United States
              Environmental Protection
              Agency
            Environmental Monitoring and Support
            Laboratory
            Research Triangle Park NC 27711
EPA-600 4-79-047
Augusi 1979
              Research and DevaSopment
SERA
Measurement of
Perchloroethylene  in
Ambient  Air

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination of traditional grouping  was  consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental  Health Effects Research
      2.  Environmental  Protection Technology
      3.  Ecological Research
      4.  Environmental  Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports

This report has been assigned to the ENVIRONMENTAL MONITORING series.
This series describes research conducted to develop new or improved methods
and  instrumentation for the identification and quantification of environmental
pollutants at the lowest conceivably significant concentrations.  It also includes
studies to determine the ambient concentrations of pollutants in the environment
and/or the variance of pollutants as a function of time or meteorological factors.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
MEASUREMENT OF PERCHLOROETHYLENE IN AMBIENT AIR


                      by
G. F. Evans, R. E. Baumgardner, J. E. Bumgarner
P. L. Finkelstein, J. E. Knoll and B. E. Martin
Environmental Monitoring and Support Laboratory
        Environmental Protection Agency
 Research Triangle Park, North Carolina 27711
                      and
  A. L. Sykes, D. E. Wagoner and C. E. Decker
          Research Triangle Institute
 Research Triangle Park, North Carolina 27709
ENVIRONMENTAL MONITORING AND SUPPORT LABORATORY
      OFFICE OF RESEARCH AND DEVELOPMENT
     U. S. ENVIRONMENTAL PROTECTION AGENCY
 RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711

-------
                               DISCLAIMER
     This report has been reviewed by the Environmental Monitoring and
Support Laboratory, U.S. Environmental Protection Agency, and approved
for publication.  Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
                                   it,

-------
                                  FOREWORD
     Measurement and monitoring research efforts are designed to anticipate
potential environmental problems, to support regulatory actions by developing
an in-depth understanding of the nature and processes that impact health and
the ecology, to provide innovative means of monitoring compliance with regu-
lations and to evaluate the effectiveness of health and environmental  pro-
tection efforts through the monitoring of long-term trends.   The Environmental
Monitoring and Support Laboratory, Research Triangle Park, North Carolina,
has the responsibility for:  assessment of environmental monitoring technology
and systems; implementation of agency-wide quality assurance programs  for air
pollution measurement systems; and supplying technical support to other
groups in the Agency including the Office of Air, Noise and  Radiation, the
Office of Toxic Substances and the Office of Enforcement.

     This study was conducted at the request of the Office of Toxic Substances
for use in health risk assessment.  A system for measurement of perchloroethy-
lene in ambient air was developed and evaluated.  Field monitoring was con-
ducted and ambient perchloroethylene concentrations reported for three     -i
metropolitan areas.  Precision and accuracy of the reported  data were  char-
acterized through implementation of a quality assurance program.


                                             Thomas R. Mauser
                                                 Di rector
                                       Environmental  Monitoring and
                                            Support Laboratory
                                    iii

-------
                                ABSTRACT


     Perchloroethylene (i.e., tetrachloroethylene) is an organic solvent
widely used in dry cleaning and industrial metal degreasing operations.
In March 1978, in response to a carcinogenic risk study by the National
Cancer Institute, the U.S. Environmental Protection Agency undertook
a program to measure perchloroethylene concentrations in ambient air.
This program was initiated by the Office of Toxic Substances and supported
by the Office of Air Quality Planning and Standards.  The research was
conducted by the Environmental Monitoring and Support Laboratory with
contractual assistance from the Research Triangle Institute.

     Short-term field studies were conducted in three major metropolitan
areas, selected on the basis of the number, density, and size of perchloro-
ethylene  emission sources as well as the proximity of these sources to
centers of high population density.  Dry cleaning, a ubiquitous activity
scattered throughout any metropolitan area, increases in volume propor-
tionately with population density.  Hence, New York City, with the greatest
population density in the U.S., was selected as a study area.  Metropolitan
Houston was chosen primarily because the Diamond Shamrock plant, located in
suburban Deer Park, is one of the largest perch!oroethylene producers  in the
nation.  Finally, metropolitan Detroit was included because of the number of
metal degreasing operations located in the area.

     Ten monitoring sites were established within each of the three metropol-
itan areas.  Most site locations were selected to represent the air quality
to which the population is typically exposed (i.e., commercial and residen-
tial areas); however, a few source-specific sites were included in the study
design.  A combination of existing sites operated by state and local agencies
and new sites established expressly for this study was utilized.  Twenty-
four hour integrated samples were collected on activated charcoal at each
site for a period of 10 consecutive days.   In addition, meteorological data
were obtained at one of the monitoring sites in each city and a comprehen-
sive quality assurance plan was maintained throughout the program.

     Observed perchloroethylene concentrations ranged from 0.2'to 10.6 ppb
in New York City, from below detectable (<0.1) to 4.5 ppb in Houston,  and
from below detectable to 2.2 ppb in Detroit.  The higher concentrations
tended to occur where source strengths were greatest and significant day-of-
week variations were apparent.


     This report covers a period from March 1, 1978, through February  28,
1979, and work was completed as of February 28, 1979.
                                     IV

-------
                                  CONTENTS


Abstract	     iv

Figures	     vi

Tables	    vii

Acknowledgments  	   viii

     1.   Introduction	      1
     2.   Summary and Conclusions  	      2

     3.   Experimental Procedures  	      3
               Selection of Sampling Sites 	      3
                    New York	      3
                    Houston  	      6
                    Detroit  	      9
               Collection and Handling of Field Samples   	      9
               Laboratory Analytical Procedures  	     14

     4.   Results and Discussion	     16
               New York, New York  	     17
               Houston, Texas  	     30
               Detroit, Michigan 	     43

     5.   Quality.Assurance  	     56
               Precision of the Analytical Technique 	     56
               Repeatability of the Measurement Method 	     58
               External Quality Assurance  	     58
               Assessment of Static Contamination and Breakthrough .  .     64
               Estimation of Concentration Intervals 	     64

References	     68

Appendices	     69

     A.   Method for the Determination of Perch!oroethylene
           in Ambient Air  	     69
     B.   Meteorological Data Summary  	     77

-------
                                   FIGURES
Number
Page
 1   Sampling Site Locations in New York, New York 	    5
 2   Sampling Site Locations in Houston, Texas 	    8
 3   Sampling Site Locations in Detroit, Michigan  	   11
 4   Diagram of Perch!oroethylene Sampler  	   13

     Ambient Concentrations of Perchloroethylene in:

 5   New York, New York - Friday, 8/18/78  	   20
 6   New York, New York - Saturday, 8/19/78	21
 7   New York, New York - Sunday, 8/20/78  	   22
 8   New York, New York - Monday, 8/21/78	23
 9   New York, New York - Tuesday, 8/22/78 	   24
10   New York, New York - Wednesday, 8/23/78 	   25
11   New York, New York - Thursday, 8/24/78  	   26
12   New York, New York - Friday, 8/25/78  	   27
13   New York, New York - Saturday, 8/26/78  	   28
14   New York, New York - Sunday, 8/27/78	29
15   Houston, Texas - Saturday, 9/16/78  	   33
16   Houston, Texas - Sunday, 9/17/78  	   34
17   Houston, Texas - Monday, 9/18/78  	   35
18   Houston, Texas - Tuesday,  9/19/78 	   36
19   Houston, Texas - Wednesday, 9/20/78 	   37
20   Houston, Texas - Thursday, 9/21/78  	   38
21   Houston, Texas - Friday, 9/22/78  	   39
22   Houston, Texas - Saturday, 9/23/78  	   40
23   Houston, Texas - Sunday, 9/24/78  	   41
24   Houston, Texas - Monday, 9/25/78  	   42
25   Detroit, Michigan - Friday, 10/27/78  	   46
26   Detroit, Michigan - Saturday, 10/28/78  	   47
27   Detroit, Michigan - Sunday, 10/29/78  	   48
28   Detroit, Michigan - Monday, 10/30/78  ...  	   49
29   Detroit, Michigan -'Tuesday, 10/31/78 	   50
30   Detroit, Michigan - Wednesday, 11/1/78  	   51
31   Detroit, Michigan - Thursday, 11/2/78 	   52
32   Detroit, Michigan - Friday, 11/3/78 	   53
33   Detroit, Michigan - Saturday, 11/4/78 	   54
34   Detroit, Michigan - Sunday, 11/5/78 	   55
                                     vi

-------
                                   TABLES
Number                                                                 Page
 1   Summary of Ambient Perchloroethylene Concentrations 	    2
 2   Sampling Site Locations in New York, New York 	    4
 3   Sampling Site Locations in Houston, Texas 	    7
 4   Sampling Site Locations in Detroit, Michigan  	   10
 5   National Weather Service Data from New York, New York 	   18
 6   Ambient Concentrations of Perchloroethylene in New York, N.Y. . .   19
 7   National Weather Service Data from Houston, Texas 	   31
 8   Ambient Concentrations of Perchloroethylene in Houston,  Texas . .   32
 9   National Weather Service Data from Detroit, Michigan  	   44
10   Ambient Concentrations of Perchloroethylene in Detroit,  Michigan.   45
11   Reanalysis of Desorbed Samples  	   57
12   Analysis of Duplicate Field Samples 	   59
13   Analysis of External  Quality Control Samples  	   61
14   Analysis of Quality Control Samples by Level  	   62
15   Analysis of Tenax Field Samples 	   63
16   Analysis of Charcoal  Tube Field Blanks  	   65
17   Individual Analysis of Front and Back Charcoal Sections  	   66
                                    vii

-------
                              ACKNOWLEDGMENTS

     The authors received much valuable guidance and technical  direction
from the following individual employees of the U.S. Environmental  Protection
Agency (EPA):  John Smith, Office of Toxic Substances; Ken Greer and George
Wahl, Office of Air Quality Planning and Standards; Seymour Hochheiser,
Environmental Monitoring and Support Laboratory; Ray Werner, Regional  Office II;
Jerry Reagan, Regional Office V; and Don Payne, Regional  Office VI.

     In addition, the EPA is indebted to the following agencies for their
substantial contributions to the design and conduct of the field monitoring
activities:  the New York State Department of Environmental Conservation,
the New York City Department of Air Resources, the Texas  Air Control Board,
the City of Houston Department of Public Health, and the  Wayne  County
(Michigan) Health Department.
                                   viii

-------
                                  SECTION I

                                INTRODUCTION
     In March 1978, the Environmental Monitoring and Support Laboratory
(EMSL) at Research Triangle Park (RTF), North Carolina, undertook a research
program to investigate perchloroethylene (PERC) levels in the ambient air of
our nation's urban areas.   This research was initiated by the Office of
Toxic Substances (OTS) and was partially funded by the Office of Air Quality
Planning and Standards (OAQPS) in support of concurrent health risk assess-
ment and regulatory activities.  Planning and management for this program
were accomplished through a task force comprised of representatives from
each component of EMSL with active participation and technical review pro-
vided by OTS, OAQPS, and Research Triangle Institute (RTI).

     Although some grab-sample measurements of PERC (i.e., tetrachloro-
ethylene) in ambient air have been previously reported in the literature (1),
a well-defined methodology for sampling and analysis at the anticipated
ambient levels (sub-ppb) was not available at the inception of this program.
Therefore, RTI developed methodology for the collection and quantitative
analysis of PERC in 24-hour integrated samples over the concentration range
of 0.10 to 10.00 ppb.   A method based on adsorption by activated charcoal,
desorption by carbon disulfide (CSpJ/methanol, separation by gas chromatog-
raphy and detection by electron capture detector (ECD) was developed and
tested under both laboratory and field operating conditions.   A description
of the method is given in  Appendix A.

     EMSL designed and conducted a field monitoring program,  incorporating
such factors as the nature, size, and density of emission sources within
major population centers and the likely impact of prevailing  meteorological
conditions.  Short-term monitoring studies were conducted in the greater
metropolitan areas of New York City,  Houston,  and Detroit.  Within each
study area,  10 PERC monitoring locations and a single meteorological  mon-
itoring location were established with the assistance of personnel from the
appropriate  EPA regional office and state and local agencies.

-------
                                  SECTION 2

                           SUMMARY AND CONCLUSIONS


     Ambient PERC concentrations observed in the field monitoring program
are summarized in Table 1.  All samples collected in New York City exceeded
the detection limit of the measurement method (0.10 ppb), half exceeded 1  ppb,
and a single observation was slightly greater than 10 ppb.  In the Houston
and Detroit areas, by contrast, about 90 percent of all measurements were
less than 1 ppb.  Nearly half (46 percent) of the Houston samples and 12
percent of the Detroit samples fell below the detection limit.


          TABLE 1.  SUMMARY OF AMBIENT PERCHLOROETHYLENE CONCENTRATIONS


   Cl-.         NQ    Cumulative Frequency (%) less than:   Concentration (ppb)
                     0.10 ppb*    1.00 ppb    10.00 ppb    Min   Max   Median
New York
Houston,
Detroit,
, NY
TX
MI
95
96
100

45.
12.
0
8
0
49.
90.
90.
5
6
0
98.
100.
100.
9
0
0
0.16
<0.10
<0.10
10.61
4.52
2.16
1.00
0.11
0.35

*Detection Limit

     The data suggest that ambient PERC concentrations in urban areas are
generally proportional to population density  (the populations per square
mile of Houston and Detroit are approximately 11 percent and 47 percent,
respectively, of that in New York City (2)).  Maximum concentrations occur
in the vicinity of point sources of PERC emissions such as industrial-scale
dry cleaning plants and PERC manufacturing facilities.  A distinct day-of-
week concentration pattern exists with the higher concentrations occurring
midweek, tapering off to a minimum on Sunday.  Ambient levels of PERC     ' -
increase during periods of atmospheric stagnation caused by light winds and
limited vertical mixing.  Background concentrations occurring at nonurban
and upwind locations appear to be minimal (<0.10 ppb).

     The precision of the analytical method employed, expressed as a
coefficient of variation for the total measurement system (including sample
collection, handling, and preparation) is approximately 16 percent.  The
accuracy (i.e., mean recovery efficiency) of the measurement method is
estimated to be 70 percent.

-------
                                 SECTION 3

                          EXPERIMENTAL PROCEDURES
     The intent of the field monitoring program for PERC was to establish a
reliable data base reflecting the ambient concentrations of this organic
solvent to which people are routinely exposed in urban environments.  Since
dry cleaning establishments account for most emissions (1), traces of PERC
would be expected in the atmosphere of almost any populated area.  Other
emission sources include industrial degreasing operations and, of course,
RERC manufacturing facilities.

     Program resources permitted the conduct of a short-term monitoring
effort within each of three geographic localities.  Assuming that the
intensity of dry cleaning activity is proportional to population density,
New York City (which has the highest population density in the nation (2))
would be expected to exhibit relatively high ambient PERC concentrations.
Indeed, preliminary studies (1,3) confirm this expectation.  Metropolitan New
York, therefore, was selected for inclusion in the field monitoring program.
The greater Houston area was chosen primarily because the Diamond Shamrock
plant (located in suburban Deer Park) is one of the largest PERC producers
in the nation.  Finally, metropolitan Detroit was included on the basis of
the number of metal degreasing operations located in the area.

