EPA-600/6-75-004
DECEMBER 1975
SCIENTIFIC AND TECHNICAL ASSESSMENT REPORT
ON
VINYL CHLORIDE AND POLYVINYL CHLORIDE
Program Element 1AA001
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
WASHINGTON, D. C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental Protection Agency, have
been grouped into series. These broad categories were established to facilitate further development and
application of environmental technology. Elimination of traditional grouping was consciously planned to
foster technology transfer and a maximum interface in related fields. These series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
9. Miscellaneous Reports
This report has been assigned to the SCIENTIFIC AND TECHNICAL ASSESSMENT REPORTS (STAR)
series. This series assesses the available scientific and technical knowledge on major pollutants that would be
helpful in possible EPA regulatory decision-making regarding the pollutant or assesses the state of
knowledge of a major area of completed study. The series endeavors to present an objective assessment of
existing knowledge, pointing out the extent to which it is definitive, the validity of the data on which it is
based, and uncertainties and gaps that may exist. Most of the reports will be multi-media in scope, focusing
on single media only to the extent warranted by the distribution of environmental insult.
EPA REVIEW NOTICE
This report has been reviewed by the Office of Research and Development, EPA, and approved for
publication. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
DISTRIBUTION STATEMENT
This report is available to the public from Superintendent of Documents, U.S. Government Printing Office,
Washington, D.C. 20402.
Report No. EPA-600/6-75-004
11
-------
PREFACE
Although this report is issued in the Scientific and Technical Assessment Report Series, it differs in several
respects from the comprehensive multi-media format that the Series will usually have because it was nearly
completed prior to the creation of the STAR series in August 1974.
The document was prepared by a task force convened under the direction of Dr. F. Gordon Hueter, Special
Studies Staff, U. S. Environmental Protection Agency (EPA), Environmental Research Center (ERC),
Research Triangle Park (RTF), N. C. Assembly, integration, and production of the report were directed by
the Special Studies Staff, ERC-RTP.
In a preliminary assessment of the environmental problems associated with vinyl chloride and polyvinyl
chloride, an EPA Task Force in August of 1974 under the direction of the Office of Toxic Substances
determined that emissions of vinyl chloride monomer were primarily an air pollution problem. Accordingly,
the Office of Air and Solid Waste Management was given the responsibility for an in-depth evaluation of the
problem. This report was proposed as a part of this evaluation. The objective was to review and evaluate the
current knowledge of vinly chloride in the environment as related to possible deleterious effects upon
human health and welfare. An extensive literature review of the toxicology of polyvinyl chloride was not
attempted since the primary concern of this report is the vinyl chloride monomer. Information from the
literature and other sources has been considered generally through June 1, 1975. A more extensive review
of sources, emissions, air quality, and control technology will be available in the Standard Support-
Environmental Impact Statement for Vinyl Chloride.
In this report, concentrations have been expressed in parts per million (ppm) by volume with the metric
equivalent in parentheses. The conversion factor at 25° C and 1 atmosphere of pressure is 1 ppm = 2560
//g/m3.
The following persons served on the Task Force:
From ERC-RTP:
James R. Smith, Chairman
Kenneth Bridbord Jean French
Paul E. Brubaker J.H.B. Garner
Joseph Bufalini Choudari Kommineni
David L. Coffin William Lonneman
Anthony V. Colucci Gordon C. Ortman
James Davis Frank Scaringelli
Dale Denny Bruce Turner
From Office of Air Quality Planning and Standards:
Josephine S. Cooper
John Crenshaw
Michael Jones
From National Institute of Environmental Health Sciences, DHEW:
Robert Drew
iii
-------
From Office of Research and Development, EPA:
Robert McGaughy
The substance of this document was reviewed by the National Air Quality Criteria Advisory Committee
(NAQCAC) of the Science Advisory Board in public session on November 14, 1974. Members of the
NAQCAC were:
Arie J. Haagen-Smit - California Institute of Technology, Chairman
Mary O. Amdur — Harvard University
David M. Anderson - Bethlehem Steel Corporation
Anna M. Baetjer — Johns Hopkins University
Thomas D. Crocker — University of California
Samuel S. Epstein - Case Western Reserve University
James McCarroll - University of Washington
Eugene P. Odum — University of Georgia
Elmer P. Robinson - Washington State University
Morton Sterling — Wayne County Michigan Health Department
Arthur C. Stern — University of North Carolina
Elmer P. Wheeler — Monsanto Company
John T. Wilson — Howard University
Ernst Linde - Executive Secretary
The document was also reviewed at a meeting with representatives of Federal Departments and Agencies
held on December 13, 1974. Officials attending the meeting were:
From Department of Commerce:
Daniel M. Sweger, National Bureau of Standards
Barry C. Cadoff, National Bureau of Standards
V. Hartwell, Office of Environmental Affairs
From Department of Defense:
Major Rothman, Aerospace Medicine Division, USAF
Douglas E. Rector, Bureau of Medicine and Surgery, USN
Leigh E. Deptis, Bureau of Medicine and Surgery, USN
J. E. Shultz, Office of the Chief of Naval Operations, USN
David Lillian, Environmental Hygiene Agency, U. S. Army
Charles Buck, Environmental Hygiene Agency, U. S. Army
John J. Sugrue, Environmental Control Office, Headquarters,
U. S. Army Material Command, U. S. Army
From Department of Health, Education, and Welfare:
Charles H. Powell, National Institute for Occupational
Safety and Health
Frank L. Mitchell, National Institute for Occupational
Safety and Health
R. E. Shapiro, Food and Drug Administration
From Department of Justice:
Joan Cloonan, Land and Natural Resources Division
iv
-------
From Department of the Treasury:
Loy A. Hanes, Bureau of Alcohol, Tobacco, and Firearms
From U. S. Atomic Energy Commission:
Carl G. Welty, Division of Operational Safety
From General Services Administration:
Harold J. Pavel, Repair and Improvement Division
From National Science Foundation:
Marvin E. Stephenson, Division of Environmental Systems
and Resources
From Veterans Administration:
William A. Schmidt, Office of Construction, Research Staff
From Consumer Product Safety Commission:
Rita Orzel, Office of the Medical Director
A review of the document was conducted by a Task Force from the Office of Research and Development
on December 5, 1974. Members of the Task Force were:
Robert E. McGaughy, Chairman Lawrence Plumlee
Alan Carlin Richard A. Rhoden
Vincent DeCarlo S. Sidney Verner
Irene Kiefer David Yount
Gunter Zweig
Dr. Samuel S. Epstein also participated in the review of the document, as a representative of the National
Air Quality Criteria Advisory Committee of the Science Advisory Board.
Review copies of this document also have been provided to other governmental agencies and to industrial
and public interest groups.
All comments and criticisms have been reviewed and incorporated in the document where deemed
appropriate.
-------
CONTENTS
Section Page
PREFACE iii
LIST OF FIGURES vii
LIST OF TABLES vii
LIST OF ABBREVIATIONS AND SYMBOLS ix
ABSTRACT x
1. SUMMARY 1
1.1 HEALTH EFFECTS 1
1.2 ECOLOGICAL EFFECTS 3
1.3 PRODUCTION AND USE 3
L4 EMISSIONS 3
1.5 EXPOSURE LEVELS 3
1.6 MEASUREMENT TECHNIQUES 4
1.7 CONTROL TECHNOLOGY 4
1.8 PHOTOCHEMICAL REACTIONS 4
2. CHEMICAL AND PHYSICAL PROPERTIES 5
2.1 Physical Properties 5
2.2 Chemical Properties 5
2.3 References for Section 2 6
3. MEASUREMENT TECHNIQUES 7
3.1 Sampling Methods 7
3.2 Sample Preparation 8
3.3 Analytical Methods 8
3.4 Automated Monitoring 12
3.5 Integrated Samples Using Adsorbents 13
3.6 References for Section 3 13
4. ENVIRONMENTAL APPRAISAL 17
4.1 Sources 17
4.2 Concentrations in Ambient Air 18
4.3 Dispersion Model Estimates of Ambient Air Concentrations 32
4.4 Reported VC Measurements in Water and Food 38
4.5 Vinyl Chloride Emissions from Solid Waste Incineration 39
4.6 Transformation, Transport, and Removal 41
4.7 References for Section 4 41
5. ENVIRONMENTAL EXPOSURE AND RECEPTOR RISK 43
5.1 Exposure 43
5.2 Risk to Human Health 44
5.3 References for Section 5 45
VI
-------
Section Page
6. UNDESIRABLE EFFECTS 47
6.1 Toxicology 47
6.2 Threshold Limit Values 70
6.3 Human Effects 72
6.4 Ecological Effects 97
6.5 Vinyl Chloride-Related Compounds and Other Chemical Carcinogens 99
6.6 References for Section 6 103
7. CONTROL TECHNOLOGY AND REMEDIAL ACTIONS 113
7.1 Introduction 113
7.2 Monomer Production 113
7.3 Polymer Production 114
7.4 References for Section 7 115
TECHNICAL REPORT DATA AND ABSTRACT 116
LIST OF FIGURES
Figure Page
4.1 Location of Vinyl Chloride and Polyvinyl Chloride Plants 19
4.2 Location of Sampling Sites for Vinyl Chloride Measurements in Niagara Falls 33
LIST OF TABLES
Table Page
2.1 Physical Characteristics of Vinyl Chloride 6
3.1 Possible Interferences With Vinyl Chloride Analysis 9
3.2 Column Materials and Liquid Substrates Separating Vinyl Chloride 10
3.3 Sensitivity of Selected Detectors Used in Gas Chromatography 11
4.1 Vinyl Chloride Monomer Plant Emission Data 20
4.2 Polyvinyl Chloride Polymer Plant VC Emissions Data 21
4.3 Vinyl Chloride Concentration in Grab Samples Taken near Region I Plant 24
4.4 Vinyl Chloride Concentration in Integrated 24-Hour Samples Collected by Charcoal Absorber
Near Region I Plant, 1974 24
4.5 Vinyl Chloride Concentration in Grab Samples Collected near Region IV Plant, March 1974 .. 25
4.6 Vinyl Chloride Concentration in Grab Samples Collected near Region IV Plant, May 1974 ... 26
4.7 Vinyl Chloride Concentration in Integrated 24-hour Samples Collected by Charcoal Absorber
near Region IV Plant, 1974 26
4.8 Vinyl Chloride Concentration in Grab Samples Taken near Region VI Plant, 1974 27
4.9 Vinyl Chloride Concentration in Integrated 10-minute Samples Collected by Charcoal Absorber
near Region IX Plant, 1974 28
4.10 Vinyl Chloride Concentration in Integrated 24-hour Samples Collected by Charcoal Absorber
Near Region IX Plant, 1974 29
4.11 Site Distribution for Vinyl Chloride Monitoring Study 30
4.12 Vinyl Chloride Concentrations, Cumulative Frequency Distribution 31
4.13 Ambient Air Samples from 57th Street, Niagara Falls 34
4.14 Ambient Air Samples from Vicinity of Goodyear Chemical 35
4.15 Ambient Air Samples from Vicinity of Chemical Complex Located on Buffalo Road 35
4.16 Ambient Air Samples outside Niagara Falls Industrial Area 36
vii
-------
Section Pa8e
4.17 Ambient Air Samples from Buffalo Area 36
4.18 Calculated 1-hour Average Concentrations of Vinyl Chloride Monomer at Selected Downwind
Distances from a Plant with Multiple Emission Sources, Plant A 37
4.19 Calculated 1-hour Average Concentrations of Vinyl Chloride Monomer at Selected Downwind
Distances from a Plant with Multiple Emission Sources, Plant B 38
4.20 National Organics Reconnaissance Survey Results for Selected Compounds 39
4.21 Range of Vinyl Chloride Concentrations in Some Categories of Consumer Products 40
4.22 Variation of Combustion Products of Polymer A with Temperature 40
5.1 Relative Importance of Vinyl Chloride Sources for the General Adult Population 43
5.2 Proportional Mortality of Liver Angiosarcoma among Vinyl Chloride Workers 44
6.1 Summary of Acute Effects of Vinyl Chloride Exposure in Experimental Animals 48
6.2 Summary of Chronic Effects of Vinyl Chloride Exposure to Experimental Animals 50
6.3 Types of Tumors Observed in Male Wistar Rats Exposed to 30,000 ppm (79,500 mg/m3) of
Vinyl Chloride 51
6.4 Tumor Incidence in Male Wistar Rats Exposed to Vinyl Chloride 52
6.5 Basic Study Design of Maltoni and Lefemine 54
6.6 Contaminants Found in the 99 Percent Pure Vinyl Chloride Used in the Maltoni and Lefemine
Experiments 56
6.7 Carcinogenic Effects of Inhaled Vinyl Chloride 83 Weeks following Exposure (Exp. BT1) ... 58
6.8 Carcinogenic Activity of Inhaled Vinyl Chloride 69 Weeks following a Reduced Exposure
Period (BT3) 60
6.9 Carcinogenic Activity of Inhaled Vinyl Chloride in Wistar Rats 9 weeks following Termination
of Exposure (BT7) 61
6.10 Carcinogenicity of Inhaled Vinyl Chloride in Swiss Mice 9 weeks following Exposure (BT4) . . 62
6.11 Carcinogenicity of Inhaled Vinyl Chloride in Golden Hamsters 18 weeks following Exposure . 64
6.12 Preliminary Experiment on the Transplacental Carcinogenicity of Inhaled Vinyl Chloride (BT5) 65
6.13 Study on Oncogenic Activity of Oral Vinyl Chloride 66
6.14 Interim Summary of Tumors in Mice Exposed to Vinyl Chloride for 8 months 66
6.15 Tumors Presently Correlated to VC Exposure (by Inhalation) on Experimental Rodents and
Man 67
6.16 Summary and Conclusions of Various Aspects of Vinyl Chloride Carcinogenicity Drawn by
Maltoni and Lefemine 68
6.17 A Summary of Results of the Metabolism Studies of Hefner et al 68
6.18 Reported Cases of Liver Angiosarcoma among PVC Workers and Non-PVC Workers 73
6.19 Historical Data, Cases of Hepatic Angiosarcoma, Connecticut, 1935-1973 74
6.20 Observed Deaths/Expected Deaths and Standardized Mortality Ratios in VC Workers with
Exposure Indices of 1.5 or Greater, by Duration of Exposed Employment 77
6.21 Summary of Dow Mortality Study 80
6.22 Comparison of Mortality Studies among Workers Exposed to VC 83
6.23 Synopsis of Anamnestic, Clinical, Biochemical, Peritoneoscopic, and Histologic Data of 50 PVC
Workers 88
6.24 Summary of Occupational Findings Relating Nonmalignant Liver Damage to Vinyl Chloride
Exposure . 94
6.25 Comparison of the Toxicity Levels of Three Concentrations of Five Fumigants on Several Plant
Species 98
6.26 Production and Use in the United States of Chemicals Related to Vinyl Chloride and Polyvinyl
Chloride „ 100
6.27 Age Adjusted Death Rates from Lung Cancer in Great Britain, Norway, and the United States . 103
vm
-------
LIST OF ABBREVIATIONS AND SYMBOLS
ACGIH
Ag+
AKT
atm
BSP
°C
cal
CH2:CHCI
or
H2C=CHC1
CH3
CH4
Cl-C
C1CH2-
CH2C1
cm
cm'1
cm"
CNS
DDE
DDT
BCD
EDC
EPA
ERC
°F
FID
ft
ft3
g
gal
GC
HC1
hr
IARC
°K
kg
km
Ib
LDH
American Conference of
Governmental Industrial
Hygienists
Silver ion
Alanine-ketoglutarate
transaminase
Atmosphere
Bromsulphalein
Degree Celsius
Calory
Vinyl chloride monomer
Methyl
Methane
Chlorinated hydrocarbon
1,2-dichlorethane
Centimeter
An expression of wavelength
used in infrared spectros-
copy (=1/X)
Square centimeter
Central nervous system
Dibromoe thane
Dichlorodiphenyltrichlorethane
Electron capture detector
Ethylene dichloride
U.S. Environmental Protection
Agency
Environmental Research Center
Degree Fahrenheit
Flame ionization detector
Foot
Cubic foot
Gram
Gallon
Gas chromatography
Acetylene
Hydrogen chloride
Hour
International Agency for
Research on Cancer
Degree Kelvin
Kilogram
Kilometer
Pound
Lactic acid dehydrogenase
LD5 o Dose lethal to 50 percent
of the recipients
m Meter
m2 Square meter
m3 Cubic meter
MFOS Mixed-function oxidase system
mg Milligram
ml Milliliter
mm Millimeter
mo Month
mph Mile per hour
N Newton
NADPH Nicotinamide-adenine
dinucleotide hydrogenase
NAQCAC National Air Quality Criteria
Advisory Committee
ng Nanogram
NIOSH National Institute of Occupa-
tional Safety and Health
nm Nanometer
OSHA Occupational Safety and Health
Administration, U.S. Depart-
ment of Labor
PCB Polychlorinated biphenyls
ppb Part per billion
ppm Part per million
psia Pound per square inch,
absolute
psig Pound per square inch,
gauge
PVC Polyvinyl chloride
RTF Research Triangle Park, N.C.
sec Second
SCOT Serum glutamic-oxaloacetic
transaminase
SGPT Serum glutamic-pyruvic
transaminase
SMR Standard mortality ratio
STAR Scientific and Technical
Assessment Report
TCD Thermal conductivity detector
THF Tetrahydrofuran
TLV Threshold Limit Value (for
occupational exposures)
TWA Time-weighted average
VC Vinyl chloride
wk Week
yr Year
jug Microgram
IX
-------
ABSTRACT
Vinyl chloride is a chemical of widespread industrial and commercial use. Occupational experience and
experimental evidence strongly indicate that it is a carcinogen. Additionally, there is experimental evidence
that indicates that it may be a teratogen and mutagen. Precise dose-response relations between vinyl
chloride and liver angiosarcoma, and other cancers in man, are not available, as they are not for any other
chemical carcinogens. An increased incidence of liver angiosarcoma, excessive liver damage, and
acroosteolysis has been reported among vinyl chloride workers, and the frequency and severity of the liver
pathology is related to the length of exposure. The principal route of exposure for people living near vinyl
chloride (VC) and polyvinyl chloride (PVC) plants is thought to be air inhalation. Sources of increased
importance for the general population include food and water.
Tumors at multiple and diverse sites have been observed in all species of experimental animals tested for
carcinogenicity by inhalation and ingestion of vinyl chloride. Industrial studies suggest an increased risk of
human cancer at multiple sites. An excess incidence of liver angiosarcoma, an extremely rare tumor in man,
was observed among VC/PVC workers and reproduced in experimental animals with very similar pathology.
Liver angiosarcoma was observed in two species of experimental animals after inhalation exposures of VC at
the lowest doses tested, 50 ppm (128,000 //g/m3), and by ingestion at 16 mg/kg.
In addition to the health effects of VC, this document also considers the sources, distribution, and control
technology. Emissions of VC from vinyl chloride and polyvinyl chloride plants are estimated to exceed 100
million kilograms annually, about 90 percent of which is from PVC plants. Installation of currently
available controls, most of which are a basic part of the processing system and serve to recover the reactant
or product, may be adequate to reduce vinyl chloride emissions from VC/PVC plants in the order of 90
percent. /
-------
SCIENTIFIC AND TECHNICAL ASSESSMENT REPORT
ON VINYL CHLORIDE AND POLYVINYL CHLORIDE
1. SUMMARY
This report presents a review and evaluation of the available current scientific data relative to the health and
welfare implications of environmental pollution resulting from the production and use of vinyl chloride and
polyvinyl chloride. The commercial importance of vinyl chloride (VC) lies primarily in the manufacture of
polyvinyl chloride (PVC) resins, which are subsequently manufactured into a large number of useful plastic
products.
Derived from petrochemical feedstock and chlorine, VC is a synthetic chlorinated olefinic hydrocarbon
monomer. Ft is a gas at ambient temperature and atmospheric pressure, but is normally shipped and stored
as a liquid under pressure. It is flammable, explosive, only slightly soluble in water, and about twice as
dense as air.
The concentration of VC entrapped in PVC is dependent upon the production process and can range from
0.1 to about 8 thousand parts per million (ppm). VC Can be liberated, particularly when heated, during
fabrication.
1.1 HEALTH EFFECTS
Occupational exposure studies have strongly implicated vinyl chloride as a human chemical carcinogen
which manifests itself in multiple tumor sites. One of these tumors is a rare liver tumor, angiosarcoma.
Similar toxicology studies have verified the occurrence of tumors in other body organs such as the brain
and lungs. Other manifestations in humans include acroosteolysis, a degenerative disease affecting bones
and finger tips, and liver dysfunction. Experimental studies have shown the potential of VC to be a
mutagen and teratogen.
Although actual VC exposure levels responsible for these effects in humans are not precisely known, limited
measurements around VC/PVC production facilities indicate that contiguous populations are being exposed
to levels of vinyl chloride of potential public health concern. The bases for this concern include, but are not
limited to, the following: two community cases of liver angiosarcoma, which are of questionable
relationship to vinyl chloride; four confirmed cases of liver angiosarcoma in workers exposed to vinyl
chloride, either in final product fabricating plants or during VC manufacture, at levels of exposure as low as
1 to 10 ppm (2560 to 25,600 jug/m3), which may be within an order of magnitude of levels observed in
ambient air; pathologic noncarcinogenic liver damage in workers exposed to VC in fabricating plants or in
post-PVC polymerization phases, which is similar to noncarcinogenic damage seen in workers exposed to
much higher levels of VC; and liver damage based upon BSP retention studies of workers involved in
polymer processing and workers exposed to TWA concentrations of VC of 50 ppm (128,000 jug/m3) for 40
hours a week. None of these data alone provide conclusive evidence that such effects will occur in the
general population, but when viewed together they provide a reasonable basis for concern.
The latency period following the onset of occupational exposure is estimated to be about 13 to 20 years. In
this regard, there are four reported cases worldwide of liver angiosarcoma following exposure of 3 to 6
years duration: two of these cases have been confirmed by pathologists.
Health implications of residual VC in PVC dust particles have been only superficially studied. Much of the
dose-response data on exposure effects of vinyl chloride comes from animal studies using exposure levels of
50 ppm (128,000 Mg/m3) and above.
1
-------
1.1.1 Sources and Exposure Mechanisms
The principal route of exposure for persons living in the vicinity of VC emission sources is thought to be air
inhalation, although exposure can occur from ingestion of food and water, and from skin contact. There is
evidence to indicate that vinyl chloride can exist in drinking water, certain foods, beverages, cosmetics, and
other consumer products. Incomplete combustion of PVC products in municipal incinerators can result in
the emission of VC as entrapped monomer. Use of vinyl chloride as a propellant in aerosol products
recently has been discontinued, so that this source of exposure should decline; however, residual
VC-containing products may still be available on the market. Other potential sources of indoor exposure,
such as migration of monomer from plastic products, have not been studied. Exposure conditions in the
vicinity of PVC product fabricating plants are not yet known.
1.1.2 Human Exposure
Our present knowledge of adverse health effects associated with human exposure to vinyl chloride comes
primarily from recent occupational observations, complemented by laboratory animal data.
Between 1949 and 1966, an excess incidence of liver damage (nonmalignant) and acroosteolysis was
reported among vinyl chloride workers in Europe. Studies in Germany revealed evidence of liver pathology
in a high percentage of PVC production workers with a history of employment ranging from 1.5 to 21
years, but exposure levels responsible for this damage are not known. Since early occupational health
studies often reported acute toxic effects (dizziness, headaches, nausea, etc.), it can be assumed that peak
exposure levels of several thousand parts per million were experienced at times. Air monitoring data in one
group of PVC plants during the period 1950-1959 indicate that time-weighted (8-hour) average exposures in
these facilities were in the range 120 to 385 ppm (307,200 to 985,600^g/m3). This may not be typical of
exposure in all PVC plants. Peak exposures probably exceeded 1000 ppm (2,560,000 ;ug/m3).
Studies in Europe and the United States since 1966 tend to confirm the earlier findings in Europe. These
recent studies include observations of liver damage among workers not directly involved in actual
production of PVC. The frequency and severity of liver pathology among PVC workers have been related to
the length of exposure, that is, liver damage is most common in workers with an exposure history in excess
of 10 years,,
To date 15 cases of liver angiosarcoma, a rare form of liver cancer that is considered fatal, have been
confirmed among workers with a history of exposure to vinyl chloride in the United States, and 12 such
cases have been confirmed in European countries and Canada. Additionally, 11 cases have been reported
but have not yet been confirmed. Most, but not all, of these reported cases have been among workers
involved directly in PVC production. Cases of liver angiosarcoma have been reported in one worker from
the United States and three from Europe exposed to VC, but not directly involved in PVC production. Two
community cases of liver angiosarcoma have been reported in persons living in the vicinity of industrial VC
emission sources.
While the focus of attention has been on liver angiosarcoma, it should again be noted that a number of
industrial studies suggest that the risk of developing other cancers, particularly lung and brain cancer, as
well as liver dysfunction and other disorders, also has been related to VC exposure.
1.1.3 Laboratory Exposure
Acute animal toxicity to vinyl chloride was first reported in 1938. Toxic manifestations in experimental
animals and man included eye irritation, cardiac irregularities, and increased motor activity, leading to
tremor and loss of muscular coordination and finally to narcosis. Short-term acute human experiments
(intermittent 5-minute exposures separated by 6 hours over a period of 3 days) with concentrations ranging
up to 20,000 ppm (51,200 mg/m3) produced acute toxic effects at levels about 8000 ppm (20,480
mg/m3).
2 VINYL/POLYVINYL CHLORIDE
-------
Chronic toxic effects due to vinyl chloride in experimental animals include cancer, and damage to the liver,
spleen, kidney, lungs, brain, and nerve bundles. Some of the pathological lesions observed in these animal
experiments were similar to those later observed in humans engaged in the production and handling of vinyl
chloride.
Multiple tumors, including angiosarcoma of the liver and hepatocellular carcinomas, have been observed in
rats, hamsters, and mice exposed to vinyl chloride. In rats and mice, liver angiosarcoma has been produced
by exposures as low as 50 ppm (128,000 jUg/m3), the lowest level for which studies have been completed.
1.2 ECOLOGICAL EFFECTS
PVC products are not readily biodegradable. In experimental studies on vegetation, symptoms for ethylene
and VC exposure between 10 to 100 ppm (25,600 to 256,000 Mg/m3) were identical. Vegetational damage
around VC manufacturing or processing plants has not been documented.
1.3 PRODUCTION AND USE
The principal use of VC is in the production of PVC, and the principal use of PVC is in the production of a
wide variety of useful plastic materials such as floor tile, phonograph records, pipes, and electric insulation.
VC also has been used as an aerosol propellant, but this practice has been discontinued. VC was first
synthesized in 1837, but the production of vinyl chloride in the United States began in the 1930's. The first
important use was in the manufacture of synthetic rubber. Production levels increased rapidly after World
War II—the beginning of the industrial chemical era that produced over 20,000 new chemical products.
Vinyl chloride production in the United States was less than 45 million kilograms in 1943 and increased to
2.4 billion kilograms in 1973. Based on recent projections, the annual growth rate in the polyvinyl chloride
industry is expected to be in the order of 6 percent per year up to 1980, and the annual growth rate in the
vinyl chloride industry is expected to be about 3 percent.
In the United States, VC is produced at 17 plants and PVC at 40 plants. Approximately 940 workers are
engaged in VC production, and approximately 5600 in PVC production.
1.4 EMISSIONS
Only a very limited amount of VC emission data from industrial sources were available when this report was
written. VC loss estimates of approximately 4 percent have been reported, based primarily on material
balance studies. Losses to the outdoor atmosphere from industrial sources may occur at a large number of
points in the manufacturing processes and will vary depending on the manufacturing facility.
Currently, emissions of vinyl chloride from VC and PVC plants are estimated to exceed 100 million
kilograms annually. About 90 percent of all vinly chloride atmospheric emissions are believed to emanate
from PVC plants. Data are being obtained on emissions of VC from fabrication plants and from fabricated
products, but analyses are not completed at this time. Incineration (without scrubbing) of PVC products
results in the emission of hydrogen chloride gas, and under poor combustion (less than 500°C), the
entrapped monomer.
1.5 EXPOSURE LEVELS
Data on exposure levels of vinyl chloride in ambient air are limited. Atmospheric measurements in the
vicinity of VC/PVC production sources indicate that concentrations are well below 1 ppm (2560jug/m3) in
over 90 percent of the cases. One peak value (grab sample) of 33 ppm (84,480 Mg/m3) has been reported at
0.5 kilometer from the center of one plant. Exposure from other sources (water, food, and other products)
has not been quantified.
Summary 3
-------
1.6 MEASUREMENT TECHNIQUES
Available atmospheric vinyl chloride data have been obtained using a variety of sampling and analytical
techniques with varying degrees of sensitivity • and accuracy, Consequently, the data are not directly
comparable in all cases. Standard sampling and analytical procedures have not been established and
practiced. Continuous monitoring methods suitable for field use are presently limited to infrared
spectrometry. The Wilkes Scientific MIRAN Portable Gas Analyszer has been successfully used in the field.
It is limited to a lower detectable limit of 1 ppm (2560 ;Ug/m3).
1.7 CONTROL TECHNOLOGY
Currently available technology may be adequate to reduce vinyl chloride emissions from VC plants in the
order of 90 percent and from PVC plants by greater than 75 percent. Control of emissions from PVC plants
is a more difficult problem, which may require complex process changes. Means of controlling emissions
from PVC product fabrication processes are being studied by EPA's Office of Air Quality Planning and
Standards.
1.8 PHOTOCHEMICAL REACTIONS
Only limited laboratory studies regarding photochemical reactions of vinyl chloride have been made. Vinyl
chloride does undergo atmospheric reactions in the presence of nitrogen oxides and solar radiation,
although the reaction rate is slower than with other hydrocarbons known to be in the atmosphere. Reaction
products of vinyl chloride photooxidation include carbon monoxide, formaldehyde, formic acid, formyl
chloride, and hydrogen chloride. The half-life of vinyl chloride in laboratory photochemical chamber
experiments has been reported to be 6 hours. The half-life of VC in the ambient atmosphere is not known.
VINYL/POLYVINYL CHLORIDE
-------
2. CHEMICAL AND PHYSICAL PROPERTIES
2.1 PHYSICAL PROPERTIES
The principal physical characteristics of vinyl chloride are given in Table 2.1.l Vinyl chloride (VC) is a
chloroolefinic hydrocarbon with a density slightly more than twice that of air, a molecular weight of 62.5,
and the structural formula shown below:
H
c = r'
/
H "H
Since VC boils at -13.9 °C, it is a gas at normal atmospheric temperature and pressure. It melts at -160°C
Vinyl chloride is highly flammable with a flash point of-78°C (-108°F). The explosive limits are from 4 to
22 percent VC in air by volume. The presence of a chlorine atom in the ethylene molecule changes the
dipole moment from 0 to 1.45 Debye units. The corresponding saturated hydrocarbon, chloroethane has a
dipole moment of 2.05. Analysis of VC usually reveals trace amounts of organic impurities, such as
acetylene, 1,3-butadiene, methyl chloride, vinylidine, and vinyl acetate. Vinyl chloride can be man-
ufactured starting with ethylene or ethyl chloride. The presence of chlorine and a double bond, along with
the phenomenon of resonance, cause the reactivity of VC to be less than ethylene and ethyl chloride.
VC is soluble in alcohol, very soluble in ether and carbon tetrachloride, but sparingly soluble in pure water.
The quantity of VC that dissolves in water will depend on the partial pressure of the gas above the solution.
If the partial pressure of the gas above the water is reduced, VC will escape into the gas phase. Chemical
reactions may occur with water impurities, which may tend to inhibit escape of vinyl chloride. Certain salts
do have the ability to combine with VC. For example, soluble silver and copper salts increase the solubility
of VC in water by forming complexes In addition to the above salts, VC, like other olefins, will complex
with ferrous chloride, platinous chloride, iridium dichloride, mercurous chloride, and a host of other salts.
Hence, the residence of VC in water could be affected by the presence of certain salts.
2.2 CHEMICAL PROPERTIES 2 6
The most important reactions of the olefinic hydrocarbons are related to additions of various compounds
to the double bond—for example, hydrogen peroxide, halogens, haloacids, halohydrins, oxides of nitrogen,
sulfuric acid, and ozone. Only a few, namely, hydrogen peroxide, oxides of nitrogen, sulfuric acid, and of
course ozone, should be of importance in ambient air. The case of formation of free radicals of importance
under the conditions of photochemical activity is allylic > 3° >2° > 1° >CH3 > vinyl. However, the
stability of the free radical is in the reverse order. In some polluted air, particularly in the presence of
ozone, reactions would be expected to occur.
In the specific case of VC, the halogen atom attached to the carbon-to-carbon double bond is generally
inert. When forced to react, hydrogen chloride is extracted from VC, with the resulting formation of
acetylene. Similarly, the hydrogen atoms attached to double bonded carbon atoms are highly stable in
substitution reactions. The order of reactivity of the hydrogen atom is allylic > 3° > 2° > CH4 > vinylic.
-------
Table 2.1. PHYSICAL CHARACTERISTICS OF VINYL CHLORIDE1
Formula
Molecular weight
Vapor pressure, 21.1°C
Specific volume, 21.1°C
Boiling point, 1 atm
Specific gravity, gas 15°C, 1 atm (air = 1)
Density of liquid, -20°C
Critical temperature
Critical pressure
Critical density
Latent heat of vaporization at boiling point
Latent heat of fusion at melting point
Specific heat
Liquid 20°C
Gas 25° C, 1 atm
Viscosity of liquid, -20°C
Flammable limits in air
Autoignition temperature
Dielectric constant, 17.2°C
Surface tension, -20°C
Refractive index, n"10D
Solubility in water, 24°C, 1 atm
Conversion factors, 25°C, 1 atm
1 ppm
1 mg/liter
CH2:CHCI
62.50
34 psig (2.4 kg/cm2 gauge)
6.2ft3/lb (387.0 ml/g)
7.0°F (-13.9°C)
2.15
0.9834
317.1°F (158.4°C)
774.7 psia 52.7 atm or 54.4 kg/cm2
absolute
0.370 g/ml
79.84 cal/g
18.14 cal/g
0.38 cal/g
0.205 cal/g
0.278 centistoke (0.2734 centipoise)
4.0 to 22.0 percent (by volume)
881.6°F (472°C)
6.26
22.27 dynes/cm
1.4046
0.11 g/100 g water
2.56 mg/m3
391 ppm
The importance of vinyl chloride lies in its ability to polymerize readily in the presence of ultraviolet light
or peroxides. The product is polyvinyl chloride (PVC), a highly useful plastic containing the basic structure:
I Cl F
: — c
C
I C
' r
'
i
I H H P
2.3 REFERENCES FOR SECTION 2
1. Braker, W. and A. L. Mossman. Matheson Gas Data Book, 5th Edition. East Rutherford, N. J.,
Matheson Gas Products, 197L p. 561.
2. Morrison, R, T. and R. N. Boyd. Organic Chemistry, 2nd Edition. Boston, Allyn and Bacon, Inc., 1970.
3. Fieser, L. F. and M. Fieser. Organic Chemistry. Boston, D. C. Health and Co., 1944.
4. Karrer, 0. Organic Chemistry. New York, Elsevier Publishing Co., 1947.
5. Porter, C. W. and Stewart. Organic Chemistry. New York, Ginn and Co., 1943.
6. Wheland, G. W. Advanced Organic Chemistry. New York, Chapman and Hall Ltd., 1948.
6 VINYL/POLYVINYL CHLORIDE
-------
3. MEASUREMENT TECHNIQUES
3.1 SAMPLING METHODS
3.1.1 Grab Samples
The least expensive approach to monitoring vinyl chloride monomer (VC) concentrations is to collect field
samples in a suitable manner and return them to a central laboratory for analysis. Using this approach,
samples are collected throughout a suspected problem area with a minimum of power, a minimum of
equipment, and with unskilled personnel. Grab samples are collected, as described in the interim
procedures,1 in Tedlar bags or stainless steel canisters. Varying degrees of loss, from 0 to 10 percent per
day, have been reported when VC in air was stored in Tedlar bags. Leaky bags may be responsible for losses.
Wall losses and permeability of the VC through the walls of the plastic bag do not appear to be a problem
with Tedlar bags tested at concentrations of about 10 ppm (25,600 ^ig/m3) or above. VC in pure air appears
to be stable. In the presence of nitrogen dioxide, which absorbs solar radiation at about 290 nm, secondary
reactions involving ozone (produced by the photolysis of nitrogen dioxide) and VC occur. It may be
possible to spike the air sample with a free radical or ozone scavenger to stabilize the VC. Direct
photoexcitation of VC is not expected to occur because solar radiation below 290 nm does not reach the
lower atmosphere.
Evacuated stainless steel canisters are more rugged, and more easily stored and transported than Tedlar
bags. These canisters need only a silicone septum through which a needle can be inserted to evacuate the
system to a low pressure. The needle is withdrawn and the septum seals itself, thus maintaining a vacuum
until a sample is ready to be taken. At the sampling site, a needle is again inserted, and the air sample is
allowed to fill the canister. The needle is withdrawn, and the septum seals itself again. At the laboratory, an
aliquot of the sample is removed with a gas-tight syringe and injected directly into a gas chromatograph or
other measuring device.
The above procedure yields a short-term concentration. Because the discontinuous nature of the emissions
produces pockets of high VC concentrations, this method does not provide an average dosage over a
prescribed period. Continuous monitoring is possible with the Wilkes Scientific MIRAN Portable Gas
Analyzer, which has a minimum detectable level of 1 ppm (2560 /ug/m3).
3.1.2 Liquid Scrubbers
Very little information is available concerning the use of liquid scrubbers for the collection of VC. The
physical properties of VC are such that it is not easily trapped by a liquid unless some complexation
reaction can be produced. Some salts have been reported to complex VC, but they have not been
thoroughly investigated for this purpose. For monitoring air pollutants, liquid scrubbers introduce
collection, handling, and stability problems that render the technique impractical.
3.1.3 Solid Scrubbers
Solid scrubbers are more easily handled and transported, and have fewer collection problems than liquid
scrubbers. Activated charcoal has been useful for the collection of gases and vapors including VC. The
capacity of charcoal for VC is limited; hence, problems have resulted from the use of small tubes and large
sampling volumes. It is imperative that all newly purchased charcoal be reactivated under nitrogen to
maintain its absorption capacity and to remove impurities that may interfere with the analysis. Charcoal
-------
was selected as the collection medium in the interim procedure in order to obtain time-weighted averages.
Multiple sections were specified to ascertain the quantity of charcoal required under field conditions. It is
not yet known how the procedure will respond under field conditions. The quantity of charcoal required to
collect VC under the most adverse conditions-such as relative humidity close to 100 percent-needs to be
determined.
Other solid scrubbers may be more suitable for VC collection. Hollis and Hayes2 reported long retention
times of low molecular weight hydrocarbons and halogenated hydrocarbons on porous polyaromatic
polymer beads. Williams and Umstead3 determined a number of halogenated hydrocarbons by first
concentrating the sample on Porapak Q & S. The materials were thermally desorbed at 100°C for analysis.
A microcoulometer with a silver cell was used to determine VC at the 10-part-per-billion (ppb)
(25.6-/;g/m3) level. Lonneman4 used carbowax under cryogenic conditions to concentrate the sample and
analyze concentrations of 100 parts per trillion by gas chromatography. More recently, Bellar5 successfully
concentrated VC from aqueous solution by adsorption on carbonsieve B. Quantitative recoveries were
obtained from aqueous solution containing from 5 nanograms (ng) to 5 micrograms (jug) of VC.
All solid scrubbers should be evaluated under simulated field conditions. Commercial permeation devices6
are available that will generate low levels of VC in air. The resulting mixture is diluted with humid air. In
this manner, collection and recovery efficiencies can be more definitively established. The stability of VC
on storage in the presence of reactive pollutants in the atmosphere is not known. Some investigators have
reported that VC might be polymerizing on these scrubbers. Lajos7 demonstrated that hydroquinone
improved recovery of VC from charcoal without reducing its adsorption capacity.
3.2 SAMPLE PREPARATION
Grab samples present no preparation problems if the amount to be analyzed is greater than 0.02 ppm (51
Atg/m3), since aliquots are injected directly into a gas chromatograph. If the sample size is sufficient, air
samples can be concentrated to detect levels of 0.2 ppb (0.51 jug/m3). Brown1 reports good recoveries from
charcoal by extracting with carbon disulfide. This is an excellent solvent for gas chromatography when
using the flame ionization detector. This detector gives no response to carbon disulfide under the usual
operating conditions, and solvent interfaces are eliminated. Keenan,8 however, observed that appreciable
quantities of VC evaporated into the head space above the liquid when carbon disulfide was used to extract
VC. Total recovery of VC in both the liquid and the gas phase was only 80 percent. When Keenan8
extracted VC with tetrahydrofuran (THF), he obtained a recovery of 88 percent, with less diffusing into
the head space than was evident with carbon disulfide.
3.3 ANALYTICAL METHODS
In selecting methods suitable for measuring VC concentrations in ambient air, two factors must be
considered. It is desirable that the method employed be capable of measuring in the part-per-million to the
part-per-billion range and, because emissions may be discontinuous, it is desirable that the method be
capable of responding to high concentration peaks as well as to low, background concentrations.
