dichloromethane 5.63 1.30
Tetrachloromethane 6.57 1.52
els-I,2-Dichloroethene 7.17 1.66
Trichloromethane 7.42 1.71
1,1-Dichloroethane 7.42 1.71
1,1,2-Trichloroethene 7.77 1.79
1,1,1-Trichloroethane 8.13 1.88
1,1,2,2-Tetrachloroethene 9.35 2.16
1,2-Dichloroethane 10.28 2.37
1,1,2-Trichloroethane 18.2 4.20
a/ Chromatographic columns and conditions: Dual 6 ft x 4 mm glass columns
packed with Durapak OPN on Porasil C, 80/100 mesh; column held at 60°C
for 2 min, then programmed to 100°C at 20°C/min.
15
-------
OQ
l-t
n>
o
tr
n
rr
O
OQ
H
o
i-h
(B
0)
o
n
O.
a
w
(D
h-1
(D
O
rt
(D
O.
O
o
(a
n
cr
o
3
en
tetrachloromet-hane
•*— chloromethane
•*— chloroethene
*— 2-propanone
•*— diethylether
1, 1-dichloroethene,
chloroethane
trans-1,2-dichloroethene
cis-1,2-dichloroethene
1, 1-dichloroethane, trichloromethane
1,1,2-trichloroethene
1,1, 1-trichloroethane
1, 1,2,2-tetrachloroethene
1,2-dichloroe thane
1,1,2-trlchloroethane
-------
80
60
<
LU
<
40
20
0
J
I
0
50 100 150 200
NANOGRAMS VDC
250
300
350
Figure 2. Calibration curve for vinylidene chloride
17
-------
A mixture of VDC and chloroethane in CS? was brominated by an es-
tablished procedure discussed later. Chromatograms of the unbrominated and
brominated solutions on the OV-101 column are shown in Figure 3. VDC and
chloroethane were not eluted from the OV-101 column. After bromination,
the 1,2-dibromo-l,1-dichloroethane peak appeared on the OV-101 column.
Experiments run with the individual compounds confirm these conclusions.
As a further test, a mixture of 12 chlorinated hydrocarbons was brominated.
The chromatograms are shown in Figure 4. No peaks prior to bromination
were present on the OV-101 chromatogram. After bromination, the VDC, di-
ch lo ro ethene s and the trichloroethene peaks diminished or disappeared on
the OPN column and appeared as noted on the OV-101 column. These peaks
were identified by experiments run with the individual alkenes.
The cis- and trans-dichloroethenes and the trichloroethene were not
completely brominated. This was not unexpected in that the bromination
conditions were not optimized for these compounds. Tetrachloroethene did
not appear to be brominated.
Although the bromination scheme was not intended as a means of quan-
titation, a calibration curve was prepared to show that the percent of
bromination was constant. Two-milliliter samples ranging from 5 to 1,000
ppm VDC were brominated as prescribed and chromatographed on the OV-101
column. A graph of peak height versus ppm VDC was linear. A sample of
pure 1,2-dibromo-l,1-dichloroethane could not be obtained to establish
the actual percent conversion} however, the fact that all the VDG disap-
pears on bromination indicates that the conversion is quantitative.
The brominated VDC was also chromatographed on a second column which
could be used for further confirmation. The separation of the 12 chlori-
nated hydrocarbons, after bromination, was performed using a 10 ft x 1/8
in. stainless steel column packed with 10% SP-2100 on 80/100 mesh Supel-
coport. The column temperature was 140 C and the nitrogen carrier flow
was 38 ml/min. As before, only the brominated di- and tri-chloroethenes
were separated from the solvent on this column.
Water Samples of VDC - The chromatographic conditions suggested by T. Bellar
were modified for use at MRI. A Varian Aerograph® 2400 gas chromatographic
equipped with a hydrogen flame ionization detector and a Bendix flasher
connected directly to the column inlet was used for the analysis. A
6 ft x 1/8 in. stainless steel column packed with 0.2% Carbowax 1500 on
80/100 mesh Carbopack C (Supelco, Inc.) was used under the following
conditions as the primary column.
Carrier: Prepurified nitrogen at 40 ml/min
Column Temperature: 60°C for 3 min, then programmed at 10°C/min
to 140°C and held 5 min at 140°C
Detector Temperature: 245° C
18
-------
\o
Brom incited 1,1-Dichloroethene
oa
3
D
-------
to
O
OQ
O C
Hi >-(
n>
!-•
N3 -P-
»-• n
o y
H 1-1
H- o
» g
03 05
It IT
ft) O
O< CQ
^ g
a. ca
H
o cr
O (B
M hh
1-1 O
a" H
O 0)
3
o> a>
3
o a-
3
o>
O hh
< rt
I (T>
)-• M
O
i-* cr
1-1
o o
o g
i_i i-..
C 3
O
3
C
D~
O
5'
Q
nT
Q.
Brominated 1, l-Dichloroethene
Bromi noted Cis-
and Trans-Dichloroethene
CD
s
Q
Brominated Tri
Chloroethene
-------
The chromatographic behavior of nine chlorinated hydrocarbons, 2-
propanone, and diethyl ether plus VDC was determined. These 11 compounds
were studied as possible interferents in the VDC analysis. The absolute
retention times and retention times relative to VDC are listed in Table 7.
A chromatogram showing the elution peaks of these 11 compounds and VDC is
shown in Figure 5. Chloromethane and chloroethane were not run on the
VOA system but had been directly injected onto the column previously and
were found to have retention times shorter than VDC.
Air Samples
Desorption Studies - Since it was established in the meeting at Cincinnati
that charcoal was the most promising trapping medium, this material was
used in all the air sampling experiments. The two charcoals evaluated were
Fisher coconut charcoal, 6/14 mesh; and Pittsburgh PCB, 12 x 30. Both were
activated and cleaned at 400°C for 1 hr in a stream of nitrogen.
The optimum conditions for desorption of VDC from charcoal were
studied briefly. Tubes containing 0.75 g charcoal were loaded by injecting
15 p.1 of 5 ppt VDC in CS2 into a stream of nitrogen, flowing through the
tubes. The VDC was desorbed with 1.5 ml CS2 at room temperature and at
dry ice temperature for 30 min. The percent recovery of VDC from each
charcoal at the two temperatures is summarized below.
% Recovery
30° C -78° C
Fisher 99 104
Pittsburg 108 104
Furthermore, the percent recovery of VDC was constant for both charcoals
at both temperatures from 15 to 90 min desorption time. As a precaution,
however, desorption will be performed at dry ice temperature for at least
45 min.
Elution Volume and Breakthrough Studies - A series of laboratory experiments
were conducted to test the elution volume of the charcoal adsorption tubes.
Charcoal tubes were loaded as before except that the nitrogen flowing
through the tubes was continued from 1 to 6 hr at flow rates varying from
0.2 to 1 liter/min. The VDC was then desorbed as described above and the
percent recovery determined. The results are summarized in Table 8.
Although the recoveries using the Pittsburg charcoal were slightly
higher, no significant difference was observed between the two. It was
decided to use the Fisher charcoal for future studies due to its easy
availability.
21
-------
Table 7. RETENTION TIMES ON CARBOWAX 1500 COLUMN
Compound
Chloroethene
Dichlorome thane
2-Propanone
1, 1-Dichloroethene
1, 1-Dichloroethane
trans-l,2-Dichloroethene
Diethyl ether
£is-l,2-Dichloroethene
T r ich lo rome th an e
1 , 2 -D ich lo ro ethane
1,1, 1-Trichloroethane
Tetrachlo rome thane
tR - (min)
1.7
3.2
3.6
4.35
5.65
5.65
5.9
6.4
6.95
7.35
8.3
8.4
tR/tR (VDC)
0.39
0.74
0.83
1.00
1.30
1.30
1.36
1.47
1.60
1.69
1.91
1.93
ji/ Chromatographic column and conditions: 6 ft x 1/8 in. stainless steel
column packed with 0.2% Carbowax 1500 on Carbopack C, 80/100 mesh;
column held at 60°C for 3 min, then programmed ato 140°C at 10 C/min.
22
-------
OQ
l-l
I
o
3-
o m
o> ft
1-1 o
cr OQ
o i-t
a o>
Ul
o
o
o
o
o
l-tl
o
3"
i-"
o
l-l
H-
3
0>
n-
o
D.
1-1
o
o
0)
1-1
cr
o
3
-------
Table 8. LABORATORY RECOVERY STUDIES OF VDC ON CHARCOAL
Recovery
Nitrogen flow Fisher Pittsburg
0.2 liter/min, 1 hr 96 109
0.5 liter/min, 1 hr 89 104
1.0 liter/min, 1 hr 93 102
1.0 liter/min, 6 hr 91 95
Extensive breakthrough experiments were then conducted using an in-
halation chamber at the MRI Deramus Field Station. An empty chamber was
operated with an atmosphere of approximately 15 ppm (v/v) VDC. The air in
the chamber was drawn from outside the building and passed through the
chamber at approximately 650 liters/min. Air was sampled through a port
in the side of the chamber using the train shown in Figure 6.
Sampling rates of 0.13, 0.20, 0.48, and 1.0 liters/min were used.
The adsorption tubes were packed with 2 g of charcoal and air was drawn
through them continuously for 24 hr. After 8 and 15 hr, the lower tubes
were replaced with fresh tubes for the first three experiments. All tubes
were then desorbed with CSo and analyzed for VDC. The lower tubes were
analyzed to determine the time and extent of breakthrough. The analysis
conditions were such that 10 ^g in any of the lower tubes would be read-
ily detected. The results are summarized in Table 9.
At a .sampling rate of 1 liter/min, a breakthrough of 37% occurred
after sampling 21 ppm VDC for 23.0 hr. No breakthrough had occurred at
0.48, 0.25, and 0.18 liter/min. The percent loading (wt/wt) of the top
and bottom tubes at 1 liter/min and for the top tube at 0.48 liter/min
was 3.6, 2.1, and 2.5%, respectively. The maximum loading before break-
through occurs appears to be approximately 2.57o. Sampling performed where
the VDC level is suspected to be greater than 10 ppm (v/v) should be done
at 0.48 liter/min. Sampling at sites where the VDC level is expected to
be much lower may be done at 1 liter/min in order to collect a larger
sample. Air at 10 ppb (v/v) sampled for 24 hr at 1 liter/min would pro-
vide approximately 60 |ig VDC for analysis. This quantity of VDC desorbed
with 4 ml of CS2 would be easily determined by the GC analysis procedure.
24
-------
Air Flow
Inhalation Chamber
X
[
[
F
>
^ B
1 I
] o
1
: F
U \
n .*
\ „
Pur
1
c
( D
D
I
F
LJ \
ri ^
ID
r
Top Charcoal Tubes
1
Bottom Charcoal Tubes
Flowmeters
_J KI.. .11 . \i .. i. .. .
T Needle Valves
Figure 6. Sampling Train
25
-------
Table 9. FIELD SAMPLING BREAKTHROUGH STUDIES^/
Top
Bottom,
Bottom,
Bottom,
Top
Bottom,
Bottom,
Bottom,
Top
Bottom,
Bottom,
Bottom,
Top
Bottom,
Flow rate Volume of
Tube (jfc/min) Air (4)
0.18 250
0-8 hr
8-15 hr
15-23.5 hr
0.25 367
0-8 hr
8-15 hr
15-23.5 hr
0.48 677
0-8 hr
8-15 hr
15-23.5 hr
1.00 1,380
0-23 hr 1.00
VDC found
(mg) UR/A
18.9 75.6
ND
ND
ND
30.3 82.6
ND
ND
ND
49.3 72.8
ND
ND
ND
71'2 \ 81.8
41.7 J
ppm
(v/v)
19.2
-
-
-
20.8
-
-
-
18.4
-
-
-
20.7
Note: ND = Not detected.
a/ Air temperature - 70 to 72°F.
Humidity - 39 to 42%.
26
-------
Recovery at Various Loading Levels (Collected from Air) - An experiment
was performed to determine the percent recovery at various levels of load-
ing. Since the real VDC level in the chamber was not known at the time,
the volume of air sampled was recorded. The volume ranged from 0.5 to
100 liters. The amount of VDC collected was determined and is listed in
Table .10 along with the volume of air sampled. These data are also graphed
in a log-log form in Figure 7. The three data points from the breakthrough
experiment have been added to the graph although these samples were col-
lected with a different sampling train over a longer time period. The fact
that a linear curve is observed is evidence that the percent recovery was
essentially constant over the range of 35 to 49,300 vtg VDC. Previous exper-
iments had also indicated that recovery was essentially complete.
Table 10. VDC COLLECTED
Volume of air VDC found
sampled
0.50 35
1.25 83
2.50 134
12.5 960
25.0 1,670
100 5,400
VDC Stability on Charcoal - The final experiment was designed to establish
how long VDC trapped on charcoal is stable. Sixteen tubes were loaded at
the inhalation chamber with 3.00 liters of air each. After loading,' the
tubes were plugged with corks, wrapped in aluminum foil, and stored in a
freezer at 4°C. The results of duplicate samples analyzed periodically for
up to 16 days after preparation are listed in Table 11. The recovery was
constant for up to 16 days of storage.
VDC Stability in CS2 - The stability of dilute solutions of VDC in CS2 was
determined. Three standards, each containing 100 ppm VDC and 1,1-dichloroethane
(DCEA) were prepared and stored in a freezer, a refrigerator, and at room
temperature. DCEA was added as an internal standard. The solutions were
analyzed over a period of 7 days and compared to a comparable standard pre-
pared fresh daily. The results are plotted in Figure 8 as the ratio of the
areas of VDC and DCEA and as the parts per million VDC found.
27
-------
NJ
00
1000 r-
100
J2
~o
^ '°
<
14-
o
E
1.0
0.1
10
100
I
i Collected 10/5/76, AM
i Collected 10/5-10/6
I
J
1000
VDC Found, Micrograms
10,000
100,000
Figure 7.. Recovery of vinylidene chloride versus volume of air sampled
-------
Table 11. STABILITY OF VDC ADSORBED ON CHARCOAL
Days after loading VDC found —
2 187
5 192
8 195
12 180
16 195
aj Average of two analyses,
29
-------
u
u
4>
1.0
0.8
0.6
0.4
0.2
0.0
TOO
a.
a.
c- 80;
S 60
"c
§ 40
o
U
u 20
Q
0
0
a = Stored at -18°C
b = Stored at 4° C
c = Stored at 25° C
4
Days
Figure 8. Stability of vinylidene chloride in carbon disulfide
30
-------
Although some scatter was observed, two trends were observed. The
ratio of peaks was relatively constant at all temperatures while the ac-
tual level of VDC decreased about 20% in the 7 days. This suggests that
both VDC and DCEA were being lost at about the same rate, presumably by
volatilization. As a result, standards are prepared fresh daily and sam-
ples were analyzed within 4 days.
Bromination of VDC - The presence of VDC in air samples must be confirmed
and attempts to find a second GC column for confirmation were not success-
ful. An alternate approach based upon the bromination of VDC to form 1,2-
dibromo-1, 1-dichloroethane was developed.
A preliminary experiment established that excess Br2 was easily re-
moved from the CS2 by extraction with 0.5 M NaOH. A solution of 0.5 M Br2
in CS2 was completely debrominated by extracting twice with an equal volume
of 0.5 M NaOH.
The two principal factors that were expected to affect the extent of
bromination of VDC were (a) bromine concentration and (b) time. A solution
of 100 ppm VDC in CS2 was used to test these two variables. Initially, an
increasing quantity of Br2 was added to 2 ml of 100 ppm VDC in CS2» Bromi-
nation was allowed to proceed 30 min after which the excess B^ was removed
by extraction with 0.5 M NaOH. The C$2 was then dried by passage through
a microcolumn of Na2S04 and analyzed for brominated VDC. The results shown
in Figure 9 indicate that a 400-fold molar excess is required for maximum
bromination. Subsequently an experiment to test the effect of bromination
time was performed at a Br2/VDC ratio of 470. The results shown in Figure
10 indicate that the bromination reaction was quite rapid and was judged
to be essentially complete after 30 min. The optimum bromination conditions
were considered to be a 400 molar excess of Br2 and a reaction time of
30 min.
Water Samples
Purging Conditions - The volume of nitrogen required to completely purge
the VDC from the water sample was determined using 5 ml of a 24-ppb solu-
tion. Quantitative sparging was achieved between 5 and 10 min using a gas
flow of 38 ml/min. Sparging for 15 min at 38 ml/min resulted in a 30% loss
of VDC. The optimum procedure was judged to be a purge time of 10 min at
38 ml/min.