     Meetings were held with personnel from the appropriate EPA Regional
Office and state and local agencies in each of these cities for the purpose
of designing a monitoring study.  A network consisting of 10 PERC sampling
sites and a single meteorological monitoring station was established within
each metropolitan area.  Field sampling was conducted by EMSL's Monitoring
and Analytical Chemistry Branch (MACB) for 10 consecutive days within each
of the study areas.  Exposed samples were delivered to RTI whose personnel,
operating under Contract No. 68-02-2722, performed and reported the chemical
analyses, including internal quality control.  An external quality assurance
plan was developed by EMSL's Quality Assurance Branch (QAB) and maintained
throughout the study period.  Data assessment and project leadership were
provided by EMSL's Statistical and Technical Analysis Branch (STAB).


SELECTION OF SAMPLING SITES

New York

     The ten sampling locations selected in the New York metropolitan area
are listed in Table 2 and displayed on an area map in Figure 1.  As indicated

-------
              Table 2. SAMPLING SITE LOCATIONS IN NEW YORK, NEW YORK
            Site
Address (Borough)
Class
Type   Elevation (M)
 1.  Battery Park Fireboat Sta.

 2.  Police Dept.

 3.  Queensboro Bridge

 4.  Central Park Arsenal

 5.  City College of NY

 6.  Greenpoint Treatment Plant

 7.  Bowery Bay Treatment Plant

 8.  Brooklyn Public Library

 9.  Boro Hall

10.  Coney Is. Treatment Plant



 Borough Code
 M:  Manhattan
 Q:  Queens
 B: Brooklyn
1 West Side Elevated (M)
Pitt & Broome St. (M)
59th & 2nd Ave. (M)
64th & 5th Ave. (M)
W. 140th & Covenant (M)
Greenpoint & Humbolt (B)
Berrian Blvd & 41st (Q)
Flatbush & Grand Army (B)
Queens Blvd & 82 (Q)
Knapp St. & Ave. Z (B)
Class Code
S: Existing State Site
C: Existing City Site
N: New Site
N B/C
N R/C
N R/C
S B/C
S R/C
S I/C
C I/R
S R/C
C R/C
N B/R
Type Code
I: Industrial
C: Commercial
R: Residential
B: Background
6
8
8
14
23
6
20
18
15
9


-------
"..v . -.
•X'..  •..  •/ •.
UPPER •  -  •'/"'.
NEW YORK ."-
BAY . -.  •
  NOTE:  Meteorology monitored at Site 6

     Figure 1. Sampling site locations in New York, New York.

-------
in the table, a combination of existing sites operated by the state and
local agencies and new sites established expressly for this study was
employed.  Five sites were selected on the island of Manhattan and an
additional five were distributed throughout the boroughs of Queens and
Brooklyn to the east.  The sites were selected to provide a variety of
expected local emission and meteorological conditions.  In the table, each
site is assigned a primary and secondary classification according to the
characteristics of the surrounding area (industrial, commercial, residential,
or background).  Because of the high-rise nature of New York's urban struc-
ture, all monitoring sites were located on building rooftops at the approx-
imate elevations shown in the table.  In an effort to obtain representative
samples, care was taken to select sites not dominated by taller buildings
nearby and to position each sampler away from any potentially interfering
structure on the rooftop.

     Sites 2 and 3 are located in a predominantly residential area char-
acterized by clusters of high-rise apartment buildings and would, therefore,
be expected to possess a high density of relatively small sources of PERC
emissions in the form of neighborhood (i.e., commercial) dry cleaning
establishments (William Seitz, Neighborhood Cleaners Assn., N.Y.C., personal
communication).  The other Manhattan sites represent a mix of residential,
commercial, and background conditions.

     Site 6 and, to a lesser extent, Site 7 are located in heavily indus-
trial sections of the Borough of Queens and would be expected to reflect
emissions from metal cleaning operations and industrial-scale dry cleaning
plants (although much fewer in number, one industrial dry cleaner emits many
times the quantity of PERC emitted by a typical commercial establishment,
so that the former may be considered a point source while the latter falls
into the category of an area source (4)).  An industrial-scale cleaner,
the Klink plant, is known (Hugh Tipping, NYC Dept. of Air Resources, personal
communication) to be located just a few blocks to the southwest of Site 6.
Sites 8 and 9 are in residential and commercial neighborhoods.

     Site 6 was chosen to monitor the meteorological conditions prevailing
during the sampling period because it is situated roughly in the center of
the geographic area of interest and offers a relatively unobstructed expo-
sure.  Because southwesterly winds were considered most likely, Sites 1 and
TO were expected to provide measures of the background concentrations of
PERC in ambient air.

Houston

     The monitoring network established in the greater Houston area is
described in Table 3 and Figure 2.  Once again, a combination of agency-
operated and new sites was required to provide the desired regional coverage.
Samplers were placed on the rooftops of one- or two-story buildings, on
trailers, or at ground level.

     Sites 1 through 6 are situated in predominantly residential and/or
commercial neighborhoods within the Houston city limits.  However, an
industrial dry cleaning plant, Mechanics' Uniform Supply Co., was known

-------
                   Table 3. SAMPLING SITE LOCATIONS IN HOUSTON, TEXAS
       Site
   Address (City)
Class     Type     Elevation (M)
 1.  Region VI Lab

 2.  Fire Station

 3.  Fire Station

 4.  Water Treatment Plant

 5.  Fire Station

 6.  Port Houston Terminal

 7.  Universal Steel

 8.  Pasadena Health Dept.

 9.  Deer Park City Hall

10.  State Trailer
6608 Horn wood St. (H)
Aberdeen & Stella Link (H)
Alabama & Cummins (H)
San Jacinto & Rothwell (H)
Kress & Lyons (H)
Clinton & Mississippi (H)
Sheldon & DeZavalla (C)
Shaw & Charles (P)
Center & Helgra (D)
4510 Aldine Mail Rd.
R
C
C
C
C
C
N
C
C
s
C/R
R/C
R/C
C/l
R/C
I/R
I/R
R/C
I/R
B/R
4
6
6
Ground
6
Ground
Ground
9
5
4
City Code

H:  Houston
P:  Pasadena
D:  Deer Park
C:  Channel View
Class Code

S:  Existing State Site
C:  Existing City Site
R:  Existing Regional Site
N:  New Site
 Type Code

 I:  Industrial
 C: Commercial
 R: Residential
 B: Background

-------
00
                 c.     PLACE
                 Slte 1J I         HOUSTON

                     BELLAIRE
I &,te 8 I PASADENA
                                                    NOTE:  Meteorology monitored at Site 8.
                                                                                                           Diamond
                                                                                                           Shamrock
                                          Figure 2. Sampling site locations in Houston, Texas.

-------
 (Vernon Jennings, City of Houston Dept. of Public Health, personal com-
munication) to be operating at a location just two blocks east of Site 3
during the sampling period in September 1978.  The plant has since moved to
another location.

     Sites 7 and 9 were used to bracket the Diamond Shamrock plant,
located along the ship canal in Deer Park.  This plant is one of the largest
PERC production facilities in the country.  Site 7 was established on the
premises of an industrial compound north of the plant site, while the neigh-
borhood immediately surrounding Site 9 to the south is residential and
commercial.

     Site 8 was chosen to serve as the meteorological monitoring station for
the study.  Since either southeasterly or northeasterly winds are expected
in the Houston area during the fall, this site also provided a further
downwind point from which to assess the impact of emissions from the Diamond
Shamrock facility.  Site 10 is located in a residential area north of the
city and was included to serve as a reference point for background measure-
ments during northerly winds.


Detroit

     Table 4 and Figure 3 depict the sampling locations selected in the
Detroit metropolitan area.  The Wayne County Health Department operates a
comprehensive air quality monitoring network in and around Detroit, and it
was possible to utilize 10 of these existing sites in the PERC monitoring
program.  All sampler placements were made on rooftops of monitoring
trai1ers.

     Sites 1 through 5 are distributed within the older,  more densely
populated and industrial  section of the city.  Estimates  of perch!oro-
ethylene usage by source (Dr. Peter Warner, Wayne County  Health Dept.,
personal communication) suggest that the higher PERC levels in ambient air
would occur in this section of the city.   The five other  sites are scattered
throughout less populated areas lying to the west and south of the downtown
section.  Site 3 was chosen for meteorological  monitoring because,  being
within a city park and openly exposed, it was likely to be representative of
the entire area under study.


COLLECTION AND HANDLING OF FIELD SAMPLES

     Field samples were collected by MACB/EMSL/EPA for 10 consecutive days
within each metropolitan  study area.   An attempt was made to schedule
sampling in the New York area to coincide with  the annual  peak in dry cleaning
volume occurring in the early fall.   Sampling was conducted in 1978 from
8/18 to 8/27 in New York, from 9/16 to 9/25 in  Houston, and from 10/27 to
11/5 in Detroit.

     In each study, 24-hour integrated PERC samples were  collected in
duplicate on charcoal  cartridges at the 10 preselected monitoring locations.

-------
              Table 4. SAMPLING SITE LOCATIONS IN DETROIT, MICHIGAN
       Site
  Address (City)
Class    Type   Elevation (M)
 1.  High School Grounds

 2.  Grade School Grounds

 3.  Highland Park

 4.  City Playground

 5.  Detroit Public Library

 6.  Stoepel Park

 7.  U. of Michigan

 8.  Newburgh Substation

 9.  City Playground

10.  Madonna College


 City Code
 Dt:   Detroit
 Db:   Dearborn
 RR:  River Rouge
 L:   Livonia
Linhurst & Strasburg (Dt)
Goethe & Lemay (Dt)
Davison & Oakland (Dt)
Stanton & Marquette (Dt)
Fort & Rademacher (Dt)
Auburn & Schoolcraft (Dt)
Hubbard & Evergreen (Db)
Cherry Hill & Lotz
Genessee & Chestnut (RR)
Levan & Martin (L)
W
W
W
W
W
W
W
W
W
W
I/C
C/R
B/l
C/R
C/R
B/R
R/C
B/R
C/R
R/l
4
4
4
4
4
4
4
4
4
4
Class Code
W:  Existing Wayne Co. Sites
  Type Code

  I:  Industrial
  C:  Commercial
  R:  Residential
  B:  Background
                                           10

-------
NOTE: Meteorology monitored at Site 3.
           Figure 3. Sampling site locations in Detroit, Michigan.

-------
Special handling'procedures were used to assure non-contamination of the
sampling cartridges before and after sampling.  The PERC cartridges used
were manufactured by SKC, Inc.,  and each cartridge contained a front and
back section of charcoal.  The cartridges supplied were all from lot #107
and contained 100 mg and 50 mg of charcoal in the front~and back sections,
respectively.  The sampling flow rate was set at 250 cm /min and measured
before, during, and after sampling.  The flow through the cartridges was
controlled by micro-needle valves, and the flow rate was measured using a
calibrated rotameter.

     A field sampler capable of collecting duplicate samples was used to
collect the 24-hour integrated samples of PERC on the charcoal cartridges.
Each sampler contained a pump, 2 micrometer needle valves, an elapsedvtime
meter, and a 7-day timer.  The pump was capable of maintaining 250 cm /min
flow through two cartridges with an excess of 20 in. Hg vacuum.  Two samplers
were required at each site to allow automatic start and end at 12 midnight.
A diagram of the sampler components appears in Figure 4.   A calibrated
rotameter was used to set the flow through the cartridge and to check the
flow rate during and after sampling.  The rotameter was calibrated both
prior to and after field use.

     After sampling, the cartridges were kept cool by freezer storage prior
to shipment to RTP.  The cartridges were shipped in a Trans Temp shipping
container capable of maintaining sub-freezing temperature during shipment.
The cartridges are supplied with tapered glass seals.  Prior to use, the
taper was broken.  After sampling, the cartridges were sealed with plastic
caps, placed in a culture tube, and sealed with a Teflon-lined cap.  Also,
the culture tubes were wrapped with aluminum foil to reduce sample loss due
to irradiation by light.

     A MRI meteorological (met) station was used to collect wind speed, wind
direction, and temperature at a representative site in each of the study
areas.  The unit was assembled and oriented with true north.  Each day the
met system was checked for proper orientation and time synchronization,
and the chart was dated and time recorded.  At the end of each phase of the
study, the data were reduced to hourly observations.

     The following instructions were followed by MACS personnel in the
conduct of each field study:

1.   Place two samplers at each preselected site in such a manner as to
     insure the collection of a representative sample.

2.   Prior to placing the cartridge on the sampler, break off the cartridge
     ends and place the charcoal tubes on the sampler which will operate the
     next day.  Note:  The charcoal tubes must be placed onto the sampler
     with the larger section of charcoal facing down.  Adjust the flow rate
     to 250 cm /min and set timer to come on at midnight.  Record sample
     flow rate, elapsed time meter reading, and vacuum pump operating vacuum
     on the Daily Check Sheet.  If the vacuum reading falls below 20 psig,
     check the system for leaks and, if necessary, replace the pump.


                                    12

-------
                                       RAIN
                                       BOX
                                            SAMPLE TUBES
                                                                  1.5 M
Figure 4. Diagram of perchloroethylene sampler.
                      13

-------
3.   Return to the site the next day.  Check the flow rate through the
     cartridges that are collecting a sample for this 24-hour period and
     record on data sheet.  Perform step 2 on the sampler to operate for
     the next 24-hour period.

4.   Check the flow rate through the cartridges that operated the previous
     24 hours and record on data sheet.  Remove the cartridges and cap the
     ends immediately with the caps provided.  Place the cartridges in the
     Teflon-capped culture tubes.  Wrap the culture tube with aluminum foil.
     Identify tube with sample tube number, site number, and date.  Place
     exposed cartridges in a freezer until return to RTP.

5.   After 2 or 3 days of sample collection, the exposed cartridges should
     be removed from the storage freezer and placed in the Trans Temp ship-
     ping container for return to RTP.

6.   The exposed cartridges packed in the shipping container are to be
     shipped Federal Express.  Because a 1-day delivery time of the field
     samples is required, no cartridges are to be shipped on Friday,
     Saturday, or Sunday.  Any cartridges (exposed or unexposed) remaining
     after the 10 days of sampling should returned by the field personnel to
     RTP using the shipping containers.  The exposed cartridges must remain
     refrigerated until receipt.  If the return trip occurs over a weekend,
     the cartridges should be stored in a suitable freezer until they can be
     delivered to RTP on the following Monday morning.  Upon arrival at RTP,
     all samples are to be labeled and immediately placed in storage at 0°C.

     A sample Daily Check Sheet with specific instructions appears in
Appendix A (Table A-l).


LABORATORY ANALYTICAL PROCEDURES

     Laboratory analysis for PERC was performed by RTI on the samples
collected in each of the three field studies.  Upon receipt, all samples
were labeled and immediately stored at 0°C.  Storage experiments have
demonstrated that PERC is stable on charcoal tubes for at least 1 month at
0°C, and all field samples were analyzed within 3 weeks of receipt.