Designing or describing a measurement technique for a particular purpose requires that three major criteria
be satisfied: sensitivity, accuracy, and specificity, In addition, practicality and economics are important
considerations in the development of new analytical methods. As a general rule, it is most desirable to
measure a pollutant or chemical species directly in the matrix or phase—gas, liquid, or solid—in which the
material is generally encountered. This rule precludes any loss or transformation of the material to a
nondetectable form. ;
3.3.1 Spectrophotometry
VC absorbs infrared radiation in the gas phase. The absorption bands at 941 or 917 cm"1 have been used to
quantify VC. Because interfering substances (Table 3.1) are present in ambient air,8 the spectrophotometric
method is not entirely specific for VC. Multiband measurement and data processing techniques are available
8 VINYL/POLYVINYL CHLORIDE
-------
Table 3.1. POSSIBLE INTERFERENCES WITH VINYL CHLORIDE ANALYSIS8-3
Compound
Acrylonitrile
Allyl chloride
Chlorobromomethane
Chloroform
Ethylene
Ethylene dichloride
Freon-1 1
Freon-12
Freon-1 13
Melhacrylonitrile
Methyl chloroform
Methyl chloride
Methyl methacrylate
Perchloroethylene
Styrene
Tetrahydrofuran
Trichloroethylene
Toluene
Vinyl acetate
Vinylidine chloride
Vinylidine fluoride
Vinyl chloride analytical bands,
cm"1
1626
W
W
W
W
M
s
s
1020
M
S
W
W
s
917
W
S
s
W
s
M
S
W
s
M
M
M
W
W
S
719
W
W
W
s
W
W
s
M
S
W
aW = weak; M = moderate; S = strong.
to correct for these interferences, but additional instrumentation is required. The Fourier transformation
system is an excellent example of a refinement in this technique. The cost, however, of this type of system
with the refinement would be prohibitive for routine monitoring. Infrared analyzers are not sufficiently
sensitive for trace quantities of VC in air since effective optical paths of 20 meters are required to achieve a
lower limit of detection of 1 ppm (2560 ^g/m3). Accuracies of ±10 percent are attainable when the
analyzer is properly calibrated with standard gas mixtures. Although the technique is adaptable to
continuous monitoring, it is impractical as a multipoint detector of the type generally required to
characterize pollutant levels in a problem area. Economics dictate the use of this technique as a research
tool or as a laboratory instrument. Air samples, either instantaneous or integrated, can be collected,
concentrated if necessary, and returned to a central laboratory for analysis by infrared spectrophotometry.
3.3.2 Gas Chromatography
Gas chromatography (GC) is an analytical technique that separates a complex mixture into its component
parts by partitioning the chemical material between a gas and a liquid or solid. The technique is highly
popular because of its versatility in solving analytical problems. A wide variety of materials and conditions
are available that can be used to achieve separations effectively and inexpensively, even for closely related
compounds.2'9"21
3.3.2.1 Column Material—A list of column materials that have been used to separate vinyl chloride and
related compounds is shown in Table 3.2. This by no means is a complete list; there are other systems that
can be designed. It is difficult to select the best column material based on the available literature because
Measurement Techniques 9
-------
Table 3.2 COLUMN MATERIALS AND LIQUID SUBSTRATES
SEPARATING VINYL CHLORIDE
Column materials and liquid substrates
Porapak, Q.
Silicone oil DC 550
Silver nitrate/ethylene glycol
30% silicone oil and polyethylene glycol
Disodecyl phthalate/carbowax
Carbowax 4000
25% di butyl ph thai ate
5 to 15% silicon rubber SE-30
Silicone grease
Porapek, -S
Poly (methyl phenyl siloxane)
Tricresyl phosphate
30% dioctyl sebacate
Carbowax 1500 or carbopack A
References
Forris
Levadie
Smith
Vyakhirev
Hannon
Newman
Martur
Hinshaw
Esposito
Koenig
Popova
Vlasov
Zalinyan
Brown
18
22
23
9
24
25
11
12
13
14
16
20
21
1
quantitative data on column efficiencies and height equivalent to a theoretical plate are not generally
provided, nor are the objectives of the reported method always similar to the EPA objectives. It is
particularly important that VC be separated from hydrocarbons and Freons. Alternatively, more specific
detectors must be used in combination with GC.
3.3.2.2 Detectors—Detectors that are used in combination with GC columns are also varied. Highly
selective and highly sensitive detectors which will detect quantities of material down to 10"12 grams
(picograms) are available. Completely automated GC instruments are commercially available for
environmental monitoring. To measure VC, all that need to be changed on some of these instruments are
the column materials and operational parameters. With rare exceptions, measurements are not made
continuously, but are made by taking instantaneous samples at short intervals.
The flame ionization detector (FID) is a general purpose detector which responds to most organic
compounds, has a wide linear range of several orders of magnitude, and a sensitivity that enables
measurement as low as parts per billion. The response to a chemical compound generally varies with the
number of carbon atoms. However, certain carbon atoms yield either reduced response or no response when
the carbon atom is attached to atoms other than hydrogen, for example, chlorine, oxygen, and sulfur. The
FID is insensitive to almost all inorganic gases and compounds. The minimum detectable concentration for
VC using a 10-ml sample of gas is 0.01 ppm (25.60 jUg/m3). When coupled to a GC column to achieve
separation, the FID has been the detector of choice because of its sensitivity and minimal cost for the
analysis of complex organic mixtures. The GC-F1D combination has been used under field conditions, but
power requirements and the need for hydrogen gas reduce the practicality of the instrument for routine
monitoring.
The thermal conductivity detector (TCD) is mentioned here only for historical purposes. The detector
measures changes in the heat capacity of the carrier gas, usually helium or hydrogen, when materials elute
from the column. The sensitivity is low when compared with other available detectors. It is not suitable for
trace analysis. In addition, the TCD responds to water vapor. This response causes problems in identifying
and measuring compounds of interest.
A third group of detectors falls under the general classification of direct current ion chambers. These
include argon ionization, helium ionization, micro cross-section detectors, and, most important of all,
10
VINYL/POLYVINYL CHLORIDE
-------
electron capture detectors. The argon detector consists of two or three parallel electrodes and a radioactive
source, usually strontium-90, that excites the argon carrier gas. When chemical compounds elute from the
column, they are ionized by the excited argon. Under a voltage gradient of up to 1000 volts per centimeter
these ions produce an increase in current flow across the plates or electrodes which is proportional to the
concentration of the eluting material,. The sensitivity is good, but the method is nonspecific and
temperamental.
The helium detector is similar in design to the argon detector, except that helium is used as the carrier gas.
Voltage gradients as high as 2000 volts per centimeter can be applied across the plates. Tritium is frequently
used as the excitation source. This detector is also temperamental, highly sensitive, and nonspecific, but is
usually recommended for detecting traces of inorganic gases by gas-solid chromatography.
The electron capture detector (BCD) is similar to the other DC-ion chambers in design. Nitrogen or argon is
used as the carrier gas and tritium or nickel-63 is used as the radioactive source. Low voltages (5 to 25 volts)
are applied across the plates, usually in a pulsating mode, to eliminate polarization of the electrodes. The
detector is specific and highly sensitive to halogenated materials and other materials that absorb electrons.
It has a smaller dynamic range and is more temperamental than the FID. The sensitivity of the BCD for VC
is poorer than that of the FID because of the presence of a single chlorine atom in themolecule. ' °'2 6 More
recent data indicate the sensitivities tabulated in Table 3.3.26
The GC-ECD combination has the advantage of requiring only a gas cylinder of nitrogen. Because of its
specificity, complete resolution of VC by the GC column is no longer mandatory. Battery operated
GC-ECD instruments have been manufactured commercially.
The microcoulometer is a highly sensitive and very accurate detector of chloride ion. The specificity of this
detector is increased when used with gas chromatography. Chlorinated hydrocarbons, such as VC, are
pyrolyzed as they elute from the column to form gaseous hydrogen chloride which reacts with the solution
to precipitate silver chloride and disturbs the electrical balance at the positive, silver electrode. The
coulometer regenerates silver ions (Ag+) until the electrical balance is restored. The detector will respond to
any substance which precipitates silver. However, depending on column and pyrolysis conditions, these
potential intereferences can be eliminated. With electrochemical efficiency of close to 100 percent, the
coulombs generated to restore the balance are proportional to the quantity of chloride ion in accordance
with Faraday's law. The detector is highly accurate because the coulomb is a primary standard, and hence,
standard reference materials are not absolutely essential. The sensitivity of the detector for VC is of the
order of a few nanograms. Electrical power requirements make this system impractical for field use, but it is
excellent for laboratory use.
Another electrochemical detector is the conductivity device developed by Coulson for use with gas
chromatography.27'28 The conductivity detector measures either water-soluble ions or gases that produce
soluble ions when they react with water. The effluent material is either oxidized or reduced in a small
furnace prior to reaching the detector. Depending on the mode of operation, the detector response can be
restricted to hydrogen chloride, sulfur dioxide, or sulfur trioxide. High sensitivity is attainable because of
the solubility of these gases in water and the high mobility of the hydrogen ion produced. Although
sensitivity on the order of a few nanograms is possible, the ultimate sensitivity depends on the geometry or
Table 3.3. SENSITIVITY OF SELECTED DETECTORS USED IN GAS CHROMATOGRAPHY26
(grams)
Compound
Vinyl chloride
Trichloroethylene
Thermal
conductivity
detector
2 x 1CT6
2.2x10-*
Argon'
detector
1.9 x 10"9
1.0 x 10~*
Flame
ionization
detector
2.2 x 10"9
8.5 x 10"9
Electron
capture
detector
2.3x10~9
2.0 x 10~n
Measurement Techniques
11
-------
the cell constant. More recently, a conductivity cell has been designed by Hall29 which yields higher
sensitivities than the Coulson detector. Again, specific detectors reduce the demand on the column to
resolve compounds with elution times that are close to those of VC Power requirements reduce the
practicality of this detector under field conditions.
The Beilstein test is a classical flame test for detecting halogens in the presence of copper. This test is the
operating principle employed in a flame photometric detector that has been used with a gas chromatograph.
The detector uses a photomultiplier tube to measure the characteristic spectrum produced by halogenated
substances. The sensitivity of this detector has been enhanced by use of iridium metal.30 It may be useful
without GC as a continuous monitor for gaseous halogenated substances. By using a filter to remove
inorganic halogens, this measuring technique would give an index of the total quantity of halogenated
materials in a given area However, to achieve specificity for VC, a GC column would be required.
Practicality for field use is equivalent to that of other flame detectors.
Chemiluminescence detectors are used in continuous monitors that measure the quantity and type of light
that is produced by reacting certain compounds with ozone-the determination of ozone by reacting with
ethylene or the determination of nitric oxide by reacting with ozone. Recently, the monitor has been
adapted by EPA scientists to measure VC concentrations. However, a GC column is required to obtain
specificity. Current evaluation of the monitor indicates that a sensitivity of a few parts per billion is
achievable. .
The alkali flame ionization detector and stacked thermionic detectors have also been used as GC detectors
to measure halogenated compounds. The mass spectrometer when connected to a GC column is particularly
useful in identifying an unknown compound and for unequivocal confirmation of VC. All of these
detectors have sensitivities within one or two orders of magnitude of each other. The necessity of one or
more gas cylinders and power requirements limit their utility for field use.
3.3.3 Other Analytical Methods
Coulometry has been used to measure olefinic hydrocarbons by reaction with electrogenerated bromine.
This technique is not useful for measuring VC in ambient air because of the long reaction time required for
the bromination of olefins. In addition, reducing substances such as sulfur dioxide would interfere by
consuming bromine. Oxidant would cause a negative interference. Wet chemical methods,1 >3' based on the
bromination of VC and then the titration of excess bromine, have also been developed. The sensitivity is
only 0.1 mg. Olefins, aromatic compounds that readily add bromine, and reducing agents interfere with this
wet chemical method.
Vinyl chloride in air has been analyzed colorimetrically following collection on activated carbon.28 The VC
was extracted and oxidized to formaldehyde, and the formaldehyde determined by reaction with
chromotropic acid. Ethylene and methanol interfere with the method. Sensitivity is only a few micrograms.
Polarography32'33 has been used to measure VC by bromination at the dropping mercury electrode. When
this procedure was applied to volatile VC from plastics, it gave a higher value than obtained by the analysis
for total chloride. Hence, interfering volatile materials are present. Sensitivity for VC was only 70
micrograms per milliliter.
3.4 AUTOMATED MONITORING
Completely automated monitoring instruments, which are commercially available, can be easily modified to
measure vinyl chloride. For example, the carbon monoxide-methane analyzer-which includes a precolumn,
a gas chromatographic column, and a flame ionization detector—may be used to quantitate vinyl chloride
by simply changing the column packing material and operational parameters. The precolumn removes most
of the higher molecular weight hydrocarbons and preserves the integrity of the main column. Using this
type of instrument, 5 to 10 samples can be analyzed per hour over a 24-hour period.
12 VINYL/POLYVINYL CHLORIDE
-------
Portable gas chromatographs that are less expensive than those described above are commercially available
for monitoring VC, but require an attendant. Both the FID and the BCD are available with these
instruments and have a reported sensitivity of 0.1 ppm (256 jug/m3) VC.
3.5 INTEGRATED SAMPLES USING ADSORBENTS
Integrated values for 8-hour to 24-hour sampling periods can be obtained by using commercially available
collection columns. These columns can become a part of a personal monitor, having a self-contained pump
worn by a worker, or be part of a permanent installation. During the course of monitoring, a steady stream
of sample is drawn through the column containing an adsorbent material, generally activated carbon. The
column is capped and returned to a central laboratory for analysis. A flasher technique is used whereby the
tube is heated to drive off the collected compounds for gas chromatographic analysis. Another procedure
calls for extraction of the vinyl chloride with carbon disulfide. The resulting solution is then analyzed
chromatographically using a flame ionization detector. Both methods are sensitive to 1 ppb (2.56
jug/m3).34-35
3.6 REFERENCES FOR SECTION 3
1. Brown, D. EPA Region IV, Athens, Ga. Personal Communication with Quality Assurance and
Environmental Monitoring Laboratory, National Environmental Research Center, Research Triangle
Park, North Carolina. 1974.
2. Hollis, O. L. and W. V. Hayes. Gas-liquid Chromatographic Analysis of Chlorinated Hydrocarbons with
Capillary Columns and Ionization Detectors. Anal. Chem. 54:1223-1226, 1962.
3. Williams, F. W. and M. E. Umstead. Determination of Trace Contaminants in Air by Concentration on
Porous Polymer Beads, Anal. Chem. 40:2232, 1968.
4. Lonneman, W. A. Measurements of Vinyl Chloride from Aerosol Sprays. U. S. Environmental
Protection Agency, National Environmental Research Center, Research Triangle Park, North Carolina.
Unpublished, April 1974.
5. Bellar, T, National Institute of Occupational Safety and Health, Cincinnati, Ohio. Personal
Communication with B. W. Gay, Chemistry and Physics Laboratory, National Environmental Research
Center, Research Triangle Park, North Carolina. 1974.
6. O'Keeffe, A. E. and G. C. Ortman. Primary Standards for Trace Gas Analysis. Anal. Chem.
3S(6):760-763, 1966.
7. Lajos, R. Separation of Vinyl Chloride from Gases. Chemical Abstracts. 50:36690r, 1974.
8. Keenan, R. R. G. D. Clayton Associates, Southfield, Mich. Private Communication with F. P.
Scaringelli, U. S. Environmental Protection Agency, Research Triangle Park, North Carolina. 1974.
9. Vyakhirev, D. A., Z. S. Smolyan, L. E. Reshetnikova, N.D. Demina, M. I. Vlasova, and A. A. Karnishin.
Analysis of Vinyl Chloride by Gas Chromatography. Tr. Pa. Khim, i Khim. Tekhnol. (U.S.S.R.)
4:490-497, 1962. (Chem. Ab. 57:4040f).
10. demons, C. A. and A. P. Altshuller. Responses of Electron Capture Detector to Halogenated
Substances. Anal. Chem. 38(1): 133-136, 1966.
Measurement Techniques 13
-------
ll.Martur, V. G., S. A. Antipova, and V. S. Kozlova. Analysis of Mixtures of Fluoro and Chloro
Derivatives of Ethane and Ethylene on the KhL-3 Laboratory Chromatograph. Ukr. Khim. (Kiev).
J2(4):391-392, 1966. (Chem. Ab. <55:4667e).
12. Hinshaw, L. D. Gas-Chromatographic Determination of Chlorinated Hydrocarbons in 1,2-dichloro--
ethane. J. Gas Chromatog.4(8):300-302, 1966.
13. Esposito, G. G. and M. H. Swann. Identification of Aerosol Propellants in Paint Products by Gas
Chromatography. J. Paint Technol. 59(509):338-340, 1967.
14. Koenig, H. Separation, Detection, and Quantitative Determination of Aerosol Propellants by Gas
Chromatography. Fresenius Z. Anal. Chem. (Wiesbaden). 2J2(6):427-432, 1967. (Chem. Ab.
6S:65487n),,
15. Balandina, L. A. and A. I. Subbotin. Chromatographic Analysis of Products of High-temperature
Chlorination of Ethylene. Zavod. Lab. (U.S.S.R.). 34(2):154, 1968. (Chem. Ab. <5
-------
29. Hall, R. C. A Highly Sensitive and Selective Microelectrolytic Conductivity Detector for Gas
Chromatography. J. Chromatog. Sci. 72:52-160, 1974.
30. Bowman, M. C. and M. J. Beroza. Indium-sensitized, Flame-photometric Detector for Gas Chromatog-
raphy of Halogen Compounds. J,, Chromatog. Sci. 9:44-48, 1971.
31. Tsendrovskaya, V. A., K. I. Stankevich, and I. S. Reisig. Selection of a Method for Detemining Volatile
Substances Separated from Some Plastics. Primen. Polim. Mater. Izdelii Nikh (U.S.S.R.). No. 1,
418-425, 1969. (Chem. Ab. 75:64701k).
32. Ryabov, A. V. and G. D. Panova. Application of the Polarographic Method in Analysis of Unsaturated
Organic Compounds. Doklady Akad. Nauk (U.S.S.R.). 99:547-549, 1954. (Chem. Ab. 49:5212g).
33. Meshkova, O. V., V. N. Dmitrieva, and V. D. Bezuglyi. Polarographic Analysis of Waste Waters from
Poly-(vinyl chloride) Production. Khim. Prom. (Moscow). 47(4):271-273, 1971. (Chem. Ab.
75:21063m).
34. Patton, J. C. Bendix HS-10 Flasher and Personal Monitoring System—A Summary Report of Current
Data. Bendix Process Instruments Division. Rbnceverte, West Virginia. August 26, 1974.
35. Tentative Method for the Determination of Vinyl Chloride in the Atmosphere by 24-hour Integrated
Sampling. Quality Assurance Branch, Environmental Monitoring and Support Laboratory, U.S.
Environmental Protection Agency, Research Triangle Park, North Carolina. September 11,1974.
Measurement Techniques 15
-------
-------
4. ENVIRONMENTAL APPRAISAL
4.1 SOURCES 1-2
Commercial processors in the United States employ several basic methods in the manufacture of vinyl
chloride (VC) monomer and polyvinyl chloride (PVC) polymer resins.
Vinyl chloride monomer production processes employ one of the following:
• The acetylene plus hydrogen chloride reaction.
• The direct chlorination of ethylene and dehydrochlorination.
• The balanced direct and oxychlorination of ethylene and dehydrochlorination.
The two general methods for the production of VC are the acetylene plus hydrogen chloride reaction:
HC==CH + HC1 -H2C CHC1 (1)
and the thermal dehydrochlorination of 1,2-dichloroethane:
C1CH2 CH2C1—>H2C CHC1 + HC1 (2)
In the second method, VC plants are integrated with an ethylene dichloride production unit. The overall
processes differ primarily in the manner in which the ethylene dichloride is produced.
Nine producers in the United States operate balanced plants in which ethylene is chlorinated by a mixture
of hydrogen chloride and air to produce ethylene dichloride. Part of the hydrogen chloride used for this
process is in the form of recycled products from the thermal dehydrochlorination of ethylene dichloride.
One producer uses the balanced oxychlorination process with the exception that oxygen is used for the
oxychlorination reaction sequence.
Three producers use an integrated process in which the ethylene dichloride is produced only by the direct
chlorination of ethylene. Hydrogen chloride is recovered from the dehydrochlorination step, but is not
recycled into the process.
Polyvinyl chloride is produced at approximately 40 plants in the United States using one of the following
processes:
• Suspension polymerization (78 percent of total production).
• Emulsion polymerization (13 percent of total production).
• Bulk polymerization (6 percent of total production).
• Solution (3 percent of total production),
17
-------
Gaseous VC is emitted at both VC and PVC resin plants. It is distributed into the atmosphere surrounding
the emissions source in patterns that depend on the amount of VC released, the nature of the plant area
from which it is released, and the meteorological conditions.
In 1974 approximately 2.6 billion kg of VC and 2.1 billion kg of PVC resin were produced in the United
States. VC manufacturers operated at approximately 85 percent of capacity in 1974. Emission data so far
supplied to EPA by industry indicate that the total VC escaping to the atmosphere exceeds 100 million kg
per year.
Vinyl chloride losses from the average VC plant are estimated to be about 0.45 kg/100 kg of VC produced,
and from the average PVC plant approximately 4 kg/100 kg of PVC produced.
Two additional features of the industry are significant in terms of potential VC concentrations in the
atmosphere near plants. VC plants are clustered primarily in areas along the Texas and Louisiana Gulf
Coast, and some PVC plants are located close to, or even adjacent to, the. VC production sites. This
"clustering" of plants is greatest in the Pasadena-Deer Park, Texas, region and in the Baton Rouge,
Louisiana, area (Figure 4.1).
Vinyl chloride monomer-producing companies are listed in Table 4.1. Included in this table are the
companies and their geographical locations, population figures for adjacent communities, and calculated VC
emission levels, assuming losses of 4.25 percent. Polyvinyl chloride producers are similarly listed in Table
4.2.
4.2 CONCENTRATIONS IN AMBIENT AIR
Few data are available to characterize the concentration of vinyl chloride in ambient air. In view of the
potential danger to human health associated with exposure to vinyl chloride, a preliminary field study was
initiated in early 1974 through EPA's Regional Offices to obtain more extensive and reliable data on
ambient levels in relation to industrial emission sources. Although air monitoring procedures were
established by EPA for this study, the varying degrees of resources and expertise available to the individual
Regional Offices produced data that are difficult to compare on a nationwide basis. Based on these limited
data, however, ambient concentrations of vinyl chloride exceeded 1 ppm (2560 jug/m3) less than 10 percent
of the time in residential areas located in the vicinity of plants producing VC or PVC.
Most of the data presented in this section are from instantaneous (grab) samples collected at varying
distances from an emission source. As anticipated because of the discontinuous nature of the production
processes and resultant emissions, the sampling revealed a wide range of concentrations. The maximum
concentration of vinyl chloride observed in ambient air was 33.0 ppm (84,480 /Kg/m3) at a distance of 0.5
km from the center of the plant. In the following discussion, the VC or PVC production plant is identified
only by the EPA Regional Office4 that conducted the sampling.
4.2.1 Region I Plant
Summary data on vinyl chloride concentrations in grab samples taken at varying distances from the center
of a production facility in Region I are reported in Table 4.3. Samples taken at sites A through P were
collected at 2-hour intervals using Tedlar bags or syringes. The remaining sampling sites are those at which
detectable vinyl chloride odors existed; these sites were chosen in an attempt to obtain maximum
concentrations. The frequency with which the concentration of vinyl chloride exceeded 1 ppm (2560
/ig/m3) is shown in the last column. More than 90 percent of the values obtained at this plant were below
the minimum detectable concentration (0.06 ppm or 153 /Kg/m3).
The average vinyl chloride concentrations collected at four plant sites in a 24-hour period by charcoal
scrubbers are shown in Table 4.4. All data shown in the table were corrected to standard ambient air
conditions of 25°C and 1 atmosphere. On May 10, 1974, the average concentration exceeded 1 ppm (2560
Hg/m3) at site A, which was located 0.3 km from the center of the plant. The site, however, may have been
closer to the actual vinyl chloride emission point.
18 VINYL/POLYVINYL CHLORIDE
-------
Environmental Appraisal
19
-------
Table 4.1. VINYL CHLORIDE MONOMER PLANT EMISSION DATA
Company and
location
Allied Chemical Corp.
Baton Rouge, La.
American Chemical Corp.
Long Beach, Calif.
Continental Oil Co.
Lake Charles, La.
Dow Chemical Corp.
Freeport, Tex.
Plaquemine, La.
Oyster Creek, Tex.
Ethyl Corporation
Baton Rouge, La.
Pasadena, Tex.
B. F. Goodrich Co.
Calvert City, Ky.
Monochem, Inc.
Geismar, La.
P.P.G. Industries
.Lake Charles, La.
Guayanilla, P.R.
Shell Chemical
Deer Park, Tex.
Norco, La.
Tenneco Chemical, Inc.
Pasadena, Tex.
Total
City
population3
165,963d
358,633e
77,998
11,997f
7,7399
— f
1 65,963d
89,277h
31,627'
7,739
77,998
—
12,773h
-
89,277h
Production
capacity,
June 1974,
106 kg/yr
155
75
330
80
155
320
120
70
455
135
135
225
410
320
115
3100
Estimated
VC
emissions,13
106 kg/yr
0.7
0.4
1.5
0.4
0.7
1.4
0.5
0.3
2.0
0.6
0.6
1.0
1.8
1.4
0.5
14
Type
of
process0
B
B
B
DC
DC
B
B
DC
B
B
B
-
B
B
A
a1970 Census data. The extent of exposure of these populations to VC is not known.
Extrapolated figures based on estimated atmospheric emission loss of 0.45 percent of VC produced in monomer plants and
plant operation at full capacity.
cBalanced (B)—combination of direct chlorination and oxychlormation process in which the hydrogen chloride produced in
cracking is recycled to the oxychlorination process. Direct chlorination (DC). Acetylene (A).
dBaton Rouge Parrish (Countyl-302,031.
eLos Angeles County-7,032,075.
f Brazoria County-108,312.
9Plaquemine Parrish (County)—25,225.
"Harris County-1,741,912.
' Population given is for Paducah, Ky.
20
VINYL/POLYVINYL CHLORIDE
-------
<
Q
co
co
O
Q.
oc
w
O
a.
HI
Q
cc
3
X
o
o
a.
X!
CO
1 >
'c/j O)
I ~*
Its
0 *"
.t m" t
CJ |^ ,>
o ^"o
Q.
"S "In
CD O>
>- o
H feL
CD
^
Q
O "5
a
a
c
Q
CO
CJ
o
Company and 1
CM ffl
Csi O
o in
CO CO
c
g
c
(a
Q.
00
•o
CM O
co m
^-" O)
co m
CJ
c
_-
CO
CJ
Products and Chem
Calvert City, Ky.
Pensacola, Fla.
<
CO
Csi
o
CM
C
0
Suspensi
CD
CO
CO
CO
00~
in
CO
uffer Chemical Co.
Long Beach, Calif.
ff>
CD CN
csi co
in o
CO 00
c
11
£ 5
Q.
co -g
00 §
CSI O
Csi CO
*-~ CST
CO
rden. Inc.
Illiopolis, III.
Leominster, Mass.
o
00
oq q
•sr •*
0 0
CM O
C
§1
11
00 c5
r^- *~
in co
co" to
CO
m
ntinental Oil Co.
Aberdeen, Miss.
Oklahoma City, Okh
8
q q
csi in
o in
in CM
c
§!
II
^ CO
csi r~
o r-«
CM" CN"
Q.
L-
imond Shamrock Co
Delaware City, Del.
Deer Park, Tex.
w
b
CM
CO
O
00
c
§1
|i
3?
OO CO
CO
CO
0)
in"
CO
CO
O)
D
. 0
Q. DC
4-*
£ DQ
ffi
CN q
•- in
01 in
q co
CM" in"
CM
6
o
0}
estone Tire and Rub
Perryville, Md.
Pottstown, Pa.
u.
CM
CM
in
in
c
0
Suspensi
CO
CO
CM
8
m
neral Tire and Rubbi
Ashtabula, Ohio
0>
Environmental Appraisal
21
-------
Q
oo
z
o
CO
LLJ
O
I-
a.
cc
o
a.
UJ
Q
OC
O
I
O
_l
o
CN
CJ
C >-
El a>
cu *eb
o •-
.tTin !r
CO ^ ^
CO *~ •*
CJ J" *£>
D_
H— .£J
CU CU
> O
1- a
t
g
£" JO
o ^
o
0-
c
g
Company and locat
o o co oo CD
CM f CM ^T CM
O O LO O LO
LO O CD CM CD
"~ ""
C
• — cf
V) ,-. ^~-
C .— O5
a-5^
3 ID
CO cu jQ
OJ _
CO O CM t-
co »- r-. CD
CD co ^~ CNJ
k. •. X 1.
CO CM «- CM '
LO CD «-
CO CO
. F. Goodrich Co.
Long Beach, Calif.
Henry, III.
Louisville, Ky.
Avon Lake, Ohio
Pedricktown, N. J.
CQ
o
^
8
""
LO
CO <-
r~" LO"
00
6
O
1_
oodyear Tire and Rubbf
Plaquemine, La.
Niagra Falls, N.Y.
CD
CM
*-
O
CO
c
0
IS)
C
CU
a
CO
CO
co"
a
o
O
reat American Chemical
Fitchburg, Mass.
CD
CO
C)
CD
C
o
't/>
c
CU
a
c/>
to
(V
1
Q.
O
CJ
C
0) (/>"
O 3
|J
^
P,
CM
«3
CD
C
CO
C
.2 r-
Suspens
emulsioi
1
co"
CO
LO
C
CU
o
lonsanto Company (to c
Springfield, Mass.
s
p co
CO O
LO r~
c
CO
c
o -^
cu — -
^ 3
<- LO
O) f^
o^ o
«-" oo"
Q.
ccidental Petroleum Cor
Burlington, N.J.
Hicksville, N.Y.
O
p oq
r-' T-
LO LO
CM ^
c
Emulsio
CM CM
LO" CD"
LT5
antasote Co.
Passaic, N.J.
Point Pleasant, W. Va.
OL
co
^
LO
C
g
't/j
c
CU
Q.
c/>
CO
CO
CO
in
co"
obintech. Inc.
Plainsville, Ohio
oc
22
VINYL/POLYVINYL CHLORIDE
-------
p
8
<3-
i
X
-E tf
. 0
.C Q.
U QJ
C it
•p
co
OJ
CO
0
00
c
g
c
CD
Q.
t/1
CO
If
CM
O_
CN
CO
uffer Chemical Co
Delaware City, Dei
CO
CO
O CM f
CO •-' •*
in o o
r-» co «-
c
c .2
g _£
c i
|-o
to co
^
C5 «— p»
«-" co" o>
<- 00
0
c
ineco Chemicals, 1
Burlington, N.J.
Flemington, N.J.
Pasadena, Texas
eB
r-
-a
c
CO
O —
•<£ c
c o
cu .^
s^
co1 8
00 00
O CO
en co
co" CD"
CO <-
CO
>
•5-
ion Carbide Corp.
Texas City, Tex.
South Charleston,
c
D
p
oi
o
in
T3
C
CO
c
o c
£.0
QJ t/>
a "5
3 1
CO cu
<0
CO
Q)
C CO
g .y
o -o
c c
t-' £
g 8
« •£
O cB
> CO
o a-
^ c
ii
*-• n
— E
Q. -5
0 ^
OJ c
£ o
.n >.
M_ CO
o c
£ 1
(/) "m
§
O
u
(U
i
ID
C
O
'•p
o
-g
0
a
OJ
E
"o
a
O)
c
^
_J
O
*-
Q>
O
OJ
Q.
q
^
0
c
g
$
E
aj
-o
»
CO
E
tt
cu
c
n
1970 Census data.^ The extent of expo
The statement on type of process is a pr
Extrapolated figures in 106 kg/yr based c
co -Q
u
c
0
4-1
O
.£.
O
Q.
cu
duction considered the same as homopol
Population given is for Paducah, Ky.
Los Angeles County— 7,032,075.
•D
cu *
Oklahoma County-526,805.
HarrisCounty-1,741,912.
Baton Rouge Parrish (Countyl-302,031
Henry County-53,217.
Plaquemme Parrish (County)— 25,225.
CM
co
co"
CM
co
c
o
o
c
o
01
c
CD
c
g
u
c
o
u
(U
13
c
3
C
CO
a
i
t/>
LU
,- 0) .C .- — , -* _
Environmental Appraisal
23
-------
Table 4.3. VINYL CHLORIDE CONCENTRATION IN GRAB SAMPLES TAKEN NEAR
REGION I PLANT
Site
A
B
C
D
E
F
G
H
1
J
K
L
M
N
0
P
TT
UU
VV
ww
XX
YY
ZZ
Distance,3
km
0.3
0.2
0.3
0.2
0.6
0.6
0.8
0.8
1.1
1.6
1.9
9.8
1.0
1.0
1.1
0.8
0.2
0
0
0.2
0.3
0
0.3
Number
of
samples
17
16
9
9
11
9
8
10
12
8
7
8
6
4
5
6
1
2
1
3
1
1
1
Maximum
concentration,
ppmb
6.0
0.30
0.22
0.9
0.6
0.24
NDe
ND
0.40
ND
0.24
ND
ND
ND
ND
0.32
ND
0.24
ND
2.6
0.16
5.7
ND
Mean
concentration,0
ppm
0.52
0.06
0.06
0.15
0.14
0.09
ND
ND
0.05
ND
0.06
ND
ND
ND
ND
0.11
ND
0.22
ND
1.48
0.17
6.15
ND
Number
>1ppmd
2
0 •
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
1
0
aDistance from center of plant to sampling site.
b1 ppm = 2560";Ug/m3.
cValues corrected to standard temperature of 25°C.
Number of vinyl chloride monomer concentrations above 1 ppm.
eNot detectable.
Table 4.4. VINYL CHLORIDE CONCENTRATION IN INTEGRATED 24-hour
SAMPLES COLLECTED BY CHARCOAL ABSORBER NEAR REGION I PLANT, 1974
Site
A
B
C
D
Concentration, ppma
May 9
0.021
0
0
0.141
May 10
1.15
0.005
0.009
0.010
May 13
0.165
0.020
0.029
0.090
a1 ppm = 2560 jUg/m3.
24
VINYL/POLYVINYL CHLORIDE
-------
4.2.2 Region IV Plant
At the present time, only the data from the initial study and summary data are available for the plant
studied in Region IV. In March 1974, 3 values out of 48 measured exceeded 1 ppm (2560 Atg/m3) (Table
4.5). The highest concentration was 2.2 ppm (5632 /zg/m3). The data collected in May 1974 around this
plant are summarized in Table 4.6. One instantaneous value of 33 ppm (84,480 jug/m3)was observed at a
distance of 0.5 km from the plant, and three mean values exceeded 1 ppm (2560 jug/m3). The data from
the 24-hour integrated samples indicated the highest value to be 0.55 ppm (1408 |itg/m3) (Table 4.7).
4.2.3 Region VI Plant
Summary data from grab samples taken near an EPA Region VI plant are shown in Table 4.8. The vinyl
chloride concentration exceeded 1 ppm (2560 jLtg/m3) in only three of the samples. At the plant property
line, the concentration ranged from below the detectable level to 7.8 ppm (19,968 jug/m3).
4.2.4 Region IX Plant
Data on vinyl chloride concentrations near a plant in EPA Region IX are presented in Table 4.9. Although
more extensive than those for the other Regions, the data provided are not comparable because the samples
are 10-minute averages on charcoal instead of instantaneous samples. Under these conditions, concentra-
tions of vinyl chloride as high as 3.4 ppm (8704 jug/m3) were found at a distance of 5 km from the plant.
Only 12 of the 180 determinations exceeded the 1-ppm (2560-/Lig/m3) level, however. The overall mean
value was 0.24 ± 0.44 ppm (614 + 1126 jug/m3). Integrated 24-hour values are given in Table 4.10.
Table 4.5. VINYL CHLOR IDE CONCENTRATION IN GRAB SAMPLES COLLECTED NEAR
REGION IV PLANT, MARCH 1974
Site
A
B
C
D
E
F
G
H
1
J
K
L
M
N
Distance,3
km
0
0.6
0.6
0.6
0.8
0.8
0.8
0.5
1.3
1.0
1.0
1.3
4.8
1.0
Number
of
samples
6
3
4
6
3
15
2
1
1
1
2
1
2
1
Maximum
concentration,
ppmb
2.2
0.58
1.26
0.29
0.24
0.39
—
—
—
—
—
—
—
-
Mean
concentration,
ppmb
0.84
0.40
0.33
0.12
0.16
<0.17
<0.17
<0.17
<0.17
<0.17
<0.17
<0.17
<0.17
<0.17
Number
>1 ppmb
2
0
1
0
0
0
0
0
0
0
0
0
0
0
a Distance from center of plant.
b1 ppm =2560/jg/m3.
Environmental Appraisal
25
-------
Table 4.6. VINYL CHLORIDE CONCENTRATION IN GRAB SAMPLES COLLECTED
NEAR REGION IV PLANT, MAY 1974
Site
A
B
C
D
E
H
I
L
N
P
Q
R
S
T
U
V
Distance,3
km
0.0
0.6
0.6
0.6
0.8
0.5
1.3
1.3
1.0
0.6
0.8
0.5
0.6
1.0
1.0
0.2
Number
of
samples
6
21
21
19
2
3
2
2
1
8
12
83
1
3
3
1
Maximum
concentration,
ppmb
1.7
5.6
5.8
2.8
0.10
1.2
1.6
NDC
ND
0.08
1.7
33.0
1.2
0.57
ND
ND
Mean
concentration,
ppmb
0.33
1.6
0.71
0.54
0.07
0.50
1.00
ND
ND
0.06
0.34
3.15
1.20
0.21
ND
ND
aDistance from center of plant.
b1 ppm = 2560jUg/m3.
CND = not detectable.
Table 4.7. VINYL CHLORIDE CONCENTRATION
IN INTEGRATED 24-hour SAMPLES COLLECTED
BY CHARCOAL ABSORBER NEAR REGION IV PLANT, 1974
Site
MSD
NFL
SOP
DPT
CP
Distance,3
km
0.6
0.6
0.3
0.2
1.0
Mean
concentration,
ppmb
0.16
0.07
0.55
0.10
0.01
a Distance from center of plant.
b1 ppm = 2560 jUg/m3.
26
VINYL/POLYVINYL CHLORIDE
-------
Table 4.8. VINYL CHLORIDE CONCENTRATION IN GRAB SAMPLES TAKEN NEAR
REGION VI PLANT, 1974
Distance,3
km
0.0
0.8
1.2
1.6
3.2
4.0
4.8
Concentration, ppmb
May 7
2.069
0.045
7.814
0.001
0.002
ND
0.002
0.002
0.002
0.001
May 8
3.218
0.666
0.003
0.336
NDC
0.003
0.002
0.003
May 9
0.095
0.078
ND
0.168
0.001
0.023
0.168
0.181
ND
ND
Maximum
7.8
0.34
0.17
0.002
0.18
ND
0.002
a Distance from a chosen center of plant emissions.
b1 ppm = 2560/Jg/m3.
cConcentration below detectable level.
Environmental Appraisal
27
-------
CO
IT
O
00
CO
fc E
_Q Q.
| a
i A
OOOO
'-'-'-CslOCNIOOOO
E
,3. | co op LO CN LO q co
§ >< OOOO'-'-COCM*-CMO'-O»-O
CC
X
u
CO
Q
> ooooooooooooooo
o
LU
_l
_l
8
^_ oo *t CD LO csio^r co ^- rt LD co
00 dddd'-'ddd'-OOOO'-O
Q. ' "~
o q^-qqqqoqqqqqpqq
< jj c, ooooooooooooooo
co £ t
^ °J P'~00'-ICN'-qt-qqeSiqqqq
c ^ ddddddciciddddddd
•;= z .2
trf *•• ~, ^•^•coojoo O5«—co oo o^-ro
o_i £ °° —-. — --—.------ ___ _
r-O. ^ ^ OOOOO'-OOO'-O'-OOO
Q V a>
frj J> °
LU — j-
bz ^ ^ ooooooooooooooo
CC— S ,~ COtCNCMOinCDOOtDCDCN _. .
— "CP? PPP°. ^ PPCJPP0'!'"'"^'^
ujUJ c1^ ddddddddddd^-'ddd
cc |
LO LnooooLnoto coujLnr^nLntoro o
^^ ddddddcodddo'dddd x
. . f —. -~. „ . . «, . u, ^ 00 CO »~ O5 o
(5 ^ "". '~p'~^CNio.p'~~pp'~cv?p
° o'ddddddddcM'o'ddv-'d ^
ff ^ o'ddddddc-iddddo'dd |
.,. ,., OOOJLn^-OOOOCOOOOOO^-CMOO O c
LU CN qcMqqqqcoq<-qqqqo<- cu
O'O'P'P'O'P'O'O'O'P'O'P'O'^'O' c
. ^
O r; Lnqqqco^qcN^qqqqqq
in ddddddddddddddd
DC ° q<-CNCNCMC\iqqo.<-'-CNiq^q c
Q ddddddddddddddd ^
O
CD
5 a
•°. '
cu"
cj
S ^- • • • ^~- *""• • • • • ^
-------
Table 4.10. VINYL CHLORIDE CONCENTRATION IN INTEGRATED 24-hour SAMPLES
COLLECTED BY CHARCOAL ABSORBER NEAR REGION IX PLANT, 1974
Distance,3
km
1.1
1.3
3.1
4.3
4.5
5.3
Concentration, ppmb
May 7
0.08C
0.08
NDd
ND
0.27
0.07
May 8
0.07C
0.08
0.06
0.06
0.65
0.04
May 9
0.1 Oc
0.05
0.05
0.05
0.27
0.05
aDistance from center of plant.
b1 ppm = 2560/lg/m3.
