United States
Environmental Protection
Agency
Office of Water
Regulations and Standards (WH-553)
Washington DC 20460
December 1982
EPA-440/4-85-015
Water
SEPA
An Exposure
and Risk Assessment
for Tetrachloroethylene
-------
DISCLAIMER
This is a contractor's final report, which has been reviewed by the Monitoring and Data Support
Division, U.S. EPA. The contents do not necessarily reflect the views and policies of the U.S.
Environmental Protection Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
-------
-101
REPORT DOCUMENTATION
PAGE
1. REPORT NO.
EPA-440/4-85-015
I 4. Title and Subtitle
An Exposure and Risk Assessment for Tetrachloroethylene
I 7. Authort,) Gilbert, D.;
Wallace. P.: Wechsler. A.; and YPP °
9. Performing Organization Name and Address
Arthur D. Little, Inc.
20 Acorn Park
Cambridge, MA 02140
>; ^
12. Sponsoring Organization Nam* and Addr***
Monitoring and Data Support Division
Office of Water Regulations and Standards
U.S. Environmental Protection Agency
Washington, D.C. 20460
3. Recipient's Accession No.
s. Report oat* Final Revision
December 1982
8. Performing Organization Rept. No.
10. Project/Task/Work Unit No.
11. ContracUC) or Qrant(G) No.
(C) C-68-01-3857
(G)
C-68-01-5949
15, Supplementary Notes
Extensive Bibliographies
13. Typ* of Report & Period Covered
Final
,
14.
18. Abstract (Limit: 200 words)
This report assesses the risk of exposure to tetrachloroethylene. This study is part
of a program to identify the sources of and evaluate exposure to 129 priority
pollutants. The analysis is based on available information from government, industry
and technical publications assembled in July of 1980.
The assessment includes an identification of releases to the environment during
production, use, or disposal of the substance. In addition, the fate of tetrachloro-
ethylene in the environment is considered; ambient levels to which various populations
of humans and aquatic life are exposed are reported. Exposure levels arp estimated
and available data on toxicity are presented and interpreted. Information concerning
all of these topics is combined in an assessment of the risks of exposure to tetra-
chloroethylene for various subpopulationp.
17. Document Analysis a. Descriptors
Exposure
Risk
Water Pollution
Air Pollution
b. tdent!fler»/Open-End«d Terms
Pollutant Pathways
Risk Assessment
Effluents
Waste Disposal
Food Contamination
Toxic Diseases
Tetrachloroethylene
Perchlorethylene
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard. 12th Floor
Chicago, IL 60604-3590 W
e. COSATI Field/Group Q6F 06T
1* Availability Statement ~ '" "
Release to Public
19. Security Class (This Report)
Unclassified
20. Security Class (This Page)
Unclassified
21. No. of Pages
137
22. Price
$14.50
OPTIONAL FORM 272 (4-77)
(Formerly NTIS-35)
Department of Commerce
-------
EPA-440/4-85-015
July 1980
(Revised December 1982)
AN EXPOSURE Ala) RISK ASSESSMENT FOR
TETRACKLOROETHYLENE
bv
Diane Gilbert
Muriel Goyer, Warren Lynan, Gary Magil, Pamela Walker
Douglas Wallace, Alfred Wechsler, and Jack Yee
Arthur D. Little, Inc.
U.S. EPA Contract 68-01-3857
68-01-5949
Charles Delos
Project Manager
U.S. Environmental Protection Agencv
Monitoring and Data Support Division (WH-553)
Office of Water Regulations and Standards
Washington, D.C. 20460
OFFICE OF WATER REGULATIONS AND STANDARDS
OFFICE OF WATER AND WASTE MANAGEMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20^60
uo..paiow
-------
FOREWORD
Effective regulatory action for toxic chemicals requires an
understanding of the human and environmental risks associated with the
manufacture, use, and disposal of the chemical. Assessment of risk
requires a scientific judgment about the probability cf harm to the
environment resulting from known or potential environmental concentra-
tions. The risk assessment process integrates health effects data
(e.g., carcinogenicity, teratogenicity) with information on exposure.
The components of exposure include an evaluation of the sources of the
chemical, exposure pathways, ambient levels, and an identification of
exposed populations including humans and aquatic life.
This assessment was performed as part of a program to determine
the environmental risks associated with current use and disposal
patterns for 65 chemicals and classes of chemicals (expanded to 129
"priority pollutants") named in the 1977 Clean Water Act. It includes
an assessment of risk for humans and aquatic life and is intended to
serve as a technical basis for developing the most appropriate and
effective strategy for mitigating these risks.
This document is a contractors' final report. It has been
extensively reviewed by the individual contractors ?nd by the EPA at
several stages of completion. Each chapter of the draft was reviewed
by members of the authoring contractor's senior technical staff (e.g.,
toxicologists, environmental scientists) who had not previously been
directly involved in the work. These individuals were selected by
management to be the technical peers of the chapter authors. The
chapters were comprehensively checked for uniformity in quality and
content by the contractor's editorial team, which also was responsible
for the production of the final report. The contractor's senior
project management subsequently reviewed the final report in its
entirety.
At EPA a senior staff member was responsible for guiding the
contractors, reviewing the manuscripts, and soliciting comments, where
appropriate, from related programs within EPA (e.g., Office of Toxic
Substances, Research and Development, Air Programs, Solid and
Hazardous Waste, etc.). A complete draft was summarized by the
assigned EPA staff member and reviewed for technical and policy
implications with the Office Director (formerly the Deputy Assistant
Administrator) of Water Regulations and Standards. Subsequent revi-
sions were included in the final report.
Michael W. Slimak, Chief
Exposure Assessment Section
Monitoring & Data Support Division (WH-553)
Office of Water Regulations and Standards
iii
-------
TABLE OF CONTENTS
Pa_ge_
LIST OF FIGURES viii
LIST OF TABLES x
I. TECHNICAL SUMMARY 1
A. Risk Considerations 1
B. Materials Balance 2
C. Environmental Fate 3
D. Exposure 5
E. Effects 6
. INTRODUCTION 9
[II. MATERIALS BALANCE 11
A. Introduction and Methodology 11
3. Overview 11
C. Production 15
D. Drycleaning and Textile Processing IS
E. Metal Degreasing 19
F. Other Uses OQ
G. Transport •)-,
H. Summary ^
References ^
v
DISTRIBUTION OF TETRACKLOROETHYLENE IN THE ENVIRONMENT -5
A. Introduction 25
3. Physical and Chemical Properties 25
C. Measured Concentrations in the Environment 31
1. Introduction 31
2. Data from Selected Surveys 31
a. Water ^
b. Air 42
c. Biota _<,£
d. Foodstuffs ££
J. Summary / ,.
D. Environmental Pathways and Fate 53
1. Overview 53
2. Behavior in Air 55
3. Behavior in Water 57
4. Behavior in Soils and Sediments 50
5. Biodegradaticn f,?
-------
TABLE OF CONTENTS (Continued)
V
Page
E. Concentration Estimates Based on Analvtic Models 54
1. Overview " 64
2. Equilibrium Partitioning 65
3. Atmospheric Dispersion of Releases from
Drycleaning Operations 71
a. Development of Emission Source Parameters 7i
b. Assessment Methodology and Values for Short-Term
Concentration Estimates 72
c. Methodology and Values for Long-Term
Concentration Estimates 75
4. EXAMS Concentration Estimates 81
a. Introduction 81
b. Results 83
F. Summary 35
References 88
EFFECTS OF TETSACHLOROETHYLENE ON HUMANS AND AQUATIC BIOTA 93
A. Human Toxic ity 93
1. Introduction 93
2. Metabolism and Bioaccumulation 93
3o Animal Studies 94
a. Carcinogenicity 94
b. Mutagenesis 95
c. Teratogenesis 96
d. Other Toxicological Effects 96
4. Human Studies 99
5. Overview 100
B. Effects on Aquatic Organisms 100
References 103
VI. EXPOSURE TO TETSACHLOROETHYLENE 107
A. Human Exposure 197
1. Introduction 107
2. Exposure Situations 107
a. Populations Exposed Through Ingestion 107
b. Populations Exposed Through Inhalation 108
c. Populations Exposed Through Dermal Absorption HO
3. Results of Exposure Calculations ' 114
B. Exposure of Aquatic Biota 114
References ]_]_6
-------
TABLE OF CONTENTS (Continued)
Page
VII. RISK CONSIDERATIONS 117
A. Risks Associated with Hunan Exposure 117
1. Introduction 117
2. Quantitative Carcinogenic Risk Estimation
a. Calculation of Human Equivalent Doses
b. Estimation of Human Risk 122
3. Other Human Risks Associated with PCE Exposure 126
3. Risks to Aquatic Systems 126
References 127
APPENDIX A. DESCRIPTION OF OCCUPATIONAL ENVIRONMENTS IN WHICH
TETRACKLOROETHYLENE IS USED 129
-------
LIST OF FIGURES
Figure
—-—'-— Page
1 Materials Balance for Tetrachloroethvlene 14
2 Location of Tetrachloroethvlene Producers with
Respect to EPA Designated River Basins 17
3 Frequency of Identification of Organics in U.S.
Surface Waters 73
4 Frequency of Concentrations of Tetrachloroethvlene
Found in 204 Samples of U.S. Surface Waters ' 34
5 Vertical Profiles of Tetrachloroethvlene (PCE),
Temperature, and Dissolved Oxygen at the Deepest
Point of Lake Zurich, Sx^itzerland 35
6 Concentration of Tetrachloroethvlene in Marine Waters
and Sediments (Liverpool Bay, Great Britain) 39
7 Maximum Concentrations of Tetrachloroethvlene
Detected in the Leachate From, or Groundx^aters
Near, Five Waste Disposal Sites 40
8 Atmospheric Concentrations of Tetrachloroethvlene
at Selected Sites in the Eastern U.S. ' 43
9 Atmospheric Concentrations of Tetrachloroethvlene
in the Los Angeles Basin 44
10 Diurnal Variations in the Ground Level Atmospheric
Concentrations of Tetrachloroethvlene (C Cl,) and
Other Halocarbons 2 4 45
11 Typical Concentrations of Halocarbons, Including
Tetrachloroethvlene (C0C1 ), and Ambient Temperature
Vs Altitude - 4 47
12 Major Pathways of Tetrachloroethvlene 54
13 Volatilization of Tetrachloroethvlene and Tetrachloro-
ethvlene—Oil Mixtures from Liquid Pools and Domestic
Refuse g-,
14 Summary of Environmental Fate of Tetrachloroethvlene 66
15 Schematic of Environmental Compartment Selected for
Estimation of Equilibrium Partitioning of Tetrachloro-
ethvlene gg
viii
-------
LIST OF FIGURES (Continued)
Figure
16 Stability Wind Rose for Niagara Falls used in Long-
Term PCE Concentration Estimates for Commercial and
Industrial Drycleaning Operations 77
17 Simulated PCE Concentration Isopleths for an
Industrial Point Source 79
18 Detail of Simulated PCE Concentration Isopleths
for an Industrial Source 80
-------
LIST OF TABLES
Table
1 Sunmary of Production and Consumption of Tetrachloro-
ethylene, 1978 12
2 Summary of Environmental Releases of Tetrachloro-
ethylene (Estimated 1978) 13
3 Domestic Tetrachloroethylene Production Capacity
and Disposal Practices, 1976 ' j_g
4 Important Physical and Chemical Properties of
Tetrachloroethylene 9g
5 Degradation of Tetrachloroethylene Under Various
Conditions on
6 Concentrations of Tetrachloroethylene in U.S.
Drinking Waters 36
7 Summary of STORET Data for Tetrachloroethylene Ambient
Concentrations in U.S. Surface Waters 37
8 Tetrachloroethylene in Wastewater Treatment Systems 41
9 Reported Concentrations of Tetrachloroethylene in Fish 48
10 Concentration of Tetrachloroethylene in Foodstuffs 50
11 Ranges in Concentration of Tetrachloroethylene in the
Environment 5I_
12 Tropospheric Half-Life of Tetrachloroethylene 55
13 Decomposition Rates of Tetrachloroethylene in Aerated
Water in the Dark and in Natural Sunlight 58
1^ Values of the Parameters Used for Level I Calculation
of Equilibrium Concentrations of Tetrachloroethvlene
Using MacKay's Fugacity Method 53
15 Level I Calculations of Equilibrium Concentrations of
Tetrachloroethylene Using MacKay's Fugacity Method 70
16 Atmospheric Emission Parameters for Three Categories
of Drycleaning Operations used in Atmospheric Dispers-'on
Model ' 7 3
-------
LIST OF TABLES (Continued)
21
rase
17 Dispersion Model Parameters and Estimated PCE Concentra-
tions in Air Near Industrial and Commercial Sources 76
13 Summary of Results of EXAMS Modeling of PCE Concentra-
tions in Aquatic Ecosystems 3^
19 Incidence of Hepatocellular Carcinoma in PCE-Treated
36C3F1 Mice 94
20 Incidence of Toxic Nephropathy in B6C3F1 Mice Given
PCE by Gavage for 78 Weeks 93
Incidence of Toxic Nephropathy in Osborne-Mendel Rats
Given PCE by Gavage for 78 Weeks ag
22 i The Toxic Effects of PCE on Aquatic Organisms JOT
23 Estimated Human Exposure to Tetrachloroethylene in
Drinking Water 1Q9
24 Estimated Human Exposure to Tetrachloroethylene in Food m
25 Estimated Exposure of Humans to Tetrachloroethylene
Via Inhalation T^O
26 Analysis of Occupational Exposure to Tetrachloroethylene 113
27 Summary of Estimated Human Exposure to Tetrachloroethyiene 115
28 Adverse Effects of Tetrachloroethylene on Mammals 113
29 Carcinogenic Response in Mice Exposed to Tetrachloro-
ethylene ,0,
30 Estimated Excess Lifetime Cancers per Million Population
Exposed co Tetrachloroethylene at Various Exposure Levels I->A
31 Ranges of Carcinogenic Risk to Humans Due to Estimated
Exposure to Tetrachloroethylene i_i-
Material Balance of Tetrachloroethylene Used in
Occupational Environments , -,,,
I jO
Use of Solvents in Vapor Degreasin^ •,-,.->
A-2
-------
ACKNOWLEDGMENTS
The Arthur D. Little. Inc., task manager for this studv was Diane
Gilbert. Other major contributors were Muriel Cover (human" effects),
Warren Lyman and Gary Magi! (environmental fate) , Pamela Walker and
and Jack Yee (materials balance), Douglas Wallace (biotic effects and
exposure), Diane Gilbert (human exposure and risk), and John Ostlund
(risk analysis).
-------
CHAPTER I.
TECHNICAL SU^C-IARY
The Monitoring and Data Support Division, Office of Water Regula-
tions and Standards of the U.S. Environmental Protection Agencv is
conducting risk assessments for pollutants which may enter and tra-
verse the environment thereby leading to exposure to humans and other
biota. This report is the assessment for tetrachloroethylene
(perchloroethylene or ?CE) using available data and quantitative
models where possible to evaluate overall risk. The results of this
work are intended to serve as a basis for developing suitable
strategies for reducing the risk, if such action is indicated.
A. RISK CONSIDERATIONS
1. Risk Considerations for Humans
Exposure levels to individuals have been estimated for different
exposure conditions. Bose/response extrapolations, based on four
models, have been applied to these exposure levels using data from
positive carcinogenic results in one study in B6C3F1 mice to estimate
risk levels. Risk estimates of excess individual lifetime tumor in-
cidence associated with PCE intakes due to nonoccupational inhalation
range from negligible to 8 x 10 , corresponding to subpopulations
exposed to background PCE concentrations in air and high ambient
concentrations near drycleaning facilities, respectively. Estimated
excess individual lifetime cancer risk due to continuous lifetime con-
sumption of drinking water contaminated at the average cbservsd PCE
levels is in the negligible to 6 x 10~; range. At the highest PCE
concentrations observed in drinking water, estimated excess individual
lifetime cancer risk is on the order of 6 x 10~J.
Considerable controversy exists regarding the most appropriate
method for extrapolating human equivalent doses frcra aninal data. Due
to this uncertainty, the range of risk estimated by the various
extrapolation models may under- or overestimate the actual risk to
man. Cverestimation appears more likely due to the conservative
assumptions utilized in the calculation of human equivalent doses. In
addition to the problems inherent to risk extrapolation, the resales
or carcinogenic studies with rats were negative, although poor
survival was observed. Thus, additional uncertainty is^added to these
risk estimates.
Other than carcinogenic risks, the risk associated with chronic
exposure to FCE cannot be quantified. The effects of chronic exposure
to PCE in humans have not been well characterized, making assessnenr
of long-term, low-level exposure to PCE difficult. Tests with labora-
tory animals have established lowest observed-effect levels of 336
mg/kg body weight over a 2-year period. These level? are orders of
magnitude above estimated levels of human environmental exposure. No
indications of teratogenic effects of PCE have been resorted. Acuta
-------
human exposure to PCE is of concern at high exposure concentrations.
In general,3little or no effects have been observed at a concentration
of 700 mg/n PCE. Mininai effects (lightheadness, impaired coordination)
oecome evident at 1300 mg/m with more definite indications of CNS denrs=
sion (cental conrusion, lassitude) observed as the concentration ir—e = se
.Sidney impairment and liver damage have been reported in humar.s follow—*
acciaental exposure to PCE but are not well documented.
2. Risk Considerations for Aquatic Biota
Based upon the limited data available, aquatic biota do not
appear to be generally at risk due to exposure to tetrachloroethvlene.
The lowest concentration at which an effect was observed on a
freshwater species was 840 ug/1, a chronic value for the fathead
minnow. This concentration is alnost two orders of magnitude lar<^r
tnan typical observed ambient concentrations; the highest level observed
was 147 -jg/1. tffluent concentrations, however, as high as 5500 ug/1
nave been reported. In addition, two fish kills indicate the potential
for aquatic risks in the vicinity of a discharge of effluent contain-
ing tetrachloroethylene.
B. MATERIALS BALANCE
Production and Consumetion
Tetrachloroethylene is a synthetic organic chemical mostly pro-
duced and used domestically. In 1978 (the most recent -rear for
complete industry data), 329,000 HT were produced at ll" plants, nos«-ly
S«tnnnGulf °f 'Le*±c° «gicn. After imports (17,000 MT)'and exports '
(29,000 MT), about 317,000 MT remained in the United States for the
following major industrial uses:
Textile Cleaning 217,000 68
Metal Cleaning 55,000 17
Fluorocarbon Production 38,000 12
Other 7,000 2
PCE's properties as a solvent for fats, oils, greases, and waxes
nave led to its widespread use by the drycleaning industry by textile
manufacturers, in metal cleaning operations (degreasing), and" a ver" "
small quantity (1.5 to 2%) as stain removers for home use. Fluoro-"
carbon production is the only main consumptive use of PCE in which it
is converted to other substances. Other minor uses include food pro-
cessing, aerosol specialty products, and as a solvent in various
industrial and manufacturing processes.
2. Environmental Releases
The total amount of PCE released to the environment in 1-78
estimated to be about 254,000 MT, fully 80% of available domestic
supply. Of this amount, about 60,^ was released in the form of air
-------
emissions, about 40% was disposed of on land, and less than 1% was
discharged to POTW and directly to water.
The drycleaning industry releases an estimated 121,000 MT of PCE
to the atmosphere each year, or 78% of all atmospheric releases of PCE.
This industry emits roughly one-half of the PCE it uses, mostly in the
form of evaporative losses. Other atmospheric releases occur from
metal degreasing uses and production of fluorocarbons.
Direct releases of PCE to water have only been accounted for in
the metal degreasing industry and are believed to total approximately
40 MT per year. Drycleaners are believed to release about 10 MT to
sewer systems. Other releases to POTW'5 were estimated to amount to
about 400 MT of PCE each year.
Most of the processes in which PCE is used as a solvent involve
in-hcuse recycling through evaporation and condensation. Many smaller
scale operations, such as machine shops, that use PCE for metal
degreasing also recycle the solvent when it becomes dirty, but this
work is done by independent recyclers. Recycling processes produce
much of the solid waste that contains PCE in the form cf greasv
sludges and saturated filters; in addition, solid waste is generated
by drycleaners in the form of used filter cartridges. The volume of
PCE disposed of on land was estimated to be about 103,000 MT in 1975,
10% generated by the metal degreasing industry and 90% by drycleaners
(recycling wastes are included in these figures). This amount, which
represents one-third of the available PCE supply, is disposed of with
either industrial or municipal wastes.
Transportation is not reported as a significant source of PCE
loss with the exception of an occurrence of a major spill. Some
evaporative losses will occur during loading and transfer. These
losses, however, are negligible compared with other sources.
C. ENVIRONMENTAL FATE
The principal properties that control the fate of PCE in the
environment are solubility, volatility, and photodegradation. PCE is
quickly volatilized and then eventually photodegraded so that these
are the dominant fate processes; if dispersed in water, however, it is
soluble and can be transported over distances, especially when there
is limited possibility for volatilization, as in grour.dwater.
1. Air
PCE volatilizes rapidly and degrades in the atmosphere throueb
the action of sunlight, with a half-life of about 2 days. Although
this is a fairly short half-life in comparison wirh other chemicals,
it is sufficiently long to allow dispersion of PCE from concentrated
sources, such as drycleaners. Levels of PCE monitored at the vents-cf
commercial drycleaners are typically between 6,800 and 630,000 ;g/m3.
Levels at adjacent buildings have been estimated co be about 2,500 ug/
and, with increasing distance, the PCE concentraticr. drops t? arbier>t
levels of about 1 to 14 ug/m"5 in urban areas. Dispersion cf PCE is
-------
also demonstrated by the identification of PCE (0.1 - 0.5 y^/m3) at
remote sites. °'
Local dispersion models were used to analyze atmospheric fate of
PCE both in the immediate vicinity of and at somewhat greater dis-
tances from^major sources such as drycleaners. The analysis indicated
that typical industrial dryciear.ing sources cculd be responsible for
levels of up to 76,300 yg/m at a distance of 0.37 km under stable
(.worst case) atmospheric conditions; under unstable atmospheric
conditions, the level was 21,800 yg/mJ at a distance of 0.14 km from
the source. Commercial drycleaning facilities were found to cause
lower dispersed PCE levels in the range from 2,600 yg/mJ (0.14 km) to
9,200 yg/m (0.37 km). Modeling of FCE concentrations over a larger
area and from the same sources.indicated concentrations ranging from
1,100 yg/m (50 m) to 4.6 yg/m (5 km), for the industrial source.
These numbers are comparable to levels estimated near commercial
sources and to the background levels in urban areas.
The highest concentrations of PCE in air (6,780,000 ug/mJ) were
detected inside drycleaning establishments. These facilities are not
only the largest users of PCE, but also the largest dischargers.
b. Water
Only a small amount (1%) of total PCE production is discharged
to water. In surface waters, PCE concentrations between undetected
and 147 yg/1 have been measured. Generally, however, concentrations
are below 10 ng/1, and are detectable in less than 10 percent of
the samples. The higher concentrations have been documented near
several industrial discharges, which themselves mav contain 0.1 to
5.5 mg/1
Drinking water has been found to contain up to about 5 tr.g/'l, although
less than IQ'-i of sampled surface-water-originated supplies gerieraliv* con-
tained measurable levels. Some drinking water wells have been found to be
highly contaminated—with concentrations exceeding 10,000 ug/1. In
highly permeable soil from a densely populated area on Long°Island,
12% of the tested wells have been contaminated with PCE and many have
been closed. Piping installed in 72 Massachusetts towns was found to
be^the source of PCE to drinking water at levels of up to 5 ms/'i. The
PCt^originated in a resinous liner applied to the pipes to prevent
acid leaching of asbestos.
The probable pathway for groundwater (and eventually surface
water) contamination originates with the land disposal of 40% of total
PCE wastes on land. As it is unlikely that the typical disposal sites
are protected from leachate generation and losses through percolation
to the surrounding soils, a significant portion of this~?CE could find
its way to the water table. This is probable for two reasons: (.1)
PCE is soluble and if sufficient water passes through the wastes, PCE
can be transported, and (2) large volumes of PCE are often disposed of
at once, so that if the waste is buried too deeplv for the PCE to
volatilize immediately, the liquid migrates downward. This ieachate
can enter groundwater supplies used for irrigation or drinkir.2 water
-------
or the groundwater can enter surface water supplies as a contaminated
recharge.
Wastewaters contain PCE at concentrations up to about 2,400 '-ig/1;
typically, concentrations are below 100 ug/1. Treatment of wastewater
appears to be successful at most POTVs, with better than 90% removal
usuaiiy coserved. .he removal process is predominately volatilization
with some biodegradation by acclimated microorganisms. The stoichiometr-'
of chlorination of wastewaters does not favor PCE formation from available
substances. This is borne out by existing data concerning levels in POTT.-:
influent and effluent.
The fate of PCE in water was modeled using EPA's EXAI-IS for five
aquatic systems. Two loading rates were applied—one at 0.004 kg/hr,
simulating POTW and industrial scale discharges; the other was 160
kg/hr, corresponding to a groundwater contribution that was heivilv
contaminated with disposal site leachata. Volatilization, sediment
content, and transport were found to be the controlling parameters.
In the river environments, transport dominated the other fate processes.
3. Land
Comparatively little is known about PCE's behavior in soils: how-
ever, modeling of partitioning indicates that PCE is not tightly bound
to soils and can thus be expected to be relatively mobile, either
volatilizing to the atmosphere or percolating to groundwater, where it
is resistant to hydrolysis. Further investigations are warranted, es-
pecially to study the situation of leachate migration and attenuation
from landfills containing PCE wastes.
D. EXPOSURE
1. Kumar. Populations
The critical exposure route for the general population is inhala-
tion, due to the volume and large number of emission sources spread
throughout population centers and the tendency of PCE to partition the
atmosphere.
The ranges of exposure to PCE have been estimated for several
different environmental scenarios. Total exposure by inhalation is
primarily determined by an individual's proximity to kev sources.
Typical nonoccupational inhalation exposures were estimated tc range
between O.CJl mg/day (remote area, background level) to a maximum of
137 mg/day near drycleaning facilities. The use of coin-operated
laundry facilities cne-half hour per day could result in an estimated
intake or 41 mg/day PCE. Occupational exposure may account for between
165 mg and 1150 mg per S-hour workday, assuming a 5Cf= retention.
Exposure via inggstion of PCI is generally low compared
inhalation. Most water supplies and foods that have been oamol
-------
con,ain_less tnan 3 ug/1 of PCE; typical estimated intakes are on the
order or less tnan 1.0 ug/day for water and 1.2 ug/day *o- food i
.iigr.iv contaminated water source (5.0 mg/1) , such as a distr*bu*Jon
!!:j/!;e° Wltn a resinous liner, could result in an intake of up to 10
Dermal absorption cannot be estimated at this time due to lac- o-"
aata_on absorption rates. This exposure route is not expected to be "
significant as the rrequency and duration of exposure would be low.
2. Aquatic Populations
Measured surface water concentrations are generally below 10 u^/l
'-•'^C fate aodels suS§est that typical effluent discharges (modeled bv
i.\A..bJ may cause localized PCE concentrations in rivers of 0.004 "»/j. '
and in lakes and ponds of 0.52-10 yg/1. Higher input loads, occurring
in the case of leachate contamination, could cause water concentrations
or 0.16-,.6 -g/1. These data are not inconsistent with available monitor-
ing data.
E. EFFECTS
1. Humans
•or*,
exposure has been linked to an increased incidence of heoat^c
carcinoma in 36C3F1 mice at a dose of 386 ng/kg bodv weight bv aa-age
out the efrect may reflect a secondary response to PCE-induced ° '
cnemical hepatectomy in this species. Carcinogenicitv tests with -ats
were inconclusive. Mutagenic findings are varied but'pos'-fve r-su^s
in mammalian cell transformation studies and host-med-'ated assavs
implicate PCE as a mutagen. There are nc indications of tarato^c
errects associated with PCE exposure.
Pronounced toxic nephropathy was seen in mica and rats chroni-
cally exposed to 386 mg/kg body weight and 471 ng/kg bodv wei?ht "?C-
respectively, by gavage for 78 weeks. " ""
PCE is readily absorbed through the lungs. Approximate'^ 57': ~f
innalea PCE is retained but most of this amount (80-100S) is subse-
quently exhaled, unchanged in expired air. A respiratorv balf-lif- of
63 aours has been estimated for man; urinary clearance of aporox—
mate_y 2% or retained PCE as trichloracetic acid has an approximate
nair-life or 1-a hours. These values suggest accumulation"of PCE mav
cccur with repeated exposure.
In laboratory animals, acute oral LD values ran^e f-r^ 3 9R.1
mg/Vg in the rat to 8,850 mg/kg in the mouse. Acute exposure to'?C^
is characterized by depression of the central nervous svstem, ar^ '—
liver and kidney damage. ' "'
In man, the predominant effect: cf PCE exposure by inha'a^cn
Uess than or equal to 30 mg/m ) is depression cf the" central -erv—3
system, caaractarized by vertigo, confusion, inebriation-like
-------
symptoms, trenors and numbness. Accidental exposure to ?CE has also
been linked to kidney impairment and hepatotoxic effects. The lack o:
long-term exposure data maKss assessment of long-term, low-level
exposure to ?CE difficult. However, the pronounced nephrotoxicitv in
rodents and increased incidence of hepatocellular carcinoma in mice
raise concerns for the human health aspects of prolonged exposure to
t w L. *
The OSKA standard for occupational exposure is an 8-hour time-
weighted average of 15 mg/m , with an acceptable ceiling (i.e., verv
short-term levels) of 30 mg/ni . The 10~° risk level for human cancer
resulting iron PCE exposure from drinking water is estimated as 0.2
tr.g/1 by the U.S. EPA.
2. Aquatic Biota
Acute toxicity to freshwater species occurs in the range of 4.3
to 21.4 mg/1 in flow-through studies, with rainbow trout being the
most sensitive species tested. A chronic value for fathead minnow of
0.84 mg/1 was reported.