     Each 150 mg charcoal tube was scored with a triangular file and broken
above the glass wool and retainer.  The glass wool and retainer were dis-
carded and the charcoal was poured into a H.P. mini-vial.  One ma of 25
percent CS«/methanol was added to the vial, which was then crimp sealed with
a Teflon-lined cap.  The vials were ultrasonicated five minutes and allowed
to stand for 1 hour.  One y& of standard or sample was injected into the
gas chromatograph (GC).  All injections were made using the solvent flush
technique.  Two y& of methanol were first drawn into the syringe after
washing in a two-stage cleaning procedure.  A 1-uj, space was left between
the solvent and the sample.  Reagent and charcoal tube blanks were run and
considered in preparation of the calibration curve.
                                     14

-------
     The analysis was performed on a Perkin Elmer 3920 GC equipped with a
nickel 63 ECD.  The column utilized was a 6.35 mm o.d. glass column, 2 mm
i.d. x 1.8 M packed with 0.1 percent SP-1000 on Carbopack C 60-80 mesh
supplied by Supelco, Inc.  A 5 percent methane in argon was used as the 3
carrier gas and filtered through molecular sieve at a flow rate of 37 cm /min,
The ECD was maintained at 218°C, standing current setting of 0.5, and oven
was held isothermally at 125°C.  The septums used were a low bleed type W
from Applied Science Labs.  Septums were changed every day to ensure a leak
tight system.  The glass column was sealed with 6.35 mm Graphloc ferrules
from Applied Sciences.  A 1 y& injection was used for all analyses.  When
the ECD was installed, a standing current vs pulse frequency curve was
established and showed adequate sensitivity and a non-contaminated cell.
The usual attenuation of 512 gave a 5 percent  full  scale deflection with a
0.2 ppb (490 ng/ml) standard.

     A strip chart recorder provided a visual  copy of the chromatogram for
inspection and was operated at a speed of 1  cm/min.   Peak area integration
was acquired on a Spectra-Physics Minigrator and included retention time  and
peak area counts.  All chromatograms were verified and sample concentrations
were obtained from a standard curve that was verified daily with a minimum
of three standards.

     Standards were prepared by injecting 15 y£ of pure perchloroethylene
into a 50 ma volumetric flask and bringing to volume with 25 preent CS? in
methanol.  The CS« was Baker-analyzed brand, and methanol was obtained from
Burdick and Jackson.   This dilution gave a 490-yg/m£ perchloroethylene
standard or 200 ppb equivalent ambient air sample.   This stock solution was
then diluted to 4.9 y/m£, 2.44 yg/m£, and 0.244 yg/m£.  The ppb concentra-
tions were calculated assuming 24-hour sampling at 250 cm  or 360 liters.
The linear concentration range for the ECD was between 0.49 yg/m£, or 0.49 ng
total weight, using a 1  yz injection, to ~ 10 ng total weight.   A standard
curve was prepared by injecting 1  y£ of each standard and a blank of
CS2/methanol, and plotting the area counts vs  concentration.

     A complete description of the measurement method employed  appears  in
Appendix A.
                                     15

-------
                                SECTION 4

                           RESULTS AND DISCUSSION
     The ambient air quality and meteorological data collected in the field
monitoring program are presented in this section.  In addition to the onsite
meteorological measurements made by EPA, concurrent records of the U.S.
National Weather Service (NWS) were obtained for each study area.  These
data, collected at municipal and regional airports, provide verification for
the onsite measurements and a broader picture of the prevailing weather
patterns and processes during the measurement of ambient PERC concentrations.

     Since ambient PERC concentrations are determined as 24-hour integrated
values  (midnight-to-midnight), the meteorological data are summarized on a
daily basis for intrepretative purposes.  The individual entries in the
tables describing meteorological conditions in the three metropolitan areas
are defined below.

  t  Wind Direction - The most representative wind direction (a subjective
     evaluation in 45° increments) over a period of time during which the
     wind direction was fairly constant.  This is based primarily on the NWS
     site closest to the center of the sampling area (e.g., LaGuardia in
     the New York area).  Some adjustment, however, is made for the passage
     of a front or wave across the area.  Major mesoscale differences, such
     as a sea breeze, are noted in the comments.

  0  Time - The period of the day over which the wind direction was
     constant.  01: is the hour average ending at 1 a.m., for continuous
     data, or the 00:53 observation at NWS sites.

  §  Wind Speed - The range of speeds in knots (KTS) which occurred for a
     given time period.

  •  Temperature - The minimum and maximum temperature for the day measured
     at the NWS site nearest the center of the sampling area.

  •  Mixing - A subjective evaluation (good, fair, or poor) of the dispersive
     ability of the atmosphere over the sampling area for the 24-hour period.
     Good implies clear skies during the day, good visibility, and moderate
     to strong wind speeds; poor implies a stagnation; and fair is everything
     else.
                                     16

-------
 NEW  YORK,  NEW  YORK

      The NWS continuously monitors meteorological conditions  at  each  of
 the  three  major  airports serving the New York City metropolitan  area
 (i.e.,  Newark, LaGuardia, and  Kennedy).  These  locations  form a  triangle
 which encompasses  the monitoring network established for  the  PERC  field
 study.  A  compilation of the NWS data from these sites for the PERC
 sampling period  appears as Table 5.

      A  comparison  of the meteorological data collected by EPA with the  NWS
 data summarized  in Table 5 revealed that a serious discrepancy in  observed
 wind direction existed throughout the study period.  Since the difference
 was  somewhat systematic and almost diametric, an explanation  may be that
 the  anemometer located at Site 6 was impacted by a local  eddy effect,
 resulting  in a bias in recorded wind direction.  At any rate, the  NWS
 data were  considered the more reliable measure  of the wind patterns affect-
 ing  the metropolitan New York area during the course of the study.  The
 NWS  data from  LaGuardia and the EPA data from Site 6 appear in Appendix B.

      The ambient concentrations of PERC observed in the New York area are
 summarized by  site and date in Table 6.  Measurable  quantities  (i.e.,
 -0.10 ppb) were  detected at each site and on each day for which  data are
 available.  Individual concentrations range from 0.16 to  10.61 and average
 1.33 ppb.

      The daily average concentrations (right hand column) reveal a marked
Iday-of-week pattern in which the highest levels of PERC occur during mid-
-week (i.e., Tuesday through Thursday) with lower concentrations  prevailing
 Friday  through Monday.  Minimum concentrations were observed  on  the two
 Sundays included in the sampling period.  These findings  are  consistent
 with the expected  activity pattern of perch!oroethylene emission sources.

      The average concentrations by sampling site are shown across  the
 bottom  of  the table.  Minimum ambient PERC levels were found  at  Sites 1
 and  10  which are both characterized as background locations (Table 2).
 The  highest concentrations were consistently observed at  Site 6  and, to
 a lesser extent, Site 7.  As discussed in Section 3, these sites are
 located in heavily industrialized sections of Queens, and at  least one  point
 source  of  PERC (an industrial dry cleaning plant) lies within the  immediate
 vicinity of the  Greenpoint Treatment Plant (Site 6).  The other  six sites,
 all  in  residential and/or commercial neighborhoods, exhibited less variation
Mn concentration for the 10-day study period.  The ambient PERC  data are
 displayed  on an  area plot for each day of the study in Figures 5 through 14.

      A  meaningful  evaluation of the meteorological processes of  diffusion and
 transport  on a day-to-day basis is complicated by the ubiquitous nature of
 PERC emissions and the constraint of time-averaged concentration measure-
$nents.  Generally, however, it appears that instances of  relatively high
 •ambient PERC concentrations throughout the study area are accompanied by
 light and  variable winds and less than good mixing conditions (e.g., 8/26/78).
                                     17

-------
                 Table 5. NATIONAL WEATHER SERVICE DATA FROM NEW YORK, NEW YORK, 8/18/78 - 8/27/78
00
8/18/78 - 8/27/78
Date
Aug:
18

19


20

21


22


23


24


25

26

4
27


Wind
Direction
NW
NE
NE
S to W
SW
SW
NWtoN
NtoNE
NE
SEtoS
N
NE
S to SW
SW
NW
W
W
SW
NE
NE

N
NE
StoSW
SW
NE
SEtoS
Time

01-21
22-00
01-09
tO-18
19-00
01-08
09-00
01-12
13-18
19-00
01-06
07-15
16-00
01-04
05-17
18-00
01-05
06-20
21-00
01-00

01-09
10-17
18-00
01-06
07-14
15-00
Wind Speed
(KTS)
5-10
5-10
4-8
5-10
5-10
3-6
8-15
8-12
8-14
3-7
4-7
5-7
6-10
5-7
5-8
5-7
4-7
6-11
11-14
5-15

4-8
5-8
5-10
4-7
8-10
8-14
Temp. Min/Max
(»F. °C )
71/82
22/28
68/88
20/31

71/81
22/27
65/80
18/27

65/83
18/28

67/85
19/29

71/86
22/30

61/71
16/22
61/76
16/24

66/80
19/27

Mixing

G

G


G

G


G


F


F


F

F


F


                                                                                                        Comments
                                                                                                     Sea Breeze effect seen closer
                                                                                                     to ocean
                                                                                                     Sea Breeze dose to shore
Sea Breeze near coast, more
Southerly winds
                                                                                                     Some Sea Breeze
                                                                                                     Cloudy with haze
                                                                                                     Overcast, occasional fog and
                                                                                                     drizzle

                                                                                                     Winds light and variable all
                                                                                                     day
                                                                                                     Haze most of day

-------
   Table 6. AMBIENT CONCENTRATIONS OF PERCHLOROETHYLENE IN NEW YORK,
                           NEW YORK, 8/18/78 - 8/27/78
^\^^ Site
Date ^\^
Friday
8/18/78
Saturday
8/19/78
Sunday
8/20/78
Monday '•
8/21/78
Tuesday
8/22/78
Wednesday
8/23/78
Thursday
8/24/78
Friday
8/25/78
Saturday
8/26/78
Sunday
8/27/78
Site
Averaqe
1
ND
0.72
0.29
0.40
i
1.03
1.28
1.10
-
0.60
0.50
0.45
0.71
2
ND
0.69
0.17
1.30
1.04
1.86
1.33
2.03
1.21
0.67
1.14
3
0.33
0.64
0.26
1.27
1.76
1.44
2.11
1.48
1.01
0.87
1.12
4
0.32
0.86
0.26
0.41
1.74
1.29
1.64
0.88
0.80
0.46
0.87
5
1.08
0.83
0.38
0.83
1.70
2.09
1.37
1.71
1.06
0.46
1.15
6
2.13
2.32
0.75
2.53
10.61
6.44
4.27
3.00
4.10
ND
4.02
7
0.83
0.82
0.49
0.49
1.75
2.89
4.42
0.92
1.60
0.66
1.49
8
0.47
1.22
0.37
0.90
1.71
1.77
1.72
1.98
2.48
0.58
1.32
9
0.16
0.55
0.28
0.29
0.81
1.64
2.36
0.72
1.05
0.47
0.83
10
ND
0.30
ND
0.43
1.00
1.16
1.06
0.35
0.91
0.56
0.72
Daily
Average
0.76
0.90
0.36
0.89
2.32
2.19
2.14
1.37
1.47
0.58
1.33
NOTE: All results are twenty-four-hour integrated values expressed in ppb.
      NO= No Data

      BD= Below Detectable (estimated at 0.05 ppb for computation of averages)
                                        19

-------

                                         r*>2f^5&--
  NEW YORK./MAMUATTA1.V';
        /•• .  ./ MANHATTAN y
       &-\\l   r-_-i  /•/
UPPER • '  '.••••.x*7
NEW YORK :_-'.*
BAY .
 	'   *  *'•  *rv
  NOTE: All results are twenty-four-hour integrated values expressed in ppb.
        ND - NO DATA
        BO -BELOW DETECTABLE
Figure 5. Ambient concentrations of perchloroethylene in New York^
New York, Friday, 8/18/78.

                             20

-------
     WESTS?/          /

    EW ^""./MANHATTAN
      /A"!I   I 0.86 1  /j
NOTE: All results are twenty-four-hour integrated values expressed in ppb.
       ND - NO DATA
       BD - BELOW DETECTABLE
Figure 6. Ambient concentrations of perchloroethylene in New York,
New York, Saturday, 8/19/78.
                              21

-------
 NOTE: All results are twenty-four-hour integrated values expressed in ppb.
        ND - NO DATA
        BD - BELOW DETECTABLE
Figure 7. Ambient concentrations of perchloroethylene in New Ygrk,
New York, Sunday, 8/20/78.
                              22

-------
UPPER • '
NEW YORK '-'
BAY .-. .
 ...»  •
 NOTE: All results are twenty-four-hour integrated values expressed in ppb.
       ND - NO DATA
       BD - BELOW DETECTABLE
 Figure 8. Ambient concentrations of perchloroethylene in New York,
 New York, Monday, 8/21/78.
                              23

-------
UPPER  • •
NEW YORK;.-*.'
           * * m A
BAY . •. .•
 NOTE: All results are twenty-four-hour integrated values expressed in ppb.
       NO - NO DATA
       BD - BELOW DETECTABLE                            ^
 Figure 9. Ambient concentrations of perchloroethylene in New York,
 New York, Tuesday, 8/22/78.

-------
                         m&Sk
UPPER • '  •/'
NEW YORK ;.-•/.
 NOTE: All results are twenty-four-hour integrated values expressed in ppb.
      ND - NO DATA '
      BD - BELOW DETECTABLE
Figure 10. Ambient concentrations of perchloroethylene in New York,
New York, Wednesday, 8/23/78.

                      25

-------
UPPER • *  •
NEW YORK :-'.'
BAY . -. .•
 NOTE:  All results are twenty-four-hour integrated values expressed in ppb.
        ND - NO DATA
        BD - BELOW DETECTABLE                            ^
Figure 11. Ambient concentrations of perchloroethylene in New York,
New York, Thursday, 8/24/78.

                              26

-------
 NOTE: All results are twenty-four-hour integrated values expressed in ppb.
        ND - NO DATA'
        BD - BELOW DETECTABLE
Figure 12. Ambient concentrations of perchloroethylene in New York,
New York, Friday, 8/25/78.

                              27

-------
UPPER • '
NEW YORK •-'.'
BAY . •.  •
 NOTE: All results are twenty-four-hour integrated values expressed in ppb.
       ND - NO DATA
       BD - BELOW DETECTABLE
 Figure 13. Ambient concentrations of perchloroethylene in New York,
 New York, Saturday, 8/26/78.

                              28

-------
                             «Bar*S^g^&&fla
  NOTE: All results are twenty-four-hour integrated values expressed in ppb.
        ND - NO DATA
        BD - BELOW DETECTABLE
Figure 14. Ambient concentrations of perchloroethylene in New York,
New York, Sunday, 8/27/78.