31
-------
100
50
o
(U
a.
0 200 400 600 800 1000
H
[VDC]
Figure 9. Bromination of vinylidene chloride: Effect of excess Br,
150
100
_D)
'oj
O
0)
0.
50
10
20 30
Time, Minutes
40
50
60
Figure 10. Bromination of vinylidene chloride: Effect of time
32
-------
Standards and Calibration - A 12 ppt stock solution of VDC was made by in-
jecting 1 ml of VDC (density = 1.218 g/ml) into 99 ml of 1-propanol. This
solution was sotred in a refrigerator at 0°C. Working standards were pre-
pared daily by injecting 25 nl of the stock standard into 100 ml of tap
water resulting in a 3 ppm standard of VDC. In preparing VDC standards in
water and in collecting samples, no headspace can be allowed to exist.
After standard and sample preparation were carried out in the necessary
volumetric flasks, clean glass vials were filled to overflowing with the
respective solutions and sealed with caps containing Teflon®-rubber lam-
inate septa (Pierce Chemical Company), so that ero headspace occurred.
Aliquots of from 10 to 100 |il of the 3 ppm solution in water were in-
jected with 100 lil syringe into 5 ml of water in the VGA system and the
VDC was purged as described previously. A calibration curve was generated
on the Carbowax 1500 column and a plot of peak height (in mm versus ppb
VDC) is shown in Figure 11. The detection limit was about 2 ppb and the
curve was linear from 2 to 60 ppb.
Sample Storage and Stability - An 18-ppb standard of VDC in water was made
by pipetting 1.5 ml of the 3 ppm standard in water into 250 ml of water in
a volumetric flask. Five milliliters of the solution were analyzed immediately
and the rest left to stand open in a hood. Samples were transferred to the
VOA system by pouring the solution into the back of a 10 ml glass syringe
to overflowing, then replacing the plunger and inserting it to the 5 ml
level. This technique is used to insure that no headspace exists above the
water during the transfer. A 5 ml aliquot of the open solution was analyzed
on the VOA system after 25 min, 45 min, 2, 3, 4, and 5 hr. A calibration
curve was prepared from the same stock solution.
The results showed a rapid loss of VDC from open aqueous solutions
and confirm the requirement that water samples and standards must be stored
with zero headspace.
Stability in another study of stability of VDC in water, a standard
was prepared and poured to overflowing in five vials which were sealed with
no headspace with TefIon®-rubber discs. One sample was analyzed immediately
on the VOA system using the syringe technique described above and was found
to be 12.5 VDC. The remaining four samples were stored in a refrigerator
(at 4°C) and analyzed at intervals over a 9-day period.
The results of this study are listed in Table 12. It is apparent that
r O
VDC samples can be stored at 4 C for at least 9 days if no headspace is
permitted.
33
-------
300 r
10
20 30 40 50
Parts per Billion, 1, 1-Dichloroethene
70
Figure 11. Calibration curve for vinylidene chloride by VOA technique
-------
Table 12. STABILITY STUDY
Sample age (days) Concentration (ppb)
0 13
3 15
6 11
9 16
Solid Samples
It was anticipated that a variety of commercial products manufac-
tured from PVDC would be analyzed for residual monomer. The two general
approaches are to either dissolve the product in an appropriate solvent or
to thermally desorb the VDC from the product in a closed vessel and analyze
the head space. Because of the wide variety of products with differing
properties that will be tested, the second approach was judged to be the
more promising.
A sample of PVDG/PVC powder was obtained from Aldrich Chemical Com-
pany, Milwaukee, Wisconsin, and used to evaluate the protocol. Samples of
Saran Wrap, produced by Dow Chemical and purchased at a local grocery store,
were also used in the evaluation. One- to 3-g quantities of the product
were placed in a 14.6 ml crimp-seal vial which was then sealed. The vials
were heated at 50, 75, 100, and 150°C for 15 to 90 min after which a 100-^.1
sample of head space gas was removed and analyzed. The gas chromatographic
conditions were as follows.
Column: Glass 6 ft x 4 mm packed with 80/100 mesh Durapak OPN on
Porasil C
Detector: Hydrogen flame ionization
Carrier: Nitrogen at 38 ml/min
Column Temperature: 90°C
Injector Temperature: 200°C
Detector Temperature: 275°C
The results obtained using the powder polymer are graphed in Figure
12. After approximately 30 min at 150°C, the polymer showed obvious signs
of decomposition and VDC was no longer found in the head gas. At 50'and 75°C,
very little VDC was released. Similarly, Saran Wrap showed extensive decom-
position at 150°C and little VDC release at 50°C. The optimum conditions
were judged to be a heating time of 90 min at a temperature of 100°C.
35
-------
lOOr
80
60
o
-------
This agrees well with the conditions employed by Dow Chemical of heat-
ing 60 min at 80°C.—' Samples of a PVDC/PVG powder obtained from Aldrich
Chemical Company, Milwaukee, Wisconsin, and Saran Wrap produced by Dow
Chemical and purchased at a local grocery store were analyzed by this
procedure. The results are summarized in table 13.
Table 13. ANALYSIS OF SOLIDS
Sample
Sample
weight
Sample
volume
Saran Wrap
Saran Wrap
1.86 g
2.00 g
1.4 cm 3
1.5 cm 3
VDC (
PVDC powder
PVDC powder
PVDC powder
1.00 g
1.00 g
1.00 g
0.7 cm 3
0.7 cm 3
0.7 cm 3
10.4
9.3
10.0
60
56
9.9 Average
58 Average
The standard deviation and relative standard deviation for the PVDC
powder analysis was 0.6 Hg/g and 6%, respectively. The VDC level of the
Saran Wrap was higher than expected. However, code numbers on the box in-
dicate that the wrap was produced in 1974 when residual monomer levels
were higher than present-day levels.
37
-------
SECTION IV
SITE SELECTION
CRITERIA FOR THE SELECTION OF SAMPLING SITES
The objective of this task was to determine the exposure of the gen-
eral public to 1,1-dichloroethene in populated areas adjacent to the most
probable sources, from fabricated products, and from chlorinated drinking
waters. The selected sampling locations were representative of the total
industrial locations that are sources of VDC. The selected fabricated
products and drinking waters were also representative of the total potential
exposure by those media.
Criteria for the Selection of Industrial Sites
The selection criteria for choice of industrial locations were in-
tended to achieve representative sampling, but at the same time, provide
sites that were most likely to have detectable levels of VDC present.
The criteria for the selection of industrial sites for sampling were:
* Estimated emission of VDC
* Polymerization technology
* Geographical location
Estimated Emissions of VDC - An estimate of the annual emissions of VDC
from the various segments of the monomer synthesis, polymer synthesis and
polymer processing industries has been reported. A summary of the estimated
annual emissions is given in Table 14. According to this estimate, the ma-
jority of the emissions, 82.6%, are associated with monomer synthesis,
while 16.7 and 0.7% occur with polymer synthesis and polymer fabrication,
respectively.
Monomer is presently produced in the United States by Dow Chemical
and PPG Industries at the plant locations listed in Table 15. Although
PPG produces almost twice as much as Dow, approximately 155 million pounds
38
-------
Table 14. MONOMER AND POLYMER SYNTHESIS AND POLYMER PROCESSING (1974)2/3/
Co
vO
Process
Monomer synthesis
Polymer synthesis
1. Latex for barrier coatings
2. Latex for miscellaneous coating
3. Synthetic fibers
4. Coating resin for cellophane
5. Extrusion resin (emulsion process)
6. Extrusion resin (suspension process)
Fabrication or polymer processing
1. Coating cellophane
2. Coating plastics, paper and glassine
3. Extrusion
4. Miscellaneous coating
Annual VDC
1,000 kg
1,523k/
308
55c/
68-.
73-
82
12
18
13.7
0.7
7.3
0-2 /
5.5-
emissions
1,000 Ib
3,355k/
679
12 0 ,
C.I
160d/
182-'.
ct /
27f
4(£X
30.4
1.6
16.4
0.4 .
12. 0»
VDC consumption
(million
112
20
15
16
25
21
15
Ib)
Total
1,845
4,064
a/ This VDC emission inventory does not include emissions of VDC from the conversion of vinylidene chlo-
fride into 1,1,1-trichloroethane,because this conversion is not considered within the scope of this
study.
b/ New emission-control technology will be installed in one of the plants during the latter part of 1975<
On this basis, the annual VDC emissions should drop to 611,000 Ib (277,000 kg).
c/ While all emissions listed in this table are estimates, the degree of certainty varies. These esti-
mates are based on minimal data and are, therefore, more uncertain.
d/ These products result in vinyl chloride emissions. The annual emissions are 1,800 Ib.
e/ These products result in vinyl chloride emissions. The annual emissions are 105,120 Ib or 47,700 kg.
f/ These products result in vinyl chloride emissions. The annual emissions are 43,435 Ib or 19,700 kg.
-------
Table 15. PRODUCERS OF VINYL1DENE CHLORIDE MONOMER^/
Manufacturer Plant location Production
PPG Industries Lake Charles, Louisiana 170 to 175 x 10^ Ib
Dow Chemical Freeport, Texas 90 to 95 x 10^ Ib
Plaquemine, Louisiana 90 to 95 x 10 Ib
or 89% is used captively for the synthesis of methyl chloroform. The re-
maining 20 to 25 million pounds plus 20 to 25 million pounds from Dow Chem-
ical are sold to the merchant market. Dow retains approximately 70 million
pounds for their own polymer synthesis. The estimate of emissions has been
reported for the monomer industry, however, specific plants were not iden-
tified in the report. The average industry emission rate, based upon 1974
data, was 1.3 Ib of VDC/100 Ib of VDC manufactured. This may currently be
less because one of the plants was scheduled to install improved emission
control systems in late 1975. The emissions for the industry would then
drop to 610,000 Ib annually. Nevertheless, the plants producing monomer
were given high priority in the selection of sampling sites.
The polymer synthesis industry can be conveniently divided into five
groups based upon the end-use of the polymer produced. The emissions esti-
mates have been based upon these five subcategories and are presented in
Table 16. While the manufacturers within each subcategory were known, the
emissions for the specific plants were not available. The largest companies
within each area were known and as such were favored as potential sampling
sites. Table 16 also lists the top one or two companies within each category.
The synthetic fiber industry is unique in that they manufacture the VDC co-
polymer and then fabricate it into a filament. These filaments are marketed
to the apparel and carpet industries where nonflammable fabrics are desired.
Emissions from the fabrication segment of the industry account for
only a small fraction of the total and are distributed among a large num-
ber of fabricators. The estimated emissions from four types of fabricator
industries are listed in Table 17.
The three companies that manufacture coated cellophane and the loca-
tion of their plants are given in Table 18. The sizes of the companies are
in the order listed. The emissions from cellophane plants appear to be
fairly low.
40
-------
Table 16. ESTIMATED ANNUAL EMISSIONS OF VDC IN POLYMER SYNTHESIS^
3/
Polymer category
Barrier-coating latex
Miscellaneous coatings, latex
Synthetic fiber
Cellophane coating resin
Thermoplastic resins for
extrusion
a/ Estimated.
Leading plants
W. R. Grace, Owensboro, KY
Morton Chemical Company
Ringwood, IL
Dow Chemical, Midland, MI
Standard Brands Chemical
Industries
Cheswold, DL
Kensington, GA
Tennessee Eastman
Kingsport, TN
Monsanto, Decatur, AL
Dow Chemical
Dow Chemical
Total annual emissions
for the industry (Ib)
120,000
150,000
Emission of VDC (Ib) per
100 Ib of VDC polymerized
0.60
160,000
182,000
27,000 (emulsion)
40,000 (suspension)
IS/
0.7
0.13
0.26
-------
Table 17. ESTIMATED ANNUAL EMISSIONS OF VDC IN POLYMER FABRICATION-
3/
Fabrication area
Glassine coating
Paper coating
Paperboard coating
Plastic coating and
miscellaneous
Leading firm (total
No» of companies)
Dixico
Dallas, Texas
(13 glassine and paper)
Crown Zellerbach
Portland, Oregon
(13 glassine and paper)
Unknown
(6)
3M
Decatur, Alabama
(polyester)
Allied
Pittsburg, Pennsylvania
(nylon)
Hercules
Covington, Virginia
(polypropylene)
Curwood
New London, Wisconsin
(polyethylene)
(12)
Consumption of
latex, dry bases (Ib)
7,500,000
5,500,000
500,000
7,000,000
Annual VDC
emissions (Ib)
5,900
4,400
400
2,160
-------
Table 18. ESTIMATED ANNUAL EMISSIONS OF VDC
IN CELLOPHANE COATING3-/
Industry consump- Estimated annual
tion of latex VDC emissions
Leading firms (dry basis, Ib) (lb)
Du Pont; Clinton, Iowa
Tecumseh, Kansas
FMC; Fredricksburg, Virginia 26,000,000 1,560
Marcus Hook, Pennsylvania
Olin; Pisgah Forest,
North Carolina
Covington, Virginia
Barrier-coating latex is used by the packaging industry to prepare
heat-sealable coatings on paper and glassine, on paperboard and on plas-
tic films. The final products are used largely to package food products.
Of the over 30 companies reported to be barrier-coating latex users, two,
Du Pont and Olin, are also manufacturers of the latex. The estimated an-
nual emissions for the companies comprising the four types of coating
substrates are given in Table 19 along with the company(s) that is a ma-
jor factor in the market. A fabricator in the glassine or paper coating
area is more likely to have a detectable level of VDC emissions.
Thermoplastic PVDC resin is primarily extruded into film products by
the companies listed in Table 20. Amtech is the only company producing
PVDC monofilament. Cryovac is the major consumer of extrusion resin produced
by emulsion while Oscar Meyer is a larger consumer of suspension resin.
No further breakdown of the emissions was reported. However, the industry
as a whole appears to be a minor factor in the total emissions of the
industry.
Polymerization Technology - A desirable site selection criterion was based
upon the level of residual monomer remaining in the VDC copolymers. This
level is influenced by whether a stripping or monomer reduction step was
included in the polymerization process. Fabricators using polymers having
a high residual monomer level would be expected to have higher VDC emis-
sions. It appears, however, that monomer reduction is the industry prac-
tice and that it was impractical to apply this criterion.
43
-------
Table 19. ESTIMATED EMISSIONS OF VDC IN THE PROCESSING
OF PVDC POLYMERS^
Annual VDC emissions VDC emissions kg/100 kg
Process k£ JLb (lb/100 Ib) PVDC polymer
Coating cellophane 710 1,560 0.06
Coating plastics, paper 7,320 16,380 0.03
Glassine and paperboard
extrusion 200 430 0.001
Miscellaneous coating 5,450 12,000 0.04
Total 13,680 30,370
44
-------
3/
Table 20. MAJOR MANUFACTURERS OF EXTRUDED PVDC PRODUCTS"'
Manufacturer
Cryovac
(Division of W. R. Grace
and Company)
Oscar Meyer
Dow Chemical
American Can Company
Amtech, Inc.£/
Union Carbide Corporation
Plant location
Simpsonville, South Carolina
Cedar Rapids, Iowa
Camarilla, California
Iowa Park, Texas
Madison, Wisconsin
Chicago, Illinois
Davenport, Iowa
Philadelphia, Pennsylvania
Nashville, Tennessee
Los Angeles, California
Midland, Michigan
Cleveland, Ohio
Odenton, Maryland
Centerville, Iowa
Resin type
Emulsion and
suspension
Emulsion and
suspension
Suspension
Emulsion
Suspension
Suspension
a/ Manufacture monofilament.
45
-------
Geographical Distribution - The geographical location of the industrial
plants selected for sampling was important so that the results from sample
analysis can be used to establish whether exposure to VDC is a nationwide
problem. The implementation of this criterion was restricted, however, by
the following consideration: (a) efficiency of time and cost; and (b) the
selection of sites most likely to have detectable levels of VDC.
Criteria for the Selection of Fabricated Productjs
Products were obtained for analysis that were representative of the
five different types of polymerization industries.
Criteria for the Selection of Drinking Water Samples
Drinking water samples were requested from cities which were repre-
sentatives of the major types of raw water sources in use in the United
States. The selection was strongly influenced by whether: (a) VDC had been
reported in the drinking water; and (b) the city was included in the National
Organic Reconnaissance Survey (NORS). The first five cities selected for
the NORS study of volatile organics were:
Miami, Florida
Seattle, Washington
Ottumwa, Iowa
Philadelphia, Pennsylvania
Cincinnati, Ohio
VDC had been found in the drinking waters of Cincinnati, Miami and Philadelphia
but was not found at Ottumwa and Seattle. The second five cities to be sur-
veyed were:
Tucson, Arizona
New York, New York
Grand Forks, North Dakota
Terrebonne Parish, Louisiana
Lawrence, Massachusetts
No results have been reported for these cities.
