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)
------- |