Saltwater species tested include the mysid shrimp and barnacle
naupii, with respective LC_0 values of 10.2'rag/l (96 hr) and 3.5 mg/1
(48 hr). The sheepshead minnow and the algae tested were less
sensitive. A chronic value for mysid shrimp was fousd to be 0.450
mg/1, the lowest reported effect level.
No criteria have been established by EPA for the protection of
either freshwater or saltwater aquatic life.
-------
CHAPTER II
INTRODUCTION
The Office of Tater ?».egulations and Standards, Monitoring and Data
Support Division of the Environmental Protection Agency is conducting
a program to evaluate the exposure to and risk of 129 priority pollutants
in the nation's environment. The risks to be evaluated included poten-
tial harm to human beings and deleterious effects on fish and other
biota. The goal of the task under which this report has been prepared
is to integrate information on cultural and environmental flows of spe-
cific priority pollutants and estimate the risk based on receptor expo-
sure to these substances. The results are intended to serve as a basis
for developing suitable regulatory strategy, especially for sources to the
nation's waters, for reducing the risk, if such action is indicated.
This document is an assessment of the risks associated with tetra-
chloroethylene in the natural and human environments. It includes
summaries of comprehensive reviews of the production, use, distribution,
fate, effects and exposure to tetrachloroethylene and the integration
of this material into an analysis of risk. In this report the terms
"tetrachloroethylene" and "PCE" are used interchangeably. Tetrachlcro-
ethylene is also sometimes called "perchloroethylene" or "tetrachlorc-
ethene."
Tetrachloroethylene is a heavy, volatile, ordorless liquid, with
excellent properties as a solvent. These properties have led to PCE's
widespread use (Chapter III, Materials Balance) in cleaning, manufac-
ture and repair operations of all types and sizes. Such facilities are
practically ubiquitous in the populated areas of the U.S.. and face
and exposure analyses have attempted to cover the extremes, as well as
typical concentration levels in these situations.
Chapter IV presents the results of media specific fate models used
to predict concentration levels of PCE in the air within close proximity
of significant sources of PCE, equilibrium concentrations resulting
from free exchange of PCE between air, soil, water and sediment; and
are presented. The most current research on effects of PCE upon human
and non-human receptors, a description of the populations exposed to
PCE, and a statement of the risk of this exposure comprise Chapters
V, VI, and VII. The Appendix describes occupational environments in
which PCE is used.
Throughout the report, data have been given in metric units
(mg/1, mg/kg, ug/raj, etc.). In the case of data for atmospheric con-
centrations, which are commonly given as ppb or ppt, the conversion
factor is 1 ppb = 5.78 ;:g/m3 at 1 atm. and 25°C. Because few researchers
include this level of detail (atm and °C) in their data: the above
conversion factor has been used consistently in this work. Mo further
mention will be mace of this artifact in converting data to the metric
unit svstem.
-------
CHAPTER III.
MATERIALS BALANCE
A. INTRODUCTION AND METHODOLOGY
In this section, a materials balance is developed for tetrachloro-
ethylene (or PCE). The materials balance considers PCE as it is released
from the cultural environment to its first point of entry into the
natural environment. Potential sources of PCE releases were identified
by a review of activities in which the material participates from its
production and use in various forms to its ultimate disposal.
For each major source of pollutant release, the amount of material
released was estimated, and the environmental compartments (air, land,
and water) initially receiving and transporting the material were identi-
fied, as were the locations at which the pollutant loadings take place.
There are many uncertainties inherent in this type of analysis: not all
current releases have been identified, past releases were not well docu-
mented, and future releases are difficult to predict. Nevertheless,
sufficient information is available to indicate in general terms the
nature, magnitude, location, and time dependence of pollutant loading of
the environment with PCE.
In developing the materials balance for PCE, data and information
were obtained from a recently completed study on tetrachloroethylene
sponsored by the U.S. Environmental Protection Agency (Versar, Inc.,
1980) and from other readily available literature, supplemented as neces-
sary by contacts with industry experts.
B. OVERVIEW
Tetrachloroethylene is a synthetic organic chemical. In 1973, 329,100
metric tons (MT) of the compound were produced (U.S. International Trade
Commission, 1979), 17,200 MT were imported, and 29,000 MT were exported
(U.S. Bureau of Census, 1978). The supply available for domestic con-
sumption totalled 317,300 MT.
Approximately 58% of the supply is used for drycleaning and textile
processing. About 17% is used for metal degreasing, and about 12% is
used for the production of fluorocarbons (MRI, 1977). The remainder is
used in various miscellaneous applications. A summary of the production,
use, and environmental releases is presented in Tables 1 and 2 and
Figure 1.
Growth in production and consumption of PCE has declined in recent
years; in 1973 the annual growth rate was about 3%; restrictions
on^the use of other solvents may create expanded markets for PCE (Versar,Inc.,
1979a; U.S. EPA, 1979). This is evident in the vapor degreasing applica-
tion, in which the use of the preferred solvent, tricnloroethvlene', is being
curtailed and replaced by tetrachloroethvlene.
11
-------
TABLE 1. SUMMARY OF PRODUCTION AND CONSUMPTION
OF TETRACHLOROETHYLENE, 1978
Supply/Consumer
Production
Capacity (MT)
Supply Consumption
(MT) (MT)
Domestic Capacity (1976)
Domestic Production (estimated
1978, at 60% of Capacity)
Imports
Exports
Textile Cleaning
Metal Degreasing
Fluo racarbons
Other
Total
(548,900:)
329,1002
17,2003
^ersar, Inc., 1979a, for trade year 1976.
^U.S. International Trade Commission, 1979.
3U.S. Bureau of Census, 1978b.
''U.S. Bureau of Census, 1978a.
5MRI, 1977.
29,000^
216,9002,5
55,4002,5
38,1002,5
6,900'
12
-------
TABLE 2. SUMMARY OF ENVIRONMENTAL RELEASES
OF TETRACHLOROETHYLENE (ESTIMATED 1978)
Release (MT/Yr)
Source
Production
Textile Cleaning
Metal Degreasing
Fluorocarbons
Misc. Other Uses
Total8
Air
3,0001.^
120, 6002
19,8002
602
6,530
149,990
Direct
Aquatic
! ft 3
ft2
itf*
*2
ft
10+
POTW
ft 3
4-o
10 -
40 6
ft2
3707
4209
Land
ft3
92,900-
10,7002'b
ft 2
ft
103,600+
}JRB Associates, 1979.
2Versar, 1979a
3U.S. EPA, 1976.
^GCA Corp., 1980
5Versar 1979b.
5Versar 1981. These data are preliminary and reflect indirect, i.e.,
untreated discharges.
This number was determined by the difference between the total releases
to POTWs and the estimated releases for Textile Cleaning and Metal
Degreasing.
8Total releases = 256,770 MT.
9Arthur D. Little, Inc., based on data from Feiler, 1980.
-Insufficient data available at the present time to quantify these
releases; some are believed to be negligible, and others are unknown.
Indicates these releases may be larger than indicated but cannot be
quantified.
13
-------
SOURCES
346,300 MT
ENVIRONMENTAL
COMPARTMENTS
254,020 MT
Imports
17.200MT
Domestically Produced
329.100 MT
Dry Cleaning
216,900 MT
Water and POTW
430 MT
Land
Water
Air
Note: Boundaries between receiving medium are often undefined and/or change; releases to
water or land often result in atmospheric concentrations.
FIGURE 1 MATERIALS BALANCE FOR TETRACHLOROETHYLENE
CONSUMED
346,30
Exports
29,000 MT
Metal
Cleaning
55,400 M
Other
6,900 MT
Fluorocarbons
38,100 MT
Legend:
14
-------
About 89 percent of the PCE supply is consumed in distributed use
as a drycleaner or solvent and is discharged to the environment as a
consequence of its use. Virtually all of the releases are atmospheric
or are in the form of solid wastes. Although some aquatic discharges from
drycleaning and organic chemical manufacture have been reported, they are
insigificant compared with the atmospheric emissions and are likely to be
volatilized quickly to the atmosphere. An estimated 150,000 MT of'airborne
emissions and 100,000 MT of solid waste occur each year as the result of
the production and use of PCE.
Although much of the PCE discharged to water and sewer is volatilized
almost immediately, this occurrence is a fate process (see Chapter 4) and
these discharges are reported as releases to the water compartment in the
materials balance (Arthur D. Little, Inc. Estimate).
C. PRODUCTION
Tetrachloroethylene is manufactured domestically in eleven plants
operated by eight companies. Table 3 lists these plants and their lo-
cations, nameplate capacities, and disposal practices; the locations of
the production sites are shown in Figure 2. PCE is produced in con-
junction with other halogenated organics, and the reported plant capa-
cities are quite variable from year to year since the same process can
be tuned to produce other products. Total U.S. production in 1978 was
329,100 MT (U.S. International trade Commission, 1978).
Tetrachloroethylene is manufactured by three processes: thermal
chlorination of one to three carbon alkanes, catalytic chlorination of
ethylene dichloride, and chlorination of acetylene (Versar, Inc., 1979a)."
Because of the flexibility of the processes involved, reported applica-
tions of the three techniques used in producing tetrachloroethvlene
vary. Versar indicates that about one-half of the total U.S. production
is via the thermal chlorination of alkanes. About 46% of the total
production is via the chlorination or oxychlorination of ethylene di-
chloride. Approximately 3% of the production is via the chlorination
of acetylene (Versar, Inc. 1980).
Feedstock and thus operating parameters vary among the producers,
usually depending on what types of other products are produced at the
site. The thermal chlorination process uses a variable feedstock at
high temperatures. The catalytic chlorination of ethylene dichloride
operates at lower temperatures, and process parameters can be adjusted
to produce trichloroethene rather than tetrachloroethylene. The
chlorination of acetylene, uses an aqueous catalytic mechanism at much
lower temperatures than either of the other two routes.
15
-------
TABLE 3. DOMESTIC TETRACHLOROET1IYLENE PRODUCTION
CAPACITY AND DISPOSAL PRACTICES, 1976
Manufacturer/Location
Diamond Shamrock, Deer Park, TX
Dow Chemical, Freeport, TX
Pittsburg, CA
Plaquemine, LA
E.I. Du Pont de Nemours,
Corpus Christie, TX
^Production Capacity (MT)1
90,700
54,400
9,100
68,000
72,700
22,700
Ethyl Corporation, Baton Rouge, LA
Occidental Petroleum Corp. (Hooker), Taft, LA 18,100
PPG Industries, Lake Charles, LA 90,700
Stauffer Chemicals, Louisville, KY
Vulcan Chemicals, Ceismar, LA
Wichita, TX
TOTAL DOMESTIC CAPACITY
31,800
68,000
22,700
548,900
Reported Disposal"
Practices
NI - Not Identified
'''Facility has potential for incineration
Versar,1979a
2MRl,1977
Packaged, sealed; transported
for incineration
NI*
NI*
NI*
NI
Deep Well Injection
NI
Waste gas/Still residue liquids
incinerated
Gravity fed into land fill areas
Landfill*
Landfill*
-------
I97S DOMESTIC PRODUCERS PCE
1. Diamond Shamrock, Dear Park. Tx.
2. Dow Chemical. Frrapoct. Tx.
3. Oow Chemical, Plttsbiug. Ca.
4. Oow Chemical, Plaquamlne, La.
5. E.I. DuPont. Corpus Christl. Tx.
6. Elhyl Cu(p.. Baton Rouge, La.
7. PPG Industries. Lah, Charles. La.
8. Stauffer Chemicals. Louisville. Ky.
9. Vulcan Chemicals. Gelemar, La.
10. Vulcan Chemicals, Wichita. Tx.
11. Occidental Petroleum (Hooker), Tafl, La
! 5
10
Source: Arthur O. Little, Inc.
FIGURE 2
LOCATION OF TETRACHLOROETHYLENE PRODUCERS WITH
RESPECT TO EPA DESIGNATED RIVER BASINS
-------
The annual releases of PCE from production facilities are difficult
to assess, and widely divergent values have been reported, ranging from
a low of 291 MT (JRB Associates, 1979) to a high of 4,550 MT (Versar,
1979a). There is general agreement that virtually all releases durin^
production are airborne emissions, primarily from process vents, dis-°
tillation vents, and fugitive emissions (U.S. EPA, 1977)o
The wide range of estimated air emissions—291 to 4,550 MT per
year—is probably due to the assumptions made in deriving these values.
The low figure is probably unrealistic, as is the high value, which
represents a loss of 1.5% of product from the process (U.S. EPA, 1976).
In summary, environmental releases from production facilities are vir-
tually all airborne and are estimated to total around 3,000 MT per year.
Disposal practices for solid and liquid production wastes vary. One
plant reports deep well injection of tetrachloroethylene-bearing sludges.
Several place the material in containers and deposit it in landfills or dump it
directly in such locations. Incineration of solid waste and residual
liquids is currently practiced and is mentioned in the future plans of
several manufacturers (MRI, 1977). However, the high boiling point of
PCE and increasing energy costs may decrease the attractiveness of
disposal via incineration (MRI, 1977). At present, insufficient data
exist to reliably estimate this category of discharge.
D. DRY CLEWING AXD TEXTILE PROCESSING
PCE is miscible with most organic liquids and is a superior solvent
for greases, fats, waxes, and oils. Because of these characteristics and
the compound's stability and non-agressiveness towards dyes and pig-
ments, it has been used for many years in the textile industry for dry-
cleaning. The compound is also used in textile manufacturing for scour-
ing, sizing, and desizing operations and as a carrier for finishes and
dyes.
About 68% (216,900 MT) of annual domestic PCE consumption is for drycleanin;
and textile processing (MRI, 1977). The facilities are widely distrib- '
uted and range in size from coin-operated laundromats, often found in
shopping centers, to commercial drycleaning establishments, to large
textile manufacturing plants where the crude fabric material is processed
to a finished fabric product. There are currently about 25,000 commercial
drycleaners in the U.S., 75% of which use tetrachloroethene as a dry-
cleaning solvent. About 500 industrial facilities utilize both tetrachloro-
ethene and other petroleum-based solvents (W. Fisher, International
Fabricare Institute, personal communication, January, 1979).
18
-------
The atmosphere is the primary environmental receptor of releases
during textile cleaning. In the textile industry (as well as the metal
cleaning industry), some solvent recovery is practiced; however, it is
estimated that about 121,000 MT of tetrachloroethene is lost directly
to the atmosphere each year from this application. Although it is clear
that much of the PCE consumed for these purposes is eventually discharged
to the environment, the amount of solid or liquid waste that is disposed
of in landfills is dependent upon the modes of disposal employed by the
waste generators. An estimated 92,900 MT are present in spent solvent
solutions or sludges that result from solvent recovery processes.
Emission estimates are derived from consideration of parameters discussed
in several sources [(Radian Corporation, 1979; U.S. EPA, 1975, 1976,
1978, 1979; and U.S. EPA, (personal communication)].
Filter cartridges for drycleaning solvents are replaced weekly at
smaller commercial cleaning establishments. Solvents in industrial and
larger commercial operations are not filtered but are distilled to re-
move contaminants. Resulting solvent wastes can be as high as 95% oil
and grease. When practical, these wastes are placed and sealed in
55-gallon drums and sold to contracting recyclers. During separation
of the solvent from the oil and grease, a small amount of tetrachloro-
ethylene is released to the air (W. Fisher, International Fabricare
Institute, personal communication, 1980).
Carbon adsorption systems are being used increasingly in drycleaning
operations to treat waste materials (MRI, 1977). In the process, sol-
vents are routed through "chillers" to reduce' relative temperature and
then adsorbed onto carbon. Approximately 95% of the material is regulated
with steam and'the resulting condensate contains tetrachloroethvlene
as still bottoms. Current knowledge on the application of this"process
is limited and, therefore, the pollutant loading of the environment with
tetrachloroethylene from these systems is difficult to quantify. (Arthur
D. Little, Inc. Estimate).
Finally, there are some limited discharges of PCE to POTWs and water
as a result of its use in the drycleaning industry. Available data in-
dicate that 10 MT is discharged to POTWs annually, although actual re-
leases may be larger (Versar, 1974).
E. METAL DECREASING
Because tetrachloroethylene is nonflammable (i.e., no flash point)
and has good degreasing characteristics, it is used in vapor degreasing
applications in the metal cleaning industries as a replacement for tri-
chloroethane (U.S. EPA, 1979). The principal disadvantage in using PCE
in vapor degreasing is its relatively high boiling point, with associat-d
high energy costs and long cooling cycles (Versar, Inc., 1979). PCE
is also used in cold cleaning metal operations.
Approximately 17% (55,400 MT) of domestic tetrachloroethylene con-
sumption is used for metal cleaning (MRI, 1977; Radian Corporation, 1979)
19
-------
(See Appendix A). Metal cleaning operations represent a very distributed
use of tetrachloroethylene. There are approximately 24,000 vaoor deceas-
ing and about 900,000 cold cleaning facilities at which this solvent"ml?
be used, including gas stations, machinery manufacturers, and other "
metal working activities (U.S. EPA, 1979). A nationwide survey conducted
in 1976 by the Dow Chemical Company for the U.S. EPA, indicated that
2188 metal working plants reported using tetrachloroethylene for vapor
degreasing and cold cleaning (MRI, 1977). At the present time, metal
degreasing is estimated to release nearly 20,000 MT per year into the
atmosphere, 10 MT to water, 40 MT to POTW, and about 11,000 MT to land.
The U.S. EPA reports that all organic solvent cleaning operations
are to be subject to regulations of the Resource Conservation and Re-
covery Act (RCRA). Within the framework of RCRA, waste solvents and
still bottoms may be disposed of by distillation and incineration, land-
tilling, or storage in surface impoundments (GCA Corporation, 1979). As
mentioned previously, incineration is becoming less attractive because
or rising energy prices.
RCRA allows organic solvent cleaners generating less than 100 Kg of
waste per month of discharge into any state-authorized landfill area°
without using containment methods (GCA Corporation, 1979).
Approximately 37% of the metal cleaning plants recover waste sol-
vents at their own locations and 70% of these dispose of these sludges
in sanitary landfills. The remaining 63% of metal cleaning operations
sell their wastes to solvent reclaimers and 20% of these discharge the
materials to^landfills (U.S. EPA, unpublished). The U.S. EPA estimates
that 15-62.5% of the solvent consumed results in waste. If 50% of
tetrachloroethylene consumption is assumed to result in sludge and with
the breakdowns given above, on-site recovery appears to result in dis-
charges of 7,200 MT to landfills, while contract recvclers dispose of
3,500 MT similarly (Versar, 1979b).
The largest release from cold metal cleaning operations is thought
to be from waste solvent evaporation. The volume of evaporative emis-
sions from a vapor degreaser is significantly less than that from a cold
cleaner of similar capacity because vapor degreasing wastes have a higher
boiling point, volatilizing less rapidly, and because vapor degreasing
solvents contain expensive halogens, which are recycled (Bollinger-and
Schumaker, 1977). The U.S. EPA indicates that distillation is used to
recover c:ivent wastes in approximately one-half of all cpen-top vapor
degreasers (GCA Corporation, 1979).
F. OTHER USES
The production of fluorocarbons is the only consumptive use of fetra-
chloroethene and comprises about 12% (38,100 MT) of the industrial demand
(MRI, 1977, Radian Corporation, 1979). The fluorocarbons derived from
tetrachloroethene are used as solvents, grain fumigants, or as an anthel-
mintic in veterinary medicine (Versar, Inc., 1979a). The manufacture of
tluorocarbons is estimated to contribute about 60 MT of atmospheric
tetrachloroethene emissions (JRB, 1979).
20
-------
The remainder of the PCE supply, about 6,900 MT, is used in various
miscellaneous applications (Radian Corporation, 1979), such as use as a
specialty solvent in food processing and in aerosol specialty products.
All of the tetrachloroethylene consumed for miscellaneous uses is assumed
to be dissipated in the atmosphere. Tetrachloroethylene itself is also
used in very small quantities as an anthelmintic (Radian Corpoation,
1979; Versar, 1979a). In sludge form, it is occasionally applied to
road surfaces as a dust preventative.
G. TRANSPORT
Because of its chemical characteristics, tetrachloroethylene is
shipped in stainless steel, aluminum, or carbon steel tanks, usually (if
not always) at full strength concentration. It is then removed by pump
or air pressure. Corrosion does not pose a problem because all commer-
cial tetrachloroethylene is reported to contain stabilizers (MRI, 1977).
Following completion of transport, the tanks are flushed x^ith water,
steamed, rinsed and dried. The cleaning waters contain residual tetra-
chloroethylene in unidentified concentrations and are most probably
discharged to local sewers (MRI, 1977). Because of the limitations on
data regarding transport of the material, it was not possible to esti-
mate associated environmental releases, although it is expected to be
small with the exception of spills.
H. SUMMARY
Tetrachloroethylene is released to all environmental compartments,
as illustrated in Figure 1 and Table 2. About 88% of the total domestic
consumption is distributed (versus captive) use, and much of the material
is discharged to the environment as a consequence of its use. The
environmental compartment receiving the largest pollutant load of PCE is
the air, which receives nearly 150,000 MT of the substance directly and
indirectly each year. The sources are frequently point sources (e.g.,
drycleaners, production). Area sources contribute significantly to the
atmospheric sources due to tetrachloroethylene's highly volatile nature
resulting in indirect releases. The land compartment receives over
100,000 MT of PCE annually. Aquatic discharges of tetrachloroethylene
are reported in negligible amounts.
Although tetrachloroethylene consumption in the United States grew
at an average annual rate of approximately 5.3% during the 1971-1974
period, there has been a steady decline in production since 1972 (-4.3%),
and sales declined by 7.8% for the 1973 to 1977 period. Since that
time production and use have been stabilized somewhat. Further regu-
lation of the drycleaning industry could alter the demand for PCE.
21
-------
REFERENCES
Bellinger, J.C. and J.L. Schumaker. 1977. Control of Volatile Organic
Emissions from Solvent Metal Cleaning. Report No. EPA-450/2-77-022,
U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards, Research Triangle Park, N.C.
GCA Corporation. 1980. Organic Solvent Cleaners - Background Informa-
tion for Proposed Standards. Report No. EPA-450/2-78-045a, U.S. Environ-
mental Protection Agency, Triangle Park.
Feiler, H. 1980. Fate of Priority Pollutants in Publically Owned
Treatment Works. Interim Report. Burns and Roe Industrial'Services
Corp. , Paramus, N.J.
JRB Associates. 1979. Materials Balance - Task 14 Chlorinated Solvents.
Draft Report. Contract No, 68-01-5793, U.S. Environmental Protection
Agency.
Midwest Research Institute (MRI). 1977. An Assessment of the Need for
Limitations on Trichloroethylene, Methyl Chloroform, and Perchlorethylene.
Vol. I Contract No. 68-01-4121, U.S. Environmental Protection Agency,
Washington, D.C.
Radian Corporation. 1979. Organic Solvent Use Study. Contract No.
78-03-2776, U.S. Environmental Protection Agency.
United States Bureau of Census, U.S. Department of Commerce. 1978a.
U.S. Exports. FT 410/Dec. 1978, Year End Total (Item 5113300).
United States Bureau of Census, U.S. Department of Commerce. 1978b.
U.S. Imports for Consumption and General Imports. FT 135/Dec 1978 -
Year End Total (Item 5174045).
United States Environmental Protection Agency (U.S. EPA). 1975. Pre-
liminary Study of Selected Potential Environmental Contaminants Optical
Brighteners, Methyl Chloroform, Trichloroethylene, Tetrachloroethylene,
Ion Exchange Basins. Report No. PB-243-910.
United States Environmental Protection Agency (U.S. EPA). 1976. Air
Pollution Assessment of Tetrachloroethylene. Report No. ?B-25673i~
United States Environmental Protection Agency (U.S. EPA). 1977.
Criterion Document - Tetrachloroethylene Interim Draft No. 1 (14 December),
Criteria and Standards Division, Office of Water Planning and Standards,
Washington, D.C.
United States Environmental Protection Agency (U.S. EPA). 1978. Source
Assessment Reclaiming of Waste Solvents. State of the Art- Report"^
PB-282-934.
-------
United States Environmental Protection Agency (11,3. EPA). 1979. Source
Assessment Solvent Evaporation - Degreasing Operations. Report NoEPA~
600/2-79-0195. P
United States International Trade Commission. 1979. Synthetic Or^an^c
Chemicals, United States Production and Sales, 1978. ~ °
United States International Trade Commission. 1978. Synthetic Organic
Chemicals, United States Production and Sales, 1977. p. 360.
Versar, Inc. 1980. Environmental Material Balance for Tetrachloroethv-
lene- Contract No. 68-01-3852U.S. Environmental Protection Agency,
Washington, B.C.
Versar, Inc. 1979a. Production and Use of Tetrachloroethene. Contract
No. 68-01-3852. U.S. Environmental Protection Agency.
Versar, Inc. 1979b. Sources of Waste Chlorinated Hydrocarbons from
Degreasing and Associated Solvent Reclamation Operations Draft Report,
U.S. Environmental Protection Agency.
Versar, Inc. 1981. Data Summary for Metal Finishing Industry. Draft
Report. U.S. Environmental Protection Agency.
-------
CHAPTER IV
DISTRIBUTION 0? TETRACHLOROETHYLENE
IN THE ENVIRONMENT
A. INTRODUCTION
This chapter describes the important physicochemical properties of
tetrachloroethylene (Section B), provides detailed information on
observed ambient levels of the chemical in all environmental compart-
ments (Section C), describes the important pathways and degradation
routes in the environment (Section D), and gives the results of three
different modeling efforts that were designed to assist in the assessment
of the major transport pathways (Section E). A summary statement cover-
ing all aspects is provided in Section F.
Modeling of the fate of PCE in environmental media was undertaken
in order to illustrate important aspects of the chemical's fate and
transport in selected environmental scenarios. One model predicts the
expected concentrations (and amounts) of tetrachloroethylene in each
environmental compartment (air, water, soil, sediments/biota) under
the assumption that all phases are in equilibrium. By comparing the
predicted concentrations (preferably the ratio of concentrations between
two compartments) with measured concentrations (ratios), it is possible
to determine qualitatively how close to equilibrium the various phases
are, and in which direction interphase transfers of PCE are occurring.
A second model examines the expected atmospheric dispersion of
tetrachloroethylene downwind from a major point source such as a dry-
cleaning or solvent degreasing establishment. The results of this
analysis provide insight into the levels of exposure to PCE that mav
result from these major sources of PCE emissions. The third model
used is EPA's EXAMS, which analyzes aquatic fate in various environ-
mental scenarios.
B. PHYSICAL AND CHEMICAL PROPERTIES
At room temperature, tetrachloroethylene is a colorless, volatile,
heavy liquid with a pleasant ethereal odor. It is nonflammable and
incombustible, is fairly stable (the most stable of all chlorinated ethanes
and ethylenes), is fairly insoluble in water, but is an excellent
solvent for a variety of organic substances (e.g., fats, oils, tars,
rubber, and gum). It is these characteristics that have made the
chemical very useful as a drycleaning and metal degreasing solvent.
The important physiochemical properties are given in Table
<4 .
^Cleaning grades of the chemical contain from 0.01% to 0.1% by
weight of stabilizers; industrial grades contain up to about 0.35^weight %.
25
-------
TABLE 4. IMPORTANT PHYSICAL AND CHEMICAL PROPERTIES
OF TETRACHLOROETHYLENE
General Properties
Molecular Weight
Boiling Point
Melting Point
Vapor Phase
Autoignition Temperature
Flash Point
Flammable (explosive) Limits
Decomposition Temperature
Vapor Specific Gravity
Saturated Air Specific Gravity
Concentration in Saturated Air
Vapor Pressure:
-20.6°C
+2.4°C
13.8°C
26.3°C
40.0°C
Vapor Viscosity
Odor Threshold
Atmospheric Conversion Factor
Liquid Phase
165.83
121.2°C (at 760 mm Hg)
-22.4°C
none
none
none
700°C
5.76 (air = 1)
1.12 (air = 1)
2.43% by volume (25°C)
1 mm Hg
5 mm Hg
10 mm Hg
20 mm Hg
41 mm Hg
0.0099 centipcise (60°C)
5 ppm3'10, 50 ppm3
1 ppm =6.78 mg/m3 (25aC,
760 mm Hg)
Specific Gravity
Viscosity
Surface Tension
Refractive Index
Dielectric Constant
Dipcle Moment
Heat of Vaporization
Heat of Fusion
Specific Heat
1.623 (20/4°C)
0.84 centipoise (25°C)
31.3 dynes/cm (20°C)
1.5055 (20°C)
2.353 (15°C)
0
50.1 cal/g (at boiling point)
2.525 kcal/g-mole
0.205 cal/g-°C (20°C)
26
-------
TABLE 4 (Continued)
Binary and Tertiary Systems
Solubility in Water
Solubility of Water in Solvent
Solubility in Octanol
Also Soluble in
Will Dissolve
Will not Dissolve (to an
Appreciable Extent)
Octanol/Water Partition
Coefficient
Henry's Law Constant
Air/Water
150 mg/1 (25°C)J
400 mg/1 (25°C)2
140 mg/1 (25°C)-
165 mg/1 (25"a)*
91 mg/1
63_ mg/1
0.008 g/lOu g solvent
Infinite
Ethyl ether, Ethyl alcohol,
benzene, chloroform, others
Oils, fats, tars, rubber,
gum, sulfur, iodine, mer-
curic chloride, aluminum
chloride, ammonia (0.4% by
weight at room temperature),
benzoic acid, and other
organic acids
Sugar, glycerol, protein
7246
7597 (calculated)
4008
0.82 yg/l (air) (20'cr
yg/1 (water)
Several values are listed because of significant differences in
some reported values. Value from Dilling (1977) is an average of
four values from the literature (at 25°C or corrected to 25°C). The
variation of solubility with temperature is given by Antropov et al.
(1972) over the range of about 5°C-80°C.
General Sources:
Specific Sources:
Except where noted, the following references were
used: U.S. Environmental Protection Agency (1977);
Franklin Institute (1975); Fuller (1976); Lapp et al.
(1977); NIOSH (1976); Walter et al. (1976).
(1) U.S. Environmental Protection Agency (1977); Franklin
Institute (1975); Fuller (1976); Lapp et al. (1977);
NIOSH (1976).