POTENTIAL SAMPLING SITES
Industrial Plants
With the criteria presented in the previous section as a guide, six
industrial sites were selected and recommended for sampling. The plants
are listed in Table 21 along with their location and other relev nt pro-
duction information.
46
-------
Table 21. LOCATION AND PRODUCT PROFILE OF RECOMMENDED INDUSTRIAL SITES
Producer
Monomer
Dow Chemical, USA
PPG Industries
Polymer
Dewey and Almy
(Grace)
Production site
EPA
region
Plaquemine, Louisiana VI
Lake Charles, Louisiana VI
Owensboro, Kentucky
IV
Dow Chemical, USA Midland, Michigan
Tennessee Eastman Kingsport, Tennessee
IV
Fabricators
Dixico
Dallas, Texas
VI
Type of facility
Producer of monomer
Producer of monomer; user
of VDC to produce methyl
chloroform
Major producer of barrier-
coating latex
Only producer of thermo-
plastic and solvent-
soluble polymers.
Major producer of
miscellaneous use poly-
mers. Fabricater of ex-
trusion resins.
Producer of modacrylic
fibers
Major user of barrier-
coating latex
Products
'Chlorinated hydrocarbons, plastics
and resins, glycols
Chlorinated hydrocarbons including
chloroethane
PVC/PVDC resins, polyvinyl acetate
resins, styrene-butadiene resins,
polybutadiene-styrene latex
Chlorinated hydrocarbons, brominated
hydrocarbons, medicinals, pesticides,
plasticizers, plastics and resins,
general organic chemicals
General organic chemicals, plasticizers,
plastics and resins, synthetic fibers
(acrylic, modacrylic, polyester,
cellulosic)
Coated glassine
-------
Two monomer production plants, PPG at Lake Charles, Louisiana, and
Dow at Plaquemine, Louisiana, were selected, based largely on the fact that
over 80% of the VDC emissions were believed to come from the monomer segment
of the industry. Dow-Plaquemine was chosen rather than Dow-Freeport because
it is in a more populated area and does not produce chloroethane.
The plants selected for sampling which synthesize polymers were Dewey
and Almy (Division of Grace Chemical Company), Owensboro, Kentucky; Dow
Chemical, Midland, Michigan; and Tennessee Eastman, Kingsport, Tennessee.
Dewey and Almy was selected because it is the major producer of barrier-
coating latex while Dow Chemical at Midland, Michigan, is the only pro-
ducer of thermoplastic resin used for extrusion and solvent-soluble resin
used for coating cellophane. Dow is also a major producer of resin for
miscellaneous applications. The synthetic fiber industry is unique in that
it is a polymer producer and a fabricator. The polymer is produced and then
fabricated into a fiber at a single plant. Potential for emissions exists
in the polymerization process as well as in the fabrication operations.
Tennessee Eastman was chosen for sampling as the major producer of modacrylic
fibers.
The plant selected to be representative of the fabrication segment of
the industry could be a cellophane coater, user of barrier-coating latex
or a user of thermoplastic resins. The emission data in Table 14 suggested
that both the total emissions and the emissions per pound of PVDC polymer
were small for cellophane coating and extrusion. The greatest emissions
resulted from the use of barrier-coating latex. Of the two major users of
barriercoating latex, Dixico, in Dallas, Texas, was in a more favorable
geographical location for sampling during the winter months.
Fabricated Products
Fabricated products were requested from a wide spectrum of the fab-
ricators.
Drinking Water Supplies
Drinking water samples were collected at pre- and post-chlorination
points from the five cities listed in Table 22. Three were included in the
first five-city NORS survey and two were in the second survey.
The geographical locations and EPA regions of the industrial plants
and the water sample locations are shown in Figure 13.
48
-------
Table 22. PROPERTIES AND PREVIOUS RESULTS OF FINISHED WATERS SELECTED FOR SAMPLING
City
Cincinnati, Ohio
Lawrence, Massachusetts
Miami, Florida
New York, New York
Philadelphia, Pennsylvania
Type of
supply
Surface
y
Ground
y
Surface
Type of Inclusion in the Results of
raw water NORS study NORS study
Industrial waste Yes
b/ Yes
Natural waste Yes
b/ Yes
Municipal waste Yes
a/
£/
0.1 ng/ liter VDC
£./
< 0.1 pg/liter
VDC
a/ VDC detected but not quantified.
b_/ These data have not yet been obtained.
c/ No results have been announced.
-------
L/i
O
S cot--5"
JV t-f. «•• —
jr -! cr u>
•*J 3
cr: 7 !
0-1 |
H 3
• Industrial, Monomer Synthesis
D Industrial, Polymer Synthesis
• Industrial, Polymer Fabricator
A Water Sample
Figure 13. Geographical Location of Recommended Sampling Sites
-------
SECTION V
DISCUSSION OF RESULTS
The six selected industrial sites were sampled between January 4, 1977
and March 23, 1977. The actual sampling schedule is shown in Table 23; pre-
sampling site visits were conducted on the day prior to sampling. A summary
of the air and water samples collected at the six industrial sites is given
in Table 24 and a summary of the air sampling parameters follows in Table
25. The results from the six plants are discussed below followed by the
results from analysis of drinking waters and solid products.
DOW CHEMICAL, PLAQUEMINE, LOUISIANA
Field Sampling
A presampling survey was conducted on January 3, 1977 immediately
prior to the actual sampling. The plant lies mostly in Iberville Parish
and is located in a rural area on the west bank of the Mississippi River.
It has levees to the north, east, and southeast. A small Goodyear Tire and
Rubber Company plant which produces polyvinylchloride resins is located
immediately southeast of the plant. About 1-1/2 miles to the north-northwest
is a plant owned by Copolymer Rubber and Chemical Corporation, producing
ethylene-propylene copolymers.
The Dow facility manufactures chlorinated hydrocarbons, plastics, resins,
and glycols. Approximately 95 million pounds of VDC are produced annually
by Dow at Plaquemine and at Freeport, Texas.
The plant uses water from the Mississippi River. The processed waste-
water is returned to the river at a point approximately 1/4 mile downstream
of the intake. The winds in this area generally blow from the southeast,
south, and north during the month of January and from the southeast, east,
and north on an annual basis.
51
-------
Table 23. SAMPLING SCHEDULE
Date
1/4-1/5
1/25-1/26
2/8-2/9
2/22-2/23
3/8-3/9
3/22-3/23
Company
Dow Chemical
PPG
Dixico
Dewey and Almy (Grace)
Tennessee Eastman
Dow Chemical
Location
Plaquemine, Louisiana
Lake Charles, Louisiana
Dallas, Texas
Owensboro, Kentucky
Knoxville, Tennessee
Midland, Michigan
Product or operation
VDC monomer
VDC monomer
Fabricator
VDC polymer
VDC polymer
VDC polymer
-
and fabricator
-------
Table 24. FIELD SAMPLING SUMMARY
Site
Dow -Louisiana
PPG
Dixico
Dewey and Almy
Tennessee Eastman
Dow-Michigan
Table
Air samples Total samples
(stations x components) (number/ type)
13 x 2 26 Air
2 Water
12 x 2 28 Air
2 Water
7x2 14 Air
10 x 2 20 Air
2 Water
9x2 18 Air
4 Water
13 x 2 26 Air
4 Water
25. SUMMARY OF AIR SAMPLING PARAMETERS
Site
Dow -Louisiana
PPG
Dixico
Dewey and Almy
Tennessee Eastman
Dow -Michigan
Average sampling
volume (^) Average sampling Rate
(no. of samples) time (hr) (^/min)
1,370 (13) 24.0 0.95
1,280 (12) 23.6 0.90
1,330 (7) 23.5 0.94
842 (10) 18.2 0.77
1,330 (9) 25.2 0.88
1,155 (13) 23.2 0.83
53
-------
Field sampling was carried out during the 24-hr period from 0800
January 4, 1977 until 0800 January 5, 1977. Thirteen air samplers were lo-
cated at 12 sites upwind, downwind, and lateral with respect to the plant.
Soil samples were taken north and south of the plant. Water samples were
taken upstream and downstream from the point of discharge.
Air Sampling - Air sampling was conducted using 13 samplers at 12 sites.
Two charcoal tubes were used in series for each sampling train. The sites
were established upwind, laterally, and downwind of the plant with the
majority initially being located downwind. Samplers were positioned at
a height of 5 ft from ground level on public land such as right-of-ways
and levees. Sampling sites were located at approximately 0, 1/4, 1/2, and
2/3 miles from the plant perimeter. Locations of the 12 sites are shown
in Figure 14. Site descriptions and air sampling data are given in Table 26.
Water Sampling - Two grab water samples were obtained January 4, 1977. The
upstream sample was taken from the Mississippi River at a point 0.9 miles
upstream from the point of discharge. The downstream sample was taken 6 ft
from the river bank, approximately 100 ft downstream from the point of dis-
charge. The locations are indicated in Figure 14.
Meteorological Conditions - A summary of the weather conditions existing
during the air sampling period is given in Table 27.
Analysis of Air Samples
VDC Levels - The results of the analysis of the 13 top charcoal tubes are
listed in Table 28. VDC was detected in only seven of the samples, and then
only at low levels. The amount of VDC found was close to the analytical
limit of detection and made the analysis fairly difficult. For quality con-
trol, several unused charcoal tubes which had been shipped to the site were
also analyzed and showed no evidence of VDC. The general absence of VDC
raised concerns about the stability of the samples. For this reason, an
aliquot of Sample No. 2 was spiked at 5 ppm (wt/v) VDC and analyzed re-
peatedly for 3 days. The VDC peak height on the chromatograms remained con-
stant. The results demonstrated that any VDC in a CS2 extract would be sta-
ble for at least 3 days, and that the absence of VDC in the sample extracts
is not due to sample unstability.
The results of the analysis of the back-up tubes are also listed in
Table 26. Evidence of VDC was found only in Samples Nos. 2 and 13, collected
at North 0 miles, and West 0 miles, respectively. The amount of breakthrough
was about 33% for Sample No. 2 and 20% for Sample No. 13. Sample 2 had the
highest flow rate, 1.22 liters per minute for any site at which VDC was detected.
54
-------
•-WEST BATON ROUGE PARISH ,.<
/«S '^:
'. \ //<;•••.•••••••• H-•'•'/'N;^^ '..:'•• 'V.Morrison
- .--' y^-tf&« ^\- \ \-
'
V^i-ii-f4r;^f*^ *7
is'**» '\V47 .-'"V 1J> '"'•^>^^1''^i''-^7r??-T/^~4i
'X. .>^, X ii'W4>o.«jl«f)(3.8L
-v 48 ^<- ••--.:•.. sv. -AxV-^ •''•"• •'1>^1 *'i
\ / -^,- \v- '""•' ':-'^ ••• °W;- ."'-•:4^\n *rt
'""'10
0 1000 2000
i i i i I I
Scale - Feet
Figure 14. Sampling Locations and Concentrations of VDC in Air at
Dow Chemical, Plaquemine, Louisiana
55
-------
Table 26. AIR SAMPLING DATA FOR DOW CHEMICAL, U.S.A., PLAQUEMINE, I.A.
Area
North transect ,
North transect ,
South transect,
South transect ,
South transect ,
South Southwest
South Southwest
1/2 mile
0 mile
2/3 mile
1/2 mile
1/4 mile
transect, 2/3 mile
transect, 1/4 mile
Southwest transect, 1/4 mile
Southwest transect, 0 mile
West Southwest
West Southwest
East transect,
transect, 0 mile (1)
transect, 0 mile (2)
0 mile
0 mile
Sample
no .
1
2
3
4
5
6
7
8
9
10
11
12
13
Totaling
Exact location
On La. 988, 0.3 miles north of Eliza
100 ft east of gate 3
New Erwin St. and Warren St.
Homestead St. and Warren St.
0.2 miles south La. 988 and La. 1148 on Warren St.
East Frontage Road for La. 1 at McDaniel St.
0.2 miles south of La. 1148 on Brownie St.
East Frontage Road for La. 1, 0.2 miles south of
La. 1148
0.2 miles west of Brownie St. on La. 1148
La. 1148 and the east Frontage Road of La. 1
La. 1148 and the east Frontage Road of La. 1
East of La. 988, 1 mile north of La. 1148
for La. I
sampling (hr)
24.
24.
23.
23.
23.
24.
23.
24.
23.
23.
23.
24.
24.
6
6
8
8
8
0
5
1
5
6
6
2
6
Sampling
rate
1
1
1
1
0
1
0
0
0
1
0
1
0
(f/min)
.00
.22
.09
.10
.47
.06
.58
.95
.67
.39
.64
.39
.89
Total
volume (M^)
1.
1
1
1
0
I
0
1.
0
1
0
2,
1.
.471
.794
.561
.567
.663
.525
.900
.376
.946
.961
.843
.012
.237
Sample
height (ftl
5.
5,
5
5,
5
5
5
5
5
5
5
5
5
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
Ln
-------
.Table 27. WEATHER CONDITIONS DURING SAMPLING AT DOW CHEMICAL, USA, PLAQUEMINE, LA
Time
January 4
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
January 5
0100
0200
0300
0400
0500
0600
0700
0800
0900
Wind
Speed (Kts)
5
6
7
10
4
5
4
5
10
9
9
7
6
5
00
6
5
6
00
5
00
3
4
00
6
8
10
Direction
SE
E
ESE
SSE
ESE
SW
NW
ESE
SE
SE
S
SSE
SSE
SE
SSW
S
SSE
NNE
NE
NNE
N
N
NNW
Precipitation
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
Very light raini/
Very light rain
Very light rain
Very light rain
Very light rain
Very light rain
Very light rain
Very light rain
Very light rain
Very light rain
Very light rain
a/ Reported as trace by the National Weather Service.
57
-------
It is possible that some VDC would also breakthrough the second charcoal
tube. Assuming that the same proportionality holds, the amount of VDC lost,
1.2 and 0.6 H-g at Sites 2 and 13, respectively, would be barely detectable
and would result in less than a 167<, reduction in the total VDC reported.
The total quantity (^g) found at each site is considered to be the sum of
the top and bottom tubes, neglecting any possible loss due to breakthrough.
All top samples were brominated and analyzed for 1,2-dibromo-l,1-
dichloroethane. The results are used only qualitatively due to the unavail-
ability of a pure standard. Table 28 shows the results arranged to indicate
confirmation of VDC by both the disappearance of a chromatographic peak on
the OPN column following bromihation and the appearance of a new peak on the
OV-101 column also following bromination.
As expected, all the samples believed to contain VDC vrere confirmed.
Most of the samples and blanks which had been shown not to contain VDC also
did not contain 1,2-dibromo-l,1-dichloroethane. However, Samples Nos. 10,
11, and 12 were positive after bromination although analysis on the primary
column did not indicate the presence of VDC. These are considered as false
positives and cannot be explained.
3
The levels of VDC found in the air, as M-g/m given in Table 28 are
shown in Figure 14. Figure 15 shows a wind rose pattern depicting the wind
behavior during the time when the samples were being collected. At the begin-
ning of the sampling period the wind was directly from the north, thus many
of the sites were established to the south of the plant. Midway through the
sampling, the wind shifted and was blowing from the southeast. The levels
found to the south decreased as expected with increasing distance. The
lower concentration at the southwest and west-southwest sites correlated
well with the wind patterns. The highest concentration of 9.6 p>g/m was
found at Site No. 13, located on the property line, approximately 1-1/2 miles
west-northwest of the VDC production area. This site was the predominately
downwind site. The predominately upwind site, No. 12, had no detectable VDC.
An attempt was made to confirm the presence of VDC by GC/MS. Sample
No. 13 was chosen for GC/MS analysis as it had the highest level of VDC.
Mass fragments from m/e 45 to 200 were collected and the presence of VDC
was tested by an extracted ion current plot [EICP] of ions 61, 96, 98, 63
as shown in Figure 16. The results were too inconclusive to be considered
as a confirmation.
Identification of Other Compounds - The analysis of the air samples by
GC-FID, had shown qualitatively the presence of six to eight additional
compounds. One sample, No. 13, was analyzed by GC/MS in order to confirm
the presence of VDC and, if possible, establish the identity of the other
peaks.