                             29

-------
HOUSTON, TEXAS

     Meteorological data from the NWS reporting stations in the Houston
area are summarized in Table 7.  In this instance, excellent agreement was
found between the onsite meteorological measurements and those obtained from
NWS, and very distinct meteorological patterns were observed during the
Houston sampling period.  Winds were predominantly from the southeast
through the first 4 days and, following a 2-day transitional period, north-
easterly winds dominated the final 4 days of sample collection.

     The ambient PERC concentrations are summarized in Table 8.  Forty-three
samples (46 prcent) were below the detection limit for the measurement
method (0.10 ppb).  The maximum concentration observed was 4.52 ppb.  An
overall average, computed by substituting one-half the detection limit for
the indeterminate samples, is estimated at 0.39 ppb.

     The average concentrations by day demonstrate a day-of-week pattern
similar to that observed in New York.  Again, higher levels tended to occur
during midweek (Tuesday through Thursday).  Although the average concentra-
tion computed for Friday, 9/22/78, appears to be relatively high, this value
is inordinately affected by a single observation (Site 9).  The minimum
daily average concentration occurred on Sunday, 9/24/78.

     The distribution of PERC concentrations by site is fairly uniform with
the exception of Sites 3, 7, and 9.  It will be remembered from the discus-
sion of siting criteria that Site 3 is in the vicinity of an industrial dry
cleaning plant while Sites 7 and 9 were selected to bracket a major PERC
production facility.  The minimum average concentration occurred at the
suburban location (Site 10), but this value was not appreciably lower than
those associated with urban locations removed from known point sources of
PERC emissions.  Absolute quantitative comparison among sites, however, is
handicapped by the number of indeterminate (i.e., below detectable)
observations.

     Figures 15 through 24 combine PERC concentration data with the
meteorological data collected at Site 8 on an area plot, for each day of the
study.

     During the first 4 sampling days  (9/16/78 through 9/18/78), winds
were consistently from the south through southwest and mixing conditions
were generally good.  Under these conditions, emissions from Diamond
Shamrock appear to have had more impact at Site 7, which is north and
slightly west of the plant, than at Site 9 (equidistant, but southwest
of the plant).  With the exception of Site 3, the other sites exhibited
less than or barely detectable PERC concentrations through this 4-day
period.  The next 2  days (9/20/78 and 9/21/78) were characterized by light
and variable winds, rain, and diminished mixing conditions.  Measurable
quantities of PERC were observed at all sampling locations on each of these
2 days.  Finally, a strong northeasterly wind system was established which
persisted through the last 4 days of sample collection (9/22/78 through
9/25/78).  During this period, relatively high PERC concentrations were
observed at Site 9 and Site 3, while PERC concentrations at the sites in

                                     30

-------
Table 7. NATIONAL WEATHER SERVICE DATA FROM HOUSTON, TEXAS, 9/16/78-9/25/78
Date
Sept:
16
17
18
19
20
21
22
23
24
25
Wind
Direction
SE
SW
SE
SE
SE
SE
calm
NE
SE-SW
NE
SE
NE
N
NE-E
NE
NE
NE
NE
Time
01-08
09-13
14-00
ALL
ALL
ALL
01-05
06-08
09-00
01-06
07-19
20-00
01-09
10-18
19-00
ALL
ALL
ALL
Windspeed
(KTS)
0-3
3-8
5-10
5-15
5-10
5-10
5-8
0-12
0-5
5-12
5-7
5-12
10-15
8-12
5-10
6-12
6-12
Temp. Mil
<«F,«
72/90
22/32
74/89
23/32
74/92
23/33
75/92
24/33
74/88
23/31
73/89
23/32
74/88
23/31
71/82
22/28
69/85
21/30
64/84
18/29
                                                                       Mixing
                                                                         G


                                                                         G
                                                                          F


                                                                          F
                                                                          F


                                                                          G
   Comments
                                                                                Very light rain on the N. side
                                                                                of town
Scattered rain showers, some
quite heavy; 2.6" accumulation
in town

Light rain on North side of town
                                                                                Fog & Haze with a trace of rain,
                                                                                morning and evening
Fog & Haze all day

-------
 Table 8. AMBIENT CONCENTRATIONS OF PERCHLOROETHYLENE IN HOUSTON,
                           TEXAS, 9/16/78 - 9/25/78
^\^Site
Date ^\.
Saturday
9/16/78
i
Sunday :
9/17/78
Monday
9/18/78
Tuesday
9/19/78
Wednesday
9/20/78
Thursday
9/21/78
Friday
9/22/78
Saturday
9/23/78
Sunday
9/24/78
Monday
9/25/78
Site
Average
1
BD
BO
0.10
0.26
0.38
0.77
0.13
0.12
BO
0.16
0.21
2
0.10
BD
BD
BD
0.24
0.64
BD
0.22
BD
0.20
0.17
3
0.71
0.11
1.25
0.42
0.60
4.52
2.46
0.52
0.36
2.05
1.30
4
NO
BD
0.14
0.14
0.57
0.43
0.10
BD
BD
BD
0.18
5
NO
BO
0.10
0.20
0.67
0.42
BD
BD
BD
BD
0.18
6
ND
NO
BD
0.39
0.73
0.29
BD
BD
BD
BD
0.21
7
0.37
1.38
2.37
3.32
151
0.21
BD
BD
BD
BD
0.94
8
BD
BD
BD
BD
0.49
0.38
BD
BD
BD
BD
0.13
9
BD
BD
BD
BD
0.12
0.27
3.19
BD
0.26
0.55
0.46
10
0.28
0.11
BD
0.27
0.19
0.22
BD
BO
BO
BD
0.13
Daily
Averaae
0.23
0.21
0.42
0.52
0.55
0.82
0.62
0.12
0.10
0.33
0.39
NOTE:  All results are twenty-four-hour integrated values expressed in ppb.
      ND= No Data
      BD= Below detectable (estimated at 0.05 ppb for computation of averages)
                                      32

-------
00
00
                          I         HOUSTON

                        BELLAIHE
                            o"To
                                       NOTE:  All results are twenty four hour integrated values expressed in ppb.
                                              ND - NO DATA
                                              BD - BELOW DETECTABLE
Diamond
Shamrock
                                 Figure 15. Ambient concentrations of perchloroethylene in Houston Texas
                                 Saturday, 9/16/78.

-------
W. UNIV.

  PLACE


BELLAIRE

    BD
              NOTE:  All results are twenty-four hour integrated values expressed in ppb.

                     NO - NO DATA
                     BD - BELOW DETECTABLE
Diamond
Shamrock
         Figure 16. Ambient concentrations of perchloroethylene in Houston, Texas,
         Sunday, 9/17/78.

-------
to
en
                          I          HOUS1 ON

                        BELLAIRE
                                      NOTE:  All results are twenty-four hour integrated values expressed in ppb.
                                             IMD - NO DATA
                                             BD - BELOW DETECTABLE
Diamond
Shamrock
                                 Figure 17. Ambient concentrations of perchloroethylene in Houston, Texas,
                                 Monday, 9/18/78.

-------
W.UNIV.

  PLACE


BELLAIRE
    BD
              NOTE: All results are twenty-four hour integrated values expressed in ppb.
                     IMD - NO DATA
                     BD - BELOW DETECTABLE
Diamond
Shamrock
         Figure 18. Ambient concentrations of perchloroethylene in Houston, Texas,
         Tuesday, 9/19/78.

-------
     PLACE
038 ' I          HOUSTON

   BELLAIRE
                 NOTE: All results are twenty four hour integrated values expressed in ppb.
                        NO-NO DATA
                        BD - BELOW DETECTABLE
Diamond
Shamrock
            Figure 19. Ambient concentrations of perchloroethylene in Houston, Texas,
            Wednesday, 9/20/78.

-------
CO
00
                                      NOTE:  All results are twenty-four hour integrated values expressed in ppb.
                                             ND - NO DATA
                                             BD - BELOW DETECTABLE
Diamond
Shamrock
                                 Figure 20. Ambient concentrations of perchloroethylene in Houston, Texas,
                                 Thursday, 9/21/78.

-------
u>
10
                                      NOTE:  All results are twenty-four hour integrated values expressed in ppb.

                                             ND - NO DATA

                                             BD - BELOW DETECTABLE
Diamond
Shamrock
                                Figure 21. Ambient concentrations of perchloroethylene in Houston, Texas,
                                Friday, 9/22/78.

-------
45
O
                         PLACE
                    012 ' |         HOUSTON

                       BELLAIRE
                                      NOTE: All results are twenty-four hour integrated values expressed in ppb.
                                            ND - NO DATA
                                            BD - BELOW DETECTABLE
Diamond
Shamrock
                                Figure 22. Ambient concentrations of perchloroethylene in Houston, Texas,
                                Saturday, 9/23/78.

-------
WMJIMIV.

  PLACE


BELLAIRE

    BD
              NOTE: All results are twenty-four hour integrated values expressed in ppb.
                     ND - NO DATA
                     BD - BELOW DETECTABLE
Diamond
Shamrock
         Figure 23. Ambient concentrations of perchloroethylene in Houston, Texas,
         Sunday, 9/24/78.

-------
NJ
                       .PLACE
                   Q1BJ|         HOUSTON
                       BELL A1 RE
                                     NOTE: All results are twenty-four hour integrated values expressed in ppb.
                                           ND - NO DATA
                                           BD - BELOW DETECTABLE
Diamond
Shamrock
                               Figure 24. Ambient concentrations of perchloroethylene in Houston, Texas,
                               Monday, 9/25/78.

-------
the northeast sector (Sites 4 through 8) were at or below the detection
limit.  At the background site (Site 10) PERC concentrations were also
below detectable throughout this 4-day period, indicating PERC concentration
in air masses entering the study area from the north is minimal.  Site 3,
located in the immediate vicinity of an industrial-scale dry cleaner, was the
only site at which measurable quantities of PERC were detected on each day of
the study regardless of meteorological conditions.


DETROIT, MICHIGAN

     A summary of the NWS data obtained from Detroit appears in Table 9.
With the exception of the first 2 days, winds were generally light and
variable with fair to poor mixing conditions throughout the sample collection
period.

     Table 10 presents the ambient concentrations of PERC found in samples
collected in the Detroit area.  Twelve of the one hundred samples fell below
the detection limit of the method (0.10 ppb).  The maximum observation was
2.16 ppb and the overall average was estimated (again, using one-half the
detection limit for indeterminate samples) at 0.46 ppb.

     Although the two highest average daily concentrations occurred on the
second Friday and Saturday in the study period, this seeming anomaly may be
explained by the very poor mixing conditions on those days (Table 9).
Otherwise, the familiar pattern of higher mid-week PERC concentrations was
maintained.  All  of the subdetectable samples were collected on a Friday,
Saturday, or Sunday, and the minimum average concentration once again
occurred on a Sunday.

     The sampling site averages appear to be grouped into two fairly distinct
classes.  Sites 1 through 4, located in downtown Detroit, averaged approxi-
mately twice the levels observed in the less densely populated and industri-
alized suburbs (Sites 5 through 10).  The minimum average concentration
occurred at Site 8, the most rural  sampling site location.

     Daily plots showing the observed PERC concentrations and the corre-
sponding distributions of wind direction and speed (monitored at Site 3)  are
shown in Figures 25 through 34.   Again, excellent agreement exists between
the meteorological  measurements made by NWS and EPA.

     The absence of well-defined point sources and sustained meteorological
systems makes transport evaluation  more difficult than was  the case in the
Houston study.  However, it does appear that PERC concentrations in air
masses moving from the southwest are at subdetectable levels prior to
encountering sources which lie within the metropolitan area (e.g., 10/27/78
and 11/5/78).  Also, relatively high ambient PERC concentrations are
apparently attained under conditions of meteorological  stagnation (e.g.,
11/3/78 and 11/4/78).
                                     43

-------
Table 9. NATIONAL WEATHER SERVICE DATA FROM DETROIT, MICHIGAN, 10/27/78 - 11/5/78
Date
Oct:
27
28
29
30
31
Now:
1
2
3
4
5
Wind
Direction
SW
SW
NW-N
N
E/Variable
E
SE/Variable
S
SW
N
NE/N

N/Variable
SE
SW
SW
W
Calm/Variable
Calm
S
Variable
SW/S
Variable
S/SW
Time
ALL
01-05
06-00
01-10
11-15
T6-00
01-10
n-oa
01-09
10-14
15-00

01-13
14-20
21-00
01-12
13-20
21-00
01-09
10-20
21-00
ALL
ALL
Windspeed
(KTS)
8-20
8-10
5-15
2-5
0-5
3-8
0-4
5-12
0-8
8-10
5-10

0-4
5-8
4-8
8-12
5-10
0-3
5-10
0-4
0-10
5-15
Temp. Mirr/Max
40/59
38/54
18/32
36/63
41/61

35/55
41/67
43/68
45/71
51/72
Mix
G
G
F
F
F

F
F
P
P
F
                                                                               Comments
                                                                           Fog and haze in the morning
                                                                            Fog and haze alt day
                                                                           Fog and haze all day
                                                                           Fog and haze all day

-------
                     Table 10. AMBIENT CONCENTRATIONS OF PERCHLOROETHYLENE IN DETROIT,
                                               MICHIGAN, 10/27/78 - 11/5/78
cn
^•vSJte
Date ^x.
Friday
10/27/78
Saturday
10/28/78
Sunday
10/29/78
Monday
10/30/78
Tuesday
10/31/78
Wednesday
11/1/78
Thursday
11/2/78
Friday
11/3/78
Saturday
11/4/78
Sunday
11/5/78
Site
Average
1
0.22
0.36
BD
0.58
2.16
0.80
0.88
1.63
1.26
0.25
0.82
2
0.19
0.20
BD
0.37
0.59
0.73
0.71
1.65
0.61
0.30
0.54
3
1.29
0.37
BD
0.45
1.10
0.85
0.83
1.31
0.96
0.25
0.75
4
0.42
0.31
BD
0.66
0.64
1.00
0.74
1.50
1.03
0.23
0.66
5
0.12
0.29
0.14
0.11
0.47
0.75
0.34
0.65
0.69
0.15
0.37
6
0.15
0.26
BD
0.42
0.33
0.47
0.22
0.74
0.49
0.17
0.33
7
0.10
0.23
0.13
0.53
0.46
0.42
0.30
0.57
0.40
BD
0.32
8
BD
BD
BD
0.18
0.20
0.23
0.18
0.39
0.33
BD
0.17
9
0.23
0.50
0.20
0.18
0.45
0.59
0.38
0.68
0.35
0.11
0.37
10
0.24
BD
BD
0.27
0.29
0.32
0.22
0.61
0.68
0.23
0.30
Daily
Average
0.30
0.26
0.08
0.38
0.67
0.62
0.48
0.97
0.68
0.18
0.46
                    NOTE:  All results are twenty-four-hour integrated values expressed in ppb.
                          ND= No Data
                          BD- Below Detectable (estimated at 0.05 ppb for computation of averages)

-------
                                                         HAMTRAMCK
     NOTE:  All results are twenty-four-hour integrated values expressed in ppb.

            ND - NO DATA
            BD - BELOW DETECTABLE
    Figure 25. Ambient concentrations of perchloroethylene in Detroit, Michigan,
-•? Friday, 10/27/78.