(2) Chiou et al^ (1978); U.S. Coast Guard (1974).
(3) Neely (1976).
(4) Dilling (1977).
(5) Antropov (1972).
(6) Environmental Protection Agency (1977); Fuller (1976)
(7) Neely et al_._ (1974).
(8) Chiou et al. (1978).
(9) Fuller (1976); Neely (1976).
27
-------
The following chemicals may be used (in various combinations)as
stabilizers:
Amines (e.g., allyl amines) Epibromohydrin
Methylmorpholine N-methylpyrrole
Epichlorohydrin Allyl glycidyl ether
Stabilized tetrachloroethylene is inert to air, water, light, and
common construction metals at temperatures up to 140°C. In the absence
of moisture, oxygen, and catalysts, the compound is stable to a
temperature of about 5008C. At 700°C, PCE decomposes upon contact with
active carbon to yield hexachloroethane and hexachlorobenzene (Lapp
et al., 1977). V
In the absence of stabilizers, tetrachloroethylene will react with
a variety of chemicals under various conditions. A summary of the
available information is shown in Table 5. Oxidation (under ambient
conditions) is seen to take place only slowly, unless light or some
other catalyst or reaction initiator is present. Ultraviolet light can
lead to fairly rapid decomposition in air (half life, t , is about
1/2 day); an intermediate stage in this reaction is thought to be the
synthesis of peroxy compounds as shown below. Compound (a) undergoes
rearrangement to form trichloroacetyl chloride and oxygen. Compound
(b) decomposes to give two molecules of phosgene, a highly poisonous
gas (Lapp et al., 1977; Fuller, 1976).
C12C=CC12
C12C - CC12 2C12CO
I I Phosgene
0-0
(b)
Tetrachloroethylene is decomposed by contact with hot metals,
certain inorganic acids, hot carbon, and certain alkaline metals or
compounds of them. Unstabilized tetrachloroethylene can be corrosive to
metals; this has obvious implications for the "terminal" disposal of
waste solutions and sludges (containing the chemical) in unlined metal
drums.
28
-------
TABLE 5. DEGRADATION OF iETRACHLOROETHYLENE UNDER
VARIOUS CONDITIONS
CONDITIONS
RATE OF REACTION
REACTION PRODUCTS
Air, sunlight
Oxygen, UV light
Oxygen, no light
Ultra zero air,1 50%
relative humidity,
light
As above, plus added
N02
As above (no N02),
plus hydrocarbons
Air, added NO (5 ppm),
UV light
Air, added N02 (16.8
ppm), IP" light
High oxygen pressure,
chlorine
Ozone
Hydroxyl ions
(in atmosphere)
Alkyl peroxy radicals
N02
S03, 150°C
- 2 days
NA2
No reaction
100% decomposi-
tion in 1.5-2
hours
100% decomposi-
tion in — 8 hours
100% decomposi-
tion in 21-31
hours
t, ~ 14.2 hours
8.3 hours
V
NA
Slow, t, - 11
years
Rapid, t, — 8
days ^
Slow, tj~ 0.6
years
Rapid
NA
Chlorine, hydrogen chlo-
ride, trichloroacetic
acid
Trichloroacetyl chloride,
phosgene
Phosgene and others
Phosgene, carbon tetra-
chloride
Phosgene
NA
Phosgene, formic acid, tri-
chloroacetyl chloride, carbon
monoxide, hydrogen chloride
Trichloroacetyl chloride,
phosgene
Phosgene, trichloroacetyl
chloride
NA
NA
Te trachlorodini troe thane
Trichloroacetyl chloride
"Ultra zero air" is a term used by some commercial suppliers of bottled
air to describe a high level of purity. Unfortunately there is no
standard and the meaning varies considerably.
Not available.
29
-------
TABLE 5 (Continued)
CONDITIONS
_RATE OF REACTION
REACTION PRODUCTS
Excess hydrogen,
220°C, reduced Ni
catalyst
700°C, contact with
active carbon
In water, 150°C
In aerated water,
25 °C, dark
In aerated water,
ambient temperatures
(- 20°C to + 40°C),
natural sunlight
Strong inorganic acids;
e.g., HaSOi^ + HN03,
fuming
Butyl lithium in petro-
leum ether
Molten potassium
In presence of
dibenzoyl peroxide
Contact with iron at
450°C, zinc at 400°C,
aluminum at 4008C
Human metabolism (anal-
ysis of urine)
NA
Ammonia, high pressure NA
NA
Slow
t, — 8.8 months
t, <. 6 months
NA
Explosive-
Explosive
NA
NA
Hydrogen chloride, ele-
mental carbon
Ammonium chloride, ele-
mental carbon
Hexachloroethane, hexa-
chlorobenzene
Trichloroacetic acid,
hydrogen chloride
NA
NA
Trichloroacetyl chloride,
some tetrachlorodinitro-
ethane
NA
NA
Will yield copolymers with
styrene, vinyl acetate,
methyl acrylate, acrylo-
nitrile
Phosgene: 37, 17, and 3
mg/g of tetrachloroethylene,
respectively
Trichloroacetic acid, tri-
chloroethanol
From NIOSH (1976)
Sources: Lapp &t_ _al. (1977), except where otherwise noted.
30
-------
C. MEASURED CONCENTRATIONS IN THE ENVIRONMENT
1. Introduction
Long-term usage of tetrachloroethlvene in the U.S. — primarily for
drycleaning and metal cleaning — has led to continuous losses of the
chemical to the environment in all parts of the country. Losses to the
environment in the U.S. were estimated to amount to 256,750 MT in 1979;
about 60% of the losses are released directly to the air from point
sources (Versar, Inc., 1980), and 40% to land. (Losses to water and
POTWs are less than 1% of the total pollutant release.)
A relatively large amount of data exists concerning the concentra-
tions of tetrachloroethylene in various environmental compartments.
Data from U.S. sites are presented in the following figures and tables,
with the exception of the data on marine waters, sediments, and biota,
which are from Great Britain because data for the U.S. are lacking.
2. Data from Selected Surveys
In order to obtain a better understanding of the ambient concen-
trations of tetrachloroethylene and the variability associated with
the measurements, data have been selected from a number of surveys.
The data should be viewed as examples only; to generalize from
these data could result in specious results. The principal cautions
are associated with the following:
• The data are usually from a small number of samples,
sampling dates, and/or locations, and do not, therefore,
adequately represent averaged temporal or spatial
distributions.
• The concentrations measured are frequently near or
below the detection limit of the analytic method used.
Furthermore, the detection limit may vary, not only from
study to study but also within a study for different
samples.
a. Water
1. Drinking Water
The data base for PCE concentrations in drinking water includes
three studies by the U.S. EPA: The Ten Cities study (U.S. EPA 1975),
the National Organics Monitoring Survey (NOMS) (U.S. EPA 1978), and the
Community Water Supply Survey (CWSS) (Brass 1981). The results of these
31
-------
studies are presented in Table 6. In both NOMS and the CWSS, reported
frequencies of detection for PCE were less than 10%. The means for the
positive values ranged from 0.8 yg/1 to 3.4 yg/1, although the medians
were less than 0.5. yg/1.
PCE has been found in well water all over the U.S. Of 36 finished
drinking water supplies pumped from the ground, 22% contained PCE at
detectable levels (>0.2 yg/1). The mean of the positive values was 2.1
yg/1 and the high was 3.1 yg/1 (U.S. EPA 1980a). In a collation of State
data on the ground water quality as sampled from 2940 existing wells in
17 states, the percentage occurrence of PCE in the ten states where it
was detected ranged from 1% to 48%. It must be kept in mind that this
sampling is biased towards contaminated supplies. As an example of the
data, 372 wells tested in Nassau County, N.Y. were found to contain PCE in
57 (15%) (many of these wells were closed by the County Board of Health).
The maximum concentration determined in Nassau County was 375 us/1
(U.S. EPA 1980s).
In March 1980, levels of PCE in the drinking water of 72 Massachu-
setts communities were found to reach 5 mg/1 and averaged between 1.5 mg/1
and 2.5 mg/1. These levels were caused by PCE leaching from a resinous
liner of concrete pipes in public water systems. The resin has been
applied to prevent leaching of asbestos fibers from the pipes by acidic
wastes (Massachusetts DEQE 1980). Since then, even higher levels of PCE
have been documented. Wakeham et al., (1980) found a peak of 18 mg/1
in a water pipe in Falmouth, MA, but the line did not serve any house-
holds, and thus was not used for drinking water. In this study, house-
holds receiving contaminated water could have been exposed to PCE levels
as high as 2.2 mg/1.
2. Surface Water
Reported concentrations of PCE in surface water from various surveys
are depicted in Figures 3-6 and Table 7.
In a 1977 national survey of U.S. surface water in 14 industrialized
river basins, Ewing et al. (1977) detected tetrachloroethylene in 38% of
a total of 204 samples. Ninety-six percent of positive samples contained
PCE at concentrations less than 5 -_g/l. Only two percent of the samples
exceeded 10 -jg/l. Figures 3 and 4 depict the results of the survey/
Ambient surface water data were retrieved from the U.S. EPA's STORE!
system for the period preceeding and through 1981. Out of a total of 870
samples, only 9% were unremarked data. Ambient concentrations are sum-
marized in Table 7 by major river basin for unremarked and remarked
measurements. The mean and maximum unremarked levels for all major
river basins were 8.5 ug/1 and 142.0 yg/1, respectively. Remarked
levels averaged around 10 ug/1, an apparently commonly'used detection
limit.
-------
Number of Samples in Which Compound was Identified
ffj
s§
s- s
o 9
if
5-fr
-
& £.
x 5'
m
5'
to
re
ID
»j
-J
o
s
m
w
""
a
» o
m ^
s 5
m H
25
H
O
O
Tl
o
3
§ 8
NJ
O
o
T~
g g
Diethyl Hexyl Phthalate/
/ Dibutyl Phthalate/
Tetrachloroethylene
• Terpineoi/
to
/1,2-Dichloroethane
'Methyl Myristate/
7/////////V
Benzene
"
Dichloro-/,
methane //
'//////
Toluene//,
7/////A
-------
U
3)
"5.
1 23456789 10
Concentration (ng/\)
Source: Ewing et al., 1977.
FIGURE 4 FREQUENCY OF CONCENTRATIONS OF TETRACHLOROETHYLENE
FOUND IN 204 SAMPLES OF U.S. SURFACE WATERS
Note: Source cited states that tetrachloroethylene was identified in 77 of 204 samples taken at
various locations in 14 industrialized river basins. Only 69 data points were reported
however, and these are represented in the chart above. The number of samples in which
tetrachloroethylene was not detected is not shown. The percent figures shown - (Sample
Frequency/204) x 100.
34
-------
5 10
C[m9/ll
Source: Schwarzenbach et al., 1979.
FIGURE 5 VERTICAL PROFILES OF TETRACHLOROETHYLENE (PCE), TEMPERATURE, AND
DISSOLVED OXYGEN AT THE DEEPEST POINT OF LAKE ZURICH, SWITZERLAND
Note: Presence, in diagrams 1, 4, 5 and 6, of thermal stratification and
higher concentrations of tetrachloroethylene in the lower depths.
Number in parentheses below date in each diagram indicates number
of replicate samples collected at each depth.
The European abbreviation for PCE is PER and appears in this
figure.
35
-------
TABLE 6. CONCENTRATIONS OF TETRACHLOROETllYLENE IN U.S. DRINKING WATERS
Concentration (ug/1)
Study
NOMS 2 3
Phase II *
NOMS
Phase III
Quenched
Terminal
4
cwss
Surface Supplies
Ground Supplies
Ten Cities5'6
Miami, FL
Ottumwa, I A
Philadelphia, PA
Cincinnati, OH
Tucson, AZ
New York, NY
Lawrence, MA
Grand Forks, ND
Seattle, WA
Terrebonne Parish, LA
Frequency of , Median of
Detection Means of Positives All Data Observed Range
48/111 (43%) not quantified
8/106 (8%) 1.1 <0.2
9/105 (9%) 0.81 <0.2
3/106 (3%) 2.8 <0.5 <0. 5-3.0
18/330 (5%) 3.4 <0.5 <0.5-30.0
*v
0.1; 0.1; 1)
0.2; D
0.4; D
0.1; 0.3; D
<0.01
0.05; 0.46
0.07; D
0.2
ND
ND
0.187 0.07 ND-0.5
9D = Detected; ND = Not Detected.
~U.S. EPA (1978).
.Detection limits for Phase IT—unreported; Phase IIJ, 0.2 iic/J
'Brass (1981).
6ll.S. EPA (1975).
-.Values are actual data from ten cities.
-------
OJ
TABLE 7. SUMMARY OK STORET DATA TOR TETRACHLOROETHYLKNE AMBIENT
CONCENTRATIONS IN U.S. SURFACE WATERS (AS OF END OK
YEAR J981),
Has in
//
01
02
03
04
05
06
()7
08
09
JO
11
12
13
I 4
15
17
22
Name
Northeast
North Atlant ic
Southeast
Tennessee River
Ohio River
Lake Erie
Upper Mississippi River
Lake Michigan
Missouri River
Lower Mississippi River
Colorado River
Western (,'ulf
Paei fie Northwest
Ca I i fornJa
Creat Basin
Hawa i i
Lake Superior
Summary
No. of
Samples
0
26
3
2
26
5
1
2
4
2
0
2
5
o
0
o
0
78
Mean
1.4
19.6
67.5
0.05
3.8
63
2.0
32.0
5.4
99
1.7
8.5
S,D,
2.5
11.2
54.4
0.2
1.3
-
-
45.8
6.5
60.8
2.4
24.4
Max,
9.2
27.0
1 06 . 0
0.7
5.0
63
2.0
100 . 0
10. 0
142.0
5.9
142.0
No. of
Samples
4
78
39
31
48
22
31
46
198
107
25
46
84
8
12
1
13
792
Kemarked Data
i Mean S.I).
1.5 0.6
2.7 3.8
6.5 2.3
9.7 1.7
3.9 3
55.4 50
14.4 21.9
24.3 40.8
10.8
4.8
9.3
1.2
8.1
10.0 0
10.0 0
10.0 0
30.0 0
10.0
Max
2.0
10.0
10.0
10.0
10.0
100.0
50.0
100.0
50.0
50.0
10.0
10.0
10.0
10.0
10.0
10.0
30.0
1 00.0
^Weighted average (by number of samples) for all remarked codes (K, LI, and L) .
-Standard deviation not determined when more than one remarked erode reported for a basin.
Source: U.S. EPA (1980b).
-------
basins. In the vicinity of the E.I. DuPont plant, Corpus Christi TX
PCE was not detected; downstream of the PPG Industries facilitv Lake'
It
s a
to
waters (Person and McConnel , 1975)
38
-------
.3?
a
E
-------
Sitel
Key:
Max Concentration
(ppb)
• Site 2
• Site 3
• Site 4
Site 5
J_
10
100
1,000
Concentration (jug/I)
Source: Touhill, Suckrow and Associates, Inc., 1979.
10,000
100,000
FIGURE 7 MAXIMUM CONCENTRATIONS OF TETRACHLOROETHYLENE DETECTED IN THE
LEACHATE FROM, OR GROUNDWATERS NEAR, FIVE WASTE DISPOSAL SITES
Note: The researchers evaluated composition data (from a variety of published and unpublished
sources) on leachates, and contaminated ground and surface waters in the proximity of
27 sites known to contain hazardous wastes. Tetrachloroethylene was listed as a con-
taminant for five of these sites; it may have been present — but not analyzed for — at
additional sites. Sites 1, 2 and 5 involved the collection of groundwater samples while
"leachate" was collected at sites 3 and 4. Thus one can probably assume, as a rough
approximation, that about one quarter of the unsecure hazardous waste disposal sites
in the U.S. may involve some surface or groundwater contamination by tetrachloro-
ethylene. Such sites (with tetrachloroethylene) probably number in the thousands.
40
-------
TABLE 8. TETRACHLOUOETHYLENE IN WASTEWATER TREATMENT SYSTEMS
Concentration (|j.g/l)
'I'ap Water
1 0
Cincinnati, Oil ND
St. Louis, MO1 2
Hartford, CT1 N[>
Atlanta, CA] 2
Rye Meads, UK7'
DavyhuJme, l)K^
Salt ford, UK*
Countess Wear, UK
Minworth, UK/f
Indianapolis, 1N^ -
Cincinnati, 011^
Lewis ton, PA^
Atlanta, GA5
St. Louis, MO5
Potts town, PA
Grand Rapids, Ml5
C
Flint, Ml5
Hartford, CT5
1 Lev ins et al. (1979 a-d) all data
Various
Sewer Sites3 Influent
2.8 + 2.4(6) 1
20 + 10(8) 45
8.0 + 9.6(7) 26
53 + 70(9) 239
5
6
30
2412
46
51
5
45
305
115
9
15
26
— L
^
are a ve races from several samo
Primary
Sludge
—
_
-
5
5
66
38
17
293
61
1601
958
14
1642
32
le nnalvfips »
Digested
Sludge
_
_
-
_
_
__d
-
-
10
ND
423
ND
_
•)\/pr "\— ft rla\j
Effluent
-
0.5
ND
2
144
0.5
5
3
3
134
26
0
3
1
0
r
Percent
Removal
—
96
100
93
94
99
90
40
93
J -J
65
77
100
80
96
s l/
100
sampling periods.
ND: None Detected .
3 Simple average of average values at each site given along with standard deviation. Number in
parenthesis indicates number of sites sampled. Sites were at varying distances (upstream) from POTW.
Sewer sites sampled did not cover all sections of POTW collection area and thus area with significant
may have been missed.
,tetrachlorocthylene discharges
rBrown and Phil, 1978.
Burns and Roe, unpublished data, 1979.
-------
b. Air
The most recent data on levels of PCE in urban air were developed
by Singh _et. al. (1979, 1980). These researchers sampled seven cities
in the western half of the U.S. and determined mean concentrations
between 2 yg/mj and 10 yg/m3. The average of these mean values was
4.3 yg/m and the overall range of observations was 0.23-51.56 yg/m3.
Concentrations of PCE that have been measured in the atmosphere are
shown in Figures 8-10. Ground-level sampling (Lillian .et al. 1975) in the
Eastern U.S. (Figure 8) included an average value from a remote mountain
site of 0.50 yg/nr> (0.07 ppb) and average concentrations at urban sampling
sites between 1.02 yg/nT (0.15 ppb) in Wilmington, OH, and 1.15 yg/m3
(1.7 ppb) in Bayonne, N.J. The highest value reported was 68 yg/m3 in the
New York City sampling.
In a Los Angeles sampling program (Figure 9), PCE concentrations &r
ground level were generally (50% of the time) less than 13.6 yg/m3 (2 ppb).
The recorded high in this study was 30.5 yg/m3 (4.5 ppb) (Simmonds, at. al.
1974). These values do not differ greatly from those values measured in
eastern cities, nor from the data of Singh et_al_. (1979, 1980).
Pellizari et al. (1979) investigated levels of PCE in four highly
industrialized areas: Niagara Falls, NY; New Jersey; Baton Rouge, LA';
and Houston, TX. They found average values for these areas ranging from
0.12-210 yg/m , with the high found in the New Jersey area. These authors
also investigated levels near a chemical waste disposal site and found
trace levels to 395 yg/mj, with a median of 1.2 yg/m3 for 23 measure-
ments in various locations around the site.
Measured variations in PCE concentrations over the course of a day
are presented in Figure 10, which shows the results of three sampling pro-
grams at an urban site, suburban site in the Los Angeles Basin, and a
remote (mountain) site (Lillian _et al. 1975). In each case, the level
changes dramatically near evening. In central New York City, two distinct
concentration peaks were observed, reaching 68 yg/m (10 ppb) and 64 yg/m3
(9.5 ppb) at 9:30 a.m. and 6:30 p.m., respectively. The PCE levels were
less than 1.4 yg/m (0.2 ppb) at night and about 2.2 yg/m3 (0.34) during
the day. At both the suburban and the remote sites, only one unique peak
was Documented during the sampling day. These values were 68 and 47
yg/m for the suburban and remote sites, respectively. Of significance
is the observed drop in air levels to below detectable limits for the
10 hours 12:00-10:00 a.m. at the remote site. This is a clear indication
that PCE is found at increasing levels during the day due to importation
from upwind sources. The suburban site shows a late night and early
morning drop, and may be reflecting changes in a nearby urban area
(Los Angeles).
-------
-p-
OJ
-i r
T
Minimum Values < 0.02
Sea Girt, NJ
6/18-19/74
New York, NY
6/27-28/74
Sandy Hook, NJ
7/2-5/74
Delaware City, DL
7/8-10/74
Baltimore, MD
7/11-12/74
Wilmington, OH
7/16-26/74
J 1 I t i I I I
White Face Mountains, NY
9/16-19/74
Bayonne, NJ
3/73-12/73'
1 1 1 [ i I I I
0.01
J 1—I I I
o.i 1.0
Groundlevel Atmospheric Concentrations (ppb)
(Minimum, <- or |—, mean •, and maximum —) values)
10.
Source: Lillian et al., 1975
FIGURE 8 ATMOSPHERIC CONCENTRATIONS OF TETRACHLOROETHYLENE AT SELECTED SITES IN THE EASTERN U.S.
Note: Samples were collected at only one site in each city or location. The number of samples collected during the indicated
sampling period was not specified. The means reported are of determinations where detectable levels were measured In
a short term monitonncj effort in New York city, levels as high as 9.8ppb were observed. A large clean oceanic air mass
had moved into the Baltimore area during the sampling period and this, the authors suggest, probably accounts for the
generally low levels of all halocarbons measured ut this site.
-------
.5
Groundlevel Atmospheric Concentration (ppb)
Source: Simmondset at.. 1974
FIGURE 9 ATMOSPHERIC CONCENTRATIONS OF TETRACHLOROETHYLENE IN
THE LOS ANGELES BASIN
Note: Samples (total of 58) were collected at about 41 sites in the Los Angeles Basin
during September 22 and 28, and October 4, 1972. In general, different sites
were selected for each sampling day although some sites were used more than
once on a sampling day. All measurements have been lumped together in the
chart above. Winds headings were near 225° for most of the sampling periods
with wind speeds in the range of 5—10 knots. The average of all measurements
was 1.25 ppb. The highest daily average (2.2 ppb) was obtained on a day with
visible smog and generally stable inversion conditions.
44
-------
100
Location: West San Gabriel Valley, CA
_ u> -
• -
0.1 '-.
/ V
,w
23 Cc'ooer 24 Ocrocer
Source: Simmonds et al.. 1974.
I*X 2OOO 2200 MOO
Source: Lillian et al.. 1975.
UJC4TKW1 • N£W fOM OTr 14S'» ft » UXWStDH A* I
MOO l«00 WOO ZOO JJOO MOO
T!«E(HOU»S)
Source: Lillian et al.. 1975.
FIGURE 10 DIURNAL VARIATIONS IN THE GROUND LEVEL ATMOSPHERIC CONCENTRATIONS
OF TETRACHLOROETHYLENE (C2CI4) AND OTHER HALOCARBONS
Note: The large diurnal fluctuations seen at ground level for tetrachloroethylene
are presumably due to shifting wind patterns in combination with the
influence of major emission sources. Photochemical degradation couid
result in a diurnal fluctuation of 1-15% (i.e., a decrease of this amount
at the end of a sunlit day).
45
-------
Areas with high population densities in Amsterdam were studied for
PCE levels in exhaled air of residents near drycleaning establishments.
Using the relationship that ambient air levels are 2.5 x exhaled levels
as discussed by these authors, ambient air levels were calculated from
the data of Verberk and Scheffers (1980). It was estimated that workers
in 12 shops were breathing, on the average, 182,500 yg/m3, while resi-
dents living above the shops breathed 12,250 yg/m3. One home away
estimated PCE levels were 2500, two houses away — 550 and across the
?!£«?*' estimated levels were less than 250 yg/m3 (Verberk and Scheffers,
19oO) .
Figure 11 shows that as altitude is increased, the levels of most
halocarbons drop significantly. In particular, PCE concentrations change
from about 2.7 yg/mj (0.4 ppb) at 500 m to 0.07 yg/m3 (0.01 ppb) at 3000 m.
It follows the same general shape as the temperature curve, showing an
abrupt change at about 1700 m and 8°C, the location of a distinct
inversion layer. The declining concentrations above the inversion are
interpreted to mean that local emissions are the only source of atmospheric
PCE at this site (Simmonds et al. 1974). Whether this is true as one
moves East has yet to be determined.
c. Biota
The only data for PCE levels in biota are British and for fish in
salt or briny water. In Table 9, the observed levels in flesh range from
0.3 yg/kg to 11 yg/kg, while for liver the range is from 1.0 yg/kg to
41 yg/kg. The slightly higher range for the latter may be attributed t
the higher fat content of liver. Average water concentrations were
found to be about 0.5 yg/1 (Pearson and McConnell, 1975). Thus it -'
plausible to assume that some degree of bioaccumulation is taking p"
However, the lower levels of the local food chain were not sampled
plankton, filter feeders, nekton.
d. Foodstuffs
The only data for PCE levels in foodstuffs are from Gr
(McConnell, et al. 1975). The highest levels are observe*-'
high fat contents, i.e., butter, margarine, eggs, oils,
(5-13 yg/kg) . Levels in tea and coffee were also high .5
vegetables, and meats are in a lower range of values.
given in Table 10.
3. Summary;
Table 11 summarizes ranges of PCE concent-
mental media, foodstuffs, and biota abstracte ,-;
above. Monitoring of PCE in the environment ;
levels are found in close proximity to sou- . X'
as opposed to degradative processes may b
fate of PCE. The fact that, despite la*
intensive users of PCE, monitored leve"
20 ug/nr3 (except in industrial locatic a''4
indicates that degradative processes may
PCE's fate. The behavior of PCE in water n.
although volatilization may dominate the loss
46
-------
Distribution of atmospheric halocarbons
0.01 -
I
I
i OX 2000 3000
Altifuae, fn
4000
Source: Simmonds et al.. 1974.
FIGURE 11 TYPICAL CONCENTRATIONS OF HALOCARBONS, INCLUDING
TETRACHLOROETHYLENE (C-CIJ, AND AMBIENT TEMPERATURE
VS ALTITUDE
Note: Data are for the inglewood, CA area. The concentration
. of each compound is seen to decrease with altitude up to
the 1700 meter level where a significant inversion layer
was observed. The concentration of tetrachtoroethylene
continues to decline with increasing altitude above this
point. Simmonds et al. believe that the C-CI. measured
represents emissions only from stationary sources in the
Los Angeles Basin.
47
-------
TABLE 9. REPORTED CONCENTRATIONS OF TliTRACHLOKOIJTHYLENE TN FISH
SPECIES
ORGAN
liSJil clavata (ray)
Pleuronectes platessa (plaice)
Platycthys flesus (flounder)
Limanda 1imanda (dab)
00 Scomber scombrus (mackerel)
Limanda JLt!l!ii!K^Jl (dab)
Pleuronectes platessa (plaice)
Solea solea (sole)
Aspl trigla cuculus (red gurnarch)
Trachurus J;raj^iujirus (scad)
.T?"_isopterus hisOLUJ (pout)
Squalus acanjtliicis (spurdog)
Flesh
Liver
Fl esh
Liver
Flesh
Liver
Flesh
Liver
Flesh
Liver
Flesh
Flesh
Flesh
Flesh
Cuts
Flesh
Guts
Flesh
Flesh
SOURCE l
Liverpool Bay
Liverpool Bay
Liverpool Bay
Liverpool Bay
Liverpool Bay
Liverpool Bay
Liverpool Bay
Liverpool Bay
Liverpool Bay
Liverpool Bay
Redcar, Yorks
Thames Estuary
Thames Estuary
Thames Estuary
Thames Estuary
Thames Estuary
Thames Estuary
Thames Estuary
Thames Estuary
Thames Estuary
CONCENTRATION
0.3 - 8
14 - 41
4-8
11-28
2
1
1.5 - 11
15-30
1
ND'
5.1
3
3
4
1
1
2
4
2
1
-------
TABLE 9 (Continued)
SPECIES
Scomber scrombrus (mackerel)
Clupea sprat tus
Cadus morrhua (cod)
ORGAN
Flesh
Flesh
Flesh
Air Bladder
SOURCE1
Torbay , Devon
Torbay, Devon
Torbay , Devon
Torbay, Devon
CONCENTRATION2
(Mg/kg)
1.6
< 0.1
3.6
coverage concentrations found in water were 0.5 Mg/1.
Wet tissue.
o
JNot detectable.
Source: Pearson and McConnell (1975).
-------
TABLE 10. CONCENTRATION OF TETRACHLOROETHYLENE IN FOODSTUFFS
FOODSTUFF
TETRACHLOROETHYLENE
CONCENTRATION
(yg/kg)
Dairy Products
Fresh Milk
Cheshire Cheese
English Butter
Eggs
Meat
English Beef (steak)
English Beef (fat)
Pig's Liver
Oils and Fats
Margarine
Olive Oil (Spanish)
Cod Liver Oil
Vegetable Cooking Oil
Castor Oil
Beverages
Canned Fruit Drink
Light Ale
Canned Orange Juice
Instant Coffee
Tea (packet)
Wine (Yugoslav)
Fruits and Vegetables
Potatoes (South Wales)
Potatoes (Northwest England)
Apples
Pears
Tomatoes*
Black Grapes (imported)
Fresh Bread
0.3
2
13
ND
0.9
1.0
5
7
7
2
0.01
3
2
ND
ND
3
3
ND
ND
0.7
2
2
1.2
ND
1
ND - Not detected '~
-Tomato plants were grown on a reclaimed lagoon at Runcorn Works of ICI,
Source: McConnell et al._ (1975).