58
-------
Table 28. VDC IN AIR SAMPLES FROM DOW CHEMICAL, PLAQUEMINE, LOUISIANA
Sampling
station
1
2
3
4
5
6
7
8
9
10
11
12
13
Sampling-/
time
0755-0828
0747-0824
0835-0825
0840-0829
0855-0832
0820-0820
0910-0837
0809-0816
0838-0800
0823-0801
0823-0801
0805-0815
0730-0808
Confirmation by broraination
Volume
(m3)
1.47
1.79
1.56
1.57
0.66
1.53
0.90
1.38
1.96
0.84
0.84
2.01
1.24
Type of sample
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
VDC
(Hgjk/
ND*'
ND
5
2
6
ND
6
ND
5
ND
1
ND
2
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
10
2
J«ss of VDC-/
NA6-/
NA
Yes
Yes
Yes
NA
Yes
NA
Yes
NA
Yes
NA
Yes
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Yes
Yes
Appearance of
bromlnated VDCC/
NA
NA
Yes
Yes
Yes
NA
Yes
NA
Yes
NA
Yes
NA
Yes
NA
NA
NA
NA
NA
Yes
NA
Yes
NA
Yes
NA
Yes
f/
Presence of
ch lor oe thane
.
-
No
No
No
-
No
-
No
-
No
-
No
-
-
-
-
-
-
-
-
-
-
-
No
No
VDC
(tig/m3)
ND
3.9
3.8
3.8
7.6
0.6
2.2
ND
ND
ND
ND
ND
9.7
a/ 1/4/77 to 1/5/77.
b/ Based on analysis on Durapak OPN column.
£/ Based on analysis on OV-101 column.
d/ ND = Not detected.
e/ NA = Not applicable.
j[/ Sample lost by evaporation.
-------
NNW
NW
WNW
NNE
ENE
SSE
SE
Wind Speed,
Knots per Hour
Figure 15. Wind Patterns During Sampling at Dow Chemical, Plaquemine, Louisiana
60
-------
Vinylidene Chloride
96
98
63
rp M r pTTiprrrrmr|T
7650
1 1 1 1 1 1 1| n r
7700
7750
7800
7850
Figure 16. Extracted ion current plots - Dow air sample No. 13
-------
The GC-FID analysis had shown two small peaks eluted before the sol-
vent. An attempt was made to analyze these peaks on the GG/MS by not vent-
ing the solvent until it had appeared that they had entered the spectrom-
eter. We were not able, however, to identify any of the peaks eluting before
the solvent was vented. After 2 min the vent was closed for the remainder
of the chromatographic run. This venting, however, did not completely prevent
CS2 from entering the spectrometer and contributing strongly to the total
ion current. Because of the residual CS2, it was not possible to obtain
meaningful mass spectral data without using EICPs. From 3 to 4 m/e values,
characteristic of the fragmentation of a specific compound, were plotted
for the chromatographic run. The appearance of maxima for these ions coinci-
dentally, at the retention time of a reference compound, or at the same
spectrum number, in the correct intensity ratio, was considered as evidence
for the presence of the compound. By this approach, seven compounds in addi-
tion to VDG were identified in Dow Chemical air Sample No. 13. The eight
compounds are noted on the sum plot for Sample No. 13 given in Figure 17.
Table 29 lists the compounds identified, the ions monitored with their ex-
pected and observed intensity ratios, and whether the compound is a company
product. The confirmation of benzene was complicated by the presence of an
intense 76 ion from G&2 over the entire chromatogram.
The following compounds were shown by the EICP's to be absent; methyl
chloride, ethyl chloride, trans-dichloroethylene, methylene chloride, cis-
dichloroethylene, methyl Chloroform, ethylene dichloride, 1,1,2-trichloro-
ethane, and 1,2-dichloropropane.
Analysis of Water Samples
VDC Levels - The downstream and upstream water samples, W2 and W3, were
first analyzed by flame ionization-gas chromatography. A sample size of
10 ml was used for the VGA procedure in conjunction with the Carbowax 1500
primary column. A small amount of VDG, approximately 2 ng, was found in the
downstream sample, W2. In addition to the VDC, at least 18 other substances
were detected. No VDC was detected in the upstream sample, W3, although at
least four other components were present.
Both samples were then analyzed by GC/MS using identical gas chroma-
tographic and analysis conditions. Examination of the mass spectra at the
retention time of VDC failed to indicate the presence of VDC in either
Sample W2 or W3.
An EICP current of the four largest fragment ions, 61, 96, 98, and 63,
was obtained. It as not possible to discern the presence or absence of VDC
in either sample due to the background noise.
62
-------
30000-
Lo
eoooo-
10000-
7700. 7750
SPEC* 17-480 - 17BM JH DOU.HIB.U ,0.81 3»OPIt.601116 2-17-7?
''""I
7800 7850
STEP SPECS'! INT> 1000
7900
Figure 17. Sum plot of Dow air sample No. 13
-------
Table 29. SUMtoRY OF GC-MS ANALYSIS OF DOW AIR SAMPLE NO. 13
Compound
Vinylidene chloride
Chloroform
Tr ich loroe thy lene
Tetrachloroethylene
Benzene
Toluene
1 , 2-Dlchloropropane
ra/e values
61, 96, 98, 63
119 117, 121, 82
83, 85, 87, 47
130, 132, 95, 97
166, 164, 129, 131
78, 76, 53, 79
91, 92, 65
63, 62, 65, 76
Intensity ratios
1.00:0.53:0.36:0.28
1.00:0.98:0.25-0.13
1.00:0.55:0.07:0.14
1.00:0.95:0.85:0.59
1.00:0.78:0.75:0.65
1.00:0.06
1.00:0.51:0.10
1.00:0.44:0.30
Spectrum
No.
17750
17783
17796
17796
17814
17813
17851
17840
Observed
intensity ratios
1.00:-:-:-
1.00:0.7-0 3--
1.00:0.70:0.13:0.6
1.0:1.0:1.0:1.0
1.0:0.5:0.7:0.7
a/
1.00:0.50:0.35
1.00:0.86:0.21
Conclusion
Present (?)
Present
Present
Present
Present (?)
Present
Present
Product
Yes
Yes
No
No
Yes
No
No
Yes
a/ Ratio cannot be calculated as ro/e 78 peak saturated.
-------
The samples were then analyzed by GC/MS operated in the multiple ion
scan (MIS) mode. Figure 18 shows the MIS plot obtained for 3 ng of VDC at
m/e values of 61, 96, 93, 98, and 100. VDC eluted at 5.7 rain (about 3.3
min in the Figure) with the five ions in an intensity ratio of 1.00: 0.47:
0.31: 0.24: 0.02, respectively. Figure 18 also shows the MIS plot of the
same five ions for 10 ml of the downstream sample, W2. The relative intensity
of the five ions monitored in Sample W2 was found to be 1.00: 0.47: 0.33:
0.24: and 0.03 which agrees well with the standard. This fact along with
the matching retention times confirms the presence of VDC in the effluent.
The relative intensity of the standard to the sample indicates that W2 con-
tained about 4 ng VDC. The concentrations of VDC in the water from the GC/
MS/MIS and the GC-FID analyses are given in Table 30 and agree quite well.
The upstream sample was not analyzed by GC/MS/MIS.
Identification of Other Compounds - The GC-FID analysis for VDC had shown
the presence of about 18 additional compounds in downstream water Sample
W2. The sum plot from the GC/MS analysis showed about 13 distinct peaks.
Initially EICPs were used to establish the identities of as many peaks as
possible. The NPDSS permit was used as a guide in selecting the compounds
for screening.
Using this approach, 13 compounds in addition to VDC were identified.
Table 31 lists the compounds which were established as being present along
with the ions plotted plus their expected and actual intensity ratios. Also
noted is whether the compound is listed as a plant product or is included
in the plant's NPDES permit. The identified compounds are noted on the sum
plot in Figure 19.
All but one peak in the sum plot were identified by EICPs plots. Its
spectrum was obtained with background subtracted and searched against the
Cyphernetics file using a Bieman search. The following matches and similarity
indices were found:
Compound Similarity index
Methylcyclohexane 0.397
Cycloheptane 0.318
3-Methyl-hexanol 0.307
1,1-Dimethylcyclopentane 0.305
The EICPs also established that the following compounds were not in
the sample: vinyl chloride, methyl chloride, ethyl chloride, and 1,1-
dichloro ethane.
65
-------
M/E 61
6i
I
i I
2
MINUTES
(b)
Figure 18. Multiple Ion Scan of 3 ng VDC and 10 ml of Dow Water Sample W-2
Scale Expanded x 10
a. 3 ng VDC
b. Dow Water Sample W-2
66
-------
Table 30. VDC CONCENTRATIONS IN WATER FROM DOW CHEMICAL, PIAQUEMINE, LA.
Sample Location Volume GC/FID GC/MS ug/L
W2 100 ft downstream 10 ml 2 ng 4 ng 0.2
of outfall
W3 0.9 miles upstream 10 ml N.D. N.D. N.D.
of outfall
-------
Table 31. SUMMARY OF GC-MS ANALYSIS OF DOW WATER SAMPLE W-2
00
Compound
Methylene chloride
Vlnylldene chloride
trans -Dich loroethy lene
els -Dlch loroethy lene
Chloroform
Ethylene dlchloride
Methyl chloroform
Carbon tetrachloride
1,2-Dichloropropane
Benzene
Trichloroe thy lene
1,1, 2-Trichloroethane
Tetrachloroe thy lene
Toluene
Methyl cyclohexane
m/e values
84, 86, 51, 88
61, 96, 98, 63
61, 96, 98, 63
61, 96, 98, 63
83, 85, 87, 47
62, 64, 98, 100
97, 99, 117, 119
119, 117, 121, 82
63, 63, 65, 76
78, 77, 52
130, 132, 95, 97
83, 97, 85, 99
166, 164, 129, 131
91, 92, 65
b/
Intensity ratios
1.00:0.60:0.33:0.10
1.00:0.53:0.36:0.28
1.00:0.84-0.48-0.37
1.00:0.99:0.84:0.48
1.00-0. 55-0. 07-0. 14
1.00:0.31:0.08:0.06
1.00:0.62:0.19:0.19
1.00:0.98:0.25:0.13
1.00:0.44:0.30:0.11
1.00:0.17:0.17
1.00:0.95:0.85:0.59
1.00:0.98:0.64:0.62
1.00:0.78:0.75:0.65
1.00:0.51:0.10
b/
Spectrum
No.
6354
6371
6390
6396
6401
6405
6415
6417
6436
6438
6440
6451
6499
6510
6462
Observed
intensity ratios
1.00:0.54:0.35:0.13
2/
1.00:0.67:0.67:-
1.00:0.57:-:-
1.00:0.56:0.28:0.09
1.00:0.53:0.25:0.16
1.00:0.78:0.21:0.21
1.00:1.00:-:-
Peaks saturated
Peaks saturated
1.00:1.20:1.20:1.00
1.00:0.88:0.50:0.63
1.00:0.83:0.77:0.77
1.00:0.48:0.20
Conclusion
Present
Present*!/
Present (?)
Present (?)
Present
Present
Present
Present (?)
Present
Present
Present (?)
Present
Present
Present
Present
Product
Yes
Yes
No
No
No
Yes
No
Yes
Yes
No
No
Yes
Yes
No
No
NPDES
permit
Yes
Yes
No
No
Yes
Yes
- No
Yes
Yes
No
No
Yes
Yes
No
No
a/ Presence of VDC confirmed by MIS.
b/ Determined by a Btemann search of the Cyphernettcs file.
-------
V
c
o
8-
Q.
O
O
Q
C-4
30000-n
vO
20000-
XQQOQ-
SPEC* &31Q -
6100 6H50 6500 6550 6600
JO DOU Lfl. EFF. 1OHL CU1500 J.G. 1-13-7? STEP SPEC^l IttI
Figure 19. Sum plot of Dow water sample W-2
1 0 u 0
-------
The upstream Sample W3 was also analyzed by GC/MS using identical
procedures. The only peak which was observed on the sum plot was tentatively
identified as hexane.
PPG INDUSTRIES, LAKE CHARLES, LOUISIANA
Field Sampling
A presampling survey was conducted on January 24, 1977 immediately
prior to the actual sampling. The plant lies in Calcasieu Parish and is lo-
cated just west of Lake Charles in the area bordered by 1-10 and 1-210. Olin
Chemical has a plant located immediately to the northeast which produces
ammonia, carbon monoxide, carbon dioxide, hydrazines, hydrogen, organic iso-
cyanates, nitric acid, phosgene, polyether polyols, urea, sodium nitrite,
and toluene 2,4-diamine. Certain-Teed Products has a plant located immediately
to the northwest which produces polyvinylchloride resins.
The PPG facility produces chlorine, ethyl chloride, ethylene dichlor-
ide, hydrochloric acid, hydrogen, perchloroethylene, sodium hydroxide, sodium
silicates, 1,1,1-trichloroethane, vinyl chloride monomer, and vinylidene
chloride monomer. Approximately 175 million pounds of VDC are produced annu-
ally by PPG in Lake Charles.
The plant uses water from the Calcasieu Ship Channel which is then dis-
charged, by way of a canal, into the Bayou d1 Inde. The wind blows predomin-
antly from the north, southeast, south, and north-northwest during January
and predominantly from the south, southeast, and north on an annual basis
at Lake Charles.
Field sampling was carried out during the 29-hr period from 0700 January
25, 1977 until 1200 January 26, 1977.
Air Sampling - Air sampling was conducted using 12 samplers, each with two
charcoal tubes in series, at nine different sites. The sites were established
upwind, laterally, and downwind of the plant. Samplers were positioned at
a height of 4 ft from ground level on public land such as right-of-ways.
Samplers were located at approximately 0, 1/4, 1/2, 2/3, and 2 miles from
the plant perimeter. Locations of the nine sites are shown in Figure 20.
Site descriptions and air sampling data are given in Table 32. Duplicate
samples were collected at the paired Sites 3 and 4, 6 and 6, and 9 and 10.
At Sites 4 and 6, charcoal tubes were used which had been preloaded with
approximately 285 ^g VDC.
Water Sampling - Two grab water samples were obtained on January 25, 1977
and on January 27, 1977. The downstream sample was taken from the PPG canal
at a point just before it enters the Bayou d1 Inde. The upstream sample was
taken from the west fork of Bayou d1 Inde where it crosses under 1-10. The
locations are indicated in Figure 20.
70
-------
' -'
•..-.•:>-•"•! jMaplewood
'•"M^-—
! •:,
«•••. ...i* _ E^ .1.9 _ '
•o f A • • VV-.
. \ '» U'V.vr .« /» J
0 1000 2000 )''•/
-------
Table 32. AIR SAMPLING DATA FOR PPG INDUSTRIES, IAKE CHARLES, LA.
Area
Northeast transect, 0.5 miles
North transect, 0.5 miles
North transect, 0 miles
North transect, 0 miles
South transect, 0 miles
South transect, 0 miles
Southwest transect, 0 miles
East transect, 2 rolles
Southeast transect, 0.25 miles
Southeast transect, 0.25 miles
West transect, 0.5 miles
Northwest transect, 0.67 miles
Sample
no.
1
2
3
4
5
6
7
8
9
10
11
12
Total
Exact location
US-90 and Columbia Southern Road
0.5 miles north of gate
150 ft north of gate
150 ft north of gate
West end of 1-210 bridge
West end of 1-210 bridge
0.2 miles north of 1-210 bridge
North end of old US-90 bridge
Coon Island
Coon Island
Junction of 1-210 and Maplewood Road
0.5 miles east of 1-210 on Maplewood Road
sampling (hr)
24
23
23
23
23
23
23
24
23
23
23
23
.00
.78
.85
.82
.22
.20
.12
.13
.92
.92
.03
.50
Sampl i ng
rate ( /mln)
0
0
0
1
0
0
1
0
0
0
0
0
.936
.887
.960
.012
.930
.983
.043
.977
.920
.250
.965
.932
Tola 1
volume (M3)
1.
1.
1.
1.
1.
1.
1.
1.
1.
0.
1.
1.
346
266
374
447
296
368
447
413
320
359
333
314
Sample
helsht (ft)
4
4
4
4
4
4
4
4
4
4
4
4
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
-------
Meteorological Conditions - A summary of the weather conditions existing
during the air sampling period is given in Table 33.