C12-hour samples.
dNot detectable.
4.2.5 Discussion of Regional Monitoring Data
Since EPA Regional Offices exercised a great deal of latitude in implementing the prescribed procedures, it
is difficult to compare data from different plants. All values obtained are expected to be lower than the real
value. Grab samples are expected to lose quantities of vinyl chloride monomer because of continuing
reaction in the sampling container and leaks. Vinyl chloride monomer losses from these containers are
estimated to range from 0 to 10 percent per day,, Moreover, losses of VC when using charcoal absorbers are
expected to be greater than from the containers. Collection efficiency and recoveries have not been
definitively established, but preliminary data indicate recoveries of 71 to 76 percent when extraction by
carbon disulfide is used to recover the VC from charcoal. Samples at seven data points were collected in
parallel and analyzed by two different laboratories. The resulting values disagreed markedly. The relative
standard deviation about the mean ranged from 5 to 140 percent,
4.2.6 Subsequent Vinyl Chloride Monitoring Study
In November 1974, EPA initiated a vinyl chloride monitoring study5 to obtain measurements of vinyl
chloride concentrations and supporting meteorological data at different types of plants emitting vinyl
chloride. The primary objective of the study was to obtain information to refine an atmospheric diffusion
model for vinyl chloride. To achieve the objective of the study, three types of plants were chosen for
monitoring purposes: (1) a PVC plant with the processing equipment enclosed in a building (B.F. Goodrich
Chemical Company, Louisville, Kentucky), (2) a PVC plant with processing equipment not enclosed in a
building (Continental Oil Company, Aberdeen, Mississippi), and (3) a vinyl chloride plant (Shell Chemical
Company, Norco, Louisiana). The distribution of the sampling sites is described in Table 4.11.
The monitoring method consists of a 24-hour sampling procedure which collects vinyl chloride on
charcoal absorbers. The vinyl chloride is subsequently extracted with carbon disulfide and resulting
solutions are measured chromatographically using a flame ionization detector.
In order to determine bias of the analytical performance, charcoal tubes containing VC which were
prepared by the National Bureau of Standards were analyzed with the regular field samples as unknowns.
The average bias was -6 percent below standard with a standard deviation of 6 percent. Analysis of field
duplicates provided an estimate of uncertainty which includes variables in addition to analytical variability.
The average mean difference determined was -1 percent with a standard deviation of 25 percent.
Environmental Appraisal
29
-------
Table 4.11. SITE DISTRIBUTION FOR VINYL CHLORIDE MONITORING STUDY
Location
Louisville,
Kentucky
Narco,
Louisiana13
Aberdeen,
Mississippi
On or off
plant
property
On
On
Off
Off
On
On
Off
Off
On
On
Off
Off
Distance
from point
of reference,3
meters
< 250
250 - 400
< 1000
> 1000
< 250
250 - 500
< 1000
> 1000
< 250
250 - 300
< 1000
> 1000
Total
No. of
sites
2
1
7
7
3
(6)
7
(4)
3
(5)
2
7
3
2
3
No. of sites in quadrant
0-90°
1
6
4
1
(1)
3
1
(5)
2
3
90-180°
1
1
(1)
(4)
1
1
1
1
180-270°
1
1
1
(2)
4
1
2
1
270-360°
2
1
(1)
3
2
1
aThe stack was chosen as the center of reference for the VC plant at Narco. For the PVC plants at Louisville and Aberdeen,
a center of reference was chosen which was estimated to be the approximate center at the emission sources. Since there are
multiple point sources in the plant area, a given observation may be located more closely to a point source than the distance
indicated from point of reference.
"Numbers in parentheses are additional sites operated for approximately 3 weeks.
The study was initiated in November 1974. The monitoring at Aberdeen was discontinued during the last
week in March 1975 and the equipment utilized to expand the network in Norco. The additional stations
and their distribution are also given in Table 4.11. The study in Norco was discontinued in May 1975 and
the stations were moved to Louisville, Kentucky, where monitoring continued until the middle of June.
Data from the intensified study in Louisville have not been processed to date and are not included in this
report.
During the time of the testing program, the economic recession caused the three plants being monitored (as
well as most of the remaining VC and PVC plants) to operate at substantially reduced capacities. This is
particularly true of the two PVC plants, which for a large part of the program operated at half capacity;
therefore, measurements taken during this time may not reflect the values that might have been obtained
under full capacity. Data on operating parameters were collected from the plants and are being compiled.
The results are summarized in Table 4.12. As can be observed, the cumulative frequency distribution is not
symmetric and the geometric mean is a closer estimate of the median, or midpoint, of the distribution than
the arithmetic mean.
30
VINYL/POLYVINYL CHLORIDE
-------
CO
z
g z
< K
EC D
t- CD
O Q
0 >
LLJ O to
Q H~
Geometric
Arithmetic
Percentile
>
S
>
a>
•o
§
c
CO
u
2
X
a:
S
O
05
O
CO
o
r^
o
CO
o
LO
o
t
o
co
0
CN
O
C
S
o£
Z 0
Distance from
point of
reference,13
meters
c t >
.2 0 ~ t
+-1 i_ C 0)
S°f&
— i o a
CO r- CO «-
co LO LO co
LO co ro p
r~' o o iri
CM CO T-
co co rv CM
LO ri- 00 CM
t — * —
O> LO O5 •*
§i~-° co «-'
CO rf «-
CO "3" LO O
O t- LO CN
l-~ CO LO CN
co co
CM co
00 CO •* CO
O «- CO CO
co r» ro o
O CO O O
o o o o
LO •* co m
co •* co a>
CN CN
0
o S o o
10 ^ 0 0
CN O O
S"
CN
V V /\
oT >
Ms S
o
CN
CN CO LO LO
r~ co ro
CO CN
CO LO p rf
CN CN i-^ CO
CO CO CM
CN
LO LO i- CO
^- r^ ro LO
O_ C0_ O)
r~-~ CN"
CN
r- T- LO «sf
"* CO CN
•^ co ro co
00 CM
o o
» 0
«-; CN O •*
00 •* CO i CO
p r-; P~- r~;
CO ^ C3 CO
CD r^ r-
co
CN r^ «- co
CN ro co CM
CO !--_ r-
p ro LO CD
CJ 1^ LO t^
LO LO (O T-
co ro
o o CN r^
co co «a- oo
LO_ •*_ «-__ t-
r--" co «-"
CN
't CO CO O
CN <- LO CM
P, ^ "~
co LO ro O)
IO OO ^~
CO r-_
co r~ o LO
CO CO CM
CN CO
•^ ro r- co
CO CN «-
r- LO
O CN r^ O
r» r~-
CO
O CN • '8 ° °
< S
t/> QJ
ro (A
5 o
JT °
O CU
11
«!
O ro
.c o
u o
VI
CO CD
5 -D
CD >
O CO
£ E
fe S
03 .Q
•O .2 CO
c a ./>
ro Q) "
CO .C T3
= *• Q)
O)
II
O o
c o
o c
- a3>
a> 3
a o
ii «
•5 E
o
LL
M S
'55 £
"E o S "
'ill i
cu c c >
?i&S
o E E c
_, S o a)
ro co
2 "5
CD w CJ
ifs
" E.»
oo
a ^
a c
CO CO
3!
o
-i "5
O ">
ID §
in s
CM v
0)
-D 3
s s
!£
I!
Environmental Appraisal
31
-------
To date, a total of 1903 24-hour VC measurements, not including duplicates, have been obtained. Of these,
21 (1.1 percent) exceeded a VC concentration of 1 ppm (2560 /ag/m3) and 47 (2.5 percent) exceeded a
value of 0.5 ppm (1280 jug/rn3). The arithmetic average for all values at all sites was less than 0.005 ppm
(13 jug/m3). This would not necessarily be a representative area value for the plant vicinity, since it does
not reflect existing meteorological conditions. The maximum values obtained from the sampling points
located outside the plant boundary show that at a distance of less than 1000 m no value exceeded 0.5 ppm
(1280 jug/m3) and for a distance of greater than 1000 m no value exceeded 0.1 ppm (256 /jg/m3). However,
it cannot be stated unequivocally that these concentrations are not being exceeded at locations not sampled
or during other times of the year.
EPA has conducted a monitoring program to determine the vinyl chloride concentrations in the vicinity of
PVC fabrication plants. These data currently are being analyzed and a report will be published by the Office
of Air Quality Planning and Standards.
4.2.7 Ambient Air Measurements of VC in the Niagra Falls Area
EPA conducted a 6-day survey in the Niagara Falls area to determine the concentration of vinyl chloride in
the atmosphere. Samples were collected in residential areas near chemical plants, and in downtown Niagara
Falls. The sampling sites are shown in Figure 4.2. Time integrated samples were collected in Tedlar bags and
analyzed by gas chromatography with flame ionization detection. The results of the survey are shown in
Tables 4.13 through 4.17.
The highest concentration measured was at Site I—the residence of an individual reported to have liver
angiosarcoma, which was, however, later diagnosed as anaplastic carcinoma. Two of the five samples taken
in and around the chemical complex contained measureable concentrations of vinyl chloride. Samples taken
upwind of the chemical complex, in downtown Niagara Falls, and in Buffalo showed no detectable vinyl
chloride.
4.3 DISPERSION MODEL ESTIMATES OF AMBIENT AIR CONCENTRATIONS
Estimates of vinyl chloride concentrations downwind of two processing plants were made using available
emission estimates and representative meteorological conditions. The estimates were made using the
Gaussian dispersion model given in EPA's Workbook of Atmospheric Dispersion Estimates.6 Concentration
estimates were for 1-hour averaging times. Two hypothetical plants, designated A and B, were considered,
with two sets of emission conditions for each plant.
Wind speed was held constant at 2 meters per second for all calculations. Because concentration is inversely
proportional to wind speed, a higher wind speed would decrease the estimates; a lower wind speed would
increase the estimates. Wind speeds lower than 2 meters per second occur fairly often—on the order of 3 to
15 percent of the time—at most locations.
Three different atmospheric stabilities were considered. Neutral stability occurs when cloudy skies prevail
either during the day or at night. Neutral stability also occurs during the transition from unstable daytime
conditions to stable nighttime conditions and vice versa. Slightly stable and moderately stable conditions
both occur at night with clear skies and light winds. These three meteorological conditions can be expected
to occur during quite a number of hours in a given year.
Receptor locations in these model calculations were placed at positions downwind of the approximate
center of the sources and off the plant property.
For Plant A, a computation was made considering the emissions from five point sources of varying heights
and from one area source. These sources are all located within 200 meters of each other. From Table 4.18,
it is seen that the predicted maximum hourly concentrations do not differ greatly for the three stabilities,
and are the highest, 4 ppm (10,240 Mg/m3), for moderately stable conditions. Maximum 24-hour
concentrations would be much lower..
32 VINYL/POLYVINYL CHLORIDE
-------
V)
"co
co
O5
CD
CD
S
V)
CD
0)
O)
•a
.c
o
O
4—
03
c
"o.
E
cc
O
'•M
03
O
O
CN
3
01
Environmental Appraisal
33
-------
Table 4.13. AMBIENT AIR SAMPLES FROM 57th STREET, NIAGARA FALLS
(Sampling Site No. 1}
Samples from
Sampling
Site No. 1
1
2
3
4
5
6
7
8
9
10
11
12
13
Remarks
8:00 p.m. to 8:00 a.m., 6/22, wind north-northwest 10-17 mph.
8:00 p.m. to 8:00 a.m., 6/22, wind north-northwest 10-17 mph
(sample taken indoors in the upstairs hallway).
12:00 midnight to 5:00 a.m., 6/24, wind north-northeast 5-10 mph.
5:00 a.m. to 10:00 a.m., 6/24, wind north-northeast 7-1 2 mph.
12:00 noon to 4:00 p.m., 6/24, wind north-northeast 7-12 mph.
4:00 p.m. to 8:00 p.m., 6/24, wind north-northeast 7-12 mph.
8:00 p.m. to 8:00 a.m., 6/24 - 6/25, wind north 5-10 mph.
9:00 a.m. to 1:00 p.m., 6/25, wind north 5-10 mph.
1:00 p.m. to 6:00 p.m., 6/25, wind north-northeast 5-12 mph.
7:00 p.m. to 1 :00 a.m., 6/25 - 6/26, wind north-northwest 5-10 mpf
1:00 a.m. to 9:00 a.m., 6/26, wind north-northwest 5-10 mph.
10:00 a.m. to 2:00 p.m., 6/26, wind northwest 5-10 mph.
2:00 p.m. to 6:00 p.m., 6/26, wind north-northwest 5-10 mph.
Vinyl chloride
concentration, ppba/b
ND
ND
40
9
ND
9
6.1
ND
ND
i. 6.6
27.5
5.5
6.6
aND = not detectable.
b1 ppb = 2.560 ^ig/rn3.
34
VINYL/POLYVEMYL CHLORIDE
-------
Table 4.14. AMBIENT AIR SAMPLES FROM VICINITY OF GOODYEAR CHEMICAL
(Sampling Sites IMo. 2 through 7)
Sampling
Site No.
2
3
4
5
6
7
Remarks
4:00 p.m. to 4:15 p.m., 6/21, wind north-northwest 5-12 mph.
11:00 a.m. to 1 1:10 a.m., 6/24, wind north-northeast 7-12 mph.
11:15 a.m. to 1 1:25 a.m., 6/24, wind north-northeast 7-12 mph.
12:00 noon to 12:30 p.m., 6/24, wind north-northeast 7-12 mph.
3:05 p.m. to 3:35 p.m., 6/25, wind north-northeast 5-10 mph.
10:00 a.m. to 1 1 :00 a.m., 6/26, wind north 5-10 mph.
Vinyl chloride
concentration, ppba
3
0
0
0
12.7
0
31 ppb = 2.560 jug/m3.
Table 4.15. AMBIENT AIR SAMPLES FROM VICINITY OF CHEMICAL COMPLEX
LOCATED ON BUFFALO ROAD
(Sampling Sites No. 8 through 12)
Sampling
Site No.
8
9
10
11
12
Remarks
11:25 a.m. to 1 1:55 a.m., 6/24, wind north-northeast 7-12 mph;
sample taken while walking 3/4-mile distance on Buffalo Road.
11:15 p.m. to 1 :30 p.m., 6/25, wind north-northeast 5-10 mph.
2:15 p.m. to 2:30 p.m., 6/25, wind north-northeast 5-10 mph.
4:00 p.m. to 4:35 p.m., 6/25, wind north-northeast 5-10 mph.
3:00 p.m. to 4:00 p.m., 6/25, wind northeast 5-10 mph; sample
collected over 1 .2-mile distance along River Road.
Vinyl chloride
concentration, ppba
0
0
0
28.6
2.5
ppb = 2.560;Ug/m3.
Environmental Appraisal
35
-------
Table 4.16. AMBIENT AIR SAMPLES OUTSIDE NIAGARA FALLS INDUSTRIAL AREA
(Sampling Sites No. 13 through 15)
Sampling
Site No.
Remarks
Vinyl chloride
concentration, ppba
13
14
15
10:40 a.m. to 11:00 a.m., 6/25, wind north-northeast 5-12 mph.
11:50 a.m. to 12:10 p.m., 6/25, wind north-northeast 5-12 mph;
sample taken over three block area of Main Street downtown
Niagara Falls area.
3:00 p.m. to 3:15 p.m., 6/21, beauty shop, large room, 30 ft by
30 ft, well-ventilated. Sample included a 1-second burst of
aerosol hair spray into the room. (Three ppm Freon-12 was
observed in sample.)
0
0
1 ppb = 2.560jUg/m3.
Table 4.17. AMBIENT AIR SAMPLES FROM BUFFALO AREA
Sample
No.
Remarks
Vinyl chloride
concentration, ppba
1
12:45 p.m. to 1:00 p.m., 6/25, wind north-northeast 7-12 mph,
in area of Dearborn Street.
1:10 p.m. to 1:30 p.m., 6/25, wind north-northeast 7-12 mph,
in area of Fargo Street.
3:30 p.m. to 4:00 p.m., 6/25, wind north-northeast 7-12 mph,
in area of Elmer Street.
0
'1 ppb = 2.560 jUg/m3.
36
VINYL/POLYVINYL CHLORIDE
-------
Table 4.18. CALCULATED 1-hour AVERAGE CONCENTRATIONS OF VINYL CHLORIDE
MONOMER AT SELECTED DOWNWIND DISTANCES FROM A PLANT
WITH MULTIPLE EMISSION SOURCES, PLANT Aa'b
Distance
plant,
km
0.25
0.4
0.5
0.8
1.0
2.0
3.0
5.0
Concentration, ppmc
Neutral stability
No spill
3.5
2.9
2.6
1.7
1.4
0.6
0.4
0.2
Spill
18.2
29.3
28.2
19.4
14.9
6.0
3.4
1.6
Slightly stable
No spill
3.8
3.4
3.2
2.5
2.1
1.1
0.7
0.3
Spill
3.8
4.4
6.1
10.9
11.8
8.2
5.4
3.0
Moderately stable
No spill
4.0
3.9
3.7
3.2
3.0
1.9
1.3
0.8
Spill
4.0
3.9
4.0
7.1
9.5
11.8
9.3
5.9
aEmission conditions'
Source
type
Point
Point
Point
Point
Point
Aread
Spill6
Emission
rate, g/sec
18.9
0.63
6.3
8.8
0.5
44.1
3783.3
Height of
emission, m
22.9
15.2
7.6
30.5
38.1
6
15.2
"All calculations assume 2.0 m/sec windspeed.
C1 ppm = 2560jUg/m3.
Emissions from a 110- by 1 70-m building through vents and
windows about 6 m above the ground.
eSpill of 2270 kg of vinyl chloride released in 10 minutes at
a height of 15.2 mat 338" K (65"C).
Concentrations resulting from this plant were also estimated for the time at which a reactor is aborted and
over 2270 kg of VC is vented to the atmosphere in about 10 minutes. Venting of this type can occur
approximately 20 times each year in a PVC plant. Other emissions were assumed to remain the same as in
the above calculation. Under these conditions, a maximum hourly concentration of 29 ppm (74,240
/jg/m3) was predicted to occur under neutral stability conditions at 400 meters. Since the other sources
contribute less than 3 ppm (7680 A*g/m3) at this point, the vented release contributes about 26 ppm
(66,560 /Ltg/m3). Since the release occurs over only a 10-minute period, a much higher concentration with
instantaneous peaks 5 to 10 times this concentration might be expected at 400 meters as the pollutant
cloud passes,, Concentrations at least five times higher might occur over a6-to 10-minute averaging time at
400 meters according to these estimates. Beyond 5 km, the impact of a spill would be minimal. For both
the spill and nonspill situations, the populations most affected would be those residing within about 2 km
of the plant.
For Plant B, one point source and two area sources were considered under two different conditions, average
emissions and peak emissions (Table 4.19). Three emission sources are assumed to be located within 300
meters of each other in this calculation. In this case, the maximum concentrations are not increased greatly
by an increase of a factor of three in the emissions from the elevated point and by a 2-minute spill from the
area source. However, concentrations are nearly doubled at great distances downwind. For Plant B, the
maximum concentration, 3.7 ppm (9472 Mg/m3), is almost the same as that from Plant A under normal
conditions, 4 ppm (10,240 /ig/m ).
Environmental Appraisal
37
-------
Table 4.19. CALCULATED 1-hour AVERAGE CONCENTRATIONS OF VINYL CHLORIDE
MONOMER AT SELECTED DOWNWIND DISTANCES FROM A PLANT
WITH MULTIPLE EMISSION SOURCES, PLANT Ba
Distance
from
plant.
km
0.2
0.3
0.4
0.5
0.8
1.0
2.0
3.0
4.0
5.0
10.0
15.0
20.0
Concentration, ppm
Neutral stability
Average15
emissions
2.2
1.8
1.6
1.4
1.0
0.8
0.4
0.3
0.2
0.1
<0.1
<0.1
<0.1
Peakc
emissions
2.8
2.4
2.2
2.1
1.7
1.5
0.8
0.5
0.3
0.3
0.1
<0.1
<0.1
Slightly stable
Average15
emissions
2.8
2.5
2.2
1.9
1.5
1.3
0.7
0.5
0.3
0.3
0.1
<0.1
<0.1
Peakc
emissions
3.3
3.0
2.7
2.4
2.1
2.0
1.3
0.9
0.7
0.5
0.2
0.2
<0.1
Moderately stable
Average13
emissions
3.1
3.1
2.9
2.7
2.2
2.0
1.2
0.9
0.7
0.5
0.2
0.2
0.1
Peakc
emissions
3.6
3.7
3.5
3.2
2.7
2.4
1.9
1.5
1.3
1.0
0.5
0.3
0.2
aAII calculations assume 2.0 m/sec windspeed. 1 ppm = 2560 ;Ug/m .
^Average emissions:
Point source - 24.0 g/sec at 33.5 m.
Area source 1 — 21.4 g/sec from 150- by 150-m building. Assume emission at 6 m.
Area source 2 — 14.4 g/sec from 180- by 15-m building. Assume emissions at 6 m.
cPeak emissions'
Point source — 90.5 g/sec at 33.5 m.
Area source 1 — Same as under average emissions.
Area source 2 — Same as under average emissions plus a 2-mmute spill of 24,000 g (2000 g/sec).
4.4 REPORTED VC MEASUREMENTS IN WATER AND FOOD
Vinyl chloride has been found in municipal water supplies. EPA, in 1974, initiated an extensive water
survey covering 80 public water supplies.7 These 80 supplies provide a reasonably representative sample of
the nation's community drinking water supplies that chlorinate their water and represent a wide variety of
raw water sources, treatment techniques, and geographical locations. One major objective of the survey was
to characterize, as completely as possible using existing analytical techniques, the organic content of ten
finished drinking water supplies which represent five major categories of raw water sources in the United
States today.7 Preliminary data for five of the ten cities are presented in Table 4.20. The remaining analyses
should be complete in December 1975. The present data were obtained in most cases from analysis of a
single "grab" sample taken from each supply. Sampling conducted at other times of the year might yield
different results. The sources of the vinyl chloride found in the Miami and Philadelphia water supplies have
not been identified.
EPA is supporting a test program to determine the migration of VC from PVC water pipe. The available
results indicate that migration does occur, and that it is a linear function of the residual vinyl chloride level
in the pipe itself. Details of this study will be published by the EPA Water Supply Research Laboratory,
Cincinnati, Ohio.
38
VINYL/POLYVINYL CHLORIDE
-------
Table 4.20. NATIONAL ORGANICS RECONNAISSANCE SURVEY
RESULTS FOR SELECTED COMPOUNDS3
Compound
Organochlorine
pesticidesb
Organophosphate
pesticides
Polychlorinated
biphenyls
Herbicides
Haloethers
Vinyl chloride
Raw
Finished0
Carbon chloroform
extract -m
Miami,
Florida
2ng/1
NF
NF
NF
NF
1.2jug/1
5.6jug/1
0.9 mg/1
Seattle,
Washington
1 ng/1
NF
NF
NF
NF
NF
NF
0.1 mg/1
Ottumwa,
Iowa
2 ng/1
NF
NF
NF
NF
NF
NF
0.7 mg/1
Philadelphia,
Pennsylvania
NF
NF
NF
NF
0.4;ug/1d
NF
0.27 yug/1
0.4 mg/1
Cincinnati,
Ohio
1 ng/1
NF
NF
NF
NF
NF
NF
0.7 mg/1
Concentrations: mg/1 = milligram per liter = part per million;;Ug/1 = microgram per liter = part per billion,
ng/1 = nanogram per liter = part per trillion. NF = none found.
bOnly dieldrm found.
cThe reason this value is higher than the raw value is unknown at this time.
^Represents Bis-2(chloroethyl) ether.
There are limited data on the migration of VC from PVC containers into food, alcoholic and nonalcoholic
beverages, and cosmetics, The extent of migration depends upon the residual monomer in the PVC, the
length of time of storage, and industrial processes used. Based upon the available migration data, the oral
daily human intake of vinyl chloride for Europeans is less than 100 jug.8 In a study of an analytical method
to determine VC concentrations in edible fats, Fuchs et al.9 found 0.0021 mg/kg (21 ppb) in fats which
had been stored in containers manufactured from PVC.
Preliminary data have been made available by the Food and Drug Administration (FDA) on vinyl chloride
concentrations in some consumer products. The range of concentrations reported is shown in Table 4.21.
These data were reported to FDA, but they have not been verified by the agency at this time. The samples
were collected prior to January 1, 1975, and therefore may not be representative of products currently
available. These data are not adequate to determine a dietary exposure to VC.10
4.5 VINYL CHLORIDE EMISSIONS FROM SOLID WASTE INCINERATION
Only limited data are available on vinyl chloride emissions from the incineration of plastics. In a study by
the University of Michigan, vinyl chloride was identified as a combustion product from the incineration of
plastics.11 The quantities of combustion products of a representative PVC homopolymer varied as a
Environmental Appraisal
39
-------
function of temperature as shown in Table 4.22. The quantity of vinyl chloride also varied with the type of
plastics and their polymers.
No data could be found on the concentration of vinyl chloride in the ambient air in the vicinity of
municipal incinerators.
Table 4.21. RANGE OF VINYL CHLORIDE CONCENTRATIONS IN
SOME CATEGORIES OF CONSUMER PRODUCTS10
Product
Cosmetics'3
Mouthwashesd
Water pipe (residue)
Biologic products
Vinegar
Oil
PVC films
Sheeting
Meat products
Cap liners (food and beverage jars)
Range3
N.D.C to 4 ppm
N.D. to 7 ppm
<1 to 100 ppm
N.D.
N.D. to 8.4 ppm
<10 to 6.5 ppm
<1 to <4 ppm
<1 ppm
N.D. to 0.4 ppm
N.D.
Sensitivity
0.1 ppm
0.05 ppm
0.05 ppm
0.4 to 0.02 ppm
Unknown
Unknown
Unknown
Unknown
Unknown
—
Preliminary data—not verified by Food and Drug Administration.
bTime in bottle—3 to 48 months.
cN.D.-not detectable.
^All samples purchased prior to Nov. 1974.
Table 4.22. VARIATION OF COMBUSTION PRODUCTS OF POLYMER A WITH TEMPERATURE11
(milligrams per gram of sample)
Compound
Carbon dioxide
Carbon monoxide
Methane
Ethylene
Ethane
Propylene
Propane
Vinyl chloride
1-butene
Butane
Isopentane
1-pentene
Pentane
Cyclopentene
Cyclopentane
Hexane
Methylclopentane
Benzene
Toluene
25-
280° C
—
—
0.04
—
0.06
-
0.04
0.02
—
—
-
—
—
-
-
-
24.
0.12
280-
350° C
9.7
20.
0.20
0.33
0.12
0.11
0.08
0.25
0.04
0.03
-
0.01
0.01
0.02
0.01
0.01
-
6.6
0.18
350-
430° C
181.
46.
1.3
0.39
0.94
0.31
0.44
0.17
0.08
0.20
0.005
0.03
0.08
0.01
0.02
0.05
0.02
0.35
0.55
430-
510°C
244.
151.
1.8
—
0.41
—
0.11
0.02
—
0.02
0.001
-
0.01
-
-
0.01
-
0.16
0.03
510-
580° C
237.
181.
0.31
—
—
—
—
-
—
—
—
—
—
—
-
—
—
—
0.01
40
VINYL/POLYVINYL CHLORIDE
-------
4.6 TRANSFORMATION, TRANSPORT, AND REMOVAL
Results on the atmospheric reactions and rates of disappearance of vinyl chloride from the ambient
atmosphere are not available. Limited laboratory studies on the stability and persistence of VC in air have,
however, been completed. Studies were recently conducted by the EPA Environmental Research
Laboratory, Athens, Georgia,12 to determine the pathways by which vinyl chloride is lost from aquatic
systems,. Bacterial degradation of VC was found to be negligible, and VC did not affect bacterial growth
under test conditions. No sorption to bacteria, algae, or fungi could be detected. Data are not yet available
on sorption to inorganic particulate. Equilibrium approximations suggest that under poor transfer
conditions sorption to inorganic particulate may be significant.
VC vapor concentrations in containers made of various materials appear to be essentially constant over
periods of many days. The peak absorption of VC in the ultraviolet region is far below the solar cutoff
(approximately 290 nm) so that VC would not undergo reaction in sunlight in the absence of other reactive
chemical species. When irradiated with simulated solar radiation in the presence of nitric oxide and nitrogen
dioxide, VC in the part-per-million concentration range reacts to form a variety of products. The reaction
products identified include ozone, nitrogen dioxide, carbon monoxide, formaldehyde, formic acid, formyl
chloride, and hydrogen chloride.13
Although VC should disappear significantly in traveling over longer distances, the conversions anticipated
within a few kilometers downwind of emission sources are expected to be small. No mechanism is presently
known for removal of vinyl chloride from the air at night. Biological sinks, such as microbiological removal
in soil, may be of significance in depletion of vinyl chloride over long time periods; but such sinks would
not be expected to be important in terms of urban scale transport of vinyl chloride. Thus, for a first
approximation, VC in the immediate vicinity of emission sources can be considered a stable pollutant. The
usual meteorological dispersion equations could thus be applied to approximate concentrations in the
vicinity of emission sources. Because of strong noctural inversions during the fall and winter, VC buildup
from emission sources may be of particular concern during such periods; there are, however, no data on
this.
4.7 REFERENCES FOR SECTION 4
1. Carpenter, B.H. Vinyl Chloride —An Assessment of Emissions, Control Techniques and Cost. U.S.
Environmental Protection Agency. Washington, D,C. Publication No. EPA-650/2-74-097. September
1974. 84 p.
2. Personal communications from industry representatives responding to Section 114 letters from
personnel of Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, N.C. May 1974.
3. Bureau of Census data. From Monitoring and Data Analysis Division Computer files. U.S. Environ-
mental Protection Agency, Office of Air Quality Planning and Standards. Research Triangle Park,
N.C. 1970.
4. Federal Air Quality Control Regions. U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards. Research Triangle Park, N.C. Publication No. AP-102. January 1972. p. 9-20.
5. Burman, F.J. and G. Akland. Vinyl Chloride Monitoring Study, Preliminary Results. U.S. Environ-
mental Protection Agency, Research Triangle Park, N.C. June 20, 1975.
6. Turner, D.B. Workbook of Atmospheric Dispersion Estimates. U.S. Environmental Protection Agency.
Research Triangle Park, N.C. Publication No. AP-26, 1970. 84 p.
Environmental Appraisal 41
-------
7. Preliminary Assessment of Suspected Carcinogens in Drinking Water. An Interim Report to Congress.
Office of Toxic Substances, U.S. Environmental Protection Agency. Washington, D.C. June 1975.
8. Von Esch, G.J. and M.J. von Logten. Vinyl Chloride: A Report of European Assessment. Fd- Cosmet.
Toxicol. 13: 121-124, 1975.
9. Fuchs, G., B.M. Gawell, L. Albanus, and S. Slorach. Determination of Vinyl Chloride Monomer in
Edible Fats. National Food Administration,, Stockholm, Sweden.
10. Shapiro, R.E., Food and Drug Administration. Letter to L.A. Plumlee, Environmental Protection
Agency. Washington, D.C. July 11, 1975.
11. Bbettner, E.A., G.L. Ball, and B. Weiss. Combustion Products from the Incineration of Plastics. U.S.
Environmental Protection Agency, Cincinnati, Ohio. EPA Report 670/2-73-049. July 1973.
12. Behavior of Vinyl Chloride in Aquatic Systems. Environmental Research Laboratory. Athens, Georgia.
To be published in fall of 1975.
13. Altshuller, Paul. Chemistry and Physics Laboratory, Environmental Research Center, Research Triangle
Park, North Carolina. Memorandum on Current Knowledge of Vinyl Chloride to G.E. Schwitzer,
Director of Office of Toxic Substances. June 7, 1974.
42 VINYL/POLYVINYL CHLORIDE
-------
5. ENVIRONMENTAL EXPOSURE AND RECEPTOR RISK
5.1 EXPOSURE
Human exposure to vinyl chloride may occur from air inhalation, consumption of food, intake of water
containing VC, and skin contact. Theoretical considerations suggests that the airborne exposure route
represents the greatest source of intake for the population living in the vicinity of emission sources. The
higher exposures to VC occur in occupational situations where it is manufactured and where the monomer,
a gas at room temperature and atmospheric pressure, is converted to PVC. Workers involved in the
polymerization process and in the fabrication of the polymer (PVC) into end products may be exposed not
only to vinyl chloride in the gaseous phase, but may also inhale or ingest PVC dust containing temporarily
entrapped VC. PVC particles containing vinyl chloride then may be deposited in tissues.1 There is not
adequate information on general or occupational population exposure to PVC particles in the air. Table 5.1
gives the relative importance of various VC sources for the general adult population.
In the past, peak exposures to VC in occupational situations may have exceeded several thousand parts per
million at times, for example, during reactor cleaning operations.2 The highest time-weighted average
exposures were probably in the 250- to 500-ppm (640,000- to 1,280,000-^g/m3)range.3-4 The threshold
limit value (TLV) for vinyl chloride was initially set at 500 ppm (1,280,000 jug/m3) in 1961 based upon its
narcotic properties. After reports of liver damage due to exposures below 500 ppm (1,280,000 jig/m3) the
TLV was reduced to a time-weighted average exposure of 200 ppm (512,000 j^g/m3) for a 40-hour work
week, with a 500 ppm (1,280,000 /-(g/m3) ceiling for peak exposures.5-6 When it became evident from
animal studies in January 1974 that vinyl chloride produced angiosarcoma of the liver in animals at
exposures as low as 250 ppm (640,000 jug/m3) and when cases of liver angiosarcoma were reported from
workers in PVC production plants, an emergency ceiling TLV of 50 ppm (128,000 /ug/m3) was established.
Also the U.S. Department of Labor recommended that a permanent standard be set at 1 ppm (2560
jug/m3),,6 That recommendation has recently been finalized and a permanent standard, which calls for a
maximum 8-hour worker exposure of 1 ppm (2560 jug/m3) of VC, with peak 15-minute exposures not to
exceed 5 ppm (12,800 /ig/m3), has been promulgated (May 1974). Since establishment of the 50 ppm
(128,000 fjLg/m3) emergency standard, additional studies have shown vinyl chloride to be a carcinogen in
experimental animals at 50 ppm (128,000 jug/m3) exposure levels.6-7
Table 5.1. RELATIVE IMPORTANCE OF VINYL CHLORIDE SOURCES
FOR THE GENERAL ADULT POPULATION (mg VC)
Food3
0.002
Waterb
0.002
Ajrc
0.03
Total
0.034
aAssume 2 kg ingested daily containing an average of 0.001 ppm VC by weight, resulting in an intake of 2 jUg/day. A World
Health Organization report estimates the levels of vinyl chloride in food may be on the order of 1 £tg per person per day
based upon tentative calculations from a limited range of foods.10
Assume 2 liters ingested daily containing 1 ppb of VC. EPA preliminary studies have shown levels of VC in water to range
from 0.27 to 5 ppm in two of five cities studied.
cAssume inhalation of 20 m3 per day containing 0.5 ppb vinyl chloride by volume, which is about one half the lowest de-
tectable limit.
43
-------
The general population in the past may have been exposed to VC through the use of aerosol products (now
banned); although the extent of exposure is unknown.8
Vinyl chloride concentrations of 2 to 3 ppm (5120 to 7680 Mg/m3) have been found in manufacturing
plant aqueous effluents. Recent studies have shown VC migration from PVC water pipe-the quantity of VC
being a linear function of the residual monomer in the PVC pipe.9 Food, either beverages or solids
packaged in PVC containers, may contain vinyl chloride as a result of leaching- The full extent of such a
potential exposure is at present unknown. A World Health Organization report tentatively estimates that
human intake of vinyl chloride, based upon a limited range of food analyses, is on the order of 1 jug per
person per day.10 It has been estimated that daily oral intake of VC in Europe is less than 100 jug.1' VC
has been shown to be carcinogenic in animals by the oral route.
5.2 RISK TO HUMAN HEALTH
Specification of the possible risk to health among the general population associated with inhalation
exposure to vinyl chloride is extremely difficult, in large part due to the lack of information regarding
responses to vinyl chloride at ambient dose levels in both animals and man. Further, the data available
regarding adverse effects attributable to vinyl chloride in man do not include adequate measures of
exposure which may have been responsible for such damage.
In considering the possible risk to human health from vinyl chloride exposure, it is important to keep in
mind that with respect to those occupationally exposed, angiosarcoma of the liver, though an invariably
fatal disease, is not necessarily the only significant health effect associated with VC exposure. Other cancers
may also be involved, and% nonmalignant damage to the liver probably affects a far greater proportion of this
occupationally exposed population than those who develop angiosarcoma.
To date, angiosarcoma of the liver has been considered an extremely rare disease among the general
population. In a survey by the American Cancer Society, only one case of angiosarcoma of the liver was
recorded among 78,000 deaths.12 Shown in Table 5.2 are the results of several studies examining the
proportional mortality of liver angiosarcoma among workers exposed to vinyl chloride.13'16
Table 5.2. PROPORTIONAL MORTALITY OF LIVER ANGIOSARCOMA
AMONG VINYL CHLORIDE WORKERS
Reference
Monson et al.13
Nicholson et al.14
Holder15
Wagoner16
Total
Number
of
deaths
161a
24b
20C .
109d
314
Number
with liver
angiosarcoma
5
3
6
14
Percent
with liver
angiosarcoma
3.1
12.5
5.5
4.5
aTwenty-six deaths occurred among workers at a VCM plant in Calvert City, Ky.; 135 deaths occurred among PVC workers
at a polymerization plant in Louisville, Ky.
bDeaths occurred among workers at a B.F. Goodrich plant in Louisville, Ky.
cDeaths occurred among production employees at one manufacturing location in Michigan who worked for at least 1 year
between 1942 and 1960.
Deaths occurred among vinyl chloride polymerization workers at two PVC plants during the calendar period 1950-1973.
44
VINYL/POLYVINYL CHLORIDE
-------
It is likely that these reported cases of liver angiosarcoma occurred among workers exposed to levels of
vinyl chloride orders of magnitude greater than that which may be found in the ambient air. Compared to
the general population (1 case in 78,000), the relative risk of developing liver angiosarcoma among those
individuals with previous occupational exposure, by combining all these data, is estimated to be
approximately 3000 times greater. Such a relative risk represents a statistically significant difference (p
<<0.01) in the frequency of liver angiosarcoma among those exposed to high levels of vinyl chloride
compared to those in the general population.
In attempting to assess trends in incidence of liver angiosarcoma among the population, it is important to
recognize that other chemicals besides vinyl chloride may be contributing to such a phenomenon. Included
in such a list would be materials with chemical structures similar to vinyl chloride, as well as arsenicals and
thorotrast, both of which have already been associated with angiosarcoma of the liver.17'18
Other compounds similar in structure to vinyl chloride have been found in high concentrations in
households. These compounds, such as trichloroethylene, tetrachloroethylene, trichloroethane, carbon
tetrachloride, ethylene chloride, and various Freons, are found in indoor atmospheres in aerosol form. Their
sources are cleaning compounds, hair sprays, deodorants, personal hygiene products, inhalants, vaporizers,
etc. The toxicity of some of the aerosols are known and the others are suspect. The prolonged exposure to
the concentrations measured in homes may have potential carcinogenic!ty implications.19
Vinyl chloride may pose a greater risk to human health than has been measured thus far. The full impact of
exposure to vinyl chloride among workers may not be realized for many years, since the greatest number of
workers have had onset of exposure only in the last decade and a long latency period is an integral part of
angiosarcoma. Certain segments of the population may be at greater risk from exposure than others, such as
the very young and aged with little or no activity of the alcohol dehydrogenase enzymes and mixed fluid
oxidases which appear to play an important role in accommodating vinyl chloride metabolism. There is also
evidence from both animal and epidemiologic studies that the risk of tumor formation can be enhanced by
the presence of other chemical agents such as alcohol and certain drugs. It is also possible that other
chemicals in the ambient air may enhance the risk of adverse health effects from vinyl chloride.
Thus far angiosarcoma has been used as the major adverse effect for assessing the health risk from vinyl
chloride. There is a broader spectrum of dysfunction which has been demonstrated in animal studies and
deserves much further study in humans. The mutagenic and teratogenic potential of vinyl chloride in animal
studies gives reason to look for comparable effects in humans. The clustering of other types of neoplasms
and cancers in the general population living in close proximity to a vinyl chloride production and
fabricating plant should also be studied.
5.3 REFERENCES FOR SECTION 5
1. Volkheimer, G. Hematogenous Dissemination of Ingested Polyvinyl Chloride Particles. Ann. N.Y. Acad.
Sci. 246: 164-171, January 31, 1975.
2. Rowe, V.K. Experience in Industrial Exposure Control. Ann. N.Y. Acad. Sci. 246:306-310, January 31,
1975.
3. Daniel, R.L., Dow Chemical Company. Testimony presented at Public Hearing—Proposed Standard for
Occupational Exposure to Vinyl Chloride. U.S. Department of Labor, Washington, D.C. June 25, 1974.
4. Dernehl, C.V., Associate Medical Director, Union Carbide Corporation. Testimony presented at Public
Hearing—Proposed Standard for Occupational Exposure to Vinyl Chloride. U.S. Department of Labor,
Washington, D.C. June 25, 1974.
5. Key, M. Introductory Remarks. Ann. N.Y. Acad. Sci. 246: 5, January 31, 1975.
Environmental Exposure and Receptor Risk 45
-------
6. Vinyl Chloride: Proposed Standard. Federal Register. 39 (92): 16896-16900, May 10, 1974.
7. Maltoni, C. and G. Lefemine. Carcinogenicity Bioassays of Vinyl Chloride: Current Results. Ann. N.Y.
Acad. Sci. 246:195-218, January 31, 1975.
