6EFK
United States
Environmental Protection
Agency
Office of Health and
Environmental Assessment
Washington DC 20460
EPA-600/8-82-007
November 1982
Revised
Research and Development
Health Assessment
Document for
Acrylonitrile
REVISED
DRAFT
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Environmental Criteria and Assessment Office (MD-52)
:: November 15, 1982
:T Revised Draft of EPA OHEA Health Assessment
Document for Acrylonitrile „ n U ~?
i £ ^^^j^^^^ ffi /f ^^^^^^^^^^
Lester D. Grant, Ph.D., Director X^W^• -^ta'"^
Environmental Criteria and Assessment Office (MD-52)
Dr. Roger McClellan, Chairman
Health Committee, EPA Science Advisory Board (SAB)
Attached is a revised draft of the EPA (OHEA) Health Assessment
Document for Acrylom'trile, earlier issued as an External Review Draft
(EPA-600/8-82-007; March, 1982) for public comment and SAB review.
The attached (November, 1982) draft of the acrylonitrile document
incorporates revisions which take into account both: (1) public comments
received on the March, 1982, External Review Draft; and (2) recommendations
provided by the SAB Health Committee under your chairmanship in connection
the Committee's peer-review of the March External Review Draft at the
August 2-3, 1982, Health Committee public meeting in Washington.
The main revisions which have been incorporated into the attached
November, 1982, draft include the following:
1. The listing of Authors and Reviewers (on pp. xiii-xvii) has been
revised to indicate clearly the individuals/groups responsible for
authorship of specific chapters or sections of the document.
2. The Summary and Conclusions (Chapter 1) materials have been revised
to reflect more accurately (and consistently) key information and con-
clusions contained in later, detailed chapters of the document. Particu-
lar emphasis is placed in the Summary and Conclusions Chapter on the
discussion of key findings and conclusions regarding studies of muta-
genic and carcinogenic effects associated with acrylonitrile. This
reflects the focus of major attention at this time on the evaluation of
such effects as likely being of pivotal importance in regard to resolution
of whether or not serious health effects are associated with chronic
low-level exposure to acrylonitrile in the ambient environment. There
have also been added to the Summary and Conclusions discussion of
carcinogenicity aspects certain important caveats regarding the evaluation
of acrylonitrile as a possible human carcinogen and the estimation by
the EPA Cancer Assessment Group (CAG) of unit risk values representing
the plausible upper bound of potential carcinogenic risk associated with
acrylonitrile exposure. Specific factors, e.g., the retrospective
estimation of exposure levels for occupationally-exposed workers studied
by O'Berg (1980), contributing to uncertainties associated with the unit
risk estimates are now more clearly noted.
-fj IK-v. 3 761
-------
3. The air quality chapters addressing "Sources In the Environment"
(Chapter 5) and "Environmental Levels and Exposure" (Chapter 7) have
been updated to include the most recent information available with
respect to sources, emission estimates, and exposure estimates based on
a comparison between monitoring data and dispersion modeling data. The
references principally used in updating these chapters were those by
R. V. Crume (1982) and B. E. Suta (1982).
4. In addition to specific air quality chapters being revised and/or
updated there were considerable changes in the Chapter 13 section (13.5)
addressing "Carcinogenicity." With regard to the epidemiologic aspects
of carcinogenicity (section 13.5.2), discussions of several newly available
studies were added, which included both published and unpublished studies.
The new studies added (unpublished studies are marked with an asterisk)
include:
1. Gaffey and Strauss (1981)*
2. Kisselbach, et al. (1980)*
3. Thiess, et al. (1980)
4. Herman (1981)*
5. Stallard (1982)*
6. Weinch and Carter (1981)
In regard to the quantitative aspects of carcinogencity (Section
13.5.3) new information was added addressing the procedures and methodology
(Section 15.3.0) involved in the determination of unit risks and the
actual calculations of unit risk for acrylonitrile based on different
animal data sets and one human study (Section 13.5.3.2). The procedures
and methodology section attempted to more clearly define the philosophy
of doing risk assessment, the basis for using the'linear non-threshold
multistage model, the mathematical formulations describing the latter
model and the selection of data sets from several studies to use in the
model. There was a brief discussion of alternate methodological approaches
that have been considered by the Carcinogen Assessment Group and a
relative comparison of the carcinogenic potency of acrylonitrile to
other carcinogens based on unit risk calculations.
Lastly, the CAG appendix was revised by the Carcinogen Assessment
Group and inserted into the main text of the document as Section 13.5
(pp 13-87 through 13-172), instead of appearing as an appendix to the
overall document.
5. Time limitations precluded listing at the end of each chapter the
references for that given chapter in the attached revised document
draft. Rather all references presently appear at the end of the document
in the collated bibliography. However, this will be changed in the
final document so that references for individual chapters will appear at
the end of each chapter as appropriate.
-------
We feel that the above major revisions and other changes appearing
in the attached draft document adequately address the major substantive
issues raised by public comments and/or the SAB Health Committee in the
course of its August, 1982, review of the March, 1982 External Review
Draft of the EPA OHEA Health Assessment Document for Acrylonitrile. We
look forward to SAB review of this revised draft at the upcoming December
8-9, 1982, SAB public meeting in Washington and wish to thank you for
the Committee's excellent and helpful recommendations concerning revision
of the earlier draft of the document.
Attachment
cc: Health Committee members
-------
EPA-600/8-82-007
Revised Draft
November 1982
DRAFT
Do not cite or quote
HEALTH ASSESSMENT DOCUMENT FOR
ACRYLONITRILE
NOTICE
This document is a preliminary draft. It
has not been formally released by EPA and
should not at this stage be construed to
represent Agency policy. It is being
circulated for comment on its technical
accuracy and policy implications.
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
Research Triangle Park, North Carolina 27711
Project Coordinator: Dr. Robert Bruce
November 12, 1982
-------
DISCLAIMER
The report is an external draft for review purposes only and does not
constitute Agency Policy. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
11
-------
PREFACE
The Office of Health and Environmental Assessment has prepared this
health assessment to serve as a "source document" for EPA use. The health
assessment document was originally developed for use by the Office of Air
Quality Planning and Standards to support decision-making regarding possible
regulation of acrylonitrile as a hazardous air pollutant. However, the scope
of this document has since been expanded to address multimedia aspects.
In the development of the assessment document, the scientific literature
has been inventoried, key studies have been evaluated and summary/conclusions
have been prepared so that the chemical's toxicity and related characteristics
are qualitatively identified. Observed effect levels and other measures of
dose-response relationships are discussed, where appropriate, so that the
nature of the adverse health responses are placed in perspective with observed
environmental levels.
111
-------
TABLE OF CONTENTS
Page
LIST OF TABLES vii
LIST OF FIGURES xii
1. SUMMARY AND CONCLUSIONS 1-1
2. INTRODUCTION 2-1
3. PHYSICAL AND CHEMICAL PROPERTIES 3-1
3.1. SYNONYMS AND TRADE NAMES 3-1
3.2. STRUCTURAL AND MOLECULAR FORMULAS AND MOLECULAR WEIGHT 3-1
3-1
3-1
3-1
3-1
3-1
3-2
3-2
3-2
3-2
3-2
3-3
3-3
3-3
3-3
3-3
3-3
3-4
3-4
3-5
3-5
3-6
3-6
SAMPLING AND ANALYTICAL METHODS 4-1
4.1. AIR 4-1
4.1.1. Sampling 4-1
4.1.2. Analysis 4-5
4.1.3. Conclusions 4-9
4.2. WATER 4-10
4.2.1. Sampling 4-10
4.2.2. Analysis 4-11
4.2.3. Conclusions 4-14
4.3. WASTEWATER 4-14
4.3.1. Sampling 4-14
4.3.2. Analysis 4-14
4.3.3. Conclusions 4-16
3.3.
3.4.
3.5.
3.6.
3.7.
BOND ANGLES AND BOND DISTANCES
PHYSICAL PROPERTIES
3.4.1. Description
3.4.2. Boiling Point
3.4.3. Melting Point
3.4.4. Density
3.4.5. Refractive Index
3.4.6. Spectroscopic Data
3.4.7. Solubility
3.4.8. Volatility in Water ....
3.4.9. Volatility
3.4.10. Stability
3.4.11. Octanol-Water Partition
3.4. 12. Conversion Factor
CHEMICAL PROPERTIES
3.5.1. Reactivity
3.5.2. Polymerization
3.5.3. Reaction at the Nitrile
3.5.4. Reactions at the Double
3.5.5. Cyanoethylation Reactior
CHARACTERISTICS OF THE CHEMICAL
CONCLUSION
Coefficient
Groups
Bond
is
PRODUCT
-------
TABLE OF CONTENTS (cont.)
4.4. SOIL AND SEDIMENT 4-16
4.4.1. Sampling 4-16
4.4.2. Analysis 4-16
4.4.3. Conclusions 4-18
4.5. RESIDUE IN POLYMERS AND THE EXTENT OF MONOMER
MIGRATION OF FOOD SIMULATING SOLVENTS 4-18
4.5.1. Analysis 4-19
4.5.2. Conclusions 4-23
4.6. OTHER MEDIA 4-23
4.7. GENERAL METHODS FOR THE ANALYSIS OF ACRYLONITRILE 4-23
5. SOURCES IN THE ENVIRONMENT 5-1
5.1. PRODUCTION PROCESSES 5-1
.5.2. ACRYLONITRILE PRODUCERS 5-1
5.3. ACRYLONITRILE USES 5-2
5.4. CONSUMPTION OF ACRYLONITRILE BY PRODUCT 5-4
5.5. SOURCES OF EMISSIONS 5-8
5.5.1. Monomer and Polymer Production 5-8
5.5.2. Emissions During Transportation 5-12
5.5.3. Emissions from End-Product Usage 5-12
5.5.4. Conclusions 5-15
6. ENVIRONMENTAL FATE, TRANSPORT, AND DISTRIBUTION 6-1
6.1. ATMOSPHERIC FATE, PERSISTENCE, AND TRANSPORT 6-1
6.1.1. Atmospheric Chemical Reactions 6-1
6.1.2. Photochemical Reactions 6-2
6.1.3. Atmospheric Persistence and Transport 6-2
6.2. FATE, PERSISTENCE, TRANSPORT, AND BIOACCUMULATION IN
AQUEOUS MEDIA 6-3
6.2.1. Chemical Reactivity in Water 6-3
6.2.2. Photochemical Reaction in Water 6-4
6.2.3. Degradation of Acrylonotrile by Microorganisms 6-4
6.2.4. Bioaccumulation in Water 6-7
6.2.5. Transport in Water 6-8
6.3. FATE, PERSISTENCE, AND TRANSPORT IN SOIL 6-10
7. ENVIRONMENTAL LEVELS AND EXPOSURE 7-1
7.1. ENVIRONMENTAL LEVELS 7-1
7.1.1. Atmospheric Levels of Acrylonitrile Around Its
Major Production and Usage Facilities 7-1
7.1.2. Acrylonitrile Levels in Surface Waters 7-5
7.2. ACRYLONITRILE LEVELS IN SOILS AND SEDIMENTS 7-7
7.3. ENVIRONMENTAL EXPOSURE 7-7
7.3.1. Exposure From Air Polluted by Industrial Sources ... 7-9
7.3.2. Exposure From Drinking Water 7-9
7.3.3. Exposure From Foods 7-9
7.3.4. Exposure From Spillage During Transportation 7-14
7.3.5. Exposure From Thermal Degradation 7-15
-------
TABLE OF CONTENTS (cont.)