Analysis of Air Samples
VDC Levels - The results of the analysis of the charcoal tubes are listed
in Table 34. VDC was detected at 5 of the 10 sampling stations. VDC was
present in all five top charcoal tubes and in four of five bottom char-
coal tubes. As was the case with the Dow samples from Plaquemine, Louisiana,
some breakthrough at the 1 liter/min sampling rate did occur, ranging from
24 to 48% of total VDC found. The higher percentage of breakthrough was
associated with the higher total quantities of VDC trapped. The effect of
the breakthrough is to slightly bias the results to the low side. If a lower
flow rate had been used to reduce the breakthrough, then the total VDC col-
lected would have been less, thus lowering the quality of all the data.
All nine samples believed to contain VDC were confirmed by bromination.
Table 34 indicates the results both in terms of whether the VDC peak was
eliminated by bromination and whether a peak appeared for brorainated VDC.
In all cases, VDC was confirmed. Moreover, the bromination confirmed that
no chloroethane had been collected and therefore was not co-eluting with
VDC. Thus, no correction had to be applied to the results to account for
chloroethane.
At Sites 4 and 6, duplicate air samplers were operated concurrently
using charcoal tubes that had been spiked at Midwest Research Institute (MRI)
with VDC. The tubes were spiked from an inhalation chamber that at the time
contained numerous test animals. The tubes were expected to have been loaded
with ~ 300 ^g VDC; however, this is only approximate in that we had no know-
ledge of the homogeniety of the chamber. (A third tube was stored at MRI
as a control.) The purpose of the experiment was to show that nothing in the
air collected around an industrial chemical plant, e.g., Cl~, NO , or PAN's,
would cause a decomposition of the trapped VDC. Upon analysis, 290, 672,
and 605 \i>g of VDC were found for samples from Site 4, Site 6, and MRI,
respectively.
The objective of the experiment, to demonstrate that VDC trapped by
the charcoal was not being decomposed by some common atmospheric contaminant,
was met. The discrepancies in the micrograms VDC found were probably due
to uncontrollable variables in the spiking.
73
-------
Table 33. WEATHER CONDITIONS DURING SAMPLING AT PPG
INDUSTRIES, LAKE CHARLES, LA.
Wind
Time
January 25
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
January 26
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
Speed (Kts)
5
4
5
4
7
6
7
6
5
7
8
9
7
5
5
5
0
4
5
3
4
3
3
3
3
3
4
6
8
7
Direction
NW
NW
NNW
NE
N
N
N
NNW
W
NW
SSW
SSW
SW
SW
SW
S
N
WSW
WSW
SW
S
S
SSW
S
S
SSE
SSE
S
S
S
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
None
None
None
None
None
None
74
-------
Table 34. VDC CONCENTRATIONS IN AIR SAMPLES FROM PPG INDUSTRIES, LAKE CHARLES, LOUISIANA
Sampling
station
1
2
3
4
5
6
7
8
9
10
11
12
Field
Blank
Field
Spike
Sampling-''
time
0800-0800
0820-0807
-
0812-0803
0813-0803
0907-0820
0908-0820
0917-0824
0745-0753
1120-1115
1120-1115
0925-0827
0845-0815
-
-
Volume
(m3)
1.36
1.27
1.37
1.45
1.30
1.37
1.48
1.41
1.32
0.36
1.33
1.31
-
-
Type of sample
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoa I
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
VDC
(ug)-/
31
23
22
7
37
34
290
No sam-
ple
12.6
4
672
No sam-
pie
7.2
OTF-'
ND
NA
ND
NA
ND
NA
ND
-
ND
-
ND
ND
605
No sam-
pie
Confirmation
Loss of VDCb/
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No sample
Yes
Yes
Yes
No sample
Yes
NA^'
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Yes
No sample
by bromination
Appearance of
brominated VDCC/
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No sample
Yes
Yes
Yes
No sample
Yes
NA
ND
NA
ND
NA
ND
NA
ND
NA
ND
NA'
ND
NA
Yes
No sample
Presence of
ch loroe thane
No
No
No
No
No
No
No
-
No
No
No
-
No
-
No
-
No
-
No
-
No
-
No
'
No
-
-
-
VDC
(ug/m3)
39.7
22.8
51.8
-
f/
12.8
f/
4.9
ND
ND
ND
ND
ND
ND
f/
a/ 1/25/77-1/26/77.
b/ Based on
£/ Based on
d/ ND = not
e/ NA = not
analysis using
analysis using
detected.
applicable.
Dura pa k
OPN column.
OV-101 column.
f_/ Spiked sample.
-------
The levels of VDC found in the air, as p-g/nr' were given in Table 34
and are shown in Figure 20. Figure 21 shows a wind rose pattern depicting
the wind behavior at Lake Charles Municipal Airport during the time when
samples were being collected. The wind was from the north during the first
one-third of the sampling period, and then blew from the south for the remain-
ing time. The concentrations found correlate quite nicely with the wind
patterns. The highest level, 51.8 Hg/m3, was found at the station on the
north plant boundary at 0 mile. The sites farther north have lower concen-
trations that correlate with the wind direction. VDC, at 12.8 and 4.9 jig/m3,
was found at the two sites directly south of the plant. No VDC was found
at any of the east or west sites which were never downwind of the plant.
The presence of VDC in two air samples, Nos. 3 and 5, was confirmed
by GC/MS. Mass spectra were collected from m/e 45 to 200 and the presence
of VDC was tested by EICPs from ions 61, 96, 98, and 63. As seen in Figure
22, VDC was confirmed in air Samples Nos. 3 and 5.
Identification of Other Compounds - The analysis of the air samples by GC-
FID had indicated the presence of seven to nine compounds in addition to
VDC. These compounds were particularly noticeable in the downwind samples
such as Nos. 1, 2, 3, 5, and 7. The north and south fence line samples,
Nos. 3 and 5, respectively, were analyzed by GC/MS to confirm VDC and provide
identifications of the additional peaks. Specific compounds were identified
using EICPs. Tables 35 and 36 list the compounds found in air Samples Nos.
3 and 5, respectively, as well as the ions monitored, their expected and
observed intensities and whether the compound is a company product. The
compounds found at the two sites were the same except for the probable
presence of cis- and trans-dichloroethylene and the presence of toluene and
1,1,2-trichloroethane at the No. 5 site, south boundary. Sum plots for air
Samples Nos. 3 and 5 are given in Figures 23 and 24, respectively, showing
the compounds identified.
The following compounds were shown by EICPs not be present; methyl chlor-
ide, vinyl chloride, ethyl chloride, trans-dichloroethylene. methyl chloroform,
cis-dichloroethylene, ethylene dichloride, toluene, 1,l,2trichloroethane
and 1,2-dichloropropane.
Analysis of Water Samples
VDC Levels - The downstream water sample, Wl, and the upstream sample, W2,
were first analyzed by GC-FID. Sample Wl was found to contain 550 ppb VDC
while W2 contained no VDC. The results are given in Table 37. The sample
was then analyzed by GC/MS using the same conditions applied to the Dow
Chemical, Plaquemine, Louisiana, water sample. Figure 25 shows a sum plot
for the analysis of 1 ml of the PPG effluent, with the VDC peak identified.
The presence of VDC was confirmed by the comparison to a known spectrum as
shown in Figure 26.
76
-------
NNW
NW
WNW
Wind Speed,
Knots per Hour
NNE
NE
ENE
SSE
ESE
SE
Figure 21. Wind patterns during sampling at PPG Industries,
Lake Charles, Louisiana
77
-------
oooo-
Vinylidene Chloride
I
^
lliHnnpSTi[iii'rmT|inT|)m'piiimnfiiii|iiM[iiii|in.|i i
6750 6300
SPEC* 16737 - taa^S .Jt1 PPG,ftIR.naRTH.a,*3.0Pft.tS(JIHS- 2-17-77
I I If I'I'll I I | Illlll I
6850 6900
STEP SPECS'! INT. 1000
6950
7000
7150
Figure 22.. Extracted ion current plots for VDC - (A) PPG
air sample No. 3, (B) PPG air sample No. 5
78
-------
Table 35. SUMMARY OF GC-MS ANALYSIS OF PPG AIR SAMPLE NO. 3
Compound
Vlnylidene chloride
Carbon tetrachloride
Chloroform
Trtchloroethylene
Methyl chloroform
Tetrachloroethylene
Benzene
m/e values
61, 96, 98, 63
119, 117, 121, 82
83, 85, 87, 47
130, 132, 95, 97
97, 99, 117, 119
166, 164, 129, 131
78, 79
Intensity ratios
1.00:0.53:0.36:0.28
1.00:0.98:0.25:0.13
1.00:0.55:0.07:0.14
1.00:0.95:0.85:0.59
1.00:0.62:0.19:0.19
1.00:0.78:0.75:0.65
1.00:0.06
Spectrum
no.
16804
16842
16853
16858
16862
16873
16872
Observed
intensity ratios
1.00:0.29-0.27:0. 27
1.0: 1.0--:-
1.00:0.55:0.07:0.62
1.0:0.7:1.3:-
1.00:0.56:0.09:0.95
1.0: 1.0:1.0:1.0
1.0:0.12
Conclusion
Present
Present
Present
Present
Present
Present
Present
Product
Yes
No
No
Yes
Yes
Yes
No
-------
Table 36. SUMMARY OF GC-MS ANALYSIS OF PPG AIR SAMPLE NO. 5
Compound
Vlnylidene chloride
trans -Dlchloroethy lene
Carbon tetrachlorlde
cls-Dtchloroethylene
Chloroform
Trichloroethylene
Methyl chloroform
Tetrachloroe thy lene
Benzene
Toluene
1,1, 2-Trichloroethane
m/e values
61, 96, 98, 63
61, 96, 98, 63
119, 117, 121, 82
61, 96, 98, 63
83, 85, 87, 47
130, 132, 95, 97
97, 99, 117, 119
166, 164, 129, 131
78, 79
91, 92, 65
83, 97, 85 99
Intensity ratios
1.00:0.53:0.36:0.28
1.00:0.84:0.98:0.37
1.00:0.98:0.25:0.13
1.00:0.99:0.84:0.48
1.00:0.55:0.07:0.14
1.00:0.95:0.85:0.59
1.00:0.62:0.19:0.19
1.00:0.78:0.75:0.65
1.00:0.06
1.00:0.51:0.10
1.00:0.98:0.64:0.62
Spectrum
no.
17020
17035
17057
17065
17069
17072
17076
17087
17086
17123
17130
Observed
intensity ratios
1.00:0.22:0.22:0.33
1.0:0.2:0.2:0.2
1.0:0.7:0.3:0.3
1.00:0.2:0.2:0.2
1.00:0.56:0.08:0.56
1.0:0.9:1.0:0.8
1.00:0.75:0.16: -
1.0:1.0:1.0:1.0
1.0:0.1
1.00:0.58:0.42
1.0:1.0:0.6:0.8
Conclusion
Present
Present (?)
Present •
Present (?)
Present
Present
Present
Present
Present
Present
Present
Product
Yes
No
No
No
No
Yes
Yes
Y.es
No
No
No
00
o.
-------
•8
30000-.
oo
eoooo-
l o o o o
SPEC» 16737 - 16918 JH PP8 . RIR
6800
, 0 , 8 3.0PN , 80IH9 Z-\T-7?
6050 6900
STEP SPECS'! INT' tOOO •••
I "
6950
Figure 23. Sum plot of PPG air sample No. 3
-------
30000-,
OO
eoooo-
10000-
roog
SPECt 16950 - 17159 Jtt PPS ,flll. SOUTH,O,t5 , OP« ,
7050
2-17-77
7100
STEP SPECO-1 INT- 1000
7150
Figure 24. Sum plot of PPG air sample No. 5
-------
Table 37. VDC CONCENTRATIONS IN WATER FROM PPG INDUSTRIES,
LAKE CHARLES, LOUISIANA
Sample
No. Location Volume analyzed VDC found
Wl PPG Canal 1 ml 0.55 jug 550
W2 West Fork, 10 ml None ND
Bayou d1 Inde
83
-------
0)
c
0)
0)
TJ
10000-
do
0)
o
u
I,,
5«t 00
SPECtt 15HOO
..,,,.. .,..,.,..,,,....,.,.. ^.-irjTTTTJITTTJTITIIirTTpTT. ,...,,. ... J,. ..,....,...,,.... | ....
5*50 B500
- 15580 JH PPG 1ML EFF CU1SOO GOIMG 2-3-77
STEP
5550
INT= 100C
Figure 25. Sim plot of PPG water sample W-l
-------
I 0 0 -q
2
904
3
eftJ
1
?0 i
60-E
oo | i
m ;
£Q-jj
*
H'°1
3o4
so-f
1 04
J"
0 —
H
— 1 — 1 — T"
0
1 1
[ 1 1 1 1 [ 1 1 I 1 [ 1 1 1 1
1
50
6
1
1 1
0
i | i i i i | i i i i | i . . .
70
100-
eo-
70-
40-1
to-
to-
30-i
eon
lo-i
!
80
In i
•
• • • • • i
•
9
0
1
1
II
1
l*t CO!
00 1
ts9 3ta
1 0
30
jl
"10
r
1
' !
_!
!
;
.
ISO
SFECt 15f60 JS PPG IttL E'FF CU1500 SOIN6
STEP MflSS=lr I.--B/S
Figure 26. Mass spectrum of VDC in PPG water sample W-l
-------
Identification of Other Compounds - Nine additional compounds were iden-
tified and have also been noted on the sum plot in Figure 25. As before,
the presence (or absence) of a compound was determined using EICPs. Table
38 lists the compounds which were established as being present along with
the ions monitored, the expected and actual intensity ratios and whether
the compound is listed as a plant product. No estimate of concentration
was made for any of the compounds. The following compounds were established
by EICPs as being absent in the sample; vinyl chloride, methyl chloride,
methylene chloride, ethyl chloride, 1,1-dichloroethane, carbon tetrachloride,
1,2-dichloropropane, benzene and toluene.
The upstream water sample, W2, was analyzed by GC/MS and found to
be virtually free of any volatile components.
DIXICO, DALLAS, TEXAS
Field Sampling
A presampling survey was conducted on February 7, 1977, 1 day prior
to sampling. The plant is located at 1300 South Polk Street, Dallas, Texas,
in a residential area in the Oak Cliff subdivision of Dallas. There are no
other manufacturing plants located in the immediate vicinity. No stacks were
observed at the plant. Dixico uses barrier-coating latex and produces coated
glassine.
The winds in this area generally blow from the south, south-southeast,
and southeast in February and on an annual basis.
Field sampling was carried out during the 24-hr period from 0900 Febru-
ary 8, 1977, to 0900 February 9, 1977. Seven samplers were deployed at five
sites located upwind, downwind, and laterally with respect to the plant. No
water samples were collected.
Air Sampling - Air sampling was conducted using seven samplers located at
five sites. Two charcoal tubes were used in series for each sampling train.
The sites were established upwind of, lateral to, and downwind of the plant.
Samplers were positioned at a height of 3 ft above ground level on public
land such as right-of-ways. Samplers were located 0 and 0.1 miles from the
plant perimeter. Locations of the five sites are shown in Figure 27. Site
descriptions and air sampling data are given in Table 39. Duplicate samples
were taken at the paired Sites 2 and 3, and 5 and 6.
Water Sampling - According to the Permits Section, U.S. EPA, Region VI, the
Dixico plant has never requested a NPDES permit. Visual inspection of the
plant likewise did not reveal any outfalls. Consequently, no water samples
were collected from the Dixico plant.
86
-------
Table 38. SUMMARY OF GC-MS ANALYSIS OF PPG WATER SAMPLE W-l
00
•vj
Compound
Vinylldene chloride
trans -Dichloroethylene
cis-Dichloroethylene
Chloroform
Ethylene dichloride
Methyl chloroform
Trichloroethylene
1 , 1 , 2-Trichloroethane
Bromoform
Tetrachloroethylene
m/e values
61, 96, 98, 63
61, 96, 98, 63
61, 96, 98, 63
83, 85, 47, 87
62, 49, 64, 63
97, 99, 61, 117
95, 130, 132, 97
83, 97, 85, 132
173, 171, 175, 93
166, 164, 129, 131
Intensity ratios
1.00:0.53:0.36:0.28
1.00:0.84:0.48:0.37
1.00:0.99:0.84:0.48
1.00:0.55:0.14:0.07
1.00:0.49:0.31:0.29
1.00:0.62:0.36:0.19
1.00:1.18:1.12:0.69
1.00:0.98:0.64:0.05
1.00:0.52:0.47:0.10
1.00:0.78:0.75:0.85
Spectrum
No.