8. Gay, B.W., Jr., W. Lonneman, K. Bridbord, and J. Moran. Measurements of Vinyl Chloride from
Aerosol Sprays. Ann. N.Y. Acad. Sci. 246:286-295, January 31, 1975.
9. Preliminary Assessment of Suspected Carcinogens in Drinking Water. An Interim Report to Congress.
Office of Toxic Substances, U.S. Environmental Protection Agency. Washington, D.C. June 1975.
10. Report of a Working Group on Vinyl Chloride. World Health Organization. Lyon, France. IARC
Internal Technical Report No. 74/005. June 24-25, 1974.
11. Van Esch, G.J. and M.J. van Logten. Vinyl Chloride: A Report of European Assessment. Fd. Cosmet.
Toxicol. 13:121-124, 1975.
12. Selikoff, I.J. Testimony Presented at Public Hearing—Proposed Standard for Occupational Exposure to
Vinyl Chloride. U.S. Department of Labor. Washington, D.C. June 25, 1974.
13. Monson, R.R., J.M. Peters, and M.N. Johnson. Proportional Mortality among Vinyl Chloride Workers.
Lancet. 2:397-398, August 17, 1974.
14. Nicholson, W.J., B.C. Hammond, H. Seidman, and I.J. Selikoff. Mortality Experience of a Cohort of
Vinyl Chloride-Polyvinyl Chloride Workers. Ann. N.Y. Acad. Sci. 246:225-230, January 31, 1975.
15. Holder, B., The Dow Chemical Company. Testimony Presented at Public Hearing—Proposed Standard
for Occupational Exposure to Vinyl Chloride. U.S. Department of Labor. Washington, D.C. June 25,
1974.
16. Wagoner, J.J., National Institute of Occupational Safety and Health. Statement Presented before the
Subcommittee on the Environment, Commerce Committee, U.S. Senate. Washington, D.C. August 21,
1974.
17. DeSilvo, J., J.D. Abbott, L. Cayolla da Mutta, and M.L. Roriz. Malignancy and Other Late Effects
Following Administration of Thorotrast. Lancet. 2:201-250, 1965.
18. Regelso, W., V. Kim, J. Ospina, and J.F. Holland. Hemangioendothelial Sarcoma of Liver from Chronic
Arsenic Intoxication by Fowler's Solution. Cancer. 27:514-522, 1968.
19. Bridbord, K., P.E. Brubaker, B. Gay., and J. French. Exposure to Halogenated Hydrocarbons in the
Indoor Environment. Human Studies Laboratory, U.S. Environmental Protection Agency. Research
Triangle Park, North Carolina. Presented at a Conference on Public Health Implication of Components
of Plastic Manufacture. Pinehurst, N.C. July 31, 1974.
46 VINYL/POLYVINYL CHLORIDE
-------
6. UNDESIRABLE EFFECTS
6.1 TOXICOLOGY
6.1.1 Introduction
Although no systematic quantitative assessment of the long-term chronic toxicity of halogenated
hydrocarbons has been made, vinyl chloride, trichloroethylene, and vinylidine chloride had been considered
among the least toxic of the aliphatic chlorohydrocarbons until recent evidence of carcinogenicity.1"10 The
toxic effects associated with vinyl chloride exposure include narcosis from acute exposure and low grade
liver and kidney damage from chronic low exposures. The low exposure effects are similar to those from
exposure to other halogenated aliphatic hydrocarbons. In addition, acroosteolysis has been observed among
workers exposed to vinyl chloride. This disorder is characterized by degeneration of bones in the fingers
and has been associated primarily with direct physical contact with polyvinyl chloride and high levels of
vinyl chloride monomer.1 Acroosteolysis appears to be unique among toxic effects of vinyl chloride when
compared to the toxicity of other aliphatic chlorohydrocarbons. The multitumor response and appearance
of liver angiosarcoma in experimental animals at lower exposure levels is similar to the cancer response
reported in PVC/VC workers.2"7
Carcinogenic activity of VC has been confirmed in several species of experimental animals (rodents) and is
associated with both subacute and chronic low-level exposure when the experimental period is sufficient in
time to permit tumors to appear.2'7 The appearance of hepatic angiosarcoma in experimental animals, and
the discovery of this rare lesion in PVC/VC workers, has served to underscore the predictive value of
experimental animal toxicology.6 Marsteller et al. have recently published a literature review of the
toxicology of VC.''
6.1.2 Acute Effects
The early experimental toxicology of VC was limited to acute exposures, employed a variety of
experimental animals, and was limited to very short exposure periods. The results of these studies have been
reviewed by von Oettingen,12 Mastromatteo et al.,13 and more recently by Marsteller et al.1' In general,
these investigations, which evaluated exposures that ranged from minutes to hours, suggested that vinyl
chloride was of low order acute toxicity, anesthetic in action, and had little capacity to cause injury to the
liver or kidneys. These observations led to considering vinyl chloride for use as a general anesthetic.11
However, the anesthetic effects were often accompanied by cardiac irregularities with some suggestion of
cardiac sensitization. Cardiac rhythm irregularities were observed in electrocardiograms of dogs exposed to
100,000 ppm (256,000 mg/m3) for less than 4 hours. The effective narcotic level in mice exposed to VC
for 1 minute ranged from 86,000 to 123,000 ppm (212,480 to 214,880 mg/m3). Approximately 170,000
ppm (435,200 mg/m3) was required to induce narcosis in dogs and rabbits over the same exposure
period.12
Other side-effects indicated that more generalized systemic disturbances are associated with acute exposure.
For example, guinea pigs exposed to 5000 ppm (12,800 mg/m3) of vinyl chloride for a 30- to 60-minute
period displayed pulmonary edema and hyperemia of the kidneys and liver.12 Although all pathological
parameters appeared normal in Sherman rats exposed to VC at levels up to 100,000 ppm (256,000 mg/m3)
over a 13-day period, advanced lymphocytic hyperplasia of the spleen was observed.14 At lower
concentrations and extended exposure periods (50,000 ppm or 128,000 mg/m3; 19 days), the liver size
increased in the rats.14 Pathological examinations revealed congestion at the cellular level in the liver. The
47
-------
morphological alteration observed appears to be consistent with more recent ultrastructure studies.15
Parasitic cysts were also observed in the liver of these animals, thus confounding conclusions regarding
acute effects and systemic damage associated with vinyl chloride.
These results suggest vinyl chloride can induce physiological effects. The morphological alterations observed
provided important indications that exposure to vinyl chloride may elicit systemic pathological effects
throughout the body. Therefore, effects such as pulmonary edema, trachea! irritation, hyperemia of the
liver and kidneys, cardiac arrhythemia, and hyperplastic changes in the spleen and liver may be side effects
of acute VC exposure. Furthermore, these effects indicate that functional interaction in various organs and
systems of the body may also occur under more chronic low exposure conditions.
More recent acute studies have dealt with modulation of vinyl chloride toxicity by other biologically
foreign synthetic chemical compounds (xenobiotics).15''6 Male Sprague-Dawley rats were pretreated for 7
days by gavage (oral exposure route) with 400 ^moles/kg of body weight of either phenobarbital, a
polychlorinated biphenyl (PCB; Acroclor 1254), or hexachlorobenzene and then subjected to 50,000 ppm
(138,000 mg/m3) of vinyl chloride for 6 hours.15 This single exposure to vinyl chloride produced acute
liver damage in the pretreated animals. Animals that were not pretreated and exposed to equivalent levels of
vinyl chloride displayed no effects. Animals pretreated with 400 jumoles of 3-methylcholanthrene,
spironolactone, or pregnonolone-16-alphacarbonitrile by gavage and exposed to vinyl chloride, 50,000 ppm
(138,000 mg/m3), were also without noticeable liver damage.15
The liver injury observed in pretreated animals subsequently exposed to vinyl chloride appears to be
specifically associated with structural changes in a basic architectural component of the cell, i.e.,
endoplasmic reticulum, the subcellular site of enzymes involved in detoxification. Serum transaminase
enzymes, used clinically to assess liver damage, increased following vinyl chloride exposure in pretreated
animals and were well correlated with induction of specific mixed function oxidase enzymes.15 This
provides indirect evidence that the mixed function oxidase system may be involved in transforming vinyl
chloride into a toxic metabolic analog. Ethanol specifically and significantly enhanced the fetal toxicity of
vinyl chloride at 500 ppm (1280 mg/m3) administered to pregnant mice.16
These studies lend support to hypotheses that address synergistic interactions and indicate that vinyl
chloride liver toxicity is enhanced by some drugs (phenobarbital), alcohol (ethanol), pollutants (PCBs), and
pesticidal agents (hexachlorobenzene). While these are for the majority acute studies, they provide
important evidence of the nature of pathology that may occur under more chronic low level exposure
conditions. A summary of acute effects of vinyl chloride exposure in experimental animals is presented in
Table 6.1.
Table 6.1. SUMMARY OF ACUTE EFFECTS OF VINYL CHLORIDE
EXPOSURE IN EXPERIMENTAL ANIMALS
1. Vinyl chloride is of low order acute toxicity and anesthetic in action.
2. Anesthetic effects of vinyl chloride are often accompanied by cardiac irregularities, pulmonary edema, trachial
irritations, and hyperemia of liver and kidneys. These effects, noted in various species of experimental animals exposed
to high levels of vinyl chloride, may be a function of exposure duration.
3. Advanced lymphocytic hyperplasia of the spleen was observed in Sherman rats exposed to 100,000 ppm (256,000
mg/m3) for 13 days. All other pathological parameters were normal.
4. Pretreatment of experimental animals with drugs (phenobarbitol),alcohol (ethanol), pollutants (PCBs), Acroclor 1254,
and pesticide compounds (hexachlorobenzene) potentiates the hepatotoxic effects of VC. VC fetotoxicity is
potentiated by alcohol.
5. The mixed-function oxidase system appears to be involved in vinyl chloride metabolism and responsible for its
conversion into more toxic metabolites.
48 VINYL/POLYVINYL CHLORIDE
-------
6.1.3 Chronic Effects
While investigations of short duration with high exposures are useful in determining lethality and gross
toxicity, chronic effects associated with more continuous or repetitive exposures to lower levels of VC are
more important in evaluating possible public health hazards. Several investigations designed to assess
chronic effects associated with repetitive exposures to lower concentrations of VC began in the early
1960's.
Torkelson et al.17 reported results of studies using several species of animals (dogs, guinea pigs, rats, and
rabbits) exposed to VC levels ranging from 50 to 500 ppm (128 to 1280 mg/m3). All species exposed to
500 ppm (1280 mg/m3) of VC for 7 hours/day over a 4.5-month period were normal with respect to
outward appearance, growth, and mortality. While several hepatotoxic blood parameters (serum enzymes)
were found to be within normal limits at all levels tested, liver size increased in male rats but not in females
exposed to levels below 100 ppm (256 mg/m3). At 100 and 200 ppm (256 and 512 mg/m3), enlarged liver
sixes were observed in both male and female rats. Central lobular degeneration of the liver and renal tubular
damage in the kidneys was apparent upon microscopic examination with a dose level to 500 ppm (1280
mg/m3). Since considerable liver damage is required to alter serum enzyme levels, l8-20 histopathologic
changes such as those observed in these studies may have signalled an important parameter in evaluating
beginning liver damage not detectable by altered serum enzyme levels. Similar histopathologic effects were
observed in the liver of male rabbits exposed to 200 ppm (512 mg/m3) of VC administered for 7 hours
daily over a 6-month period.17 These effects were not observed in the females.17 These observations
suggest a difference in the response of male and female rabbits to VC, possibly from hormonal differences.
While the liver of male and female rats remained increased in size at exposures of 200 ppm (512 mg/m3) no
microscopic pathology was apparent. There were no apparent kidney disturbances noted among the
various animals exposed to 200 ppm (512 mg/m3) of vinyl chloride. Increased liver size was sustained
even when exposure concentration and duration were reduced to 100 ppm (265 mg/m3) for 2 hours daily
over the 6-month experimental period. However, while liver-to-body weight ratios at lower dose levels were
not statistically different from the ratios in the control group, a trend was suggested. A further reduction in
exposure to 50 ppm (128 mg/m3) for 7 hours a day, 5 days per week over a 6-month period produced no
evidence of liver abnormalities either in size or microscopic appearance.17 Despite the lack of statistical
significance, the investigators conducting these studies placed sufficient weight on their observations of
subtle liver damage in the several animal species studied to recommend adjustments in the established
standards for industrial exposure to vinyl chloride in 1961.'7
Attempts by Viola et al.2'21-22 to develop an animal model for investigating the pathogenesis of
acroosteolysis revealed convincing evidence of generalized systemic toxic effects in a wide variety of organ
systems. These subacute studies provide a more complete description of systemic pathological effects
associated with exposure to high concentrations of VC extended over a 12-month period. While attention
focused upon the liver of these experimental animals (Wistar rats), pathology was noted in the kidneys,
arteries, skin, bones, brain, and nerves. A fibrosclerotic reaction appeared to be a common denominator in
the biologically diverse tissues of the various organs examined. A fibrotic lesion in the liver appears to be
important in evaluating the precancerous state of liver angiosarcoma and portal cirrhosis associated with
vinyl chloride toxicity.23
These observations provide evidence that vinyl chloride permeates the body, interferes with membrane
structure, and elicits a compensatory repair response marked by endothelial proliferation. This is consistent
with morphological observations of vinyl chloride-induced liver damage in animals which were pretreated
with various xenobiotics that induce detoxification enzymes present in such cellular membranes.15 The
cellular ultrastructural alterations observed in the liver of animals exposed to vinyl chloride and predisposed
animals were similar to that associated with carbon tetrachloride. Carbon tetrachloride produces a
peroxidative degradation of structural lipids of membranes and the capacity of this chlorinated compound
to produce cellular injury is linked to its reactivity in free-radical reactions.1 5 Other reports published in
the Russian and European literature provide evidence of alterations in cardiac function, hypertension, and
increased adrenalin and neural activity in the brain of several animal species subjected to low, chronic
exposure levels of vinyl chloride.11 These results suggest the possibility of cardiac disturbances and
Undesirable Effects 49
-------
behavioral changes in humans exposed to vinyl chloride under occupational circumstances.24 Furthermore,
workers or other individuals appear to be at a greater risk of liver damage due to exposure to other chemical
agents that tend to enhance the hepatotoxic activity of vinyl chloride.
A summary of chronic systemic effects of vinyl chloride observed in experimental animals is presented in
Table 6.2.
Although attempts were made to establish a dose-response relationship, few data are available to permit
quantitative analysis of the systemic response to VC exposure. Qualitatively, it is important to recognize
the significance of repetitive exposures and exposure durations that permit observations of time-dependent
response phenomena.2"9'14"17
Table 6.2. SUMMARY OF CHRONIC EFFECTS OF VINYL CHLORIDE EXPOSURE
TO EXPERIMENTAL ANIMALS
1. Vinyl chloride is a hepatotoxin. The liver appears to be the most sensitive critical organ. Response includes increased
size and weight that may be sex-specific (being noticed more in male than female animals).
2. Central lobular degeneration was observed in the liver of male animals on microscopic exam and was not present in
females. Serum enzyme levels used clinically to assess liver damage were normal and not a good parameter to determine
early liver damage. Structural damage precedes serum enzyme changes.
3. The circulatory system is disturbed; irregularities in cardiac function; endothelial fibrosis in arteries, and alteration of
circulating white blood cell levels and platelets have been observed.
4. Hypertension, increased adrenalin activity, and increased neural activity in the brain have also been observed under
chronic exposures.
5. A fibrosclerotic reaction appears to be a common denominator observed in a variety of organs when exposure is of
sufficient concentration and duration.
6.1.4 Carcinogenicity
The first evidence of carcinogenic effects associated with exposure to vinyl chloride was reported by Viola
et al. 2.21,22 jhese investigations involved 51 (25 controls) Wistar (AR/IRE) albino male rats (150-g body
weight) exposed to 30,000 ppm (76,800 mg/m3) of VC for 4 hours a day, 5 days week, for 12 months.
Under these subacute exposure conditions, tumors were observed in skin, lungs, and bones (Table 6.3).
Although Viola's experiments were not specifically designed to investigate carcinogenicity, there was
sufficient evidence of a tumorogenic response to warrant further investigation.2 Particular attention
focused upon the following aspects of the tumorogenicity observed: (1) tumor multiplicity, with neoplastic
lesions observed in several tissues and (2) the presence of Zymbal gland tumors. These sebaceous glands,
located in the ear of rodents, are particularly responsive to other chemical carcinogens; for example,
acetyl-aminofluorine, polynuclear aromatic hydrocarbons such as 9,10-dimethyl-l,2-benzanthracene,
urethane, 4-amino-stilbenes, and benzidine.5
Since the majority of tumors observed were epidermoid carcinomas of the skin, it was concluded that the
cutaneous system was the most susceptible to the tumorogenic effects of VC (Table 6.3 and 6.4). It is
important to note the absence of liver angiosarcoma in these animals subjected to levels of VC for nearly
half of their lifespan. The exposures used in these experiments were similar to those experienced by PVC
reactor cleaners and are similar to the levels which were associated with acroosteolysis.
50 VINYL/POLYVINYL CHLORIDE
-------
Table 6.3. TYPES OF TUMORS OBSERVED IN MALE WISTAR RATS
EXPOSED TO 30,000 ppm (79,500 mg/m3) OF
VINYL CHLORIDE2'3
Wistar
rats
number
1
2,3
4
5
6
7
8
14
16,17
21
22
23
24
25
26
Total
Tumors
Skin
Mucoepidermoid
carcinoma
Epidermoid
carcinoma,
keratinizing type
Epidermoid
carcinoma,
keratinizing type
Papilloma, keratotic
type
Epidermoid
carcinoma
Epidermoid
carcinoma
Mucoepidermoid
carcinoma
Epidermoid
carcinoma
Epidermoid
carcinoma
Epidermoid
carcinoma
Epidermoid
carcinoma
Epidermoid
carcinoma
Epidermoid
carcinoma
Epidermoid
carcinoma
Epidermoid
carcinoma '
17
Lungs
Adenoacanthoma
No tumor
Adenocarcinoma
No tumor
No tumor
No tumor
No tumor
No tumor
Adenocarcinoma
No tumor
Adenocarcinoma
No tumor
Mucus-producing
adenocarcinoma
(alveolar cell
carcinoma?)
No tumor
Squamous cell
carcinoma
6
Bones
Osteochondroma
No tumor
No tumor
Osteochondroma
Osteochondroma
No tumor
Osteochondroma
No tumor
No tumor
No tumor
Osteochondroma
No tumor
No tumor
No tumor
No tumor
5
a Inhalation exposures were conducted 4 hours/day,
animals.
5 days/week over a 12-month period. Scoring only on surviving
Undesirable Effects
51
-------
Table 6.4. TUMOR INCIDENCE IN MALE WISTAR RATS EXPOSED
TO VINYL CHLORIDE3'2
vc
ppm
30,000
Controls
mg/m3
79,500
Total
animals
26
25
Skin
17
-
Tumors
Lung
6
-
Bone
5
-
Total
25
-
a Inhalation of 30,000 ppm (79,500 mg/m3) vinyl chloride for 4 hours/day, 5 days/week, over a 12-month period.
Due to the evidence of carcinogenicity found in the Viola studies and presented in 1970, a more
comprehensive and quantitative series of investigations were initiated in Europe and the United States. These
studies were designed to confirm the carcinogenicity of vinyl chloride with emphasis placed on
development of dose-response relationships in several species of experimental animals.
In Europe, additional investigations were begun in late 1971 by Maltoni and Lefemine.5 A series of 14
experiments was designed to investigate the relationship of VC carcinogenicity to the route of
administration, exposure concentration, exposure length, exposure frequency, species response variation,
and sex- and age-dependent effects. The basic study design placed particular emphasis upon controlled
exposure conditions and a definition of a tumor incidence profile from observations compiled over the
entire lifespan of the experimental animals (Table 6.5).25 Vinyl chloride used in these experiments was
analyzed as 99.9 percent pure but contained some impurities which are listed in Table 6.6.
A multitumor response was observed with tumors appearing in the circulatory system (liver, vascular bed),
excretory system (kidneys), central nervous system (brain), and skin of Sprague-Dawley rats exposed to
variable concentrations of vinyl chloride daily (4 hours) for 5 days/week over a 1-year period.6 Zymbal
gland tumors were observed 26 weeks prior to the end of the 52-week treatment period (Table 6.7). These
sebaceous gland tumors, which were observed in animals exposed to as low as 500 ppm (1325 rng/m3),
migrate to produce tumors at other body sites such as the lungs. Nephroblastomas of the kidneys and liver
angiosarcomas are observed down to 50 ppm (133 mg/m3). The kidney tumors can metastasize to the liver,
lung, spleen, and brain, while liver angiosarcoma metastasized to the lung. The neuroblastoma observed in
the brain of rats (Wistar) exposed to vinyl chloride is very similar in appearance to the modulioblastomas
that occur in humans.6 Renal nephroblastoma and liver angiosarcoma were observed in animals exposed to
10,000 ppm (26,500 mg/m3) at 7 weeks and 12 weeks, respectively, following termination of treatment
(52 weeks). The average latency period (the time from the beginning of exposure to diagnosis) for these
two tumor types increases with decreasing exposure concentration from 59 weeks for nephroblastoma and
64 weeks for liver angiosarcoma at 10,000 ppm (26,500 mg/m3) to 135 weeks for both types at 50 ppm
(133 mg/m3).
Liver angiosarcoma was observed in Sprague-Dawley rats at all levels of vinyl chloride exposure investigated.
The incidence of liver angiosarcoma decreases with decreasing exposure concentrations as the number of
animals exposed remains comparatively constant. It is important to note, however, the appearance of
angiosarcoma and other tumors at observed sites at all exposure concentrations studied. Intra-abdominal
angiosarcoma as well as angiosarcoma of the liver was reported among the animals exposed to 50 ppm (128
mg/m3). The appearance of hepatomas in animals exposed to vinyl chloride levels down to 500 ppm (1280
mg/m3) is particularly noteworthy since these tumors arise from the parenchyma! cells of the liver. Liver
angiosarcoma is associated more with blood vessels and the lining of blood vessels.
52
VINYL/POLYVINYL CHLORIDE
-------
With more than a 60 percent reduction in length of exposure, tumor incidence is reduced and the type of
tumors observed changes, but a multitumor response is still observed (Table 6.8). The central nervous
system (brain neuroblastomas) appears as equally sensitive to inhaled vinyl chloride at exposures to 2500
ppm (6625 mg/m3) administered over an 11-week period as over a 52-week period. The only tumor
observed at 50 ppm (133 mg/m3) was an angiosarcoma in the periorbital region of the eye. The Zymbal
gland tumors and renal nephroblastoma observed are associated more with higher exposure levels than in
the 52-week exposure experiment. There were no tumors of the liver observed at any exposure level of
vinyl chloride where the length of exposure was considerably reduced.
Vinyl chloride elicits a multitumor response and liver angiosarcoma in other species of rodents; Wistar rats
(Table 6.9), Swiss mice (Table 6.10) and Golden hamsters (Table 6.11). Liver angiosarcoma was observed in
Wistar rats and Syrian Golden hamsters at 500 ppm (1280 mg/m3) and in Swiss mice at 50 ppm (128
mg/m3). Preliminary analysis 9 weeks after exposure to vinyl chloride under conditions identical to those
to which Sprague-Dawley rats were subjected suggests Wistar rats and Golden hamsters may be more
resistant to vinyl chloride.
There were no tumors observed in the Wistar rats below 250 ppm (662.5 mg/m3) 9 weeks after a 52-week
exposure. Liver angiosarcoma, nephroblastoma, neuroblastoma of the brain, and Zymbal gland tumors were
observed in these animals above 250 ppm (640 mg/m3). Pulmonary tumors, mammary carcinomas, liver
angiosarcoma, vascular tumors of the circulatory system, and epithelial tumors were observed in Swiss mice
at 50 ppm (128 mg/m3).
Vinyl chloride is apparently able to traverse the placental barrier and elicit a carcinogenic response in the
offspring of pregnant Sprague-Dawley rats exposed to 10,000 ppm (26,500 mg/m3) and 6000 ppm (15,900
mg/m3) during their 12th to 18th day of gestation (Table 6.12). Subcutaneous angiosarcoma was observed
in a 24-week-old male and a 22-week-old female offspring. No other tumors were observed.
The preliminary results from Maltoni's ingestion experiments (No. BT11) have revealed the carcinogenic
activity of vinyl chloride via the gastrointestinal tract.26 This study involved four groups of eighty
13-week-old Sprague-Dawley rats .(40 males, 40 females). Vinyl chloride, dissolved in olive oil at
concentrations of 20 percent, 6.6 percent, and 1.32 percent, was administered 5 times per week to the
animals by gastric catheter. Equivalent dosages would have been 50.0, 16.6, and 3.3 mg vinyl chloride per
kg of body weight, respectively. Control animals received olive oil only. The scheduled treatment duration
was 52 weeks. The results obtained at the end of 50 weeks of treatment are presented in Table 6.13.
Caution appears warranted in interpreting these results since vinyl chloride, a gas at room temperature and
of low solubility, may have escaped from the oil. However, any loss from the indicated administered dose
would point to the ability of vinyl chloride to produce tumors at levels considerably below 50 ppm (128
mg/m3).
Angiosarcoma of the liver was observed in one animal (one of 40 treated males) that had received a total
dose of 863 mg administered over a 52-week period (16.6 mg/kg, 200-g 13-week-old males, 5 days/week, 52
weeks). Angiosarcoma of the thymus gland was also observed in one of the female animals that had received
a dosage three times as high. Although differences exist in animal weight, circadian rhythm, absorption,
organ distribution, excretion, etc., by alternate exposure routes, die dosage delivered into the
gastrointestinal tract by gastric catheter approximates the total dosage received by inhalation that induced
both liver angiosarcoma and renal nephroblastoma, i.e. 800 mg (50 ppm; moderate respiration rate of 0.100
liter/mm; 4 hr/day, 5 days/week, 52 weeks; Expt, BT1).
Although administered dosages are similar, there may be a significant difference in the period for hepatic
angiosarcoma by the two exposure routes. The average latency period reported for the appearance of liver
angiosarcoma in Sprague-Dawley rats that inhaled 50 ppm (128 mg/m3) of vinyl chloride appears to be 135
weeks, i.e. 83 weeks following the end of the 52-week treatment period. Liver angiosarcoma that appeared
as a consequence of gastrointestinal absorption was noted 2 weeks prior to the end of the 52-week
treatment period. Due to the anatomical location, the functional role of the liver in digestion, and the
potential for more efficient absorption of the gastrointestinal tract than the efficiency of the lungs, this
Undesirable Effects 53
-------
Table 6.5. BASIC STUDY DESIGN OF MALTONI AND LEFEMINE2
Exp.
no.
BT1
BT2
BT3
BT4
BT5
BT6
BT7
BT8
BT9
BT10
BT11
BT12
BT13
BT14
BT15
Treatment
Route
Inhalation
Inhalation
Inhalation
Inhalation
Transplacental
Inhalation
Inhalation
Inhalation
Inhalation
Inhalation
Ingestion
Endoperitoneal
Subcutaneous
injection
Inhalation
Inhalation
Doses of VCa
10,000,6,000,2,500,
500, 250, 50 ppm
Untreated controls
Treated controls: VA
2,500 ppm
200,150, 100 ppm
Untreated controls
10,000,6,000,2,500,
500, 250, 50 ppm
Untreated controls
10,000,6,000,2,500,
500, 250, 50 ppm
Untreated controls
10,000, 6,000 ppm
30,000 ppm
10,000,6,000,2,500,
500, 250, 50 ppm
Untreated controls
10,000,6,000,2,500,
500, 250, 50 ppm
Untreated controls
50 ppm
Untreated controls
10,000, 6,000 ppm
Untreated controls
16.6, 33.2, 50 mg/kg
body weight in olive
oil
4.25 mg in 1.0 cm3
olive oil
Controls: 1.0cm3
olive oil
4.25 mg in 1.0 cm3
olive oil
Controls: 1.0cm3
olive oil
10,000, 6,000 ppm
25, 10, 5 ppm
Untreated controls
Length
4 hr daily, 5 days
weekly, 52 wk
4 hr daily, 5 days
weekly, 52 wk
4 hr daily, 5 days
weekly, 1 7 wk
4 hr daily, 5 days
weekly, 30 wk
4 hr daily, 7 days
(12th to 18th day
of pregnancy)
4 hr daily, 5 days
weekly, 52 wk
4 hr daily, 5 days
weekly, 52 wk
4 hr daily, 5 days
weekly, 30 wk
4 hr daily, 5 days
weekly, 52 wk
4 hr daily, 5 days
weekly, 5 wk;
4 hr daily, 1 day
weekly, 25 wk;
1 hr daily, 4 days
weekly, 25 wk
5 times weekly
4,3,2, times by 2
months and
once
1 injection
4 hr daily, 5 days
weekly, 5 wk
4 hr daily, 5 days
weekly, 52 wk
appm X 2560 = ;Ug/m3
54 VINYL/POLYVINYL CHLORIDE
-------
Animals
Species
Rat
Rat
Rat
Mouse
Rat
Rat
Rat
Hamster
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Strain
Sprague-
Dawley
Sprague-
Dawley
Sprague-
Dawley
Swiss
Sprague-
Dawley
Sprague-
Dawley
Wistar
Golden
Sprague-
Dawley
Sprague-
Dawley
Sprague-
Dawley
Sprague-
Dawley
Sprague-
Dawley
Sprague-
Dawley
Sprague-
Dawley
Age,
weeks
13
13
21
11
19
(breeders)
12
(embryos)
17
11
11
11
11
13
13
21
1 day
13
No.
Female
268
280
262
250
110
30
200
420
160
150
80
45
240
Male
309
265
288
260
36
30
220
268
200
420
160
150
70
44
240
Total
577
545
550
510
146
60
220
268
400
840
320
300
150
89
480
Per
group
64-96
120-185
60-190
60-150
30-54
60
3040
32-70
100
300
120
80
60
75
43-46
120
Undesirable Effects
55
-------
Table 6.6. CONTAMINANTS FOUND IN THE 99 PERCENT PURE VINYL CHLORIDE
USED IN THE MALTONI AND LEFEMINE EXPERIMENTS25
Contaminant
Water
Acetic aldehyde
Acetylene
Allene
Butane
1,3-butadiene
Chloroprene
Diacetylene
Vinyl acetylene
Propine
Methyl chloride
Maximal levels,
ppma
100
5
2
5
8
10
10
4
10
3
100
appm X 2560 = jUg/m3.
route may prove to be a more hazardous exposure route in vinyl chloride carcinogenicity. Thus the
importance of food and beverage contamination by VC is again emphasized.
The American investigations7'27 began after the onset of the European studies and were designed to
complement the work of Maltoni and Lefemine.4"6 While the experimental design does differ to some
extent from the design of Maltoni and Lefemine, the early results from the American investigations confirm
the capacity of VC to induce hepatic angiosarcoma in mice at an exposure level of 50 ppm (128 mg/m3)
(Table 6.14),27 Angiosarcoma was also identified in the liver of male rats and hamsters exposed to 2500
ppm (6390 mg/m3) and one female rat exposed to 200 ppm (530 mg/m3).7 Tumors were identified in the
lungs, mammary glands, and skin of mice, which is consistent with the evidence of multitumor
carcinogenicity of VC found by Viola et al.2'22 and Maltoni and Lefemine.4"6 Liver angiosarcomas also
were found in Golden Syrian hamsters exposed to gaseous VC, a finding that is consistent with the
European studies. No such tumors were observed in the control animals. Although the mortality rate has
been high among experimental animals in these studies, it is important to note that angiosarcoma of the
liver has been observed in mice exposed to 50 ppm (128 mg/m3) of VC for only 26 weeks. In the
experiments reported by Maltoni and Lefemine,4"6 liver angiosarcoma was observed at 50 ppm (128
mg/m3) only after 83 weeks following termination of treatment.
The results from the American and European investigations provide convincing evidence that VC is a
chemical carcinogen. It has been shown to induce multiple tumors in various organs and systems in three
species (Sprague-Dawley rats, Wistar rats, Syrian Golden hamsters, and mice) of experimental animals at
exposure concentrations down to 50 ppm (128 mg/m3). Angiosarcoma of the liver has been observed in
rats and mice at 50-ppm (128-mg/m3) exposure concentrations in two research laboratories. Wistar rats
display a multiple carcinogenic response following inhalation of VC, with tumors appearing in several
different organs (bone, kidneys, and skin). However, this strain of rats appears to be more resistant in the
development of liver angiosarcoma below 500 ppm (1280 mg/m3).2'5 This may be due to limitations
imposed by differences in the number of animals exposed. Furthermore, the carcinogenic response in the
liver of experimental animals is not restricted to angiosarcoma, since hepatomas also have been observed in
experimental animals.6'28 Only the studies of Maltoni et al.s>26 are sufficiently advanced to provide some
information in regard to dose-response relationships of VC. The information available from several studies
indicates that VC carcinogenicity is dose-dependent. Total tumor incidence increases with increasing
concentrations of VC and also is a function of exposure duration (Tables 6.7 and 6.8). In fact, the results of
56 VINYL/POLYVINYL CHLORIDE
-------
the Maltoni et al. studies4'26 suggest that liver angiosarcoma may be more dependent upon duration of
exposure than upon the concentration of VC administered since it was not observed under reduced
exposure conditions. A comparison of the various tumors observed in experimental animals and in man is
presented in Table 6.15. This table fails to note the epidemiological evidence of brain tumors observed
among VC/PVC workers.24
Conclusions drawn by Maltoni and Lefemine on vinyl chloride carcinogenicity are presented in Table 6.16.
An additional important toxicological consideration of vinyl chloride, its metabolites, and its structural
analogs (trichloroethylene; vinylidine chloride) is the evidence of transplacental carcinogenicity. The results
of Maltoni and Lefemine demonstrate transplacental carcinogenicity of vinyl chloride (6000 to 10,000
ppm, or 15,400 to 25,600 mg/m3).25 This indicates that vinyl chloride or metabolic conversion products
traverse the placental barrier. Trichloroethylene, a structural analog and a positive carcinogen in
experimental animals, has been shown to traverse the placental barrier in humans.29 Although the study
involved administering trichloroethylene mixed with nitrous oxide as an anesthetic without specifying
dosage, levels of trichloroethylene were determined in maternal and fetal blood at the time of delivery. The
results clearly demonstrated transplacental transfer of trichloroethylene. The author also noted the effects
that may be attributed to the mode of administration. Intermittent exposure leads to the metabolic
conversion of trichloroethylene to trichloroacetic acid and trichloroethanol in the mother while
unmetabolized trichloroethylene accumulates in fetal circulation. One explanation for this accumulation is
that the fetus is very poor in enzymes capable of metabolizing halogenated compounds such as
trichloroethylene and vinyl chloride. Since this study preceded the recent evidence of trichloroethylene and
vinyl chloride carcinogenicity, no attention has apparently been given to followup studies as to tumors in
the mothers or offspring. There is evidence that vinyl chloride is a transplacental carcinogen and can
apparently reduce litter size and increase resorptions of fertilized ova in experimental animals.16 This
suggests that such effects may have heritable genetic consequences.
In vitro mutagenic bioassays have shown that vinyl chloride, vinylidine chloride, and presumed metabolites
of vinyl chloride—chloroethylene oxide, chloroacetaldehyde, and chloroethanol—are mutagenic.30"32 Vinyl
chloride exposures used in the microbial test systems were 0.2, 2.0, and 20 percent or 2000,20,000, and
200,000 ppm (512, 53,000, and 530,000 mg/m3). While these levels are more relevant to past exposures
experienced in the workplace and operating rooms, qualitatively, the results suggest potential hazards at
lower levels. These data, considered in light of the teratogenic evidence in whole animals at 500 ppm (128
mg/m3) of vinyl chloride and the enhanced effects observed with ethanol,16 indicate that these compounds
warrant serious concern for health hazards associated with exposure to vinyl chloride (and related
compounds), particularly for members of the general population predisposed by pregnancy. Preliminary
evidence of such a hazard has been detailed in the work of Infante, who has shown an increased incidence of
birth defects (spina bifida, cleft palates, etc.) in a community near a polyvinyl chloride production plant.33
Caution is warranted in drawing causal conclusions with respect to these birth defects and location of the
PVC facility since there are no monitoring data to document exposure to vinyl chloride, but this report
demonstrates the need for further investigation.
Undesirable Effects 57
-------
Table 6.7. CARCINOGENIC EFFECTS OF INHALED VINYL CHLORIDE
83 WEEKS FOLLOWING EXPOSURE (Exp. BT1)6
Treatment3
10,000 ppm
6,000 ppm
2,500 ppm
500 ppm
250 ppm
50 ppm
No treatment
Total
Animals with tumors
Animals
(Sprague-Dawley
rats)
Total
69
72
74
67
67
64
68
577
Cor-
rected
numberb
61
60
59
59
59
59
58
464
Zymbal gland
carcinomas0
No.
16
7
2
4
-
-
—
29
%d
26
12
3
7
-
-
—
-
Average
latency
time.
weeks
50
62
33
79
-
-
—
- .
Nephroblastomas6
No.
5
4
6
4
6
1
—
26
%d
8
7
10
7
10
2
—
-
Average
latency
time.
weeks
59
65
74
83
80
135
—
-
aThe animals were treated by inhalation for 4 hours daily, 5 days weekly, for 52 weeks, ppm x 2560 = |Ug/m3 .
Animals alive after 26 weeks, when the first tumor (a Zymbal gland carcinoma) was observed. The percentages are re-
ferred to the corrected number.
cMetastases to lung.
Percentage of corrected number.
eMetastases to liver, lung, spleen, and brain.
Metastases to lung.
9Several cases of breast fibroadenomas and adrenal and pituitary tumors (generally adenomas) have not been considered,
since their distribution in the different groups does not vary.
Several animals with two or more tumors.
! One angiosarcoma of the lips; one angiosarcoma of the nose; one intra-abdominal angiosarcoma (next to liver).
1 One angiosarcoma in subcutaneous fibrosmg angioma; one ossifying parauricular angiosarcoma; one intra-abdominal
angiosarcoma (next to liver).
Two intra-abdominal angiosarcomas (one next to spleen and one next to ovary); one ossifying angiosarcoma of the neck.
58
VINYL/POLYVINYL CHLORIDE
-------
Animals with tumors
Angiosarcomas
No.
9
13
13
7
4
1
-
47
Liverf
%d
15
22
22
12
7
2
-
—
Average
latency
time,
weeks
64
70
78
81
79
135
-
—
Other
sites.
no.
3'
y
3k
21
2m
1n
—
14
Sub-
cutane-
ous
angio- -
mas.
no.
4
3
3
1
—
1
-
12
Skin
carci-
nomas,
no.
3
1
1
1
4
1
—
11
Hepa-
tomas.
no.
1
1
2
3
—
—
—
7
Brain
neuro-
blasto-
mas.
no.
7
3
5
—
—
—
-
15
Other
type
and or
site,9
no.
7°
8P
49
4r
3s
9l
10U
45
Total,h
no.
38
31
32
22
16
10
6
155
One pulmonary angiosarcoma; one angiosarcoma of the uterus.
I
mOne intra-abdominal angiosarcoma (next to spleen); one intrathoracic ossifying angiosarcoma.
nOne intra-abdominal diffused angiosarcoma.
°Two Zymbal gland adenomas; three mammary carcinomas; one neurilemmoma; one ovarian cystoadenocarcmoma.
Ppour Zymbal gland adenomas; one salivary gland adenocarcinoma; two hepatic and one peritoneal angiomas.
qOne Zymbal gland adenoma; one mammary carcinoma; two ependymomas.
rOne mammary carcinoma; two lymphomas; one pulmonary fibrosarcoma.
sOne Zymbal gland adenoma; one mammary carcinoma; one lymphoma.
1 Three Zymbal gland adenomas; two mammary carcinomas; one subcutaneous angiopercitoma; three uterine adenocarci-
nomas (one with sarcomatous component).
uOne invasive acanthoma of Zymbal gland; one subcutaneous fibrosarcoma; two peritoneal fibroangiomas; two uterine
adenocarcinomas (one with sarcomatous component); one uterine leiomyosarcoma; one ovarian fibrosarcoma; one pul-
monary rhabdomyosarcoma; one lymphoma.
Undesirable Effects
59
-------
0)
I
s
o
> UJ
Q °-
UJ UJ
_j cc.
< D
1°
O <
(D §
il
O
CO
CO
0)
3
1=
vt
O
E
i
"co
p
Number of anir
Number of
•g_
CD O
II
CU O
c
2
CO
Angiosarcomas
CO
"i
CU
g
J2 CO
co Q
E CU
'c ^
I
neuro-
01
6
Nephrobla-
-o
O)
V*
CO
blastemas
tst
CU
CU
~J
stomas
carcinomas
Survivors
CO
o
1
CU
CO
CU
O r» co co — — — .