Page
7.4 CONCLUSIONS ................................................ 7-17
8 . BIOLOGICAL EFFECTS ON MICROORGANISMS ............................ 8-1
9 . BIOLOGICAL EFFECTS ON PLANTS .................................... 9-1
10. BIOLOGICAL EFFECTS ON DOMESTIC ANIMALS .......................... 10-1
11. BIOLOGICAL EFFECTS ON WILDLIFE .................................. 11-1
11.1. Insects ................................................. 11-1
12. BIOLOGICAL EFFECTS ON AQUATIC ORGANISMS ......................... 12-1
12.1. ACUTE TOXICITY ............................................ 12-1
12.1.1. Freshwater Fish .................................. 12-1
12.1.2. Marine Fish ...................................... 12-4
12.1.3. Freshwater Invertebrates ......................... 12-5
12.1.4. Marine Invertebrates ............................. 12-5
12.2. SUBCHRONIC TOXICITY ....................................... 12-6
12.2.1. Freshwater Fish .................................. 12-6
12.2.2. Freshwater Invertebrates ......................... 12-7
12.3. SUMMARY AND CONCLUSIONS ................................... 12-8
13. BIOLOGICAL EFFECTS IN MAN AND EXPERIMENTAL ANIMALS .............. 13-1
13.1. PHARMACOKINETICS .......................................... 13-1
13.1.1. Absorption and Distribution ...................... 13-1
13.1.2. Metabolism ....................................... 13-7
13.1.2.3. Summary and conclusions ............... 13-16
13.2. ACUTE, SUBCHRONIC, AND CHRONIC TOXICITY ................... 13-17
13.2.1. Acute Toxicity ................................... 13-17
13.2.2. Subchronic Toxicity in Non-Human Mammals ......... 13-43
13.2.3. Chronic Toxicity in Non-Human Mammals ............ 13-45
13.2.4. Summary and Conclusions .......................... 13-50
13.3. TERATOGENI CITY AND REPRODUCTIVE TOXICITY . .................. 13-51
13.3.1. Summary and Conclusions .......................... 13-57
13.4. MUTAGENICITY .............................................. 13-60
13.4.1. Gene Mutation Studies ............................ 13-60
13.4.2. Chromosomal Aberration Studies ................... 13-68
13.4.3. Other Tests Indicative of Genetic Damage ......... 13-71
13.4.4. Summary and Conclusions .......................... 13-75
13.5. CARCINOGENICITY ........................................... 13-87
13.5.1. Animal Studies ................................... 13-87
13.5.2. Epidemiologic Studies ............................ 13-116
13.5.3. Quantitative Estimation .......................... 13-136
13.5.3.1. Procedures and Methodology ............ 13-136
13.5.3.2. Acrylonitrile Risk Estimates .......... 13-151
13.5.4. Summary .......................................... 13-164
13.5.5. Conclusions ...................................... 13-172
REFERENCES
VI
-------
LIST OF TABLES
Table
No. Page
4-1 Various Sorbents and Trapping Media for Collection
of Acrylonitrile in Air 4-3
4-2 Recovery of Acrylonitrile from Various Solvents 4-7
4-3 Direct Analysis of Acrylonitrile 4-8
4-4 Analyses of Acrylonitrile in Water 4-15
4-5 Analysis of Acrylonitrile in Wastewaters 4-17
4-6 Analysis of Acrylonitrile Residue in Polymers and
Food-Simulating Solvents 4-20
5-1 Producers of Acrylonitrile in the United States 5-2
5-2 Distribution of Acrylonitrile in 1980 and Projected
Growth Through 1984 5-2
5-3 Primary Uses for Compounds Synthesized from
Acrylonitrile-Containing Compounds 5-3
5-4 Acrylonitrile Consumption and Project Growth of
Products Using Acrylonitrile 5-4
5-5 Producers of SAN and ABS Resins 5-6
5-6 Producers of Acrylic and Modacrylic Fibers 5-6
5-7 Producers of Nitrile Rubbers and Elastomers 5-7
5-8 Producers of Acrylamide 5-7
5-9 Estimated Atmospheric Emissions of Acrylonitrile from
Monomer Production Facilities 5-9
5-10 Estimated Acrylonitrile Emission Rates from ABS-SAN
Resin Production 5-10
5-11 Estimated Acrylonitrile Emission Rates from Acrylic
Fiber Production 5-10
5-12 Estimated Acrylonitrile Emission Rates from Nitrile
Elastomer Production 5-11
VII
-------
LIST OF TABLES (cont.)
Table
No.
5-13 Estimated Acrylonitrile Emission Rate from Adiponitrile
Production 5-11
5-14 Hazards of Acrylonitrile Transportation 5-11
5-15 Monomer Residue in End-Products of Acrylonitrile 5-14
7-1 Atmospheric Monitoring Data for Acrylonitrile 7-2
7-2 Comparison of Monitoring and Dispersion Modeling Data 7-4
7-3 Acrylonitrile Monitoring Data for Surface Waters 7-6
7-4 Acrylonitrile Monitoring Data for Sediments 7-8
7-5 Acrylonitrile Monitoring Data for Soils 7-8
7-6 Estimates of Population Exposures to Atmospheric
Acrylonitrile from Specific Emission Source
Categories 7-10
7-7 Acrylonitrile Migration Under Different Storage
Conditions 7-12
7-8 Amounts of Various Acrylonitrile Copolymers Used in
Food-Contact Applications 7-13
11-1 Lethal Dose Values for Insects Exposed to Acrylonitrile
Fumigation 11-2
12-1 Median Lethal Concentration (LC50) Values for Fish
Exposed to Acrylonitrile 12-2
13-1 Recovery of Radioactivity from Rats Given Single Oral
Doses of 0.1 or 10 mg/kg 14C-Acrylonitrile 13-2
13-2 Recovery of Radioactivity from Rats Exposed by
Inhalation to 5 or 100 ppm l4C-Acrylonitrile for
6 hours 13-3
13-3 Distribution of Radioactivity in Selected Tissues of
Rats Given 14C-Acrylonitrile 13-5
Vlll
-------
LIST OF TABLES (cont.)
Table
No. Page
13-4 Urinary Metabolites Following the Oral Administration
of C-l (Cyano) Labeled Acrylonitrile 13-14
13-5 Metabolites of 14C-Acrylonitrile Separated from
Various Fluids of Rats by High Pressure Liquid 13-18
13-6 Summary of Results of Exposures of Rats to
Acrylonitrile 13-22
13-7 Minimal Lethal Concentration of Acrylonitrile During
Four-Hour Exposure 13-24
13-8 Comparison of the Effects of Acrylonitrile and of
Hydrocyanic Acid on Various Species of Animals 13-26
13-9 Cyanide and Thiocyanate in Blood of Animals Exposed
to Acrylonitrile 13-28
13-10 Effect of Methemoglobinemia on Mortality Ratios in
Albino Rats Poisoned with Acrylonitrile, Potassium
Cyanide, and Acetone Cyanohydrin 13-33
13-11 Therapeutic Effect of SH and S-S Compounds on Acute
Acrylonitrile Poisoning 13-38
13-12 Concentration of Protein (PBSH) and Nonprotein (NPSH) SH
Groups in Normal and Acrylonitrile-Intoxicated
Animals 13-39
13-13 Cumulative Mortality of Male and Female Rats Maintained
for Two Years on Drinking Water Containing
Acrylonitrile 13-49
13-14 Significant Changes Considered to be Secondary to
Ingestion of Acrylonitrile 13-52
13-15 Incidence of Fetal Malformation Among Litters of Rats
Given Acrylonitrile 13-54
13-16 Pup Weight on Days 4 and 21 of Lactation 13-58
13-17 Tissues Examined for Histopathologic Changes in the F3b
Litter 13-59
-------
LIST OF TABLES (cent.)
Table
No. Page
13-18 Mutagenicity Tests of Acrylonitrile 13-78
13-19 Cumulative Mortality Data of Male Rats Maintained
for 2 Years on Drinking Water Containing
Acrylonitrile 13-89
13-20 Cumulative Mortality Data of Female Rats Maintained
for 2 Years on Drinking Water Containing
Acrylonitrile 13-90
13-21 Histopathologic Diagnoses and Tumor Incidences
in Male Rats Maintained for 2 Years on Drinking
Water Containing Acrylonitrile 13-91
13-22 Histopathologic Diagnoses and Tumor Incidences in
Female Rats Maintained for 2 Years on Drinking
Water Containing Acrylonitrile 13-92
13-23 Tumor Incidences in Sprague-Dawley Rats Fed
Acrylonitrile in Drinking Water 13-96
13-24 Tumor Incidence in Fischer 344 Rats Fed
Acrylonitrile in Drinking Water 13-98
13-25 Incidence of Tumors Observed in Rats During a
Three-Generation Reproductive Study 13-100
13-26 Tumor Incidence in Rats Fed Acrylonitrile Orally
by Intubation 13-103
13-27 Tumor Incidence by Rats Following Inhalation
of Acrylonitrile 13-104
13-28 Stability and Trace Impurity Analysis of the
Acrylonitrile Liquid Test 13-106
13-29 Chamber Concentrations of Acrylonitrile Vapors 13-107
13-30 Cumulative Mortality Data of Male Rats Exposed
by Inhalation for 2 Years to Acrylonitrile Vapors ... 13-108
13-31 Cumulative Mortality Data of Female Rats Exposed
by Inhalation for 2 Years to Acrylonitrile Vapors ... 13-109
13-32 Tumor Incidence in Sprague-Dawley Rats Exposed
to Acrylonitrile by Inhalation 13-110
-------
LIST OF TABLES (cont.)
Table
No. Page
13-33 Enhancement of SA7 Transformation by Treatment of
HEC with ACN 13-113
13-34 Transformation of HEC by ACN 13-115
13-35 Number (Percentage) of Rats Developing Tumors in
At Least One of the Following Target Organs:
Zymbal Gland, Tongue, Stomach, Brain, and
Spinal Cord by Dose and by Sex 13-155
13-36 Relative Carcinogenic Potencies Among Suspect
Carcinogens Evaluated by the Carcinogen
Assessment Group 13-161
13-37 Carcinogenicity of Acrylonitrile in Rats 13-165
13-38 Epidemiologic Studies Reviewed in Acrylonitrile
Risk Assessment 13-168
XI
-------
LIST OF FIGURES
Figure
No. Page
5-1 Flow Diagram for Acrylonitrile 5-5
6-1 Biological Oxidation of Acrylonitrile in Aqueous Systems ... 6-5
13-1 Proposed Pathways for Acrylonitrile Biotransformation 13-10
13-2 Proposed Scheme for Metabolism of Acrylonitrile by
the Rat 13-12
13-3 Distribution of Acrylonitrile, Cyanide, and Thiocyanate
in the Blood after a Single Injection of
Acrylonitrile 13-31
13-4 Effect of Sodium Thiosulfate on the Distribution of
Acrylonitrile, Cyanide, and Thiosulfate 13-32
13-5 Distribution of Acrylonitrile, Cyanide, and Thiocyanate
in the Blood after a Single Injection of
of Acrylonitrile 13-35
13-6 Effect of E-Cysteine on the Blood Concentrations of
Acrylonitrile, Cyanide, and Thiocyanate 13-36
13-7 Histogram Representing Frequency Distribution of the
Potency Indices of 54 Suspect Carcinogens Evaluated
by the Carcinogen Assessment Group 13-160
XII
-------
The EPA office of Health and Environmental Assessment (OHEA) is respon-
sible for the preparation of this health assessment document. The OHEA Envi-
ronmental Criteria and Assessment Office (ECAO/RTP) had overall responsibility
for coordination and direction of the document, preparation and production
effort (Dr. Robert M. Bruce, Project Manager). The chapters addressing physi-
cal and chemical properties, sampling and analysis, air quality and toxicity
data were written by Syracuse Research Corporation. The principal authors of
these chapters are listed below.
Dr. Dipak K. Basu
Life and Environmental Sciences Division
Syracuse Research Corporation
Syracuse, New York
Dr. Robert S. Hsu
Life and Environmental Sciences Division
Syracuse Research Corporation
Syracuse, New York
Dr. Michael W. Neal
Life and Environmental Sciences Division
Syracuse Research Corporation
Syracuse, New York
Mr. Joseph Santodonato
Life and Environmental Sciences Division
Syracuse Research Corporation
Syracuse, New York
Dr. Richard H. Sugatt
Life and Environmental Sciences Division
Syracuse Research Corporation
Syracuse, New York
The OHEA Carcinogen Asssessment Group (CAG) was responsible for prepara-
tion of the sections on carcinogenicity. Participating members of the CAG are
listed below (principal authors of present carcinogenicity materials are
designated by *).