-------
 NOTE: All results are twenty-four-hour integrated values expressed in ppb.
        ND - NO DATA
        BD - BELOW DETECTABLE
Figure 26. Ambient concentrations of perchloroethylene in Detroit, Michigan,
Saturday, 10/28/78.

-------
00
                                   NOTE: All results are twenty-four-hour integrated values expressed in ppb.
                                          NO - NO DATA
                                          BD - BELOW DETECTABLE
                                 Figure 27. Ambient concentrations of perchloroethylene in Detroit, Michigan,
                                 Sunday, 10/29/78.

-------
(0
                                  NOTE: All resulu are twenty-four-hour integrated values expressed in ppb.
                                         ND - NO DATA
                                         BD - BELOW DETECTABLE
                                Figure 28. Ambient concentrations of perchloroethylene in Detroit, Michigan,
                                Monday, 10/30/78.

-------
01
o
                                                                                                           GftLMI bli IU1I ?02K

                                                                                                             WIND SPEED, nuili
                                   NOTE: All results are twenty-four-hour integrated values expressed in ppb.

                                          ND - NO DATA
                                          BD - BELOW DETECTABLE
                                  Figure 29. Ambient concentrations of perchloroethylene in Detroit, Michigan,
                                  Tuesday,  10/31/78.

-------
  NOTE: All results are twenty-four-hour integrated values expressed in ppb.

         ND - NO DATA
         BD - BELOW DETECTABLE
Figure 30. Ambient concentrations of perchloroethylene in Detroit, Michigan,
Wednesday, 11/1/78.

-------
(SI
K>
                                   NOTE: All results are twenty-four-hour integrated values expressed in ppb.

                                          ND - NO DATA
                                          BO - BELOW DETECTABLE
                                 Figure 31. Ambient concentrations of perchloroethylene in Detroit, Michigan,
                                 Thursday, 11/2/78.

-------
to
                                                                                                            CAlMI bli IU1I
                                                                                                             WIND SPEED, mpli
                                   NOTE: All results are twenty-four-hour integrated values expressed in ppb.
                                           ND - NO DATA
                                           BD - BELOW DETECTABLE
                                  Figure 32. Ambient concentrations of perchloroethylene in Detroit, Michigan,
                                  Friday, 11/3/78.

-------
01
                                  NOTE: All results are twenty-four-hour integrated values expressed in ppb.
                                         ND - NO DATA
                                         BO - BELOW DETECTABLE
                                 Figure 33. Ambient concentrations of perchloroethylene in Detroit, Michigan,
                                 Saturday, 11/4/78.

-------
in
en
                                   NOTE:  All results are twenty-four-hour integrated values expressed in ppb.

                                          ND - NO DATA

                                          BD - BELOW DETECTABLE
                                 Figure 34. Ambient concentrations of perchloroethylene in Detroit, Michigan,
                                 Sunday, 11/5/78.

-------
                                  SECTION 5

                              QUALITY ASSURANCE


     A quality assurance plan was initiated by QAB/EMSL/EPA combining
(1) internal quality control procedures implemented by RTI to determine the
precision of sample analysis, and (2) external quality assurance by QAB to
provide information on measurement accuracy.  Quality control checks on
sampling methodology consisted of repeated checks of sampler flow rate
during sample collection and collection of collocated samples at each site.
Internal quality control procedures for the analytical measurements performed
by RTI consisted of daily calibration of the analytical instrumentation,
using standard solutions of perchloroethylene, and reanalysis of selected
desorbed field samples.  External quality assurance consisted of inserting
with each day's field samples quality control samples of known perchloro-
ethylene concentration.  In addition, several blank field samples were
analyzed and individual analyses of front and back sections were conducted
for a portion of the charcoal tubes exposed in the field studies.


PRECISION OF THE ANALYTICAL TECHNIQUE

     To determine the precision of the analytical technique (i.e., GC/ECD),
20 samples were reanalyzed after desorption and the results are shown in
Table 11.  Since the within-pair variability (measured by the standard
deviation) appears to be an increasing function of level (measured by the
average concentration), a simple pooling of the variance estimates is not
valid for these data.  Rather, least squares procedures were used to estimate
an intercept and slope for standard deviation as a linear function of average
concentration.  The resulting intercept estimate was not significantly dif-
ferent from zero, so it may be safely assumed that the standard deviation
(i.e., measurement error) is proportional to the average concentration (i.e.,
measurement level), at least within the range of these data (0.1-6.0 ppb).
The constant of proportionality, called the "coefficient of variation" (CV =
(a/y)100), may be estimated as the slope of the linear relationship between
standard deviation and average concentration with no intercept term.  It is
also possible to compute the variance of this parameter estimate so that a
confidence interval about the true coefficient of variation may be con-
structed.  As shown in Table 11, the estimated coefficient of variation for
the analytical technique is 6.6 percent, with 95 prcent confidence that the
true value falls between 4.9 and 8.3 percent.  The precision of the desorp-
tion portion of the analytical procedures is more difficult to obtain and
must be estimated from analysis of external quality control samples.
                                     56

-------
                 Table 11. REANALYSIS OF DESORBED SAMPLES
City
New York. NY



. _" •
_^~—
_X'
f-
Houston, TX



Detroit, Ml




Sample
,„**• NY-007
NY-25
NY-118
NY-175
NY-195
NY-162
NY-137
a-'"'
QC-28
H-354
H-433
H-458
QC-38
D-3
D-23
D-45
D-109
D-131
D-166
D-204
D-220
Run 1 (ppb)
BD
1.22
1.75
0.92
0.91
0.46
3.77 ^^
,~. """"^
1.40
0.14
0.64
3.19
5.84
BD
BD
BD
0.29
0.32
0.74
0.25
0.23
Run 2 (ppb)
0.16
1.34
1.79
0.85
1.07
— "*" 0.46
3.95

1.20
0.14
0.68
3.00
5.12
BD
BD
BD
0.24
0.33
0.74
0.22
0.22
BD: Below Detectable
                          Coefficient of Variation:  CV=( O/fji  ) 100 = 6.6%
                          Confidence Interval: Prob (4.9 < (tffJ. ) 100< 8.3) = 95%
                                       57

-------
REPEATABILITY OF THE MEASUREMENT METHOD

     The repeatability of the measurement method, combining sampling
methodology and all analytical procedures, can be determined from analysis
of collocated (i.e., duplicate) samples.  Although duplicate samples
were collected at each site each day, only a portion of these samples
was analyzed.  Table 12 shows the analysis for 28 pairs of duplicate samples
from New York, Houston, and Detroit.  The coefficient of variation of the
measurement method, determined by the same procedure used to estimate analy-
tical precision, is estimated at 16.2 percent, with 95 percent confidence of
falling between 13.2 and 19.2 percent.  Since, in addition to analytical
imprecision, this estimate includes variability due to sample collection,
handling, and desorption, an estimate of the combined relative variability
which may be ascribed to these factors is obtained as:
                    CVTotal * ^Analytical = 14-8 Percent
EXTERNAL QUALITY ASSURANCE

     Quality control samples were prepared by QAB using blank sample tubes
filled with activated charcoal.  These tubes were obtained from the same
charcoal lot as were the field sample tubes.  These tubes were exposed to
known concentrations of perch!oroethylene in a manner similar to that used to
collect field samples.

     A permeation tube filled with perch!oroethylene and with a known
output was used to generate the PERQ necessary to spike the quality control
tubes.  A flow of nitrogen at 65 cm /min was passed over the permeation tube
and then to a manifold.  The tube to be spiked was connected to the manifold
and the flow was allowed to pass through the tube for a specified time.  The
tube was then removed and capped.  Tubes were checked at random to verify
that the flow rate was constant through the system when the tube was on the
manifold.

     Standard Reference Materials (SRM's) for perch!oroethylene are not
available in the concentration range of interest (0.1 ppb to 10 ppb).  It was
necessary to verify that the permeation tube used to provide external quality
control samples was generating the calculated perchloroethylene concentration.
The permeation tube was first weighed on Cahn Model 100 Electrobalance which
had been calibrated against NBS weights.  Once the weight loss had been
determined for the permeation tube, the tube and delivery systems were taken
to a GC-Mass spectrometer operated by MACB/EMSL.  Samples analyzed from the
permeation system indicated that PERC was the only compound present within
the sensitivity of the instrument.  This analysis verifies the presence of
perchloroethylene and that no impurities were present to account for any
significant portion of the weight loss determined for the permeation tube.

     To assess the accuracy of the analytical measurement method, one or two
quality control samples were analyzed along with field samples on each


                                     58

-------
                Table 12. ANALYSIS OF DUPLICATE FIELD SAMPLES
City
New York, NY









Houston, TX






Detroit, Ml








Sample Date
8/18/78
8/20/78
8/21/78
8/22/78*
8/23/78
8/23/78
8/23/78
8/24/78
8/24/78
8/25/78
9/17/78
9/19/78
9/20/78
9/21/78
9/22/78
9/23/78
9/24/78
9/25/78
10/27/78
10/28/78
10/29/78
10/30/78
10/31/78
11/1/78
11/2/78
11/3/78
11/4/78
11/5/78
Site
6
1
9
6
1
4
5
2
3
10
9
10
7
4
3
9
1
2
7
9
10
5
6
3
4
8
1
2
Sample a (ppb)
2.13
0.29
0.29
10.61
1.28
1.29
2.09
1.33
2.11
0.35
BD
0.27
1.51
0.43
2.46
BD
BD
0.20
0.10
0.50
BD
0.11
0.33
0.85
0.74
0.39
1.26
0.30
Sample b (ppb)
3.06
0.33
0.31
3.92
1.14
0.97
1.82
1.37
1.69
0.49
BD
0.17
1.85
0.41
3.19
0.12
BD
0.17
0.13
0.50
BD
0.14
0.32
0.93
0.48
0.46
1.04
0.27
* Omitted in computations
BD: Below Detectable
                           Coefficient of Variation:  CV = (  a///  ) 100 = 16.2%
                           Confidence Interval:  (13.2 < ( o/(i  ) 100  < 19.2%) = 95%
                                         59

-------
analysis day.  Table 13 shows the analysis date for these samples, the
quantity in yg thought to be loaded on the tube by QAB and the quantity found
on the tube when analyzed by RTI.  The percent recovery (%R = (amount found/
amount loaded)100) is given in the last column.  The overall percent recovery
for 49 quality control samples analyzed from 8/25/78 to 11/27/78 was 70.2
percent, with a standard deviation of the mean of 1.7 percent.


     The quality control results are shown stratified by spike level in
Table 14.  The following statistics appear for each level:  sample size (yg),
mean (x), standard deviation (S), standard deviation of the mean (Sc), bias
(B), percent recovery (%R), and coefficient of variation (CV).  With the
exception of the lowest spike level (0.80 yg), it appears that percent
recovery and coefficient of variation are essentially independent of level,
with approximate values of 70 and 8 percent respectively.  While it is not
possible at present to determine whether perchloroethylene loss occurs in the
adsorption as well as the desorption mechanism, it is believed that the
latter accounts for most of the consistent thirty percent negative bias.

     The coefficient of variation in this instance includes the variability
in spiked sample preparation, desorption, and analytical determination.
Recalling that the coefficient of variation for the latter factor alone has
been estimated at 6.6 percent  (Table 11), it appears that the variability
introduced in the preparation of quality control samples and subsequent
desorption is relatively minor.  The lowest level of spiking included in the
external quality assurance program approaches the minimum detectable limit
for the measurement method (0.50 yg/tube is equivalent to a 24-hour sample
collected at an ambient concentration of 0.20 ppb).  Both accuracy and pre-
cision of the measurement method appear to deteriorate somewhat as this lower
limit is approached.

     Further external quality assurance procedures were performed by MACB/EMSL.
First, two desorbed samples from the New York study were rerun in the EPA
laboratory using the analytical method developed by RTI.  The results (0.37
vs 0.43 ppb and 1.20 vs 1.27 ppb) suggest reasonably good interlaboratory
reproducibility for the method.  Secondly, to provide a completely independent
field check on the total measurement system (sampling and analytical),
several of the routine charcoal tube duplicates in the New York and Detroit
field studies were replaced by tubes filled with Tenax (an alternative
adsorbant).  The Tenax tubes were analyzed for PERC by MACB using a different
analytical scheme (i.e., a flash desorption followed by GC and mass spectro-
scopy (GC/MS).  The comparative results as shown in Table 15 provide further
substantiating evidence of the validity of the RTI values.  These paired
data are correlated with a coefficient of 0.82, with the average Tenax
result exceeding the average charcoal result by 21 percent.

     Finally, four desorbed field samples were supplied to PEDCo Environmental
for species confirmation.  Based on an analysis by GC/MS (Gregg Fusaro, PEDCo
Environmental, Inc., personal communication), PEDCo confirmed the presence
of perchloroethylene in each sample.
                                      60

-------
            Table 13. ANALYSIS OF EXTERNAL QUALITY CONTROL SAMPLES
 Analysis Date

 8/25/78

 8/28/78
 «-
 9/29/78

 8/30/78
 8/31/78

 9/1/78

 9/5/78

 9/6/78

 10/2/78

 10/3/78

 10/4/78

 10/5/78

 10/6/78
 10/9/78

 10/11/78
 10/12/78

 10/13/78

 11/9/78
 11/10/78

 11/13/78
11/14/78
11/15/78

11/16/78

11/17/78
11/20/78

11/21/78

11/22/78
11/27/78
 Sample

 QC-A
 QC-B
 QC-C
 QC-D
 QC-E
 QC-F
 QC-G
 QC-H
 QC-I
 QC-J
 QC-K
 QC-L
 QC-M
 QC-N
 QC-O
 QC-28
 QC-29
 QC-28
 QC-29
 QC-30
 QC-31
 QC-32
 QC-33
 QC-34
 QC-35
 QC-36
 QC-37
 QC-38
 QC-39
 QC-40
 QC-41
 QC-42
 QC-43
 QC-44
 QC-45
 QC-46
 QC-50
 QC-47
 QC-48
QC-49
QC-51
QC-52
 QC-53
QC-54
QC-55
QC-56
QC-57
QC-58
QC-59
p.g Loaded

    6.40
    3.20
    1.60
    8.00
    0.80
    3.20
    8.00
    8.00
    3.20
    3.20
    1.60
    6.40
    3.20
    3.20
    3.20
    3.20
    1.60
    4.80
    1.60
    1.60
    4.80
    0.80
    8.00
    0.80
    8.00
    1.60
    0.80
    8.00
    1.60
    0.80
    8.00
    0.80
    3.20
    0.80
    8.00
    8.00
    4.80
    3.20
    1.60
    1.60
    1.60
    4.80
    4.80
    3.20
    1.60
    3.20
    7.80
    3.00
    7,80
g Found