^- CN CN T- en CO CN
in o un CN co «-
- CN
i^ en r- r-
• LT5 CO CO
O) -C i — — , i -*
CO l"» 1 CM CN 1 CO
CO CN CN 1 1
1
co po co CM r-
*_~ \ \ It-1
—
en T- en co CN
1 1 1
1
1
UO •* ^- CN ^t
1 1 CN 1 .-
co -~ -~ -
«- r^ CN r
CO •* «- ~
J
1
1
1
en CD co r- in co co
T- CO CO «- t- CN
o o o o o o o
co co CD CD co CD en
E E E E E E £
Q. Q. a a a a c
a a a. a a a B
o o o o o o S
o o o o LO in 2:
o o in m CN +-1
o" co" CN o
£
00
CO
5
-
CN
T—
o
CN
CN
CO
1
T—
PO
CN
CO
IT)
CN
0
in
IT)
"co
4-i
0
4 hours/day, 5 days/week, for 17 weeks, ppm x 2560 = /j:g/m3 .
c
g
CD
"co
c
.Q
0)
CO
CU
CU
86 weeks are given in parentheses.
drenal and pituitary tumors {generally adenomas) have not been considered, since their distribution in the different groups doi
CD
CD
n experiment BT1
The animals wer
Tumors found ii
CO .
D
CO
CO
breast fibroadeno
Several cases of
u
crt
tumor
with two or more
not vary.
Several animals
"O
us angiosarcoma.
One subcutaneo
CD >.
osarcoma.
O)
c
CD
j5
O
CU
6
4-
il papilloma; one renal adenoma.
IU
CD
C
CU
(—
r fibrosarcoma; or
:in carcinomas; one subcutaneous angioma; one mammary carcinoma; one forestomach papilloma; one cranial osteoma.
JC
O
1
nd fibroangioma;
One parauricula
One Zymbal gla
O)
.c .
Two lymphoma
_ .
o
.0
1
D
.Q
0
CU
nd adenoma; one
CO
O)
1
N
CD
C
O
0
.c
Q.
_»
nd adenoma; two
n
O)
"TO
I
N
0)
c
O
^
60
VINYL/POLYVINYL CHLORIDE
-------
- cc
I- LU
O
< g
o o
1
o
cc
<
o
ai
co
JD
O
E
D
•t-1
.C
+-*
g
J2
CO
'E
CO
"5
CD
.0
1
CO
o
1-
03 OJ
Q. 4-»
> '«
1—
aJ -2
_C T3
6 g
t/)
III
OD
6 «
^_
CD
JO
^} CO
2 E
.c o
0) *"
z
V)
CD
_g "0 |
E ! .E
N CO
H— "^
° « ID
o3 "5 2
•0 E fe
£ c S
z ro 5
'> ^
1_ *—
W >
CD
O
CO
C
CD
E
•»-»
CO
CD
H
oo CD in «-
"TT^^ ' '
r- <- •-
0 T3
CM | | 1 - 1 1
-T i i i i i
TTT i i i i
CO CM ^
CM I 1 «- 1 I 1
CO ZZ. •"
^ 1 T I 1 1 1
iC ^ ^ >-
. . ^_ . .
1 1 1 1 1 1 1
00 CO 00 CO CD CM *fr
«- «- >- CM CO
O O O O O O O
CO CO CO CO CO CO «*
E E E E E E I
a Q. a a a Q. c
a Q. a a. a a £
o o o o o o S
o o o o in in ¥.
C3 o_ in LO CM ^
o" co CM" ^
o
co
"
S
CO
CO
t—
• —
T
CD
CO
in
"^" "
O
T—
1
i —
'
O
CM
CM
"CD
O
II Jd
o c
(O CD
ffi«
II
-I"
|H
LO *:
C >
o ^
II
01 ro
2 E
a> _ c ^
fe ~ -2 •*
> -a ra o
c — ra
1111
C fe M c
ro O fM ro
o E § a,
j: 3 3 c
I- H H O
co .Q o "O
Undesirable Effects
61
-------
Table 6.10. CARCINOGEIMICITY OF INHALED VINYL CHLORIDE IN SWISS MICE 9 weeks
FOLLOWING EXPOSURE (BT4)6
Treatment3
1 0,000 ppm
6,000 ppm
2,500 ppm
500 ppm
250 ppm
50 ppm
No treatment
Total
Animals (Swiss mice)
Total
M
30
30
30
30
30
30
80
260
F
30
30
30
30
30
30
70
250
T
60
60
60
60
60
60
150
510
Corrected number6
M
22
26
23
29
29
27
74
230
F
28
28
30
29
29
30
67
241
T
50
54
53
58
58
57
141
471
Survivors
M
—
—
-
-
3
1
24
28
F
—
—
-
2
1
9
34
46
T
—
-
—
2
4
10
58
74
?The animals were treated by inhalation for 4 hours daily, 5 days weekly, for 30 weeks, ppm x 2560 = JUg/m3 .
Animals alive after 16 weeks, when the first tumor (a mammary carcinoma) was observed. The percentages are referred to
the corrected number.
Adenomas, some of which undergoing malignant transformation.
In females.
Several cases with two or more tumors.
Four liver fibroanglomas; three subcutaneous angiomas; one heart fibroangioma; one ossifying interscapular angioma.
^Four liver angiomas; two liver f ibroangiomas; one subcutaneous angiosarcoma; one renal fibroangioma; one thymic
.angioma.
Three liver angiomas; three subcutaneous angiosarcomas, one subcutaneous angioma; two intra-abdominal angiosarcomas;
. two renal angiosarcomas; one pulmonary angioma.
' One liver angioma; three liver fibroangiomas; two subcutaneous angiosarcomas; one subcutaneous angioma; one subcuta-
neous fibroangioma; one intra-abdominal fibroangioma; three intra-abdominal angiosarcomas; one angioma of the caecum;
one renal angiosarcoma; one pulmonary fibroangioma; one testicular fibroangioma.
J Five liver angiomas; five liver fibroangiomas; two intra-abdominal angiosarcomas; one pulmonary angioma; one scrotal
angioma.
In summary, the experimental evidence indicates that vinyl chloride can act as a mutagen in microbial assay
systems, can act as a transplacental carcinogenic agent, and can elicit teratogenic effects, which are
potentiated by ethanol, at 500 ppm (1280 mg/m3). Since most carcinogenic agents are mutagenic in the
bioassay systems used, vinyl chloride may elicit heritable genetic effects in subsequent generations. This
fact is consistent with the concerns of the Environmental Mutagen Society.34
The earlier reports from Maltoni and Lefemine5 noted the use of vinyl acetate as a control in their studies
of vinyl chloride. Exposure of Sprague-Dawley rats to vinyl acetate up to 2500 ppm (6400 mg/m3) was not
followed by the appearance of any tumors. A report from a group of European toxicologists discussed the
relationship between the chemical structure of vinyl chloride and its toxicological effects.3 s They noted
that a large number of compounds that contain double bonds may lead to free radical formation and have
been shown to be carcinogenic in experimental animals. Therefore, vinyl acetate may be metabolized by a
different mechanism than that of vinyl chloride.
6.1.5 Metabolism and Pharmacodynamics
Early studies indicate that vinyl chloride was not metabolized, but was absorbed, distributed throughout
the body, and eliminated essentially unchanged by the pulmonary and urinary excretory routes.12'13
These conclusions tended to support those of the earlier toxicological studies, which indicated a low order
of acute toxicity for vinyl chloride. With the more recent evidence of carcinogenicity, however, more
extensive investigations of the metabolism and pharmacodynamics of vinyl chloride have been
initiated.35-36
62
VINYL/POLYVINYL CHLORIDE
-------
Animals with tumors
Pulmonary tumors0
No.
35
38
30
38
31
1
4
177
%
70
70
57
66
53
2
3
-
Average
latency
time.
weeks
36
38
43
41
42
56
53
-
Mammary carcinomasd
No.
13
8
9
7
11
10
_
58
%
47
28
30
24
32
33
_
-
Average
latency
time.
weeks
31
33
35
37
39
37
_
-
Liver
angio-
sarcomas.
no.
8
5
11
11
11
1
—
47
Vascular
tumors of
other type
and/or site,
no.
9f
99
12h
16'
14*
11k
—
71
Epithelial
tumors of
the skin,
no.
3'
6m
3n
1°
_
_
—
13
Other
type
and/or
site.
no.
2P
4q
2f
1s
3»
1U
1V
14
Total,
no.
38
39
31
42
38
16
5
207
Two liver angiomas; two liver fibroangiomas; one subcutaneous angiosarcoma; two subcutaneous angiomas; one subcut-
aneous fibroangioma; one intra-abdominal angioma; one intrathoracic fibroangioma, one angioma of the mterscapular
fat pad.
Two squamous carcinomas; one invasive acanthoma.
mFive squamous carcinomas; one acanthoma.
nOne squamous carcinoma; two acanthomas.
°One acanthoma.
^One Zymbal gland adenoma; one forestomach papilloma.
''One lymphoma; one subcutaneous leiomyosarcoma; one forestomach papilloma; one Harderian gland adenoma.
r One forestomach papilloma; one parotid gland mixed tumor.
5 One Zymbal gland adenoma.
tOne Zymbal gland adenoma; one lymphoma; one Leydig cell tumor.
"One parotid gland adenocarcinoma.
vOne lymphoma.
In 1955 von Oettingen1 2 determined VC levels in the blood of cats subjected to acute exposure conditions.
Exposure to 100,000 ppm (25,600 mg/m3) for less than 4 hours produced vinyl chloride concentrations of
15 to 17 mg/100 mg blood in the animals. Respiratory arrest occurred at blood levels of VC of 27 to 30
mg/100 mg blood and cardiac arrest at levels exceeding 40 mg/100 mg blood. Approximately 82 percent of
the inhaled VC was eliminated immediately from the lungs in these experiments.
The observation by Viola et al.2'21 in 1969 using Wistar rats exposed to 10,000 ppm (2560 mg/m3) for
60 minutes, tended to support conclusions which identify the lungs as the principal excretory route of vinyl
chloride. The concentration of vinyl chloride decreased rapidly in expired air, blood, urine, brain, liver and
kidneys during the first hour following exposure.2 There was essentially no detectable level of vinyl
chloride in these animals at 3 hours following exposure. Analysis of the distribution of vinyl chloride
among the formed elements and the fluid media of the blood indicate red blood cells have a greater affinity
for vinyl chloride than serum.
Knittle et al.37 conducted measurements of VC in the subcutaneous fat depots of three control subjects
and 11 workers exposed to the gas during the manufacture of PVC. Significant levels were found in workers
exposed for periods of 5 years or more but none was found in the control group nor in two workers whose
exposure history was less than 1 year. The results indicate that measurement of VC in fat stores could
provide a basis for a practical screening for workers exposed to VC gas.
The metabolism and pharmacodynamics of vinyl chloride under more controlled conditions have recently
been studied by Hefner et al.35>36 A summary of the results from these studies, which involved male
Sprague-Dawley rats (Spartan strain) exposed to initial concentrations of 51 to 1167 ppm (130.6 to 2997.5
mg/m3) with exposure ranging from 52.5 to 356.3 minutes, is presented in Table 6.17.
Undesirable Effects
63
-------
to
01
V)
2
X
01
CJ
Z
o
_l
o
0.
V)
1
00
C/5
GC
LU
e/3
X
Z
LU
Q
O
a
LU
a
E
o
X
o
£
z
Q
1 LJ
U-l
X
Z
11
o
o
z
LU
O
O
O
cc
o
T—
r™
CO
0)
z
|2
umors
•H*
^
.t;
g
vt
CD
E
'c
CO
H-
0
CD
.0
E
^2.
CD
O
1-
CD CD
Q. ^
>- 'to
5 "^
O 5
J.
CD ^ ^^
o ° § S8
•M — t> C
|a™ E
O Q. c
LL CD
V)
CO
E
0
a.
E
-1
to
c
O
C
CD
CD
, -O
• — t/)
D. CD
C 0) £
!* _g o
w .y ^
CO
CD 0 E
> '5) o
^] c Jf
ro co
to
^
QJ °
H- +-• .>
1 1 5l
t ^ c
2 — . fa
c§ o
— • r—
4-J
C
(U
E
4-J
CD
CD
K
CO LT> ^ ^- CM LO CN
i- i - i i
) n f ^ \ \ 1
1 - 1 «- - «- 1
1 «- 1 1 1 - 1
CO CM T- *- i- CO CM
1 1 1 «- 1 1 1
03 00 *— CO 00 *~ LO
«- T- t- CO
LO CM CO CO CM CO O
CO CO CO 00 CO 00 (^
E E E E E E £
a a a Q. a a c
a. a. a. a. a. a. £
o o o o o o S
O O O O LO LO i-
O O LO IO CM -^
o CD CN" o
T— *i.
LO
CN
CM
CO
^•
CM
CO
i —
IT)
00
CO
CN
"co
£
O
8
CN
X
a.
a
o
CO
£.1
> s
CD *••
TJ *-
" S
^.1
ro "ro
«£
=j ?
r- O
cn
t fc
-Q '« h
-a o o
is c §
i S E g g
IsiSs
iwOO
O "D
-------
LL
o
>-
1-
o
z
111
C?
o
CARCIT
_i
Z CD
UJ —
0 £
rf 1—
j£
a.
M UJ
Z 0
$°=
£3
•^ .J
UJ I
I U
1- _J
z >
o z
^ ^
•^ >
w O
z^
en <
UJ X
a. •%_
X -
UJ
>
cc
<
z
5
€£
I]
HI
cc
a.
CM
CO
_c
^1
1=
ith tumors
limals wi
C
CO
4—
o
l_
03
_Q
E
D
-^
^_
CO
•M
£
•coma
5
0
'o>
C
<
C/)
15 >
•M
C £
"9
O QJ
! i- =3
1 Q CD
1 -Q P
E Q.
= Hi
2
CD
O.
£
i—
CD
^
•M
O
i_
CD
_c
+j
o
0)
•4-1
't/>
i_
O
"D
C
CD
C/9
CD
•M
'(/)
CD
>
_l
s
.c
a
CD
2
T3
C
CO
O)
"to
^2
N
(/)
•H-i
1—
(/)
CD
E
o
tt
CO
J2
CO
CO
O
C
'o
k_
CO
u
CO
urvivor
CO
lo
s
H
n
4-»
C
CD
E
•M
CD
a>
K
1 1 CM r-
1 1 1 1
1 I^-S-
1 1 1 1
1 1 1 1
1 1 1 1 i
1 1 <- 1
oo CM i- ir>
CO «-
O O •* CM
CO CO ID CO
z z??
CD CD •£ '^
"D "O Q. n
CD CU tS' Li
0) CD 4^ .{2
.a ^ 'o 'o
E E E E
a a a a
a a a a
o o o o
o o o o
C3 O_ O^
.C
O
.C
IN
Undesirable Effects
65
-------
Table 6.13. STUDY OF ONCOGENIC ACTIVITY OF ORAL VINYL CHLORIDE26
(results at end of 50 weeks)
Group/treatment,
mg VC/kg body
weight3
1 750
11/16.65
III/3.33
Control
Total
Survivors
Total
67
68
66
75
276
Female
34
37
39
39
149
Male
33
31
27
36
127
Angiosarcomasb
1 (thymus)c
1 (liver)d
-
-
2
aA total of 320 Sprague-Dawley rats, 40 males and 40 females in each group, were treated.
No tumors other than angiosarcomas were found.
c Female rat.
dMale rat. The tumor was identified at 49 weeks after onset of treatment. Lung metastasis.
Table 6.14. INTERIM SUMMARY OF TUMORS IN MICE EXPOSED
TO VINYL CHLORIDE FOR 8 months"
Exposure
group
Control3
50 ppmb-c
200 ppmb-c
2500 ppmb-c
No. of mortalities
with neoplasms
Total
(9)
4
15
30
Male
(7)
1
3
6
Female
(2)
3
12
24
Type and location of tumor
Alveologenic
adenomas.
lung
0
2
12
28
Angio-
sarcoma,
liver
0
2
11
28
Adeno-
squamous
carcinoma.
mammary
gland
0
2
3
6
Metastasis
of
mammary
tumor
to lung
0
2
1
1
aSeven male and two female control mice have been examined histologically. None had tumors in spleen, liver, kidneys,
heart, or lung.
Only mice with grossly visible tumors have been examined histologically.
cppm X2560= ffl/m3.
66
VINYL/POLYVINYL CHLORIDE
-------
o
CO
LU
DC
u) 0
Q 111
LU O
H O
< CC
* 2
O Lu
CO
DC
O
in
to
.Q
10
Tumors
'stomach
omas and
ithomas
£ = to
o a o
a. a «
> C/>
a 1
if
ammary
rcinomas
5 S
= •5 -a
§?§
Sis
CT 3 Q.
C/J +J (U
Sebaceous
carcinomas
Nephro-
blastomas
to i/j «
EtO Qj
c •*-•
8 .1 'i8
S e ^
O to -t-*
'S, -i-i O
C c fr
< to 0
Lymphomas
and
leukemias
£
O o>
r°j
£ c
O to
I5
H-S
(/)
6 | i
1 o -I
c CJ
rf J: s-
^ CO O
l/J
in
01
8
Q.
CO
+
+
+" +
+ + +
+ +
+
+ + +
+ T
+ T
+ T
+ + + +
k.
CD
S w
4- 3 E C
co O co to
DC S I S
Undesirable Effects
67
-------
Table 6.16. SUMMARY AND CONCLUSIONS OF VARIOUS ASPECTS OF VINYL
CHLORIDE CARCINOGENICITY DRAWN BY MALTONI AND LEFEMINE5-6,25
1. Under our experimental conditions, VC produced tumors in the three animals species studied: rats, mice,
and hamsters.
2. The range of induced tumors varies to some extent from species to species. When given by inhalation,
VC produced: in rats, Zymbal gland carcinomas, nephroblastomas, angiosarcomas and angiomas of the
liver and other sites, skin carcinomas, hepatomas and brain neuroblastomas; in mice, lung adenomas,
mammary carcinomas of a peculiar type, angiosarcomas and angiomas of the liver and other sites, skin
epithelial tumors; in hamsters, liver angiosarcomas and, as early evidence seems to suggest, skin
trichoepitheliomas, lymphomas and forestomach papillomas and acanthomas. Liver angiosarcomas have
been observed in all three animal species.
3. In the BT1 and BT4 experiments, VC shows acarcinogenic effect at 50 ppm (128,000/ug/m3).
4. From the BT1 experiment (the only one completed) a dose-response relationship clearly emerges, as far
as angiosarcomas and nephroblastomas are concerned, in the lower dose ranges: i.e. from 500 to 50 ppm
(1,280,000 to 128,000 /ug/m3) for angiosarcomas, and from 250 to 50 ppm (640,000 to 128,000
jug/m3) for nephroblastomas.
5. A comparison of the results available at the present moment in rats exposed for 52 weeks and 17 weeks
(BT1 and BT3 experiments) shows that the neoplastic response, particularly as far as angiosarcomas and
nephroblastomas are concerned, is affected by the length of exposure to VC.
6. The comparison of the results obtained in rats of two different strains, i.e., Sprague-Dawley (BT1) and
Wistar (BT7), at the present moment, seems to suggest that the strain factor considerably affects the
neoplastic response.
7. The onset of two subcutaneous angiosarcomas and of one Zymbal gland carcinoma in the offspring of
breeders exposed during pregnancy for 7 days appears to indicate a transplacental effect of VC.
8. Blood vessel ectasis and endothelial hyperplasia, associated or not with cellular atypia, are often
observed in the liver and in other organs and tissues in treated animals, with or without angiosarcomas.
Therefore, the effect of VC on blood vessels and endothelium should be considered systemic.
Table 6.17. A SUMMARY OF RESULTS OF THE METABOLISM
STUDIES OF HEFNER ETAL.35
1. VC is quite readily metabolized to polar metabolites which are excreted predominantly in the urine of
rats exposed via inhalation to an initial concentration of 50 ppm (128 mg/m3). Smaller amounts of 14c
activity are excreted in the expired air as carbon dioxide and in the feces. Very little is excreted in
expired air as unchanged VC.
2. A significant but small amount of 14c activity is retained in tissue, particularly liver, as long as 75 hours
post exposure.
3. Metabolites excreted in the urine appear to be conjugated with glutathione and/or cysteine through
covalent linkage to the sulfhydryl group. This is consistent with the reduction of the nonprotein free
sulfhydryl levels in the livers of exposed rats. Preliminary in vitro experiments4 have shown that direct
conjugation of vinyl chloride with cysteine or glutathione in aqueous solutions occurs to a small degree
but very slowly.
4. Monochloroacetic acid also appears to be a urinary metabolite of VCM, when rats were exposed to 5000
ppm (12,800 mg/m3) for an extended time.
68 VINYL/POLYVINYL CHLORIDE
-------
These investigations lead the authors to speculate that a metabolic threshold for vinyl chloride may exist.
The implication that there is a carcinogenic threshold, however, cannot be supported. Sprague-Dawley rats
exposed to concentrations below 100 ppm (256 mg/m3) appear to metabolize vinyl chloride fairly readily.
Exposure to levels in excess of 200 ppm (512 mg/m3) reduces metabolism considerably. This also suggests
that the metabolic pathway available at vinyl chloride levels below 100 ppm (256 mg/m3) can be saturated.
The fact that a pathway may be saturated above a given concentration does not mean that below that
saturation concentration alternative metabolic pathways are inoperative. These investigators further
conclude that vinyl chloride appears to be metabolized by alcohol dehydrogenase since it can be inhibited
by pyrazole (1,2-pyrazole), and ethanol. The metabolism of vinyl chloride did not appear to be inhibited by
SKF-525-A, a drug used to block the activity of some microsomal enzymes (mixed function oxidases)
which are important in steroid metabolism and detoxification of biologically foreign chemical compounds.
Metabolism by the mixed functional oxidase system is suggested when the alcohol dehydrogenase pathway
was saturated somewhere above 200 ppm (512 mg/m3). While these are preliminary conclusions, they
suggest that a carcinogenic epoxide metabolite would not be formed by the alcohol dehydrogenase
pathway. Attempts to infer a carcinogenic threshold for man are not justified. Such attempts would not
adequately consider the genetic variation that exists among the heterogeneous human population compared
to the genetic homogenicity of experimental animals. It also assumes that the alcohol dehydrogenase
pathway is open and can accommodate vinyl chloride metabolism at each instant and each exposure. It
would also have to accommodate the host of other foreign compounds, nutrients, etc., that are continually
passing through the pathway. Public health considerations would also involve those very young and aged
individuals who have little or no activity of the alcohol dehydrogenase enzymes and mixed function
oxidases. Experimental evidence of transplacental carcinogenicity tends to support this concern.
Furthermore, these investigators noted that vinyl chloride is excreted largely in the urine as B-hydroxy-
cysteine that is in itself suggestive of metabolic transformation by an epoxide intermediate. The elimination
of foreign compounds by the formation of an epoxide intermediate followed by conjugation with
thiol-containing compounds is common to the metabolism of a number of chemical carcinogens.
P.L. Grover, P. Sims, and their coworkers have tested a number of carcinogens for in vivo and in vitro
formation of epoxides, which have been shown to be alkylating agents of nucleic acids and proteins.35'41
They have presented evidence for formation of epoxides for a number of carcinogens including pyrene,
benzofajpyrene, phenanthrene, benz [a] anthracene and dibenz[a,h] anthracene. Epoxides of polycyclic
hydrocarbons have been shown to be mutagenic to T2 bacteriophage, to bacteria, to Drosophila sp. and to
mammalian cells. They also produce malignant transformations of cells in culture.42
Recent studies conducted by Reynolds et al.15 indicate the potentiation of vinyl chloride toxicity by
agents that induce the activity of the detoxifying mixed-function oxidase enzymes. These studies have
obvious public health considerations since these inducing agents were drugs and pollutants such as
polychlorinated biphenyls and pesticides. Individuals exposed to these agents as well as alcohol appear to
be at greater risk to liver damage through exposure to vinyl chloride than those who are not. Postulated
metabolic mechanisms of carcinogenicity of VC and structurally related compounds have been reviewed by
van Duuren38 and investigated by Reynolds et al.1 5
6.1.6 Toxicity of Polyvinyl Chloride
The free radical content and the level of residual VC in PVC resins and plastic end products could affect
their toxicity and their potential for carcinogenic activity. Volkheimer43 reported that PVC particles up to
70 micrometers in diameter could be transported and deposited throughout the tissues of experimental
animals.
Another potential problem area related to vinyl chloride is the composition and toxicity of products
produced by incineration of PVC. Again, this is a potentially widespread opportunity for general
population exposure to other hazardous materials. Disposal of PVC by incineration is a common practice,
and incomplete combustion can result in releasing entrapped vinyl chloride and plasticizers, in addition to
combustion products (HC1).44
Undesirable Effects 69
-------
Sokal et al.4S reported respiratory symptoms in workers, employed as meat wrappers, exposed to fumes
from polyvinyl chloride film cut with a hot wire. Symptoms included dyspnea, coughing, and wheezing.
Tests of pulmonary function generally showed obstructive defects with decreased vital capacity and
increased residual lung volume responsive to bronchodilator therapy. The cause of the symptoms was not
definitely established; however, the authors suspected some ingredient of the fume from the heated
polyvinyl chloride.
Jaeger and Hites46conducted a study to determine whether di-2-ethyhydroxyl adipate (DEHA),a plasticizer
used in PVC plastic food wrap, would pyrolytically evaporate when the film was heated. The results did
show DEHA to be a pyrolysis product from PVC film when heated at temperatures from 275 to 350°F.
Van Hauten et al.47 found that the concentration of hydrochloric acid and particulate produced by a
typical meat wrapping machine, using polyvinyl chloride film, varied significantly with wire temperature
and operating conditions. Approximately 75 percent of the particulate produced by the machine was found
to be DEHA.
In a study conducted by the University of Michigan, vinyl chloride was found to be a pyrolysis product
from PVC plastics at combustion temperature below about 500 °C.
6.2 THRESHOLD LIMIT VALUES
Although occupational health studies associated with polyvinyl chloride production and/or exposure to
vinyl chloride monomer began in 1930, there were essentially no reports of possible systemic effects or
serious health adversities until 1949. Chronic "epithelial" hepatitis was diagnosed in Russian resin
fabricators engaged in processing PVC resins.11 The possible etiological agents listed included primary
ingredients used for polymerization, compounds released from the resin during processing, and plasticizers.
A threshold limit value (TLV) of 500 ppm (1280 mg/m3) time-weighted average was established in 1959 by
the American Conference of Governmental Industrial Hygienists (ACGIH) as the industrial hygiene
standard for VC in the United States This standard apparently was based solely upon the fire and explosive
hazards that are possible at a minimal level of 2.5 percent (25,000 ppm or 64,000 mg/m3) by weight of VC
in air.
The results of acute toxicity studies provided evidence of pulmonary congestion (edema), with damage
noted in the liver, kidneys, and tracheal epithelium, as well as narcotic effects associated with high exposure
levels of short duration. The evidence of organ damage and/or systemic effects observed under acute
exposure conditions, led to the chronic (low level) inhalation studies of Torkelson et al.1 7 in 1961. Based
on observations in several species of experimental animals that included evidence of liver and kidney
pathology at the TLV (500 ppm or 1,280,000 yug/m3), and the absence of these effects at 50 ppm (128,000
Hg/m3) following 6 months of investigation, these authors recommended a change in the industrial hygiene
standard. They suggested limiting occupational vinyl chloride exposures to less than 100 ppm (256,000
jug/m3) with a time-weighted average not to exceed 50 ppm (128,000 jug/m3) in air of the PVC workroom.
The ACGIH's Committee on the Threshold Limit Values therefore changed the industrial hygiene standard
for VC from a 500-ppm maximum time-weighted average (TWA) value to a 500-ppm (1,280,000-jug/m3)
ceiling level.
Vinyl chloride disease, or acroosteolysis, was first reported in 1966.1' This disease involved a progressive
skeletal deterioration of the fingers accompanied by interference in peripheral nerve response and
diminished blood circulation (Raynaud-like syndrome). While other reports have appeared in subsequent
years, occupational acroosteolysis in PVC reactor cleaners has been well documented by large scale
epidemiological studies conducted between 1969 and 1972. l>n Prior to 1970, available standard
textbooks and review articles have stressed the safety of polymer processing and noted only a minimal risk
of narcosis associated with inhalation of VC.48"5'
70 VINYL/POLYVINYL CHLORIDE
-------
The studies on occupational exposure to VC that were influential in adjusting permissible exposure levels
were those of Baretta et al.52 and Kramer and Mutchler.53 (The study by Kramer and Mutchler is discussed
in more detail in Section 6.3.3.) These investigations involved determining levels of vinyl chloride in the air
of the work area and correlating them with results of a systematic screening of employees using a variety of
clinical parameters. The mean concentration of vinyl chloride in the work environment in these studies was
found to be 160 ppm (409,600 Mg/m3) with a range of 30 to 170 ppm (76,800 to 435,200 jitg/m3).
Vinylidine chloride at a level of 5 ppm (12,800 jug/m3) was noted as a co-contaminant. No differences were
observed in blood pressure, hemoglobin levels, or electrocardiograms; nor was there evidence of
morphological anomalies (acroosteolysis). Still, there was indication of some degree of liver damage among
PVC employees at time-weighted average (TWA) exposures of 300 ppm (768,000 jug/m3). These
observations led the investigators to conclude that there was a definite risk of liver damage at vinyl chloride
levels of 300 ppm (768,000 /xg/m3) TWA in the presence of 5 ppm (12,800 /zg/m3) of vinylidine
chloride.s3
An adjustment of the industrial hygiene standard to a TLV of 200 ppm (512,000 Mg/m3) in 1972 was
apparently influenced by evidence of human liver dysfunction, and the availability of monitoring data from
at least one PVC production facility.52"54
Due to the increasing incidence and concern regarding acroosteolysis, Viola, in 1970 and 1971, undertook
experimental studies to develop an animal model to explain this and other adverse effects observed in
humans exposed to high levels of vinyl chloride.21'22 In the course of these acute studies, carcinogenic
effects were observed. The observations were presented in 1970, followed by a detailed publication in
1971. Although the experimental design was deficient with regard to carcinogenic investigations, the
evidence warranted further study. Subsequent investigations were initiated in Europe and in the United
States, using lower exposure levels and purer compounds.5'7 Preliminary results from these efforts
confirmed the carcinogenicity of vinyl chloride in several species of experimental animals; a dose-response
dependency of total tumor incidence was observed; and positive effects were detected at exposure levels
down to 250 ppm (640,000 Mg/m3).4-5
Due to the history of liver dysfunction in PVC employees and in the chronic experimental animal studies
observed earlier, particular concern was aroused by the appearance of angiosarcoma, rare in experimental
animals and man, in the same critical organ, the liver.4'5 Concern with respect to VC followed the findings
that from 1968 to 1973 four employees at a PVC plant had died of either liver angiosarcoma or other liver
cancers of unknown type.8 A fifth individual in the same plant died in late 1973 of cirrhosis of the liver.
Investigation of the exposure history of these individuals revealed that the deceased employees had an
average exposure period of 19 years to VC, and 10 years to vinylidine chloride. These workers had been
engaged in operations where VC concentrations may have greatly exceeded the 1972 TLV of 200 ppm
(512,OOOMg/m3).
The Occupational Safety and Health Agency (OSHA) in January 1974 set an emergency standard for
industrial exposure at 50 ppm (128,000 /ig/m3).55 Results from the American and European chronic
studies soon revealed the induction of liver angiosarcoma and other tumors at a 50 ppm (128,000 jug/m3)
exposure level of vinyl chloride.4'5 '7 Industrial epidemiological investigations identified additional cases of
liver angiosarcoma among American and European PVC workers. An occupational standard of 1.0 ppm
(2560 jug/m3) (or detectable levels) was then proposed by OSHA.56 Subsequently, a permanent 1 ppm
(2560 /zg/m3) TWA occupational standard (8 hours per day; 5 days per week), with a peak 15 minute
excursion not to exceed 5 ppm (12,800 jMg/m3), was promulgated in May 1974.
The incidence of angiosarcoma among employees involved in the manufacture of VC and PVC resins
substantially exceeds the estimated national incidence level, as discussed in Section 5. Interest has been
expressed regarding possible community exposures among those whose residences are situated near VC/PVC
production plants and resin fabricating facilities. There is similar concern with respect to possible
community exposure surrounding fabricating plants using PVC resins containing residual VC. Vinyl chloride
has been detected in the ambient air near vinyl chloride and polyvinyl chloride production sites.
Undesirable Effects 71
-------
6.3 HUMAN EFFECTS
Our knowledge of undesirable effects associated with vinyl chloride exposure in man comes primarily from
occupational situations. These effects include an increased risk of cancer of multiple organ sites including
angiosarcoma of the liver. Angiosarcoma of the liver, observed today in workers exposed to VC, probably
was the result of very high occupational exposures received many years ago. The latent period for
angiosarcoma of the liver has been estimated at 15 to 20 years following onset of exposure.57'58 A long
latency period is an integral part of the natural history of the disease; therefore, the full impact from past
vinyl chloride exposure among workers may not be realized until many years from now since the greatest
number of workers have had onset of exposure in the last decade. For example, of the confirmed cases of
liver angiosarcoma, information as to date of diagnosis or death, where available, indicates that only 2 of 27
cases died prior to 1965 and that 15 of 27 cases had died or were diagnosed in 1970 or later.58 A
"confirmed" case has been microscopically confirmed, whereas a "reported" case is reported merely on the
results of a pathologic examination. Accordingly, estimates of cancer risk from vinyl chloride, based upon
data available today, may well understate the magnitude of this problem. In this regard, increased awareness
and improved diagnostic procedures may in part also contribute to future increases in reported cases of liver
angiosarcoma.
6.3.1 Confirmed Cases of Angiosarcoma
To date, surveys have reported 17 occupational cases of liver angiosarcoma in the U.S.A;58'59 of these
cases 15 have been confirmed as angiosarcoma of the liver by pathologists at the National Cancer
Institute.60 Of these confirmed cases, 14 have been among workers in PVC polymerization plants, and one
of the confirmed cases involved an accountant employed at a vinyl cloth plant. The accountant is presumed
to have had a lower level of exposure than the PVC workers. In addition to these U.S. cases, 21
occupational cases of liver angiosarcoma have been reported from European countries and Canada, 12 of
which have been confirmed microscopically (Table 6.18). Nine of the confirmed cases were among workers
in the PVC polymerization industry and 3 were among nonpolymerization workers. A summary of these
reported occupational liver angiosarcoma cases is shown in Table 6.18.
The period from onset of initial exposure to diagnosis or death is 10 years or greater in all known instances.
Similarly, the years of exposure among these individuals preceding development of clinical disease is, with
two exceptions, in excess of 10 years.5 s
With respect to the general population, cases of liver angiosarcoma were reported among individuals who
had resided in the vicinity of industrial vinyl chloride emission sources. A review of these cases by
pathologists at the National Cancer Institute has confirmed the diagnosis of liver angiosarcoma in two of
the three instances. One case of a woman in Buffalo, New York, that was originally believed to be liver
angiosarcoma61 has now been rediagnosed as anaplastic carcinoma rather than a sarcoma.62 The two
other cases, both from Connecticut, represent confirmed angiosarcoma,62'63'25 but these cases are not
identical in all respects to the pathology which has been observed among PVC polymerization workers
60,25,64 (see Table 6.19). The implications of these dissimilarities between community and occupational
cases are not fully understood.
6.3.2 Reported Studies
As a result of the reported cases of liver angiosarcoma among vinyl chloride workers, a number of studies
have compared the mortality experience of these workers to that of the general population.
6.3.2.1 Tabershaw I Cooper Study—Tabershaw/Cooper Associates conducted a mortality study of workers
in the vinyl chloride industry.64'9 The objectives of this study were threefold: (1) to contrast the mortality
experience of individuals employed in vinyl chloride plants with that of the general population, (2) to
examine mortality patterns among vinyl chloride workers in relation to estimated occupational exposure,
and (3) to compare mortality patterns among vinyl chloride workers with those for other occupational
groups.
72 VINYL/POLYVINYL CHLORIDE
-------
Table 6.18. REPORTED CASES OF LIVER ANGIOSARCOMA AMONG
PVC WORKERS AND NOIM-PVC WORKERS59-3
Country
Canada
Canada
Canada
Canada
Czechoslovakia
Czechoslovakia
France
Great Britain
Great Britain
Italy
Norway
Rumania
Sweden
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
United States
W. Germany
W. Germany
W. Germany
W. Germany
Great Britain
Italy
Sweden
United States
United States
W. Germany
Case
No.
01b
02°
03°
04b
01b
02°
01C
die
03
02C
01C
Olb
01C
01C
02c
03C
04C
05C
06C
0?c
08C
OQC
1QC
11C
12c
13C
16C
17
01C
02C
04C
05C
02c,d
01M
02c-9
14". i
15C.J
03C,k
Birth
date
00-00-00
00-00-01
06-00-37
11-13-29
12-23-15
06-23-27
10-17-23
08-19-33
05-25-15
01-15-24
01-25-12
00-00-29
05-03-22
05-06-20
11-08-31
08-16-13
05-27-09
11-17-18
12-01-21
11-04-27
05-06-31
07-26-31
06-24-30
00-00-00
00-00-00
09-08-14
06-15-34
11-27-11
00-00-13
00-00-25
07-16-30
First
VCorPVC
exposure
00-00-00
00-00-46
02-00-66
00-00-57
03-00-50
08-14-51
1 2-09-48
11-15-55
11-28-45
07-06-52
06-19-44
01-17-62
08-00-44
10-07-46
09-09-54
06-12-51
10-14-46
09-13-49
08-19-44
05-08-50
06-23-55
10-14-57
10-01-57
00-00-00
00-00-00
00-00-46
00-00-65
00-00-45
08-18-38
00-00-00
00-00-00
Angio-
sarcoma
diag-
nosed (DX)
00-00-00
12-00-72
00-00-74
12-13-72
12-20-71
02-00-70
03-03-73
05-00-70
12-19-73
08-19-67
04-09-64
02-00-74
00-00-68
08-00-61
03-01-74
05-00-68
03-00-70
05-02-69
05-00-74
00-00-69
10-11-74
09-25-70
09-19-68
00-00-00
00-00-00
02-00-70
04-19-71
05-15-72
06-00-73
07-00-72
02-00-68
Age
at
DX
43
71
38
43
56
43
49
36
58
43
52
45
45
41
43
55
61
50
53
41
43
40
38
44
49
55
36
61
60
47
43
Yr from
1st exp.
to DX
19
26
8
15
22
19
22
14
28
15
20
12
24
15
17
17
23
20
30
17
19
11
13
17
11
24
6
27
36
00
14
Total
y
exp.
19
20
4
6
21
18
16
13
28
15
18
12
18
15
17
17
23
15
30
4
19
11
13
11
11
11
3
23
00
00
14
Date of death
00-00-67
12-00-72
12-24-74
12-00-72
01-04-72
10-20-70
03-03-73
09-28-71
12-19-73
01-07-68
04-09-64
Alive
03-23-68
08-29-61
Alive
05-10-68
03-16-70
05-02-69
07-04-74
03-27-69
Alive
12-14-71
01-25-69
00-00-00
Alive
12-00-70
04-16-71
08-16-72
07-03-73
02-15-73
10-10-73
aOO indicates unknown data.
^Awaiting details
cMicroscopically confirmed angiosarcoma of the liver.
^Pouring PVC oil mixture onto fabric bases.
Production of PVC sacks.
'Angiosarcoma involving liver, lung, and pericardium. Although difficult to determine, primary site seems to be
pericardium.
SProduction of vinyl chloride.
"Machine operator covering electrical wire with PVC plastic insulation.
'Diagnosis, sarcoma (possibly "angiosarcoma"), liver. Possibility of generalized neoplasm of the
reticuloendothelial cell system cannot be ruled out.
'Accountant at plant making PVC fabric.
kLoading pesticide cans with VC propellant.
Undesirable Effects
73
-------
I
cc
<
CO
g
C5
<
o
!_
3
to
O
J2
to
Medical history
H—
0 4;
0) eg
ra CO
Q "°
i/)
'S To 'g
aj .£ c
Q 0 ^
t/>
'55
— o
O c
Z.2
73
X
CO
CO
0)
O)
•^
to
O
Z
u
0
^_
o
"O
c
g
cu
i
1
o
I
CO
in —i cu
T— > O) tO
o <» E 5
_c E ro ..
a. 'z o >
^ o = £
< CL m X
to .— •
1-month history of anorexi
with abdominal pain and
back pain. Firm epigastric
mass. Died 6 days after ad-
mission with carcinoma-
tosis and pulmonary embol
o
in
en
CO
8
CSJ
CO
g
'01
c
to
•2 i
to o
Q. o
o> >-
X c/i
LL
CD
r~
(Excluded from series)
i
M—
O
to
E «
O C13
c QJ
'o o
CD m
0 Q.
^
o>
CD
CM
(Excluded from series)
to
E
0
c
CO
•o
"D to
1 5
'o c
i_ Q
to 2
O XI
LL
00
Ifl
CO
^
O. ' —
CO
73 '^
'i- c
CQ a>
CN ^. ps.
PS. "* ^
«- c\i u ^
2 E en *-' ^
c = >- o <- II
to c "- <-, CO V,
r— '~ QJ n l^ QJ
CD -, 1- u2 «rf 'P
.h — 0 0 C! CO
LL. < g CJ ?- DC
D 0
O ~
10 CD ° m ^
I CN «—' -° !E
Q i ^ to co
a? ""- u "^ ^ a
ro=) Lo o ^ ™
t~l * - ,_ C
O> ro —^ "^ Co
*"" ^- /-^ "^ C "to 1~
— — §°ooE^
CQ < 00 £ ° H -1
to"
2-month history of diarrhe
anorexia, and 20-lb weight
loss. Intermittent abdomi-
nal pain. Nontender, firm
epigastric mass. Died after
9 days with spontaneous
ruptured liver leading to
shock. Past history of
alcohol intake.
fs>
CD
CO
CSI
CD
UT)
CN
r—
6
*CJ3 '
c to
to
•— P .9 '">
ra O "o ?