Xlll
-------
Roy Albert, M.D. (Chairman)
Elizabeth L. Anderson, Ph.D.
Larry D. Anderson, Ph.D.
Steven Bayard, Ph.D.*
David L. Bayliss, M.S.*
Chao W. Chen, Ph.D.
Herman J. Gibb, M.S., M.P.H.
Bernard H. Haberman, D.V.M., M.S.
Charalingayya B. Hiremath, Ph.D.*
Robert McGaughy, Ph.D.
Dharm V. Singh, D.V.M., Ph.D.
Todd W. Thorslund, Sc.D
The OHEA Reproductive Effects Assessment Group (REAG) was responsible for
the preparation of sections on mutagenicity. Participating members of REAG
are listed below (principal authors of present mutagenicity sections are
indicated by *).
Vicki Vaughn-Dellarco, Ph.D.*
Jack R. Fowle III, Ph.D.
K.S. Lavappa, Ph.D.
Sheila Rosenthal, Ph.D.
Carol Sakai, Ph.D.
Peter Voytek, Ph.D.
xiv
-------
The following individuals provided peer-review of this draft or earlier
drafts of this document:
U. S. Environmental Protection Agency
Donald Barnes, Ph.D.
Office of Toxic Substances
David Berg
Office of Environmental
Engineering and Technology
H. Matthew Bills
Acting Director for Monitoring Systems
and Quality Assurance
Office of Research and Development
Robert M. Bruce, Ph.D.
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
Lester D. Grant, Ph.D.
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
Gary E. Hatch, Ph.D.
Office of Health Research
U.S. EPA
Richard N. Hill, M.D., Ph.D.
Office of Toxic Substances
U.S. EPA
Raelyn Janssen
Office of Health Research
U.S. EPA
Steven Nesnow, Ph.D.
Office of Health Research
U.S. EPA
Joseph Padgett
Office of Air Quality Planning and Standards
U.S. EPA
Nancy Pate, D.V.M.
Office of Air Quality Planning and Standards
U.S. EPA
Shabeg Sandhu, Ph.D.
Office of Health Research
U.S. EPA
Jerry F. Stara, D.V.M.
Office of Health and Environmental Assessment
U.S. EPA
xv
-------
John Todhunter, Ph.D.
Assistant Administrator
Office of Toxic Substances
U.S. EPA
Herbert Wiser, Ph.D.
Acting Director
Office of Environmental
Engineering and Technology
U.S. EPA
Other Agencies
William C. Brumley, Ph.D.
Bureau of Foods
Food and Drug Administration
Health and Human Services
Michael T. Flood, Ph.D.
Bureau of Foods
Food and Drug Administration
Health and Human Services
Chiu A. Linn, Ph.D.
Toxicology Division
Food and Drug Administration
Human Health Services
T. P. McNeal, Ph.D.
Bureau of Foods
Food and Drug Administration
Health and Human Services
Terry Troxell, Ph.D.
Division of Food and Color Additives
Food and Drug Administration
Health and Human Services
Consultants and/or Reviewers
Julian B. Andelman, Ph.D.
Professor of Chemistry
Graduate School of Public Health
University of Pittsburgh
Pittsburgh, Pennsylvania
Rudolph J. Jaeger, Ph.D.
Consulting Toxicologist
7 Bogert Place
Westwood, New Jersey
George Hoffman, Ph.D.
Professor, Biology Department
College of Holy Cross
Worchester, Maine
xvi
-------
Edmond J. LaVoie, Ph.D.
Head, Metabolic Biochemistry Section
Naylor Dana Institute for Disease Prevention
American Health Foundation
Valhalla, New York
Richard R. Monson, M.D., Sc.D.
Associate Professor of Epidemiology
School of Public Health
Harvard University
Daniel Straus, Ph.D.
Associate Professor
Biomedical Science and Biology
University of California
Riverside, California
Gary Williams, M.D.
Associate Director
Naylor Dana Institute for Disease Prevention
American Health Foundation
Valhalla, New York
xvil
-------
1. SUMMARY AND CONCLUSIONS
Acrylonitrile is a clear, colorless, and highly flammable liquid that has an
unpleasant and irritating odor (Fassett, 1963). The boiling point of acrylo-
nitrile is 77.3CC; the melting point is -83.55°C, and the density of the liquid
at 20°C is 0.8060 (Groet, 1978). Acrylonitrile is soluble in water between 7.2
and 9.1 weight % at temperatures of 0°C and 60°C, respectively. The open cup
flash point of acrylonitrile is 0°C, and the explosive limits are between 3.0 and
17$ by volume in air at 25°C (Steere, 1968). Synonyms for acrylonitrile include
2-propenenitrile, cyanoethylene, and vinyl cyanide. Acrylonitrile has a
molecular weight of 53-06 and a molecular formula of C-,H,N. The structural
formula is given below.
HxC=C-C=N
H' L~n .
H
Acrylonitrile monomer production capacity in the United States is approxi-
mately 1,128,000 million grams. Of the 862,000 million grams of acrylonitrile
produced in 1980, approximately 77$ (661,000 Mg) will be used domestically; the
remainder is exported (CMR, 1980). Acrylonitrile is used primarily as a raw
material in the synthesis of acrylic and modacrylic fibers, acrylonitrile-
butadiene-styrene (ABS) and sytrene-acrylonitrile (SAN) resins, nitrile rubbers,
adiponitrile, acrylamide, and barrier resins (Suta, 1979) • A small percentage of
the acrylonitrile produced is used as a chemical intermediate.
Acrylonitrile is emitted to the atmosphere during monomer production, poly-
mer production, transport, and end-product usage; however, the major sources of
acrylonitrile emissions are monomer and polymer production facilities. Of the
estimated total of 3,856 Mg emission per year, monomer ABS-SAN resin, and arylic
fiber production facilities emit 802 Mg, 1H2H Mg, and 1276 Mg, respectively of
1-1
-------
acrylonitrile in the atmosphere per year. The atmospheric half-life of
acrylonitrile has been estimated to be between 9 and 10 hours (Suta, 1979), which
is long enough to allow transport of acrylonitrile from emission sources to
nearby populations. In natural waters, acrylonitrile may be removed by either
chemical decomposition or microbial degradation (Going et al., 1979; Mills and
Stack, 1955). Evaporation may also lower the concentration of acrylonitrile in
water; the calculated half-life for acrylonitrile by evaporation alone from
water 1 m deep is 795 minutes (Dilling, 1977). The percent of acrylonitrile
removal through each of the processes will depend on the characteristics of the
aquatic system. For example, in biologically active treatment ponds, less than
0.1$ of the acrylonitrile was found to be removed from the aquatic media due to
volatilization while most of the acrylonitrile removal took place via
biodegradation.
The acrylonitrile levels in the vicinity of acrylonitrile production and
polymer manufacturing plants were investigated by Hughes and Horn (1977) and
Going et al. (1979). In general, arylonitrile levels around user plants may be
greater than around producing plants. Acrylonitrile was detected in the air at
distances up to 5 km from a user facility; however, the concentrations of
acrylonitrile were dependent on meteorological conditions and the production
stage within the plant at the time of sampling. No acrylonitrile was detected in
the soil near these plants. Variable low levels of acrylonitrile were generally
detected in the water downstream from the plants, except for high levels of 35 to
1300 ug/1 detected in some samples near wastewater discharge points.
Acrylonitrile has also been detected in drinking water, although the levels were
not quantified (Kopfler £t al., 1976). The inhalation exposure of acrylonitrile
in the vicinity of a plant site estimated by dispersion modelling does not agree
well with the experimental monitoring data obtained from the same site (Going et^
1-2
-------
al., 1979; Suta, 1979). There are insufficient data with which to determine the
human intake of acrylonitrile through food and drinking water.
Limited data suggest that both aerobic and anaerobic microorganisms are
capable of degrading acrylonitrile, especially acclimated microorganisms.
Certain isolated bacteria can tolerate 10,000 ppm acrylonitrile and use it as a
sole source of nitrogen. In natural water, a concentration of 50 ppm may inhibit
aerobic microbial degradation of acrylonitrile. The breakdown products of
aerobic microorganisms may include ammonia and organics, followed by nitri-
fication of ammonia.
Acrylonitrile has been shown to affect some terrestrial and aquatic plants
at exposure concentrations of 9 to 100 mg/1, Acrylonitrile is toxic to aquatic
animals at exposure concentrations in the low milligrams per liter range. The
reported acute LC50 values for fish ranged between 10.1 and 70 mg/1. Subchronic
exposure of fish for 30 to 100 days resulted in LC50 values of about 2 mg/1, with
no evidence that a threshold concentration had been reached. Although the only
tested invertebrate, Daphnia magna, had the lowest acute LC50 value (7.6 mg/1),
this species was not adversely affected by chronic exposure to 3.6 mg/1 through-
out its whole life cycle.
Use of acrylonitrile as a fumigant has shown that the vapor concentrations
required to kill 95% or more of many species of pest insects is between about 1
and 10 mg/1. No other information was found concerning the effects of acrylo-
•
nitrile on wildlife.
Acrylonitrile is readily absorbed in animals following ingestion or inhala-
tion, while dermal absorption is poor and occurs at about 1f of that of the
lungs. Following absorption of radiolabeled acrylonitrile, the radioactivity
disappears in a biphasic manner, with a half-life for the first phase of 3.5 to
3.8 hours and the second phase of 50 to 77 hours (Young et al_., 1977). The
1-3
-------
predominant route of elimination is through the urine. The routes of elimination
are dose-related with the percent eliminated through the urine less for small
doses as compared to larger doses, while the relative amount retained by the
carcass is greater for the small dose as compared to a larger dose. Acrylo-
nitrile is metabolized to cyanide, which is transformed to thiocyanate and by
cyanoethylation of sulfhydryl groups to S-(2-cyanoethyl)cysteine, followed by
elimination of these metabolites in the urine. Other minor metabolites are
formed from acrylonitrile. The toxicity of acrylonitrile is caused by both the
acrylonitrile molecule itself and its metabolites.
Acrylonitrile intoxication in humans results in irritation of the eyes and
nose, weakness, labored breathing, dizziness, impaired judgement, cyanosis,
3
nausea, and convulsions. The TLV of acrylonitrile is 45 mg/m for humans.
Acrylonitrile also causes severe burns to the skin. In experimental animals,
there is considerable species variation in susceptibility to acrylonitrile intox-
ication; the guniea pig is the most resistant and the dog is the most sensitive.
In animals, effects of intoxication include respiratory changes, cyanosis, convul-
sions, and death. In rats, the LD50 for acrylonitrile is between 80 and 113
mg/kg (Knoblock et al., 1971; Smyth et al., 1969). There is some evidence that
acrylonitrile produces abnormal function of both the peripheral and central
nervous systems and that acrylonitrile causes damage to the adrenals. With
subchronic exposure of animals to acrylonitrile, some signs of functional dis-
orders of the liver and kidney are observed. Chronic exposure of dogs and rats
results in unthrifty appearance, weight loss, and early death. Some of these
signs may be related to low food and water consumption resulting from the un-
pleasant taste of acrylonitrile in the water. Pathological changes in the rats
believed to be treatment related included hyperplasia and hyperkeratosis of the
1-4
-------
squamous epithelium of the nonglandular portion of the stomach, proliferation
of glial cells in the brain, and mammary gland hyperplasia in females.