15459
15474
15480
15484
15489
15497
15518
15527
15549
15565
Observed
intensity ratios
1.00:0.66:0.39:0.24
1.00:1.00:0.50:0.38
1.00:0.77:0.59:0.36
1.00:0.67:0.08:0.08
1.00:0.73:0.28:0.28
1.00:0.50:1.12:0.33
1.00:0.93:1.02:0.69
1.00:1.02:0.68:0.11
1.00:0.46:0.47:0.25
1.00:0.76:0.79:0.83
Conclusion
Present
Present
Present
Present
Present
Present
Present
Present
Present
Present
Product
Yes
No
No
No
Yes
Yes
Yes
No
No
Yes
NPDES
permit
a/
a/
a/
a/
a/
a/
a/
a/
a/
a/
a/ NPDES permit does not list specific compounds".
-------
UNITED STATES
PARTMENT OF THE INTERIOR
GEOLOGICAL SURVEY
oo
CD
Figure 27. Sampling locations and concentrations of VDC (ug/m3) in air at
Dixico, Dallas, Texas
-------
Table 39. AIR SAMPLING DATA FOR DIXICO, DALLAS, TEXAS
Area
North transect 0.1 mile
North transect 0 mile
North transect 0 mile
South transect 0.1 mile
East transect 0 mile
East transect 0 mile
West transect 0.1 mile
Sample
no.
1
2
3
4
5
6
7
Exact location
Corner, Polk St. and Nolte Dr.
North property line, 20 ft east of
Polk St.
North property line, 20 ft east of
Polk St.
Corner, Polk St. and Elmdale Dr.
Junction, Tyler St. and Vernon Ave .
Junction, Tyler St. and Vernon Ave.
South side Lebanon Ave., 0.1 mile
west of Polk St.
Total
sampling (hr)
23.6
23.6
23.6
23.25
23.6
23.4
23.25
Sampling
rate (4 /rain)
0.982
1.069
1.057
1.107
0.573
0.940
0.875
Total
volume (nr')
1.390
1.512
1.496
1.544
0.811
1.321
1.220
Sample
height (ft)
3
3
3
3
3
3
3
00
vO
-------
Meteorological Conditions - A summary of the weather conditions existing
during the air sampling period is given in Table 40.
Analysis of Air Samples
The results of the analysis of the air samples are listed in Table 41.
No VDC was detected at any of the site locations shown in Figure 27. The
wind rose pattern in Figure 28 shows that the sampling sites were correctly
positioned to sample VDC if it had been present. Based upon the calibration
curve for VDC, it can be estimated that the level of VDC in the air would
have to be less than 1 H-g/m .
No attempt was made to establish the identity of any other compounds
in the air sampler.
Analysis of Water Samples
The Dixico plant did not have any known accessible outfalls; consequent-
ly, no water analyses were performed.
DEWEtf AND ALMY, OWENSBORO, KENTUCKY
Field Sampling
A presampling survey was conducted on January 21, 1977. The plant is
situated in Daviess County, Kentucky, along the southern bank of the Ohio
River and about 3.2 miles east of the junction of US-60 and Kentucky Highway
212 off US-60. About 1.1 miles to the southwest is a power station operated
by Owensboro Municipal Utilities. Stacks located at the plant are barely
visible from the road and are no more than about 25 ft high.
The Dewey and Almy facility produces polybutadiene-styrene latex, poly-
vinyl acetate resins, polyvinyl chloride-vinylidene chloride copolymer resins,
and styrene-butadiene copolymer resins.
Processed wastewater is discharged into the Ohio River at a point along
the northeastern edge of the plant 35 ft out into the river and about 10 ft
below the surface.
Winds in this area, according to data from Evansville, Indiana, blow
from the northwest, west-northwest, and south-southwest in February and from
the south-southwest, northwest, and southwest on an annual basis.
Field sampling was carried out during the 23-hr period from 1600 Febru-
ary 22, 1977 to 1500 February 23, 1977. Ten air samples were deployed at
seven sites upwind, downwind, and laterally with respect to the plant. Water
samples were taken upstream and downstream from the point of discharge.
90
-------
Table 40. WEATHER CONDITIONS DURING SAMPLING AT
DIXICO, DALLAS, TEXAS
Wind
Time
Tuesday, February 8
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
Wednesday, February 9
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
Direction
ESE
SE
SE
SSE
SE
SSE
W
W
SSW
S
SSW
SSE
SSE
SSE
SSE
S
SSE
S
SSE
SSE
SSE
SSE
ESE
SE
SE
SSE
S
Speed (kts)
10
9
10
6
9
11
6
4
10
8
8
10
12
10
10
10
8
9
7
6
6
4
7
7
7
6
7
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
None
None
None
91
-------
Table 41. VDC CONCENTRATIONS IN AIR SAMPLES FROM DIXICO, DALLAS, TEXAS
vO
Sampling
station
1
2
3
4
5
6
7
Blank
No. 1
Blank
No. 2
Sampling
time
0915-0850
0905-0840
0905-0840
0950-0905
0925-0900
0935-0900
1000-0915
-
-
Vo lume
(m3)
1.39
1.512
1.496
1.544
0.811
1.32
1.22
-
-
Type of sample
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Confirmation by bromination
VDC Appearance of Presence of
(ug) Loss of VDC brominated VDC chloroethane
ma/
ND -
ND -
ND -
ND -
ND -
ND -
ND -
ND -
VDC
(ug/m3)
ND
ND
ND
ND
ND
ND
ND
ND
ND
a/ ND = not detected.
-------
NW
WNW
NNW
NNE
NE
SSE
Wind Speed,
Knots per Hour
Figure 28. Wind patterns during sampling at Dixico, Dallas, Texas
ENE
ESE
93
-------
Air Sampling - Air sampling was conducted using 10 samplers at seven sites.
The sites were established upwind, laterally, and downwind of the plant with
the majority being located downwind. Samplers were positioned at a height
of 4 ft from ground level on public land such as right-of-ways and riverbanks,
Sampling sites were located at approximately 0, 1/4, 1/3, and 2/3 mile from
the plant perimeter. Locations of the 10 stations are shown in Figure 29.
Site descriptions and air sampling data are given in Table 42. Sampling
along the banks of the Ohio River was terminated early in the interest of
safety.
Water Sampling - Two grab samples were obtained February 23, 1977. The up-
stream sample was taken from the Ohio River at a point approximately 2,000 ft
upstream from the point of discharge and about 50 ft from the south bank.
The downstream sample was taken approximately 2,000 ft downstream from the
point of discharge and about 50 ft from the south bank. The locations are
indicated in Figure 29.
Meteorological Conditions - A summary of the weather conditions existing
during the air sampling period is given in Table 43.
Analysis of Air Samples
The results of the analysis of the charcoal tubes are listed in Table
44. No VDC was detected at any of the sites shown in Figure 29. The wind
rose pattern in Figure 30 shows that the sampling Sites 1 and 8 were down-
wind of the plant during the sampling period. Based upon the calibration
curve for VDC, it can be estimated that the level of VDC in the air was less
than 1 u,g/m3.
The upwind and downwind samples did not show any notable difference
when analyzed by GC using flame ionization detection. Therefore no GC/MS
analysis of the air samples was performed to provide identification of other
compo und s.
Analysis of Water Samples
Water samples had been collected from the Ohio River approximately
2,000 ft upstream and downstream of the plant at a distance of 50 ft from
the piant-side bank.
The samples were analyzed first by GC-FID, and no VDC was found. The
downstream sample was then analyzed by GC/MS using the multiple ion selection
(MIS) mode. Again, no VDC could be detected.
The upstream and downstream water samples were virtually identical by
GC-FID analysis. Consequently no GC/MS analysis of the samples for components
other than VDC was attempted.
94
-------
vO
_L._
.. .L.J
C>* i
16-
jtf*
X
, 21
' In Light
\22'l
/
[W-q /, Production Area ,„—. .
/ \ ^N r' /~JC _ .^V ^
JOO^Th^N
T-"T-f^X.4 .
7: '. e T
o
i-1 v. While Chapel A ^,
Memornl GlrdensS \ /
0 1000 2000
! I I I I
Scale - Feet
Figure 29. Sampling locations and concentrations of VDC (ug/irP) in air
at Dewey and Almy, Owensboro, Kentucky
-------
Table 42. AIR SAMPLING DATA FOR DEWEY AND ALMY, OWENSBORO, KENTUCKY
o\
Area
North transect 0.75 mile
East-northeast transect
0.4 mile
South transect 0.6 mile
East transect 0.3 mile
East transect 0.3 mile
West transect 0.6 mile
West transect 0.6 mile
West-northwest transect
0 mile
West transect 0 mile
West transect 0 mile
Sample
no.
1
2
3
4
5
6
7
8
9
10
Exact location
South bank of Ohio River, east of
Yellow Creek
Drive-in theater, 0.2 mile east of
plant entrance
Entrance to road, 0.3 mile west of
plant entrance
South side US-60 immediately south
of plant
South side US-60 immediately south
of plant
North side of Ohio River straight
across from plant
North side of Ohio River straight
across from plant
South side of Ohio River Immediately
west-northwest of plant
South side of Ohio River immediately
west of plant
South side of Ohio River immediately
west of plant
Total
sampling
(hr)
17.83
20.20
19.83
19.67
19.67
16.17
16.17
16.92
17.5
17.5
Sampling
rate
tfl/min)
0.810
0.869
0.649
0.810
0.809
0.650
0.830
0.750
0.720
0.810
Total
vo lume
(m3)
0.867
1.052
0.772
0.956
0.955
0.631
0.805
0.761
0.756
0.851
Sample
height
(ft)
4
4
4
4
4
4
4
4
4
4
-------
.Table 43. WEATHER CONDITIONS DURING SAMPLING AT
DEWEY AND ALMY, OWENSBORO, KENTUCKY
Wind
Tuesday, February 22
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
Wednesday, February 23
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
Direction
SW
SSW
SW
S
SW
SSW
S
S
S
S
S
S
SSE
S
S
S
S
S
S
SSE
SSE
S
S
S
S
S
S
SSW
SSE
S
Speed (kts)
t
, 20
21
20
20
20
13
12
12
12
12
11
12
7
13
15
14
12
15
12
13
13
17
17
20
18
22
23
14
13
10
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
None
None
None
None
None
None
97
-------
Table 44. VDC concentration in air samples from Dewey and Almy,
Owensboro, Kentucky
Sampling
station
No. 1
No. 2
No. 3
vD
00
No. 4
No. 5
No. 6
No. 7
No. 8
No. 9
Sampling-
time
1645-1035
1810-1429
1820-1410
1835-1415
1835-1415
1720-0850
1720-0850
1745-1040
1700-1030
Volume
(m3)
0.867
1.052
0.772
0.956
0.955
0.631
0.805
0.761
0.756
Type of
sample
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup ch ar co a 1
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Confirmation by bromination
. Loss of Appearance of Presence of
M-g— VDck/ brominated VDO0- chloroethane
ND^ -
__
ND
ND
ND
ND
ND
ND
ND
ND
. 3
M-g/m
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
-------
Table 44. (Concluded)
Sampling
station
No. 10
Blank No. 1
Blank No. 2
Sampling— Volume Type of .
time (m ) sample Jig—
Top charcoal ND
1700-1030 0.851 Backup charcoal
Top charcoal ND
Backup charcoal
Top charcoal ND
Backup charcoal
Loss of Appearance of. Presence of
VDC£' brominated VDQ— chloro ethane (J.g/m
ND
__ — ND
ND
ND
ND
ND
vO
a/ 2/22/77 to 2/23/77.
b/ Based on analysis on Durapak OPN column.
£/ Based on analysis on OV-101 column.
d/ Not detected.
-------
NW
WNW
Wind Speed,
Knots per Hour
NNW
NNE
NE
SSE
SE
Figure 30. Wind Patterns During Sampling at Dewey and Almy
ENE
ESE
100
-------
TENNESSEE EASTMAN, KINGSPORT, TENNESSEE
Field Sampling
A presampling survey was conducted on March 7, 1977. The Tennessee
Eastman plant is located within the city limits of Kingsport, Tennessee, on
the north bank of the south fork of the Holston River. The Holston Army Am-
munition plant lies to the northwest. The Tennessee Eastman plant produces
a large number of organic chemicals and modacrylic fibers. There are a series
of seven discharge points from the plant into the Holston River including
one from a water treatment facility.
Field Sampling was carried out during the 27-hr period from 9:00 a.m.
March 8, 1977 to 12:00 p.m. March 9, 1977. Nine air samplers were deployed
at nine sites upwind, downwind, and laterally with respect to the plant.
Water samples were taken at four points upstream and downstream from the
plant.
Air Sampling - Air sampling was conducted using nine samplers at nine sites.
The sites were established upwind, laterally, and downwind of the plant.
Samplers were positioned at a height of 4 ft above ground level on public
land such as rights-of-way and schools. The sampling train consisted of two
charcoal tubes in series. Sampling sites were all located within 1/4 mile
of the plant perimeter. Sampling Site No. 1 was located within 15 ft of an
existing, permanent air sampling station. Locations of the nine stations
are shown in Figure 31. Site descriptions and air sampling data are given
in Table 45.
Water Sampling - A total of 12 grab water samples were taken at four sites
from each of three bridges on March 9, 1977. Samples Wl were taken from the
bridge where Highway 93 crosses the south fork of the Holston River. These
samples were upstream of all outfalls from the plant except outfall No. 008,
which is scheduled to be discontinued by July 1, 1977, according to the
Tennessee Eastman NPDES permit. Samples W2 and W3 were collected from the
south and north end of the bridge, respectively, where South Wilcox Drive
crosses the Holston River. This point is downstream of all outfalls except
No. 002 from the wastewater treatment plant. Sample W2 was directly down-
stream of outfall No. 003. Sample W4 was taken from the C. P. Edwards Bridge
about 1 mile downstream from the wastewater treatment facility. The locations
of the outfalls and the points where the water samples were collected are
shown in Figure 31.
Meteorological Conditions - A summary of the weather conditions existing
during the air sampling period is given in Table 46.
101
-------
o
NJ
Scale - Feet
Residential
Kingsport
Tennessee Eastman
Wastewater Treatment Plant
Tennessee
Eastman! \
II |\
{.Production \
Residential
Wilcox Road
River
3
Figure 31. Concentrations of VDC (p.g/m ) in Air at Nine Sampling Stations
at Tennessee Eastman, Kingsport, Tennessee
-------
Table 45. AIR SAMPLING DATA FOR TENNESSEE EASTMAN COMPANY, KINGSPORT, TENNESSEE
o
CO
Area
Northeast transect 0.5 mile
Northeast transect 0.25 mile
Northeast transect 0 mile
East-northeast transect 0.25 mile
South transect 0.25 mile
West-southwest transect 0 mile
West-southwest transect 0.25 mile
West transect 0 mile
East transect 0 mile
Sample
No.
1
2
3
4
5
6
7
8
9
Exact location
Fair Oaks St. & Sevier St.
Circle on Kenwood St.
Lincoln St. & Konnarock St.
Sherwood St. & Sherwood Ext.
Mo re land Dr. & Highway 93
Circle Dr. & Mason Rd.
North side Long Island School
Siler St. & River Rd.
Lincoln St. & Highway 93
Total
sampling
(hr)
25.2
24.9
25.8
24.9
25.2
25.3
25.2
25.5
25.1
Sampling
rate
(^/min)
0.87
1.00
0.87
0.83
0.78
0.95
0.86
0.91
0.87
Total
volume
(m^)
1.31
1.50
1.31
1.24
1.18
1.45
1.30
1.40
1.31
Sample
height
(ft)
4
4
4
4
4
4
4
4
4
-------
Table 46. WEATHER CONDITIONS DURING SAMPLING AT
TENNESSEE EASTMAN, KINGSPORT, TENNESSEE
Wind
Time
March 8
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
March 9
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
1200
Speed
(Kts.)
3
3
00
6
6
8
6
9
7
6
4
00
00
00
00
3
00
3
3
00
00
00
3
00
4
4
3
4
Direction
NE
ESE
-
ssw
wsw
wsw
w
w
sw
ssw
sw
-
—
-
-
ENE
_
E
ESE
-
-
-
ENE
-
E
NE
S
SSE
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
None
None
None
None
104
-------
Analysis of Air Samples
VDC Levels - The results of the analysis of the charcoal tubes by GC-FID
are listed in Table 47. No VDC was detected at any of the site locations
shown in Figure 31. The wind behavior for the day of sampling, shown as a
wind rose in Figure 32, indicates that Stations 9, 3, and 4 were the down-
wind stations. The chromatograms of these samples, particularly No. 9, in-
dicated the presence of significant concentrations of various components.