50
-------
TABLE 11. RANGES IN CONCENTRATION OF TETRACHLOROETHYLENE
IN THE ENVIRONMENT
Number of
Sites 2
Concentration
Typical Range
High Value
AIR
Background (remote sites, ground
level) 15-25
Urban Locations (ground level) ^ 20
Troposphere (0-20,000 ft.) 1 area
Stratosphere (20,000-30,000 ft.) 1 area
Near Manufacturing and User Sites 4
Near Chemical Dump Site 1
Work Areas inside Drycleaning
Establishments (DCE) >100
Above Drycleaning Establishment 12
Adjacent to DCE 12
Two Buildings Away from DCE 12
Across the Street from DCE 12
WATER
Drinking Water (surface) >100
Drinking Water (ground) <400
Marine Waters . > 2
Rain Water
Sewer Waters (4 cities) 30
Municipal Waste Waters (treated) 7
Waste Water from Drycleaning
Establishments 3
Contaminated Wells, Leachate several
SEDIMENTS, SLUDGES
Marine Sediments (Liverpool Bay) 1
POTW Sewage Sludge6 2
<0.1 - 0.5
1.0 -14.0
0.07-0.14
0-0.07
0.12-2.10
1.2
47,500-237,000
12,250
2,500
550
250
(yg/1)
ug/m
0.61
68.0
0.15
0.12
394
>6,780,000
<0.2- 3
<0.5- 4
0.1-.8
(.04;3
2-50
4- 5
3
5000
2.6
0.15*
200
10
5,000-110,000
1,000,000
>10,000
(yg/kg^
0.02-1
290, 7, 61
4.8
BIOTA7
Marine Fish (flesh)
Marine Fish (liver)
Marine Algae
Marine Invertebrates
Birds (various parts, eggs)
Mammals
Humans (various parts)
FOODSTUFFS
Various Foods
Number of Species'
15
15
6
13
8
2
(8 subjects)
2
Number of Foods
25
(yg/kg)
1-5
5-20
13-20
1-9
1-25
0.6-3
<.5-6
11
56
22
15
39
19
29
ND-3
13
51
-------
Notes to Table 11
- The concentrations given in this table are generally representative
of the available data resulting from actual measurements. Whenever
possible, data from measurements in the U.S. were used.
2 Approximate number of sites (species, goods) for which data were
available. Exact number not always specified in original reports.
Estimated using a water/air partition coefficient of 1.22 (see
Table 4) and an assumed average air concentration of 0.03 ppb.
Datum is from one site in England near an organochlorine manufacturing
plant.
Wells and leachate are presumably contaminated by nearby industrial
waste disposal sites.
5 Three values given are for: (1) primary sludge at plant A (290yg/kg),
(2) secondary sludge at plant A (<7Ug/kg) and (3) combined sludge at
plant B (61ug/kg).
The concentration of tetrachloroethvlene in the waters from which
these species (excluding humans) were taken averaged 0.5 ppb.
52
-------
D. ENVIRONMENTAL PATHWAYS AND FATE
1. Overview
Figure 12 provides a schematic overview of the potential environ-
mental pathways (transport and degradation) of tetrachloroethylene.
The major sources may be grouped as follows:
• Point-source atmospheric emissions
(e.g., manufacturing sites, drycleaning and solvent
degreasing establishments, solvent reclaimers).
• Area-source atmospheric emissions
(e.g., chemical dump sites, sewer system manholes, waste
water treatment impoundments [lagoons, aeration basins,
etc.]).
• Area-source discharges to land
(sites used for disposal of waste solvent and solvent
sludges from reclaiming operations).
• Point-source discharges to sewers and surface waters
(waste solvent or condensate from carbon control
systems discharged from user facilities, e.g., dry-
cleaners and metal degreasing establishments).
The direct releases of PCE waste streams to air or surface waters
appear not to contain any other chemicals that materially affect trans-
port and fate. The one exception is for the waste sludges from manu-
facturer, user, and solvent reclaimer facilities. These wastes will
differ significantly in their tetrachloroethylene content, in the nature
of the other wastes present, and in the actual manner of disposal. All
of these factors may alter the rate of escape (via volatilization or
leaching) from the sludge into other environmental compartments (air,
soil, ground water), and they may alter the relative amounts that are
transported to other compartments, but there will be no influence on
the major degradation pathways.
When finally released to the environment, tetrachloroethylene fol-
lows a few important transport and degradation pathways (see Figure
12). Because the atmospheric lifetime is on the order of a few days,
long-distance aerial transport (hundreds to thousands of kilometers) is
possible; photochemical degradation during sunlight periods is the only
significant degradation pathway. Minor amounts may be removed from the'
atmosphere by wet and dry fallout. Tetrachloroethylene in well-mixed
surface waters will volatilize fairly rapidly (half-life -1 day) into
the atmosphere; photochemical degradation provides a minor loss pathway.
The chemical can easily be transported to deep soils and ground waters,
and to deep surface waters and sediments, and in these compartments the
chemical will have a relatively long residence time, perhaps on the order
53
-------
Oi
Transport from
Distant Sources
(background concentration
(~20-70ppt)
Photochemical Degradation
(t
J4
Area-Source
Atmospheric
Emissions
(from dumps,
sewers, etc.
"2—10 days with constant
sunlight)
Volatilization
day for well-mixed
surface waters)
Point-
Source
Atmospheri
Emissions
55% of total)
Wet and Dry
Fallout
Photochemical Degradation in Witter
(l,, ~6 months)
Transport to I Jeep Waters and Sediments
(Inno residence time)
Note: Surface water discharges to large water body not shown for clarity, but should lie considered an important pathway.
Source: Arthur D. Little, Inc.
FIGURF. 12 MAJOR PATHWAYS OF TETRACHLOROETHYLENE
-------
of several years or decades unless the turnover or mixing time in the
compartments is shorter. While photochemical degradation plays no
part in these deep compartments, biochemical and chemical degradation
may cause gradual reduction of PCE levels.
The soil-to-groundwater pathway mentioned above is a key link in a
potentially significant pathway leading to drinking water contamination.
The complete pathway is:
Land disposal of tetrachloroethylene wastes
Leaching to groundwaters Transport with groundwaters
to wells and reservoirs transport to water treatment
plant and distribution system.
Several factors combine to make this pathway one of concern: (1)
over 100,000 MT /yr of PCE are disposed on land in the U.S. (Chapter III)
(2) numerous land disposal sites are presumably involved; (3) the mobil-
ity of PCE in soils is relatively high; (4) PCE is not readily degraded
while in the soil/groundwater compartment; and (5) conventional treat-
ment at water supply treatment plants will generally be ineffective in
removing PCS from the water supply.
The following subsections provide a more detailed discussion of the
transport and fate in each major environmental compartment.
2. Behavior in Air
Once PCE is in the atmosphere, aerial transport plays a major role
in the chemicals distribution throughout the environment, at least on
a regional basis. The compound is, however, subject to relatively rapid
chemical or photochemical degradation so that it does not continually
accumulate in the atmosphere and does not, itself, reach the upper
stratosphere* (ozone layer) in sufficient concentrations to affect the
ozone concentration (Lapp e_t al>_, 1977; Ross et_ _al^_, 1977).
Tropospheric attack on tetrachloroethylene may be by oxygen atoms,
hydroxyl free radicals, or ozone molecules; principal reaction products
from tropospheric degradation (Table 5) would include trichloroacetic
acid, phosgene, chlorine, hydrogen chloride, and other chemical species.
Rates of reaction, or associated half-lives, for a number of these reac-
tions (under laboratory conditions) were shown in Table 5. These data
indicate that a relatively short tropospheric half-life, perhaps 1-10
days, is possible. Other estimates, based both on laboratory work and
on an atmospheric budget model, indicate longer tropospheric half-lives
(see Table 12).
"This may not hold true for some of the degradation products
55
-------
TABLE 12. TEOPOSPKERIC HALF-LIFE OF TETRACHLOROETHYLEXE
Half Life
Method
Reference
0.9 days'
1.5 days1
1 day
4 days
10 days
12 weeks
21 weeks
Experiment with simulated
atmosphere containing
initially 3.9 mg/m3
Experiment with simulated
atmosphere containing
initially 13 mg/m3
Estimate
Atmospheric budget
OH concentration and rate
constant
Laboratory rate
Atmospheric budget model
Billing et_ al. (1976)
Billing et al. (1976)
Lapp £t_ al. (1977)
Lapp et_ al. (1977)
Lapp et. al. (1977)
Pearson and McConnell (19
Lapp et al. (1977)
Under conditions of bright sunlight.
56
-------
Various studies have been conducted on the atmospheric reactions
of tetrachloroethylene. One study showed that oxidation of the chemical
may be initiated by a chlorine atom with a long-chain free radical pro-
cess resulting; the major products of reaction at high oxygen pressure
were found to be trichloroacetyl chloride (75%) and phosgene (25%). The
quantum yield for oxidation was 300; oxygen was found to have an inhib-
iting effect on the photochlorination reaction (Lapp _et_ al. , 1977) .
The photooxidation may also be initiated directly by sunlight, in
which case a C-Cl. ra<*ical is formed along with a chlorine radical:
ci2c = cci2 llght ci2c = cci- + ci-
The latter may then initiate the chain-reaction process mentioned
above. A study of this light-initiated reaction found a linear relation-
ship between the rates of trichloroacetyl chloride formation, phosgene
formation, and absorbed light. Yields were found to be independent of
the tetrachloroethylene pressure (Lapp et al. , 1977).
Both hydrocarbons and nitrogen oxides (at least N02) have an inhib-
iting effect on the photooxidation of tetrachloroethylene. (See Table
5 for some data.) In one study --not necessarily contradictory —
tetrachloroethylene decomposition rates increased as the concentration
of tetrachloroethylene and NO were (simultaneously) increased; a tenfold
increase in the concentration of both species resulted in a 3.0-fold
increase in the decomposition rate (Dilling et_ al., 1976) .
A significant fraction of tetrachloroethylene in the atmosphere
may be associated with water droplets and dust particles, especially
organic particles. (Table 4 has data on distribution coefficients.)
From the atmosphere, tetrachloroethylene could enter the hydrosphere by
direct transfer (dry impact) , washout by rain, or dry fallout of particles
with adsorbed tetrachloroethylene.
3. Behavior in Water
Tetrachloroethylene undergoes hydrolysis very slowly in the presence
of water. The reaction products from hydrolysis
C12C = CC1? + H90 • C13COOH + HC1
are trichloroacetic acid and hydrochloric acid.
The half-life for chemical degradation to be expected in natural
water bodies has been reported from two different studies. In one (see
data in Table 13), a half-life of 8.8 months was associated with a test
system containing aerated water at 25°C with no light. With natural
sunlight and ambient temperatures (which ranged from 20°C to +40°C),
57
-------
TABLE 13. DECOMPOSITION RATES OF TETRACHLOROETHYLENE IN
AERATED WATER IN THE DARK AND IN NATURAL
SUNLIGHT1
Concentration
Dark Reaction
Initial
Dark
Light
Six Months
Dark Light
Twelve Months2
Dark Light
_1 Half Life,
k3, mo Months
1.00 1.00 0.63 0.52 0.35 0.24
0.41 0.25
0.079 + .002
8.8
Dark reaction is at 25°C; natural sunlight reaction carried out at
ambient temperatures (-20°C + 40°C).
2Duplicate test results.
Decomposition constant calculated on the assumption of a first order
reaction.
58
-------
the chemical degradation half -life was about 6 months (Billing et al. ,
1975) . The second study resulted in an estimate of 72 months for" the
chemical degradation half-life in water (Pearson and McConnell, 1975).
The difference in the two estimates may be due to the fact that in the
latter case the researchers corrected for the loss of tetrachloroethylene
into the air space over the liquid samples (Lapp et al. , 1977).
The main process for the removal of tetrachloroethylene from shallow
surface waters is volatilization (Billing £t al. , 1975). Laboratory
experiments have measured the rates of evaporation from a stirred beaker
(250 ml beaker, 1 mg/1 of chemical in 200 ml water, solution depth 6.5
cm, still air, 25°C, stirred at 200 rpm) yielding half-lives for evapor-
ation of 26 + 3 minutes (Billing et al . , 1975) and 24+3 minutes (Neely,
1976). The results have been found to agree well with theoretical pre-
dictions derived from interphase transfer processes (Chiou et al. , 1978;
Neely, 1976). The calculated half-life corresponding to the~~experimental
situation described above is 26.5 minutes (Neely, 1976).
The theoretical model referred to above allows a rough estimate of
the half-life for evaporation (tly/2) into still air from any well-agitated
surface water of depth (d) by the following equations:
(d in cm)
(d in cm)
The good agreement between the experimental and theoretical half-
lives (at 25°C) is considered somewhat fortuitous. The second equation
(for use at 1.5°C) underestimated the measured half-life under the lab-
oratory conditions described: 27.0 minutes predicted, 37.5 minutes
measured (Neely, 1976) .
The evaporation rates for tetrachloroethylene were also measured
under conditions more nearly like those found in the environment. Addi-
tion of various contaminants (clay, limestone, sand, salt, peat moss, and
kerosene) to the water had relatively little effect on the evaporation
rate. However, an increase in the wind speed across the top of the
beaker from 0 + 0.2 mph to 2.2 + 0.1 mph caused a significant increase
in the evaporation rate; after 20 minutes, the solute evaporation was
about 17% greater with the higher wind (Billing et al . , 1975) .
A somewhat different study — involving laboratory aquaria — also
sheared the significance of evaporation in the loss of tetrachloroethylene
and other chlorinated hydrocarbons from aquatic systems (Jensen and
Rosenberg, 1975). Natural seawater was used in this study, which was
59
at
1
at
t
25°C
:, 12 (min.)
1.5°C
• ,„ (min.)
= 4.08 d
= 4.15 d
-------
carried out to investigate the degradability of the chemicals. In experi-
ments with both open and closed systems, the researchers found that
evaporative losses (from the aquarium) were greater than losses by
degradation. No significant differences in degradation were noted
between a closed system kept lighted and one kept in darkness. Losses
of tetrachloroethylene from the "open" system (a partially covered
aquarium, 40-1 capacity, filled with 20-1 of seawater, held at 11°C
to 12°C) amounted to just over 50% after eight days.
In addition to the predictive equations discussed above, one other
approach to predicting the rate of volatilization from surface waters
has been described by Smith et al. (1980). Using this approach we
estimate approximate half-lives for volatilization from a "typical"
river, lake, and pond as 1.3, 5.0, and 6.4 days, respectively. By
comparison, when a well mixed depth (d) of 5 m is inserted in the
Neely equation above, the half-life (tj/o) = 4.08 (500) min =1.4 days.
The predictions for rivers could probably be taken as an upper limit
for POTWs with aeration basins or other well mixed impoundments. If,
for example, one assumed a volatilization half-life of 0.5 days for
such POTWs and a residence time in the aeration basin of 1-2 days,
a volatilization loss of 75% to 94% would be expected. This is in
line with a 81-96% volatilization loss that may be inferred from the
measurements of tetrachloroethylene in the influent, aqueous effluent
and sludge from one POTW (Feiler, 1979).
4. Behavior in Soils and Sediments
The movement of tetrachloroethylene through soils or sediments has
not been studied, although movement would clearly be possible as a
result of leaching (transport in solution) and/or volatilization
(transport in the vapor phase in unsaturated soils). It may be
presumed that, when water is present, a partitioning exists between
the two phases, which may or may not be at equilibrium. Chemical and
biological degradation play a very minor role, if any, and the chemical
would be expected to persist in deep soils and groundwaters. There are
data on the concentration in sediments from one area, Liverpool Bay
(see Figure 6), but there there is no direct evidence of any trans-
formations that may be taking place. The Liverpool Bay data showed
no direct correlation between concentrations in sediment and that of
the overlying water at the time of sampling (Pearson and McConnell,
1975).
The persistence of tetrachloroethylene in soils is not well docu-
mented. It is possible, however, that certain reactions, presumably
biochemical, could occur in cultivated soils that could assist in
the loss of PCE from the soil compartment. Such reactions have been
seen for chemically similar compounds, including ethylene dibromide,
l,2-dibromo-3-chloropropane, and 1,3-dichloropropane. None of these
compounds x-ras detected in food crops grown in soil pretreated with
the chemicals; no noticeable tendency for bioaccumulation was observed;
and the chemicals were found to disappear fairly rapidly from the
soil. An increase in the amounts of inorganic bromide and chloride
ion was, however, found in the plants (National Academy of Sciences,
1975).
60
-------
Tetrachloroethylene has the potential to move from land disposal
sites, through the soil, to underground aquifers. Two case histories,
involving the contamination of water-supply wells for two towns and an
industrial park, have been reported (Jarema, 1977). In one case, the
water drawn from a 250-ft deep well was thought to have been contami-
nated by inadvertent spillages (during loading and unloading operations)
at an adjacent industrial site; concentrations of tetrachloroethylene
up to 50 yg/1 were found in the raw water from the contaminated well.
In the second case, discharge of waste solvents into a dry well resulted
in the contamination of one town well located approximately 200 yards
away; the concentration of tetrachloroethylene in the (raw) well water
was 4 ug/1. Wells from a nearby industrial park were found to have
a PCE concentration of 0-600 yg/1 (Jarema, 1977).
For tetrachloroethylene in topsoils or in landfills that have not
been sealed, volatilization may be a significant loss mechanism. Atmo-
spheric concentrations near one dump site in Edison, NJ, showed signif-
icantly elevated levels of the chemical; the highest concentration^
from a sample collected on the dump site, was nearly 400,000 ng/m3
(-v58 ppb) (Pellizzari, 1978). This value was three to four orders of
magnitude higher than atmospheric concentrations at more distant or
remote sites.
There appear to be little or no data on the composition and disposal
practices of solvent sludges that are disposed of on land. The solvent
content can apparently vary from just a few percent to nearly 100 per-
cent, depending in large part on the extent and efficiency of solvent
recovery operations at the user sites. The initial solvent sludge
will obviously contain the oils, greases, dirt, etc., that the solvent
was used to remove. It appears likely, hox^ever, that such wastes are
commgled with other wastes (e.g., lubricating oils in the case of
automotive repair shops) before final disposal, so that the final
wastes have a variety of characteristics. The extent to which such
wastes nave been disposed of in corrosion-resistant drums and/or in
secure landfills is unknown, but is thought to be very small. Current
RCRA regulations prohibit durmmed disposal of PCE except for small
generators (<100 kg/mo).
Wastes from drycleaning establishments are also sludges containing
a large amount of PCE. However they are generally collected in dis-
posable filter cartridges which are disposed of along with other
municipal wastes.
61
-------
The fate of the PCE depends upon the ultimate disposal of the
municipal wastes. If they are buried deeply beneath other wastes or
covered with soil, the potential exists for groundwater contamination
through leachate migration. PCE contained in wastes left on the sur-
face would tend to be volatilized.
Jones and McGugan (1977/78) have documented that a significant
amount of the tetrachloroethlyene in land-disposed solvent sludges can
escape to the atmosphere. In one set of experiments conducted by
these authors, the rate of evaporation of the chemical from a pulverized
domestic waste was compared with the rate of evaporation from a liquid
pool (of the chemical) under the same environmental conditions. In a
second set of experiments with the pool and pulverized waste, the rates
of evaporation of the chemical were measured after a tetrachloroethylene-
oil mixture was added. The data from these tests, shown in Figure 13,
indicate that as much as 20-30% of the tetrachloroethylene that is land-
disposed may be volatilized within 4-5 hours after disposal. Over a
period of several days, the amount lost by this mechanism could clearly
be much higher.
Further tests were conducted by the same research group on the
potential for the leaching of tetrachloroethylene through domestic
waste to groundwaters (Jones et al^_, 1977/78). When the solvent, in
the form of filtration residues from drycleaning processes, was added
to domestic wastes* at a loading of about 200 mg/1, leachate did not
contain concentrations of the chemical above the detection limit of
2 mg/1. This held true even for one experiment in which the possibility
for volatilization was eliminated by a cap placed over the waste con-
tainer. After the leach tests with this capped container, about 15%
of the tetrachloroethylene initially added was recovered from the
waste by extraction with 68 1 of petroleum spirit. In contrast, pet-
roleum extract of a column that had been left open to the air did not
contain detectable levels of tetrachloroethylene. It appears from
these tests that the combination of volatilization and adsorption of
tetrachloroethylene (in waste sludges) applied to domestic refuse can
significantly reduce the potential for groundwater contamination.
5. Biodegradation
One biodegradability test on PCE has been carried out with a static
culture flask screening procedure (Tabak et al., 1980). This test
utilizes BOD dilution water containing 5 mg/J of yeast extract as the
synthetic medium, a 7-day static incubation at 25°C followed by 3 weekly
subcultures (totaling 28 days of incubation), and incorporating settled
domestic wastewater as a microbial inoculum. Tests were carried out in
glass stoppered reagent bottles (to minimize volatilization losses) and
in darkness (to minimize photodegradation losses). Volatility controls,
*Fifty kg of fresh, untreated domestic wastes placed in steel columns
about 56 cm in diameter and 90 cm deep.
62
-------
. Tetrachloroethylene
Only
II. Tetrachloroethylene
Plus Domestic Waste
IV. Tetrachloroethylene
'lus Oil and Domestic Waste
III. Tetrachloroethylene and Oil Only
0 20406080100
200 300
Time (Min.)
400
500
Source: Jones and McGugan, 1977/78.
FIGURE 13 VOLATILIZATION OF TETRACHLOROETHYLENE
AND TETRACHLOROETHYLENE-OIL MIXTURES
FROM LIQUID POOLS AND DOMESTIC REFUSE
Note: An open square tray 1m x 1m x 3cm (deep) was used to measure
volatilization from liquid pools. An open square tray 1m x 1m x
0.5m (deep) was used for the experiments with domestic waste.
Weight loss was the parameter monitored in all experiments. The
domestic waste, when placed in the tray, had a density of
~ 350 kg/m3. The weights of material used for each test were
as follows:
Test
(See Figure)
1
II
III
IV
Tetra.
20.2
29.0
16.3
16.3
Weight (kg)
Waste
_
155
—
167
Oil
_
7.75
7.2
The conditions under which the two sets of tests were run were
as follows:
Mean wind speed (m/sec)
Ambient Air Temperature(°C)
Solvent Pool Temperature(°C)
Waste Surface Temperature(0C)
I and II
~ 10
~ 12
~ 15
-12-20
III and IV
~2.5
-22
- 19-35
- 18
63
-------
utilizing a non-biological system (medium-test compound without inoculum),
were held for a period of 10 days and then analyzed to determine volatil-
ization losses. The concentration of PCE used in both the inoculated
and volatility-control tests was 5 mg/1. The volatility-control showed
a 23% loss of PCE (at 25°C, 10-day period) from the flask, due (presumably)
to volatilization. The original culture, and first, second, and third
subcultures showed PCE losses of 45%, 54%, 69%, and 87%,respectively.
These percentages are uncorrected for volatilization losses. Tabak et al.
(1980) conclude from these tests that PCE under went "significant degTa-
dation with gradual adaptation."
Two analytical studies have considered the fate of PCE during waste-
water treatment, a process which may involve biodegradation as a mechanism
for pollutant removal.
Burns and Roe, Inc. (unpublished, 1979) sampled influents, final
effluents and at intermediate treatment stages of nine publicly owned
treatment works (POTW's) throughout the U.S. (see Table 8). Percent
removals of PCE at these 9 POTW's were between 40 and 100%, and averaged
69%. In almost every case, PCE concentrations in primary sludges were
much higher (by factors of 2 to 182) than influent levels, while levels
in digested sludge were either "not detected" or much lower than primary
sludges. Because primary sludge does not involve biological activity
while digested sludge does, these data indicate that PCE may be bio-
degraded by aquatic micro-organisms. Acclimation was not studied in
these analyses. An alternative explanation involves volatilization
during secondary treatment.
A study of five treatment works in the United Kingdom (Brown and
Phil, 1978) also gives evidence for biodegradation of PCE. The researchers
did not sample biologically produced sludges. However, the high percent
removals and the fact that PCE levels in effluents and primary sludges
were of the same order of magnitude are indicative of an active removal
mechanism for PCE following the physical settling process which pro-
duces primary sludge. Brown and Phil state that both biodegradation
and volatilization were important processes for PCE removal and felt,
without hard data however, that the latter process might remove more
PCE from wastewater during treatment than the former.
Although not shown in Table 8, Burns and Roe, Inc. data included
pre- and post-chlorination effluent concentrations. In no case was the
chlorinated effluent higher in PCE, which indicates that as the "per"
chlorinated form of ethylene, it is least likely to be produced by
treatment.
E. CONCENTRATION ESTIMATES BASED ON ANALYTIC MODELS
1. Overview
Analytic models are used to increase understanding of complex
systems in which many interacting physical processes and chemical
reactions determine the fate of given substances. The major path-
64
-------
ways describing environmental fate of PCE are shown schematically in
Figure 14. The pollutant releases from the six sources are weighted
in order to indicate which compartments receive the largest inputs.
The pathways through which PCE moves are shown in broken lines to
indicate that, despite our knowledge of both the input amounts and
the dominant fate processes, the proportion of the PCE input that
follows each pathway is unknown. Three major pathways are shown in
the figure:
(1) Landfills groundwater surface water - air
(2) Landf ills air
(3) POTW - surf ace water -air
Three models have been selected to consider these pathways and the
source emissions/discharges.
The first is a general environmental partioning model, which
demonstrates the tendency of a chemical to accumulate, disperse or
degrade in all compartments and which compartments will be most
affected. The second model performs an atmospheric dispersion
analysis of the emissions from two types of drycleaning operations;
the large commercial and the point source industrial facility. These
sources were selected because they contribute the largest air emissions
of PCE. While nearly comparable amounts of PCE are conveyed to land-
fills, about 30% volatilized in simulation experiments. In addition,
the levels of PCE in air monitored near a chemical dunp were more
than three orders of magnitude less than levels measured near commerical
sources. Thus due to the lack of information on exact volatilization
rates of PCE from waste disposal sites, it was not possible to model
releases from this source. The three pathways in Figure 14 that lead
to the surface water compartment are analyzed by use of the US EPVs
EXAMS model.
2. Equilibrium Partitioning
As an initial step in hazard or risk assessments for toxic chem-
icals, in the planning of laboratory and field tests, and in the
interpretation of data on ambient concentrations, it is important to
understand the likely transport and fate of the chemical. Which en-
vironmental compartment (air, water, soils, sediments, biota) will
be most affected? Which degradation pathways (photolysis, hydrolysis,
biodegradation, etc.) will be most important? Rough guesses'can often
be made by simple inspection of the chemical's properties and reaction
rate data (if such are available) or by the use of mathematical models
that seek to yield defensible and quantitative estimates for dynamic
situations. Unfortunately, realistic chemical fate models usually
require extensive input information on both the chemical and environ-
mental compartments of concern (such are not always available), and a
-------
PRODUCTION
MISCELLANEOUS
SOLVENT USE
FLUROCARBON
PRODUCTION
METAL DEGREASINS
DRYCLEANING
OTHER
M.T. / year — Key Quantity
105
,4
— 10
103
- 102
111
; impossible to estimate at this time
FIGURE 14 SUMMARY OF ENVIRONMENTAL FATE OF TETRACHLOROETHYLENE
66
-------
computer for solving the lengthy calculations. A simple, initial esti-
mate of environmental partitioning is thus desired; one involving an
approach with minimal data requirements and capable of solution with
a hand calculator. Such an approach has recently been proposed by
Mackay (1979) in a treatment based upon the fact that in a system at
equilibrium, the fugacity of the pollutant must be the same in all
phases.
A three-tiered approach has been proposed by Mackay (1979) . In
Level 1 (the approach used here) all environmental compartments (phases)
are assumed to be directly or indirectly connected, and at equilibrium.
The compartments considered are air, surface water, suspended sediments,
bottom sediments, soil, and aquatic biota. The Level I calculations
require that these compartments be roughly described (volumes, temper-
ature, sediment and biota "concentrations," etc.) and the model output
will clearly depend on the nature of the "environment" selected. The
compartment-specific parameters chosen here (somewhat arbitrarily) are
listed in Table 14, A schematic diagram of the selected "environment"
is shown in Figure 15.
The Level I calculations do not consider degradation, or transport
into or out of the selected environment. A relatively small number of
chemical-specific parameters (also listed in Table 14) are required
for the equilibrium partitioning. If one desires an absolute 'estimate
of the equilibrium concentrations in each phase, it is necessary to
estimate the totaf amount of the chemical that is likely to be in the
selected environment.* We have taken this amount to be 11 moles/km2
(1.82 kg/km^), or just 11 moles in our compartment whose surface is
1 km^. This amount is equivalent to the pollutant releases in_the
U.S. over a 12-day period, divided by the area of the 48 contiguous
states. Implicit in the selection of this quantity is an assumed
atmospheric half -life (due to photochemical degradation) of about 4
days in constant sunlight or 8 days when diurnal light cycles are
considered.
Details of the calculational methods are provided bv Mackay (1979)
and are not repeated here. The calculations were carried out for
three different temperatures (0°C, 10°C and 20°C) in order to assess
the importance of this parameter of equilibrium partitioning. The
results of the calculations are given in Table 15. Not indicated
by the numbers in this table is the prediction that 99.8% of the
chemical resides in the atmosphere (at equilibrium) in the selected
environment.
^ _ It is interesting to compare the numbers in Table 15 with measured
amuient concentrations (Table 11) . The atmospheric concentration is
estimated with some degree of accuracy but it appears, not surprising
that measured concentrations in other phases (especially surface waters
and _ biota) are higher than would be expected under equilibrium con-
fr0m°-°072 <*8/kg:ug/k8) at
*Note that predicted ratios of concentrations between two phases will
not be affected by the number selected.
67
-------
TABLE 14. VALUES OF THE PARAMETERS USED FOR LEVEL I CALCULATION
OF EQUILIBRIUM CONCENTRATIONS OF TFT^fYI.OROETHYLENE
USING MACKAY's FUGACITY METHOD
Chemical-Specific Parameters
_. Parameters Values At:
20°C 1Q°C Q°C
• Solubility (mg/L) :
• Vapor pressure (mm Hg)
140
14.4
104
8.04
82.7
4.32
• Adsorption coefficient (K ) for:
P
a) suspended sediments 29 34 38
b) sediments 29 34 38
c) soils 5,8 6.8 7.7
g/m'
• Octanol/water partition coefficient (all temperatures): 724
(Used for estimating a bioconcentration factor for aquatic biota.)
• Total amount of chemical in compartment: H moles/km2 (1.82 kg/km2)
(Equivalent to total U.S. environmental losses over 12 day'period,
divided by area of 48 contiguous states.)