Acrylonitrile adversely affected pup survival following exposure of
pregnant rats and, in one study, produced teratogenic events. In a three-
generation study in which rats were exposed to 500 ppm acrylonitrile in the
drinking water, there was reduced pup survival in the first generation (Beliles
et al., 1980). This was a maternal effect inasmuch as fostering the pups on
untreated dams eliminated the poor survival. Reproductive capacity was un-
changed in the other generations, and the offspring showed no adverse effects
on development. Similarly, rats exposed by inhalation to 40 or 80 ppm of
acrylonitrile for 6 hours a day on days 6 to 15 of gestation had no statisti-
cally significant changes in reproductive success or fetal development (Murray
et al. , 1978). Only the pups of rats administered acrylonitrile per ££ (65
mg/kg) for days 6 to 15 of gestation had an increase in malformations (Murray
et al., 1976). This increase was in total malformations, with no statistically
significant increase occurring in any single malformation. It was concluded
that these fetal abnormalities were the result of acrylonitrile and not the
result of toxicity in the dams. Although several studies have been conducted
to evaluate the ability of acrylonitrile to cause adverse tetratogenic, embryo-
toxic, and reproductive effects, the limitations of the available data do not
allow for a full assessment of these effects. However, there are data that
indicate that acrylonitrile has the potential to cause fetal malformations in
rats, in a non dose-related manner.
There is evidence that acrylonitrile as an epoxide metabolite causes
point mutations in bacterial test systems (Vemitt et al. 1977, De Meester et
al. 1978, and Kier 1982) and there is suggestive evidence that acrylonitrile
produces a positive response in the sex-linked recessive lethal mutation assay
1-5
-------
in Drosophila Melanogaster (Benes and Sram 1969). Chromosomal damage was not
detected in plants or whole animals when treated with acrylonitrile. In vitro
DNA binding studies indicate that acrylonitrile in the presence of a rat liver
activation system has the potential to react with DNA as well as its epoxide
metabolite, 2-cyanoethylene oxide (Guengerich et al., 1981). Acrylonitrile
also induces sister chromated exchange in CHO cells but requires metabolic
activation (Ved Brat and Williams, 1982). From the data base available it
appears that acrylonitrile may have the potential to bind to DNA and cause
genetic damage. If the pharmacokinetics of this chemical substance in humans
results in metabolic products that can intereact with DNA, it may cause so-
matic mutations in humans.
Acrylonitrile is not a direct acting carcinogen and hence the localiza-
tion and nature of the effects depend on its metabolism. It is probable that
the proximal carcinogen is 2-cyanoethylene oxide, since it has been demon-
strated as a reaction product with calf thymus DNA. However, the metabolite
has not been tested directly for its carcinogenicity. It has been shown to be
produced in the liver and possibly circulates to other organs. However,
studies have not been done to determine where else in the body this metabolite
is produced. There appears to be a clear difference between animals and
humans in their tumorigenic response to acrylonitrile: no lung tumors have
been produced in animals and no brain tumors have been observed in humans.
There are no human studies on the metabolism of acrylonitrile and there are no
pharmacokinetic studies that would be relevant to the characterization of
dose-response relationships at low levels of exposure.
The carcinogenicity of acrylonitrile has been studied in seven cancer
bioassays in rats: four in drinking water, one by gastric intubation, and two
by inhalation. In addition, ten epidemiologic studies of cancer incidence
have been reported. A short description of these studies is presented below.
1-6
-------
Quast et al. (1980a) administered acrylonitrile in drinking water to
Sprague-Dawley rats for 2 years at dose levels of 35, 100, and 300 ppm. A
statistically significant incidence of tumors was observed in the central
nervous system, Zymbal gland, stomach, tongue, and small intestine in both
male and female rats, as well as in the mammary gland of female rats. The
occurrence of central nervous system and Zymbal gland tumors in Sprague-Dawley
rats was further confirmed in four other studies: a three-generation repro-
duction study performed at Litton-Bionetics by Beliles et al. (1980); three
studies by Biodynamics, Inc. (1980a, b, c) in which acrylonitrile was adminis-
tered in drinking water and via gastric intubation; and an inhalation study by
Quast et al. (1980b).
A second inhalation study by Maltoni et al. (1977) exposed rats to atmos-
phere containing 5, 10, 20, and 40 ppm acrylonitrile 4 hours per day, 5 days
per week, for 12 months. Marginal increases in tumors of the mammary gland in
females and the forestomach in males were observed, although the sensitivity
of this test was limited by the relatively low dose levels and the short
duration of exposure.
Ten epidemiologic studies of the association between acrylonitrile expo-
sure and cancer incidence have been reported: five published [Monson (1981),
O'Berg (1980), Thiess et al. (1980), Werner and Carter (1981), Delzell and
Monson (1982] and five unpublished [Gaffey and Strauss (1981), Herman (1981),
Kiesselbach et al. (1980), Stallard (1982), and Zack (1980]. Six of these
studies present no evidence of carcinogenic risk from exposure to acryloni-
trile. However, all suffer from problems in the design or methodology, i.e.,
small cohort size, insufficient characterization of exposure, short follow-up,
and relatively youthful cohort. Because of these probelms none of these
studies can be cited as adequate evidence that acrylonitrile is not carcino-
genic.
1-7
-------
Data presented in the remaining four epidemiologic studies consistently
demonstrate a statistically significant risk of lung cancer in various sub-
groups of the populations studied. All four have problems with the method-
ology, definition, and/or size of the population, whether or not exposure to
other carcinogens occurred, and short follow-up intervals. In three of the
four studies [Delzell and Monson (1982), Thiess et al. (1981), Werner and
Carter (1981)], the problems were sufficient to cast doubt on the finding of
significantly elevated risks of lung cancer reported in each study. In the
fourth study by O'Berg, the problems were insufficient to obscure the signif-
icant finding of lung cancer. After adjusting for latent factors and evaluat-
ing the contribution due to smoking, the finding of a statistically signifi-
cantly elevated risk of lung cancer remained. Thus, one study appears adquate
and three are suggestive, while the remaining six are inadequate to address
the issue of a risk of lung cancer.
In addition to lung cancer, two other findings of concern are the signi-
ficantly elevated risk of lymph system cancer found in the Thiess et al. study
(4 observed versus 1.38 expected, P <0.05)and the significantly elevated risk
of stomach cancer found in the Werner and Carter study (5 observed versus 1.9
expected, P < 0.05). These findings provide additional suggestive evidence of
the carcinogenicity of acrylonitrile.
This level of animal evidence would be regarded as "sufficient" evidence
of carcinogenicity according to the International Agency for Research on
Cancer (IARC) classification scheme. The human evidence for the carcinogeni-
city of acrylonitrile would be regarded as somewhere between "sufficient" and
"limited", using the IARC classification. Therefore, in combining the human
and animal evidence, acrylonitrile would be placed in group 2A, which IARC
characterizes as "probably carcinogenic in humans, where the evidence for
human carcinogenicity is almost sufficient".
1-8
-------
To provide a rough estimate of the potency of acrylonitrile relative to
other chemicals and a crude indication of population risks associated with
known exposures, unit risks estimates have been calculated. The unit cancer
risk for air is defined as the lifetime cancer risks occurring if an individual
3
is exposed to air containing 1 (Jg/m continuously for a lifetime. The linear
non-threshold dose extrapolation model has been used to give a rough but
plausible upper-bound of risk; that is, the true risk is not likely to be
higher than the estimate but could be lower. For a discussion of the limi-
tations and uncertainties of the procedure the reader is referred to Section
13.5.3.1.
Three unit risk estimates for air are calculated; one based on a human
occupational study (O'Berg 190a, b) and two based on rat cancer bioassays
(Quast et al. 1980a, b). The upper-bound lifetime risk of cancer associated
with a lifetime inhalation exposure of ± (Jg/rn is 6.8 x 10 from the human
study and 1.5 x 10 from the rat inhalation study. The value based on the
rat drinking water study is 1.2 x 10 (or 2 x 10 if the equivalent human
dose is assumed to be mg/kg/day rather than surface area) but this study is
less reliable because of the inappropriate route of exposure.
The estimate based on the human study is uncertain because of the rela-
tively weak documentation of the available exposure estimates of the acrylon-
itrile workers. The air concentration had not been measured when the workers
experienced their heaviest exposure and was estimated 12 years after the end
of the exposure period. However, in spite of these difficulties, the esti-
mates are consistent with those of the animal studies.
The upper-bound risk estimate for 1 (Jg/1 of acrylonitrile in drinking
water is 1.2 x 10 , based on the Quast, et al. (1980a) drinking water study
in rats.
1-9
-------
There is evidence that acrylonitrile is a human carcinogen. This con-
clusion is based on: 1) findings of three positive drinking water rat bio-
assays and one positive rat gastric intubation study; 2) statistically signif-
icant positive findings of respiratory cancer in four epidemiologic studies;
3) the positive mutagenic evidence in bacteria and sister chromated exchange
tests; 4) .in vitro evidence of interaction acrylonitrile and/or its metabo-
lites with DNA; and 5) acrylonitrile's structural similarity to vinyl chlor-
ide, a known animal and human carcinogen.
The carcinogenic potency of acrylonitrile is in the third quartile among
54 suspected carcinogens evaluated by the Carcinogen Assessment Group.
Using the International Agency for Research on Cancer (IARC) classifica-
tion scheme, this level of evidence in animals and humans would be considered
sufficient for concluding that acrylonitrile is likely to be a human carcino-
gen with rank of 2A.
1-10
-------
2. INTRODUCTION
EPA's Office of Research and Development has prepared this health assess-
ment to serve as a "source document" for Agency use. This health assessment
was originally developed for use by the Office of Air Quality Planning and
Standards to support decision-making regarding possible regulations of acrylo-
nitrile under Section 112 of the Clean Air Act. However, based on the ex-
pressed interest of other agency offices, the scope of this document was
expanded to address acrylonitrile in relation to sectors of the environment
outside of air. It is fully expected that this document will serve the infor-
mation needs of many government agencies and private groups that may be in-
volved in decision-making activities related to acrylonitrile.
In the development of the assessment document, existing scientific liter-
ature has been surveyed in detail. Key studies have been evaluated and sum-
mary and conclusions have been prepared so that the chemical's toxicity and
related characteristics are qualitatively identified.
The present document represents an up-to-date data base. The document
considers all sources of acrylonitrile in the environment, the likelihood for
its exposure to humans, and the possible affect on man and lower organisms
from absorption. The information found in the document is integrated into a
format designed as the basis for performing risk assessments. When appro-
priate, the authors of the document have attempted to identify gaps in current
knowledge that limit risk evaluation capabilities.
2-1
-------
3. PHYSICAL AND CHEMICAL PROPERTIES
3.1 SYNONYMS AND TRADE NAMES
Chemical Abstracts Name: 2-propenenitrile
CAS No.: 107-13-1
EPA Toxic Substances List No.: R037-2101
RTECS No.: AT52500
Standard Industrial Code: 2822; 2821
The compound is also known as acrylonitrile (AN), cyanoethylene, propene-
nitrile, and vinyl cyanide (VCN). Fumigant formulations containing acrylo-
nitrile with names Acrylon, Carbacryl, Fumigrain, Ventox, and ENT-51 are no
longer manufactured in the United States.
3.2 STRUCTURAL AND MOLECULAR FORMULAS AND MOLECULAR WEIGHT
Hvt=C-GEN
H' H
C H3N Molecular Weight: 53.06
3.3 BOND ANGLES AND BOND DISTANCES
A molecule of acrylonitrile is planar; all the bond angles are close to
120°. Estimated bond distances are as follows (Wilcox and Goldstein, 195*0:
C-H: 1.09 A; C-C: 1.H6 A; C=C: 1.38 A; and C N: 1.16 A.
3.1 PHYSICAL PROPERTIES
3.1.1 Description
Acrylonitrile is a clear, colorless, and highly flammable liquid that has an
unpleasant and irritating characteristic odor (Fassett, 1963).
3.1.2 Boiling Point
77.3CC at 1 atmosphere (Groet, 1978).
3.1.3 Melting Point
-83.55°C (Groet, 1978).
3-1
-------
3.1.1 Density
djj°( liquid) : 0.8060 g/ml; vapor density: 1.83 (air=l) (Greet, 1978).
3.1.5 Refractive Index
on
n£u : 1.3911 (Weast, 1976).