 4.34
 2.93
 1.32
 5.42
 0.49
 2.44
 5.86
 6.25
 2.44
 2.44
 1.46
 4.56
 2.00
 2.27
 1.67
 3.42
 1.17
 3.51
 1.17
 1.22
 3.18
 0.67
 5.25
 0.49
 5.60
 1.07
 0.63
 5.52
 1.16
 0.55
 4.85
 0.33
 1.84
 0.35
 4.94
 5.43
 3.54
 2.13
 1.25
 1.29
 1.15
 2.95
 3.12
 2.19
 1.26
 2.56
 3.57
 2.13
 4.64
% Recovery

   68
   92
   83
   68
   61
   76
   73
   78
   76
   76
   91
   71
   63
   71
   52
  107
   73
   73
   73
   76
   66
   84
   66
   61
   70
   67
   79
   69
   73
   69
   61
   41
   58
   44
   62
   68
   74
   67
   78
   81
   72
   61
   65
  68
  79
  80
                                                                                     59
                                              Overall Recovery = 70.2 ±1.7 percent
                                            61

-------
      Table 14. ANALYSIS OF QUALITY CONTROL SAMPLES BY LEVEL
Loaded: 0.80 /jg
Found: 0.49 /jg
0.67
0.49
0.63
0.55
0.33
0.35





n: 7
"X: 0.50
S: 0.13
S-: 0.05
X
B=X-ju -0.30 jug
%R=- X 100: 62.5%
CV = "x X 10° 26.0%
1.60 jug
1.32 yug
1.46
1.17
.17
.22
.07
.16
.25
1.29
1.15
1.26

11
1.23
0.10
0.03

-0.37
76.9%
8.1%
3.20 jug
2.93 jug
2.44
2.44
2.44
2.00
2.27
1.67
3.42*
1.84
2.13
2.19
2.56
11
2.26
0.35
0.11

-0.94
70.6%
15.5%
4.80 /jg
3.51 jug
3.18
3.54
2.95
3.12







5
3.26
0.26
0.11

-1.54
67.9%
8.0%
6.40 /ug 8.00 /jg
4.34 /Jg 5.42 M9
4.56 5.86
6.25
5.25
5.60
5.52
4.85
4.94
5.43



2 8
4.45 5.46
0.16 0.43
0.11 0.14

-1.95 -2.54
69.5% 68.3%
3.6% 7.9%
omitted in computations
                                62

-------
Table 15. ANALYSIS OF TEN AX FIELD SAMPLES
City
New York, NY





Detroit, Ml






Average
Sample Date
8/21/78

8/24/78



10/28/78
10/29/78
10/30/78
10/31/78
11/1/78
11/2/78
11/3/78

Site
3
7
7
8
9
10
4
8
6
7
9
5
3

Charcoal Result (ppb)
^•x
1.3
0.5
4.4
1.7
2.4
1.1
0.3
<0.1
0.4
0.5
0.6
0.3
1.3
1.23
Tenax Result (ppb)
2.0
1.5
3.2
3.3
2.9
1.8
0.4
0.1
0.5
0.5
0.7
0.3
1.1
1.52
                  63

-------
ASSESSMENT OF.STATIC CONTAMINATION AND BREAKTHROUGH

     As previously explained, the samplers employed in the field studies
were controlled by a timer to operate on a midnight-to-midnight cycle.  This
sampling schedule required that a fresh charcoal tube be positioned in the
standby mode for - 12 hours prior to the 24-hour sampling period and an.
additional 12 hours subsequent to this period.  In an effort to simulate
this static exposure, a charcoal tube with one end open was left for a 24-
hour period at each monitoring site used in the field studies.  These tubes
were labeled as field blanks and returned to the analysis laboratory with
the exposed field samples.  A portion of the field blanks were analyzed and
the results appear in Table 16.  Since all results are below the minimum
detectable limit, static contamination is not considered to occur to any
significant extent.

     The NIOSH charcoal tubes used for sample collection consist of two
sections of activated charcoal separated by a section of urethane foam.  The
front section contains 100 mg charcoal while the back section contains 50 mg.
In the analytical procedure followed, the charcoal from the two sections is
mixed prior to desorption (Appendix A, Section 1.0).  This procedure was
waived, however, for about 10 percent of the field samples for which indi-
vidual analytical determinations were made for the front and back sections
of each tube.  These results appear in Table 17.

     The purpose of this special treatment was to determine whether or not
appreciable quantities of perchloroethylene appeared in the back section of
any field sample.  Evidence of such perchloroethylene "breakthrough" would
raise the possibility of sample loss and render the results unreliable.  Of
the 24 tubes analyzed individually, only one contained a detectable level of
perchloroethylene in the back section.  This particular sample was collected
at the Greenpoint Treatment Plant in New York City on 8/22/78 and was the
highest concentration observed in the field studies (10.61 ppb).  Of note is
that the duplicate for this sample was analyzed (Table 12), and the result-
ant concentration was a much lower value (3.92 ppb).  The variability in
this pair was not in the population of duplicate results and, hence, was not
included in the statistical computations.  This maximum value, then, must be
considered questionable on the basis of the results of the quality assurance
program.


ESTIMATION OF CONCENTRATION INTERVALS

     The estimates of overall method repeatability and accuracy derived in the
quality assurance program may be used to construct a confidence interval about
the true value associated with any measured perchloroethylene concentration.
Such an interval estimate consists of a lower and upper bound which will
bracket the true concentration value for a preselected frequency (e.g., 95
percent) of trials.
                                     64

-------
Table 16. ANALYSIS OF CHARCOAL TUBE FIELD BLANKS
City
New York, NY
Houston, TX
Detroit, Ml


Sample Date
8/18/78
8/19/78
8/21/78
9/16/78
9/18/78
9/21/78
9/23/78
10/28/78
10/29/78
10/30/78
10/31/78
11/01/78
11/02778
11/04/78
Site
3
6
6
1
1
1
7
9
10
6
5
8
2
Concentration (ppb)
<0.10
<0.10
<0.10
,^0.10
<0.10
<0.10
<0.10
< 0.10
< 0.10
<0.10
< 0.10
< 0.10
<0.10
< 0.10
                     65

-------
Table 17. INDIVIDUAL ANALYSIS OF FRONT AND BACK CHARCOAL SECTIONS
City
New York, NY





Houston, TX







Detroit, Ml







Sample Date
8/20/78
8/22/78
8/23/78
8/24/78
8/25/78
8/26/78
9/17/78
9/18/78
9/20/78
9/20/78
9/21/78
9/22/78
9/23/78
9/24/78
9/25/78
10/27/78
10/28/78
10/29/78
10/30/78
10/31/78
11/1/78
11/2/78
11/3/78
11/4/78
Site
6
6
6
6
6
6
2
7
4
7
4
3
9
10
2
9
9
3

" ~~~ 6
3
4
8
1
Front (ppb)
0.75
10.40
6.44
4.27
3.00
4.10
<0.10
2.37
0.57
1.85
0.41
3.19
0.12
<0.10
0.20
0.23
0.50
<0.10 s
^r ^ • - 0.14
0.32
0.93
0.48
0.46
1.04
Back (pob)
<0.10
0.21
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
< 0.10
< 0.10
<0.10
< 0.10
<0.10
<0.10
<0.10
< 0.10
< 0.10 A'
                            66

-------
     Approximate lower and upper confidence limits may be computed as follows:


                  x[l * VzH

where X = measured PERC concentration (ppb)

   t /9 = student's-t statistic (2.048 for o= 0.05 with 28 degrees of
    a/^      freedom)
      R = method recovery expressed as a fraction (0.702)
     CV = method coefficient of variation expressed as a fraction  (0.162)

For example, to construct an interval estimate for the median  PERC con-
centration observed in New York City (1.00 ppb),
± 2.048
                                 (0.162)]
         =  [1.09 ppb,  1.76 ppb].
                                      67

-------
                                 REFERENCES
1.   Mitre Corporation, "Air Pollution Assessment of Tetrachloroethylene,"
     February 1976.

2.   U.S. Department of Commerce, "Statistical Abstract of the United
     States," 1970.

3.   Rutgers University, "Atmospheric Freons and Halogenated Compounds,"
     November 1976.

4.   U.S. Environmental Protection Agency, "Study to Support New Source
     Performance Standards for the Dry Cleaning Industry," May 1976.
                                     68

-------
                                APPENDIX A

         METHOD FOR THE DETERMINATION OF AMBIENT PERCHLOROETHYLENE


1.0  PRINCIPLE AND APPLICABILITY

1.1  Ambient PERC is adsorbed onto activated coconut shell charcoal.  The
PERC is then desorbed with 25 percent CS^/methanol  and analyzed by gas
liquid chromatography using and ECD.

1.2  The method is applicable to 24-hour sampling in the vicinity of PERC
sources and in areas where low levels are expected.  The samples are
collected and returned to the laboratory for analysis.


2.0  LOWER DETECTABLE LIMIT AND RANGE
                                                              q
2.1  The lower limit of detection is estimated to be 0.68 yg/m  (0.1 ppb)
assuming a 360-liter air sample at 250 cm /min sampling rate.  The range
of the method is 0.68 to 68 yg/m  (0.1 - 10 ppb) and can be extended to
higher values with shorter sampling periods or by dilution of desorbed
samples.


3.0  INTERFERENCES

3.1  It must be emphasized that any compound which  has the same retention
time as perchloroethylene at the conditions described in this method would be
a potential interference.  For this reason, it is important that confirmation
of perchloroethylene by GC/MS be obtained for a selected number of samples.


4.0  PRECISION AND ACCURACY

4.1  Quality control tubes were prepared by an independent laboratory (QAB-
EPA) with a permeation device in the range of 0.8 yg - 8.0 yg (equivalent to
0.33 ppb - 3.27 ppb in a 360-liter air sample).  Forty-nine tubes were analyzed
with an overall sample recovery of 70 percent and a mean standard error of
2 percent.

4.2  Reanalysis of 20 desorbed samples in the range of 0.1 ppb - 5.84 ppb
produced a coefficient of variation of 6.6 percent.

4.3  Analysis of 27 duplicate field samples in the  range of <0.1 ppb - 3.2 ppb
resulted in a coefficient of variation of 16 percent.

                                     69

-------
4.4  Analysis, of 23 front and back tube sections (i.e., 100 mg and 50 mg,
respectively) for breakthrough in the range of < ,0.1  - 6.44 .ppb showed less
than the minimum detectable level (i.e., < 0.1 ppb) in back sections of all
tubes.
4.5  Analysis of 14 charcoal tube field blanks showed less than the minimum
detectable level of the method (i.e., < 0.1 ppb).

5.0  APPARATUS
5.1  Sampling
     5.1.1     Charcoal Tubes—NIOSH standard 150«-mg tubes available from
     SKC, Incorporated, Pittsburgh, Pennsylvania 15220.
     5.1.2     Sampling Pump—A calibrated sampling pump whose flow can be
     determined accurately and will sample at least i liter per minute.
     5.1.3     A1r Flow Meter—Rotameter or other type of device for measur-
     ing air flow rate, 0 - 500 cm-/minute.
     5.1.4     Tubing—All tubing must be Teflon tubing.
     5.1.5     Elapsed Time Metei—To determine period of sampling.
     5.1.6     Timer—For automatic on/off operation of sampler.
     5.1.7     Calibration Kit—Calibrate?! wet test meter or soap bubble
     flow meter for calibration of sampling pumps.
5.2  Analysis                                                 /
     5.2.1     Gas Chromatograph—With Electron Capture Detector (ECD)
     5.2.2     A mechanical or electronic Integrator to determine peak area
     and a recorder for a visual copy of the chromatqgram.
     5.2.3     Chromatographic Column—1.8-meter glass column, 2-mm i.d.,
     packed with 0.1 percent SP-1000 on Carbopack C 80/100 mesh.  The
     packing material is available from Supelco, Inc., Bellefone, Pennsyl-
     vania 16823.
     5.2.4     Syringe—5.0 yji for GC injection and a 50 yfc for preparing
     standards.
     5.2.5     Pipets—1.0 ma Mohr, graduated in 0.1 mi.
     5.2.6     Sample Vials and Crimper—2.0 m£ vials with Teflon-lined
     caps.
     5.2.7     Ultrasonic Cleaner—Used to desorb sample.
                                     70

-------
     5.2.8     Volumetric Flasks—10.0 m£s, 200 m£s.
     5.2.9     Miscellaneous Lab Supplies—Vial rack, pipette, bulb, tri-
     angular file, beakers.
6.0  REAGENTS
6.1  Sampling
     6.1.1     Charcoal Tubes—Commercially available as discussed in
     Section 5.1.1.
6.2  Analysis
     6.2.1     Five percent methane in argon for chromatographic carrier gas
     and a filter/dryer.  Specify ECD Grade methane in argon.
6.3  Calibration
     6.3.1     Tetrachloroethylene (perchloroethylene),  ACS—Analytical
     reagent grade.
6.4  Sample Desorption
     6.4.1     Carbon Disulfide,  Baker "Analyzed"
     6.4.2     Methanol, Burdick  and Jackson,  "distilled in  glass"

7.0  PROCEDURE
7.1  Sampling
     7.1.1     Sampler Location—Ideally,  the  charcoal  tube  should be located
     at a level  high enough above ground to eliminate contact of the  incoming
     air with vegetation or physical  obstructions.
     7.1.2     Twenty-four Hour Sampling
               7-1.2.1  Immediately before sampling,  the ends of the  tube
     should be broken to provide  an opening at least  one-half the internal
     diameter of the tube.
               7.1.2.2  The smaller section of charcoal  (50  mg)  is used  as a
     back-up and should be positioned nearest  the sampling pump.
               7.1.2.3  The charcoal  tube  is connected  to the sampling pump
     with an appropriate length of Teflon  tubing.
               7.1.2.4  Air being sampled  should not  be  passed through any
     hose or tubing before entering the charcoal  tube.
                                    71

-------
               7.1.2.5  The flow, time, and/or volume must be measured as
     accurately as possible at the initiation and termination of sampling.

               7.1.2.6  The charcoal tubes should be sealed with the supplied
     plastic caps immediately after sampling.

               7.1.2.7  One tube should be handled in the same manner as the
     sample tube, except that no air is sampled through this tube.  This
     tube should be labeled as the field blank.

7.2  Storage of Samples

     7.2.1     All samples are immediately stored in a freezer or in a con-
     tainer with dry ice.

7.3  Preparation of Samples

     In preparation for analysis, each charcoal tube is scored with a file
in front of the first section of charcoal and broken open.  The glass wool
and retainer wire is removed with the use of a short piece of wire with a
hook on the end.  Both sections are poured into a 2-nut vial and 1 ms, of
CSg/methanol mixture is carefully pipetted into the vial.  The cap is then
crimped onto the vial.  If breakthrough studies are being conducted, each
section of charcoal is poured into separate vials and analyzed individually.
If any detectable amount is found in the back half, then breakthrough has
occurred and results are not reliable.

7.4  Desorption of Samples

     Samples are placed in an ultrasonic cleaner with sufficient water to
cover three-fourths of the vial.  After 5 minutes the samples are set aside
for 1 hour and occasionally agitated.