Compartment-Specific Parameters
• Temperature: 20°C, 10°C and 0°C
q
• Concentration(S) of suspended sediments: lOg/m
6 "}
• Concentration(S) of soils and sediments: 2 x 10 g/m
• Volume fraction (B) of aquatic biota: 50 x 10 m3/m3
• Fraction (y) of aquatic biota equivalent to octanol: 0.2
• Accessible volume for each subcompartment:
1. Air: 1 km x 1 km x 3 km (high) = 3 x 109 m3
2. Surface water: 1 km x 0.05 km x 3m (deep) = 1.5 x 105 m3
4. Sediments: 1 km x 0.05 km x 10 cm (deep) = 5 x 103 m3
6. Soils: 1 km x 0.95 kin x 14 cm (deep) - 1.3 x 105 m3
(Note: in the preliminary calculations, the suspended sediments and
aquatic biota are assumed to have the same "accessible volume" as
the surface water subcompartment.)
Parameters for specific compartments may be selected to reflect the
nature and size of any area of concern. The values used here are
considered to be reasonable rather than typical.
68
-------
Atmosphere
Soil
Surface Water
Aquatic Biota and
Suspended Solids
Bottom Sediments
FIGURE 15
SCHEMATIC OF ENVIRONMENTAL COMPARTMENT SELECTED FOR ESTIMATION
OF EQUILIBRIUM PARTITIONING OF TETRACHLOROETHYLENE
Note: Diagram is not to scale. Dimensions and accessible volumes of each
subcompartment given in Table 15.
69
-------
TABLE IS. LEVEL I CALCULATIONS OF EQUILIBRIUM CONCENTRATIONS OF
TETRACHLOROETHLYENE USING MACKAY's FUGACITY METHOD
Concentration (ng/kg)3- at:
Compartment
Air
Surface waters
Suspended solids
Sediments
Aquatic biota
Soils
20°C
90
0.65
19.
37.
94.
7.5
10°C
90
0.84
28.
57.
120.
11.
0°C
90
1.2
46.
92.
170.
19.
•'-Two significant figures are reported for the sole purpose of
allowing a better assessment of the effect of temperature on
the calculations. The estimates should not, however be con-
sidered significant to this extent in an absolute sense. One
significant figure or only an order of magnitude will be
reasonable for most chemicals.
70
-------
It is more likely that soils (in remote sites) would be in equilib-
rium with the air, and here the model predicts a soil concentration on
the order of 10 mg/kg. Unfortunately, there are no data from soil
analyses with which to test this prediction. Similarly, it is likely
that surface waters and suspended sediments in the real world would be
near equilibrium, and here the model predicts the ratio of concentrations
(sediment/water) to be about 30. Again there are no data with which to
check the prediction.
The numbers in Table 15 indicate that, for a system at equilibrium,
seasonal cycles involving temperature changes of about 20°C will change
the concentrations in all compartments, except air, by a factor of two.
The highest concentrations are associated with the colder temperatures.
3. Atmospheric Dispersion of Releases from Drycleaning Operations
As previously presented, the overwhelming source of atmospheric
emissions of PCE is the textile drycleaning industry. Within this
industry, however, there are three identifiable categories of dry-
cleaning operations, and the quantity of atmospheric emissions, and
the specific nature of the emission and dispersion characteristics of
each of these categories vary significantly. Commonly used analytic
and algorithmic methods for estimating ambient levels of an air contami-
nant were adapted and applied to model the fate of pollutant releases
from drycleaning plants in two of the industrial categroies. Estimates
have been made of both short-term (i.e., a few hours) concentrations in
the very near vicinity of a plant and long-term (i.e., annual average)
concentrations over a larger geographical area. In order to carry
out these calculations, several assumptions and estimations have been
necessary and these are discussed in the following paragraphs.
a. Development of Emission Source Parameters
In order to fulfill the input data requirements of the computer
program used to calculate long-term concentrations with respect to
emission source parameters, several information resources were reviewed.
The estimation of FCE emission rates was considered to be the most
important single factor in the use of the model. For the other param-
eters (exist gas temperature and volume in particular) reasonable
or conservative assumptions were made in conjunction with available
industry data or U.S. EPA reports from actual operating conditions at
specific drycleaning operations. To the extent possible, PCE emission
rates were calculated and compared for more than one set of throughput
estimates.
The method used to estimate PCE emission rates from the three
generic types of drycleaning establishments was based on national
statistics concerning PCE consumption as was presented previously.
An average emission rate of PCE for each type of establishment was
computed by dividing the total number of pounds of PCE released to
the atmosphere per year for each type of plant in the U.S. by the
71
-------
number of plants of that type reported to be in operation. The values
for this calculation were provided by an industry organization (W. Fischer,
IFI, 1980, private communication). Additional information from this
source made it possible to make assumptions about the normal operating
schedule of typical drycleaning plants of the various types. For the
large commercial type of facility, 13,750 plants had a total atmospheric
emission of 106.4 x 10° pounds/year. Large industrial plants (270
plants) had a total emission of 28.67 x 106 pounds/year. Performing
the calculation using the data and the assumed operating schedule data
given in Table 16 yielded the emission rates given in the table (in
units appropriate to the air quality models). Note that the emission
rate of PCS for the industrial area source and industrial point source
are the same; these two categories were defined to evaluate the effect
of differences in the other emission characteristics. Note also that
different assumptions about the operating schedule result in calculated
emission rates ranging between 1.5 g/second to 5.4 g/second. The value
of 3.6 g/second in the model for the industrial source was a reasonable
comprise.
A calculated value of the PCE emission rate for large commercial
drycleaning operations of 3.44 g/second was based on the industry survey
data given in Table 16. This was compared with the computed results for
PCE emission rates based on an EPA estimate of PCE loss of 4.1 kg
PCE/100 kg of laundry, and a daily throughput and operating schedule of
455 kg per facility per 12-hour day (Kleeberg and Wright, 1978; McCoy
1976; U.S. EPA, 1979a). This estimated value was 0.432 g/sec. or about
20% greater. The higher value was chosen as model input because it is
based on the results of a well-controlled, EPA-sponsored, comprehensive
source test and materials balance conducted at a large commercial dry-
cleaning establishment (McCoy, U.S. EPA 1979b). These specific test'
results have been used and cited by EPA in policy-making support docu-
ments (EPA, 1970; Kleeberg and Wright, 1978; McCoy, 1976). Furthermore,
the lower emission rate calculated from the industry statistics includes
a wide range of individual source characteristics, such as operator
competence, machine capacity, level of maintenance control/vapor recovery
systems, etc. This is not as likely to be true with respect to the large
industrial sources, since there are fewer sources and therefore less
variation among the sources is probable. Economic incentives for efficient
operation of an industrial source may be larger than for a commercial
operation as well.
^ Assessment Methodology and Values for Short-Term Concentration
Estimates "
Algorithmic treatment of short-term atmospheric dispersion is
usually structured by determining the elevation of the plume above
the ground surface and the horizontal and vertical dimensions of the
plume of emitted gases downwind of an idealized "point source" of
emissions (e.g., a chimney stack, or a roof-top vent). The plume di-
mensions are determined by a variety of meteorological parameters,
72
-------
TABLE 16. ATMOSPHERIC EMISSION PARAMETERS FOR THREE CATEGORIES OF DRYCLEANING
OPERATIONS USED IN ATMOPSHERIC DISPERSION MODEL
Source Description
Parameter
Emission Rate (g/sec)
Stack Height (m)
Side Width (m)
Exit Temperature (°C)
Exit Gas Velocity (m/sec) 2
Stack Diameter (m)
Industrial
Area Source
3.577
10
100
28
—
Large Commercial
Point Source
0.432
10
-
28
1.2
0.5
Industrial
Point Source
3.577
15
-
28
2.0
0.5
Emission rates were computed in general on the basis of the nation-
wide materials balance, as follows:
Emission Rate » (Total # Ibs lost to Air/yr)/(#plants)
(f/Operating hours) /year
for 270 industrial plants, we used a 52 wk/yr 6 days/wk
12 hr./day operating schedule, and a loss rate of
28.67 x 106 Ib/yr
for 18,750 commercial plants, we used a 52 wk./yr., 5 da3-T/wk.
8 hr/day schedule and a loss rate of 106.4 x 10° Ib/yr
The latter figures were adjusted (see text) to accommodate EPA data.
it gas velocities determined from typical values.
Industrial source: 825 ft. /min. through a 1.5-ft diameter vent
or duct.
Large commercial source: 500 ft^/min through same size duct.
-^Emission temperature of 85 °F.
73
-------
the emission characteristics of the source, and the relative geometries
of the emission source, and the specific locations at which concen-
trations are to be estimated.
The most widely used technique for estimating the dimensions of
the plume of emissions is to assume that the concentration distributions
in the horizontal and vertical dimensions are 'each Gaussian in nature,
and then to apply dispersion equations given by Turner (1970). The
principal dispersion equation given by Turner is:
X(x,y,z;H) =
exp - T
exp -TT
where X = ambient air concentration in yg/m
x,y,z = downwind, crosswind, and elevation coordinators
respectively for a given source - receptor geometry
cfy = horizontal plume dimension
az - vertical plume dimension
H = elevation of plume centerline above ground
surface (effective stack height)
u = wind speed
Q = emission rate (source strength) of air pollutant
in g/sec.
The equation yields estimated short-term (i.e., about 1-hour) concen-
tration levels, due principally to the level of resolution in the
meteorological data upon which the semiempirical parts of the dis-
persion equation were founded. Average concentrations for multiple,
consecutive hours can be determined by summation of individual hourly
estimates (which vary according to meteorological conditions), and
division by the number of estimates.
Several tools such as graphs and interpolation tables have been
developed to aid in the determination of individual hourly concentra-
tion estimates. For this particular study, graphs provided by Turner
(1970) were used to estimate ambient concentration for a specific
emission source strength (mass of pollutant emitted per unit time)
and ambient wind speed.
Applying this methodology to the "model" drycleaning plants des-
cribed in Table 16 at representative urban wind speed produced the
results shown in Table 17. This table shows some extremely high
values, particularly for stable atmospheric conditions: note, however,
that these estimates are for extreme, "worst-case" conditions for
very brief averaging times. Thus, Table 17 suggests a qualitative
74
-------
picture of potential worst-case ambient concentrations for very short
time periods. More significantly, it indicates the relative "spread"
of estimated maximum concentrations and the downwind distance of these
for different meteorological conditions. In terms of observed fre-
quency, the atmospheric conditions specified at the top and bottom
of Table 17 are much less common than the ones indicated in the middle.
Examination of figures provided by Turner (1970) showed that
under all atmospheric stability classes for emission sources such
as a typical commercial drycleaning plant, a relative concentration
isopleth of 10~3 m~2 extends at least 50m downwind. Applying the
commercial emission rate given in Table 16 gives, as a rough estimate,
a short-term concentration at this distance of about 150 + 50 ug/m3.
Similarly, in the immediate vicinity of an industrial drycleaner
emitting PCE at the rate given in the table, short-term concentrations would
be in the range of 900 to 1800
It is important to note that these concentration estimates are
representative of estimated levels for very short time periods. If
estimates for slightly longer time periods (such as 8, 12, or 24 hours)
were to be extrapolated from these values, PCE's persistence in air and
the dynamic fluctuations in wind direction and wind speed would have
to be considered. Also, variation in the operation of the emission
source would need to be considered, because, for example, an industrial
plant is likely to emit no PCE at all during certain periods of the day
and/or at night because of batch operation. In this context, it is
important to note that the emission rate given for the industrial
plant in Table 17 was based on a noncontinuous operation, specifically,
12 operating hours per day and 6 operating days per week. If the
industrial plant were assumed to operate continuously (24 hours/day,
7 days/week), the emission rate and the resulting "worst-case" ambient
concentration estimates would be reduced by a factor of 6x12 or 0.43.
7x24"
c. Methodology and Values for Long-Term Concentration Estimates
Rather than make 87600 hourly calculations of estimated concen-
trations at each one of a field of locations and determine annual
average concentrations for each, a computer program is used incorpo-
rating a meteorological joint frequency function. This tabulation
represents the normalized frequency of specific wind direction class,
wind speed class, and stability class joint occurrences. This is also
known as a "stability wind rose."
For this project a slightly modified version of the standard COM
dispersion model program was used (Busse and Zimmerman, 1973) . This
model has been developed by EPA, and is frequently used for estimating
long-term (seasonal or annual) quasi-stable pollutants concentrations0
at ground-level receptors using average emission rates point and area
sources. A stability wind rose appropriate for input to COM was
readily available (due to other ongoing projects), and this tabulation
was for a major urban area in the Northeastern United States. The
rose aggregated over all stability classes is shown in Figure 16.
75
vino
-------
TABLE 17. DISPERSION MODEL PARAMETERS AND ESTIMATED PCE CONCENTRATIONS IN AIR
NEAR INDUSTRIAL AND COMMERCIAL SOURCES
Meteorological Conditions
Maximum PCE
CT>
Atmospheric
Stability
Class
Highly Unstable
Unstable
Slightly Unstable
Neutral
Slightly Stable
Stable
Wind Speed
(m/sec. )
1.5
2.5
3.0
3.5
3.5
4.5
7.0
2.5
3.0
4.0
1.5
2.5
Effective
Stack
Height (m)
20
20
17
15
17
15
10
! ' 1
17
15
10
15
10
Industrial
Sources
720
475
600
648
411
432
668
576
624
1125
1128
1656
"1-0 1 — '
Large
Commercial
Source
86.4
57.0
72.0
77.8
49.4
51.8
80.2
69.1
74.9
135.0
135.4
198.7
Downwind Distance
of Maximum Concen-
trations (km)
~0.1
0.14
0.16
0.145
0.30
0.26
0.21
0.43
0.37
0.22
0.62
0.37
-------
N
e-» *-. r-
.11-3.1.
kts
FIGURE 16 STABILITY WIND ROSE FOR NIAGARA FALLS USED IN LONG-TERM
PCE CONCENTRATION ESTIMATES FOR COMMERCIAL AND
INDUSTRIAL DRYCLEAIMING OPERATIONS
77
-------
The emission source data used in this modeling was that given in
Table 16. The network of receptor locations was chosen to give adequate
spatial coverage and satisfactory resolution of concentration values.
As an aid to interpretation of the model results, a computerized graphics
system made possible the preparation of contour or isopleth naps shown
as Figures 17 and 18. These figures shox* the concentration pattern
determined by the CDM model for the industrial drycleaning plant. Figure
18 is an enlarged view in the immediate vicinity of the plant location.
The CDM program permits the estimation of ambient impacts from
emission sources which do not conform to the usual parameterization of
point sources. These types of sources are known as area sources and an
example of one would be a low one or two story building with roof to
vents or a horizontal duct venting to the atmosphere. In addition, area
sources typically have relatively cool low gas volume emissions to the
atmosphere, so that their emissions have relatively little plume rise
due to buoyancy forces (which are caused by the temperature gradient
between the atmosphere and the exhaust gases) and momentum forces (which
are caused by the mass flux and exit velocity of the exhaust gases).
The model run described below analyzed point and area source charac-
terizations of the model drycleaning plants, as described in Table 16.
The results of the initial model runs provide an indication of
annual average ambient levels of PCE that could be expected in the
vicinity of a drycleaning plant. The ambient impact from the "average"
sized plant appeared to have a negligible effect over the long-term.
In other words such sources do not appear to have an effect on the
average observed background level, as indicated by monitoring programs.
Similarly, a large commercial operation was responsible for a maximum
concentration of less than 6.8ug/m3 at a downwind distance of 500m.
This level too is lower than annual averaged background levels. The
maximum long-term concentrations of PCE near an industrial plant were
about the same, whether it was treated as a point source or as an area
source. The maximum value was about 27 ug/m^, and was predicted at a dis-
tance of 500 m from the source also.
On the initial grid of receptor locations, the maximum concentration
estimates for all three model source types, were found for receptors
close to the source. Thus it was suspected that still higher concen-
trations might be determined by the model if receptors were placed
closer to the sources than 500 m. Therefore, a second set of model
runs as obtained for a receptor grid having a finer spatial resolution
around the emission source and including receptor locations as close
as 60 m from sources.
78
-------
FIGURE 17 SIMULATED PCE CONCENTRATION ISOPLETHS FOR AN INDUSTRIAL POINT SOURCE
79
-------
FIGURE 18 DETAIL OF SIMULATED PCE CONCENTRATION ISOPLETHS FOR AN INDUSTRIAL SOURCE
80
-------
As expected, significantly higher concentrations were estimated for
the receptors close to the emission source. The highest concentration in
the immediate vicinity of the commercial plant (modeled as a point source)
was estimated at about 20 ug/m3 at a distance of 100 m. Maximum average
concentrations due to the industrial plant, modeled both as an area
source and as a point source, were also estimated to occur at approxi-
mately the same distance. For the industrial plant treated as a point
source, the maximum concentration was determined to be 136 jig/m-*, while the
area source model indicated maximum concentrations of over 920 (j.g/m^. The
validity of the area source dispersion algorithm at such close distances
is questionable, however, because of computational instabilities in the
coding and execution of the algorithm. In reality, the ambient impact
of the source, if more properly characterized as an area source, would
probably be lower at this distance than as estimated by the point source
algorithm.
4. EXAMS Concentration Estimates
a. Introduction
The EXAMS model has been developed by EPA Athens Environmental
Research Laboratory in order to help assess the behavior of a pollutant
in various characteristic aquatic systems. The output of the model
includes:
(1) simulations of steady-state concentrations and pollutant
mass distribution among water, sediment and biota in
different compartments (e.g. water column, bed sediment),
(2) percentage of system loading removed by each chemical
and biological kinetic process, and
(3) concentration die-away time following cessation of discharge.
As input, the model requires the pollutant's physicochemical properties,
environmental reaction rate constants, and loading rate to the system.
The assumptions of the model include a continuously discharging source
at a constant level, a box of water made up of a system-defined number
of well-mixed compartments, and first-order rate kinetics in all processes,
A more thorough discussion of the model is given in Lassiter, et al.
(1978) and Baughman and Burns (1980).
Three pathways are modeled using the EXAMS.
1. Metal Degreasing Effluents -Surface Water ( -Air)
2. Landfill Leachate -Groundwater -Surface Water ( -Air)
3. Drycleaning and other Discharges -POTW's Surface Water
( -Air)
81
-------
The model does not give air concentrations resulting from the introduction
of a given amount of PCE to the water column, but because the model ac-
counts for volatilization, the air compartment is shown in parentheses
as it completes the pathway.
The four scenarios modeled by EXAMS were as follows:
1. Metal Degreasing; 10,000 kg/yr direct aquatic discharge from
2,188 plants. Lacking data on effluent PCE concentrations
and effluent flow rates, the loading rate was estimated by
apportioning the discharge equally between the plants and
assuming a 6 hour/day period over which the discharge would
4.
occur,
• Loading rate = 0.004 kg/hr as discharge
• Type of receiving water bodies: rivers - clear
and turbid
2- Landfill Leachate I; amounts of PCE involved are unknown.
Levels measured in contaminated wells range between 0.4 mg/1
and 10 mg/1; thus these values are assumed to be indicative
of those in some leachate plumes from landfills,, Based on
a study of leachate plumes in Long Island (Kimmel and
Braids, 1980), the plume width varied from 0.6 km (the
width of the landfill) to 0.2 km, at which point the plume
intersected a stream. To develop a worst case scenario
we postulated a highly permeable 3.85 x 10* liter/day/m2
(105 gal/day/ft2) sand aquifers which intersects a water
body fairly close to the landfill. The cross section at
intersection is 500 m x 2 m and the metric daily discharge
is 4.075 x 109 liters. A concentration of PCS in the
leachate of 1.0 mg/1 yields the following load:
• Loading rate = 160 kg/hour as groundwater interflow
• Water bodies: river and turbid river, pond, lake
3* Landfill Leachate II; a less drastic case of leachate con-
tamination of surface water would involve a less permeable
aquifer (102 gal/day/ft2), a lower concentration of PCE
(0.2 mg/1) and an intersection xjith the x^ater body further
from the source,which reduces the cross sectional area of
the plume (120 x 1 m).
• Loading rate = 0.004 kg/hour, as interflow
• Water bodies: same as scenario 2, above.
Drycleaning and Other Discharges to POTW's; while the amount
of PCE which is treated by POTW's has been approximated,
the exact input to a POTW is unknown. This scenario is
82
-------
based on a POTW of average size (3 MGD) with effluent
PCE concentrations on the high side (10 ug/1) of monitored
values, thus a loading rate can be estimated.
• Loading rate = 0.004 kg/hr, as discharge
• Water bodies: river and turbid river
• Results correspond to those of scenario 1
The modeling results were based on use of the measured value for
the ratio of rate of volatilization of PCErrate of volatilization of
02. This relationship is given as:
K . - Kn ^02 . 0.44
vol Oo
Because the measured value (Smith, et al. , 1980) of 0.52 + 0.09 agreed
closely with the theoretical value, we opted to use the measured value
for the input parameter.
b. Results
Two runs were made to analyze scenarios 1 and 4, whose loading
rates, quite accidentally, were about equal. The model specification
for scenario 2 and 3 called for the loading to occur through ground-
water flow to surface waters. However, EXAMS only accepts discharges
to surface waters directly and scenarios 2 and 3 were modeled in this
manner instead, to at least approximate the impact of the high loading
rate on the five ecosystems.
The key results of all modeling efforts are shown in Table 18, which
shows applied and actual loading rates, maximum concentrations in the
water column, the percent of the load which volatilizes, and system self-
purification times. There is a difference between applied (input) and
actual (level accepted by EXAMS) for hydrologically closed systems —
ponds and lakes. This difference arises only when the applied load is
above the capacity of the system to dissipate it. Thus it shows the
loading rate at which all inputs are either dispersed, degraded or
otherwise transported. The excess load in this case, would "puddle"
at the bottom of the water column as PCE's specific gravity is greater
than that of water.
These are clear differences in the way these ecosystems respond
to inputs of PCE.
1. Ponds and lakes require from 70 to 210 days to purify themselves
while river systems needed between 61 and 132 hours. The
hydraulic retention time is the controlling variable.
2. The amount of organic material present in the system may be an
important factor as there are differences between the eutrophic
83
-------
TABLE 18. SUMMARY OF RESULTS OF EXAMS MODELING OF PCE CONCENTRATIONS
IN AQUATIC ECOSYSTEMS
oo
Ecosystem
River
Turbid River
Pond
Eutrophic Lake
Oljgotrophic Lake
Applied
Load
(kg/hr)
0.004
160
0.004
160
0.004
160
0.004
160
0.004
160
Actual
Load
(kg/hr)
0.004
160
0.004
160
0.004
2.4 -
0.004
58.?
0.004
58.2
Maximum
Concentration
in Water
Column (wg/1)
0.004
160
0.004
158
10
6200
0.52
7600
0.55
8100
Maximum
Concentrations
in Sediments
(Wg/kg)
0.03
1200
0.02
810
460
270,000
2.0
29,000
0.48
7100
Percent of
Load which
Volatilizes1
1.6
1.6
1.6
1.6
93.2
93.2
95.8
95.8
95.2
95.2
Percent
in Water
Column?
91.8
91.8
94.9
94.9
40
40
97
97
99.3
99.3
System Self-
Purification
Time
131.8 hours
129.8 hours
63.4 hours
61.3 hours
210 days
210 days
74.6 days
74.6 days
70.6 days
70.6 days
For PCE this percentage plus the percent of the load which is transported, add up to 100%
The percent of PCE in the water column and the percent of PCE in bottom sediments add up to 100%.
-------
and oligotrophic lake, and the turbid vs. the clear rivers.
There are insufficient data here to analyze this further, and
in any case, the lack of a complete input data base for PCE
renders these results "indicative" rather than "conclusive."
3. Downstream advection is the critical removal mechanism in short
river reaches, while volatilization is the most important factor
in ponds and lakes and in long river reaches.
4. The maximum concentrations of PCE in water and sediment are
a function of loading rates. Water concentrations ranged
between 0.004 ug/1 and 8.1 mg/1. These numbers are less than
most positive monitoring data. Sediment concentration maxima
ranged between 0.02 ug/kg and 270 mg/kg, which correspond well
with the values in Table 11.
5. Only in the pond scenario does the percent of PCE in
sediments exceed the percent in the water column (60%
and 40%,respectively) while in all other scenarios the
water column holds over 92% of the total accumulation.
Use of EXAMS, even with crudely estimated loading rates,has pro-
vided simulated concentrations that are believable when compared with
monitoring data. Because the input data to characterize the full range
of behavioral mechanisms of PCE in aquatic systems are lacking, it would
be misleading to draw further conclusions from the observations made
above.
F. SUMMARY
Physicochemical Properties - Tetrachloroethylene is a volatile liquid
and a relatively stable chemical. It is readily degraded by photochemi-
cal reactions in the atmosphere (and to a much smaller extent in sur-
face waters), but it is resistant to hydrolysis and biodegradation.
Phosgene, a highly toxic chemical, is one of the degradation products
from the photochemical reactions in air. Adsorption on soils and
sediments, and bioconcentration in aquatic biota, will take place
(concentration factors are about 30 and 100, respectively), but not
to the extent that food chain contamination is of concern.
Mobility and Persistence - Tetrachloroethylene is very mobile in the
environment. Atmospheric transport can carry the chemical hundreds to
thousands of kilometers downwind from the original emission sources.
The atmospheric residence time is estimated to be in the range of 2-10
days in sunlight. Transport through soils to deep wells has resulted
in water supply contamination at a number of sites. In deep soils and
groundwaters, as well as in deep surface waters and sediments, the
chemcial may have a residence tine of 7-14 years, unless the
turnover time is smaller for the compartment. Significant amounts
of the chemical can be transported in water (solubility is -100 mg/1
at 20°C). In well-mixed surface waters volatilization,will be an
85
-------
Ambient Levels - The entire atmosphere, all surface waters, and topsoils
will contain some tetrachloroethylene. Background concentrations are
however, below 1 yg/1 or yg/kg. Most surface waters in industrialized
or heavily populated river basins will show elevated (>1 yg/1) levels
of the chemical. Drinking water in the U.S., however, typically contains
<0.2 yg/1. Elevated levels in the atmosphere (6.8 yg/m3) 'are found
near manufacturing and user sites, near land disposal sites (where
waste solvent, solvent sludges, or POTW sludges are usually sent), and
are likely to be found near sewer vents in certain commercial/industrial
sectors. Concentrations above 10 yg/1 have been found in the leachate
from some dump sites. Aquatic biota, food, and humans contain 1-20
yg/kg of the chemical
Equilibrium Distribution Model - A relatively simple equilibrium distrib-
ution model was used to predict concentrations in various environmental
oo a°mPrrtmentS* ThS results indicate that - at equilibrium - about
99. 8xi of PCE would reside in the atmosphere. The predicted atmospheric
concentration of 0.6 yg/m3 (v/v) is fairly close to the ranges measured:
-------
EXAMS Water Distribution Model - The EPA's standard computerized model
for the determination of chemical fate in aquatic ecosystems was used
to analyze four scenarios for five ecosystems. The scenarios were:
industrial discharges to rivers at a loading rate of 0.004 kg/hr, POTW
discharges to rivers at the same rate, and modeling all ecosystems for
input in the form of contaminated groundwater which included a leachate
plume from landfills containing PCE wastes. The leachate plume contami-
nation scenario was modeled at two loading rates, 0.004 and 160 kg/hr,
the latter representing an extreme situation. The modeling results led
to the following conclusions:
1. Ponds and lakes (systems with long hydraulic retention times)
require far longer times than rivers for 99% reduction of PCE
following the cessation of loading.
2. Volatilization is the primary mechanism for PCE reduction from
the aquatic environment, provided that aqueous concentrations
are sufficient to drive that process.
3. The maximum concentrations of PCE in the water column and in
sediments are a function of the loading rate and the presence of
organic matter as sediment or as suspended solids (although
this latter relationship has not been quantified due to lack
of data). Concentrations in water were between 0.004 ug/1 and
7.6 mg/1. Sediment levels were in the range of 0.2 yg/kg and
270 mg/kg.
4. The water column generally accumulates more than 92% of the total
PCE in the aqueous systems analyzed, with the exception of the
pond where the sediments contained 60% of the accumulated PCE.
87
-------
REFERENCES
Antropov, L.I., V.E. Pogulyai, V.D. Simonov, and T.M. Shamsutdinov.
1972. Solubility in chlorocarbon-water systems, (English translation).
Russian Journal of Physical Chemistry 46(2);311-312.
Baughman, G.L., and L.A. Burns. 1980. "Transport and transformation
of chemicals in the environment," In: 0. Hutzinger (ed.) Handbook
of Environmental Chemistry (in preparation). Springer Verlag, New
York.
Brass, H. 1981. Final Community Water Supply Study (CWSS) Data
Memo. Technical Support Division, Office of Drinking Water, U.S.
Environmental Protection Agency. Cincinnati, OH.
Brown, D., and Phil, D. March 1978. Chlorinated solvents in sewage
works. Effluent and Water Journal, page 110.
Burns and Roe. 1979. Preliminary Data for POTW Studies. Effluent
Guidelines Division, U.S. Environmental Protection Agency.
Busse, A.D., and J.R. Zimmerman. 1973. User's guide for the climato-
logical dispersion model. EPA-RA-73-204. Office of Research and
Development, U.S. Environmental Protection Agency.
Chiou, C.T., V.H. Freed, D.W. Schmedding, and R.L. Kohnert. 1978.
Partition coefficient and bioaccumulation of selected organic chemicals.
Environ. Sci. Technol. 11(5) ; 475-478.
Dilling, W.L., N.B. Terfertiller, and G.J. Kallos. 1975. Evaporation
rates and reactivities of methylene chloride, chloroform, 1,1,1-trichloro-
ethane, trichloroethylene, tetrachloroethylene, and other chlorinated
compounds in dilute aqueous solution. Environ. Sci. Technol. 9(9):
833-838, ibid., 10:1275 (1976). ~
Dilling, W.L. 1977. Interphase transfer process. II. Evaporation
rates of chloro methanes, ethanes, ethylenes, propanes, and propylenes
from dilute aqueous solutions. Comparisons with Theoretical Predictions.