3-1.6 Spectroscopic Data
The A is at 203 nm with a molar extinction coefficient of 6100; a
ZDoX
compilation of infrared, Raman, NMR, and mass spectral data is available in
Grasselli and Ritchey (1975).
3.^.7 Solubility
Acrylonitrile is soluble in water, acetone, and benzene (Weast, 1976);
miscible with ethanol, carbon tetrachloride, ethyl acetate, ethylene cyanohy-
drin, liquid carbon dioxide, ether, toluene, petroleum ether, and xylene (Miller
and Villaume, 1978).
The solubility in water is given below (Groet, 1978):
0°C: 7.2 weight?
20°C: 7.35 weight?
10°C: 7.9 weight?
60°C: 9.1 weight?
3.1.8 Volatility in Water
Henry's law constant: 0.063 at 25°C (Bocek, 1976). Partial vapor pressure
(water azeotrope) : log P = 7.518 - j^'L i-e., 80 mm at 20CC (Miller and
Villaume, 1978). The half-life of evaporation of acrylonitrile from water with
an assumed 1 meter depth can be calculated, using the method of Dilling (1977),
to be 795 minutes. The method of Dilling (1977) is applicable for slightly soluble
3-2
-------
organic compounds present In a static aquatic system. Therefore, the calculated
evaporative half-life for moderately soluble acrylonitrile present in a
naturally flowing aquatic system may not be accurate.
3.4.9 Volatility
Values for the vapor pressure of acrylonitrile (mm of mercury) at different
temperatures are given below (Groet, 1978):
8.7°C: 50.3
23-6°C: 99.8
H5.5°C: 2U9.8
64.7°C: 500.3
77.3°C: 759.8
3.4.10 Stability
Flash-point (open cup): 0°C (Steere, 1968); Flash-point (closed cup): -1CC
( Patterson et al., 1976), -U.1»°C (Miller and Villaume, 1978).
Explosive limits: 3.0 to 17$ by volume in air at 25°C (Steere, 1968).
Ignition temperature: 48l°C (Steere, 1968).
3.4.11 Octanol-water Partition Coefficient
k = 0.12 (Leo et al., 1971).
3.^.12 Conversion Factor
1 ppm in air =2.17 mg/m3 at 25°C.
3.5 CHEMICAL PROPERTIES
3.5.1 Reactivity
Undergoes reactions at both the nitrile group and the double bond (Maltoni
£t al., 1977). Some of these reactions are used for the quantification of
acrylonitrile and have been discussed in Section 1. Acrylonitrile also undergoes
the following reactions (Miller and Villaume, 1978), many of which are important
commercially.
3-3
-------
3.5.2 . Polymerization
Polymerization, forming high molecular weight products, is the most
important commercial reaction of acrylonitrile. The polymerization usually
requires the presence of free radical initiators, such as peroxydisulfate. Heat
o
or light (X < 2900 A) can also initiate polymerization reactions. Oxygen and
methylhydroquinone are powerful inhibitors of the polymerization reaction.
Pure polyacrylonitrile cannot be dyed using conventional techniques.
Copolymerization of acrylonitrile with small amounts of methyl methacrylate or
vinyl pyridine introduces reactive dyeing sites. Acrylonitrile can be copoly-
merized with other monomers as well; examples of other acrylonitrile copolymers
include nitrile rubber, acrylonitrile-butadiene-styrene (ABS), and styrene-
acrylonitrile (SAN) resins. Terpolymers of acrylonitrile or methacrylontrile
are the so-called barrier resins.
3.5.3 Reaction at the Nitrile Groups
Acrylonitrile when reacted with 8H.5% H2son at 100°C produces acrylamide
sulfate, which yields acrylamide upon neutralization:
CH2=CHCN •»• H20 + H^O^ ) CH2=CHCONH2 • H^O^
CH2=CHCONH2 • H2SOU + 2NaOH ) CH2=CHCONH2 + Na2SOu + 2H20
Until recently, the above process was used exclusively for the commercial
production of acrylamide. The catalytic hydration of acrylonitrile is the cur-
rently used method for the production of acrylamide. When acrylontrile is heated
with less concentrated H^O^ or with water, acrylic acid (CH2=CHCOOH) is formed.
In aqueous NaOH, NH., is the reaction product. The hydrolysis constant for this
reaction was measured as 3-8 x 10~^ min" at 60°C and 1.39 min" at 100°C
(Linetskii and Serebryakov, 1965). Violent polymerization, however, has been
reported to occur with concentrated alkali (Steere, 1968). Acrylonitrile
allowed to react with alcohols in the presence of concentrated H-SOj.
3-4
-------
produces esters of acrylic acid. With olefins, it forms N-substituted acryl-
amides in the presence of concentrated H-SOj..
3.5.1) Reactions at the Double Bond
The double bond in acrylonitrile acts as a dienophile in the Diels-Alder
reaction. Cyclic products are produced when acrylonitrile is treated with ali-
phatic or alicyclic compounds containing conjugated carbon-to-carbon double
bonds. An example is the reaction with butadiene:
CH2=CHCN + CH2=CH-CH=CH2 - HC CHCN
HC
A-3-tetrahydrobenzonitrile
In the presence of catalyst, acrylonitrile can be hydrogenated to propio-
nitrile, which can be further hydrogenated to n-propylamine:
CH -CHCN "2 CH CH CN "2
It can also be reduced, in the presence of magnesium and methanol, in the
following manner to produce adiponitrile:
or»u _ftJPM _i OPU f\U • \Jtfr PU PU /"*M
£Un_ = L/nUN + cLH—UM + rlK ~~^"^"~ ^n_t/H^LN
2 3 22.
This is then further reduced to hexamethylenediamine, which is used in the
production of nylon.
3.5.5 Cyanoethylation Reactions
These reactions involve the interaction of acrylonitrile with compounds
containing active hydrogen. Examples of compounds containing active hydrogen
are water, alcohols, ammonia, amines, mercaptans, aldehydes, and inorganic acids
3-5
-------
and their salts (Miller and Villaume, 1978). The generalized reaction can be
written:
CH2=CHCN + AH > ACH2CH2CN
The cyanoethylation of pseudouridine, inosine, and ^-thiouridine by acrylo-
nitrile has been studied as a model for the cyanoethylation of intact tRNA
(Miller and Villaume, 1978).
3.6 CHARACTERISTICS OF THE CHEMICAL PRODUCT
Technical grade acrylonitrile is a highly purified product with greater
than 99$ purity. The major impurity is water, which is usually present at a
maximum of about 0.5$. The water improves the stability of the product. Other
possible trace contaminants include acetone, acetonitrile, acetaldehyde, iron,
peroxides, and hydrogen cyanide. Highly pure acrylonitrile may polymerize spon-
taneously. To prevent this, methylhydroquinone (35-50 ppm) is added to the
commercial product. Yellowing upon exposure to light indicates photoalteration
to saturated derivatives.
3.7 CONCLUSION
Acrylonitrile is moderately soluble in water. Because acrylonitrile does
not appreciably dissociate in water, hydrogen cyanide is not expected to be a
product of hydrolysis. Because the vapor pressure of acrylonitrile is appre-
ciably high, most atmospheric emissions from its manufacture and use should occur
as vapor. The evaporation rate of acrylonitrile from water is appreciable.
Therefore, evaporation from contaminated water surfaces can be expected to
occur.
3-6
-------
H. SAMPLING AND ANALYTICAL METHODS
The level of acrylonitrile has been determined in a number of environmental
media of interest. These include: (1) air; (2) water; (3) waste water; (1) soil
and sediment; (5) residue in polymers and the extent of monomer migration in
food-simulating solvents; and (6) various other media. All these media can
directly or indirectly affect the environmental level or human intake of acrylo-
nitrile.
The sampling method is generally dependent on the medium intended to be
monitored. The analysis of samples can be divided into two steps, namely,
pretreatment or clean-up procedure( s), when necessary, and quantification pro-
cedures. The selection of a particular identification and quantification method
is dictated by the accuracy, reproducibility, detection limit, and the possible
interference(s) of the method. The sampling and analysis of acrylonitrile in the
individual media are discussed below.
M.I AIR
4.1.1 Sampling
The collection of acrylonitrile from air has been done by two methods. The
first method consists of direct collection of samples without preconcentration.
The second method employs the concentration of acrylonitrile in a collection
medium during sampling.
In direct sample collection, the air is drawn into plastic bags (Keresztesy
•
et al., 1977) via a two-way valve by means of a sampling pump. The samples are
then transported to the laboratory for analysis.
The disadvantage with direct collection is that it does not allow any
concentration of the sample and may cause additional problems during transport
and storage of samples. Therefore, the detection limit of the method is not
satisfactory for ambient air samples, even with the most sensitive method of
4-1
-------
detection available presently. The advantage of direct collection is that the
samples can be analyzed without any pretreatment, thereby reducing analysis time
and avoiding any sample losses. It also provides a method for continuous area
monitoring of acrylonitrile and for monitoring acrylonitrile from stack and
other high concentration emission sources.
In the preconcentration method for sample collection, the air containing
acrylonitrile is passed through either a solid sorbent or a trapping liquid
medium. Table H-l lists the different sorbents and trapping media used for
collection of acrylonitrile in air. Activated carbon, silica, and porous
polymers have been used as solid sorbents. The sampling unit usually consists of
a battery operated pump and a rotometer that indicates the sample flow rate
through the sorbent. A tube that contains the solid sorbent is held vertically
at a height of 1.5 m from the ground and is connected to the rotometer and the
pump unit. Air is drawn through the sorbent tube at a certain flow rate for a
specified length of time. At the end of sample collection, the tube is closed
with caps and shipped to the laboratory for analysis.
In addition to the error that could be made in measuring the air volume
(calibrated pumps may not maintain constant air flow over the entire sampling
period), another disadvantage of the solid sorbent method is that sample loss
will occur if the breakthrough capacity for the sorbent is exceeded. Therefore,
the breakthrough capacity of acrylonitrile through the sorbent should be deter-
mined experimentally. The breakthrough capacity is dependent on the relative
humidity of air sampled and the presence of other interferences in the air. The
concentration of the compound being collected, however, affects the breakthrough
volume only slightly. A pollutant present at 20 ppm will break through at only a
slightly smaller volume than when present at 1 ppm (Russell, 1975).
4-2
-------
Table 4-1. Various Sorbents and Trapping Media for Collection of Acrylonitrile in Mr
i
CO
Sample
Plastic producing
plant
Acrylic fiber
factory
Simulated air
Plastic processing
plant
Eleven production
plants
Simulated air
Simulated air
Settling pond In
acrylonltrlle plant
Occupational air
Occupational air
Craln fumlgant air
Occupational air
Acrylonltrlle plant
—
Rubber plant
SAN plant
— — —
Sorbent
Sllochrom-2 at
-50'C
Silica gel
Activated carbon
Activated carbon
Activated carbon
Porapak N
Porapak N
Porous polymer
Tenax GC(T)
Activated carbon
Carbosleve B
95Z ethanol
Chilled water
Water
Water
Chilled water
Chilled water
Water
H.SOg & glasa beads
1Z H2S04
1Z H.SO.
I n
DMF
25Z methanol at
-15'C
Sampling
Condition
0.5 1pm for
4-6 1
Personnel
sampler for
100 nln.
15 1 in gas
bag
0.2 1pm for
8 hr.
1.0 1pm for
=24 hr.
0.1 1pm for
3-5 1
0.1-0. IS 1pm
for 3-8 1
=0.07 m3/hr.
for 5-30 mln.
0.2 1pm for
20 1 max
1 1pm for
10-20 1
1 Ipa
0.03-0.33 1pm
0.015 1pm for
1.5-3 1
1 1pm for
20-60 1
1 1pm
0.4 1pm
0.1-0.3 1pm
— — —
Desorptlon Analysis
Method Technique
Water Colorlmetrlc
Thermal at 200'C CC-FIDh
Water ' CC-FID
2T acetone in CS_ CC-PID
2Z acetone In CS. CC-PID
CS? CC-FID
Thermal at 200*C CC-FID
Thermal CC-FID
Thermal CC-FID
CHjOB CC-FID
CHjOH CC-FID
Polarographlc
Polarographlc
Colorlmetric
Colorlmetrlc
u.v.
