: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

-------