Q. 0 4= "t
v m 0 >
X u) U U
^
CO
'*
t Q. |
o M- E ">
co ^T .5" .2
"O in 13 to
'C CT1 O CJ
CO r- '5 5
1 ° 6 ^
4-1 +-l tO •*-"
C C J± •>. O
to * to 0) ~ +1,
•£o'CO£-QCO«c
O-rCO O'+jCO'^ c CO
O>~CDOv>UD5;3-t-'
<^ '> *— 'a)rj
Pj5|_"to5— Co
/rt°C---E(-n'l/I
^ -c O t- c .E 0 5
O Q. ^ ps^ c w ^o to i_'
— ^ "I Q. " " "S <° "
S < 13 1- to -ff CJ "5. ~
«-ilslil
> o 9- "o S2 n ~ -§co
ui to co y Q-co"oco j^
~O™x^g'->T^ro'
CO
in
*~
CN
CO
^
ol
01 O
c a
CD
O 5 to
+-* t 'to
CD O O
Q. a i-
QJ ^~ O
X (/i M^
^
P*^.
^r
LO
DC
O
co
X
(O
-i
74
VINYL/POLYVINYL CHLORIDE
-------
.22
H-
t^
CO
LL
00
-n oo
O CD
"co C
G CO
'<= a
O tn
UJ D
CD
C
O
J2
•a
CO
!i
^ o
.__ _y
CQ <
C C
•~ co
to D.
~ !c
•2" ™
0 ^
1 p
CO w
00
f^
00
^
CO
r^"
in
«7
CO
6
'co
c
CO
o ^
£ E
CO O
a o
0) <-
I 8
O)
00
00
CD
O
D.
.C
•? 00
S CD
00
Q.
CO
T- CD
-1 CO
pain and weakness. 10-cm
nontender, sharp liver.
00
T3 Q_
cu GO
$ 8
V 2.
CC <
D
•a- m.
00 &
£
--^ CO
S? "0
O) +-<
C O
t 0)
co 73
s.s
1- in
Diagnosed by needle
biopsy. Slow deterioration
iceman
o
Q.
CD
>
CO
en
CD
£
5 .c
— CD
2e
^ a
o. 2
00 ^
II '
^ i
CO
CO
T3
CD
CO
CD
•D
O
5
CO
CD
o
CJ
QC CD
0 "P
t! oo
cu OM
3 CD
Q- «-
-a
cl i
herman
penter 1
00
Ti r^
'>- CD
03 ^>
CD
I"
Q- r^
E 6
O c CD
CO 13 <-
O)
>
O '4=
CM- a
5 g
-5 c
^s
E 2
Q.
^. 0
CD £
Q- "V
H a
distension, general edema,
icterus, fever, shaking
chills. Rapid downhill
course with death due to
renal and hepatic failure.
Past history of alcohol
intake.
5
o
in
i-
in
00
•g
o
t t/>
CO
i3 CD
CO >
^
o
o
o
usewife
stauram
O (D
I QC
D
^*
00
00
I
D
f=j
O en
O Q.
m <
Admitted 12-2-73 with short
history of RUQ abdominal
S
CN
OM
*-
00
CD
r—
OM
"—
6
en
c
CO
o co
£ E
CD O
a o
cu >-
I «
CO
HI
in
00
D
1- ^
CO "co
in c1
CM C
O ^'
(J 0.
CO h-
pain radiating to R shoulder.
Had RUQ tenderness. Open
. — .
cq J=-
S E
C?CM
^_ ^
'c "Si
0
z
oo .
r^ i-
E
CM -3
«— CD
If
.9 -o
f 1
cu O
! resection. Deteriorated
until death 34 days later.
u.
00
CO
CO
o
u
o
Undesirable Effects
75
-------
Table 6.19 (Continued).-HISTORICAL DATA, CASES OF HEPATIC ANGIOSARCOMA,
CONNECTICUT, 1935-197363
1. The angiosarcoma of the liver of this patient is believed to be different from most of the angiosarcoma seen in VC-PVC
workers. Sinusoidal dilatation and/or megalocytosis of hepatocytes were not seen. Also there was no hepatic fibrosis
similar to that seen in the VC-PVC workers.
2. Carcinoma involving pancreas and liver, possibly primary in the pancreas. A carcinoma of this type is not present among
the known VC-PVC workers whose lesions have been reviewed.
3. The carcinoid type bronchial adenoma seen in the lung of this patient almost surely has no relationship to the hepatic
lesions (hemangiomas) in the liver. No bronchial adenomas have been observed in the VC-PVC workers examined.
4. The angiosarcoma in this patient forms capillary slit-like spaces and replaces hepatic tissues. This type of histological
pattern is found only occasionally in the angiosarcomas of the VC-PVC workers and,in those patients.always involved
multicentric areas of angiosarcomas with sinusoidal and/or papillary pattern. These types of patterns are not observed
in this patient; thus, his hepatic angiosarcoma is not thought to be characteristic of the angiosarcomas seen in VC-PVC
workers. Also, this patient's liver is definitely cirrhotic with large deposits of iron. These features were not seen in the
VC-PVC workers with hepatic fibrosis and with or without hepatic angiosarcomas.
5. The diagnosis on this case is uncertain. One histological feature, i.e., the cells of the angiosarcoma are somewhat
"f ibroblastic", is dissimilar to angiosarcomas seen in most VC-PVC workers. However several other histological features
are similar to the VC-PVC workers' hepatic lesions, namely: sinusoidal dilatation, tectorial and enveloping features of
the sinusoidal lining cells, portal tract fibrosis, and variability in size of hepatocytes. There are sufficient histological
similarities that prevent this case from being excluded as a so-called "VC-PVC type case."
6. It is uncertain whether this is an angiosarcoma of the liver or even a primary sarcoma of the liver. It is not like any of
the other tumors seen in the livers of VC-PVC workers.
7. The histological features of the hepatic angiosarcoma and portal tract fibrosis seen in this case are similar in nearly all
respects to the lesions we have seen in most of the VC-PVC workers' livers.
8. Only needle biopsies have been reviewed from this patient's angiosarcoma. The amount of material is insufficient for a
definite comparative statement, but histological features,which are believedmost characteristic of the angiosarcomas in
VC-PVC workers, are not seen.
The study population—from 33 domestic plants—was composed of 8384 individuals with at least 1 year of
occupational exposure to VC. The study population included retired and terminated as well as currently
employed workers. The vital status of these workers was ascertained as of December 31, 1972, and cause of
death was determined based upon available death certificates. Observed mortality was then compared to
expected mortality based upon the United States male population, taking into account age and time of
death. Standardized mortality ratios were computed for total mortality and specific causes of death.
Exposure categories were defined subjectively by industrial hygiene and safety personnel at each plant.
They identified those jobs and work locations with the highest exposures to VC, then classified other job
categories accordingly as medium or low. This procedure was reasonable for estimating relative exposure
within a given plant. However, it could not assure comparable exposure categories across all plants or over
time, since a low exposure in past years might be numerically equivalent to a relatively high exposure in
recent years. An exposure index was calculated for each worker as a time-weighted average of exposure
categories over the period of employment, thus defining two overall exposures, low and a high.
Followup procedures were able to define the vital status for 7128 workers (85 percent of the study
population) as of December 31, 1972. Among the 352 workers known to have died, death certificates were
obtained for 328. The mortality calculations considered only those workers who had been traced, which
assumes that the mortality experience of these workers was equivalent to that of workers not traced. The
median birth year for those traced was 1931 compared to 1920 for those not traced. The median year in
which exposure began was 1962 among those traced compared to 1953 among those not traced. The
76 VINYL/POLYVINYL CHLORIDE
-------
median duration of employment for those followed was 80 months in contrast to 44 months for those who
were not traced. Thus, while those traced had worked about twice as long as those not traced, their
employment began about 10 years later. Accordingly, the mortality experience among those not traced
might have been different from that in the study population considering the increased latent period. About
60 percent of the study population entered employment in 1960 or later, indicating that the majority of
workers in this study could not be followed long enough to assure observation of all potential long-term
effects. Included among the 7128 workers traced, however, were 854 workers with exposure of 20 or more
years and 1640 workers with exposures of 15 years or more.
After examining the effects of exposure index (low versus high), and duration of exposure (less than or
greater than 5 years) upon mortality, as well as interaction effects between level and duration of exposure,
the following observations were made in the Tabershaw/Cooper study:
• Compared to the general male U.S. population, the overall mortality of the study population was
approximately 75 percent of what would have been expected. (Note - this favorable overall mortality
frequently occurs in occupational groups—even if an industrial hazard increases the risk of death from
a particular cause—since occupational groups are usually healthier than the average population.)
• Increases in specific cause of death over the expected occurrence in the U.S. male population were
not statistically significant.
• No deaths identified as angiosarcomas of the liver were found other than those previously identified.
Standardized mortality ratios (SMR) for malignant neoplasms as a whole increased with increasing
exposures as measured by level, duration, or both (see Table 6.20). For example, 36 malignancies were
observed in the high exposure group with 5 years or more exposure compared to 26 expected cases. Among
those with greatest exposure, cancers of the liver (primarily angiosarcoma), respiratory system, and brain;
cancers of unknown primary site; and lymphosarcoma occurred more frequently than expected. These
findings were not statistically significant. However, the authors of the Tabershaw/Cooper study considered
them suggestive of a relationship between exposure to vinyl chloride and increased cancer risk at multiple
sites.
Table 6.20. OBSERVED DEATHS/EXPECTED DEATHS AND STANDARDIZED MORTALITY
RATIOS IN VC WORKERS WITH EXPOSURE INDICES OF 1.5 OR GREATER,
BY DURATION OF EXPOSED EMPLOYMENT64
Cause of death with LC.D. No.
All causes
Tuberculosis (001-019)
Tuberculosis of respiratory system (001-008)
Malignant neoplasms (140-205)
Malignant neoplasms, buccal cavity and
pharynx (140-148)
Malignant neoplasms, digestive organs and
peritoneum (150-159)
Malignant neoplasms, respiratory system
(160-164)
<60 months
exposure
Obs/exp
38/47.93
0/0.76
0/0.71
5/6.57
0/0.23
1/1.67
1/1.79
SMRa
79
0
0
95
0
76
71
>60 months
exposure
Obs/exp
119/147.81
0/1.57
0/1.48
36/26.11
0/0.99
11/7.47
12/8.50
SMRa
8!b
0
0
141
0
151
144
Undesirable Effects
77
-------
Table 6.20 (continued). OBSERVED DEATHS/EXPECTED DEATHS AND STANDARDIZED
MORTALITY RATIOS IN VC WORKERS WITH EXPOSURE INDICES OF 1.5 OR
GREATER, BY DURATION OF EXPOSED EMPLOYMENT64
Cause of death with LC.D. No.
Malignant neoplasms, genital organs
(170-179)
Malignant neoplasms, urinary organs
(180-181)
Malignant neoplasms, other and unspecified
sites (190- 199)
Leukemia and aleukemia (204)
Lymphomas (200-203, 205)
Diabetes mellitus (260)
Major cardiovascular and renal diseases
(330-334, 400-468, 592-594)
Vascular lesions affecting CNS (330-334)
Rheumatic fever and chronic rheumatic heart
dis. (400-402,410-416)
Arteriosclerotic heart disease (420)
Nonrheumatic endocarditis (421, 422)
Hypertensive heart disease (440-443)
Other hypertensive disease (444-447)
Chronic and unspecified nephritis and renal
sclerosis (592-594)
Influenza and pneumonia (480-493)
Ulcer of stomach and duodenum (540, 541)
Appendicitis (550-553)
Hernia and intestinal obstruction (560,
561,570)
Gastritis, duodenitis, enteritis and colitis
(543,571,572)
Cirrhosis of liver (581)
Hyperplasia of prostate (610)
Symptoms, senility and ill-defined conditions
(780-795)
All other diseases (residual)
Motor vehicle accidents (810-835)
Other accidents (800-802, 840-952)
Suicide (963, 970-979)
Homicide (954, 980-985)
No. of workers
Person-years
<60 months
exposure
Obs/exp
0/0.29
0/0.26
1/1.18
1/0.44
1/0.71
0/0.61
7/16.54
2/1.87
0/0.82
5/10.41
0/0.57
0/0.76
0/0.27
0/0.54
0/0.99
1/0.35
0/0.08
0/0.14
1/0.14
0/1.56
0/0.01
1/0.80
0/4.02
7/6.05
4/4.73
3/2.40
1/2.18
1240
12,828
SMRa
0
0
107
288
178
0
54C
135
0
6ia
0
0
0
0
0
362
0
0
904
0
0
158
0
146
107
158
58
>60 months
exposure
Obs/exp
1/1.41
0/1.26
7/3.51
1/1.13
4/1.84
2/2.04
62/70.46
4/8.19
2/2.04
46/47.65
1/2.32
2/3.10
2/0.81
0/1.23
0/3.13
0/1.25
0/0.9
1/0.49
0/0.41
1/5.08
0/0.13
0/2.30
6/11.88
2/7.43
2/7.96
, 4/4.62
0/2.76
1817
19,305
I SMRa
73
0
204
90
222
100
90
50
100
98
44
66
253
0
0
0
0
209
0
20
0
0
513
28
26
88
0
Standardized mortality ratios adjusted for deaths with cause unknown.
bSigmficant at 5 percent level.
Significant at 1 percent level.
78
VINYL/POLYVINYL CHLORIDE
-------
6.3.2.2 Dow Chemical Study—Dow Chemical conducted a long-term mortality study of 594 chemical
workers exposed to vinyl chloride between the years 1942-1960.65'66 The study population was defined as
production workers at one manufacturing facility who worked in areas with potential vinyl chloride
exposure. Each job classification was assigned an exposure rating of low, intermediate, or high, depending
on existing industrial hygiene data. This was the only available mortality study in which VC exposures
could be reconstructed using relative degrees of exposure. Three categories of exposure were defined based
upon estimated time-weighted average (TWA) concentrations for an 8-hour day:
• Low exposure group, TWA below 25 ppm VC (964,000 jUg/m3 ).
• Intermediate exposure group, TWA ranging from 25 to 200 ppm (64,000 to 512,000 jug/m3).
• High exposure group, TWA of 200 to 300 ppm (512,000 to 768,000 jug/m3)- Also included in the
high group were those with TWA exposures in the intermediate range but also exposed tofrequently
unpredictable excursions above 1000 ppm (2,560,000 jug/m3).
• A fourth category of indeterminate exposure was defined for individuals working in areas where
sufficient air monitoring data were not available.
Assignment to exposure groups was determined by the highest exposure experience for one or more
months. By this procedure, the lowest exposure category contained only individuals with low exposure
whereas the highest exposure group included some individuals with predominantly lower exposures.
Durations of exposure were categorized as less than 1 year and 1 year or longer. (Note: effects of exposure
well above 1 year were not adequately examined, although the analysis did consider the impact of latency
period.) Of the 594 employees in this study, 72 had histories of exposure to both VC and arsenicals. In
view of the cancer risk associated with arsenicals, the 72 with arsenic exposure were excluded from
dose-response relationships related to VC.
Expected deaths in this cohort were determined from U.S. white male mortality rates. Death certificates
were obtained for 86 of the 88 individuals known to be dead. Of the 148 individuals who had left the
company, 131 were traced. Among the individuals who had worked with arsenicals and VC, 7 of 10 deaths
were due to neoplasms, compared to 1.9 cancer deaths expected; 3 lung cancers were noted in this group.
Among workers exposed to VC, but not arsenicals, observed deaths were 91 percent of the expected deaths
based upon the U.S. white male population. No deaths due to angiosarcoma of the liver or other liver
cancers were noted in the group exposed to VC but not arsenicals. For this group as a whole, total
malignancies were only 13 observed, compared to 15.4 expected.
The effects of exposure grouping upon malignancy rate were examined for this cohort exclusive of arsenical
workers. Of 163 individuals in the high-exposure group, 27 had 20 or more years at low to high exposure
and only 19 had 10 or more years of only high exposure. Of the 13 malignancies observed in this cohort, 9
occurred in the high-exposure group, compared to 5.1 expected. Due to the small number of deaths
involved, this difference was not tested for statistical significance. To examine for possible latent effects,
the mortality experience of workers with 15 or more years since onset of exposure was studied. In this
group, nine malignancies were observed, eight in the high-exposure group. Accordingly, eight of the nine
malignancies observed in the high-exposure group occurred 15 or more years after onset of exposure. Table
6.21 summarizes the results.
The authors of the Dow Chemical Study concluded that workers exposed to VC at levels above 200 ppm
(512,000 /Mg/m3) experienced an "apparent increase in overall malignancy rate." When exposures were kept
below 200 ppm (512,000 /ug/m3) the malignancy rate decreased. Angiosarcomas of the liver were not found
at any level of exposure. Among the workers exposed above 200 ppm (512,000 Mg/m3) TWA, the increase
in overall malignancy was not statistically significant. The authors also commented as to possible
cocarcinogenic effects of other exposures with VC, particularly benzene, cigarette smoking, and arsenic.
Undesirable Effects 79
-------
Table 6.21. SUMMARY OF DOW MORTALITY STUDY65 -a
Category
All causes
All cancers
All VC exposed
Only high exposure group
< 1-yr exposure
> 1-yr exposure
High exposure group with
15 yr after onset of exposure
< 1-yr exposure
> 1-yr exposure
Observed
mortality
78
13
9
3
6
8
3
5
Expected
mortality
85.7
15.4
5.1
2.2
2.9
3.2
1.3
1.9
Standardized
mortality
ratio (SMR)
91
84
176
136
209
250
231
263
aStudy population: (1) excluded workers exposed to arsenicals; (2) years exposed—1 or more; (3) onset since first expo-
sure—no restriction; (4) size of cohort—594 total, 522 with VC exposure only; (5) number successfully followed up—577.
In reviewing these data, certain strengths in this study are evident, especially the availability of measured
vinyl chloride exposures and the successful followup of over 95 percent of the whole. However, there are
also several weaknesses. The level and duration of exposure to vinyl chloride is widely variable among
individuals within the high exposure group. Since only 1 month of exposure to levels of 200 to 300 ppm
(512,000 to 768,000 Mg/m3) TWA was required to place an individual in the high exposure group, only 9
of the 163 men in the group had exposure exclusively to high levels for periods of 10 or more years. And
only 66 of the 163 individuals had 10 years or more of exposure.
The importance of both duration and level of exposure on the development of a malignancy is reflected in
the fact that seven of the nine malignancies which did occur in this group were in men with more than 10
years of exposure to high levels. It is possible that the malignancy rate might have been even higher if more
of the men had exposure for 10 years or greater to 200 to 300 ppm (512,000 to 768,OOOAtg/m3) of vinyl
chloride.
6.3.2.3 Harvard University Study-Momon et al.24>67 conducted a proportional mortality study among
workers in a VC plant in Calvert City, Kentucky, and among workers in a vinyl chloride polymerization
plant in Louisville, Kentucky. Death certificates were used as a source of cause of deaths, and were
obtained for 142 out of 161 of the white males who were employed at these plants and were known to be
dead. When death certificates were not available, cause of death as recorded in company abstracts was used.
Causes of 161 deaths were tabulated for the period 1947-1973, and these were compared with the expected
distribution of deaths as calculated from proportional mortality ratios for U.S. white males, taking into
account age and time of death. Since mortality patterns among workers in both plants were similar, these
groups were combined for purposes of data analysis.
Overall, a statistically significant 50 percent excess in deaths due to cancer was observed. Five cases of liver
angiosarcoma were identified, in addition to one cancer of the gall bladder, one of the common bile duct,
and one unspecified case of liver cancer. All told, a 900 percent excess was observed in cancers of the liver
and bilary tract. Excluding angiosarcoma cases from this analysis, a 275 percent excess was observed in the
remaining cancers. Five cases of brain tumors also were found, as were 13 cases of lung cancer, representing
320 and 60 percent excesses above the expected frequency of these cancers, respectively. A 100 percent
excess in deaths due to suicides also was noted.
80
VINYL/POLYVINYL CHLORIDE
-------
In addition to these overall cancer excesses, an increasing trend of cancer deaths with time was observed.
No excess deaths due to cancer were observed prior to 1965. However, in the period 1965-1969, about a 50
percent excess in total cancers was observed, and in the period from 1970 on, a 100 percent excess. (This
trend is generally consistent with the clustering in recent times of reported occupational cases of liver
angiosarcoma.)
These data imply that at least two other forms of cancer, lung and brain, in addition to liver cancer, are
increased among vinyl chloride workers. The experimental design did not permit the absolute risk of death
in the study population. However, the observed excesses in specific cancers combined with the time trend
for all cancers was considered by the authors to suggest a relationship between exposure to vinyl chloride in
the work environment and cancer at multiple sites.
6.3.2.4 Mount Sinai Study—Nicholson et al.68 utilized labor union and company records to identify a
cohort of 257 individuals, each with a history of occupational exposure to VC in a polymerization plant for
at least 5 years subsequent to 1946. This cohort included all individuals employed in this plant during the
period 1946-1963. The mortality status of these individuals was evaluated from the 10th anniversary of
their employment through April 1974. The minimum 5-year exposure criterion was established to focus
upon the effects of significant durations of exposure. Beginning observations after only 10 years or more
since onset of first exposure emphasizes the possible long-term effects of VC. The majority of individuals in
this cohort, however, were exposed to VC for a period of 20 years or under.
This cohort represents a relatively young group since over half of the men were under age 37 when they
entered the cohort. Over half of the men are presently employed in the PVC production facility, although
not all in locations with VC exposure.
Of the 257 individuals in this cohort, 255 (or 99 percent) were successfully traced and their current health
status evaluated. The majority of these men were directly employed in production although maintenance
men and nonproduction workers were also included. Thus, exposures varied considerably among study
subjects. No measurements of actual exposures were available. Over half of the workers in this cohort
reported experiencing symptoms of dizziness, headache, or euphoria during work periods; 14 had
experienced episodes of loss of consciousness. The authors concluded that peak VC exposures in this
production facility may often have exceeded 1000 ppm (2560 mg/m3) and may occasionally have
approached 10,000 ppm (25,600 mg/m3).
Included among the 24 deaths identified in this cohort were three confirmed cases of angiosarcoma of the
liver. These preliminary findings suggest an excess of 25 percent in all deaths and a 131 percent excess in all
cancer deaths although in neither case did these excesses reach statistical significance. In addition to liver
angiosarcoma, one brain cancer and two lymphomas were observed, causing the authors of this study to
suspect a possible relationship between VC exposure and these rare cancers.
6.3.2.5 NIOSH Study-A. study of mortality and morbidity among current and past employees at two vinyl
chloride polymerization facilities was conducted by the National Institute of Occupational Safety and
Health (NIOSH).58 The criteria for selection of facilities studied were, in order of decreasing priority: (1)
involvement in the polymerization of vinyl chloride for at least 15 years, (2) existence of a sizeable work
force, (3) location in a state where vital statistics are easy to obtain, and (4) existence of a medical program
in the plant.
Since cancer often takes many years to become clinically evident, the study population was restricted to
individuals with 5 or more years employment and at least 10 years since the beginning of employment in
departments directly involved in polymerization of VC. The study population consisted of 930 white males.
Attempts were made to trace study members from the time they terminated employment to December 31,
1973. The authors were unable to trace 285 individuals (31 percent); hence the mortality experience of
these people compared to those who were traced is unknown. All individuals not traced were considered to
be alive and were included in the analysis, thereby making any findings of increased mortality in this study
cohort a conservative estimate of risk. Observed risks of death in the study population were compared to
Undesirable Effects 81
-------
expected risks based upon mortality rates for the general white male population of the United States.
Measurements, or estimates, of previous exposure levels to VC were not included in this study.
A total of 109 deaths were observed among these polymerization workers compared to 105 which would
have been expected. This difference is not statistically significant; however, most occupational groups have a
favorable mortality experience compared to the general population. Any deaths in the 31 percent not
traced would have increased the observed number of deaths. An evaluation of specific causes of death
indicate that, except for cancer, causes of death in the study group did not differ from those expected in
the general population. However, a statistically significant (p<0.01) 57 percent excess in cancer deaths was
observed above the expected. Excess cancer deaths were not limited to any single organ system—excesses
being observed for cancers of the respiratory system, blood forming tissues, and the brain and central
nervous system. Deaths due to liver cancer in this population were about 12 times above the expected
number, and brain cancer deaths were fivefold higher. These latter contrasts were statistically significant (p
< 0.01 and p< 0.05, respectively). The report does not state whether excesses in liver cancer other than
angiosarcomas were observed. The majority (25 out of 31) of observed cancer deaths in this study
population did not occur until at least 15 years following first exposure to VC.
6.3.2.6 Comparison of Mortality Studies—A comparison of the mortality studies shows reasonably
consistent results; that is, an overall excess in cancer mortality among workers exposed to VC for long
durations (Table 6.22). Both the NIOSH and Mt. Sinai studies, which employed the same study criteria,
suggest increased overall mortality among these workers, but neither comparison shows statistically
significant differences. In all five studies, workers exposed for 5 or more years to high levels of VC had
greater than expected frequencies (ranging from 41 to 150 percent) for all cancers; however, in only two
studies (NIOSH and Harvard) were these excesses statistically significant at the 0.05 level or lower.
In evaluating these observed mortality effects among VC workers, it must be recognized that the workplace
situation may include exposures to other carcinogens and/or liver toxins in addition to VC and that any one
of them may have contributed to the observed effects. While this situation makes it difficult to draw final
conclusions from these human studies with regard to the role played by VC in the development of liver and
other cancers, toxicologic studies have observed liver angiosarcoma and other cancers in mice, rats, and
hamsters following inhalation exposures to VC at concentrations of 50 ppm (128,000 /jg/m3) and
higher.2S'7'5 The liver angiosarcoma lesions observed in these animal studies, combined with the human
observations in industry, strongly indicate that VC exposure is etiologically related to liver angiosarcoma in
man.
With respect to levels of vinyl chloride exposure required to produce liver angiosarcoma, most, but not all,
occupational cases reported to date have occurred among PVC workers and consequently may generally
have involved TWA exposures in excess of 200 ppm (512,000 jug/m3) with peak excursions in excess of
1000 ppm (2,560,000 Mg/m3). The most definitive evidence for past exposures among these workers comes
from actual 8-hour average air measurements ranging from 120 to 385 ppm (307,200 to 985,600 Mg/m3)
and excursions of 2000 to 4000 ppm (5120 to 10,240 mg/m3) among highly exposed PVC workers at the
Dow Chemical Company from 1950-1959.59 Evidence of peak exposure excursions in excess of 1000 ppm
(2,560,000 Mg/m3) among PVC workers in past years is also derived from the frequent reports of
neurological symptoms among such workers. Existence of odors attributed to VC for much or part of the
workday at these plants would tend to support these observations since the odor threshold for vinyl
chloride is believed to be 250 ppm (640,000 Mg/m3) or higher.69
Reports of liver angiosarcoma among workers exposed to VC but not involved in the production of PVC,
however, including that of an accountant in a U. S. vinyl cloth plant would tend to argue that at least for
some individuals, liver angiosarcoma may occur at much lower exposures than encountered among PVC
workers. Cases of liver angiosarcoma are reported in a worker employed at a VC monomer plant in Sweden
and in a worker from England employed at a vinyl cloth plant.58 Data from the Dow Chemical Company69
show TWA exposures at monomer plants in the years 1973-1974 to generally be under 10 ppm (25,600
Mg/m3) although short-term exposures in excess of 100ppm(256,000 Mg/m3) have been reported. A survey
by the National Institute of Occupational Safety and Health has shown VC levels in fabricating plants to
82 VINYL/POLYVINYL CHLORIDE
-------
O
>
o
Q
LU
CO
O
Q.
X
LU
CO
CC.
CC
O
o
CO
in
CO
>
cc
O
c-
">•
"5,
«
"H
CD
e
CD
I
'15 CO
.£ *
I
CO
g
z
i/i
*o
-D
i
0
Q
CD
Q.
O
0
$
^ >
CD "a
^f
CD
CD
8
U
CD
LL
CM
r-
CM
CO
CO
Number of plants
c c
.2 .2
-M -M
0 O
'£ '5 1 1 1 CD
to c/> T—
CD CU
O O
Z Z
o
0 4-i LO LO TO ^
E CO CM CN §5
o Z
LO <£
>
o ^ ^
i- *~ O LO -^ £35
o « co •* §; o
E % 05 CO § -
j_ CD - — •
O 4-.
LO Oo> C/O ZH-
co
•Oj
1 LO
CD «-
CN CO
>- CN
CO
co r-
O LO
§6 °3 V>
Q. JJ C
^ X ^ 0 —
•^ CD ~3
— 0 5 co- > 5
•* +; a.
*- *~ ':: X
«- <- g CD
.15 CD _^
CD "O CO
> H_ CU
'J3 O "O
la 8
,3 C
Q CD CD
DC < <
CO
en en
CD CD
(/> tf>
to CD
>• >"
—
—
en en
CD CD
CD O
> Z
CD CD
to CD
CD -C
Multiple cancer sit
suggested
Angiosarcoma of t
liver found
v- in
q q
o o
V V
Q. a
+^ *j
CO CO
c c
CO CO
o o
(D ro
(/3 C/5
Undesirable Effects
83
-------
range from 1 to 12 ppm.71 In the case of workers at fabricating plants, vinyl chloride exposures may result
in part from release of trapped monomer in the PVC during processing and/or from inhalation of PVC dust
containing entrapped monomer. While it is difficult to reconstruct exposures to vinyl chloride in these
instances, it is likely that exposures for the workers involved in the fabrication process are considerably less
than those for workers involved directly in the production of the monomer or the polymer.
6.3.2.7 Infante Study in Ohio33-A study of the distribution of congenital anomalies was undertaken in
residents living in three northeastern Ohio communities of Painesville, Ashtabula, and Avon Lake, where
vinyl chloride production facilities are located. The population of the three communities ranges from
24,000 in Ashtabula to 12,000 in Avon Lake. The vinyl chloride production facility in Ashtabula began
operations in 1954; the one in Avon Lake in 1946. Of the two plants located in Painesville, one began
operation in 1946 and the other began in 1967.
The number and rates of children with congenital malformations in the three cities were computed and
these observations were compared to an expected value derived from the congenital malformation rate in
the entire state. The difference between the observed and expected numbers of malformations in each city
was significant at p <0.01 level (Chi Square). When all three cities were combined, the difference between
observed and expected was significant at p < 0.001.
When birth malformations in the three cities were compared to congenital malformations for residents
living in the remainder of the three counties in which the cities were located, the difference between the
number of malformations per 1000 live births between the cities and the counties was significant at p
<0.001.
Malformation rates were computed for nine cities in the vicinity of the index communities and two of the
nine had significantly greater numbers of malformations than expected. One community, Geneva was
located 12 miles from Ashtabula and a second community, North Ridgeville, was located 8 miles from
Avon Lake.
Significant excess of defects of the central nervous system, upper alimentary tract, genital organs, and club
foot were observed in the study communities.
The observed differences could not be attributed to differences in race, maternal age, or reporting.
The number of deaths from central nervous system (CNS) tumors in the white population aged 45 years
and older for the period 1958-1973 was significantly greater (p <0.01) in the combined communities of
Ashtabula, Avon Lake, Painesville, and North Ridgeville than in the one comparable group for the state as a
whole. The excess number of CNS tumor deaths in white males in Painesville and North Ridgeville
corresponds with an excess of CNS anomalies among stillbirths and live births in the two communities.
Limitations of the study:
• No consideration was given to other factors which can contribute to mutagenesis, teratogenesis, and
carcinogenesis such as genetic factors, exposure to background radiation, other industrial exposures,
and experience with infectious agents such as certain viruses.
• No information on exposure to vinyl chloride either in the occupational setting or in ambient air is
presented in the paper. It is not known whether the plants have used the vinyl chloride monomer or
polymer.
• It is not known whether the congenital anomalies occurred in families with occupational exposure to
vinyl chloride or in families without occupational exposure which would then suggest ambient
exposure.
84 VINYL/POLYVINYL CHLORIDE
-------
Despite these limitations, the findings in this study suggest that exposures to vinyl chloride either from an
occupational setting and/or ambient air may contribute to excess risk of anomalies particularly of the
central nervous system,, This area deserves further study.
A subsequent study by the Cancer and Birth Defects Division of the Center for Disease Control72
confirmed a moderate increase in CNS malformations in Painesville, Ohio, but could not establish any
association between cases and vinyl chloride exposure.
6.3.2.8 SUVA Study (Organization of Insurance Carriers in Switzerland)73—During the period from
February to August 1974, the 62 persons involved in vinyl chloride fabrication and the 33 persons involved
in vinyl polymerization in Switzerland were subjected to in-depth examinations. No evidence of illness was
found among the workers. The average length of exposure was 13 years in the fabrication plant and 17
years in the polymerization plant.
No information was provided on exposure levels to vinyl chloride nor was there a description in the report
of the type of clinical assessment carried out on the workers.
6.3.3 Nonmalignant Effects
In addition to the carcinogenic effects of VC, a considerable body of evidence has become available relating
to nonmalignant effects, including reactions of the liver. The vast majority of evidence in this regard comes
from observations among industrially exposed individuals.
Lester et al.,74 in 1963, conducted animal and human acute toxicity experiments with VC. Three men and
three women were exposed for 5-minute periods twice each day, separated by a 6-hour interval, for three
successive days to VC concentrations up to 20,000 ppm (51,200 mg/m3),. Acute toxic effects (dizziness,
nausea, dulling of visual and auditory cues, and headaches) were observed at concentrations above 8000
ppm (20,480 mg/m3).
Kramer and Mutchler53 correlated clinical and environmental measurements for 98 healthy male workers
exposed to VC for periods up to 25 years. Exposure indices were based upon actual air measurements since
1950 and expressed as cumulative dosage (ppm-years) and career time-weighted averages, considering the
time each worker spent in critical job classifications. History, physical examination, and laboratory tests
were determined on exposed workers in other departments. Of 21 clinical parameters studied, six—systolic
and diastolic blood pressure, BSP retention, icterus index, hemoglobin, and beta-protein—showed
significant correlations (p <0.05) with exposure variables, cumulative TWA, and cumulative dose. The best
correlation (coefficient of multiple determination, 0.4) was between exposure and BSP retention, which is a
measure of liver cell damage. Based upon these observations, the authors considered the possibility that
"...repeated exposure to vinyl chloride at TWA levels of 300 ppm (768,000 /ig/m3) or above for a working
lifetime together with a very low level of vinylidene chloride may result in slight changes in certain
physiologic and clinical laboratory parameters. The possibility of some impairment in liver function must be
considered even though no overt clinical disease was evident in any of the individuals studied."53
The BSP test is an insensitive index of overall liver cell function. Therefore, when the BSP test shows
abnormal results, the liver is already extensively damaged. At present, it is a widely used and useful liver
function test. BSP (Bromsulphalein) is taken up rapidly by liver cells, concentrated and stored within the
cytoplasm, and conjugated enzymatically with glutathione. In diseases that produce hepatic cell
dysfunction, significant quantities of unconjugated BSP may reenter the blood stream and be retained in
the body.74
It is noteworthy that a good dose-response relationship between BSP retention and career TWA exposure
was observed over the entire range of exposures examined. Based upon the derived regression equation, BSP
retentions of 12.5 percent were expected among those with TWA's of 300 ppm (768,000 )ug/m3) and 5.6
percent among those with TWA's of 100 ppm (256,000 jug/m3). A BSP retention in excess of 5 percent is
considered to be abnormal in clinical medicine and suggests that substantial damage to liver cells may have
Undesirable Effects 85
-------
occurred,75' Judging from the data as presented, a small fraction of individuals with career TWA's of 50
ppm (128,000 Aig/m3) had abnormal BSP retention tests suggesting that liver damage had occurred.
Exposure to other liver toxins such as alcohol was not adequately considered in this study.
Observations of liver damage among workers who fabricate PVC plastic into finished products as well as
among workers who convert VC monomers into the polymer indicate that injury to the liver among
exposed workers is important and that such damage may occur at lower levels of exposure than is usually
encountered in the production of PVC.
These data suggest that exposure to 50 ppm VC (128,000 ;ug/m3) is associated with increased BSP
retention and the evidence is greater in VC/PVC workers than in the general population.
In German studies, enlarged livers and spleens, as well as abnormal results of tests of liver function, were
found in PVC production workers.76'77 Microscopic examinations of biopsy specimens revealed evidence
of liver pathology in a high percentage of cases. These workers had a history of employment ranging from
1.5 to 21 years. No measures of past VC exposure were available, and it is not known whether adequate
comparisons were made with control groups (see Table 6.23). Further, Table 6.23 shows increased BSP
retention in workers involved in processing the polymer, which is consistent with the findings of Kramer
andMutchler.53
Following these initial reports of liver damage in PVC production workers from Germany, additional
studies were carried out in 50 individuals with varying durations of exposure to VC during the production
of PVC.78 These studies indicated that there was a relation between the duration of exposure to VC and
the severity of liver damage, as determined by microscopic examination of biopsy specimens. The most
severe evidence of liver pathology was among 16 workers with exposure in excess of 10 years. All exhibited
evidence of liver abnormality, and two cases of liver angiosarcoma were observed. All of the German
workers, with exposure lasting 3 years or less, had some form of liver damage, although generally not as
severe as that found in workers with longer exposures. Five workers who were involved in postpolymeriza-
tion of VC were examined, and all five showed signs of minimal damage to the liver parenchymal cells.
Although these studies do indicate a relationship between exposure duration and histologic evidence of liver
damage, the lack of exposure data on these workers makes it difficultto determine what levels of exposure
may have been responsible for such damage. Failure to compare exposed workers with a suitable control
group not exposed to vinyl chloride and failure to consider the effect of alcohol intake are other limitations
which deserve mention.
Examinations of 70 out of 128 workers in a PVC production plant revealed evidence of extensive
abnormalities based on biochemical indicators and other tests.79 These workers were employed an average
of 7.7 years in the industry (range 6 months to 21-3/4 years). Upper abdominal complaints were present in
42 of 70 workers, and symptoms such as tiredness, dizziness, parasthesias, and arthralgia were frequently
reported. Thrombocytopenia, increased BSP retention, and splenomegaly were present in a majority of
cases, 81, 67, and 57 percent, respectively. Reticulocytosis was also common (41 percent), and abnormal
liver enzymes, esophageal varices, and leucopenia were also observed. Unfortunately, effects of exposure,
both level and duration, were not evaluated in this study. Further, the frequency of abnormal findings
among workers not exposed to vinyl chloride was not studied so that it is difficult to accurately judge the
effects of such exposure. Findings such as splenomegaly, thrombocytopenia, and increased BSP retention
in the majority of instances does, however, suggest that damage in excess of the expected frequency among
the general population had occurred in these workers, though these changes were not necessarily specific
for vinyl chloride.
Additional studies were carried out among workers in Germany employed in PVC processing plants.80 Such
workers would have had a somewhat lower exposure than those involved in the direct polymerization of
PVC from the monomer. Medical examinations were conducted on 15 such workers who were employed an
average of 5 years, ranging from L5 to 13 years. Seven complained of pressure and/or pain in the upper
abdomen. Thrombocytopenia and increased BSP retention were found in 7 of 25 workers, although not
86 VINYL/POLYVINYL CHLORIDE
-------
necessarily concommitantly. One worker had an enlarged spleen. In biopsy specimens from four workers,
one showed histologic evidence of liver damage similar to, although less severe than, that observed in PVC
production workers.
Creech and Makk81 studied liver disease among PVC workers at the B. F. Goodrich plant in Louisville,
Kentucky. A total of 1183 employees had blood tests to screen for evidence of liver damage. These workers
included individuals not involved in the direct production of PVC, such as maintenance personnel,
administrators, and secretaries. On initial screening tests, 315 of 1183 (26.6 percent) showed at least one
abnormal blood test, and 41 (3.5 percent) had two or more abnormal tests. Among the 315 tested for a
second time, 75 had a persistent abnormality. The most common observed abnormality in this test was an
elevated alkaline phosphatase, although increased bilirubms and serum glutamic-oxaloacetic transaminase
(SGOT's) also were observed. Based upon this initial battery of screening tests, 116 individuals were given
more extensive blood tests; some abnormalities were found in 59 (about 50 percent) of those examined.
Seven of these individuals had major abnormalities that required additional test procedures. The highest
percentage of abnormal batteries of tests (10.9 percent) occurred among PVC production workers; although
abnormal batteries also were found in other production workers, in maintenance workers, and in clerical
personnel.
Depending upon results from the battery of tests, more elaborate diagnostic procedures such as liver scans,
hepatic arteriograms, and liver biopsies were initiated. Of 17 individuals undergoing such tests, 11 cases of
portal fibrosis, indicating severe damage to the liver, were discovered. Two cases occurred among workers
not directly involved in the production of PVC. Angiosarcoma of the liver was found in 2 of the 11
workers—both were involved in PVC production.
While this study in Louisville did document evidence of damage related to vinyl chloride, i.e., liver
angiosarcoma, the extent to which less severe liver damage may or may not be related to vinyl chloride is
not at all clear from this study. Although there is suggestive evidence that less severe damage may also have
occurred, adequate comparisons were not made with matched control groups, and measurements of vinyl
chloride exposure were not made. Accordingly, level of exposure could not be related to observed effects.
Failure to correlate abnormal tests with duration of exposure or with latent period since onset of exposure,
and lack of consideration of alcohol intake are additional limitations in this study.
Miller et al.82 examined the work force at a PVC production plant in Niagara Falls. Duration of exposure
and alcohol intake were considered, but exposure was not measured directly. A total of 354 workers was
examined, 267 currently employed in a vinyl chloride polymerization plant (encompassing the entire work
force), and 87 former workers. Hepatosplenomegaly was observed in a high percentage of current and
former workers (15.0 and 3.4 percent, respectively), with the most frequent occurrence in each category
among workers exposed for 20 or more years. Hepatosplenomegaly was observed among 6 percent of the
current workers with not more than 2 years' exposure, but was less frequent among former workers. Nearly
one-third of the current workers exposed for 20 or more years showed enlarged livers or spleens.