CC-FID
Colorlnetric
Tltrimetrlc
Colorimetrlc
Colorlmetrlc
CC-FID
GC-FIO
Conf Irmat Ion
Technique
None
None
None
None
None
Rone
None
None
None
None
Rone
Rone
None
None
None
None
None
None
None
None
None
None
None
Detection
Limit
0.1 ppm
0.25-0.5 ppm
0.03-0.31 ppm
< 0.3 ug/a3
< 1 ppb
< 0.26 ppm
10 ng
Range 5-135
mg/m3
Range 40-750
0.5-5 mg/50 ml
1.08 x 10"5 H
0.5 mg/m3
Range 10-100
ppm
50 ppb
1.0 ug/nl
20-300 ug/ml
< 0.85 ug/1
0.5 mg/m3
...
Reference
Rrynska. 1970
Korthova et^ al. , 1974
Sakural e^ a^. , 1978
CagnoTi & Pogner,
1979
Marrs ft. al. , 1978
Going et a_l. , 1979
Russell. 1975
Campbell & Moore.
1979
Hughes ft Horn, 1977
N10SH, 1977
Barrett. 1974
Rogaczewnka, 1964
Berck, 1962
Najiarova & Nakrap
1978
RogacKewnka, 1976
Brleger ct^ •!. . , 1952
Hughes ft Horn. 1977
Russklkh. 1973
Gunther & Bllnn. 1955
Babanov, 1960
Arato & Blttera, 1972
Bablna, 1979
Fasltta a Llcclardello,
1977
1pm - liters per minute
CC-FID - gas chromatography-flame lonlzatlon detectors
-------
For air nearly saturated with moisture, the breakthrough volume for acrylo-
nitrile in Porapax N (H" x 1/1" tube) was determined to be 3 to 5 liters.
Breakthrough volumes can be increased by using longer sampling tubes. These
tubes, however, will result in increasing the back pressure in £he tube.
Activated carbon is the most widely used sorbent for the collection of
acrylonitrile in air. The adsorption parameters for activated carbon will vary
depending on the nature of the carbon and the treatment it received. With 6-10
mesh BC-AC granular carbon, Sansone et al. (1979) determined the adsorption
capacity and adsorption rate constant for acrylonitrile to be O.IOH g/g and 116.0
min. , respectively. With acrylonitrile that contained 50$ relative humidity,
Nelson and Harder (197*0 determined the adsorption capacity to be 0.17** g/g. The
breakthrough times for 1$, 10$, and 99$ passage from a 58 g charcoal in a
respirator cartridge were determined to be U8.5 min., 61.1 min., and 168 min.,
respectively (Nelson and Harder, 197*0. The types of activated carbon used in
the above experiments were different.
The effect of humidity on sample recovery by 1.5 g activated carbon was
studied in detail by Going £t al. (1979). He observed that introducing a
Drierite drying tube to absorb moisture did not significantly improve recoveries
at relative humidities equal to or exceeding 70$; however, the average recovery
for acrylonitrile with air volumes of 750 liters/g or greater remained acceptable
at 75 i 9.1$ at relative humidities up to 60$, with or without the drying tube.
NIOSH (1977b) determined that the maximum amounts of acrylonitrile that
could be collected with a sorbent tube (8 cm x 5 mm) from 100 ppm acrylonitrile in
air were 6 mg, 15 mg, and 22 mg when the relative humidities greater than 95$,
equal to 50$, and less than 5$, respectively.
Segmentation of the carbon tubes into a front and back section remains the
most widely used method for the determination of breakthrough capacity limit
4-4
-------
during sample collection. If the second tube retains more than a certain
predetermined percentage of acrylonitrile, it indicates that the breakthrough
limit has been exceeded.
It is important to determine the stability of acrylonitrile in the sample
tubes during transportation and storage. Marrs e_t al^ (1978) showed that acrylo-
nitrile would remain stable on the carbon tubes at room temperature for a
minimum of 5 days. Going et al. (1979) demonstrated that acrylonitrile would
remain stable up to 8 days on charcoal tubes stored at -17°C and up to 24 days on
tubes stored at -78°C.
The collection of acrylonitrile in air by liquid trapping media has been
used rarely in recent years because of the inconvenience in handling and trans-
porting the impingers and the need for cooling the trapping media during sample
collection. The collection efficiencies and the storage stabilities of acrylo-
nitrile in these media are not always known. The collection efficiency of
chilled water as a trapping medium was determined to be over 98$ (Berck, 1962;
Brieger, 1952). Virtually no data are available regarding the storage stability
of acrylonitrile in this trapping medium, but it is expected to be reasonably
stable in water and 1$ H-SCL, particularly since Going et_ al. (1979) demonstrated
that acrylonitrile was stable for about 23 days in neutral distilled water and at
a pH of 4. Acrylonitrile collected in aqueous methanol and ethanol and stored at
3°C can be expected to be fairly stable, based on the evidence of Going e_t al.
(1979), who demonstrated that acrylonitrile in carbon disulfide remained stable
for over 12 days when stored at 3°C.
4.1.2 Analysis
1.1.2.1 Pretreatment
Acrylonitrile collected on solid sorbents requires a desorption procedure
before identification and quantification. Thermal desorption and solvent
4-5
-------
desorption are two commonly used methods. Thermal desorption normally uses gas
chromatographic methods for identification and quantification. In this pro-
cedure, the sorbent tube is heated between 100°C and 200°C in the injection port
of a gas chromatograph. To avoid peak broadening, the position of the sample
tube in the gas chromatograph injection port should be in the reverse order as
that used during sampling. The thermal desorption method is appropriate for use
with porous polymers, particularly with Porapak N. Several investigators have
used this technique (Campbell and Moore, 1979; Russell, 1975; Hughes and Horn,
1977). The advantages of thermal desorption are that it avoids manual sample
pretreatment and the recovery of acrylonitrile is almost quantitative (Campbell
and Moore, 1979; Russell, 1975). The method also has a higher sensitivity than
the solvent desorption method, where only a fraction of the eluted acrylonitrile
can be injected into the gas chromatograph. The disadvantages of the thermal
desorption method include its inability to afford replicate analysis of the same
sorbent tube and its tendency to cause other gas chromatography separation
problems such as shortening retention time and broadening eluted peak due to the
presence of adsorbed moisture on the sorbent column (Russell, 1975).
The solvents that have been used for desorption of acrylonitrile from acti-
vated carbon are acetone, methanol, carbon disulfide, and 2% acetone in carbon
disulfide. The selection of the desorption sorbent is dictated by two considera-
tions, namely, the sorbent desorption efficiency and its compatibility with gas
chromatography. For example, if column separation is not adequate, methanol that
has a high response on flame ionization detectors (FID) will produce a large peak
shadowing the acrylonitrile peak; therefore, it is not very compatible with flame
ionization detectors. Although carbon disulfide appears to be compatible with
flame ionization detectors, it is a poor solvent for nitrogen/phosphorous detec-
tors. Acetone is the solvent of choice in the latter case.
4-6
-------
The recovery efficiencies of acrylonitrile from activated carbon with
various solvents are given in Table 1-2.
Table 1-2. Recovery of Acrylonitrile from Various Solvents
Solvent
% Recovery
Reference
Methanol
Acetone
2% acetone in CS
2% acetone in CS
CS2 (2 ml)
CS2 (U ml)
ca. 50$
73.5 ± 5.3$
95.5 ± 7.9$
91*
58$
75$
Going et al., 1979
Marano et al., 1978
Gagnon and Posner, 1979
Silverstein, 1977
Silverstein, 1977
Silverstein, 1977
It should be recognized that the recovery of acrylonitrile is dependent on
the nature of the activated carbon and the extent of loading. Using three
different activated carbons and a variable loading of 2 ug to 200 |ig acrylo-
nitrile, Going eta^. (1979) determined that the CSp desorption efficiency varied
from 53$ to almost 100$. It would appear from Table 4-2 that a 2$ acetone in CS
is the best solvent for elution of acrylonitrile from activated carbon in terms
of both recovery and GC-FID compatibility.
When the acrylonitrile is collected in liquid trapping media, the samples
from high concentration sources usually do not require any pretreatment prior to
the detection and quantification procedures.
1.1.2.2 Identification and Quantification
The methods utilized for the identification and quantification of acrylo-
nitrile collected by sorbent or trapping techniques are shown in Table 1-1 (see
1.1.1). The methods for the analysis of acrylonitrile collected without precon-
centration appear in Table 1-3. The infrared techniques are used exclusively for
samples that need no preconcentration. Although a number of methods including
4-7
-------
Table 4-3. Direct Analysis of Acrylonitrile
Sample
Simulated occupa-
tional air
Occupational air
Acrylonitrile in N,,
Simulated air
Gases generated by
heating ABS plastic
Occupational air
Simulated air
Occupational air
Sample Introduction
into Detector Unit
Sample probe & pump
Introduction of air
from gas bag into
evacuated cell
Introduction to a
vacuum cell at 8 torr
Pump & rotometer
Forced air
Passive diffusion
Direct injection
Direct injection
Analysis Technique
IR-microcomputer
IR
IR lasers
Detector tube
Detector tube
TM
Abcor GASBADGE
GC-NPD
GC-FID
Comf irmation
Technique
None
None
None
None
GC-MS
None
MS
None
Detection
Limit
0.2 ppm
0.5 ppm
0.03 ppm
Range 1-120
ppm
Range 0.8-19
ppm
100 ppb
0.5 ppm
References
Jacobs & Syr Jala,
1978
AIHA, 1970
Sweger & Travis, 1979
Kobayashi, 1956
Grote et^ al. , 1978
Silverstein, 1977
Marano et al. , 1978
AIHA, 1970
-p.
I
CO
GC-MS = gas chromatography-mass spectrometry
GC-NPD = gas chromatography-nitrogen/phosphorus detectors
MS = mass spectrometry
GC-FID = gas chromatography-flame ionization detectors
-------
colorimetric, titrimetric, and polarographic were used for the analysis of
acrylonitrile in the past, the GC method is most extensively used at the present
time. The reason for this is the lower interference and higher sensitivity of
detection obtained from GC-flame ionization detection (FID) and GC-
nitrogen/phosphorous detection (NPD). The chromatographic columns found to be
most appropriate for use with thermally desorbed acrylonitrile were Porapak N
(Russell, 1975) and Porapak Q (Campbell and Moore, 1979) . In the case of solvent
desorption, the columns that were found to be suitable were Durapak OPN/ Poracil
C (Going et al., 1979), SP-2100 (Grote et al., 1979) for carbon disulfide;
SP-1000 (Marano et al^., 1978) for acetone; and SP-1000 (Gagnon and Posner, 1979)
and TCEP (Marrs et al.., 1978) for methanol. The details of other chromatographic
columns are given in subsection M.7.
M.I.3 Conclusions
The detectors available at present often do not have sufficient sensitivity
to allow the detection of acrylonitrile in ambient atmospheric samples, if these
samples have been collected without preconcentration. The low level of acrylo-
nitrile in atmospheric samples virtually mandates the use of a preconcentration
device during sample collection. Of the two preconcentration methods presently
available, namely, solid sorbents and trapping media, the former method is
preferable to the latter. The collection of acrylonitrile by solid sorbents
affords convenience in handling, shipping, and storage of the samples.
Activated carbon and porous polymer Porapak N are the two best sorbents
available for the collection of acrylonitrile in air. The breakthrough capacity
for Porapak N (3 to 5 liters) is lower than that for activated carbon (over
1000 liters) . Consequently, Porapak N cannot be used for 2U hour sampling. Even
with much lower sample volume, Porapak N will afford a detection limit comparable
to that obtained from a larger volume of air collected by activated carbon, for
4-9
-------
the following reasons. The efficiency of thermal desorption for acrylonitrile
from Porapak N is quantitative, while it is poor with activated carbon; there-
fore, activated carbon requires solvent desorption. Whereas the entire sample
can be injected into the quantitative gas chromatography column in the case of
thermal desorption, in the case of solvent desorption, only a fraction of the
total sample can be injected.