7.5  Daily Calibration of Gas Chromatograph

     A standard curve is prepared each day by first injecting 1 y£ of blank
solvent, CSp/methanol, before any samples are desorbed.  A detectable quan-
tity of perch!oroethylene may be present in the CS«» and this will be desig-
nated as the reagent blank.  This amount should never be greater than the
0.1 ppb standard which is injected next.  A 1.0 ppb and a 2.0 ppb standard
is injected and the area counts of all of these are plotted versus concentra-
tion.  A linear regression analysis of the data should produce a straight
line with the intercept at the area counts of the blank (usually above zero
area counts).  The curve prepared in this manner corrects for any perch!oro-
ethylene found in the blank solvent, if it is constant.  The correlation
coefficient of the calibration equation should never be less than 0.997.

7.6  Injection Technique

     The solvent flush technique is used to eliminate any blow back.  The
syringe is first flushed with methanol several times to wet the barrel and
plunger.  Two microliters of methanol are drawn into the syringe to increase

                                     72

-------
the accuracy and reproducibility of the injected sample volume.  The needle
is removed from the solvent, and the plunger pulled back about 0.5 yA to
separate the solvent flush from the sample with a pocket of air to be used
as a marker.  The needle is then inserted through the desorption vial septum
and immersed in the sample, or the standard solution.  A 1.0 yfcaliquot is
withdrawn and measured from end to end in the syringe barrel.  After the
needle is removed from the sample and prior to injection, the plunger is
pulled back a short distance to minimize evaporation of the sample from the
tip of the needle.  Duplicate injection of each sample and standard should
be made.

7.7  Gas Chromatograph Conditions
                                                        3
         Carrier gas, 5 percent methane in argon:  37 cm /min
         Nickel 63 ECD temperature:  218°C
         Electron capture standing current:  0.5
         Oven temperature:   125°C
         All transfer lines should be at least 170°C
         Attenuation:  512

7.8  Preparation of Standards

     Standards are prepared by injecting 15 y£ of pure perch!oroethylene
into a 50 ma volumetric flask and bringing to volume with 25 percent
CS2/methanol.  This is equivalent to a 200 ppb standard, or 490 yg/m£.   This
stock solution is then diluted to 20 ppb (49 yg/mji).  The 20-ppb standard
then is diluted to 2.0 ppb  (4.9 yg/m£) and 1.0 ppb (2.44 yg/nm).  The 1.0
ppb standard is then diluted to 0.1 ppb (0.244 yg/mfc).  The ppb concentra-
tions were calculated assuming 24-hour sampling at 250 cm /min, or 360
liters.  The dilutions are  always prepared with 25 percent CSp/methanol.

7.9  Electron Capture Standing Current Versus Pulse Frequency Curve

     This procedure demonstrates the sensitivity of the detector and will
indicate cell contamination.  This should be conducted initially and if
problems of sensitivity occur.


8.0  CALCULATIONS

8.1  Parts Per Billion Concentration of Sample

     From the calibration curve, find the area counts obtained from the
sample which corresponds to the ppb.  This ppb concentration is assuming
a 360-liter sample.  If the sample volume differs from 360 liters, then
corrections must be applied.  For example, from the calibration curve a
sample was found to contain 1.0 ppb, but the sample volume was only 300
liters.  The same yg weight of pollutant is in less air volume; therefore,
the concentration is actually higher.

                                     360£  _ ,  9
                                     300*  " '•*


                                    73

-------
               1 ppb x 1.2 = 1.2 ppb actual concentration

If the sample volume is greater than 360 liters, then the opposite is
true.

8.2  Parts Per Billion
                          v Pm 10*

where R = 0.08205 liter-atm/mole °K
      T = 2.98.16°K (25°C)
      V = volume of air
      P = 1 atm
      m = 166 g/mole molecular weight of perchloroethylene

                          0.082051, -atm/mo1e-°K x 298.16 ('
                               1 (atm) x 166 (g/mole)
                 m

          „„.   6.789 yg rtv. 6.789 x 10"3 yg
          PPb	or-*2- or	**-
                   nr              *

                6.789 x 10"3 yg „ OCAn = 2>44
9.0  QUALITY CONTROL

9.1  A multifaceted quality control program should be employed to insure
the integrity of all perchloroethylene results.  A calibration curve pre-
pared every day insures that the GC system is functioning properly.  In
order to insure that the 25 percent CS^/methanol solvent is not contami-
nated, every bottle of each reagent should be screened initially.  Daily  „
analysis of each mixture insures that it is also not contaminated.  A^'
valuable part of the quality control program is the analysis of charc01l
tube samples that have been loaded with perchloroethylene by an independent
laboratory using a permeation system.  These quality control tubes^should
be analyzed at least once a day with a quality control chart being used
to identify any outliers and indicate the need for correcti^iction.
Other quality control procedures include analyzing dupTlclte samples,
repeat injections, blanks, and front and back halves to check breakthrough.


10.0 DAILY RECORD OF FIELD OPERATIONS

     All Daily Check Sheets (Figure A-l) are to be completed and signed by
the field operator on a daily basis.  Duplicate copies of the Daily Check
Sheets must be made.  One copy will be sent with the samples to Barry
Martin, MD-76, EPA Annex, RTP, N. C. 27711.  The second copy is to be
retained by the field personnel and included in their trip report upon
their return to RTP.
                                    74

-------
10.1   City - Name the city where the study is located.
10.2   Date(s) Sampled - The data the sample was collected.
10.3   Operator - The field operator should sign his name.
10.4   Temperature - Record maximum and minimum temperature if available.
If not, record whatever temperature is available.
10.5   Relative Humidity - Record relative humidity.
10.6   Precipitation - Record if cloudy, clear, scattered showers, heavy
rain, etc.
10.7   Barometric Pressure - Record barometric pressure.
10.8   Wind - Record calm, steady, or gusty wind conditions.
10.9   Note - Add any pertinent comments.
10.10  Site No. - Each site will be assigned a number.
10.11  Type Sampler - Single, duplicate, or tandem.
10.12  Tube No. - Serial number of sample cartridge.
10.13  Time Period - Hours sampled (12-12)
10.14  Elapsed Time Meter - Record the meter reading at the beginning and
end of the sampling period.
10.15 Time Run, Min. - Subtract beginning meter reading from end meter
reading.
10.16  Flow Checks - Check the flow at the beginning of the sampler period
with a calibrated rotameter.   Using calibration curve supplied with rota-
meter, record flow on data sheet.   Repeat for middle (if any)  and ending
flow checks.  NOTE:  Flows are measured with cartridge  in place.
10.17  Vacuum - Record pump operating vacuum.
10.18  Average Flow Rate - Average the beginning,  middle, and  ending flow
rate, and enter on check sheet.
10.19  Total Volume.Sampled - Multiply the Average Flow Data by the Time Ran
                                     75

-------
     CITY
     Dates Sampled
     Operator 	
TEMPERATURE - MIN
Relative Humidity
Precipitation 	
MAX
BAROMETRIC PRESSURE,
Wind
                     Note
                                                Cloud  Conditions
SITE TYPE
NO. SAMPLER
Q/N
o/IN







•

TUBE
NO.











TIME PERIOD











ELAPSED
TIME METER
RCTTM CMD
DtulM bINU









TIME RUN
MINUTES









i

FLO
r>M
url
BEGIN









W CHECK
1^/MTM
/NUN
MID








l
S

END









VACUUM
READING











AVG. FLOW
RATE
o
CMJ/MIN









TOTAL
VOLUME
•3
CMJ







l

CT>
                              Figure A - 1. Daily check sheet for 1978 PERC study conducted by EPA,RTP.

-------
         APPENDIX B





METEOROLOGICAL DATA SUMMARY
             77

-------
                                  TABLE B-l.  METEOROLOGICAL DATA, 8/18/78
oo

NWS*
Hour
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
£2
23
24
DATA - LA
Wind
Direction
330
360
350
320
340
330
330
350
330
320
360
340
310
310
360
010
320
310
340
360
360
040
050
050
GUARDIA AIRPORT, NEW
Wind
(°) Speed (kts)
•08
07
07
06
10
08
11
10
09
10
08
10
07
08
09
08
12
08
07
05
05
08
10
08
YORK, NY
Temperature
C°F)
76
75
73
72
72
71
72
74
76
77
78
79
8T
81
81
81
81
80
77
78
76
76
73
72
EPA DATA -
Wind
Direction (°)
60
120
150
150
130
130
150
150
160
150
120
130
130
140
150
140
140
140
100
110
120
no
120
150
SITE 6 - NEW
Wind
Speed (mph)
3.5
3.5
5.0
4.0
4.0
5.5
6.0
6.0
7.0
8.0
7.5
6.5
6.0
6.0
5.0
6.0
7.5
6.0
6.5
5.0
4.5
2.5
4.0
5.0
YORK, NY
Temperature
(°F)
82
81
79
78
76
76
75
75
76
77
80
, 82
o 86
^ 88
90
90
90
90
88
87
84
82
82
81

    *These data are taken from NWS hourly observations and supplied by the National Climatic Center.

-------
                                  TABLE B-2.  METEOROLOGICAL DATA, 8/19/78
vo

NWS*
Hour
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
DATA - LA
Wind
Direction
030
070
060
040
310
030
010
010
040
330
320
330
300
060
020
140
170
180
210
200
200
190
200
210
GUARDIA AIRPORT, NEW
Wind
(°) Speed (kts)
05
05
07
06
04
05
05
06
04
07
06
08
04
05
07
10
09
11
16
14
12
11
09
08
YORK, NY
Temperature
(°F)
72
72
70
68
68
69
72
75
77
79
80
82
'84
85
88
85
82
79
77
77
76
77
76
76
EPA DATA -
Wind
Direction (°)
170
180
180
190
190
190
190
200
180
130
100
70
90
70
90
no
330
300
330
330
330
330
340
330
SITE 6 - NEW
Wind
Speed (mph)
4.0
4.0
3.5
2.0
2.0
1.0
2.0
2.0
2.5
2.5
3.0
4.0
4.5
4.0
3.5
3.0
8.0
10.0
9.0
10.0
9.5
7.0
6.0
5.5
YORK, NY
Temperature
(°F)
79
78
76
75
75
72
71
72
75
81
85
87
88
93
96
97
93
85
81
81
79
78
78
78

    *These data are taken from NWS hourly observations and supplied by the National  Climatic  Center.

-------
                                    TABLE B-3.   METEOROLOGICAL DATA, 8/20/78
00
o

NWS*
Hour
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
£3
24
DATA - LA
Wind
Direction
140
170
170
180
210
270
300
310
330
360
360
350
330
340
340
350
360
360
360
010
010
360
020
010
GUARDIA AIRPORT, NEW
Wind
(°) Speed (kts)
07
08
17
06
06
07
07
09
10
11
14
11
14
14
11
10
09
09
10
14
13
16
14
13
YORK, NY
Temperature
(°F)
76
76
75
75
75
76
76
79
81
80
79
77
78
78
79
80
8T
79
79
78
75
74
72
71
EPA DATA -
Wind
Direction (°)
340
330
330
300
350
350
340
60
100
140
150
130
130
120
130
130
140
140
140
150
140
150
150
150
SITE 6 - NEW
Wind
Speed (mph)
4.0
3.0
3.5
3.5
3.0
2.5
2.5
3.0
5.0
7.0
6.5
7.5
9.0
10.0
8.0
7.0
6.0
5.0
6.0
6.0
5.0
7.5
8.5
8.5
YORK, NY
Temperature
(°F)
78
78
78
78
78
78
78
78
80
82
85
83
81
80
82
82
85
88
87
85
83
82
79
78

     *These data are taken from NWS  hourly observations  and  supplied  by the National  Climatic Center.

-------
                                  TABLE B-4.  METEOROLOGICAL DATA, 8/21/78
00

NWS*
Hour
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
DATA - LA
Wind
Direction
020
360
020
360
010
010
030
360
040
060
060
050
060
050
050
060
040
070
050
160
200
130
280
030
GUARDIA AIRPORT, NEW
Wind
(°) Speed (kts)
12
09
13
11
14
12
12
10
13
14
12
13
10
10
11
10
05
07
07
07
07
05
05
06
YORK, NY
Temperature
(°F)
7.1
69
67
66
66
65
66
68
71
75
76
78
79
80
79
79
79
78
74 -
74
73
72
72
68
EPA DATA -
Wind
Direction (°)
170
170
170
160
160
160
150
150
150
160
160
170
150
150
150
150
110
160
160
150
220
300
300
330
SITE 6 - NEW
Wind
Speed (mph)
8.0
9.0
7.5
7.0
6.5
6.5
6.0
6.5
6.0
6.5
6.5
6.0
5.5
5.5
4.5
5.0
4.0
4.5
4.5
4.0
4.5
5.0
4.0
3.0
YORK, NY
Temperature
(°F)
75
73
71
71
70
69
68
68
70
72
75
80
82
84
86
87
87
87
87
83
79
75
75
73

    *These data are taken from NWS hourly observations and supplied  by  the  National  Climatic Center.

-------
                                     TABLE B-5.  METEOROLOGICAL DATA, 8/22/78
00
ro

NWS*
Hour
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
DATA - LA GUARDIA
Wind
Direction {°) 5
360
020
010
330
330
360
010
040
060
040
040
030
320
050
01Q
270
180
170
170
170
190
200
210
220
AIRPORT, NEW
Wind
>peed ffcts)
'05
06
05
07
05
04
05
06
07
07
06
07
07
05
OS
08
08
09
06
08
08
10
08
07
YORK, NY
Teuyierature
(°F)
68
66
66
66
65
66
68
72
74
76
78
80
80
SI
82
82
80
77
76
75
74
73
72
71
EPA DATA -
Wind
Direction (*)
190
170
140
160
160
ISO
150
1*0
160
160
210
210
90
90
120
130
90
120
300
300
300
330
45
45
SITE 6 -
Uind
Speed (•
2,0
1,5
2.5
2.5
2.0
3.5
4.0
3.5
3.0
3.5
3.5
2.5
5,0
4,5
4.0
3.5
5.0
4.0
7.5
8,0
5.5
6.0
4.5
4.5
NEW YORK, NY
Temperature
i*) PF)
72
71
70
70
69
60
69
69
71
75
81
85
86
86
W
92
91
91
86
80
78
78
77
75

       *These data are taken from NWS hourly observations and supplied by the National Climatic Center.

-------
                               TABLE B-6.  METEOROLOGICAL DATA, 8/23/78
oo
co

NWS*
Hoar
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
DATA - LA
Wind
Direction
220
220
240
280
280
300
300
340
310
320
300
320
290
320
330
330
280
270
270
270
260
240
300
250
GUARDIA AIRPORT, NEW
Wind
(°) Speed (kts)
07
05
05
07
08
08
06
06
05
10
08
10
07
05
05
07
07
08
05
09
06
07
08
03
YORK, NY
Temperature
(°F)
70
69
69
68
67
68
70
75
78
79
81
83
84
84
85
85
85
83
81
80
79
79
76
75
EPA DATA -
Wind
Direction (°)
45
45
45
,45
60
100
120
110
100
150
150
150
150
no
90
100
90
90
no
70
60
60
60
60
SITE 6 - NEW
Wind
Speed (mph)
3.5
3.5
3.0
3.0
3.0
2.0
1.5
1.5
1.5
3.5
4.5
5.0
4.5
5.5
5.0
5.0
5.5
5.0
5.0
5.0
4.0
4.5
4.0
2.5
YORK, NY
Temperature
(°F)
75
74
73
72
71
71
69
69
75
81
86
88
90
91
93
96
97
95
93
90
87
85
84
82

      *These data are taken  from  NWS  hourly observations  and supplied by the National  Climatic Center.