Hepatosplenomegaly generally was higher in each exposure category among those with a history of
significant alcohol intake compared to those with no significant intake. In each group, the frequency of
hepatosplenomegaly was related to duration of work exposure. Tests of liver enzymes were also abnormal,
even among those with exposure of not more than 2 years' duration, and these abnormalities were generally
more frequent with longer exposures. Elevated alkaline phosphatase, the most frequently observed
biochemical abnormality, did not correlate well with ethanol intake, but did correlate significantly with
duration of exposure to vinyl chloride.
In addition to finding evidence of liver dysfunction in these workers, the study showed that abnormal
pulmonary function and chest X-rays also were associated with longer exposures to VC.83
While the association of abnormal findings with duration of exposure to VC suggests an effect relationship,
these results were not compared with the frequency of abnormalities in a matched control group.
Undesirable Effects 87
-------
Table 6.23. SYNOPSIS OF AN AMNESTIC, CLINICAL, BIOCHEMICAL,
PERITONEOSCOPIC AND HISTOLOGIC DATA OF 50 PVC WORKERS11'3
6
z
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
.+;
~co
c
o
•(-»
CO
z
Ger
Ger
Ger
Ger
Ger
Ger
Ger
Ger
Ger
Ger
Ger
Ger
Gr
Ger
Ger
Gr
Gr
Ger
Ger
Turk
Gr
Ger
Gr
Gr
Gr
Ger
Ger
Gr
Turk
Gr
Turk
Turk
^
cu*
01
<
50
56
47
46
51
51
52
53
44
39
48
27
40
52
35
45
42
54
31
33
30
34
36
30
35
30
31
31
33
30
36
33
C
_g
CO
o
H-
C/T
CO
o
.Q
Q
PM
PM
PM
PM
PM
PM
PM
PM
PM
PM
PM
PM
PM
PC
PM
PM
PM
PM
PC
PM
PM
PM
PM
PM
PM
PM
PM
PM
PM
PM
PC
PM
PM
PM
PM
o
E
^
CD*
w
H3
O
Q.
X
OJ
21/9
18/6
18
17/6
17
16
15/3
14/6
13/9
13/9
13
12/6
12/3
0/9
11
9
8/6
6/8
4/0
6/6
5/9
5/9
5
5
5
5
4/9
4/9
4/9
4/0
1/0
4/0
3/9
3/9
3/6
CO
T3
O)
CD*
CO
.£
—
O
-C
o
fj
<
-
16-32
—
8
48-64
16
8
8
8
16
16-32
—
28
—
16
16
8
16
16
8
8
16
16
8
16
16-32
8
—
8
4
7
-
N
CD O
C +"1
~£ £
^ w
CD
O
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
CO
c
1 J5
~g £
CO O
CD .—
a TJ
Q.
D
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
-Q
"co
OT
CD
C7
O
CO
a
CD
I
1.5
2
3
3
1
2
1.5
1.5
3
2
2
10
1.5
1.5
3
1.5
6
1.5
1.5
3
_a
~CO
cu
E
o
c
CD
Q.
CO
2
5
1.5
(2)
(11)
1.5
11
3
(10)
1.5
6
6
3
1.5
E
o
C O
1?
.__ t:
-Q Q
E*~
A
u Ci
CD
CO
1.1
1.2
2.0
1.1
1.1
1.1
1.4
1.8
1.7
1.8
2.2
'E
LO
C i_
O 0)
c To
+J --P
CD ^^
i— ^
Q_ /\
C/3 ~~~~
CD
8.7
6.1
15.6
10.2
9.1
11.3
16.7
5.6
17.6
6.2
13.0
15.1
8.8
13.8
6.9
9.6
21.5
7.6
8.5
7.5
8.3
7.1
25.6
6.4
22.5
6.8
=
D
E
00
^Z
CM
A
O —
(n
CO
15
13
13(22)
19
25
15
15
31
15
13
16
19
13
25(24)
(20)
15
15
21(21)
90
32(27)
15
15
23
25
15
25
23
E
^
E
ON?
CN
CM
I — A
Q_ • — •
CD
CO
15(23)
19(24)
15(47)
17
22
13
23
15
13
16
17(31)
13
23(63)
(27)
(28)
13
23(40)
25
190
30(40)
15(24)
15
25
58
17
34
17
CU
to
CO
Q.
O
_C
a
CD
C
' —
CO
\/
<
92
83
85
50
58
66
68
88
VINYL/POLYVINYL CHLORIDE
-------
icroosteolysis
<
+
+
+
+
c
o
c
CD
O
c
.c
Q.
CD
•a
3
CO
c
CO
tr
(+)
+
(+)
(+)
+
(+)
co
c
o
'(/
_CD
^c
CO
+
+
+
+
+
Peritoneoscopy
s-
"CD
I.
CD
O
CO
13
CO
G/Nod
U/G
Nod
G
U
G
U
G
G/Nod
G
G
G
U
G
G
G
CO
1
.Q
CO
C/1
Q.
CO
0
R
R
R
R/S
R/S
R/S
?
R/S
R
R/S
R/S
R/S
R
R
S
R
?
S
C/R
VI
CO
CD
CO
CO
a
CO
o
A/I
?
A
A
?
1
1
A
1
A
A
?
A
Not examined
U
U
G
G
R
R
R
R
S/PC
R
R
1
R
A
A
A
A
A
Histology
M—
0 J2
ollagenization
sinusoidal wal
O
(+)
+
+
+
+
+
+
(+)
+
(+)
+
+
(+)
+
+
+
+
+
+
(+)
+
+
+
+
+
(+)
+
+
!_
O .j
~d °
•M '£i "CD
C co 0
CD i-
§ ~ ™
"E
LU
+
++
+
+
+
(+)
++
(+)
+
+
(+)
(+)
+++
Not examined
(+)
(+)
(+)
(+)
(+)
(+)
+
+
(+)
(+)
+
+
(+)
+
+
(+)
'in
O
.a
"to-
a
CU
CO
+
+
(+)
+
+
+
(+)
+
(+)
++
(+)
+
(+)
(+)
.c
Q.
CO
D)
C
'o
X!
CD
N
'co
!§
o.
CO
Normal (liver
scan)
15x 11 x 12
Markedly en-
larged (liver
scan)
12x7x 11
14x 12x6
12x9x9
Splenectomy
(15x 12x7)
10x8x4
10x8x 5
24 x 12x7
15x 11 x 7
14x5x 12
Normal (liver
scan)
23 x 14x9
13x 10x5
12x9x 5
Normal (liver
scan)
Splenectomy
(500 g)
23x 10x 18
18x 11 x 12
14x 12x 13
15x9x 10
14x9x 10
17 x 12x 14
13x 10x 10
14x 6x8
13x9x 14
16x 10x4
14x 13x6
Normal (liver
scan)
14x6x 11
13x5x 11
c
CD
CD
c/1
niargement of
LU
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
CO
CU
sophageal vari
LLJ
+
+
+
+
+
+
+
+
+
n
E
•- E
c -5
m r*\
hrombocytopi
O150X 10'
h-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Undesirable Effects
89
-------
Table 6.23 (continued). SYNOPSIS OF ANAMNESTIC, CLINICAL,
BIOCHEMICAL, PERITONEOSCOPIC AND HISTOLOGIC DATA OF 50 PVC WORKERS11-a
o
z
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
>
~co
c
o
CO
^
Ger
Ger
Gr
Gr
Gr
Gr
Turk
Gr
Gr
Ger
Gr
Turk
Gr
Gr
Gr
Gr
Gr
Ger
^
^~
QJ
cn
<
32
40
31
38
47
30
30
33
41
41
36
33
48
44
39
45
37
36
c
o
+-J
O
t/l
TO
+
+
+
+
+
+
+
+
+
+
+
^D
-^
cn
QJ
/.
O
CO
QJ
I
6
1.5
3
3
1.5
1
1
3
7
6
8
.a
>*
"co
cn
QJ
E
o
c
QJ
Q_
CO
4
1
E
o
C 0
-Q 7
D cn
.5 P
f- T
— A
3 O-
QJ
CO
1.1
C
E
LO
C t.
O CD
C c^
aj °^
i_ j_n
a. A
CO ~~
CD
12.5
8.6
5.2
9.7
8.4
6.1
6.3
5.1
9.1
15.1
8.8
7.0
,E
^D
E
CO
CN
I A
O -—-
fn
CO
19
19(20)
(20)
13
13
17
19(22)
17
28
13(33)
15
(20)
(22)
(88)
28(20)
E
^r
c
c
CN
CN
I — A
Q_ CJ-
fH
CO
16
15(31)
15(24)
13
17(47)
17
30(23)
13(80)
17
(47)
(72)
28(38)
QJ
CO
CO
a.
in
O
_c
a
QJ
C
-^
CO
~^~
<
56
66
58
110
Abbreviations used are: Nationality: Ger, German; Gr, Greek; Turk, Turkish; Job Classification: PM, VC polymerization;
PC, processing of the polymer; Peritoneoscopy: U, slightly irregular or undulated; G, granular to finely nodular; Nod,
"Centimeters below costal margin.
90
VINYL/POLYVINYL CHLORIDE
-------
•—
_>
Acroosteo
+
c
o
c.
CD
E
o
c
0)
a
a>
Raynaud 1
(+)
+
vi
o
c
C/3
+
+
Peritoneoscopy
u,_
CD
Surface rel
U/G
U
U
G/Nod
G/Nod
'is*
O
.a
03
Q.
CD
CJ
S
R
C
C
(S)
C/R
C/S
C/R/PC
S
R
(R)
(R)
R/PC
S
C
if)
cu
(/)
c/)
m
Capsular v<
1
*
A
A
A
A
A
A
A
Histology
O <"
c 15
.2 5
to —
N -^j
II
= .E
O ">
CJ
+
+
+
+
+
+
+
(+)
+
+
+
~S •<-
"c c
CO g _
C to CJ
Enlargeme
prolifer;
littoral
+
+
(+)
w
(+)
+
+
(+)
(+)
+
(+)
+
(+)
(+)
(+)
(+)
+
(+)
i/)
o
^
4—
lo
Q.
CD
+
t-
Q.
CO
O)
c
'C
>
-°
CD
i E
CD O
Q.
CO
14 x9x7
Slightly en-
larged (liver
scan)
13x7x9
10x9x7
12x6x8
15x I2x ?
16x6x8
13x7x 12
11 x 7x 10.5
12x9x7
11x8x5
12x8x6
Normal (liver
scan)
9x6x6
12x8x9
Normal (liver
scan)
11x5x7
12x6x 11
QJ
"a
^
o
*->
c
Enlargeme
+
+
+
+
+
+
+
+
+
+
+
+
+
to
Q)
U
cu
>
Esophagea
+
— .
CO
•IE
CD CO
o 2
>- V
Thromboc'
(>150;
+
+
+
+
+
+
+
+
+
+
+
+
+
coarsely nodular, C, comma-like or stellate; R, finely to coarsely reticular; S, small scar-like patches to broader concave
postnecrotic scars; PC, patchy "perihepatitis cartilaginea," A, augmented; B, increased; Histology: (+), minimal; +,
slight; ++, moderate; +++, marked.
Undesirable Effects
91
-------
Not all U.S. studies observed effects among workers exposed to VC, One negative study involved an
evaluation of health surveillance data on 335 workers exposed to VC at Dow Chemical Company.84 This
survey was based upon a multiphase screening program. The study population was comprised of production
employees who had worked for at least 1 year between 1942 and January 1972 in areas with potential VC
exposure and who were also employees at the Dow Midland Division between February 1967 and March
1974, the period of the multiphasic health screening program.
Exposure categories were based upon industrial hygiene data compiled from 1950 on, using estimated
time-weighted average concentrations for an 8-hour day. The high exposure group was defined as those with
exposures above 200 ppm (512,000 /ug/m3) for a duration of 1 month or longer; the intermediate group
had exposures from 25 to 200 ppm (64,000 to 512,000 /xg/m3); and the low group had exposures under 25
ppm (64,000 ;ug/m3). A fourth exposure category, undefined, was established for those individuals without
sufficient industrial hygiene data. A subjective evaluation indicated that most individuals in this latter group
were exposed in the low to intermediate range. The participation rate in the multiphasic screening program
was about 80 percent in all exposure groups. Measured exposure data were available for most of the
workers. By definition, the high exposure group may have included individuals with predominantly
low-level exposures throughout the majority of their work experience, but the lowest exposure category
would not have included workers with a history of high-level exposure to vinyl chloride.
Because of revisions in the multiphasic screening program in 1970, the data obtained before and after this
date were analyzed separately. A control group of matched pairs for sex, age, smoking history, and month
of examination, and, where possible, date of hire was included in the analysis. The parameters studied prior
to 1970 included tests of pulmonary function, blood pressure, white blood count, total bilirubin, serum
glutamic-pyruvic acid transaminase (SGPT), and alkaline phosphatase. The only statistically significant
difference (p <0.05) between exposed and matched-pair control groups was for decreased diastolic blood
pressure in the high-exposure group. No differences between exposed and control groups were noted in
terms of pertinent historical questions including shortness of breath, chronic cough, jaundice, gastrointes-
tinal trouble, numbness in hands or feet, cancer, anemia, or blood problems.
The health surveys after 1970 included similar historical information and laboratory tests for pulmonary
function, hemoglobin, white blood count, serum glutamic-oxaloacetic transaminase (SCOT), lactic acid
dehydrogenase (LDH), total protein, protein albumin, and protein globulin. Statistical analysis of alkaline
phosphatase was not performed due to changes in laboratory procedures. No statistically significant
differences between exposed and control groups were found. The availability of measured exposure data
and the comparability of laboratory results with the inclusion of matched-pair controls for each exposure
category in this investigation are important factors not found in some other studies. Information was not
available as to possible toxic chemical exposure in the matched-pair controls, and the impact of duration of
exposure was not considered. Based upon the available data, the authors concluded "... below 200 ppm
(512,000 jug/m3) nothing of statistical significance has been observed."84 It is unlikely that the
high-exposure group was really exposed to TWA of 200 ppm (512,000 /ug/m3) over their full employment
period, since high exposures of only 1 month's duration were sufficient to place an individual in this
category. An earlier study at Dow Chemical Company suggested that liver damage may have occurred
among some workers with TWA exposures of 50 ppm (128,000 jug/m3).53
Kotin85 reported the results of a study of currently employed workers exposed to VC at Air Products and
Chemicals facilities in Calvert City, Kentucky, and at Pace, Florida. Also included were results of a death
certificate survey of PVC employees who had left the company, or had died while employed at the
company. A detailed medical history and physical examination (with X-rays and laboratory tests) were
performed for each employee.
At Pace, 13 of 201 employees examined showed abnormal findings. Persistent minor abnormalities were
found on retesting of three employees. These findings did not justify further immediate retesting, although
retesting was scheduled 90 days later. The remaining six employees were given liver scans, all of which were
normal. One case of acroosteolysis was found at Pace.
92 VINYL/POLYVINYL CHLORIDE
-------
At Calvert City, 97 of 291 employees examined showed an abnormality on initial testing, partly due to
equipment malfunction. Following re test, 29 employees showed persistent abnormalities; 16 were
considered equivocal, not justifying further immediate additional testing, but indicating retesting 90 days
later. Among the 13 employees who were immediately retested, six were found to need further laboratory
and physical examinations. Of these, two cases of Gilbert's disease and two cases of gallbladder disease were
found, including one with coexistent hepatitis. An additional case of chronic persistent hepatitis without
gall bladder disease was found. Examinations of the death certificates did not indicate any relationship
between exposure to vinyl chloride and cause of death. In light of these results, except for the one case of
acroosteolysis, the authors concluded that there was not evidence to identify VC as a causative agent in
disease. No cases of angiosarcoma were identified in this survey. Neither information on age, level and
duration of exposure, or a definition of an abnormal test was given. Measurements of vinyl chloride
exposure were not made, nor is there any indication as to what percentage of workers examined was
suspected to have high vinyl chloride exposures. Workers who had left the plants were not included in the
death certificate survey. The frequency of abnormal findings on physical examinations, particularly
enlarged livers and spleens, was not indicated. Accordingly, an adequate latent period following onset of
first exposure may not have been present to allow effects such as liver angiosarcoma to be evident.
Dernehl70 surveyed 36 PVC plants to determine if there were factories in which no cases of liver
angiosarcoma were observed. Plants in which angiosarcoma had occurred were not included in the analysis.
Similarly, plants with operating experiences of 10 years or less were not considered. The remaining 16
plants included 2372 persons currently employed and 1471 previously employed. Included were 787
persons who had worked more than 10 years, and 104 who had worked more than 20 years with VC. Also
included were 1402 persons with a time since initial exposure to VC of more than 10 years, and 416
persons with 20 years or more. Of these 16 plants, only 2 had conducted measurements of vinyl chloride
concentrations prior to 1970. The presence of vinyl chloride odors was used to estimate past exposures in
the other plants. The odor threshold for vinyl chloride is above 250 ppm (640,000 jug/m3) and perhaps as
high was 2000 ppm (5120 mg/m3) for unacclimated workers. On this basis, 4 companies indicated that
vinyl chloride odors were detectable for most of the work day prior to I960; only 2 of 16 reported odors
most of the day between 1960 and 1970; and no plants reported odors this frequently after 1970. All 16
plants reported the occasional presence of VC odors at some time in the past. Since January 1, 1974, 3285
examinations were conducted on employees in these companies, including 872 retirees. This constitutes a
followup rate of nearly 90 percent of all employees, but only 60 percent of those who were previously
employed. Of these, 3249 were given liver profile tests as included in the SMA-12, that is,bilirubin, SCOT,
LDH, and alkaline phosphatase. An abnormality in one of these four tests was reported in 15 percent of
those examined, but this fell within the limits of abnormality observed among 900 Union Carbide
employees not exposed to VC and 400 office personnel with no known occupational exposure to the
chemical. Similarly, the occurrence of abnormalities in two tests (2.7 percent) and in three tests (1.3
percent) were comparable to those found in control groups. No definition of an abnormal test was given.
Based upon these observations, the author concluded that "Examinations of these men have failed to show
the existence of abnormal liver function tests in greater proportion than would be found in a control
population. There is no case of angiosarcoma of the liver among these 1402 men even though their
exposure time is sufficient for disease to have occurred..."70 It should be noted that less than 10 percent of
these workers had direct exposure of more than 20 years'duration and 416 workers had a time lapse since
first exposure of 20 years or more.
Considering the lack of information on control and exposed groups and the many laboratories participating
in these analyses, it is difficult to draw any conclusions with regard to the frequency of abnormal liver
function tests among these vinyl chloride workers. The fact that only 60 percent of former workers who
have had the greatest durations of exposure were included in these examinations of liver function tests is a
serious shortcoming, as is failure to examine the influence of duration of VC exposure upon abnormal liver
function.
A summary of the data showing nonmalignant effects of vinyl chloride is shown in Table 6.24. The results
of these studies and their limitations have been discussed above.
Undesirable Effects 93
-------
Table 6.24. SUMMARY OF OCCUPATIONAL FINDINGS RELATING NONMALIGNANT
LIVER DAMAGE TO VINYL CHLORIDE EXPOSURE
Study group
(number studied)
PVC production
workers76 /77
PVC production
workers64 -
liver biopsy
study78
PVC
polymerization
workers63 —
liver biopsy
study80
PVC production
workers79 >9S
PVC processing
workers81-84
PVC production
workers and
non-PVC
production
workers (1 183
total)81
Current and
former PVC
production
workers (354
total)82'53
Level of
exposure
Unspecified
Unspecified
Unspecified
Unspecified
Unspecified
Unspecified
Unspecified
Duration of
exposure
11/2 to 21 years
3 years and under
10 years and more
Unspecified
6 months to 21%
years (average
7.7 years)
V/z to 13 years
Unspecified
-2 years
5-10 years
20 years or more
Observed effects
Enlarged livers and spleens. Abnormal
BSP retention. Biopsy of liver showed
portal fibrosis.
Biopsy showed mild liver damage in all
workers.
Relation between duration of exposure
and severity of damage with most severe
liver pathology observed in workers with
10 or more years of exposure.
5 of 8 workers examined showed evidence
of mild damage to liver parenchyma based
on liver biopsy.
Upper abdominal complaints, lethargy.
and paresthesias common complaints.
Thrombocytopenia, increased BSP reten-
tion, and splenomegaly found in majority
of workers.
7 of 15 workers complained of pressure or
pain in upper abdomen and had throm-
bocytopenia and increased BSP retention.
Biopsy showed mild liver damage similar
to that in PVC production workers.
1 16 of 1 183 (about 10 percent) showed
significant biochemical abnormalities.
Abnormal liver function tests found in
non-PVC production workers. 1 1 cases
of portal fibrosis found on liver biopsy.
2 of which were in workers not directly
involved in PVC production.
Hepatosplenomegaly observed in 6% of
workers exposed not more than 2 years.
Sharp increase in hepatosplenomegaly.
Nearly one-third of workers had hepa-
tosplenomegaly. Elevated alkaline phos-
phatase correlated with duration of
exposure. Abnormal lung function tests
found.
94
VINYL/POLYVINYL CHLORIDE
-------
Table 6.24 (continued). SUMMARY OF OCCUPATIONAL FINDINGS RELATING
NONMALIGNANT LIVER DAMAGE TO VINYL CHLORIDE EXPOSURE
Study group
(number studied)
PVC production
workers53
PVC and non-
PVC production84
workers (335)
PVC workers
(492) ss
PVC workers
(3843)70
PVC production
workers (594)6S
Level of
exposure
TWA expo-
sures up to
300 ppm
TWA expo-
sures of
25-200+ ppm
Unspecified
Estimated as
greater than
250 ppm
prior to 1960
TWA expo-
sures (8 hr
day)
< 25 ppm
25-200 ppm
200-300+ ppm
Duration of
exposure
Up to 25 years
1 year and
greater
Unspecified
Includes ex-
posed 20 years
and more
< 1 year
> 1 year
Observed effects
Abnormal liver function (BSP retention)
correlated with TWA exposures. Evidence
of abnormal BSP retention at TWA expo-
sures of 300 ppm and suggestive evidence
of BSP retention in some workers exposed
to TWA of 50 ppm.
No adverse effects related to angiosar-
coma by any of the criteria studied below
200 ppm.
One case of acroosteolysis only evidence
of injury attributable to VC.
No increased abnormalities above levels
in control groups.
No deaths due to angiosarcoma or other
liver cancer. For workers exposed to
>200 ppm, apparent increase in overall
malignancy, not statistically significant.
These observations of liver injury among PVC workers, and particularly among workers not directly
involved in PVC production, have potentially important implications with respect to the health of the
general population exposed to vinyl chloride. In reviewing these findings it is, however, important to
recognize that other toxic agents, either work or nonwork related, such as liver toxic drugs or alcohol,
could have contributed to many of these abnormal findings. Further, some of the biochemical screening
tests are not specific for liver injury, though others, such as BSP, are. Ideally, one would like to know the
prevalence of liver injury among comparable nonindustrially exposed populations before drawing final
conclusions regarding the effect of exposure to vinyl chloride upon the liver from the above studies. Most
studies were lacking in this regard. In spite of difficulties with the present studies such as noted above, there
is suggestive evidence presented for at least minimal liver damage associated with vinyl chloride, which may
be observed with durations of exposure under 2 years. Though there is reason to believe that cessation of
exposure to vinyl chloride would cause a reversal in some of this damage, there is also evidence that in some
people, this damage is not fully reversible and may even progress further. For example, one vinyl chloride
production worker examined by liver biopsy at the National Institute of Health showed persistent and
perhaps progressive liver pathology 2-1/2 years after the cessation of exposure despite an absence of
abnormalities in biochemical tests of hepatocellular function. There is also concern that the histologic
changes in the liver observed in PVC workers may represent premalignant changes that would increase the
risk of developing angiosarcoma in future years.
Undesirable Effects
95
-------
Any final conclusions regarding the implications of these findings for the general population who are
exposed to levels of vinyl chloride in the air generally much lower than in occupational situations must
await completion of additional studies. However, several observations do suggest that exposure to vinyl
chloride in the air may pose some risk to health at these lower levels.
6.3.4 Ongoing Research in Health Efi'ects from Vinyl Chloride
An attempt has been made to determine what studies are presently being conducted on VC. The following
is a listing, obtained through personal communication,86'87 of research that is presently underway, as well
as preliminary findings as of June 1, 1975. This probably does not represent a complete listing and there
may well be other studies of which the authors of the report were not aware.
6.3.4.1 Studies by the Center for Disease Control* 7 -Drs. Henry Falk and John Herbert of the Cancer and
Birth Defects Research Division are reviewing 300 reported cases of hepatic angiosarcoma and correlating as
much background information on the victims as is available to determine possible causative agents. Drs.
Hans Popper and Lou Thomas will review the pathological reports and specimens for verification of
diagnosis.
Dr. William Flynn and Mr. Lawrence Edmonds are doing a follow-up of the Infante study on birth defects
in two communities in Ohio, Their study will attempt to determine if the nearby VC plants have any
impact on the observed unusually high occurrences of central nervous system birth defects.
6.3.4.2 Studies Funded by the Manufacturing Chemists Association*n —&. 24-month study, using Dr.
Maltoni's protocol, but restricted to 50 ppm and higher doses, has been in progress for over 18 months. The
12- and 18-month data have been correlated, and can be provided. Dr. J. C. Calandra of Industrial Bio-Tech
Laboratories is conducting the study.
Drs. B. A. Schwetz and P. J. Gehring of Dow Chemical Company are directing a metabolism study. It
has been completed, and results have been promised.
6.3.4.3 Harvard University-Peter Bent Brigham Hospital Study*7~A study on endoplasmic acute liver
injury by vinyl chloride has indicated that a single 6-hour inhalation exposure to vinyl chloride
monomer (5 percent) produces extensive vacuolization of centrolobular liver parenchyma and focal
midzonal necrosis in the hepatic lobuole in rats pretreated with 0.1 percent sodium phenobarbital in
drinking water. Ultrastructurally, vacuolization consists of dilation of cysternae of rough endoplasmic
reticulum and, in the same cells, smooth endoplasmic reticulum coalesces into discrete aggregates
resembling denatured membranes. The findings support the hypothesis that vinyl chloride is hepatotoxic
because it is converted into a toxic metabolite by components of the mixed-function oxidase system
(MFOS) of liver endoplasmic reticulum.
In another study, preliminary findings indicate that single 6-hour inhalation exposures to vinyl chloride
monomer (5 percent) produce acute liver injury in rats pretreated with sodium phenobarbital (400 n
mole/kg, daily, for 7 days, by gavage). Pretreatment with Arochlor 1254 in the same manner appeared to
render animals exquisitively sensitive to VC, as shown by an increase in serum glutamic-oxalacetic
transaminase (SCOT) of approximately 5 (for phenobarbitol) to 10 (for VC) times normal, respectively. If
the activation of VC to a hepatotoxin occurs during metabolism, an initial reaction would most likely occur
via the multimolecular MFOS, the enzyme system responsible for conversion of most xenobiotics to more
readily excretable metabolites. This metabolism mechanism may lead to toxification or detoxification
reactions. Correlations between induction of specific MFOS components and the degree of VC-induced
hepatic traumas as measured by increased serum transamenase (SCOT, SGPT) at 24 hours following VC
exposure reveal significant relationships between injury and increased NADPH cytochrome P-450
reductase activity (as measured by reduction of cytochrome C) and total cytochrome P-450 content. Injury
appears related to morphologic changes in the endoplasmic reticulum. Hepatic injury following inhalation
exposure to 1,1-dichloroethylene (0.02 percent) differs strikingly from that observed after administration
of vinyl chloride, in that it appears to involve plasma membranes, mitochondira, and chromatin, but not the
96 VINYL/POLYVINYL CHLORIDE
-------
cndoplasmic reticulum. In contrast to vinyl-chloride-induced reactions, induction of cytochrome P-450
appears to protect against 1,1-dichloroethylene.
6.3.4.4 International Agency for Research on Cancer, Report of the Working Group** -The IARC has held
several meetings in an attempt to coordinate the various epidemiological studies being conducted in
different countries on the oncological hazards associated with VC exposure. A primary consideration is the
general terms of the types of studies needed to be done and the principles underlying their design. IARC
recognizes the fact that cohorts need to be established now for long-term prospective followup; also, the
failure of risk to become apparent after 10 years does not preclude the emergence of such a risk, perhaps at
a different target organ, many years later. A figure of 90 percent was tentatively proposed as the minimum
acceptable loss during followup of cohort. Attempting to unify the results of so many different studies, the
IARC is considering the establishment of a pathology review panel, with the World Health Organization
Cancer Unit, and a register of liver angiosarcomas, which would operate in conjunction with the registers of
both the United States and the United Kingdom.
6.4 ECOLOGICAL EFFECTS
The possible ecological problems associated with release of vinyl chloride into the environment are just
coming under scrutiny. For some time it has been known that the plastic polyvinyl chloride is not readily
biodegradable. Microorganisms are not able to utilize the plastic or are able to do so only after an extended
period of weathering. Although acetylene and ethylene are both capable of being reduced by microbial
activity,89-90 their chlorination seems to make them less amenable to attack by microorganisms. Available
evidence does indicate that alkynes may be oxidized by peptone-grown pseudomonas species.9 * Very little is
known regarding the biological metabolism of alkynes.92 The fact that PVC is not readily biodegradable
has led to the difficulties in disposal. Incineration, the chief means of disposal, is not without its problems.
Though hydrogen chloride is evolved when burning refuse, the amounts released do not compare to those
released when polyvinyl chloride and polyvinylidene chloride plastics are incinerated. Addition of
polyethylene and polystyrene plastics to normal base refuse containing no plastics had no effect on chloride
ion emissions because these plastics contain no chlorine. Addition of 2 percent polyurethane foam resulted
in slightly increased chloride emission up to 689 ppm (1763 mg/m3); with a 4 percent addition, emission
increased to 751 ppm (1922 mg/m3). Adding PVC to the normal refuse increased chloride ion emission to
1990 ppm or 5094 mg/m3 (0.1990 percent) for the 2 percent addition, and to 3030 ppm (7756 mg/m3)
for the 4 percent addition.93 During burning, most of the chloride present in refuse and in the
polyurethane and polyvinyl chloride materials, which were added to the base refuse in the test work, was
evolved as hydrogen chloride. No free chlorine gas or phosgene was detected.94
6.4.1 Vegetation
The effect of hydrogen chloride gas on vegetation has not been studied in any detail. This probably reflects
its unimportance as a phytotoxicant. Hydrogen chloride gas is easily scrubbed from flue gases and the major
sources are point sources; therefore, it has not been emitted into the atmosphere in large amounts. The
incineration of chlorine-containing plastics in large amounts could change this picture unless scrubbers are
used.
The effects of hydrogen chloride gas on vegetation were noted in the mid-19th century in the vicinity of
alkali plants in Europe and Great Britain.93 The highest concentration of hydrogen chloride recorded in
stack gases was 0.45 mg/m3 in 1874. No further crop damage due to this gas was reported in Great Britain
after the passage of the Alkali Act of 1906. In the United States, damage due to hydrogen chloride gas has
been reported by Weiler,95 Hindawl96 and Wood.97 In the USSR, Antipov98 reported hydrogen chloride
gas damage to ornamental plants near a chemical factory that released fumes once or twice a month.
Species which were affected included oriental poppy, daisy, belleflower, columbine, bluets, and pylox.
Only one study specifically reported the combustion of PVC as the source of the hydrogen chloride gas.97
Smoke from burning PVC insulation at a wire salvage operation in northern Pennsylvania extensively
damaged several northern hardwood species.
Undesirable Effects 97
-------
Bohne" reported hydrogen chloride gas damage to shrubs, trees and flowers near a hospital incineiator.
Not all plants are sensitive to hydrogen chloride gas. Means and Lacasse100 tested the sensitivity of 12
coniferous and broad-leaf tree species to hydrogen chloride gas. The 4-hour fumigations were conducted
under controlled conditions at a temperature of 27°C, relative humidity between 78 and 85 percent, and a
light intensity of 1.4 x 104 ergs/cm2-sec. Under these conditions, the only symptom noted on conifer
needles was a tip necrosis on white pine at 8 ppm (20,480 /jg/m3), on Douglas fir at 12 ppm (30,720
iug/m3), and on Norway spruce at 19 ppm (48,640/Jg/m3). Austrian pine and arborvitae were not injured at
concentrations of 18 and 43 ppm (46,080 and 110,080 jug/m3), respectively. Symptoms on broadleaf
species included marginal and interveinal necrosis and necrotic flecking. Tulip poplar was injured at 3 ppm
(7680 pg/m3), European black alder and black cherry were injured at 6 ppm (15,360 jug/m3), and sugar
maple and Norway maple were injured at 7 ppm (17,920 ng/m3). Red oak was not injured at
concentrations up to 13 ppm (33,280 /jg/m3).
There are few studies on the effects of vinyl chloride in the environment. In 1962 a study by Heck and
Fires' ° ' found that VC can cause significant injury to plants. Heck and Pires used five different fumigants
at three different levels and ranked them in the following order: ethylene > acetylene > propylene >
ethylene oxide > vinyl chloride (Table 6.25). The injury symptoms shown for acetylene, propylene, and
vinyl chloride were identical to those shown for ethylene. Ethylene is usually considered as a
physiologically active gas rather than a toxic gas, such as sulfur dioxide. Ethylene affects a great number of
physiological phenomena in plants-such as ripening of fruits, abscission of plant parts, proliferation of
tissue, inhibition of growth, and variations in cellular metabolism.102 Ethylene is a product of plant
metabolism, but VC has not been reported from natural sources.
6.4.2 Other Effects
The effects of VC upon microorganisms have not been studied. There has been little work done to
determine whether alkynes can be metabolized by microorganisms.91 Ethylene is adsorbed by soil;103 VC
may be also.
Table 6.25. COMPARISON OF THE TOXICITY LEVELS OF THREE CONCENTRATIONS
OF FIVE FUMIGANTS ON SEVERAL PLANT SPECIES101 -a
Toxicity
level
1
2
3
4
5
6
7
Fumigant
Ethylene oxide
Ethylene
Propylene
Acetylene
Acetylene
Vinyl chloride
Acetylene
Propylene
Ethylene oxide
Vinyl chloride
Propylene
Ethylene oxide
Vinyl chloride
Concentration,
ppm
1000
10, 100, 1000
1000
1000
100
1000
10
100
100
100
10
10
10
98
aPlants were fumigated for 7 days in each fumigant at each concen-
tration. A qualitative comparison with 1 causing the death of all
plants and 7 showing no effect.
VINYL/POLYVINYL CHLORIDE
-------
A by-product of VC production, EDC tar, has been disposed of by dumping into the North Sea. When EDC
tar, a mixture of short-chained aliphatic hydrocarbons, is dumped into the ocean, it gradually sinks. As the
tar sinks, the components gradually dissolve in the water. Therefore, relatively little of the tar accumulates
in the sediments. It also has a tendency to adhere to a large variety of substances, plankton among them,
and form a film or layer around the particles.
Studies by Jernelov et al.1 °4 indicate that marine animals rapidly accumulate EDC tars from contaminated
sea water. An accumulation factor of 2900 was estimated for shrimp (Leander adspersus) exposed to 0.01
ppm (25.6 jug/m1) EDC tar for 48 hours. Accumulation of low molecular weight compounds of EDC tar is
highest from water, whereas the high molecular weight compounds show the greatest accumulation through
the food chairt. These conclusions are in agreement with the results of studies dealing with chlorinated
hydrocarbon (Cl-C) compounds such as DDT in seawater. Dieldrin has also been shown to accumulate
rapidly through solution and much less slowly through the food chain.104 Compared to DDT, PCB, and
other Cl-C aromatic substances, (lie biological half-time is short (1 day to 3 weeks). This suggests that the
effects of EDC components might not be as severe as those of DDT, PCB, and other chlorinated
hydrocarbons.
Studies made to determine the effects of EDC tars on different stages in the life cycle of the barnacle
Balanus balanoides L showed that the stage II nauplii were ten times more sensitive than the older stage V
and VI larvae,105 Age, therefore, seems to make the barnacles more resistant to the EDC tars.
In an attempt to determine some physiological aspects of EDC tars at the cellular level, the microorganism
Kschcrichia coli was studied.10<> The death of the intact cells was shown to be due to the breakdown of the
permeability of the cytoplasmic membrane. The authors suggest that since most known biological
membranes are formed according to similar principles, the action of EDC tar on the cell membranes of
higher organisms would be similar.
6.5 VINYL CHLORIDE RELATED COMPOUNDS
AND OTHER CHEMICAL CARCINOGENS
6.5.1 Related Compounds
The compelling evidence of the carcinogenicity of VC from both an epidemological and toxicological
standpoint raises the question of the possible carcinogenicity of other related chemicals in the ambient air.
Production figures and major uses for chemicals of industrial importance with structure similar to vinyl
chloride are summarized in Table 6.26.I07~109 Compounds similar in structure and metabolism to vinyl
chloride have not been adequately studied for possible carcinogenicity. The following are examples of such
compounds:
• 1,1 dichloroethylene (vinylidine chloride)—This is a known potent hepatotoxin, which acts quite
rapidly. Workers at the B. F. Goodrich plant in Louisville who developed angiosarcoma of the liver
were exposed to 1,1 -DCE as well as to vinyl chloride. No information is available at present on
workers exposed only to 1,1-DCE for their working lifetimes. When fasted rats were exposed to 0.02
percent 1,1-DCE, serum alanine-ketogluterate transaminase (AKT) activity was elevated about 50 fold
at 2 hours after the end of a 4-hour inhalation exposure. Exposure of fasted rats to 0.1 percent VC
was without effect on serum AKT.110 There is some preliminary evidence of carcinogenicity in
animals but this has not yet been reported in the literature.
• 1,2 dichloroethylene—No data available.
• Trichloroethylene—Preliminary evaluation of a recent study in mice has shown cancer of the liver and
other organs.1!!
Undesirable Effects 99
-------
CO
u
§UJ
o E
u. O
=> Q
UJ Z
I <
I- uj
Z Q
HI CC
»o
§5
i
Q Q
O UJ
EC t-
0- <
(fi UJ
tM. OC
(O
(0
Ul
- 1-
o JH
tj c
D 0
"D —
o ~
£
3
o
5
tn
Chemical
1
Monomer for vinyl chloride copoly-
mer and polyvinylidine chloride
su'
ii
o' NO
3)
1, 1-dichloroethylen
(vinylidine chloride)
Metal degreasing, manufacturing
solvent
t-
0
csT
r-
O3
^_
r^.
O3
II
0^0
Trichloroethylene
Dry cleaning, metal, degreasing,
chemical intermediate
£
?J
O3
*^
LO
00
CO
o
II
— /° —
Tetrachloroethylene
(perchloroethylene)
Polymer production
8
O5
$
^
00
CM
00
V1
II
0=0
o
B
8
CO
c
Textile sizing, adhesives, paper
coating, polymerization aid
£
o
o5
^~
03
oo
o
0--6 _c
0-0
o
Polyvinyl acetate
Mainly as a refrigerant and aerosol
propellant, Pharmaceuticals and
toiletries
o
O
r-
O3
5~.
co
o
O
o
Chloroform
Manufacture of dichlorodifluoro-
methane (320 million kg) and
trichlorofluoromethane (120
million kg)
o
O
0)
^~^
in
in
O-o-o
o
Carbontetrachloride
100
VINYL/POLYVINYL CHLORIDE
-------
Q ¥*
+-< QJ (J _^
g
en
to
OJ
• —
o
o
0
^1-
^
x-S-x
x-6-i
^
CQ
-------
• Tetrachlorocthylene (perchloroethylene)—Various animal species were exposed for 7 hours a day to
100 to 2500 ppm (256 to 6400 mg/m3) for up to 250 days. No tumors were discovered.1 12 No
long-term followup studies of exposure in man have heen conducted.
• Epichlorohydrin—Considered to have carcinogenic potential. Mutagenicity has been demonstrated in
drosophila neurospora, E. coli, and barley.1 ' 3"' 2 '
• Carbon tetrachloride—Has produced liver tumors in the mouse, hamster, and rat following several
routes of administration including inhalation and oral ingestion. Cases of hepatomas have appeared in
man in several years after carbon tetrachloride poisoning was reported.122
• Chloroform—The carcinogenicity of chloroform has been investigated only in mice in experiments
involving a small number of animals at risk, but among these the frequency of liver tumors was high.
No long-term followup studies in men exposed to chloroform have been reported.122
• 1,2 dibromoethane—A severe irritant. Necrosis of liver and kidney is not conspicuous, although fatal
liver damage has been reported.121 Exposure of bacteria to levels of DBE exceeding 0.04 M resulted
in cellular death accompanied by cell lyses. A DBE concentration of 0.015 M was bacteriostatic for
the first hour and bactericidal for the next 4 hours. Of the metabolism processes tested, RNA
synthesis was the most susceptible to inhibition by DBE. The authors hypothesize that 1,2-DBE
possesses a neurotropic carcinogenic potential.124
• Chlorobromomethane—Has proved to be more toxic than carbon tetrachloride in acute exposures and
less toxic in chronic ones. Both liver and renal injuries have been noted.124
• Chlorodibromopropane—No data available.
• Polyvinyl chloride—Film was implanted in various locations in rats for up to 18 months. Several
tumors were observed, all in the area of the implant.12 s
• 1,1,2 trichloropropene—Rabbits were dosed orally with compound in oil at 0.1 LD^o for 6 months.