Sample No. 9 was considered to be the most likely one to contain any VDC
and consequently was analyzed by GC/MS. The presence of VDC was tested by
an EICP using the ions 61, 96, 98, and 63. The EICP for Sample No. 9 is
shown in Figure 33. The peak maxima for the four selected ions occur at the
appropriate time in the sample chromatogram and were absent in the field
blank. When compared to a standard, the results suggest that less than 1 u,g
of VDC was present in the 1.31 m3 air sample. Since this was the closes
downwind sample, the level of VDC in the upwind and farther downwind samples
appeared to be negligible.
Identification of Other Components - All air samples, when analyzed by GC-
FID, showed the presence of several other components at significant levels,
although the highest levels were at the downwind stations.
As previously mentioned, Sample No. 9 was analyzed by GC/MS. Initially,
selected ion plots were used to establish the presence or absence of specific
compounds.
Manual interpretation as well as a Cyphernetic search were also used.
Table 48 lists all the compounds tentatively identified and the basis of
the identification. The identified compounds are also noted on the sum plot
shown in Figure 34.
The two peaks after VDC appear to be structural isomers of hexane. The
next peak was matched by Cyphernetics computer search with cyclohexane and
2-hexene, having similar indices of 0.735 and 0.728, respectively. The next
cluster of peaks was shown by a series of selected ion and limited mass
range plots to be composed of five to six unresolved peaks. Identification
of the major components was rendered impossible due to the overlapping of
the peaks and the fact that most of the major fragment ions had saturated
the detector. The unsubtracted mass spectra have the appearance of being
C-6 methyl ketones. Tennessee Eastman lists 2-heptanone, methyl n-butyl
ketone, methyl n-heptal ketone, methyl isoamyl ketone, and methyl isobutyl
ketone as company products.
105
-------
Table 47. VDC CONCENTRATIONS IN AIR SAMPLES FROM TENNESSEE EASTMAN, KINGSPORT, TENNESSEE
Sampling
station
No. 1
No. 2
No. 3
No. 4
No. 5
No. 6
No. 7
No. 8
No. 9
Blank
a/
Sampling— Volume Type of
o
time (m ) sample
Top charcoal
1020-1121 1.31 Backup charcoal
Top charcoal
1030-1125 1.50 Backup charcoal
Top charcoal
1013-1118 1.31 Backup charcoal
Top charcoal
1037-1130 1.24 Backup charcoal
Top charcoal
1000-1111 1.18 Backup charcoal
Top charcoal
0927-1048 1.45 Backup charcoal
Top charcoal .
0945-1057 1.30 Backup charcoal
Top charcoal
0912-1042 1.40 Backup charcoal
Top Charcoal
1008-1115 1.31 Backup charcoal
Top charcoal
Backup charcoal
b/
li£~
-_
ND
~
ND
ND
ND
~
ND
—
ND
~
ND
—
ND
--
ND
--
ND
•*•
Loss of Appearance . Presence of
VDCJi/ brominated VDC— chlo roe thane u,g/m
ND
ND
ND
ND
__ __ ND
ND
ND
ND
ND
ND
__ __ — ND
ND
— ' • — ND
ND
ND
_- ~ ND
ND
_- — ND
ND
ND
&l 3/8/77 to 3/9/77.
_b/ Based on analysis using Durapak OPN column.
£/ Based on analysis using OV-101 column.
d/ Not detected.
-------
NNW
NW
WNW
WSW
Wind Speed,
Knots per Hour
NNE
SSE
NE
SE
ENE
Figure 32. Wind Patterns During Sampling at Tennessee Eastman
107
-------
61
roaa
7300
Figure 33. Extracted ion current plot of Tennessee Eastman
air sample No. 9 for VDC
108
-------
Table 48. SUMMARY OF GC/MS ANALYSIS OF TENNESSEE EASTMAN AIR SAMPLE NO. 9
Compound
Hexane isomer
Hexane isomer
Cyclohexane or
2 -hexene
Carbon tetrachloride
C-6 methyl ke tones
Benzene
Toluene
Basis of
identification m/e values
SIP3/ 61, 96, 98 63
MAN
GYPS!./
SIP 119, 117, 121, 82
MAN (?)
SIP 52, 51, 50, 79^
SIP 91, 92, 65, 63
Spectrum Observed intensity
Intensity ratios number ratios
1.00:0.53:0.36:0.28 7092 b/
7112
7116
7121
1.00:0.98:0.25:0.13 7129 1.0:1.0:0.2:0.2
7138,7142
1.00:0.71:0.59:0.35 7156 1.00:0.81:0.65:0.27
1.00:0.51:0.10:0.04 7187 All peaks saturated
Product or
known use
Yes
No
No
No
No
Yes
No
No
af SIP = Selected ion plots.
W Peaks were too small to make a meaningful measurement.
c/ MAN = Manual interpretation of mass spectrum.
d/ CYP " Identified by computer search of Cyphernetics file.
£/ m/e 78 peak was saturated and could not be used.
-------
v>
0)
O
leoac-
I ' I ' I ' I ' I
7200 7250
SPEC* 7Q1H - 7270 JK TEK« ERST filR MQ.9 4UL OPN JG 3-31-77
730
STEP
IN
Figure 34. Sum plot of Tennessee Eastman Air Sample No. 9
-------
Analysis of Water Samples
VDC Levels - The water samples collected at four points from the Holston
River were analyzed for VDC by GC using flame ionization detection. The
results are summarized in Table 49.
The only sample where VDC was detected by FID was W2 which had been
collected immediately downstream of outfall No. 003. Sample Wl was contami-
nated during analysis. The samples were analyzed by GC/MS using the multiple
ion scan mode and VDC was found in Samples W2 and W4. Identity was confirmed
by matching retention times and intensity ratios for the peaks monitored
with those of.a standard. The peak ratios are included in Table 49. No attempt
was made to quantify the level of VDC by GC/MS although the samples were es-
timated to contain less than 1 p.g/liter VDC.
Identification of Other Components - The upstream and downstream water sam-
ples did not show any notable differences when analyzed by GC with flame
ionization detection. Therefore no GC/MS analyses of the water samples for
any other components were performed..
DOW CHEMICAL, MIDLAND, MICHIGAN
Field Sampling
A presampling survey was conducted on March 21, 1977. The Dow Chemical
plant is located in Midland Township, Midland County, Michigan, immediately
south of the city of Midland, Michigan. The Tittabawassee River runs through
the plant complex. The city of Midland, Michigan, is on the north side of
the plant. To the east is a facility of Dow-Corning, to the south is a nu-
clear power plant under construction, and to the west is a rural area. The
Dow Chemical plant produces a large number of organic chemicals and Saran
Wrap. There are a series of five discharge points along the Tittabawassee
River.
The winds in this area generally blow from the west-southwest, west,
and northwest during the month of March and from the west-southwest, south-
west, and west on an annual basis.
Field sampling was carried out during the 30-hr period from 9:00 a.m.
March 22, 1977 to 3:00 p.m. March 23, 1977. Thirteen air samplers were de-
ployed at 13 sites upwind, downwind, and laterally with respect to the
plant. Water samples were taken at three points upstream and downstream
from the plant.
On March 22, 1977, a luncheon meeting was held with Diane M. Carlson
of the Michigan Department of Natural Resources, after which Ms. Carlson
was shown an air sampler. Ms. Carlson then accompanied the sampling crew
111
-------
Table 49. VDC CONCENTRATIONS IN WATER FROM
TENNESSEE EASTMAN', KINGSPORT, TENNESSEE
VDC Found
Sample
No. Volume analyzed
W-l 10 ml
GC-FID
*P
GC-MS-MIS
(peak ratios)—/
ND
Level
ND
W-2 10 ml 1 H>g Yes ~ 1 ppb
[1.00:0.71:0.55:0.39]
W-3 10 ml ND NA ND
W-4 10 ml ND Yes < 1 ppb
[1.00:0.63:0.42:0.33]
a/ Peak intensities for m/e 61, 96, 98, 63, for standard were
1.00:0.67:0.42:0.38.
b/ ND = Not detected.
112
-------
g©<^%
l'« -ifr~VAV'
^•o i (ND> r^\~
M 'vl D'k\L I A\
K..=;'. f-.V--:.:i I v-:-
^-^_ ^"~br«s?™__J ... ;•
4 f-^iijw^ 1 ^T^—^ r ^ ->^ . i * ^;.
Q
Figure 35. Sampling locations and concentrations of VDC (ug/nr5)
in air at Dow Chemical, Midland, Michigan
113
-------
Table 50. AIR SAMPLING DATA FOR DOW CHEMICAL, MIDLAND, MICHIGAN
Area
North transect 0.25 mile
Northwest transect 0 mile
Northeast transect 0 mile
South-southwest transect 1 mile
South transect 1 oile
South-southeast transect 1 mile
Southeast transect 1 mile.
Southwest transect 1 mile
Southeast transect 0.25 mile
Southwest transect 0.75 mile
East transect 0 mile
West transect 0 mile
Northwest transect 0 mile
Sample
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
Total
sampling
Exact location
Washington St. & Penn Central
Tracks
Ellsworth St. & Lyons St.
Bay City Rd. & Saginaw Rd.
0.7 miles east of Poseyville Rd.
on Gordonville Rd.
1.3 miles east of Poseyville Rd.
on Gordonville Rd.
1.7 miles east of Poseyville Rd.
on Gordonville Rd.
Gordonville Rd. & Tittabawassee
River
Little League baseball field on
i
E. Miller Rd.
Saginaw Rd. & Chesapeake & Ohio
Tracks
Road to Bechtel Power Plant
0.4 miles south of Bay City Rd.
on Saginaw Rd.
1.2 miles north of E. Miller Rd.
on Poseyville Rd.
Poseyville Rd. & Tittabawassee
River
(hr)
22.9
23.1
21.8
23.4
23.5
23.5
23.6
23.3
23.7
23.2
23.9
23.2
23.1
Sampling
rate
(liters/
rain)
0.95
0.92
1.04
0.90
0.85
0.80
0.67
0.73
0.68
0.91
0.89
0.92
0.78
Total
volume
(m3)
1.301
1.272
1.363
1.268
1.201
1.134
0.957
1.020
0.970
1.264
1.279
1.279
1.077
Sample
height
(ft)
4
4
4
4
4
4
4
4
4
4
4
4
4
-------
Table 51. WEATHER CONDITIONS DURING SAMPLING AT
DOW CHEMICAL, USA, MIDLAND", MICHIGAN
Wind
Time
March 22
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
March 23
0100
0200
0300
0400
0500
0600
0700
0800
0900
1000
1100
Speed
(Kts.)
17
17
16
20
18
13
17
13
17
13
10
7
7
7
7
7
10
7
7
6
7
7
7
11
12
13
9
Direction
N
NNW
N
N
N
N
NNW
WNW
NNW
WNW
NW
NW
WNW
W
W
W
WSW
W
WSW
WSW
SSW
SSW
SSW
sw
sw
SSW
SSW
Precipitation
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
Lt. Snow
Lt. Snow
Lt . Snow
Lt. Snow
115
-------
Table 52. VDC CONCENTRATIONS IN AIR AT DOW CHEMICAL, MIDLAND, MICHIGAN
Sampling
station
No. 1
No. 2
No. 3
No. 4
No. 5
No. 6
No. 7
No. 8
No. 9
No. 10
a/
Sampling-
time
1135-1030
1120-0025
1150-0937
1035-1000
1026-0957
1022-0953
1012-0950
1050-1006
1003-0945
1059-1010
Volume
(m3)
1.301
1.272
1.363
1.268
1.201
1.134
0.957
1.020
0.970
1.264
Type of
sample
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup charcoal
Top charcoal
Backup ch arco a 1
Top charcoal
Backup charcoal
b/
HD*/
ND
34
ND
ND
ND
ND
ND
ND
ND
ND
Confirmation by bromination
Loss of Appearance of. Presence of
VDG?-' brominated VDC— chloroethane M-g/m
ND
ND
ND
ND
Yes Yes No 25
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
— _- -_ ND
-------
Table 52. (Concluded)
Sampling
station
No. 11
No. 12
No. 13
Blank
a/
Sampling— Volume Type of
o D/
time (m°) sample |ig—
Top charcoal 29
0949-0941 1.279 Backup charcoal ND
Top charcoal ND
1106-1016 1.279 Backup charcoal
Top charcoal ND
1113-1019 1.077 Backup charcoal
Top charcoal ND
Backup charcoal ND
Confirmation by bromination
Loss of Appearance of . Presence of
VDQby brominated VDC— chloroethane M-g/m
Yes Yes No 23
ND
ND
ND
ND
ND
ND
ND
al 3/22/77 to 3/23/77.
£/ Based on analysis using Durapak OPN column,
£/ Based on analysis using OV-101 column.
d/ ND = Not detected.
-------
- ... < L .IV _K
63
-------
1 0 0 -3
=
:
90-i
'—
-
80-i
;
—
"* fl ~
i
60-=
D
;
£ 0-|
40 "I
30-f
£ 0 -r.
i
1 0-|
0-
i
""'""I""'""! j""i""|' • i"
'I"
1
,,t...j,..,, i
I
' " '""''
i ,i
'""I1"
It,,.
'"1
H
70-1
;
to-!
SOH
10-j
30-;
1 0-
i
i 'i"
0
.,....(...
1 L L i
50
'•'"I""'1-"
L
.
ijiimrcrdii
00 ISO ZOO 250 390
1 1 , 1 1 1 f 1 ( i t < f - 1 < f 1 1
30
eo
t 0
"""1 11"'!""
.
*
"
"
- c. n
-
-
;
-
-10
.
-
-
f
fl
0
0 50 100
SPECtt 6343 JS 6343 Jfl - 6340.JM
150
200 8 SO 30 0
STEP fl fl 3 S = 1 , I x B x S
Figure 38. Mass spectrum of VDC in Dow Chemical, Michigan, Air Sample No. 3
-------
IQQ-.
Z
90-j
:
:
8U1
E
?o-j
:
60-E
I
50-1
HO-I
30-E
EO-E
E
1 0-j
E
0
i
i1
.1
1
5
0-
,
|
1 00
,00
90-
tc-
70
C.D-
50^
HO
30H
eo-i
1 0-
t f L ;
0 50
•'""1 ""1 '
150
•
„
00 ISO £00 2SO 3BO
£ 0 0 £ B 0
30
£0
10
• 0
"i"T"i"T-i--|
3
-EO
SPECS
JS 6592.JM - 6589JH
Figure 39. Mass spectrum of VDC in Dow Chemical, Michigan, Air Sample No. 11
-------
The levels of VDC in the air, as micrograms per cubic meter, are given
in Table 52 and are in Figure 35. Figure 40 shows a wind rose pattern depict-
ing the wind behavior at the Tri City Airport, Freeland, Michigan, during
the time when samples were being collected. The wind had a western component
for over 80% of the sampling time and was evenly distributed from north-
northwest to south-southwest. This indicates that Stations 3 and 11 were
downwind of the plant and were the most likely stations where VDC would be
detected. In fact, these were the most likely stations where VDC would be
detected. In fact, these were the two locations where VDC was found; 25
at Station 3, and 23 u,g/m3 at Station 11. The similarity in levels is in
agreement with the wind patterns. No VDC was found at the upwind or crosswind
stations.
Identification of Other Components - Most of the major peaks in the two Dow
air Samples Nos. 3 and 11 were identified by EICPs. Upon close examination
of the data, it was determined that one peak eluting immediately after trans-
dichloroethylene in both of the samples remained unidentified. One subtracted
spectrum was searched against the Cyphernetics file and identified as 2-methyl-
pentane with a similarity index of 0.475. The peaks in the two Dow samples
had the same retention time as the one of the peaks identified as a hexane
in the Tennessee Eastman air sample. Tables 53 and 54 contain a complete
listing of the compounds identified in air Samples Nos. 3 and 11, the basis
of the identification, and whether the chemical is known to be a product or
is heavily used by the company. All of the identified compounds in air Samples
Nos. 3 and 11 are also noted on the sum plots shown in Figures 41 and 42,
respectively.
Analysis of Water Samples
VDG Levels - The water samples collected at three locations along the
Tittabawassee River were analyzed for VDC by GC using flame ionization detec-
tion. The analytical results are summarized in Table 55. No VDC was detected
in any of the samples. The samples were analyzed by GC/MS in the MIS and VDC
was detected in Sample W3. Identity was based up peak intensity ratios plus
matching retention times. The peak intensity data are included in Table 55.