Environ. Sci. Technol. _11(4): 495-409.
Dilling, W.L., C.J. Bredeweg, and N.B. Terfertiller. 1976. Organic
photochemistry, simulated atmospheric photodecomposition rates of
Methylene Chloride, 1,1,1-trichloroethane, trichloroethylene, tetra-
chloroethylene and other compounds. Environ. Sci. Technol. 10(4):
351-356. —
Ewing, B.B., E.S.K. Chian, J.C. Cook, C.A. Evans, and P.K. Hopke. 1977.
Monitoring to detect previously unrecognized pollutants in surface
waters. Report EPA-560/6-77-015, U.S. Environmental Protection
Agency, Washington, D.C.
88
-------
Feller, H. 1979. Fate of priority pollutants in publicly owned
treatment works. Report EPA-440/1/79-300, U.S. Environmental
Protection Agency, Washington, D.C.
Franklin Institute. 1975. Preliminary study of selected potential
environmental contaminants: Optical brighteners, methyl chloroform,
trichloroethylene, tetrachloroethylene, ion exchange resins. Report
EPA-560/2-75-002, Office of Toxic Substances, U.S. Environmental
Protection Agency, Washington, D.C. (OTIS PB 243 910).
Fuller, B.B. 1976. Air pollution assessment of tetrachloroethylene.
Report by Mitre Corporation, Technical Report No. MTR-7143, the U.S.
Environmental Protection Agency, Washington, D.C. (NTIS PB 256 731).
Jarema, R. 1977. Plainville and Plainfield's plight with pollution.
Unpublished paper prepared for the Connecticut State Department of
Health.
Jensen, S., and R. Rosenberg. 1975. Degradability of some chlorinated
aliphatic hydrocarbons in seawater and sterilized water. Water Res
9/7):659-661.
Jones, C.J., B.C. Hudson, P.J. McGugan, and A.J. Smith. 1977/78. An
investigation of the evaporation of some volatile solvents from domestic
waste. J. Hazardous Materials .2:235-51.
Kleeburg, D.F., and J.G. Wright. 1978. Control of volatile organic
emissions from perchloroethylene drycleaning systems. Report No
EPA-450/2-78-050, OAQPS, U.S. EPA.
Lapp, T.W., B.L. Herndon, C.E. Mumma, and A.D. Tippit. 1977. An
assessment of the need for limitations on trichloroethylene, methyl
chloroform, and perchloroethylene. (3 volumes). Draft Final Report
by Midwest Research Institute, Kansas City, MO, to the Office of Toxic
Substances, U.S. Environmental Protection Agency.
Lassiter, R.R., G.L. Baughman, and L.A. Burns. 1978. "Fate of toxic
organic substances." In:State of the Art in Ecological Modelling Vol. VII
International Society for Ecological Modelling, Copenhagen, Denmark.
Levins, P., K. Thurn, G. Harris, and A. Wechsler. 1979a. Sources of
toxic pollutants found in influents to sewage treatment plants. II.
Muddy Creek Drainage Basin, Cincinnati, Ohio. Final Report by A.D
Little, Inc., on Task Order No. 6, EPA Contract No. 68-01-3857, for
U.S. Environmental Protection Agency, Washington, D.C.
Levins, P., J. Adams, P. Brenner, S. Coons, K. Thurn, and J. Varone.
1979b. Sources of toxic pollutants in influents to sewage treatment
plants. III. Coldwater Creek Drainage Basin, St. Louis, Missouri
Final Report by A.D. Little, Inc., on Task Order No. 7, EPA Contract
No. 68-01-3857 for U.S. Environmental Protection Agency, Washington, D.C.
89
-------
Levins, P., J. Adams, P. Brenner, S. Coons, K. Thrun, and J. Varone.
1979c. Sources of toxic pollutants found in influents to sewage treat-
ment plants. IV. R.M. Clayton Drainage Basin, Atlanta, Georgia.
Final Report by A.D. Little, Inc., on Task Order No. 13, EPA Contract
No. 68-01-3857 for U.S. Environmental Protection Agency, Washington,
D.C.
Levins, P. J. Adams, P. Brenner, S. Coons, C. Freitas, K. Thrun, and
J. Varone. 1979d. Sources of toxic pollutants found in influents
to sewage treatment plants. V. Hartford Water Pollution Control
Plant, Hartford, Connecticut. Final Report by A.D. Little, Inc.,
on Task Order No. 13, EPA Contract No. 68-01-3857 for U.S. Environ-
mental Protection Agency, Washington, D.C.
Lillian, D., H.B. Singh, A. Appelby, L. Lobban, R. Arnts, R. Gumpert,
R. Hagne, J. Toomey, J. Kazazis, M. Antell, D. Hansen, and B. Scott.
1975. Atmospheric fates of halogenated compounds. Environ. Sci. Technol.
2(12):1042-1048.
Mackay, D. 1979. Finding fugacity feasible. Environ. Sci. Technol.
13(10):1218-1223.
Massachusetts Department of Environmental Quality Engineering. 1980.
"Tetrachloroethylene Update," August, Boston, MA.
McConnell, G., D.M. Ferguson and C.R. Pearson. 1975. Chlorinated
hydrocarbons and the environment. Endeavour 34(21);13-18.
McCoy, B.C. 1976. Study to support new source performance standards
for the dry cleaning industry. Report No. EPA-450/3-76-029, U.S.
Environmental Protection Agency.
National Academy of Sciences (NAS). 1975. Assessing potential ocean
pollutants. Washington, D.C. CSee especially pp. 109-110 and
references cited therein.)
National Institute for Occupational Safety and Health (NIOSH). 1976.
Criteria for a recommended standard... Occupational exposure to tetra-
chloroethylene. Report PB-266 583, Public Health Service, U.S. Depart-
ment of Health, Education and Welfare.
Neely, W.B. 1976. "Predicting the flux of organics across the air
water interface," In:Control of Hazardous Material Spills, Proceedings
of the 1976 National Conference, April 25-28, 1976; New Orleans, LA.°
Neely, W.B., D.R. Branson and G.E. Blau. 1974. Partition coefficient
to measure bioconcentration potential of organic chemicals in fish.
Environ. Sci. Technol. 8_(13): 1113-1115.
Newell, I.L. 1976. Naturally occurring organic substances in surface
waters and effect of chlorination. J.N. Engl. Water Works Assoc. 90
(4) .-315-340. —
90
-------
Pearson, C.E., and G. McConnell. 1975. Chlorinated C]_ and G£ hydro-
carbons in the marine environment. Proc. R. Soc. London, Ser. B.
189;305-332.
Pellizzari, E.D., M.C. Erickson, and R.A. Zweidinger. 1979. Formulation
of a preliminary assessment of halogenated organic compounds in man and
environmental media. U.S. EPA, Washington, B.C.
Ross, R.H., L.B. Yeatts, Jr., E.3. Lewis, G.A. Dailey, D.S. Harnden,
B.C. Michelson and L.M. Frogge. 1977. Environmental and health aspects
of selected organohalide compounds: An information overview. Braft
report by Oak Ridge National Laboratory, Oak Ridge, TN, to the National
Science Foundation.
Schwarzenbach, R.P., E. Molnar-Kubica, W. Giger, and S.G. Wakeham. 1979.
Bistribution, residence time, and fluxes of tetrachloroethylene and
1,4-dichlorobenzene in Lake Zurich, Switzerland. Environ. Sci. Technol.
13(11):1367-1373.
Simmonds, P.G., S.L. Kerrin, J.E. Lovelock, and F.H. Shair. 1974.
Bistribution of atmospheric halocarbons in the air over the Los Angeles
Basin. Atmos. Environ. _8:209-216.
Singh, H.B., L.J. Salas, H. Shigeishi, and E. Scribner. 1979. Atmos-
pheric halocarbons, hydrocarbons and sulfur hexafluoride: Global
distribution, sources and sinks. Science 203:899:903.
Singh, H.B., L.J. Salas, A. Smith, R. Stiles, and H. Shigeisha. 1980.
Atmospheric measurements of selected hazardous organic chemicals. Grant
805990, U.S. EPA, Research Triangle Park, NC.
Smith, J.H., B.C. Bomberger and B.L. Haynes (SRI, Inc.). 1980. Pre-
diction of the volatilization rates of high volatility chemicals from
natural water bodies. Environ. Sci. Technol. (in press).
Tabak, H.H., S.A. Quave, C.I. Mashni, and E.F. Earth. 1980. Bio-
degradability studies with priority pollutant organic compounds. Staff
Report (Braft), Water Research Bivision, U.S. EPA, Cincinnati, OH.
Touhill, Suckrow and Associates, Inc. 1979. Phase II - Technology
evaluation approach. Interim Report on Contract No. 68-03-2766 with
U.S. EPA, Cincinnati, OH.
Turner, B.B. 1970. Workbook of atmospheric dispersion estimates.
Report PB-191-482, Public Health Service, U.S. Bepartment of Health,
Education and Welfare.
U.S. Coast Guard. 1974. CHRIS, Hazardous Chemical Bata. Report CG-
446-2, U.S. Bepartment of Transportation.
91
-------
U.S. Environmental Protection Agency (U.S. EPA). 1975. Preliminary
Assessment of Suspected Carcinogens in Drinking Water. U.S. EPA,
Washington, B.C.
U.S. Environmental Protection Agency (U.S. EPA). 1977. Ambient water
quality criteria. Criterion document - Tetrachloroethylene. Interim
Draft No. 1 (14 December), Criteria and Standards Division, Office of
Water Planning and Standards, Washington, D.C.
U.S. Environmental Protection Agency (U.S. EPA). 1978. The national
organic monitoring survey. Technical Support Division, Office of Water
Supply, Washington, D.C.
U.S. Environmental Protection Agency (U.S. EPA). 1979a. Perchloro-
ethylene dry cleaning - background information for Proposed Standards
Report EPA-450/3-79-029a, Emission Standards and Engineering Division.
U.S. Environmental Protection Agency (U.S. EPA). 1979b. Water Quality
Criteria for Tetrachloroethylene. Federal Register 44:15966.
U.S. Environmental Protection Agency (U.S. EPA). 1980a. Volatile
Organics in Drinking Water. Criteria and Standards Division, Washington,
D.C.
U.S. Environmental Protection Agency (U.S. EPA). 1980b. STORET.
Monitoring and Data Support Division, Office of Water Planning and
Standards, Washington, D.C.
Verberk, M.M., and T.M.L. Scheffers. 1980. Tetrachloroethylene in
exhaled air of residents near dry cleaning shops. Environ. Res. 21:432.
Versar, Inc. 19SO. Materials Balance Report - Tetrachloroethylene.
Wakeham, S.G., A.C. Davis, R.T. Witt, B.W. Tripp, and N.M. Frew. 1980.
Tetrachloroethylene contamination of drinking water by vinyl-coated
asbestos-cement pipe. Bull. Environ. Contam. Toxicol. 25:639.
Walter, P., A. Craigmill, J. Villaume, S. Sweeny and G.L. Miller. 1976.
Chlorinated hydrocarbon toxicity: 1,1,1-trichloroethane, trichloro-
ethylene, and tetrachloroethylene. Report No. CPSC-BBSC-76-MI; PB 257
185, Consumer Product Safety Commission, Bureau of Bioinedical Science,
by Franklin Institute Research Labs, Philadelphia, PA.
92
-------
CHAPTER V
EFFECTS OF TETRACHLOROETHYLENE
ON HUMANS AND AQUATIC BIOTA
A. HUMAN TOXICITY
1. Introduction
Exposure to the chlorinated hydrocarbon tetrachloroethylene has
been associated with various hepatotoxic and nephrotoxic effects ; de-
pression of the central nervous system; and eye, nose and throat irri-
tation. A potential carcinogenic risk to man has also been suggested
by the occurrence of hepatocellular carcinoma in mice given PCE by
gavage. A discussion of these effects associated with PCE exposure
follows .
2. Metabolism and Bioaccumulation
Evidence concerning absorption of PCE is predominantly limited to
the inhalation route, although some investigation of dermal absorption
has been conducted. Ogata and co-workers (1971) estimate that approxi-
mately 57% of inhaled PCE is retained by humans. In a series of studies,
Monster (1979 a,b,c) noted total uptake of PCE was influenced more by
(lean) body mass than by respiratory minute volume of adipose tissue
(i.e. distribution volume is larger when body weight is higher).
Retained PCE is primarily eliminated unchanged in expired air
(Stewart .et. al. , 1961; Ideda, 1977). Monster (1979 a,b,c) noted 80-100%
elimination of retained PCE in the expired air of six male volunteers by
162 hours following PCE exposure; an additional 2% of the PCE uptake was
converted to trichloroacetic acid and subsequently eliminated in the
urine. The time course of PCE concentration in blood and exhaled air
showed that a long period was necessary to complete elimination of PCE.
A respiratory half -life of 65 hours has been estimated for PCE in man; the
urinary half-life for its metabolite, trichloroacetic acid, is somewhat
longer (144 hours) (Ikeda, 1977). Thus, an accumulation of PCE in the
body would occur with repeated exposure. McConnell et_ al. (1975) reported
some evidence of PCE in human tissues at extremely low concentrations
(U
With respect to dermal exposure, Riihimaki and Pfaffli (1973) found
that in ambient air, concentrations of88.5mg/m^ PCE readily penetrated
human skin; the small number of human volunteers precluded conclusive
quantification of PCE absorption. In another study, Stewart and Dodd
(1964) reported the presence of 160-260 pg/mj in the expired air of human
volunteers 5 hours after immersing their thumbs in PCE for 40 minutes.
93
-------
Schumann and Watanabe (1979) recently reported that in B6C3F1 mice
exposed to 1.47 mg/mj of 14C-tetrachloroethylene for 6 hours, 63% of the
radioactivity was excreted in the urine as nonvolatile metabolites.
An additional 12% of the radioactivity was excreted unchanged in expired
air. Similar exposure of Sprague-Dawley rats, however, resulted in the
elimination of only 19% of the radioactivity in urine, with an additional
68% of the dose eliminated in expired air. The mouse metabolized 7-8
times more PCE per kilogram of body weight than did the rat, with approxi-
mately 7-9 times more radioactivity irreversibly bound to hepatic macro-
molecules in the mouse than in the rat. No radioactivity was detected
bound to purified hepatic DNA at times of peak macromolecular binding in
the mouse.
3. Animal Studies
a. Carcinogenicity
The carcinogenicity data indicate that PCE is an apparent liver car-
cinogen in the mouse; the data in rats are inconclusive. USP-grade PCE was
administered by gavage to groups of 50 male and female B6C3F1 mice at pre-
determined maximally tolerated dose and one-half this amount 5 days per
week for 78 weeks. The dose of PCE was changed during the course" of
the experiment in order to prevent excessive loss of animals; time-
weighted average doses were 386 and 772 mg/kg* for females and 536 and
1072 mg/kg for males. The vehicle control group consisted of 20 mice
of each sex.
Hepatocellular carcinoma was found in 40% to 65% of all treated
mice compared with 0-10% incidence in controls (see Table 19). Five
hepatocellular carcinomas metastasized to the lung (NCI, 1977).
TABLE 19. INCIDENCE OF HEPATOCELLULAR CARCINOMA IN PCE-TREATED
B6C3F1 MICE
2/20
32/49
27/48
Male
(10%)
(65%)
(56%)
Female
p<.001
p<.001
0/20
1,9/48
19/48
( 0%)
(40%)
(40%)
p <. 001
p<.001
Vehicle Controls
Low PCE Dose
High PCE Dose
Source: NCI (1977)
*Note that all references to dose (given as mg/kg) are dose/kilogram
body weight.
94
-------
A concurrent study conducted with groups of 50 Osborne-Mendel rats
given time-weighted doses of 941 mg/kg and 471 mg/kg for males and 949
and 474 mg/kg PCE for females by gavage produced no significant increase
in neoplastic lesions. However, the high incidence of dose-related earlv
deaths confound the interpretation of these results. Fifty percent of "
the high dose males died by week 44 and 50% of the high-dose females
were dead by week 66 compared with a median survival time of 88 and 102
weeks, respectively, for male and female control rats. Toxic nephropathv
was observed in 79% of the treated rats but in none of the controls. Due
to poor survival and the poor response (< 5% hepatocellular carcinoma)
of this strain to the positive control, carbon tetrachloride, the re-
sults of this carcinogenicity bioassay were considered inconclusive (NCI,
-L.7 / / / •
Recently, NIOSH (1973) reported no statistically significant increase
in tumors in Sprague Dawley rats exposed to-44.3or 88.5mg/m3 PCE by in-
halation (the study duration was not given). A higher, but not statis-
tically significant incidence of adrenal pheochromocytoma was noted in
females^ at the44.3mg/m3 level only and increased mortality occurred ir
the 88. 5 mg/m-i treated males.
Thus, a single positive finding of liver carcinoma in 36C3F1 mice
has been linked to PCE exposure. The lack of a dose-response and the
high effect level, however, suggest that the observed tumorigenesis
may be secondary to rapid cellular proliferation and liver regeneration
resulting from a partial chemical hepatectomy induced bv toxic levels of
PCE.
b. Mutagenesis
Negative findings were reported in two bacterial mutagenicity assays
with PCE. Bartsch et al. (1979) noted negative results with two strains
of Salmonella typhimurium (TA100, TA1530) exposed to PCE vapor in che
presence of mouse liver microsomal activation.
In a second study, a concentration of 149 mg/1 PCE was not mutagenic
when tested with Escherichia coli K12 in the presence of a liver micro-
somal fraction (Greim e_t al. , 1975).
Cema and Kypenova (1977), however, found increased mutagenic ac-
tivity in Salmonella typhimurium TA100 without metabolic activation at
0.01-1 mg/ml concentrations. These investigators also reported positive
results in a host-mediated assay in mice with S. typhimurium strains
TA1950, TA1951, and TA1952 at 1/2 LD and LD levels. No evidence of
aose-dependence was seen. -)U
In another study, Price and co-workers (1978) reported that PCE
(97 VM) induced phenotypic transformations of F1706 rat embryo cells [(178
SCo7 ^hSS VS 124 f°Ci f°r the P°sitive control, 3-methylcholanthrene
(U.J7yM;j. Isogenic Fischer rats inoculated subcutaneously with trans-
formed cells produced undifferentiated fibrosarcomas in all seven rats
within 45 days of inoculation.
95
-------
Although the mutagenic findings are varied, the strong positive results
in mammalian cells supported by positive results in the host-mediated
assay implicate PCE as a mutagen. Further investigations are needed to
clarify the mixed response in bacterial systems.
c. Teratogenesis
Administration of 44.3 mg/m PCE by inhalation to pregnant Swiss
Webster mice 7 hours per day on days 6-15 of gestation produced a sig-
nificant decrease in fetal body weight (1.19 g vs. 1.3 g for controls)
and significantly greater incidences of subcutaneous edema (59% vs. 27%
for controls), delayed ossification of skull bones (100% vs. 69% for
controls) and split sternebrae (24% vs. 0% for controls). Similarly
treated Sprague Dawley rats exhibited a light but significant decrease
in maternal weight gain and a significant increase in the percent of
fetal resorptions (9% vs. 4% for controls) (Schwetz et al., 1975).
Nelson (1979) found that pregnant Sprague Dawley rats exposed to
133 mg/m3 PCE, 7 hours per day on days 7-13 or 14-20 of gestation had
a significant reduction in the proportion of pups born alive (no values
given). Behavioral tests indicated poorer initial performance with
respect to neuromuscular activities in PCE-exposed offspring but
differences disappeared with age. Analysis of brain neurotransmitters
in newborn and 21-day-old offspring of PCE-exposed dams indicated a
significant reduction in acetylcholinesterase levels in offspring from
both treatment regimens and a reduction in dopamine in offspring of
dams initially exposed during the second week of gestation.
PCE injected into the air space of developing chick embryos at
doses of 25-100 umol/egg influenced survival if treatment occurred on
the sixth day of incubation but was more or less ineffective in compari-
son to olive oil controls when treatment occurred on the second day of
incubation. The number of dead embryos treated on the sixth day roughly
corresponded to dose: 60%, 20% and 10% mortality for the 100, 50 and
25 M-mol groups compared with 14% mortality in olive oil-treated controls.
An increase in the embryonic length of dead embryos demonstrated de-
layed lethal toxicity due to declining dose (e.g., 1.5 cm at 100 ^mol
PCE/egg, 2.3 cm at 50 vmol/egg). Six of 61 surviving embryos (9.8%)
exposed to 5-100 ymol PCE/egg were malformed (predominantly skeletal
anomalies) compared with two of 56 (3.6%) olive oil-treated controls
(Elovaara et_ al_. , 1979). However, the extreme sensitivity of this assay
procedure frequently results in false positive results, making experi-
mental findings difficult to interpret with certainty.
d. Other Toxicological Effects
In laboratory animals, the acute oral (LD,-n) values range from
3980 mg/kg-4680 ing/kg in the rat (Withey and Hall_, 1975) to 8850 mg/kg
in the mouse (Stecher, 1968). In mammals, acute exposure to PCE is
characterized by depression of the central nervous system, cardiac
depression, decreased respiration, decreased blood pressure and exces-
sive fluid accumulation, congestion and inflammation of the lungs
(NIOSH, 1978). The liver appears to be the principal target organ of
PCE toxicity (i.e., liver enlargement, fatty degeneration and abnormal
96
-------
liver function tests) but damage to the proximal convoluted tubules of
the kidney has also been noted in mice, rats and rabbits (NIOSH, 1978;
NCI, 1977).
A significant incidence of toxic nephropathy was noted in both
B6C3F1 mice and Osborne Mendel rats in the National Cancer Institute's
bioassay on the carcinogenicity of PCE (NCI, 1977). Time-weighted
average doses of 386 mg/kg and 772 rag/kg for female and 536 mg/kg and
1072 mg/kg for male B6C3F1 mice given by gavage 5 days per week for 78
weeks results in 82-100% incidence of toxic nephropathy (see Table 20).
Similarly, 58% to 94% of Osborne Mendel rats given time-weighted oral
doses of 474 mg/kg and 949 mg/kg (females) or 471 mg/kg and 941 mg/kg
PCE (males) for 78 weeks also exhibited toxic nephropathy (see Table 21),
Kidney damage has also been noted after inhalation of PCE. Carpen-
ter (1937) reported congestion and granular swelling in kidneys of rats
exposed to 1540 mg/m3 PCE, 8 hours per day, 5 days per week for a period
of / months.
Liver injury resulting from PCE exposure was noted by Kylin et_ al.
(1963) who reported moderate fatty degeneration of the liver following a
single 3-hour exposure to 1340 mg/m3 of PCE. Exposure to this same con-
centration 4 hours per day, 6 days per week for 8 weeks enhanced the
severity of the lesions induced by PCE (Kylin _et_ al., 1965)
In a series of inhalation studies, Rowe (1952) reported loss of
coordination and equilibrium, weight loss, increased liver and kidney
weights, and central fatty degeneration and swelling of the liver in
guinea pigs exposed to 100-370 mg/n3 PCE 7 hours per day for 10-236 days.
Rabbits exposed 7 hours per day for 39 days to 370 mg/m3 PCE exhibited
central nervous system depression and slight liver toxicity but displayed
no adverse effects following exposure to 60 mg/m3 for 222 days. Similar
results were noted in two rhesus monkeys following 179 seven hour
exposures to 60 mg/m3 PCE over a 250-day period.
Reports from the Russian literature (Tsulaya et al., 1977;
Bonashevskaya, 1977 a,b) also indicate that daily exposure of rats
to 19 mg/m3 of PCE for 94 days can disrupt central nervous system
function, blood enzyme activity, the normal morphology of the liver,
lungs and mast cells, DNA synthesis by the liver and the biological
oxidation in the liver, lungs and adrenal glands.
97
-------
TABLE 20. INCIDENCE OF TOXIC NEPHROPATHY IN B6C3F1 MICE
GIVEN PCE BY GAVAGE FOR 78 WEEKS
Male Female
Vehicle Control
Low PCE Dose
High PCE Dose
0/20
40/49
45/48
(0%)
(82%)
(94%)
0/20
46/48
48/48
(0%)
(96%)
(100%)
Source; Adapted from NCI (1977)
TABLE 21. INCIDENCE OF TOXIC NEPHROPATHY IN OSBORNE-MENDEL
RATS GIVEN PCE BY GAVAGE FOR 78 WEEKS
Male Female
Vehicle Controls
Low PCE Dose
High PCE Dose
Source: Adapted from NCI (1977)
0/20
43/49
47/50
(0%)
(88%)
(94%)
0/20
29/50
38/50
(0%)
(58%)
(76%)
98
-------
4. Human Studies
In man, the predominant effect of PCE exposure (2. 155 mg/kg) is
depression of the central nervous system characterized by vertigo, im-
paired memory, confusion, irritability, "inebriation-like" symptoms,
tremors and numbness (NIOSH, 1978; U.S. EPA, 1979). PCE is a skin,
eye and respiratory tract irritant (NAS, 1977; U.S. EPA, 1979) and has
been linked to cases of peripheral neuritis (NIOSH, 1976). Kidney im-
pairment, toxic chemical hepatitis, and enlargement of the liver and
spleen have been reported following accidental exposure to PCE (NIOSH,
1973). The level of PCE permissible in U.S. working environments is
670 mg/m3 (100 ppm) (NIOSH, 1976).
Because of the widespread use of PCE in industry, the acute central
nervous system effects of PCE have been examined in some detail (Rove
It al., 1952; Carpenter, 1937; Stewart et al., 1961, 1970, 1977; Medek
and Kovarik, 1973). In general, little or no effects occur at a con-
centration of 700 mg/m3 PCE. Minimal effects (sensory changes, light-
headedness, impaired coordination) become evident at 1300 mg/m3, with
more definite indications of CNS depression (mental confusion, lassi-
tude) observed as the concentration increases. At 10,000 mg/m3 PCE,
signs of inebriation occur. Volunteers exposed to 13,400 mg/m3 PCE*
were forced to leave the chamber after 7.5 minutes (Rowe et_ a_l. , 1952).
Essentially no data, however, are available on the long-term
effects of PCE exposure'. Stewart and co-workers (1977) examined 12
human volunteers exposed to 168 mg/m3 and 670 mg/m3 PCE for 5.5 hours
per day repeated up to 53 days. No consistent neurological changes
due to PCE exposure could be documented.
Kidney impairment and liver damage have been reported in humans
following accidental exposure to PCE (NIOSH, 1978) but are not well
documented. Supporting evidence for hepatotoxic effects were presented
by Coler and Rossmiller (1953). Three of seven men occupationally ex-
posed to PCE concentrations of 1890 mg/m3 to 2600 mg/m3 had evidence
of impaired liver function. In view of the very high incidence of toxic
nephropathy in both mice (82-100%) and rats (58-94%) chronically ex-
posed to PCE and the induction of hepatocellular carcinoma in mice but
not rats exposed to PCE, the significance of long-term, low-level ex-
posure to PCE to human health is difficult to assess, but is, neverthe-
less, an area of concern. Indeed, a preliminary report on a cohort
mortality study of 330 laundry and dry cleaning" workers indicates an
increased proportion of cancer deaths, particularly of liver cancer
and leukemia. The small number of deaths, however, may have biased
the findings, and cautious interpretation of the study is needed until
additional members of this occupational group are examined (Blair et- al
1979). ——''
99
-------
5. Overview
Tetrachloroethylene, a widely used industrial solvent, is readily
absorbed through the lungs. Approximately 57 percent of inhaled PCE is
retained but most of this amount (80-100%) is subsequently exhaled
unchanged in expired air. A respiratory half-life of 65 hours has'been
estimated for man; urinary clearance of approximately 2% of retained
PCE as trichloroacetic acid has an approximate half-life of 144 hours.
These values suggest accumulation of PCE may occur with repeated ex-
posure.
_ In laboratory animals, acute oral LD50 values range from 3980 mg/k*
in the rat to 8850 mg/kg in the mouse. Acute exposure to PCE is
characterized by depression of the central nervous system, and by
liver and kidney damage.
Pronounced toxic nephropathy was seen in mice and rats chronically
exposed to 386 mg/kg and 471 mg/kg PCE, respectively, by gava^e for 78
weeks.
PCE exposure has been linked to hepatic carcinoma in B6C3F1 mice at
a dose of 386 mg/kg given by gavage. This effect may possibly reflect
a secondary response to PCE-induced hepatectomy. Carcinogenicity assays
in rats were inconclusive. Mutagenic findings are varied but positive
results in mammalian cell transformation studies and host-mediated assays
implicate PCE as a mutagen. There are not indications of teratogenic
effects associated with PCE exposure.
In man, the predominant effect of PCE exposure by inhalation
(>. 30 mg/m ) is depression of the central nervous system, characterized
by vertigo, confusion, inebriation-like symptoms, tremors and numbness.
Accidental exposure to PCe has also been linked to kidney impairment
and hepatotoxic effects. The lack of long-term exposure data makes
assessment of long-term, low-level exposure to PCE*difficult. However,
the pronounced nephrotoxicity in rodents and increased incidence of
hepatocellular carcinoma in mice raise concerns for the human health
aspects of prolonged exposure to PCE.
B. EFFECTS ON AQUATIC ORGANISMS
Data concerning the toxicity of PCE for aquatic biota are extremely
limited and have been found for only a few fresh and saltwater soecies
(Table 22).