There was evidence of changes in lymph nodes after 18 months.12S
• Vinyl alcohol polymer—Implants of polymer sponges at various locations in rats for the life span gave
many sarcomas at the site of implantation and a few tumors at other locations.
• Vinyl chloride acetate copolymer—Implants in rats gave formation of tumors only at the site of
implantation.12S
For the most part, toxicological studies of chemicals related to VC have been limited to acute studies, with
only minor emphasis on long-term or carcinogenic effects. When carcinogenic studies were undertaken, they
involved, in general, exposures that were too short or included too limited a number of animals to provide
conclusive negative results.
6.5.2 Other Carcinogens in Polluted Community Air
The array of contaminants identified in polluted community air includes other chemical and physical
agents—such as polycyclic aromatic hydrocarbons, azaheterocyclic hydrocarbons, certain metal compounds,
asbestos, and certain radionuclides-that are either proven or highly suspect carcinogenic hazards.126'127
However, very few definitive studies have been conducted to determine epidemiologically the contribution
of these ambient pollutants to human carcinogenesis. One of the observed epidemiologic characteristics of
the world-wide increase in lung cancer is the higher incidence in urban residents. Although other
factors—such as population density and occupational differences—may contribute to urban and rural
differences, an urban-rural difference in lung cancer rates persists even after correction for these factors.
102 VINYL/POLYVINYL CHLORIDE
-------
Additional support for a probable etiological role for ambient chemical carcinogens in lung cancer can be
gained from several studies undertaken to measure the effects of population migration on lung cancer
risk.128'129 These studies in migrants have shown that either increases or decreases in lung cancer are
compatible with changes in environment. The changes in rates parallel the general population
concentrations in the areas under study and persist after correction for cigarette smoking, although at a
reduced level. Moves from high pollution to low pollution regions reduced lung cancer death rates and vice
versa (Table 6.27). Within the United States and Great Britain, studies show a gradient of risk to lung
cancer from low in rural to high in urban areas. Migrants from rural to urban areas in the United States
appear to increase their lung cancer rates.
In a recent article, Kotin126'127 draws attention to certain observations on the nature of carcinogens
which should be considered when initiating studies of carcinogenesis associated with air pollutants:
• Cancer induction most frequently requires prolonged periods of exposure to carcinogenic agents.
• Cancer can be caused by several carcinogenic agents acting in combination either in an additive,
synergistic, or inhibitory relation to one another. This is particularly relevant to lung cancer where a
variety of ubiquitous environmental exposures to carcinogenic agents exist.
• The action of a carcinogenic agent in lung cancer induction may be determined by the competency of
the host's defenses at the anatomic, physiological, and biochemical levels. Polluted community air
contains a large variety of chemical and physical irritants, which though unable to cause cancer,
facilitate the action of carcinogenic agents by attenuating or destroying the effectiveness of the
mucociliary apparatus of the lining of the lung. This facilitates deposition and retention of particles
carrying carcniogenic agents. In addition, these irritants can induce changes in the epithelium
(metaplasia) which may enhance the progression of changes to cancer. These irritants may alter the
metabolic handling of carcinogenic agents and thereby enhance their cancer-inducing potency.
• There is evidence that, at the cellular level, environmental chemical cofactors of a highly nonspecific
nature may work together with chemical carginogens to increase their effectiveness.
Vinyl chloride may be viewed as an excellent example of a chemical carcinogen in air. It may be just one of
many such compounds, although the potential carcinogenicity of the others has not yet been identified.
6.6 REFERENCES FOR SECTION 6
1. Dinman, B. D., W. A. Cook, W. M. Whitehouse, H. J. Magnuson, and T. Ditcheck. Occupational
Acroosteolysis; I. An Epidemiological Study. Arch. Environ. Health. 22:61, 1971.
Table 6.27. AGE ADJUSTED DEATH RATES FROM LUNG CANCER
IN GREAT BRITAIN, NORWAY, AND THE UNITED STATES129'131
Population group
Great Britain residents
Great Britain-born U.S. residents
Norway residents
Norway-born U.S. residents
Native U.S. residents
Lung cancer death rate
(per 100,000 persons)
Males
151.2
93.7
30.5
47.5
72.2
Females
19.3
11.5
5.6
10.7
9.8
Undesirable Effects 103
-------
2. Viola, P. L., A. Bigotti, and A. Caputo. Oncogenic Response of Rat Skin, Lungs, and Bones to Vinyl
Chloride. Cancer Res. 31:516-522, 1971.
3. Riches, E. Industrial Cancers. Skin Lesions Caused by Trichloroethylene in a Metal Processing Plant.
Nurs. Mirror. 134:21-25, 1972.
4. Maltoni, C. Occupational Carcinogenesis. In: Proceedings of the Second International Symposium on
Cancer Detection and Prevention, International Congress Series No. 322. (ISBN 9021902281)
Excerpta Medica, Amsterdam. April 9-12, 1973.
5. Maltoni, C. and G. Lefemine. Carcinogenicity Bioassays of Vinyl Chloride; I. Research Plan and Early
Results. Environ. Res. 7.387-405, 1974.
6. Maltoni, C. The Value of Predictive Experimental Carcinogenicity; An Example: Vinyl Chloride.
Ambio. 4:18-23, 1975.
7. Keplinger, M. L., J. W,, Goode, D. E. Gordon, and J. C. Calendra. Interim Results of Exposure of Rats,
Hamsters, and Mice to Vinyl Chloride. Ann. N.Y. Acad. Sci. 246:219-224, January 31, 1975.
8. Creech, J. L., Jr. and M. N. Johnson. Angiosarcoma of the Liver in the Manufacture of Polyvinyl
Chloride. J. Occupational Med. 76:150-151, 1974.
9. Tabershaw, I. R. and W. R. Gaffey. Mortality Study of Workers in the Manufacture of Vinyl Chloride
and Its Polymers. J. Occupational Med. 76:509-516, 1974.
10. Irish, D. D. Halogenated Hydrocarbons; I. Alipractic. In: Industrial Hygiene and Toxicology (2nd
Ed.). Patty, F. A. (ed.). New York, Interscience Publishers, John Wiley and Sons, 1967. Vol. II, p.
1241-1332.
11. Marsteller, H. J., W. K. Lelbach, R. Muller, and P. Gedigk. Unusual Splenomegalic Liver Disease as
Evidenced by Peritoneoscopy and Guided Liver Biopsy among Polyvinyl Chloride Production
Workers. Ann. N.Y. Acad. Sci. 246:95-134, January 31, 1975.
12. Von Oettingen, W. F. The Halogenated Hydrocarbons, Their Toxicity and Potential Dangers. Public
Health Service, U.S. Department of Health, Education, and Welfare. Washington, D.C. PHS
Publication No. 414. 1955.
13. Mastromatteo, E., A. M. Fisher, H. Christie, and H. Danziger. Acute Inhalation Toxicity of Vinyl
Chloride to Laboratory Animals. J. Amer. Ind. Hyg. Assoc. 27:394-397, 1960.
14. Lester, D., L. A. Greenberg, and W. R. Adams. Effects of Single and Repeated Exposures of Humans
and Rats to Vinyl Chloride. Amer. Ind. Hyg. Assoc. J. 24:265-275, 1963.
15. Reynolds, E.S., M.T. Moslen, S. Szabo, R. Jaeger, and S.D. Murphy. Hepatotoxicity of Vinyl Chloride
and 1,1 dichloroethylene; Role of Mixed Function Oxidase System. Arch. Pathol. In press.
16. Schwetz, B. A., B. K. J. Leong, F. A. Smith, M. Balmer, and P. J. Gering. Preliminary Report:
Vinyl Chloride Teratology Study in Mice, Rats, and Rabbits. Toxicology Research Laboratory. Health
and Environmental Research. Midland, Mich. Aug. 19, 1974.
17. Torkelson, T. R., F. Oyen, and V. K. Rowe. The Toxicity of Vinyl Chloride as Determined by
Repeated Exposure to Laboratory Animals. Amer. Ind. Hyg. Assoc. J. 22:354-361, 1961.
18. Cornish, H. H. Problems Posed by Observations of Serum Enzymes Changes in Toxicology. CRC. Crit.
Rev. Toxicol. 7:1-32, 1971.
104 VINYL/POLY VINYL CHLORIDE
-------
19. Grice, H. C., M. L. Barth, H. H. Cornish, G_ V. Foster, and R. H. Gray. Correlation between Serum
Enzymes, Isoenzymes Patterns and Histologically Detectable Organ Damage. Food Cosmet. Toxicol.
9:847-885, 1971.
20. Grice, H. C. The Changing Role of Pathology in Modern Safety Evaluation. CRC Crit. Rev. Toxicol.
7:119-152, 1972.
21. Viola, P. L. Pathology of Vinyl Chloride. Sixteenth International Congress of Occupational Health.
Tokyo, Japan. Abstr. No. 38. 1969.
22. Viola, P. L. Carcinogenic Effect of Vinyl Chloride. The Tenth International Cancer Congress.
Houston, Texas. Abstr. Vol. 29. 1970.
23. Popper, H. and B. Thomas. Alterations of Liver and Spleen among Workers Exposed to Vinyl
Chloride. Ann. N.Y. Acad. Sci. 246:172-194. January 31, 1975.
24. Monson, R. R., J. M. Peters, and M. N. Johnson. Proportional Mortality among Vinyl Chloride
Workers. Lancet. 397-398, August 17, 1974.
25. Maltoni, C. and G. Lefemine. Carcinogenicity Bioassays of Vinyl Chloride: Current Results. Ann.
N.Y. Acad. Sci. 246:195-218, 1975.
26. Maltoni, C., A. Giliberti, L. Gianni, and P. Chico. Occurrence of Angiosarcoma in Rats Following Oral
Administration of Vinyl Chloride: Preliminary Report. Ospedali di Bologna, p. 65-66, 1975.
27. Preliminary Findings from a Study Funded by the Manufacturing Chemists Association. Industrial
Bio-Test Laboratory. Northbrook, Illinois. April 1975.
28. Maltoni, C. Communication sent to the Proceedings on the Proposed Permanent Standards for
Occupational Exposure to Vinyl Chloride. Occupational Safety and Health Administration, U.S.
Department of Labor, Washington, D.C. June 25, 1974.
29. Laham, S. Studies on Placental Transfer; Trichloroethylene. Ind. Med. 39:46-49, 1970.
30. Rannug, U., V. Johansson, K. Ramel, and V. Wachtmeister. Mutagenicity of Vinyl Chloride after
Metabolic Activation. Ambio. 5(5): 194-197, 1974.
31. Malavielle, C., H. Bartsch, A. Bartin, A. M. Camus, and P. Montesano. Mutagenicity of Vinyl Chloride,
Chloroethylene Oxide, Chloracetaldehyde, and Chloroethanol. Biochem. Biophys. Res. Comm.
63:363-369, 1975.
32. Bartsch, H., C. Malaveille, P. Montesano, and L. Tomatis. Tissue-Mediated Mutagenicity of Vinylidine
Chloride and 2-Chlorobutadiene in Salmonella Typhimurium. Nature. 225:641-643, 1975.
33. Infante, P. F. Oncogenic and Mutagenic Risk in Communities with Polyvinyl Chloride Production
Facilities. Presented at N.Y. Acad. Sci. Conf. on Occupational Carcinogenicity. March 24-27, 1975.
34. Committee 17 Report, Council on Environmental Mutagen Society. Environmental Mutagens Hazard
Science. 757:503-514, 1975.
35. Hefner, R. E., Jr., P. G. Watanabe, and P. J. Gering. Preliminary Studies on the Fate of Inhaled Vinyl
Chloride Monomer (VCM) in Rats. Ann. N.Y. Acad. Sci. 246:135-148. January 31, 1975.
36. Gering, P. J. Toxicology Research Laboratory, Health and Environmental Research, Dow Chemical
Co., Midland, Mich. Personal Communication. 1974.
Undesirable Effects 105
-------
37. Knittle, J. L., P. Fontanares, B. B. Aubrey, S. Daum, and L J. Selikoff. Vinyl Chloride Content of
Subcutaneous Adipose Tissue from Vinyl Chloride Polymerization Workers. Draft. Mount Sinai
School of Medicine and New York City Department of Air Resources. July 1975.
38. Van Duuren, B. L. Carcinogenic Epoxides, Lactones and Halo-ethers and Their Mode of Action. In:
Biological Effects of Alkylating Agents. Ann. N.Y. Acad. Sci. 163:633-651, 1969.
39. Grover, P. L. and P. Sims. Interactions of the K-Region Epoxides of Phenanthrene and Dibenz [a,h]
Anthracene with Nucleic Acid and Histone. Biochem. Pharmacol. 79:2251-2259, 1970.
40. Grover, P. L. and P. Sims. Enzyme-Catalyzed Reactions of Polycyclic Hydrocarbons and
Deoxyribonucleic Acid and Proteins In Vitro. Biochem. J. 7/0:159-160, November 1968.
41. Grover, P. L., J. A. Forrester, and P. Sims. Reactivity of the K-Region Epoxides of Some Polycyclic
Hydrocarbons Towards the Nucleic Acids and Proteins of BHK 21 Cells. Biochem. Pharmacol.
20:1297-1302, June 1971.
42. Arcos, J. C., M. F. Argus, and G. Wolf. Chemical Induction of Cancer (Vol. 1). New York, Academic
Press, 1968.
43. Volkheimer, G. Hematogeneous Dissemination of Ingested PVC Microparticles. Presented to the
Working Group on the Toxicity of Vinyl Chloride-Polyvinyl Chloride, the New York Academy of
Sciences, New York. May 10-11, 1974.
44. Polakoff, P. L., L. Lapp, and R. Reger. Polyvinyl Chloride Pyrolysis Products. Arch. Environ. Health.
JO:269-271, June 1975.
45. Sokol, W. N., Y. Aelony, and G.N. Beall. Meat Wrappers' Asthma, A New Syndrome? J. Amer. Med.
Ass. 22(5:639-641, Nov. 5, 1973.
46. Jaeger, R, J. and R. A. Hites. Pyrolytic Evaporation of a Plasticizer from Polyvinyl Chloride Meat
Wrapping Film. Bull. Environ. Contam. Tbx. 77:45-48, Jan. 1974.
47. Van Houten, R. W., A. L. Cudworth, and C. H. Irvine. Evaluation and Reduction of Air Contaminants
Produced by Thermal Cutting and Sealing of PVC Packaging Film. Amer. Indus. Hyg. Ass. J.
4:218-222, April 1974.
48. Wilson, R. H. and W. E. McCormick. Plastics, the Toxicity of Synthetic Resins. A.M.A. Arch. Ind.
Health. 27:536, 1960.
49. Zapp. J. A., Jr. Toxic and Health Effects of Plastics and Resins. Arch. Environ. Health. 4:125, 1962.
50. Malten, K. E. and Zielhuis. Industrial Toxicology and Dermatology in the Production and Processing
of Plastics. New York, Elsevier Publishing Co., 1964.
51. Encyclopedia of Occupational Health and Safety. International Labour Office; Geneva, Switz. Vol. II,
p. 1467, 1922.
52. Baretta, E. D., R. D. Stewardt, and J. E. Mutchler. Monitoring Exposures to Vinyl Chloride Vapor:
Breath Analysis and Continuous Air Sampling. Amer. Ind. Hyg. Assoc. J. 30:531, 1964.
53. Kramer, C. G. and J. E. Mutchler. The Correlation of Clinical and Environmental Measurements for
Workers Exposed to Vinyl Chloride. Amer. Ind. Hyg. Assoc. J. 33:19-30, 1972.
106 VINYL/POLYVINYL CHLORIDE
-------
54. Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment
with Intended Changes for 1972. American Conference of Governmental and Industrial Hygienists.
1972.
55. Emergency Temporary Standard for Occupational Exposure to Vinyl Chloride. Occupational Safety
and Health Administration, U.S. Department of Labor. Fed. Reg. 39, 12342. April 5, 1974.
56. Proposed Standard: Vinyl Chloride. Occupational Safety and Health Administration, U.S. Depart-
ment of Labor. Fed. Reg. 39, 16896. May 10, 1974.
57. Heath, C. W., H. Falk, and J. L. Creech. Characteristics of Cases of Angiosarcoma of the Liver among
Vinyl Chloride Workers in the United States. Ann. N.Y. Acad. Sci. 246:231-236, January 31, 1975.
58. Wagoner, J. K. National Institute of Occupational Safety and Health. Presented before the
Subcommittee on the Environment. Commerce Committee, United States Senate. Washington, D.C.
August 21, 1974.
59. Data from National Institute of Occupational Safety and Health to Kenneth Bridbord, Special Studies
Staff, U,,S. Environmental Protection Agency. January 31, 1975.
60. Popper, H. and L. B. Thomas. Alterations of Liver and Spleen among Workers Exposed to Vinyl
Chloride and Pathology of Angiosarcoma of the Liver among Vinyl Chloride-Polyvinyl Chloride
Workers. Ann. N. Y. Acad. Sci. 246:172-194, 268-277, January 31, 1975.
61. Biggar, R. Memo for the Record: Report of an Angiosarcoma Case. County Health Department,
Rochester, N.Y. April 9, 1974.
62. Falk, Henry. Center for Disease Control. Letter with enclosed autopsy report to Kenneth Bridbord,
U.S. Environmental Protection Agency. August 14, 1974.
63. Thomas, Louis B. National Cancer Institute. Letter to Philip Landrigan. Center for Disease Control.
July 31, 1974.
64. Epidemiologic Study of Vinyl Chloride Workers. Prepared by Tabershaw/Cooper Associates, Inc.,
Berkeley, California. Final Report, submitted to Manufacturing Chemists Association, Washington,
D.C. May 3, 1974.
65. Holder, B. Dow Chemical Company. Testimony presented at Public Hearing on Proposed Standard
for Occupational Exposure to Vinyl Chloride. U.S. Department of Labor, Washington, D.C. June 25,
1974.
66. Ott, M. G., R. R. Langner, and B. B. Holder. Vinyl Chloride Exposure in a Controlled Industrial
Environment. Arch. Environ. Health. JO:333-339, July 1975.
67. Monson, R. R., J. M. Peters, and M. Johnson. Mortality Among Vinyl Chloride Workers. Presented at
National Institute of Environmental Health Sciences Conference, Pinehurst, N.C. July 29-31, 1974.
68. Nicholson, W. J., E. C. Hammond, H. Seidman, and I. J. Selikoff. Mortality Experience of a Cohort of
Vinyl Chloride Polyvinyl Chloride Workers. Ann. N.Y. Acad. Sci. 246:225-230. January 31, 1975.
69. Daniel, Roger L. Dow Chemical Company. Testimony presented at Public Hearing—Proposed
Standard for Occupational Exposure to Vinyl Chloride. U.S. Department of Labor, Washington, D.C.
June 25, 1974.
Undesirable Effects 107
-------
70. Dernehl, Carl U. Associate Medical Director. Union Carbide Corporation. Testimony presented at
Public Hearing—Proposed Standard for Occupational Exposure to Vinyl Chloride. U.S. Department of
Labor, Washington, D.C. June 25, 1974.
71. Wagoner, Joseph. National Institute of Occupational Safety and Health.
72. Vinyl Chloride and Congenital Malformations. Morbidity and Mortality. 24:1-2, July 25, 1975.
73. Ziegler, G. Organization of Insurance Carriers in Switzerland. Letter to Dr. B. List of the Lonza Co.
February 27, 1975.
74. Lester, D., L. A. Greenberg, and W. R. Adams. Effects of Single and Repeated Exposures of Humans
and Rats to Vinyl Chloride. Amer. Indus. Hyg. Ass. J., 24:265-275. May-June 1963.
75. Harrison's Principles of Internal Medicine. M. M. Wintrobe, G. W. Thorn, R, D. Adams, I. L. Bennett,
E. Braunwald, K. J. Isselbacher, and R. G. Petersdorf (ed.). New York, McGraw-Hill, Vol. 2, p.
1524-1525.
76. Juhe, S., C. B. Lange, G. Stein, and G. Veltman. On the So-Called Vinyl Chloride Disease. Deut. Medi.
Wochenschr. (Stuttgart). 95:2034-2037, 1973.
77. Marsteller, H. J. Chronic Toxic Liver Damage in Workers Engaged in PVC Production. Deut. Medi.
Wochenschr. (Stuttgart). 95:2311-2314, 1973.
78. Gedigk, P., R. Muller, and H. Bechtelsheimer. Morphology of Liver Damage among Polyvinyl Chloride
Production Workers: A Report of 51 Cases. Ann. N.Y. Acad. Sci. 246:278-285, January 31, 1975.
79. Veltman, G., C. E. Lange, S. Juhe, and V. Bachner. Clinical Manifestations and Cause of Vinyl
Chloride Disease. Ann. N.Y. Acad. Sci. 24(5:6-17, January 31, 1975.
80. Lange, C. E., S. Juhe, G. Stein, and G. Veltman. Further Results in Polyvinyl Chloride Production
Workers. Ann. N.Y. Acad. Sci. 24618-21, January 31,1975.
8L Creech, J. L. and L. Makk. Liver Disease among Polyvinyl Chloride Production Workers. Ann. N.Y.
Acad. Sci. 246:88-94 .January 31, 1975.
82. Miller, A., A_ S. Teirstein, M. Chuang, R. Warshaw, and I. D. Selikoff. Changes in Pulmonary
Function in Workers Exposed to Vinyl Chloride and Polyvinyl Chloride. Ann. N.Y. Acad. Sci.
246:42-52, January 31, 1975.
83. Lilis, R., H. Anderson, W. J. Nicholson, S. Dawn, A. S. Fischbein, and I. J. Selikoff. Prevalence of
Disease among Vinyl Chloride and Polyvinyl Chloride Workers. Ann. N.Y. Acad. Sci. 246:22-41,
January 31, 1975.
84. Cook, R. R. The Dow Chemical Company. Testimony presented at Public Hearing on Proposed
Standard for Occupational Exposure to Vinyl Chloride. U.S. Department of Labor, Washington, D.C.
June 25, 1974.
85. Kotin, P. Consultant to Air Products and Chemicals, Inc. Testimony presented at Public Hearing on
Proposed Standard for Occupational Exposure to Vinyl Chloride. U.S. Department of Labor,
Washington, D.C. June 25, 1974.
86. Adams, A. R. Chairman, Health Subcommittee, Air Products and Chemicals, Inc. Memo to Dr. John
Knelson, Director, Health Effects Research Laboratory, U.S. Environmental Protection Agency,
Research Triangle Park, N.C. May 23, 1975.
108 VINYL/POLY VINYL CHLORIDE
-------
87. Marcus, W. L. Senior Pharmacologist, Office of Toxic Substances, U.S. Environmental Protection
Agency. Memo to Dr. F. G. Hueter, Director, Special Studies Staff, U.S. Environmental Protection
Agency, Research Triangle Park, N.C. June 26, 1975.
88. Report of a Working Group on Epidemiological Studies on Vinyl Chloride Exposed People.
International Agency for Research on Cancer. IARC Internal Technical Report No. 75/001. Lyon,
France. January 8-9, 1975.
89. Stanier, R. Y., M. Doudoroff, and E. A. Adelberg. The Microbial World (3rd Ed.). Englewood Cliffs,
N.J., Prentice Hall, 1970. p. 873.
90. Zobell, C. E. Assimilation of Hydrocarbons by Microorganisms. Adv. in Enzymol. X:443-486, 1950.
91. Traxler, R, W. and W. L. Flannery. Mechanisms of Hydrocarbon Degradation. In: Biodeterioration of
Materials. A. H. Walters and B. J. Elplver (ed.). London, Elsevier Pub. Co. Ltd. p. 44-54.
92. McKenna, E. J. and R. E. Kelbio. The Biology of Hydrocarbons. Ann. Rev. Microbiol. 79:183-208,
1965.
93. Haselhoff, E. and G. Lindau. Chlorine and Hydrochloric Acid. In: Damage to Vegetation by Fumes:
Handbook for the Identification and Assessment of Fume Damage (German). 1903. p. 230-256.
94. Kaiser, E. R. and A. A. Carotti. Municipal Incineration of Refuse with 2% and 4% Additions of Four
Plastics. Society of the Plastics Industry. New York, N.Y. June 30, 1971.
95. Weiler, A. Corrosive Damage on Foliage Organs Caused by Acids and Tarry Substances. Phytopath.
7:121-144, 1934.
96. Hindawi, I. J. Injury by Sulfur Dioxide, Hydrogen Fluoride, and Chlorine as Observed and Reflected
on Vegetation in the Field. J. Air Pollut. Contr. Ass. 7^:307-312, 1968.
97. Wood, F. A. The Influence of Smoke from the Combustion of Polyvinyl Chloride Insulation on
Northern Hardwood Forest Species. Phytopathol. 58:1073, 1968.
98. Antipov, V. G. Resistance of Perennials to Gases. Sadovodstvo (Horticulture)../: 1-2, 1956. (Russian).
99. Bohne, H. Problems of Determining the Effects of Gaseous Chlorine Emission Upon Plants.
Staub-Reinholt. Luft. 29:41-43, 1969.
100. Means, W. E., Jr. and N. L. Lacasse. Relative Sensitivity of Twelve Tree Species to Hydrogen Chloride
Gas. Phytopathol. 59(4):401, 1969.
101. Heck, W. W. and E. G. Pires. Growth of Plants Fumigated with Saturated and Unsaturated
Hydrocarbon Gases and Their Derivatives. Texas Agricultural Experimental Station, Agricultural and
Mechanical College of Texas. Publication No. MP603. 1962. p. 12.
102= Abeles, F. Ethylene in Plant Biology. New York, Academic Press, 1973.
103. Abeles, F. B0, L. E. Croker, L. E. Forrence, and G, R. Leather. Fate of Air Pollutants: Removal of
Ethylene, Sulfur Dioxide, and Nitrogen Dioxide by Soil. Science. 775:914-916, 1971.
104. Jernelov, A., R. Rosenberg, and S. Jensen. Biological Effects and Physical Properties in the Marine
Environment of Aliphatic Chlorinated By-products from Vinyl Chloride Production. In: Water
Research. New York, Pergamon Press, 1972. p. 1181-1191.
Undesirable Effects 109
-------
105. Rosenberg, R. Effects of Chlorinated Aliphatic Hydrocarbons on Larval and Juvenile Balanus
balanoides L. Environ. Pollut. 5:313-318, 1972.
106. Hagstrom, A. and S. Normark. Toxic Effects and Action of Chlorinated By-products from Vinyl
Chloride Production on Escherichia coliKU. Ambio. 5:77-79, 1974.
107. Chemical Economics Handbook. Chemical Information Services, Stanford Research Institute, Menlo
Park, California. 1967 (And Additions).
108. Synthetic Organic Chemicals, United States Production and Sales, 1969. U.S. Tariff Commission.
Washington, D.C. Publication 412. 1971. p. 206.
109. Patty, F. Industrial Hygiene and Toxicology (Vol. II, 2nd Ed.). Interscience Publishers, p. 1284.
110. Jaeger, R. Vinyl Chloride Monomer; Comments on Its Hepatotoxicity and Interaction with 1,1
Dichloroethylene. Ann. N.Y. Acad. Sci. 246:150-151, 1975.
111. Saffiotti, U. Internal memorandum of alert. Issued by National Cancer Institute.
112. Hartwell, J. L. Survey of Compounds Which Have Been Tested for Carcinogenic Activity (2nd Ed.).
U.S. Public Health Service, Bethesda, Md. Publication Number (NIH) 73-35 or PHS-149. Reprinted
1969.
113. Fishbein, L., W. Flamm, and H. Falk. Chemical Mutagens Environmental Effects on Biological
Systems. New York, Academic Press, p. 205.
114. Deering, Millik Research Corporation, British Patent 855,547 (1960), Chem. Abstr. 55, P27884d
(1961).
115. Hercules Powder Co. Paper of Increased Porosity and Absorbency. British Patent 871,205. June 21,
1961. Chem. Abstr. 55 P27884d.
116. Michel, P. Waterproof Coating for Building Materials. Belgium Patent 555,772. March 20, 1957.
Chem. Abstr. 53, P19348H.
117. Smyth, H. F., Jr., Jane Seaton, and Louis Fiocher. The Single Dose Toxicity of Some Glycols and
Derivatives. J. Ind. Hyg. Toxicol. 25:259-268, 1941.
118. Carpenter, C. P., H. F. Smyth, and V. C. Pozzani. The Assay of Acute Vapor Toxicity and the
Grading and Interpretation of Results on 96 Chemical Compounds. J. Ind. Hyg. Toxicol. 57:343-346,
1949.
119. Schultz, C. Deut. Med. Worschr. (Stuttgart). 89:1342, 1964.
120. Food Cosmet. Toxicol. 2:240, 1964.
121. Rapaport, I. A. Dokl. Akad. Nauk. (Moscow). SSR 60. 469, 1948.
122. Evaluation of Carcinogenic Risk of Chemicals to Man. International Agency for Research on Cancer.
L :53-61, 1972.
123. Gleason, M., R. Gasselen, H. Hodge, and R. Smith. Clinical Toxicology of Commercial Products (3rd
Ed.). Baltimore, The Williams and Wilkins Co., 1969. p. 48.
110 VINYL/POLYVINYL CHLORIDE
-------
124. Brem, H., J. Coward, H. Rosenkrantz. 1, 2 Dibromoethane-Effect on the Metabolism and Ultra
Structure of Eshericia Coli. In: Biochemical Pharmacology. Great Britain, Pergamon Press, 1974. Vol.
23, p. 2345-2347.
125. Hartwell, J. L. Survey of Compounds Which Have Been Tested for Carcinogenic Activity (2nd Ed.).
U.S. Public Health Service. Bethesda, Md. Publication Number (NIH) 73-35 or PHS149. Reprinted
1963.
126. Kotin, P. and H. L. Falk. The Role and Action of Environmental Agents in the Pathogenesis of Lung
Cancer. Cancer. 72:147-163, 1959.
127. Kotin, P. and H. L. Falk. Atmospheric Factors in Pathogenesis of Lung Cancer., Advances in Cancer
Research. 7:475-514, 1963.
128. Haenszel, W. Cancer Mortality Among the Foreign-born in the U.S. J. Nat. Cancer Inst. 26:37-132,
1961.
129. Haenszel, W., S. C. Marcus, and G. G. Zimmerer. Cancer Morbidity in Urban and Rural Iowa. U.S.
Department of Health, Education, and Welfare, Washington, D.C. Public Health Monograph 37,
Public Health Service Publication 462. 1956, p. 85.
130. Kotin, P. Mutagenic and Carcinogenic Problems Associated with Air Pollutants. Proceedings of the
Conference on Health Effects of Air Pollutants. GPO Serial No. 9395. November 1973. p. 603-617.
131. Reid, D. C., J. Cornfield, R. D. Markush, D. Seigel, E. Pedersen, and W. Haenszel. Studies of Disease
among Migrants and Native Populations in Great Britain, Norway and the United States: III
Prevalence of Cardiorespiratory Symptoms among Migrants and Native-born in the U.S. Nat. Cancer
Inst. Monog. 79:321-346, 1966.
Undesirable Effects 111
-------
-------
7. CONTROL TECHNOLOGY AND REMEDIAL ACTIONS
7.1 INTRODUCTION
Most of the presently used or available technologies for controlling VC emissions are a basic part of the
processing system and serve to recover reactant or product. These controls are appraised herein either by
using performance data from selected VC and PVC manufacturing plants, or by comparing emission levels
for plants with and without controls. The controls so appraised for VC production include: recycling of
vent streams, condensation with refrigeration, adsorption to carbon, incineration, absorption (scrubbing),
and venting to flares. Monomer loading and unloading involves special additional controls: vapor collection
adapters with recycling, thermal level detectors with recycling, and magnetic gauges. Polymer production
can possibly benefit from application of controls indicated for the monomer production in addition to
vacuum stripping, steam stripping, and the recovery of the stripped monomer.
A qualitative assessment of the potential applications of selected controls has been made based upon
information presented by a few U.S. industrial firms.1 The results of this assessment are summarized for
each process in the following paragraphs. All percent reductions of emissions are estimates.
7.2 MONOMER PRODUCTION
The available data on emissions from monomer production seem to point to a present total emission of
about 0.45 kg per 100 kg of VC produced. To this amount should be added a smaller, intermittent loss of
VC in the loading area.
A reduction in emissions can be obtained by refrigeration and/or absorption of vinyl chloride in the vents
by appropriate solvents (ethylene dichloride, EDC, for instance), or by combustion of the organics in vent
streams, followed by removal of the HC1 produced. Ninety-nine percent control represents the
approximate maximum reduction available for industrial point sources with present-day (1975) technology.
To ensure such a reduction, vent losses in the loading area must be controlled.
7.2.1 VC from Acetylene and Hydrogen Chloride (HC1)
This route for producing VC is based on acetylene. A second route based on ethylene will be discussed
later. Vinyl chloride from acetylene is the older technology and suffers an economic penalty. As of the
summer of 1975, no producers were known to be operating an acetylene-based plant.
The reactor vent is the main emissions source for the acetylene and HC1 process, accounting for 60 percent
of total emissions. Condensation at 4.4°C (40°F) and 0.24 X 106 N/m2 (35 psig) is now used. Addition of
refrigeration would decrease emissions by about 50 percent. If an HC1 scrubber were also used, the
combined controls should achieve 85 percent reduction. With carbon adsorption, emissions might be
reduced by 99 percent. Recycling does not appear to be applicable.
Fugitive emissions and tank car loading, unloading, and accidents account for about 25 percent of total
emissions from this process. Use of diaphragm valves, replacement of packed pump seals with pressurized
mechanical seals, use of vapor collectors on samplers, and preventive maintenance can be expected to
reduce these emissions by 50 to 95 percent.
113
-------
Incineration, with HC1 recovery, should reduce condenser vent losses 99 percent. Thermal-level detectors
combined with vent gas refrigeration and/or recycling would reduce slip gauge emissions by 95 percent.
Replacing the slip gauge with a magnetic gauge could reduce the emissions by nearly 100 percent.
Vapor-collector adapters with recycling would reduce purge losses 50 to 90 percent. Incineration should
reduce loading air emissions about 90 percent.
7.2.2 VC by Ethylene-based Technology
Ethylene-based technology consists of basically two processes which combine to produce VC from chlorine,
ethylene, and air (oxygen). In the direct process, the EDC is made from ethylene and chlorine; in the
second process, oxychlorination is used for making EDC from C22~, HC1, and oxygen. The EDC
intermediate product is cracked to VC during dehydrochlorination, and the by-product HC1 stream is
recycled to the oxychlorination reactor. The combination of processes is commonly referred to as the
balanced process.
Major emission sources of VC in this process are from the EDC light-ends column vent (9 to 20 percent),
the EDC heavy-ends tar removal column vent (18 percent), the VCM light-ends column vent (10 to 13
percent), tank car loading (10 to 20 percent), and oxychlorination reactor vent (6 to 10 percent). The
distillation vents could be controlled by incineration or condensation with recycle.
EDC light-ends column emissions could be reduced about 50 percent by using refrigerated condensers and
nearly 100 percent with carbon adsorption. The adsorbed organics would then have to be disposed of;
incineration is one means of disposal. Recycling to a post chlorination unit would be almost 100 percent
effective.
Heavy-ends column emissions are believed to be controllable by incineration (90 percent reduction), and by
adsorption (close to 100 percent reduction); again, the adsorbed organics would have to be disposed of.
VC light-ends column vent emissions would require adsorption or incinceration, either of which is capable
of nearly 100 percent reduction.
The tank car loading controls given in Section 7.2.1 would apply here also.
The oxychlorination reactor vent emissions on some processes could be reduced by using additional
chlorination. If oxygen is fed instead of air, the fluidized bed systems will produce a vent stream that is
combustible without supplemental fuel. Several experimental processes are in development stages using vent
gas recycle with oxygen feed and catalytic incineration.
7.3 POLYMER PRODUCTION
In the production of polyvinyl chloride, present monomer losses in kilograms per kilogram of product are
at least an order of magnitude higher than in the production of VC. Most producers report about a 3 to 4
percent lower PVC production than monomer intake. From some data submitted by manufacturers, 0.01 to
0.3 percent of PVC made is lost. A portion of this is emitted as fine particulate to the atmosphere. The
actual monomer emission is therefore in the order of 3-to 3.7 percent. These losses result from the batch
nature of the polymerization operation and from the drying of the polymer, if practiced. Reduction of
these losses poses a more difficult problem than those encountered in VC plants. A reduction to 80
percent of the present level of losses seems possible, but a 95 percent reduction of the total emissions in
some of the existing polymer plants without process changes might be beyond present techniques at
acceptable costs. However, if intensive stripping of the suspension at the end of the reaction is allowable,
the 95 percent reduction might be feasible.
The move to progressively bigger reaction vessels should be noted. One company has studied this
development and is proposing the use of a 454 m3 (120,000 gal) polymerization reactor as compared to the
114 VINYL/POLY VINYL CHLORIDE
-------
present typical size reactor of 19 to 38 m3 (5000 to 10,000 gal). A possible emergency blowing of such a
reactor might, however, lead to very high peak values of vinyl chloride emissions.
This report has not considered the influence of VC remaining in the polymer. Polymer is now being offered
for sale with VC residual content of less than 10 ppm (10 mg/kg). The residual monomer is mostly released
during further processing and might thus create emission problems during fabrication processes, particularly
those involving heat. EPA is now sponsoring a study to determine the emissions from fabrication processes.
7.3.1 Suspension Polymerization
Fugitive emissions throughout the suspension polymerization process account for an estimated 45 percent
of VC emissions. However, a good maintenance program and minor equipment modifications should reduce
fugitive emissions by about 50 percent.
Vacuum stripping of the crude product would be 95 percent efficient for reducing emissions from sources
downstream of the stripper. Carbon adsorption, incineration, or absorption could reduce emissions by the
same amount; condensation with refrigeration, 50 to 70 percent.
Collectively, vents from the dryer, the air conveyor, the storage site, and wash water provide 35 percent of
the total emissions. Vacuum stripping and absorption are expected to give 95 percent reduction; and
recycles to compressors, 40 to 60 percent reduction in emissions from these sources.
7.3.2 Emulsion Polymerization
The dryer vent, air conveyor vent, site storage vent, and waste-water vent appear to account collectively for
about 85 percent of total emissions from emulsion polymerization. Carbon adsorption and steam stripping
could reduce these emissions by 50 to 95 percent. Fugitive losses contribute 17 percent of the emissions;
blend surge tank vents contribute another 6 percent. Vacuum stripping, if practiced, would effect 90 to 95
percent reduction; carbon adsorption, 50 to 95 percent. Condensation with refrigeration would reduce
either source about 40 to 60 percent. Absorption is expected to reduce both losses 50 to 80 percent.
Preventive maintenance would reduce fugitive losses 25 to 50 percent.
7.3.3 Bulk Polymerization
During bulk polymerization, the VC reactor vents (25 percent), fugitive emissions (35 percent), and the
combined resin receiver, collector, and storage (20 percent) are the major emission sources. For the reactor
vent, adsorption (50 to 90 percent reduction) and incineration (50 to 90 percent reduction) are indicated
for control purposes. Intensive maintenance is believed to be capable of giving 50 to 75 percent reduction
of the diverse and ill-defined fugitive emissions. The product-collection systems vents could be
water-washed, then adsorbed (50 to 90 percent reduction) or incinerated (50 to 90 percent reduction).
7.3.4 Solution Polymerization
While limited data are at present available for this process, it is expected to have the emission characteristics
of the suspension process (Section 7.3.1) and to respond roughly to the same controls.
7.4 REFERENCE FOR SECTION 7
1. Vinyl Chloride—Assessment of Emission Control Techniques and Costs. U.S. Environmental Protection
Agency, Washington, D.C. Publication No. EPA-650/2-74-097. September 1974. 84 p.
Control Technology and Remedial Actions 115
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/6-75-004
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Scientific and Technical Assessment Report on Vinyl
Chloride and Polyvinyl Chloride
5. REPORT DATE
December 1975
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Research and Development
Environmental Research Center
Research Triangle Park, N. C. 27711
10. PROGRAM ELEMENT NO.
1AA001
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Vinyl chloride (VC) is a chemical of widespread industrial and commercial use. Occupa-
tional experience and experimental evidence strongly indicate that it is a carcinogen. Addi-
tionally, there is experimental evidence that indicates that it may be a teratogen and muta-
gen. An increased incidence of liver angiosarcoma, excessive liver damage, and acrooste-
olysis has been reported among VC workers, and the frequency and severity of the liver
pathology is related to the length of exposure. The principal route of exposure is thought to
be air inhalation. Sources of increased importance for the general population include food
and water. Tumors at multiple and diverse sites have been observed in all species of experi
mental animals tested for carcinogenicity by inhalation and ingestion of VC . An excess inci-
dence of liver angiosarcoma was observed among VC/PVC (polyvinyl chloride) workers and
reproduced in experimental animals with very similar pathology. Liver angiosarcoma was
observed in two species of experimental animals after inhalation exposures of VC at the
lowest doses tested, 50 ppm (128,000 yg/m3), and after ingestion at 16 mg/kg. In addition
to the health effects of VC, this document also considres the sources, distribution, and
control technology. Emissions of VC from VC/PVC plants are estimated to exceed 100 million
kilograms annually, about 90 percent of which is from PVC plants. Installation of currently
available controls may be adequate to reduce VC emissions from VC/PVC plants in the order
of 90 percent.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Vinyl chloride
Polyvinyl chloride
Carcinogen
Angiosarcoma
Exposure
Toxicology
Air pollution
Health effects
Environmental pollution
Environmental distribution
07C,11I 06E
07C,11I
06E
06E
06J
06T
13B
13. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
130
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
116
------- |