With activated carbon, 2% acetone in carbon disulfide is the best desorption
solvent, since it gives the maximum recovery. This solvent system is compatible
with flame ionization detectors. Carbon disulfide, however, is unsuitable when
the more sensitive nitrogen/phosphorus detectors are used. Acetone is the most
suitable solvent in this case. A number of columns including SP-2100 (Gagnon and
Posner, 1979) and Durapak OPN/Poracil C (Going e_t al., 1979) have been used as
the quantitative column when carbon disulfide or 2% acetone in carbon disulfide
was the solvent. For acetone solvent, SP-1000 was found to be a suitable
quantitative column (Marano et^ al.. 1978) .
4.2 WATER
In this section, only water samples obtained from either surface water or
treated drinking water will be discussed. Wastewaters will be discussed in
subsection 4.3.
4.2.1 Sampling
Water samples have been collected almost exclusively by the grab technique
(Kopfler et al., 1976; Wronski and Zbigniew, 1974; Going et al.., 1979). In a few
instances, multiple grab samples were composited for analysis (Going e£ al.,
1979). Proposed EPA methods require that the samples be adjusted to pH 7.0 + 0.5
and be colected in screw-cap vials with Teflon* septa without headspace. The
collected samples should be stored in a refrigerator (U.S. EPA, 1979a; U.S. EPA,
1979b; U.S. EPA, 1980c). The ASTM (1980) procedure allows headspace, but
4-10
-------
recommends immediate freezing of the samples. However a better method suggested
that the samples should be collected in brown glass bottles with Teflon-lined
caps and acidified at the site to a pH <_ 14 (Going tit al_., 1979). The samples
should be maintained at 0 to iJ°C by ice or an ice substitute during transporta-
tion (Going et al., 1979; Kopfler et al., 1976) .
In one instance (Going et^ al^., 1979) , an attempt was made to collect acrylo-
nitrile from water by the use of solid sorbents. Four different sorbents,
namely, activated carbon, Porapak N, Chromosorb 101, and Chromosorb 104, were
tried. Water spiked with acrylonitrile was passed through the sorbents at a rate
of 4 ml/min. The sorbents were then eluted with 25 ml methanol at a rate of
5 ml/min. The recoveries of acrylonitrile were poor; with activated carbon, the
recovery was 30-35 percent and the porous polymers showed zero percent recovery.
1.2.2 Analysis
lJ.2.2.1 Sample Treatment
Some water samples were analyzed without any pretreatment. Water samples
containing low levels of acrylonitrile were pretreated in order to concentrate
the acrylonitrile. Three available methods for concentration are purge-trap
(Kopfler et al., 1976; Going etal., 1979), distillation (Going et al,., 1979;
Peters, 1979), and headspace (McNeil and Brader, 1981; Markelor et al., 1981)
techniques.
Acrylonitrile can be purged from water at elevated temperatures by passing
an inert gas through it. The acrylonitrile contained in the purged gas is then
trapped in chromatographic media for subsequent analysis. The details of the
purging system were described by Going et^ al. (1979) and Kopfler e_t al. (1976) .
The system used by Kopfler et^ al. (1976) appears to be preferable to the system
used by Going e£ al. (1979), because the former allows purging of 1HO ml of water
4-11
-------
compared to 10 ml water in the latter case; however, the purge-trap efficiency
for acrylonitrile was not studied in detail in the Kopfler et aK (1976) system.
The purging conditions were studied in detail by Going et al. (1979), who
found an almost quantitative recovery when helium was passed at a rate of
20 ml/min through water heated to 85°C for 30 minutes. The effect of different
trapping systems were also studied by Going et al. (1979). It was established
that, under thermal desorption conditions, both Porapak N and Chromosorb 104 gave
quantitative desorption. The recovery from Tenax GC, however, was found to be
poor (Going et al., 1979).
Two different distillation techniques, namely steam distillation and
azeotrophic distillation have been used to concentrate acrylonitrile from the
aquatic phases. In the steam distillation technique, the sample is directly
distilled for an appropriate interval and the organics of interest is collected
in the distillate. Peters (1980) has used this technique to concentrate acrylo-
nitrile by a factor of 300 in about 15 minutes of steam distillation. The major
disadvantage of this method and the azeotropic distillation is that relatively
large sample volumes (>300 ml) are required for these methods.
In the azeotropic distillation technique, water containing acrylonitrile is
distilled with methanol, and a small volume of the azeotropic distillate contain-
ing acrylonitrile in methanol is collected for further analysis. This technique
serves as a simultaneous clean-up and concentration device for acrylonitrile in
water samples. The description of the distillation apparatus was given by Going
et al_. (1979), who obtained maximum recovery under the following conditions.
A 500 ml water sample was added to the distillation flask along with 25 ml
methanol and 5 ml 18 N H-SOj. and the content was distilled at a rate of 1 ml/min.
The first 10 ml of the distillate were collected for subsequent analysis. The
4-12
-------
percent recovery was about 90J for the combined first and second 10 ml aliquot of
the distillate.
In the headspace method, a known volume of the aqueous sample is Introduced
into a specially designed enclosed glass apparatus and ths sytem is thermo-
statically maintained at a constant temperature. After the system attains
equilibrium, a known volume of the headspace vapor is withdrawn for acrylonitrile
determination. This method in the pas thad faced problems owing to the
difficulty in establishing a calibration procedure. The partition coefficient
of a component between gas and liquid phase is dependent on the total ionic
strength of the solution. Therefore, the same concentrations of a component
present in two aqueous solutions of different ionic strengths but otherwise
identical conditions, will not produce the same equilibrium vapor pressure. This
problem of a calibration curve has been large obviated through the development of
a standard addition method. The headspace method has been utilized by Markelor
et al. (1981) and McNeil and Brader (1981) for the analysis of acrylonitrile in
aqueous solutions.
U.2.2.2 Detection and Quantification
With the exception of one case in which a titrimetric method was used
(Wronski and Zbigniew, 1971*) , the rest of the studies reviewed utilized GC
separation and GC retention data for the identification of acrylonitrile from
water samples. For direct aqueous injection, both Chromosorb 101 (Going et^ al.,
•
1979) and Chromosorb 102 (Marano et al., 1978) were used, although Going Q al.
(1979) reported better separability with Chromosorb 101. The same column
(Chromosorb 101) was used for acrylonitrile determination by the purge-trap and
azeotropic distillation techniques (Going e_t al., 1979; Kopfler e_t al., 1976;
Federal Register, 1979). Other Ge columns including carbowax 1500 and Tenax GC
have been used for the separation of acrylonitrile (Markelov e_t al., 1981;
4-13
-------
Peters, 1980). An aliquot of the distillate from azeotropic distillation was
injected directly into the GC column. In the purge-trap technique, the
acrylonitrile from the trapping column was thermally desorbed onto the
separating column. A summary of water analysis techniques is given in Table 1-1.
Although nitrogen/phosphorus detectors (including the Hall detector) and flame
ionization detectors were used for quantification, the former detectors
antrogen/phosphorous) have better selectivity and sensibility than the latter
detector. However, the best available method for the unambiguous identification
of acrylonitrile is provided by the more expensive mass spectrometric method.
1.2.3 Conclusions
The two best available methods for the determination of acrylonitrile in
water are the purge-trap and the azeotropic distillation techniques. Both
methods gave almost quantitative recovery of acrylonitrile in water. The detec-
tion limit for acrylonitrile by the purge-trap method was lower than that for the
azeotropic distillation method; however, this advantage of the purge-trap tech-
nique is somewhat offset by the experimental complexity of the method and its
inability to perform replicate analysis on the same water sample.
1.3 WASTEWATER
1.3-1 Sampling
No details regarding the sampling of wastewaters is available. Grab samples
may be suitable in certain cases. To monitor the discharges that are dependent
on process operation stages, a 21-hour composite sample is preferable.
1.3.2 Analysis
1.3.2.1 Pretreatment
Azeotropic distillation with methanol is a method used frequently for
wastewater. This technique allows concentration of acrylonitrile and reduces
the possibility of interference. In one case, solvent extractions using benzene,
4-14
-------
Table U-*4. Analyses of Acrylonitrile in Water
Sample
Collection Analysis Technique Confirmatory Detection
Method Technique Limit
Reference
Spiked water
Drinking water
Surface water
Surface water
Surface water
Surface water
Spiked water
Spiked water
Spiked water
Direct injection, MS
GC-NPD
Grab Purge-trap ( Tenax GC) , None
GC-MS
Grab Titrimetric with None
mercaptoethanol > Na_SO,
Grab Direct injection, MS
GC-FID
Grab Purge-trap, GC-FID, > MS
GC-Hall
Grab Azeotropic distillation, MS
GC-Hall
Sorbent Methanol desorption, None
adsorption GC-Hall
Steam distillation, None
GC-FID
— Headspace, GC-Hall None
10 ppb
Qualitative
0.5-0.05 ppm
1 ppm
0.1 ppb
0.1-1.3 ppb
Showed poor
desorption
efficiency >
method
abandoned
NR
NR
Marano e^ al.
Kopfler et al.
, 1978
, 1976
Wronski > Zbigniew,
197U
Going et al. , 1979
Going ejt al. ,
Going et^ al. ,
Going et al. ,
Peters, 1980
Markelovetal.
1979
1979
1979
,, 1981
MS = mass spectroraetry
GC-NPD = gas chromatography-nitrogen/phosphorus detectors
GC-MS = gas chromatography-mass spectrometry
GC-FID = gas chromatography-flame ionization detectors
GC-Hall = gas chromatography-Hall detector
NR = Not Reported
-------
ether, and isobutyl acetate were used (Ponomarev et al., 197M) . The recovery of
acrylonitrile after three extractions was not quantitative but was reproducible.
The various pretreatment methods used for acrylonitrile determination in
waste water are shown in Table M-5. Table M-5 also lists the different detection
and quantification methods and their detection limits where available. The
principles of the detection methods are discussed in subsections 1.2.2.2 and 1.7.
U.3.3 Conclusions
Although the analysis of acrylonitrile in wastewater by the purge-trap
technique has not been reported, it is a potentially appropriate method for
wastewater analysis. The usefulness of the method can be enhanced by fractional
purging at different temperatures (Kopfler et al., 1976). A better method for
acrylonitrile analysis in wastewater is the azeotropic distillation technique.
The detection and quantification can be best achieved by either the GC-FID or the
GC-NPD technique. The chromatographic columns used for water analysis should be
suitable for wastewater analysis.
1.1 SOIL AND SEDIMENT
Only one reference could be found in the literature for the analysis of
acrylonitrile in these media. The following discussion in this section is based
on the work of Going et al. (1979).
1.1.1 Sampling
Soil from the top 12 cm was removed and placed in glass bottles and stored
over dry ice until analyzed. Sediment samples were collected with a dredge and
the samples were kept on dry ice until analyzed.
1.1.2 Analysis
1.1.2.1 Pretreatment
The pretreatment of the sediment samples after the removal of excess water
was the same as that used for soil samples. The samples were extracted with
4-16
-------
Table 4-5. Analysis of Acrylonitrile in Wastewaters
Sample
Pretreatment
Analysis Technique
Detection Limit
Reference
From polymeri-
zation process
Solvent extraction
Azeotropic distillation
Azeotropic distillation
None
Azeotropic distillation
None
Polarographic
Polarographic
Polarographic
Titrimetric
Colorimetric
Titrimetric
GC-FID
10 ppm
3-5 ppm
< 0.1 ppm
< 2 ppm
< 0.5 ppm
0.1 ppm
Sevest'ysmova et^ al^., 1966
Ponomarev et al., 1974
Daues & Hamner, 1957
Stefanescu & Ursu, 1973
Lawniczak, 1977
Covic-Horvat £t_ al., 1970
Deur-Siftar & Svob, 1976
GC-FID = gas chromatography-flame ionization detectors
-------
water using ultrasonic agitation and the mixtures were centrifuged. The super-
natants were withdrawn and filtered, and the aqueous extracts were directly used
for analysis. A few samples were concentrated by the purge-trap technique
already discussed in subsection U.2.2.1. It should be mentioned that the overall
recovery of acrylonitrile from the soil and sediment samples either by direct
extraction or by the subsequent purge-trap technique was not determined.