In a bioassay with fathead minnows (Pimephales promelas). Alexander
£t al. (1978) compared the results of a flowthrough test in which the
concentration was measured with a static test in which the concentra-
tion was calculated. The 96-hour LC50 (median lethal concentration)
for PCE in the flowthrough experiment was 18.4 mg/1, while the static
test result was 21.4 mg/1. Similarly, U.S. EPA (1980a) reported a
96-hr. LC5jD of 13.5 mg/1 for fathead minnow in a flowthrough test
Because of the volatility of PCE, the flowthrough test results were
100
-------
TABLE 22. THE TOXIC EFFECTS OF PCE ON AQUATIC ORGANISMS
Concentration
__ Species (mg/1)
• freshwater Species
Rainbow trout 4 8 - 5 ft
(Salmo j>airdng£i)
Fathead minnow \b U
(JPimephales promelas)
Fathead minnow 10 / ,,, ,
fvt u i iH.4 - 21.4
(..rlmephales promelac)
Fathead minnow Q 84
Bluegill Sunfish
( Lepomi£ m^iToc]rlnis ) 1 2 • 9
Daphnia magna -n -,
— a — .1 / . /
Alga (Sclenastriim >816
c^ipr_icor nu turn)
Marine Species
fab (Limanda limanda) 5
—
Sheepshead minnow >29.4- <52 2
L jy££A££5itu£)
Mysid shrimp Q 45Q
^HZ^JiL^psJ^. baliia)
—
Mysid shrimp lo 2
(Mysi^lops^s bahia)
Barnacle naupii 3 5
OLLiHinius modestus)
Alga (Phaeodect^lum JQ 5
trie or nu turn) ~~~
Alga (^kele tonenia 504-509
— --—• —~
Test Type
and Dura t lor.
•
96 hr
flowthrough
96 hr
flowthrough
96 hr
flowthrough
chronic embryo-
larval
96 hr static
48 hr static
96 hr
96 hr
Life cycle
96 hr
48 hr
96 hr
Effect
L?50
50* IG!»S °f
equilibrium
LC50
chronic value —
hatchabi 1 i ty
survived growth and
deformities
LC
LSo
LC
ECcQ, reduction in
cell number and
chlorophyll a
t r\
LC50
LC
LL50
Chronic value
LC50
LC50
EC50» "['take of CO
50
Reference
U.S. EPA 1980a
Alexander et^ al . (1978)
Alexander et al. (1978)
U.S. EPA (1980a)
U.S. EPA (1978)
U.S. EPA (1978)
U.S. EPA (1978)
Pearson and McConnei 1 (1975)
U.S. EPA (1978)
U.S. EPA (1978)
U.S. EPA (1978)
Pearson and McConnei 1 (1975)
Pearson and McConnei 1 (1975)
U.S. EPA (1978)
-------
considered to be more accurate. Rainbow trout were considerably more
sensitive, and the 96-hour LC50 valves were 5.8 and 4.8 mg/1 with and
without the presence of the solvent dimethylforinamide (U.S. EPA 1980a).
Other freshwater organisms bioassayed for sensitivity in static
bioassays were the bluegill sunfish (Lepomis macrochirus), Daphnia
aagna. and the alga Selenastrom capricornutum (U.S. EPA, 1978). For
the bluegill, the calculated 96-hour LC5Q was 12.9 mg/1. The 48-hour
LC5Q for the daphnia was 17.7 mg/1. The median effects concentration
(£€50) for a reduction in cell number and chlorophyll-a_ mass in the
alga was greater than 816 mg/1.
The lowest effect level for a freshwater species is a chronic
value established for fathead minnow of 0.840 mg/1 based upon an embryo-
larval test (U.S. EPA 1980a);this is the only chronic study for a
freshwater species.
Among marine species, the barnacle Elminius modestus had the
lowest reported 96-hour LC5Q (3.5 mg/1), as calculated by Pearson and
McConnell (1975). The only saltwater finfish tested was the sheepshead
minnow (Cyprinodon variegatus), for which the estimated 96-hour LC50
was between 29.4 mg/1 and 52.2 mg/1 (U.S. EPA, 1978).
The lowest effects concentration for a saltwater species reported
was 0.450 mg/1, a chronic value for the mysid shrimp (Mysidopsis'bahia).
The 96-hour LC50 for this species was found to be 10.2 mg/1 (U.S. EPA,
1978). Among marine algae, Phaeodectylum tricornutum (EC50 = 10.5 mg/1)
was apparently more sensitive to PCE than Skeletonoma costatum, with
96-hour EC5n values based upon effects on chloraphyll-a of 504-509 mg/1
(U.S. EPA, 1978). ~
All available aquatic toxicity data are summarized in Table 23.
It should be emphasized that with the exception of the data of Alexander
£t al. (1978), and U.S. EPA (1980a), the LC50 values presented are
probably overestimated because of the rpaid evaporation of PCE. More-
over, since the effects of certain water parameters (e.g., hardness,
temperature) on PCE toxicity are not known, it may not be appropriate
to compare the results of unrelated studies.
Two fish kills directly attributed to PCE have been reported in
the last ten years. In 1970, discharges from a textile mill into an
estuary killed 500 fish in two days. A much more serious kill of 16,300
fish occurred in 1974 as the result of a spill of PCE from an over-
turned tanker truck into a freshwater stream. In neither case were
concentrations reported; however, it is likely that the spill produced
much higher PCE concentrations than the industrial discharges (U S
EPA Files, 1980b).
The U.S. EPA (1980a) has not established water quality criteria
for the protection of aquatic life at this time due to the inadequacy
of the data base.
102
-------
REFERENCES
Alexander, H.D., W.H. McCarty and E.A. Bartlett. 1978. Toxicity of
Perchloroethylene, Trichloroethylene, 1,1,1-Trichloroethane, and Methylene
Chloride to Fathead Minnows. Bull. Environ. Contain. Toxicol. 20:344-352.
Bartsch, H., C. Malaveille, A. Barbin and G. Blanche. 1979. Mutagenic
and alkylating metabolites of halo-ethylenes,chlorobutadienes and dichloro-
butenes produced by rodent or human liver tissues. Arch. Toxicol. 41:
249-277.
Blair, A., P. Decoufle and D. Grauman. 1979. Causes of death among
laundry and dry cleaning workers. Am. J. Pub. Health 69(5);508-511.
Bonashevskaya, T.I. 1977a. Results of morphological and functional
studies of the lungs in a hygienic assessment of atmospheric pollution.
Gig. Sanit. 2_:15-20, taken from: Chem. Abs. V37 #034038E.
Bonashevskaya, T.I. 1977b. Morphological characteristics of the adap-
tation of the liver to the effect of some chemical substances. Gig. Sanit.
4_:45-50, taken from: Chem. Abs. V87 //034051D.
Carpenter, C.P. 1937. The chronic toxicity of tetrachloroethylene. J.
Ind. Hyg. Toxicol. 19_:323, as cited in U.S. EPA, 1979.
Cerna, M. and H. Kypenova. 1977. Mutagenic activity of chloroethylenes
analyzed by screening system tests. Mutat. Res. 46;214-15.
Coler, H.R. and H.R. Rossmiller. 1953. Tetrachloroethylene exposure in
a small industry. Arch. Ind. Hyg. Occup. Med. 8:227, as cited in U.S.
EPA, 1979.
Elovaara, E., K. Hemminki and H. Vainio. 1979. Effects of methylene
chloride, trichloroethane, trichloroethylene, tetrachloroethylene and
toluene on the development of chick embryos. Toxicology 12(2);111-119.
Greim, H., G. Bonse, Z. Radwan, D. Reichart and D. Henschler. 1975.
Mutagenicity in vitro and potential carcinogenicity of chlorinated
ethylenes as a function of metabolic oxirane formation. Biochem. Pharma-
col. Z4:2013-2017.
Ikeda, M. 1977. Metabolism of trichloroethylene and tetrachloroethylene
in human subjects. Environ. Hlth. Perspect. 21:239-45.
Kylin, B. , H. Reichard, I. Suinegi, e_t al_. 1963. Hepatotoxicity of in-
haled trichloroethylene, tetrachloroethylene and chloroform. Single
exposure. Acta Phannacol. Toxicol. 20/16-26, as cited in U.S. EPA, 1979.
Kylin, B., _et al. 1965. Hepatotoxicity of inhaled trichloroethylene
and tetrachloroethylene. Long-term exposure. Acta Phannacol. Toxicol.
22_:379- , as cited in U.S. EPA, 1979*.
103
-------
McConnell, G., D.M. Ferguson, and C.R. Pearson. 1975. Chlorinated
hydrocarbons and the environment. Endeavour 34:13, as cited in U.S
EPA, 1979. —
Medek, V. and J. Kovarik. 1973. The effect of perchloroethylene on the
health of workers. Pracovni Lekarstvi 25_:339, as cited in u'.S. EPA, 1979,
Monster, A.C., G. Boersma and H. Steenweg. 1979a. Kinetics of tetra-
chloroethylene in volunteers; influence of exposure concentration and
work load. Int. Arch. Occup. Environ. Health 42;303-309.
Monster, A.C. 1979b. Difference in uptake, elimination, and metabolism
in exposure to trichloroethylene, 1,1,1-trichloroethane and tetrachloro-
ethylene. Int. Arch. Occup. Environ. Health 42;311-317.
Monster, A.C. and J.M. Houtkooper. 1979c. Estimation of individual up-
take of trichloroethylene, 1,1,1-trichloroethane and tetrachloroethylene
from biological parameters. Int. Arch. Occup. Environ. Health 42_;319-323.
National Academy of Science (NAS). 1977. Chapter VI: Organic Solutes,
pp. 769-770, in Drinking Water and Health. Washington, D.C.
National Cancer Institute (NCI). 1977. Bioassay of Tetrachloroethylene
for possible carcinogenicity. DHBJ Publication No. (NIH) 77-813. U.S.
Department of Health, Education, and Welfare, Public Health Service,
National Institutes of Health.
Nelson, B.K. 1979« Behavioral teratology of perchloroethylene. Tera-
tology 19:41A.
National Institute for Occupational Safety and Health (NIOSH). 1976.
Criteria for a recommended standard . . . Occupational exposure to tetra-
chloroethylene (perchloroethylene). DHEW Publication No. (NIOSH) 76-185.
U.S. Department of Health, Education, and Welfare, Public Health Service,
Center for Disease Control, National Institute for Occupational Safety
and Health.
National Institute for Occupational Safety and Health (NIOSH). 1978.
Current Intelligence Bulletin No. 20, Tetrachloroethylene (Perchloro-
ethylene), U.S. Department of Health, Education and Welfare, Public
Health Service, National Institute for Occupational Safety and Health.
Ogata, M. , Y. Takatsuka, and K. Tomokuni. 1971. Excretion of organic
chlorine compounds in the urine of persons exposed to vapors of tri-
chloroethylene and tetrachloroethylene. Br. J. Ind. Med. 28: 386-391
as cited in U.S. EPA, 1979. —
Pearson, C.R. and G. McConnell. 1975. Chlorinated ^ and C2 hdyrocar-
bons in the marine environment. Proc. R. Soc. Lond. B., 189:305-332.
104
-------
Price, P.J. , C.M. Hassett and J.I. Mansfield. 1978. Transforming
activities of trichloroethylene and proposed industrial alternatives.
In Vitro 14/3);290-293.
Riihimaki, V., and P. Pfaffli. 1978. Percutaneous absorption of solvent
vapors in man. Scand. J. Work, Environ. Health 4_(1) : 73-85 , taken from
Chen. Abs. V89 #03761N.
Rowe, V.K., D.D. McCollister, H.C. Spencer, E.M. Adams and D.D. Irish.
1952. Vapor toxicity of tetrachloroethylene for laboratory animals and
human subjects. Arch. Ind. Hyg. _5_:566-79, as cited in U.S. EPA, 1977.
Schumann, A.M., and P. G. Watanabe. 1979. Species differences between
rats and mice on the metabolism and hepatic macroiaolecular binding of
tetrachloroethylene. Toxicol. Appl. Pharmacol. 48_(1, Part 2) :A 89.
Schwetz, B.A., B.K.J. Leong and P.J. Gehring. 1975. The effect of
maternally inhaled trichloroethylene, perchloroethylene, methyl chloro-
form and methylene chloride on embryonal and fetal development in mice
and rats. Toxicol. Appl. Pharmacol. 32;84-96.
Stecher, P.G. (ed). 1968. The Merck Index. 8th edition, Merck & Co.,
Inc., Rahway, New Jersey.
Stewart, R.D. and H.C. Dodd. 1964. Absorption of carbon tetrachloride,
trichloroethylene, tetrachloroethylene, methylene chloride and 1,1,1-
trichloroethane through the human skin. Am. Ind. Hyg. Assoc. J. 25:439
as cited in U.S. EPA, 1979.
Stewart, R.D. H.H. Gay, D.S. Erley, C.L. Hake, and A.W. Schaffer. 1961.
Human exposure to tetrachloroethylene vapor. Arch. Environ. Health 2:
516-522, as cited in U.S. EPA, 1979.
Stewart, R.D. , E.D. Baretta, H.C. Dodd, e_t al. 1970. Experimental
human exposure to tetrachloroethylene. Arch. Environ. Health _20:255 ,
as cited in U.S. EPA, 1979.
Stewart, R.D. e_t al. 1977. Effects of perchloroethylene/drug interac-
tion on behavior and neurological function. DREW (NIOSH) Publ. NO. 77-
191, as cited in U.S. EPA, 1979.
Tsulaya, V.R., T-.I. Bonashevskaya, V.V. Zykova, V.M. Shaipak, F.M. Enaan,
V.N. Shoricheva, N.N. Belyaeva, K.N. Kumpan, K.T. Tarasova, L.M. Gush-
china. 1977. Toxicological features of certain chlorine derivatives of
hydrocarbons. Gig. Sanit. _8:50-53, taken from Chem. Abs. V37 j? 128362H.
Withey, R.J. and J.W. Hall. 1975. The joint toxic action of perchloro-
ethylene with benzene or toluene in rats. Toxicology 4:5 as cited
in U.S. EPA, 1979. ~
U.S. Environmental Protection Agency. 19SOb. Files on Fish Kills, 1980.
105
-------
U.S. Environmental Protection Agency (U.S. EPA). 1977. Criterion Docu-
ment: Tetrachloroethylene. Interim Draft No. 1 (December 10), Washing-
ton, D.C.
U.S. Environmental Protection Agency (U.S. EPA). 1978. In-depth studies
on health and environmental impacts of selected water pollutants. Report
No. 68-01-4646. Washington, D.C.
U.S. Environmental Protection Agency.(U.S. EPA). 1979. Ambient Water
Quality Criteria: Tetrachloroethylene. Office of Water Planning and
Standards, Criteria and Standards Division, Washington, D.C.
U.S. Environmental Protection Agency (U.S. EPA). 1980a. Ambient Water Quality
Criteria for Tetrachloroethylene. EPA 44015-80-073. Office of Water
Regulations and Standards, Washington, D.C.
U.S. Environmental Protection Agency (U.S. EPA). 1980b. Files on Fish
Kills.
106
-------
CHAPTER VI
EXPOSURE TO TETRACHLOROETHYLENE
A. HUMAN EXPOSURE
1. Introduction
In estimating exposure to PCE, the populations and subpopulations
exposed to the chemical were identified. The duration of exposure was
estimated and the total possible intake from such exposures was calcu-
lated. These estimated exposures are not definitive, but rather indi-
cative of the range of potential exposures.
The preceding analysis of the fate of PCE in the environment has
shown that measurable levels of PCE may occur in all environmental
media—air, water, soil and sediment. Monitoring data substantiate
this, showing a wide range of PCE levels in the human environment.
Therefore three exposure routes—inhalation, ingest ion and dermal—were
considered, indicating, where possible, which routes are most signifi-
cant.
Identifying exposed populations and estimating the duration of
exposure requires knowledge concerning the sources of PCE release to
each environmental medium and the types of human activities occurring
in proximity to each source type. Factors considered relevant to PCE
exposure are: place of residence and/or occupation, use of coin-operated
laundries, and consumption of food and water containing PCE. The expo-
sure analysis considers three settings that differ with respect to the
number of PCE sources and. therefore, concentration: urban, remote, and
near manufacturer and user sites.
Because PCE is used primarily in the work environment, occupational
exposure is of great concern. Since the associated risks are being
investigated by other federal agencies, they are not stressed in this
report. The occupationally exposed subpopulation that handles PCE
directly is considered briefly in order to place the estimated environ-
mental exposures into perspective. The facilities in which PCE is used
are also viewed as point sources to the environment. Exposures in the
vicinity of such sources are estimated on the basis of the dispersion
analyses performed in Chapter IV, monitoring data, and the work of
Verberk and Scheffers (1980).
2. Exposure Situations^
a. Populations Exposed Through Ingestion
Humans may be exposed to PCE via ingestion of contaminated foodstuffs
and drinking water. Surface water may be contaminated by direct dis-
charge of factory effluents, sewers, transportation spills, and through
107
-------
contact with contaminated ground water. Landfills are postulated as a
likely source of ground water pollution, through leaching, as well as
surface runoff.
The data on PCE concentrations in drinking water were detailed
previously in Table 6 (Chapter IV). The range in mean values for the
positive samples in surface supplies was 0.18 ug/1 (Ten Cities) to
2.8 ug/1 (CWSS), with the value for NOMS (0.81 ug/1) in between.
Medians in all three surveys, however, were below 0.5 ug/1.
Human exposure to PCE via drinking water in the U.S. has been
estimated by making a number of assumptions as shown in Table 23. For
the 117 million persons utilizing surface water supplies (Temple, Barker
and Sloane, 1977), it was assumed that 8% or 9 million persons ingest
detectable levels of PCE (^0.2 ug/1) in their drinking water. This
assumption is based on the data from NOMS, which showed that 8% of the
cities sampled contained detectable levels. Although there is no direct
correlation between the frequency of detection and the population served,
such an assumption provides a rough method for estimating nationwide
exposure. It is apparent that most persons (an estimated 108 million)
are exposed to levels less than 0.2 ug/1 PCE in drinking water from
surface supplies and thus ingest less than 0.4 ug per day.
Estimating exposure of persons utilizing ground water supplies is
more difficult since the sampling is more limited and is generally
biased toward the sampling of contaminated supplies. A very rough
approximation can be made from the CWSS data (see Chapter IV), in which
about 5% of the samples contained detectable levels of PCE (>_0.5 ug/1).
On this basis, about 4 million persons could be exposed to detectable
levels (with a mean 3-4 ug/1) of PCE in ground water supplies.
There is evidence, however, of some highly contaminated supplies as
a result of a local contamination incident (disposal site, spill, etc.)
or leaching in the distribution system. These occurrences are unpre-
dictable and the size of the population exposed cannot be estimated.
Judging from the ground water data discussed in Chapter IV, such inci-
dents appear to be fairly common. Worst case exposures of this type
are estimated in Table 23.
Ingestion of PCE may also occur in foodstuffs. Unfortunately, the
data on PCE levels in food are sparse as was shown in Chapter IV. The
only available data, a British study, were used to estimate a potential
ingestion exposure via food of 1.2 Ug/day (see Table 24). Due to the
limited data base, however, it is unknown if this estimate is represen-
tative of typical U.S. exposure.
b. Populations Exposed Through Inhalation
Exposures are estimated from a concentration, an exposure duration
and an intake rate. For inhalation exposures to PCE, the concentrations
will vary with proximity to a source or sources, and will also varv
108
-------
TABLE 23. ESTIMATED HUMAN
EXPOSURE TO TETRACHLOROETHYLENE IN DRINKING WATER
Population
Size
Estimated
Exposure
(yg/day)
Assumptions
General population 108 million
9 million
<0.4 <0.2 yg/1 in surface
water; 2 I/day
2-6 1-3 yg/1 mean of posi-
tive samples ; 8% of
population receiving
detectable levels ;
2 1/dav
General population
71 million
4 million
<1
6-8
<0.5 yg/1 in ground
water; 2 I/day
3-4 yg/1 mean of posi-
tive samples; 5% of
population receiving
detectable levels;
2 1/dav
Isolated Exposures:
Contaminated wells
Contamination from
distribution system
750 worst case - 375 yg/1
in well water; 2 I/day
10,000 worst case - 5000 yg/1
in tap water; 2 I/day
109
-------
temporally with weather conditions and source operations. Thus, when
analyzing a particular exposure scenario, such as an activity taking
place in an urban area, the exposure concentration could vary consider-
ably depending upon the locations of source and receptor (see Figures 17
and 18). Similarly, the duration of exposure can vary from zero to an
entire day. Therefore, in the following analysis of exposure scenarios,
although a PCE concentration and a characteristic exposure duration
were selected for each scenario, the selected concentration generally
represents a range of mean values for that type of environment as shown
in Chapter IV, Table 11. The intake rate, however, remains fixed: the
average adult human breathes at a rate of 1.2 m3/hour while awake and
at 0.43m /hour when sleeping (ICRP 1975). Furthermore, it is assumed
that approximately 50% of inhaled PCE is retained and absorbed (Oeata
et al., 1971).
Table 25 presents the various inhalation exposure situations
analyzed for PCE and the associated estimated exposures. These numbers
do not represent actual exposures but suggest the possible ranges of
such exposures. At this juncture, it is not possible to quantify the
numbers of people who may fall into each category with the associated
exposure levels and durations actually designated for each cohort.
However, such cohorts would include the following:
1. People living far from PCE sources (i.e., in remote areas),
2. People living in urban areas with low PCE levels,
3. People living in urban areas with higher PCE levels,
4. People living near manufacturing or user sites,
5. People living near a drycleaning establishment,
6. People working in any of the five above situations
while living in another, and
7. People using coin-operated laundry facilities equipped
with drycleaning equipment.
Table 26 describes occupational exposures, with concentration data
and workers exposed. Note that the concentrations and exposures are
given in units that are three orders of magnitude higher than those for
non-occupational exposure.
c. Populations Exposed Through Dermal Absorption
Dermal exposure may occur through use of PCE and contact with
contaminated air and water during various activities. Although informa-
tion is available concerning quantities of water that are actuallv in
contact with the skin in daily washing activities and during water sports
activities, the rate of absorption by the skin has not been documented
for PCE. The potential exposure via this route is probably very low in
comparison with that of other exposure routes.
110
-------
TABLE 24. ESTIMATED HUMAN EXPOSURE TO TETRACHLOROETHYLENE IN FOOD
PCE
Food or Food Group
Milk
Cheese
Butter
Meat
Oils and fats
Fruits and vegetables
Fruit drink
Fish
TOTAL
Concentration
(Ug/kg)1
0.3
2
13
1
7
2
2
13
Consumption
(g/day)2
266
15
6
207
8
343
29
11
Exposure
(ug/day)
0.08
0.03
0.08
0.2
0.06
0.7
0.06
0.01
1.20
!See Table 10, Chapter IV.
2USDA (1980).
3Lower end of the range of 1-5 ug/kg was taken since these samples were
taken from a relatively contaminated area (Liverpool Bay and Thames
Estuary).
Ill
-------
TABLE 25. ESTIMATED EXPOSURE OF HUMANS
TO TETRACHLOROETHYLENE VIA INHALATION
Subpopulation
Rural/remote
Urban
Concentration1
(yg/m3)
0.1-0.5
1-14
Duration
(hours/day)
24
24
Estimated
Exposure2
(ug/day)
1.1-6
11-160
Near Manufacturing
Sites & Industrial
Areas
Near Drycleaning
Facilities
Use of Coin-Operated
Laundry Facilities
with Drycleaning
Equipment
0.12-210
250-12,250
136,000
24
24
0.5
1.3-2400
2,800-137,000
41,000
1These levels generally represent a range of average concentrations
for different locations and were taken from Chapter IV, Table 11 and
Table 26 for coin-operated laundry facilities with drycleaning equipment
2A respiratory flow of 1.2 m3/hour during a 16-hour day and 0.4 m3 for
8 hours at night (ICRP 1975). A retention of approximately 50% is
assumed (Ogata et al. 1971).
112
-------
TABLE 26. ANALYSIS OF OCCUPATIONAL EXPOSURE TO TETRACHLOROETHYLENE1
Exposed Populations
and
Exposure Levels
Total Employees per Plant
Concentration in General
Workplace (TWA2, mg/m3)
Directly Exposed Employees3
Concentration of Direct
Exposure (TWA, mg/m3)
Calculated Exposures
(ing/day) '>
DRYCLEANTNC
Fabric
Commercial Industrial Coin-Op Scouring
47.5
1
203
46
34
4
203
225-970 160-970
136
650
203
DECREASING
Vapor Cold
110 5
34 136
10 4
237
136
160-970 650-1150 650
TOTAL
Total Employees
Number of Facilities
100,000
18,750
13,500 22,000 315,000 32,000 90,000 572,500
270 14,500 2,500 3,600 45,800 85,000
ta and references in Appendix A, which profiles the industry.
2 TWA - time weighted average, in this case, over the 8-hour working day.
"Total employees" is an industry figure and includes the entire available workforce. The number
employed thus cannot be derived by multiplying the number of facilities by the total employees
per model plant.
'Based upon an 8-hour workday, a respiratory flow rate of 1.2 m3/hr. and 50% retention of inhaled PCE.
-------
j. Results of Exposure Calculations
The total daily exposure ranges calculated above are summarized in
Table 27. Clearly the proximity of the human receptor to the source of
PCE is the critical element determining the extent of innalation exposure.
In urban areas non-industrial drycleaning facilities represent highly
concentrated sources of PCS. The exposure potential of these facilities
falls on local residents, users, workers and maintenance staff of these
establishments.
In general, drinking water does not appear to contribute greatly
to overall exposure to PCE. In rural areas, however, drinking water does
represent a major fraction of total estimated exposure due to the lower
ambient air concentrations of PCE in these areas. However, certain inci-
dents of contaminated water supplies have made drinking water an important
route of exposure in some cases. Although information is extremely
limited regarding levels of PCE in food, the preliminary estimate
included in Table 27 suggests that food does not represent a significant
source of PCE exposure.
B. EXPOSURE OF AQUATIC BIOTA
A formalized'exposure analysis is not possible for aquatic systems.
The two fish-kill incidents described in Chapter V graphically present
the exposure events of concern: an accidental spill and an industrial
discharge; however, ambient concentrations associated with these spills were
not reported.
The mean of unremarked ambient levels of PCE (as of 1981) in surface
water according to STORET was 8.5 yg/1. The maximum was 142 ug/1. Xinety-
one percent of all observations were at or below the detection level
(usually 10 ug/1). In a 1977 study, PCE was detected in 38% of national
samples in industrialized regions, for the most part at levels <5 ug/1;
only 2% of the samples exceeded 10 yg/1.
114
-------
TABLE 27. SUMMARY OF ESTIMATED HUMAN EXPOSURE TO TETRACHLOROETKYLENE
Estimated
Exposures
Exposure Route (ug/day)
Drinking water
general population - surface water <0.4
- ground water <1
smaller subpopulation - surface water 2-6
- ground water 7-8
Food 1.2
Inhalation1
Rural/Remote 1.1-6
Urban 11-160
Near Manufacturing Site/Industrial Area 1.3-2400
Near Drycleaning Facilities 2,800-137,000
Use of Coin-Operated Laundry Facilities
with Drycleaning Equipment 41,000
Occupational2 160,000-1,150,000
Isolated Exposures
Drinking Water - Contaminated Wells 750
Contamination in
Distribution System
up to 10,000
Assumes 50% respiratory retention and various durations of exposure
(see text).
2See Table 26.
115
-------
REFERENCES
Federal Register, Vol. 44, No. 52, Thursday, March 15, 1979, Appendix
c» Guidelines and Methodology Used in the Preparation of Health Effect
Assessment Chapter of the Consent Decree Water Criteria Documents.
International Commission on Radiological Protection (ICRP)„ 1975.
Report of the Task Group on Reference Man. Pergamon Press, Oxford.
Ogata, M., Y. Takatsuka, and K. Tomokuni. 1971. Excretion of organic
chloring compounds in the urine of persons exposed to vapors of tri-
chloroethylene and tetrachloroethylene. Br. J. Inc. Med. 28:386-391.
(As cited by U.S. EPA 1979.)
U.S. Environmental Protection Agency. 1979. "Tetrachloroethylene:
Ambient Water Quality Criteria," Criteria and Standards Division, Office
of Water Planning and Standards, U.S. EPA.
U.S. Department of Agriculture. (USDA). 1980. Nationwide Food
Consumption Survey 1977-1978. Preliminary Report No. 2 Food and nutrient
intakes of individuals in 1 day in the United States. Spring 1977.
Science and Education Administration, Washington, D.C.
Verberk, M.M., and T.M.L. Scheffers. 1980. Tetrachloroethylene in *
exhaled air of residents near drycleaning shops. Env. Research 21:
432-437.
116
-------
CHAPTER VII
RISK CONSIDERATIONS
The desired output of this risk assessment is the quantification of
risks associated with use and production of tetrachloroethylene (PCE) to
various subpopulations of humans and other classes of biota. This
requires careful: (1) identification of the populations exposed,
(2) evaluation of the ranges in each subpopulation's exposure, (3) con-
sideration of the effects levels or dose response data for the species
of concern and/or proxies for these species and (4) extrapolation of
effect levels from dose/response data for laboratory animals to the
human subpopulations at risk.
A. RISKS ASSOCIATED WITH HUMAN EXPOSURE
1. Introduction
To assess the risks of PCE exposure, several human exposure situa-
tions derived in Chapter VI were compared with dose levels of PCE that
have caused adverse effects in man and/or laboratory animals. The risks
most clearly associated with PCE exposure in laboratory animals are
carcinogenesis, CNS disturbances, kidney impairment and hepatotoxic
effects (Table 28). Quantitative estimates were made of human carcino-
genic risk based on animal data. The estimated margins of safety for
acute effects associated with various exposure situations were approxi-
mated as well. However, there is at present no basis for quantifying
the effects of chronic exposure.
One study indicated PCE is a liver carcinogen in mice; results with
rats were negative but were confounded by poor survival. Malignant
transformation of PCE-exposed rat-embryo cells to tumor-producing cells
has been demonstrated jln vitro.
Preliminary data in a retrospective human mortality study also
suggest an increased incident of cancer-related deaths, particularly of
liver cancer and leukemia in laundry and drycleaning workers. However,
the small population examined makes uncertain the findings and cautious
interpretation of this study is needed until additional members of this
occupational group are examined. Further research is needed to clarify
the carcinogenicity of PCE in man and to establish dose-response
relationships.
The chief target organs of PCE toxicity in animals are the liver
and kidney. Liver enlargement, fatty degeneration and abnormal liver
function tests as well as kidney damage, particularly to the proximal
convoluted tubules, have been linked to PCE exposure. Disruption of the
central nervous system has also been reported. However, dose-response
relationships for these effects are unclear.