-------
TABLE B-7.  METEOROLOGICAL DATA, 8/24/78
NWS*
Hour
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
124
DATA - LA
Wind
Direction
290
300
280
230
280
260
000
230
230
250
290
310
250
250
230
250
280
270
260
250
050
040
040
050
6UARDIA AIRPORT, NEW
Wind
(°) Speed (kts)
04
07
07
05
03
06
00
07
09
08
10
09
08
11
09
10
TO
10
08
09
11
12
12
14
YORK, NY
Temperature
(°F)
74
74
73
72
72
72
74
78
81
82
83
84
•85
86
86
86
85
84
82
82
77
74
72
71
EPA DATA -
Wind
Direction t°)
45
Calm
45
Calm
Calm
30
45
30
30
30
45
30
30
30
30
30
30
!45
30
45
45
45
90
180
SITE 6 - 'NEW
Wiffld
Speed (mph)
3.0
Calm
2.5
Calm
Calm
2.0
2.5
2.0
3.5
5.5
6.5
5.5
7.0
7.5
7.0
7.5
6.5
5.5
5,.5
4,5
4v5
3,5
3.0
7.0
YORK, NY
Temperature
(°F)
81
79
78
77
76
76
75
74
78
81
85
88
90
92
92
92
92
91
90
88
87
86
83
77

-------
                                   TABLE B-8.  METEOROLOGICAL DATA, 8/25/78
00
en
__ _ 
-------
                                    TABLE B-9.  METEOROLOGICAL  DATA,  8/26/78
00

NWS*
Hour
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
DATA - LA
Wind
Direction
010
010
OTO
010
310
320
360
360
OTO
070
050
040
080
050
070
060
060
150
180
180
190
210
220
230
GUARDIA AIRPORT, NEW
Wind
(°) Speed Cfe*s)
05
06
06
04
06
07
07
08
07
05
07
09
Q&
08
07
08
05
07
09
10
11
09
07
10
YORK, NY
Temperature
(f)
§1
§T
62
62
62
62
62
63
65
68
69
n
72
73
76
76
75
73
73
70
7Q
69
69
69
EPA DATA -
Wind
Direction (*)
170
190
180
180
170
Calm
70
1(50
T60
T50
170
T70
T70
180
150
180
T50
200
300
300
300
310
330
330
SITE 6 - NEW
Wind
Speed! (mfjh)
3.0
2.0
2.0
2.5
2.5
Calm
2.5
3.0
3.5
4.0
3.5
3.5
4.0
4.5
5.5
3.5
4.0
3.0
6.5
7.0
6.5
6.0
5.5
4.5
YORK, NY
Temperature
(°F)
61
62
62
62
62
62
62
65
67
69
73
77
79
80
81
86
88
86
78
75 '
73
71
71
70

      *These data are taken from NWS hourly observations and supplied  by the National  Climatic  Center.

-------
                                   TABLE B-10.  METEOROLOGICAL DATA, 8/27/78
00

NWS*
Hour
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
DATA - LA
Wind
Direction
230
230
240
240
250
250
330
040
060
060
040
030
060
060
170
140
150
180
170
170
170
170
170
180
GUARDIA AIRPORT, NEW
Wind
(°) Speed (kts)
09
05
07
05
04
04
06
07
08
10
09
08
09
09
09
09
13
13
09
07
11
12
12
10
YORK, NY
Temperature
(°F)
68
68
67
66
67
66
67
70
73
75
77
78
80
80
78
78
75
76
74
75
74
74
74
74
EPA DATA -
Wind
Direction (°)
45
30
30
45
60
60
45
90
160
180
210
210
180
150
300
300
300
300
310
300
300
300
330
330
SITE 6 - NEW
Wind
Speed (mph)
4.5
4.5
5.0
4.0
5.0
4.5
2.0
3.0
3.0
2.5
2.5
3.5
3.0
2.5
7.5
8.0
8.0
9.0
8.0
7.0
7.5
8.0
6.5
5.5
YORK, NY
Temperature
(°F)
70
70
70
69
69
68
68
68
70
73
79
84
87
90
87
81
80
78
76
75
75
75
76
76

     *These data are taken from NSW hourly observations and  supplied  by  the  National  Climatic Center.

-------
TABLE B-ll.  EPA METEOROLOGICAL DATA,  SITE 8,  HOUSTON,  TEXAS

Date

3
O
31
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
>21
22
23
24
Q/lfi/7R
o
c
o
•r*
•P
U
C ^
•r- T-
3O
160
150
160
150
160
180
190
180
160
180
210
210
210
220
180
160
130
120
160
160
160
180
160
160
Q.
•o
•O 0)
C 
Qx— .
EU_
(U 0
1— ^^
78
78
78
78
78
78
77
78
79
81
85
91
95
97
97
93
88
91
90
87
85
82
81
80
Q/17/7R
o
.c
C Q.
£ 3.
•p *""*
o -o
•a 
C J- CO)
•P- •!- -I- Q.
30 3 co
160 4.0
160 3.5
160 3.0
160 3.0
160 3.5
150 2.5
150 2.0
150 2.0
160 2.5
170 5.5
180 6.5
160 6.5
180 8.0
180 8.5
160 10.0
160 10.0
160 10.0
180 9.0
180 9.0
170 7.5
160 6.5
150 5.0
150 5.0
150 4.5

-------
                         TABLE B-12.   EPA METEOROLOGICAL  DATA,  SITE 8,  HOUSTON, TEXAS
oo

Date







^_
3

=C
01
02
03
04
05
06
w
07
08
09
w >J
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
9/21/78
o
_g s_
E Q. 3
O E -P

•P S-
O "O ft)
lO ft) "T3 ft) Ci-*1""*^
C $- E ft) E U-

3 «5 3 co 1— - —
-160 2.0 80
150 2.0 80
Calm Calm 80
Calm Calm 79
Calm Calm 78
60 2.0 78
70 3.0 78
80 3.0 78
80 4.0 80
80 4.0 82
80 6.0 86
80 6.0 87
60 6.0 85
190 5.0 78
180 2.5 85
100 7.0 86
70 6.5 85
120 6.5 89
120 5.5 87
120 3.5 87
80 2.5 86
80 4.0 82
70 3.0 81
60 4.0 81
9/22/78
o

E
O
•r-

O
•O ft)
E i.

3 5
60
60
50
30
30
30
30
30
30
30
40
40
40
60
70
60
90
100
150
60
45
45
45
45

_^
CL

*+^S

T3
tJ 0)
C ft)
•r- Q.
3CO
3.5
4.0
3.5
4.5
5.5
5.5
6.5
5.5
6.0
7.0
6.5
8.0
9.0
10.5
10.0
8.5
6.5
7.0
6.5
7.0
8.5
8.0
9.0
8.0
ft)
i.
3

fO

ft)

Six.
Q
• 1 *~- ^
80
80
79
78
78
78
78
75
77
78
82
87
88
90
91
90
81
92
91
87
85
82
80
78
9/23/78
0

E
O
•i—

U
"O ft)
C f-
.(— .|—
30
45
60
60
60
60
60
60
60
60
60
60
60
70
80
80
70
60
90
90
90
90
70
70
70

^^
a.
£
*- -•

•o
•a ft)
E ft)
•r- Q.
300
9.0
8.0
7.0
8.0
6.5
6.5
7.0
8.0
8.0
8.0
8.0
7.5
7.0
7.0
6.0
5.0
6.0
4.5
3.5
3.0
2.5
3.5
4.0
4.0
a)
i_
3
-P
Id f
£


E ft)
•i- Q.
3 CO
7.0
6.5
7.5
7.0
8.0
6.5
6.5
6.5
8.5
7.5
7.5
8.5
8.0
9.0
8.5
9.0
7.0
6.5
6.5
6.5
6.0
7.5
6.5
7.0
ai
^,
3
•P
id
s-
ft)
Q.*"^
E * '
ft) o
• ^^^
80
80
79
78
78
78
76
75
74
75
78
80
81
83
87
88
86
86
84
81
79
78
75
73
9/25/78
o

E
O
•i—
•P
U
•a 
-------
                       TABLE B-13.  EPA METEOROLOGICAL DATA, SITE 3, DETROIT, MICHIGAN
10
o

Date







s-
3
O
3=
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
10/27/78
o

c
o
•i—
4->
O
-a 
(0
S-
0)
^x-x
Su-
0
IH^ *^_^*
37
35
33
Void
Void
Void
Void
Void
Void
Void
Void
Void
Void
Void
Void
50
49
45
41
40
39
38
38
37
10/30/78
o
*""" ^c i.
C Q. 3
O E •*->
•r- ••«— • fO
4-> S-
U tO 0)
•O Q) -O 0) CL«-~.
C S- C TU EUJ_
•r~ Br» *r~ '^^ QJ ^C
2Q 3 CO 1 	 '
Calm Calm 35
Calm Calm 34
130 2.0 36
Calm Calm 36
Calm Calm 37
Calm Calm 37
Calm Calm 38
150 2.0 39
150 5.0 44
180 7.0 49
180 8.0 51
180 8.0 56
180 6.0 59
180 8.0 61
180 8.0 63
180 8.0 63
200 5.0 63
170 5.0 60
170 4.0 57
170 3.0 55
200 3.0 54
200 3.0 52
210 5.0 55
210 7.0 58
10/31/78
o
•^^ *~** <1)
-C SL
C Q. 3
O E +•>
••- -^ to
•!-> SL
o -o 
-------
TABLE B-14.  EPA METEOROLOGICAL DATA,  SITE 3,  DETROIT,  MICHIGAN

Date








i.
3
o
2T
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
•n/oi/78
0
«~* ^-^ 0)
-E i.
E Q. 3
0 E -P
•r- . —  E u.
•p- •!— T" Q. (U O
3 Q 3 CO h-^
Calm Calm 39
Calm Calm 38
Calm Calm 37
Calm Calm 36
Void Void Void
Void Void Void
Void Void Void
Void Void Void
Void Void Void
Void Void Void
Void Void Void
Void Void Void
Void Void Void
Void Void Void
Void Void Void
Void Void Void
120 3.0 52
120 3.0 49
180 4.0 48
200 3.0 45
210 3.0 44
230 2.0 42
210 3.0 41
230 3.0 40
11/02/78
o
— - ^-. (U
.£ S-
E OL 3
0 E -P
•r- <*-* nt
•P $-
O T3 
C S- CO) E LL-
•r- •!— «r- O. CD O
30 3 co |--—
230 3.0 39
230 3.0 39
230 3.0 39
230 4.0 39
230 4.0 39
230 3.0 38
230 3.0 38
230 4.0 38
230 4.0 41
260 4.0 46
260 4.0 53
260 5.0 58
270 6.0 63
270 7.0 68
260 7.0 68
260 7.0 68
260 5.0 67
250 3.0 63
Calm Calm 60
Calm Calm 58
Calm Calm 53
Calm Calm 52
Calm Calm 50
calm Calm 49
11/03/78
o
« 	 *~~. Q)
-C S-
C Q. 3
0 E -P
•r- ^ 10
•P S-
o -a Q)
•O 0) -U  — s~* Q)
-E S-
C Q. 3
0 E -P
•r- *-— (0
-P S-
O T3  »—•*-'
Calm Calm 51
Calm Calm 50
Calm Calm 50
Calm Calm 49
Calm Calm 49
Calm Calm 49
Calm Calm 48
220 2.0 49
220 3.0 51
230 4.0 58
240 5.0 66
260 6.0 70
260 5.0 72
220 6.0 73
220 5.0 73
220 6.0 74
220 4.0 72
210 4.0 70
Calm Calm 67
200 2.0 66
210 3.0 66
210 3.0 63
345 2.Q 56
Calm Calm 51
n/nR/78
o
V_< ^-x (U
-C S.
E Q. 3
0 E -P
•r- *•— <0
•P J-
O "O OJ
T3 
-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
     EPA 600/4-79-0/17
                                                           3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE

 Measurement of Perchloroethylene in Ambient Air
             5. REPORT DATE
              August. 1Q7Q
                                                           6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
G.  Evans, R. Baumgardner,  J.  Bumgarner, P. Finkelstein,
J.  Knoll, B. Martin  (EPA)  A.  Sykes, D.Wagoner, C.Decker(
             8. PERFORMING ORGANIZATION REPORT NO

             m)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Environmental Monitoring  and Support Laboratory
 Environmental Protection  Agency
 Research Triangle  Park, NC 27711
             10. PROGRAM ELEMENT NO.

                1LD763 (Toxics)
             11. CONTRACT/GRANT NO.
 12. SPONSORING AGENCY NAME AND ADDRESS
                                                           13. TYPE Of REPORT AND PERIOD COVERED
             14.!

              EPA 600/08
                                                                             CODE
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT
              Perchloroethylene  (i.e., tetrachloroethylene)  is  an  organic solvent
  widely used in dry cleaning and industrial metal degreasing  operations.   In March
   1978,  in response to a carcinogenic risk study by the National  Cancer Institute,
   the U.S. Environmental Protection Agency undertook a program to measure perchloro-
   ethylene concentrations  in ambient air.   This program was  initiated by the Office
   of Toxic Substances and  supported by the Office of Air Quality  Planning and Standards
   The research was conducted by the Envirpnmental Monitoring and  Support Laboratory
   with contractual assistance from the Research Triangle  Institute.

        Short-term field studies were conducted in each of  three major metropolitan
   areas which were selected on  the basis of the number, density and  size of perchloro-
   ethylene emission sources as  well as the proximity of such sources to centers of
   high population density.  Dry cleaning,  a ubiquitous activity scattered throughout
   any metropolitan area,  increases in volume proportionately with population density.
   Hence, New York City, with the greatest population density i.n the  U.S., was selected
   as a study area.  Metropolitan Houston was chosen primarily  because the Diamond
   Shamrock plant, located  in suburban Deer Park, is one of  the largest perchloro-
   ethylene producers  in the nation.  Finally, metropolitan  Detroit was included
   because of the number of metal degreasing operations located in the area.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS  C. COSATI Field/Group
 Toxic Substances
 Air Quality Monitoring
 Dry Cleaning
 Metal Degreasing
 Perch1oroethy1ene
 Tetrachloroethylene
 New York, New York
 Houston, Texas
 Detroit, Michigan
  68A
  43F
13. DISTRIBUTION STATEMENT
        Release to Public
19. SECURITY CLASS (ThisReport)
    Unclassified
21. NO. OF PAGES
  104
                                              20. SECURITY CLASS (Thispage)

                                                         ! f i*»H
                                                                        22. P ft ice
EPA Form 2220-1 (9-73)

-------