The level of VDC was not quantified but was estimated to be between 0.1 and
1 ppb.
Identification of Other Components - Samples Wl and W3 were analyzed by GC/MS
in an attempt to confirm VDC and to establish, if po$sible, the identity of
any additional compounds present. The upstream Sample Wl showed four major
peaks in the sum plot, while Sample W3 had about eight major and six minor
additional peaks in its sum plot.
121
-------
NNW
NW
WNW
WSW
Wind Speed,
Knots per Hour
NNE
NE
SE
ENE
Figure 40. Wind Patterns During Sampling at Dow Chemical
122
-------
Table 53. SUMMARY OF GC/MS ANALYSIS OF DOW AIR SAMPLE NO. 3
Basis of
Compound identification
Vinylidene chloride SIP a' MAN^-'
trans-Dichloroethylene SIP
2-Methylpentane GYP-'
Methylene chloride SIP
Methyl chloroform SIP
Benzene SIP
Toluene SIP
ho
OJ
Spectrum Observed intensity
m/e values Intensity ratios number ratios
61 96, 98, 63 1.00-0.53-0 36-0 28 6343 1 00-0 60-0 33-0 44
61, 96, 98, 63 1.00:0.84:0.48:0.37 6359 1.0:0.8:0.5:0.3
6361
84, 86, 51, 88 1.00:0.71:0.37:0.09 6366 1.00:0.49:0.30:0.10
97, 99, 117, 119 1.00:0.62:0.19-0.19 6396 1 0:0.7-0.2:0.1
78, 52, 51, 50 1.00:0.17-0.15:0.12 6406 1.0-0.6:0.5:0.5
91, 92, 65, 63 1.00:0.51:0.10:0.04 6437 1.0:0 54-0.17-0.08
Product or
known use
Yes
No
No
No
No
No
No
a/ SIP = Selected ion plots.
tj/ MAN = Manual interpretation of mass spectrum.
£/ GYP = Computer search of Cyphernetics file.
-------
Table 54. SUMMARY OF GC/MS ANALYSIS OF DOW AIR SAMPLE NO. II
NJ
Basis Spectrum Observed intensity
Compound identification m/e values Intensity ratios number ratios
Vinylidene chloride SIP.S/ MAN^/ 61, 96, 98, 63 1.00:0.53:0.36:0.28 6592 1.00:0.54:0.30:0.18
trans-Dichloroethylene SIP 61, 96, 98, 63 1.00:0.84:0.48:0.37 6608 1.0:0.5:0.3:0.2
2-Methylpentene CYP-/ - - 6611
Methylene chloride SIP 84, 86, 51, 88 1.00:0.92:0.37:0.09 6616 1.00-0.75-0 60-0 10
Methyl chloroform SIP (?) 97, 99, 117, 119 1.00:0.62:0.19:0.19 6647 1.0:0.5:0.2:-
Benzene SIP 78 52 51 50 1.00:0.17:0.15-0.12 6657 d/
Toluene SIP 91 92 65 63 1.00:0.51-0.10:0.04 6688 d/
Product or
known use
Yes
No
No
No
No
No
No
a/ SIP = Selected ion plots.
b/ MAN = Manual interpretation of mass spectrum.
cl CYP = Computer search of Cyphernetics file.
d/ One or more peaks saturated.
-------
Ui
1 I
ssoa
Figure 41. Sum plot of Dow Chemical, Michigan, Air Sample No. 3
-------
Figure 42. Sum plot of Dow Chemical, Michigan, Air Sample No. 11
-------
Table 55. VDC CONCENTRATIONS IN WATER FROM
DOW CHEMICAL, MIDLAND, MICHIGAN
VDC found
Sample Volume GC-MS-MIS
No. analyzed GC-FID (peak ratios)—/ Level
W-l 10 ml ND^/ ND ND
W-2 10 ml ND ND ND
W-3 10 ml ND Yes < 1 ppb
(1.00:0.40:0.26:0.20)
&l Peak intensities for m/e 61, 96, 98, 63 for standard
were 1.00:0.36:0.21:0.14.
b/ ND = Not detected.
127
-------
All of the compounds identified as being unique to Sample W3 are noted
on the sum plot in Figure 43. Table 56 contains a complete listing of the
identified compounds, the basis of the identification, and whether the com-
pound is listed as a company product.
Dimethoxymethane ..as identified by searching mass spectrum No. 5497
against the Cyphernetics file. A similarity index of 0.654 was obtained.
The authentic spectrum was retrieved from the Cyphernetics file and closely
resembled the sample spectrum. Also, dimethoxymethane has a low enough boil-
ing point (44°C) to elute at the observed retention time. While Dow does
not list dimethoxymethane as a product, it does produce the structural iso-
mer 2-methoxyethanol or methyl cellosolve.
Three brominated alkanes, bromochloromethane, dibromomethane, and 1-
chloro-3-bromopropane were first identified by a manual study of the data.
Authentic mass spectra for all three compounds were obtained from the Cypher-
netics file and in all cases excellent agreement was obtained.
•f
DRINKING WATER FROM FIVE U.S. CITIES
Field Sampling
The plan to collect drinking water samples from five major cities had
been made by the previous project officer, Perry Brunner. The plan was to
utilize local EPA or other personnel to collect and ship the water samples
to MRI. Mr. Brunner made initial communication with the appropriate personnel
to arrange for the collection of samples and establish the contact person
for MRI. MRI then made initial contact with the individual who was to do the
actual sampling. Table 57 lists the cities from which samples were collected
as well as general information about the water supply.
Special water sampling kits were constructed at MRI and sent to the
five cities. The kits contained 10, 25-ml screw cap vials that had been
cleaned at 400°C. The vials were all fitted with Teflon® septa and five were
labeled "Pre-chlorination" and five were labeled "Post-chlorination." The
kit was actually a polystyrene shipping carton that had been modified to
hold the 10 vials and two ice packs as well. Detailing sampling and shipping
instructions accompanied each sampling kit. The kits were all shipped on
March 29. Table 57 lists the samples which were collected and the date on
which they arrived at MRI.
Analysis of Water Samples
The pre- and post-chlorinated drinking water from four of the five cities
was analyzed for VDC using the previously established GC/MS/MIS procedure.
128
-------
30000
20000-
vD
toooo-
.5
X
1
_0
-C
u
-"•v, .,
A
SPEC» 5H70 *
5501 5550
5670 JP DOW til HO.3 IOML CUL5QO J.8.3-30-77
5600
STEP SPEC3«1 INT» 1COC
565C
Figure 43. Sum plot of Dow Chemical, Michigan, Water Sample W-3
* Peaks found in upstream sample
-------
Table 56. SUMMARY OF GC/MS ANALYSIS OF DOW WATER SAMPLE W-3
Compound
Methylene chloride
Vinylidene chloride
Dime thoxyme thane
Bromoch lorome thane
Chloroform
Ethylene dichloride
OJ
<-5 Dibromome thane
Carbon tetrachloride
Benzene
Trichloroethylene
l-Chloro-3-bromopropane
Tetrachloroethylene
Basis of
identification m/e values Intensity ratios
SIP^/ 84, 86, 51, 88 1.00:0.71:0.34:0.10
Mist/ 61, 96, 98, 63 1.00:0.36:0.21:0.14
CYP^/
MAN, SP-/
SIP 61 96, 98 63 1.00:0.84:0.48:0 37
SIP 83, 85, 47, 87 1.00:0.55:0.14:0.07
SIP 62, 64, 98, 100 1.00:0.31:0.08:0.06
MAN, SP -
SIP 119, 117, 121, 82 1.00:0.98:0.25:0.13
SIP 78, 52, 51, 50 1.00:0.17:0.12:0.10
SIP 130, 132, 95, 97 1.00:0.94:0.86:0.59
MAN, SP -
SIP 164, 166, 129, 131 1.00:0.78:0.75:0.85
Spectrum Observed intensity
number ratios
5476 1.00:0.77:0.45:0.15
~ 5499 1.00:0.40:0.26:0.20
5497
5508
5522 1 0-0 6-0 6-0 4
5528 1.00:0.42:0.17:0.13
5534 1.0:0. 6:0. 2:0. 1£/
5539
5550 1.0:0.8:0.1:0.1
5576 1.00:0.12:0.14:0.12
5577 1.00:1.00:1.67:0.67
5623
5665 1.00:0.80:0.80:0.95
Product
No
Yes^/
e/
Yes
No
No
No
Yes
No
No
No
No
No
a/ SIP = Selected ion plots.
b/ MIS = Multiple ion scan.
cl VDC is used as a starting material.
d/ CYP = Computer search of Cyphernetics file.
£/ Dow produces the structural isomer methyl cellosolve.
il Comparison to authentic spectrum.
£/ m/e 62 peak saturated.
-------
Table 57. DRINKING WATER SAMPLES FROM FIVE U.S. CITIES
Treatment Type of Type of Results of
City plant Contact supply raw water NORS study— Samples collected
Cincinnati, b_l J. W. Oehler Surface Industrial waste £/
Ohio "~ "~
Lawrence, b_/ T. Murphy Surface Industrial waste £/
Massachusettes
Miami, Preston D. H. Waddell Ground Natural waste 0.1 ug/
Florida
Waterford, b/ T. DeGaetano Ground d/ d/
New York "~
Philadelphia, Torresdale B. Blankenship Surface Municipal waste < 0.1 pg/
Pennsylvania
5 at pre-chlorination
5 at post-chlorination
5 at pre-chlorination
5 at post-chlorination
5 at pre-chlorination
5 at post-chlorination
5 at pre-chlorination
5 at post-chlorination
5 at pre-chlorination
5 at post-chlorination
Date
received
4/5/77
4/14/77
4/12/77
4/18/77
4/7/77
&/ "Preliminary Assessment of Suspected Carcinogens in Drinking Water: Report to Congress," EPA, Washington, D.C.
b/ No specific treatment plant designated.
£/ VDC detected but not quantified.
d/ Unable to determine.
-------
The sample from Waterford, New York, did not arrive in time to be analyzed
on the scheduled analysis date, April 15. The results of the analyses are
given in Table 58. No VDC was found in any of the samples except the pre-
(90.59 u,g/liter) and post-chlorination (0.45 ^g/liter) samples from the
Preston plant, Miami, Florida. The MIS plots for the two Miami samples and
the 1 ng VDC standard are shown in Figure 44. Confirmation of VDC is based
upon matching retention times and matching peak intensity ratio. The inten-
sity ratio data are footnoted in Table 46. The estimated detection limit,
using a 10-ml sample, was 0.005 ppb.
SOLID FINAL PRODUCTS
Field Sampling
Samples of solid final products manufactured from VDC were requested
from about 30 companies. The companies were selected to be representative
of total industrial use of VDC. The samples described in Table 59 either
received from the manufacturer or purchased locally.
Analysis of Solid Products
Four types of commercial PVDC products were analyzed for residual VDC
monomer, and the results are summarized in Table 60. VDC was detected in
the two samples of Saran Wrap but not in the monofilaments, modacrylic fibers
or the meat packaging film. The Saran Wrap, presumably produced in 1974,
was found to contain 41 ppm VDC compared to 58 ppm found in the same sample
in October 1976. The Saran Wrap, presumably produced in 1976, was found to
contain only 4.9 ppm VDC. This is consistent with the recent statement by
Dow Chemical that its Saran Wrap contains "no more than 10 ppm of unreacted
monomers."—'
The estimated detection limit for the procedure was 0.005 ppm. Con-
sequently the monofilaments, the modacrylic fiber, and the meat packaging
film were judged to contain less than 0.005 ppm VDC.
132
-------
Table 58. VDC CONCENTRATIONS IN DRINKING WATER FROM FOUR U.S. CITIES
Sample
Cincinnati pre-chlorination
Cincinnati post-chlorination
Lawrence pre-chlorination
Lawrence post-chlorination
Miami pre-chlorination
Miami post-chlorination
Waterford pre-chlorination
Waterford post-chlorination
Philadelphia pre-chlorination
Philadelphia post-chlorination
Sample
size (ml)
10
10
10
10
10
10
10
10
10
10
VDC found
(ng)
ND
ND
ND
ND
o!453/
b/
I/
ND
ND
VDC
(ug/l)
ND
ND
ND
ND
0.059
0.045
b/
ND
ND
a/ The measured intensity ratios for the ions 61, 63, 96, and 98 were
1.00:0.39:0.33:0.20 for standard VDC; 1.00:0.39:0.33:0.22 for Miami
pre-chlorination; and 1.00:0.36:0.31:0.21 for Miami post-chlorination.
W Sample not received in time for scheduled analysis.
133
-------
xse
re 61
i
1284
HI MOTES
62/ MI/ 1 NC VDC 10ML CM 1300 J P A IS 17
A
xsa
I
1 2
MIMUTE3
68/ HI/ MIAMI r LOB IDA PRE CL 1QML JC A 13 77
B
M/E 61 X38
i 2 a A
MINUTES
67/ Ml/ MIAMI FLORIDA POST CL 1OML JC A 13 77
C
(a) One nanogram
standard
(b) Miami, Florida,
pre-chlorinated
drinking water
(c) Miami, Florida,
po st -ch lo r inate d
drinking water
Figure 44. Multiple ion scan of VDC in
drinking waters
134
-------
Table 59. SOLID FINAL PRODUCT SAMPLES
Date
Sample description Source received
Saran Wrap, Purchased locally 10/76
carton coded 2474
Saran Wrap, Purchased locally 3/77
carton coded 060376
Amtech monofilaments, Supplied by 11/3/76
produced from Dow B-Copolymer manufacturer
Monsanto modacrylic fiber, Supplied by 11/24/76
1 Ib manufacturer
Oscar Meyer, Supplied by 12/9/76
films A, B, and C manufacturer
135
-------
Table 60. VDC CONCENTRATIONS IN SOLID PRODUCTS
Sample
Saran Wrap
(carton coded 2474)
Saran Wrap
(carton coded 060376)
Amtech monofilaments
Monsanto modacrylic
fiber
Oscar Meyer Film A
Oscar Meyer Film B
Oscar Meyer Film C
Source
Local grocery store
Local grocery store
Manufacturer
Manufacturer
Manufacturer
Manufacturer
Manufacturer
Sample size
range (g)
1.96-2.43
1.83-2.11
0.20-1.29
1.00-1.42
2.40-3.03
3.98-4.27
3.47-3.71
(58'
a/ Average of three samples.
b/ Analyzed October 1976.
c/ Five samples analyzed in duplicate.
d/ Analyzed in duplicate.
136
-------
REFERENCES
1. "Sampling and Analysis of Selected Toxic Substances, Task IV, Ethylene
Dibromide, EPA 560/6- 76-021.
2. Personal communication, John Cobler, Dow Chemical, USA, Midland,
Michigan.
3. "Vinylidene Chloride Monomer Emissions from the Monomer, Polymer, and
Polymer Processing Industries," A. D. Little, Inc., EPA Contract No.
68-02-1332, Task Order No. 13, April 1976.
4. Chem. and Eng. News, 21, March 17, 1977.
137
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse, before completing)
1. REPORT NO.
EPA-560/6-77-026
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Environmental Monitoring Near Industrial Sites:
Vinylidene Chloride
5. REPORT DATE
October 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
John E. Going, James L. Spigarelli
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research Institute
425 VoIker'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, D.C. 20400
13. TYPE OF REPORT AND PERIOD COVERED
Task Final 6/76-9/77
14. SPONSORING AGENCY CODE
EPA-OTS
15. SUPPLEMENTARY NOTES
16. ABSTRACT
A sampling and analysis program was conducted to determine the levels
of vinylidene chloride in air and water around six industrial sites, in se-
lected municipal drinking waters and in selected polymeric materials. The
industrial sites included monomer producers, polymer producers and poly-
mer users.
Air was collected at perimeters of the industrial sites as 24-hr com-
posite samples using charcoal adsorption tubes. Grab water samples were
collected up- and downstream, at the plant discharge and at the municipal
water treatment plants.
Q
Air levels ranged from. 52 H»g/m to nondetectable at the industrial
sites; water levels ranged from 550 ^g/liters to nondetectable at the
plant sites and from 0.06 n-g/liters to nondetectable in the drinking waters.
The air and water samples were analyzed by GC and results confirmed by
GC/MS. Other components in the samples were identified by GC/MS.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATl Held/Group
Vinylidene chloride
Air
Water
Sampling
Analysis
Monomer plants
Polymer plants
Environmental monitoring
Industrial plants
Organic
chemistry
18. DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220—1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
138
------- |