117
-------
TABLE 28. ADVERSE EFFECTS OF TETRACHLOROETIIYLENE ON MAMMALS
oo
Adverse Effect
Species
Lowest Reported
Effect Level
No Apparent
Effect Level
Hepatocellular
carcinoma
Toxic nephropathy
Mouse 386 mg/kg
by gavage
Mouse 386 mg/kg
by gavage
Rat /,71 mg/kg
by gavage
(% of incidence)
(65%)
(82%)
(88%)
Mutagenicity
(cell transformation)
Rat
(embryo cells) 97 umol in vitro
CNS disturbances
Man
1360 mg/m by inhalation 678 mg/nf
Median Oral Lethal Dose
Rat
3980 mg/kg (50%)
-------
Inhalation appears to be the dominant PCE exposure route for humans.
Ambient air levels of PCE, even in urban areas, are several orders of
magnitude lower than inhalation levels associated with toxicological
effects in humans. Exposure calculations (Table 29) indicate that non-
occupational inhalation intakes may range from 1 ug/day to 6 yg/day in
remote areas to 11-160 tig/day in urban areas. The highest non-occupational
inhalation intakes (2.8 mg/day to 137 mg/day) would occur near drycleaning
facilities. Use of coin-operated laundry facilities for 0.5 hours per
day presents the second largest non-occupational intake (41 mg/day),
but this exposure is not anticipated to occur on a daily basis. Occupa-
tional exposures were estimated to be in the 160-1150 mg/day range.
In general, drinking water does not appear to contribute greatly
to overall exposure to PCE except perhaps in rural areas.where it
constitutes a greater portion of total exposure than in urban areas.
However, certain incidents of contaminated water supplies have made
drinking water an important route of exposure in some cases. A limited
data base suggests a low level of exposure (1.2 ug/day) via foodstuffs.
2. Quantitative Carcinogenic Risk Estimation ,
Below, the potential carcinogenic risk to humans due to tetrachloro-
ethylene ingestion is estimated.
^Ideally, this problem is approached on two fronts:
1) Application of various extrapolation models to occupational
vs. ambient* human exposure data (from retrospective epide-
miological studies) to obtain an approximate dose/response
relationship.
2) Application of these same models to data from controlled
experiments on laboratory animals, and conversion of the
animal/dose/response relationship to an estimated human
dose/response.
In the first approach, the overriding uncertainty is in the data
themselves: usually the exposure levels, lengths of exposure, and even
response rates (responses per number exposed) are "best estimates,"
and, furthermore, unknown factors (background effects, etc.) may bias
the data. In the second approach, the data are usually more accurate,
but the relationship between animal and human response rates must be
questioned, and at present there is no universally accepted solution to
this problem. (In short, in the former case we have relevant data of
questionable validity, whereas in the latter we have valid data of
questionable relevance.) If both analyses can be performed and the
results corroborate each other, we gain confidence in these results.
If, on the other hand, data are not available for one of the analyses,
it is assumed that some result is better than no result, and an analysis
is performed that is based on the available data.
*(or ambient, location A vs. ambient, location B)
119
-------
In any case, a more important question is: which, if any, of the
mathematical models is accurate? For the time being, though there is
no firm basis for judgment, the models applied here are believed to tend
to overestimated the true risk.
For PCE, the only quantitative carcinogenicity data currently
available are from an NCI study on mice (discussed in U.S. EPA, 1979a).
The available data concerning human and other mammalian effects are
discussed in Chapter V. The data selected for extrapolation are listed
in Table 29.
To deal with the uncertainties inherent in extrapolation, three
commonly used dose/response models have been applied to the mouse data
to establish a range of potential human risk. Additionally, the results
from the CAG multistage model have been included. The assessment of
potential human risk is subject to several important qualifications:
• Though positive carcinogenic findings exist, there
have also been negative findings in tests with other
species. Thus the carcinogenicity of PCE to humans
is far from certain.
• Assuming that the positive findings indeed provide a
basis for extrapolation to humans, the estimation of
equivalent human doses involves considerable uncertainty.
• Due to inadequate understanding of the mechanisms of
carcinogenesis, thee is no scientific basis for
selecting among the several alternate dose/response
models (as discussed above), which yield widely
differing results.
a. Calculation of Human Equivalent Doses
To obtain a quantitative human risk estimate based on animal data,
it must first be determined what human dose is equivalent to a given
animal dose. The approach recommended by the U.S. EPA (1979b) has been
followed, which normalizes the dose rate according to body surface area.
This approach is relatively conservative, in that it results in a lower
equivalent dose than would be obtained from simple multiplication
of animal dose rate (mg/kg/day) by human body weight. Whether the
surface area or body weight ratio is the more appropriate normalization
factor is still debatable. Since the weight ratio is roughly 14 times
as large as the surface area ratio, the choice of a conversion method
suggests an uncertainty of an order of magnitude at least.
120
-------
TABLE 29. CARCINOGENIC RESPONSE IN MICE EXPOSED TO TETRACHCOROETHYLENE
Male
Mice
Fema le
Mice
Time-Weighted
Average Dose
(mg/kg/day)
0
0 (vehicle
control)
536
1072
0
0 (vehicle
control)
386
772
Equivalent
Human Dose
(mg/day)
0
0
1930
3860
0
0
1390
2780
Response
2/17
2/20
32/49
27/48
2/20
0/20
19/48
19/49
Percent
12
10
65
56
10
0
40
39
Percent Excess
Over Averaged
Controls*
-
61
51
-
36
36
Source: U.S. EPA (1979a).
P. (x) - P
*]> /v\ _ t V_
' U) t - P
-------
The calculation of a human equivalent dose was performed using the
following formula, assuming 70 kg for human weight and 0.025 kg for mouse
weight:
Human Dose = Animal Dose x 5 days x Animal Weight x Human Weight
(mg/day) (mg/kg/day) 7 days (kg) Animal Weight
From this, it was estimated that a dose of 1 mg/kg/day, 5 days per week
for a mouse is equivalent to a dose of 3.6 mg/day for a human.
b. Estimation of Human Risk
The three dose/response models used to extrapolate human risk were
the linear "one-hit" model, the log-probit model, and the multistage
model. The latter is actually a generalization of the one-hit model,
in which the hazard rate is taken to be a quadratic rather than linear
function of dose. All of these models are well known in the literature,
and a theoretical discussion may be found in Arthur D. Little (1980).
The one-hit and multistage models assume that the probability of a
carcinogenic response is described by
P (response at dose X) = 1 -
where h(x) is the "hazard rate" function. The log-probit model assumes
that human response varies with dose according to a log-normal distribu-
tion. Due to their differing assumptions, these dose/response models
usually give widely differing results when effects data are extrapolated
from relatively high doses to the low doses typical of environmental
exposure.
For the linear one-hit model, the equation
: (x) - 1
c
1 - e"Bx, where P (x) = t (x) "
1 - P
c
P (x) is the excess probability of response at dose x and is solved for
the parameter B. From our data, we find that B is approximately 3 x 10
based on the average of the dose/response data in male and female mice.
For the log-probit extrapolation, the "probit" intercept A was
determined by the following equation:
P(x) - $ (A + log1Q [x])
where $ is the cumulative normal distribution function.
122
-------
This equation makes the assumption that the log-probit dose/response
curve has unit slope with respect to the log, ,-dose. Using tables of the
standard normal distribtuion, A is found approximately equal to 3.5, based
on a mean value of A for all of the dose/response data. This value was
used to determine the probability of a response at various concentrations
according to the above equation.
The multi-stage model with a quadratic hazard rate function,
2
h(x) = ax + bx + c,
was fit to the data. To estimate the parameters a, b, and c, a maximum
likelihood method was used, aided by a computer program, which performed
a heuristic search for the best fit. The parameter b dominates for small
values of dose x and parameter a dominates for large values.
In Table 30 the risk estimates obtained from these three models, as
well as EPA's CAG estimate, have been summarized. The expected number
of cancers per million exposed population is shown for daily doses ranging
from 0.1 ug/day to 100 mg/day. These estimates represent probable upper
bounds on the true risk, due to the conversative assumptions that were
used. The gap between the estimates is large in the low-dose region;
thus, there is a substantial range of uncertainty concerning the actual
carcinogenic effects of PCE. However, present scientific methods do not
permit a more accurate or definitive assessment of human risk.
Exposure levels and doses to individuals have been estimated for
many different exposure conditions. These conditions consider inhalation
of PCE in air and ingestion of PCE in food and water. The range of risks
associated with body intakes for these exposure conditions, using four
risk models, is summarized in Table 31. In addition to the assumptions
previously mentioned, all of these risk estimates are based on the
assumption that the daily exposures occur continuously over an individual's
lifetime.
Estimated excess individual cancers due to continuous lifetime
consumption of water contaminated with PCE at average concentration
levels range from negligible to less than 6 x 10~7. At the highest
concentration observed in drinking water, the estimated risk of excess
individual lifetime cancer is on the order of 6 x 10~3.
The range of individual predicted excess lifetime cancers associated
with non-occupational inhalation intakes extends from negligible additional
risks to 9.1 x 10~5 for maximum urban exposures (Table 31). For non-
occupational intakes due to inhalation near drycleaning facilities, the
estimated excess individual risks due to continuous lifetime exposure
are on the order of 5.6 x 10~4 to 8.5 x 10~2. Continuous lifetime
exposures to concentrations likely to occur in coin-operated laundry
facilities would result in an estimated excess individual risk of life-
time cancers between 8.2 x 10~3 to 8.5 x 10~2, but exposure at these
levels is not anticipated to occur continuously throughout an individual's
lifetime.
123
-------
TABLE 30. ESTIMATED EXCESS LIFETIME CANCERS PER MILLION POPULATION EXPOSED TO
TETRACHLOROETHYLENE AT VARIOUS EXPOSURE LEVELS1
Extrapolation
Exposure
Estimated No. Excess Lifetime Cancers
(per Million Population Exposed)1
Model Level (mg/day):
Linear Model
Log-Probit Model
Mul ti-Stage Model
CAC Multi-Stage Model
0.0001
0.03
*
0.02
0.057
0.001 0.01
0.3 3
* <0.02
0.2 2
0.57 5.7
0.1
30
3
20
5.7
1
300
233
200
570
10
3000
6200
2000
5700
100
30,000
67,000
22,000
57,000
'The number of lifetime excess cancers represents the increase in number of cancers over the normal
background incidence, assuming that an Individual is continuously exposed to tetrachloroethyJene at
the indicated daily intake over their lifetime. There is considerable variation in the estimated
risk due to uncertainty introduced by the use of laboratory rodent data, by the conversion to equiva-
lent human dosage, and by the application of hypothetical dose-response curves. In view of several
conservative assumptions that were utllized.it is likely that these predictions overestimate the
actual risk to humans.
-------
TABLE Jl. RANGES OF CARCINOGENIC RISK TO HUMANS DDK TO ESTIMATED
EXPOSURE TO TETRACIiLOROETIIYl.ENE1
Exposure Koute
Drinking Water
General Population - Surface Water
Ground Water
Smnl lor Sulipopiilatlon - Surface Water
- Ground Water
Food
Inhalation
Rural Remote
Urban
- Near Manufacturing Site/Industrial Areas
- Near Dryclcanliig Facilities
- Use of Coin-Op Facilities wttli Drycleanlng Equipment
- Occupational
Isolated Exposures
Drinking Water - contaminated wells
- contamination in distribution system
Estimated
Exposure dig/day)
< 0.4
< I
2-6
6-8
1.2
1.1-6
11-160
1.3-2.300
2.800-135,OOO6
41.OOO6
165.000-1,150,000*
>50
10,000
Range in no. estimated excess
lifetime cancers
(per Million Popul.it Ion Exposed)
neg2 - -0.23
net; - <0.61
neg - 0.3
neg
0.5
neg2 - 0.071
neg
0.3
3
neg - 91
neg2 - 1.3003
560* - 85,OOO2
8.200* - 30,OOO2
33,000* - 481,OOO1
2,000* - 6.20O2
A range of probability Is given, based on several different dose-response extrapolation models. The lifetime excess incidence of cancer
represents the Increase in incidence of cancer over the normal background incidence, assuming that an individual is continuously exposed to
letrachloroethylene at the Indicated daily Intake over their lifetime. There is considerable variation in the estimated risk due to uncertainty
Introduced by the use of laboratory rodent data, by I he conversion of equivalent human dosage, and by the application of hypothetical dose-
respon.se curves. In view of several conservative assumptions that were utlllzed.it is likely that these predictions overestimate the actual risk
to humans.
2
I.og-prubit extrapolation model.
•)
CAG »fj t islage extrapolation model.
/,
Multistage extrapolation model.
Assuming a tesplratory flow of 1.2 m /hr while aw.,ke (16 In), 0.4 m3/hr while asleep, .mil i esplr.ilory retention of 50Z.
'Assumed continuous lifetime dally exposure; the leauUing risk e&t ImaLes probably torn! Lo la- higher than actually occur because exposure for a given
Individual is unlikely to be continuous.
-------
its
presented above should be considered along with the followin firsl
the human carcinogenicity of PCE is as yet unproven, second, the size" 'of
some_of the subpopulations is unknown, and third, the "median" levels
for ingestion are not statistically valid for the entire U.S. population.
— — Other Human Risks Associated with PCE Exposure
Other than carcinogenic risks, the risks associated with chronic
exposure to PCE cannot be quantified. The effects of chronic exnoiure
ion! L™ ri T6 n0t been Wel1 characterized, making assessment of
long-term, low- level exposure to PCE difficult.
Tests with laboratory animals have established lowest observed-
effect levels of 386 mg/kg body weight over a two-year period. These
-0 f ?f magnitude fbove estimated human exposure levels.
of teratogenic effects of PCE have been reported
B. RISKS TO AQUATIC SYSTEMS
ef fectson- t£StS ^ been Perfo™^ to evaluate the
ettects of PCE on aquatic organisms, some insight may be Cleaned *™m
the laboratory experiments. The lowest concentrations at wJich toxic
efrects on a freshwater species were observed was 840 yg/1 for the
Cathead minnow. This concentration is two orders of magnitude larger
than typical unremarked ambient concentrations (STORET data average
was 8.0 ug/1), x^hile the highest level observed was 142 yg/1 Effluent-
concentrations reported in STORET average 57 Ug/l, howevef the
'
-re similarly well above
for marine organisms and 4. S-l
_ Thus, the maximum effluent concentration exceeds th* rr e
sr^^.—Ss^^.MST
an effluent may be significant as & ' tl°We/er' riSKS ln the -----
126
-------
REFERENCES
U.S. Environmental Protection Agency. 1979a. Tetrachloroethylene.
Ambient Water Quality Criteria. Criteria and Standards Division,
Office of Water Planning and Standards, Washington, B.C.
U.S. Environmental Protection Agency. 1979b. Guidelines and Methodology
Used in the Preparation of Health Effect Assessment Chapter of the Con-
sent Decree Water Quality Criteria Documents. Federal Register 44(52)
Thursday, March 15, 1979.
Arthur D. Little, Inc. 1980. Integrated Exposure/Risk Assessment
Methodology contract Draft Report 68-01-3857. Monitoring and Data
Support Division, Office of Water Planning and Standards, U.S. EPA,
Washington, D.C.
127
-------
APPENDIX \
DESCRIPTION OF OCCUPATIONAL ENVIRONMENTS IN WHICH
TETRACHLOROETHYLENE IS USED*
A. INTRODUCTION
Tetrachloroethylene is, at room temperature, a colorless, sweet-
smelling, volatile liquid. It is about 1.5 times heavier than water,
and its vapor is about five and three quarters times as dense as air.
PCE is an effective solvent for a large number of natural and syn-
thetic organic substances. It is these good solvent properties, low
fire hazard, and ability to form an azeotropic mixture with water that
have made PCE popular for use in the cleaning of garments and textile
fabrics. Approximately 45% of domestic tetrachloroethylene consump-
tion occurs in the drycleaning and textile industries, with approxi-
mately 19% used as a metal cleaning solvent (Arthur D. Little, Inc.,
1977). Table A-l is materials balance for occupational use of PCE: the
data have been used in this report as a basis for fate modeling.
Tetrachloroethylene is used in drycleaning at approximately
35,000 facilities throughout the United States (NIOSH, 1979). Most of
these tetrachlorethylene users are small, independent retail dry-
cleaners. Most retail (commercial) drycleaning is done with equipment
that requires the manual transfer of garments damp with solvent. In
the textile industry, PCE is used in wool scouring and dye scouring of
knits, as well as in laboratory-scale simulation of drycleaning opera-
tions for testing of fabric wear characteristics.
The extent of use of PCE in metal cleaning is small as compared
with the use of trichloroethylene. The higher boiling point of tetra-
chloroethylene(121°C versus 87°C) requires the use of substantially
more heating for vapor degreasing than is needed with trichloroerhy-
lene. The resultant hotter vapor may be desirable for selected ap-
plications (e.g., dewaxing) but is generally undesirable as the cleaned
material has to be removed at a higher temperature. Tetrachloroethylene
is also utilized to a limited extent as a cold cleaning solvent.
PCE is a raw material in the production of some fluorocarbon
materials, e.g., Freon 113. In this application, it is utilized in a
closed chemical system.
••This appendix is a compilation of information derived from the seven
sources listed in the References.
129
-------
TABLE A-l MATERIAL BALANCE OF TETRACHLOROETHYLENE USED IN OCCUPATIONAL ENVIRONMKNTS
References
Number of Facilities
I'oundu fC.K Lost per Year
I'lIK lx>st Through Solid Waste
I'OK Lost Through Incineration
1'Cli Lost Through Atnvisphere
7. Loss in Solid Uni.lv
1 Loss in Incineration
Z Loss in Atmosphere
Dry Clo.ming
Commercial
(4)(6)
18.750
128 x 106
21.6 x 106
—
106.4
17
—
83
1 ml list riii I
(5)(7)
2/0
28.8 x 106
0.13 x 106
—
28.67 x 106
<»
—
>99
"
2
B
90
Cold
(DO)
45,800
41 x 106
2.05 x 106
8.2 x 106
30.75 x 106
5
20
75
Subtotal
49.400
135.8 x 106
3.95 x 106
15.8 x 10b
116.05 x tOb
3
12
65
Fabric
Scour Inj;
(D(3)
2.522
120.4 x 106
0.54 x 106
—
119.86 x 106
<1
—
>99
Crand Total
85,442
461.4 x 106
26.22 x 106
15.8 x 106
419.38 x 106
6
3
91
-------
1. Degreasing
a. Vapor Degreasing
Vapor degreasing is a cleaning process in which organic soil is
dissolved and removed through the condensation of hot solvent vapors
on the cold, soiled work. Industry sources confirm that a total of
30,000 vapor degreasers are currently operating in the United States,
and a fraction of these machines employ tetrachloroethylene.
The most recent and comprehensive survey of the vapor degreasing
industry was presented by Dow Chemical Company in a recent EPA report.
This survey involved contacting 2,578 plant sites engaged in a manu-
facturing activity with the metalworking industries and employing more
than 20 people. (Most vapor degreasers are believed to be employed in
these industries.) When the results of the survey are extrapolated to
the entire industry, the population of vapor degreasers in the United
States can be defined as shown in Table A-2. Dow Chemical reported thai
85% of the vapor degreasers surveyed were using open-top equipment;
the remainder were using enclosed, with conveyor systems machines.
The distribution of open-top versus enclosed machines did not appear
to be correlated with plant sizes in the United States (EPA, 1979).
On the basis of communications with vapor degreaser operators and
suppliers, it is estimated that an average of three operating per-
sonnel and three maintenance personnel are associated with each open-
top vapor degreaser. Twice as many people are likely to be associated
with each enclosed machine. These estimates account for rotation and
turn-over of personnel and occasional extensive maintenance or clean-
out requirements (Arthur D. Little, Inc., 1977).
b. Cold Cleaning
Cold cleaning (solvent degreasing) involves the use of liquid
solvent to remove soil, with solvent being directly hand-applied (rub-
bing or wiping) in some cases, while spraying or soaking is utilized
in other cases. A survey of solvent producers by Dow indicated that
41 million pounds of PCE are utilized in solvent degreasing.
c. Textile Industry Applications
Tetrachloroethylene is utilized in several processes within the
textile industry. It is found as a component of some carrier solvents
in the dyeing of synthetics, is employed as a solvent in scouring wool
and synthetics, and is utilized in simulated drycleaning operations in
order to evaluate fabric wear characteristics.
131
-------
TABLE A-2. USE OF SOLVENTS IN VAPOR DECREASING
Solvent
Number of Decreasing
Tanks
Trlchloroethyene
1,500
Tet rac hloroothy]ene
4 ,095
Methyl Chloroform
5,310
Number of Facilities
Number of Exposed
Personnel
885
8,4 00
2,700
23,169
350
32,174
SOURCE: Arthur D. Little, Inc., estimates based on Dow Survey and conversations with degreaser vendors and
operators and solvent manufacturers.
-------
2. Drycleaning
a. Overview
The "drycleaning" process, whereby garments are cleaned in a
solvent that is primarily non-aqueous, originated in Paris, France, in
the mid-19th century. This process has undergone many refinements and
changes since its origin, as gasoline, and carbon tetrachloride have
been replaced as drycleaning solvents by Stoddard solvent, tetra-
chloroethylene, and fluorocarbon 113.
The drycleaning industry encompasses three types of firms. "Com-
mercial drycleaners" are those engaged primarily in drycleaning or
dyeing of apparel and household fabrics other than rugs. This segment
of the industry is characterized by a large number of independent
businessmen, each of whom operates his own small plant. "Industrial
drycleaners" are those engaged in supplying laundered or drycleaned
work uniforms, wiping towels, dust control items, etc., to industrial
and commercial users. "Coin-operated" dry cleaning installations are
those that are often found in conjunction with coin-operated laun-
dries. Although the equipment utilized for coin-operated drycleaning
is designed to require no attendant operator, many coin-operated in-
stallations are manned with an attendant who both conducts routine
maintenance procedures and assists customers.
The drycleaning process utilized by both commercial establish-
ments and industrial plants entails removing soils from garments through
the use of a non-aqueous solvent. The drycleaning solvents currently
in use can be classified as "petroleum" solvents and "synthetic" solvents,
with the synthetic category including tetrachloroethvlene and trichlorotri-
fluoroethane (fluorocarbon 113), and the petroleum category including
Stoddard solvent and 140F solvent.
On the basis of available census data and communications with
industry sources, it is estimated that there are currently 25,000
commercial drycleaners (SIC code 7216) in the United States and that
approximately 75% of these establishments utilize tetrachloroethylene
as their drycleaning solvent. Approximately 125,000 workers are em-
ployed in commercial drycleaning, with an estimated 30,000 of these
employees actually operating the drycleaning machinery. While all
employees in a commercial drycleaning plant may be exposed to solvent
vapors, the drycleaning machine operators are most directly exposed.
Industrial drycleaners (SIC code 7218) are those industrial laun-
derers who operate their own cleaning facilities. Industry sources
estimate that there are approximately 500 industrial establishments
with drycleaning operations, and these plants utilize either tetra-
chloroethyleneor petroleum solvents in their drycleaning operations.
Industrial drycleaners employ approximately 25,000 workers, with an
estimated 2,000 employees directly exposed to solvent as either clean-
ing machine operators or maintenance personnel.
133
-------
In each drycleaning plant (commercial or industrial), several
discrete operations are performed. The following section describes
these operations and the potential of each for exposing employees to
PCE and other chemicals.
b. Process Description
i. Garment Marking
"Marking" is the process of identifying garments by attaching
tags to each garment or stamping an identification code onto an inner
surface of the^garment. Marking is the first operation performed upon
garments entering a cleaning plant and is a necessary prerequisite to
garment processing. As garments are delivered to the cleaning plant,
they are bundled; before processing can begin, garments must be iden-
tified so that, at the conclusion of the drycleaning process, orders
can be properly assembled for return to their owners. The marking
process usually involves the sorting of garments, both by "due date"
and by garment type (light or dark color, fragile or durable garments,
etc.).
There are no hazardous chemical or physical agents inherent in
the marking process. However, marking areas are often adjacent to the
drycleaning area, and marking personnel may be exposed to diffusing
solvent emissions that originate from the drycleaning equipment.
ii. Spotting
"Spotting" involves the selective application of chemicals, steam,
detergent, and/or water to loosen or remove specific stains from soiled
garments. Spotting is sometimes done prior to drycleaning (pre-spotting),
but may also be necessary following the drycleaning step to remove
stubborn stains. Depending upon the size of the drycleaning plant and
the nature of the drycleaning process, spotting can require a full-
time employee; however, this step is usually handled by the drycleaning
machine operator. Industrial drycleaning plants generally do not
conduct spotting operations.
iii. Drycleaning
Equipment/Process Categories: Drycleaning is a process during
which batches of garments are immersed in solvent and agitated within
a horizontally mounted cylinder. This "washing" step is followed by a
spin cycle to extract solvent and a drying operation to evaporate any
remaining solvent from the damp clothing.
Most commercial PCE equipment involves the use of two machines,
the first to wash and extract garments, and the second to dry. This
"transfer" equipment requires the manual handling of damp garments.
Some PCE drycleaning equipment combines the washing, extraction, and
drying steps into a single unit. This type of equipment, known as
134
-------
"dry-to-dry," is utilized by only a small number (less than 20%) of
commercial drycleaners; the need for larger capacity units to achieve
production comparable to that of a transfer plant (due to longer resi-
dence of a load in a dry-to-dry machine) has limited its spread.
Employee Exposures from Cleaning Operations: The drycleaning
machine operator generally is exposed to the highest concentration of
solvent within each plant. One major source of direct employee ex-
posure to solvent vapor is the transfer of garments. Garment transfer
usually is done manually and typically involves holding garments wet
or damp with tetrachloroethylene directly in the employee's breathing
zone. Most drycleaners comply with OSHA's current permissible expo-
sure levels of 100 ppm tetrachloroethylene over an 8-hour time-weighted
average; however, manual transfer operations result in employee ex-
posures which exceed the peak allowable concentration (300 ppm) speci-
fied for PCE by OSHA in 29 CFR 1910.1000. Furthermore, it is unknown
how many drycleaners currently meet the exposure levels recommended by
the recent NIOSH criteria documents on tetrachloroethylene.
There are several solvent sources compounding the employee ex-
posures that occur during1 garment transfer operations. Among the
common sources of solvent emissions are leaking washer and tumbler
door gaskets, tumbler aeration dampers, lint trap and button trap
doors, improperly operating water separators, and pump gaskets. These
emission sources are generally indicative of inadequate maintenance
programs; however, they highlight the need for equipment design that
minimizes routine maintenance requirements.
Another source of emissions is the premature removal of garments
from the drying cycle. This problem may result from attempts to shorten
the cycle for increased productivity or from the presence of solvent-
retaining items such as comforters.
Employee Exposures from Solvent Treatment: Other recognized
emission sources result from the techniques utilized by drycleaners to
maintain the purity of their solvents. During normal operation, sol-
vent is continuously filtered. The filter medium typically consists
of either a series of wire mesh strips coated with diatomaceous earth
or of a replaceable filter cartridge. Periodically, the filter medium
must be replaced. Where cartridge filters are utilized, they must be
drained and discarded, a process that often results in excessive em-
ployee exposures to residual solvent. Where diatomaceous earth is
employed, the filter medium is removed from its mesh carrier by a back
wash system, by air bubbling, or by mechanical agitation. The col-
lected filter medium is then either discarded (presenting a handling
problem) or, as is usually done at PCE plants, is "cooked" to recover
solvent from the filter "muck." Muck cooking can result in signifi-
cant solvent emissions since the cooker itself contains gaskets that
may leak and. if the cooking process is not properly carried out,
large amounts of residual solvent may escape into the plant environ-
ment when the cooker is opened.
135
-------
Most drycleaners also employ distillation to purify their sol-
vent. Tetrachloroehylene cleaners utilize a batch atmospheric dis-
tillation process. In addition to the potential leak points (gaskets,
joints, etc.) in the still itself, misadjustment of the still's water
separator can result in the presence of solvent in open-top wastewater
containers.
Tetrachloroehylene stills have another potential emission source;
the still's relief vent is often located inside the plant building,
and may discharge small quantities of air saturated with tetrachloroethyleneval
Activated carbon adsorbers ("sniffers") are utilized by most
industrial cleaners and by approximately one-third of the commercial
PCE drycleaners to recover solvent from washer and tumbler exhaust
lines and from the general plant environment. However, if the carbon
bed is not regenerated frequently by steam stripping, significant
quantities of solvent may be lost.
iv. Garment Finishing
The term "finishing" is employed in the drycleaning industry to
indicate the "pressing" of garments to remove wrinkles and restore
each garment to its original size, shape, and appearance. Pressing
equipment is heated with super-heated steam, and pressers may be ex-
posed to elevated heat levels throughout their work shifts. In ad-
dition, the application of heat during the garment finishing step will
drive off any residual solvent from the garment, so that pressers may
be exposed to solvent vapor.
v. Assembly
Following the finishing step, garments are sorted and assembled,
generally in plastic bags. Employees engaged in assembly operations
are exposed to the background levels of solvent and heat found throughout
the plant.
136
-------
REFERENCES
Arthur D. Little, Inc., 1977, Technology Assessment and Economic Im-
pact Study of OSHA Regulations for Tetrachloroehvlene, Trichloro-
ethylene and Methyl Chloroform. Draft report to U.S. Department of
Labor Occupational Safety and Health Administration, Washington, D.C.
Arthur D. Little, Inc., 1979, Engineering Control Technology Assess-
ment of the Drycleaning Industry. Contract 210-77-004, U.S. Depart-
ment of Health, Education, and Welfare Public Health Service Center
for Disease Control, National Institute for Occupational Safety and
Health Division of Physical Sciences and Engineering, Cincinnati,
Ohio.
Fisher, Bill—Personal Communication, International Fabricare Insti-
tute, Silver Spring, Maryland.
Lester, Dick—Personal Communication, American Laundry, Company, Cin-
cinnati, Ohio.
Sluizer, Bud—Personal Communication, Institute of Industrial Laun-
derers, Washington, D.C.
Woolsey, John—Personal Communication, International Fabricare Insti-
tute, Silver Spring, Maryland.
137
-------
U.r-;, Envhunmeniui . , oction Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Floor
Chicago, IL 60604-3590
------- |