4.1.2.2 Detection and Quantification
Gas chromatography with a Chromosorb 101 column and a Hall detector was used
for detection and quantification. The detection limit for direct aqueous injec-
tion ranged from 50 to MOO ug/kg. For the purge-trap technique, the detection
limit was ca. 0.5 ug/kg.
4.U.3 Conclusions
The desorption efficiency of acrylonitrile from soil and sediment samples
by ultrasonic agitation has yet to be determined. Although azeotropic distilla-
tion was not used, both this method and the purge-trap technique should be
suitable for the determination of low levels of acrylonitrile in these samples.
U.5 RESIDUE IN POLYMERS AND THE EXTENT OF MONOMER MIGRATION IN FOOD-
SIMULATING SOLVENTS
Several polymers of acrylonitrile are presently used in the United States as
food packaging materials. Until recently, some containers for carbonated bever-
ages had nitrile barrier resins as components; however, the use of nitrile
barrier resins in beverage containers in the United States is presently banned by
FDA. The ABS resins (terpolymer of acrylonitrile, butadiene, and styrene),
however, are currently used for such food packages as margarine tubs, fruit juice
containers, and vegetable oil bottles. In this section, the analysis of acrylo-
nitrile residue in both polymers used for food packaging and their migration to
food-simulating solvents kept in contact with the polymer will be discussed.
4-18
-------
H.5.1 Analysis
14.5.1.1 Pretreatment
1.5.1.1.1 Pretreatment for Polymers
Three methods are available for the pretreatment of polymers prior to quan-
tification. In one method, the sample is dissolved in a suitable solvent and an
aliquot of this solution is used for analysis. The solvents used for different
polymers are shown in Table ^-6. The disadvantage of direct injection of the
solution is that it accelerates the deterioration of the separating column when
GC is used for identification. The polymer build-up in the gas chromatograph
injection port can be prevented by addition of water or methanol to precipitate
the polymer and the supernatant can be injected into the gas chromatograph. The
detection limit of this method, however, is not satisfactory and was determined
to be about 10 ppm (Steinchen, 1976).
To avoid column contamination, reduce the interference arising from large
amounts of solvent, and increase the sensitivity of detection, the second method,
known as the head-space analysis, is presently used for the determination of
residual monomer in polymers.
Two approaches to the head-space analysis have been used: solid and solu-
tion. The solid approach involves the equilibration of a solid polymer sample in
the sealed tube at a constant elevated temperature. The advantage of the solid
approach is that it has tenfold more sensitivity than the solution approach
(Steinchen, 1976). The disadvantages of this approach are: (a) equilibration
with the head-space may take a long time, and (b) since polymer standards of
known monomer content are not readily available, the head-space monomer concen-
tration must be related to the original concentration in the polymer either by
assuming 100$ diffusion of the monomer into the head-space or through determina
4 -19
-------
Table 4-6. Analysis of Acrylonitrile Residue in Polymers and Food-Simulating Solvents
Sample
Plastic
Plastic
Residue In
food-simulating
solvent
Polymer
Residue In
food-simulating
solvent
Residue In
food-simulating
solvent
Residue in
food-simulating
solvent
Polymer
Residue In
food-simulating
solvent
ABS resin
Residue In
food-simulating
solvent
Polymer
Plastic
Residue In
beverages
Pretreatment
Pyrolysla & solvent
trapping
Dissolved in benzene
Direct
(a) Pyrolysls CC
(b) Dissolution in
dimethyl formamlde
(c) Dissolution In ether
Head-space equilibration
Azeotroplc distillation
Direct
Dissolved In
N,N-dlme thy lace t amide
Azeotropic distillation
Dissolved in dimethyl
formanlde or dimethyl
sulfoxide and head-space
equilibration
Head-space equilibration
Dissolved In o-dichloro-
benzene & head-space
equilibration
Dissolved in propylene
carbonate and head-space
equilibration
Head-space equilibration
Analysis
Technique
Polarographic
Tltrlmetric
GC-FID
CC-FID
CC-FID
CC-NPD
GC-NPD
CC-NPD
GC-NTD
GC-FID &
GC-NPD
CC-NPD
CC-FID
GC-NTD
CC-NTD
Confirmation Detection
Technique Limit
None 100 ppm
None < 6 ppm
None
None 10-100 ppm
None
MS < 0.01 ppm
None 0.04 ppm
None 0-5 ppm
None 50 ppb
None 0. 3 ppm
None 30 ppb
None 0.5 ppm
None °-1 PP™
None 5 PPb
Reference
Ulide & Koehler, 1967
Roy, 1977
Harkelov & Semenenko, 1976
Relchle & Tengler. 1968
Chudy & Crosby, 1977
KcNeal e£ a_l. , 1979
Brown £t al. , 1978
Brown e_t al. , 1978
Hartshorn, 1975
DIPasquale et al., 1978
DIPasquale et aJU , 1978
Stelnchen, 1976
Cawell, 1979
Cawell, 1979
-e»
CD
CC-FID - gas cliroiaatography-flarae lonlzatlon detectors
CC-NPD • gas chromatography-nltrogen/phosphorus detectors
-------
tion of equilibrium concentration using Henry's law and the appropriate parti-
tion coefficient.
The solution head-space approach has been used with a much wider range of
samples (Steinchen, 1976; Gawell, 1979; DiPasquale et al_., 1978). In this case,
the polymer has been dissolved in a suitable solvent and allowed to equilibrate
in a sealed vial at a constant temperature (60 to 90°C). Sometimes water has
been added to the solvent to enhance the equilibrium monomer concentration
(Steinchen, 1976). The advantages of this approach are: (a) head-space
equilibration is rapid, (b) calibration procedure is simplified, and (c) the
head-space gas/solution partitioning of the constituents is not appreciably
affected by changes in the solvent phase.
The third method, which is rarely used at the present time, consists of
passing nitrogen gas through the heated polymers and trapping the volatile com-
ponents in dimethyl formamide (Uhde and Koehler, 1967). The dimethyl formamide
solution is subsequently analyzed for acrylonitrile. A variation of this method
called the pyrolysis GC technique has also been used (Reichle and Tengler, 1968).
M.5.1.1.2 Pretreatment of Food-Simulating Solvent Containing Monomer
Residue
The pretreatment of food-simulating solvents for the determination of
monomer content as a result of migration from the polymer has been done in three
ways. The first method that has been used is the direct analysis of the solvent
left in contact with the polymer for a predetermined time and temperature
(Markelov and Semenenko, 1976; Brown et^ al., 1978). The disadvantage of this
method was that it had a poor sensitivity and the large amount of solvent
interfered with the small amount of acrylonitrile during the detection and quan-
tification stage (Brown et aU, 1978; Hartshorn, 1975).
4-21
-------
To avoid the solvent interference and to enhance the sensitivity of detec-
tion, either the second or the third method, namely, azeotropic distillation or
head-space analysis, is presently employed.
In the azeotropic distillation technique, the solvent is distilled with
methanol and the azeotrope containing methanol and acrylonitrile is collected
for further analysis. When water or 8$ ethanol was used as the food-simulating
solvent, the distillation was done by adding methanol directly to the solvent
(McNeal et^ al., 1979). When 3$ acetic acid was used as the food-simulating
solvent, the solution was neutralized with sodium hydroxide before distillation
with methanol (McNeal £t al., 1979). When heptane was used as the food-
simulating solvent, it was extracted with water and the water extract was dis-
tilled with methanol (McNeal _et^ al., 1979). The average recovery of acrylo-
nitrile from the azeotropic distillation of all the extracts varied between
29.7 + 2.6$ and 32.6 + 2.8$ (McNeal et al., 1979).
The third method employs the principle of the head-space equilibration
technique. The solvent was introduced in a sealed vial and allowed to equili-
brate at a predetermined temperature for a certain length of time (Chudy and
Crosby, 1977; DiPasquale et al., 1978; Gawell, 1979).
4.5.1.2 Detection and Quantification
Although a polarographic (Uhde and Koehler, 1967) and a titrimetric method
(Roy, 1977) have been used for the determination of monomer in polymers, these
methods lacked adequate sensitivity for the determination of low levels of
acrylonitrile. The method used almost exclusively at the present time is gas
chromatography, either with flame ionization detectors or with nitrogen/phos-
phorus detectors. The sensitivity and selectivity of NPD makes it a preferable
method over FID.
4-22
-------
4.5.2 Conclusions
The best available method for the determination of monomer residue in
polymers seems to be the solution head-space and GC analysis. For the determina-
tion of the extent of monomer migration in food-simulating solvents, both azeo-
tropic distillation and solution head-space equilibration with subsequent GC
analysis are the preferred methods. The preferred stationary phases that have
been used for non-aqueous injections are Carbowax-20M and Carbowax-1500 (Chudy
and Crosby, 1977; DiPasquale et al., 1978; Steinchen, 1976; Gawell, 1979). For
aqueous injections, gas-solid chromatography with porous polymer packings such
as Chromosorb-102 (DiPasquale et al., 1978), Chromosorb-101 and -108 (McNeal
e£ al., 1979) and Porapak-QS and -S (Brown et al., 1978) have been used.
U.6 OTHER MEDIA
Acrylonitrile has been determined in many other media such as tobacco,
production streams, and foods and grains consumed by humans. Since acrylo-
nitrile-containing pesticides have been voluntarily withdrawn from the market
and the other media have no direct bearing on general population exposure,
acrylonitrile analysis in these media will not be discussed.
1J.7 GENERAL METHODS FOR THE ANALYSIS OF ACRYLONITRILE
This section presents the general methods available for detection and quan-
tification of acrylonitrile without regard to the medium in which it is present.
Many of the methods have been applied for analysis of acrylonitrile present in
more than one medium. This approach has been adopted so that replication of
discussion of the same method from medium to medium is avoided.
The identification and quantification methods can be divided into two cate-
gories: one based upon the chemical reactivities of the functional groups in
acrylonitrile and the other based upon instrumental techniques.
4-23
-------
The chemical techniques that have proved useful are based on (a) hydrolysis
of the nitrile group and (b) additions to the double bond. In the hydrolysis
method, acrylonitrile is hydrolyzed to ammonia and acrylate ion by a strong base.
The resulting ammonia can be determined colorimetrically by the Nessler's method
(Aarato and Bittera, 1972; AIHA, 1970) or by NaOCl and Na-salicylate in the
presence of Na-nitroferricyanide (Rogaczewska, 1976). Alternatively, the
liberated ammonia can be determined by titrimetric method (Gunther and Blinn,
1955).
Colorimetric, titrimetric, and thin layer chromatographic procedures have
been developed based upon the addition reaction of acrylonitrile. In one colori-
metric procedure, acrylonitrile is brominated with (Krynska, I960; Russkikh,
1971; Lawniczak, 1977) or without (Russkikh, 1973; Nazarova and Nakrap, 1978)
u.v. light. The excess bromine is neutralized and the cyanogen bromide that is
formed is allowed to react with a benzidine-pyridine solution to form a colored
complex (Nazarova and Nakrap, 1978). In another method, the acrylonitrile
complex that is formed with pyridine in the presence of a basic hypochlorite
solution at 60-65°C is measured at Mil nm (Hall and Stevens, 1977).
The titrimetric procedures have been used following the reaction of acrylo-
nitrile with excess Na^SO, (Burkart £t al., 1961; Terent'ev and Obtemperanskaya,
1956; Taubinger, 1969), thioglycolic acid (Stefonescu and Ursu, 1973; Covic-
Horvat Q a^., 1970), NaHSO, (Kostin and Vidanova, 1957), dodecanethiol (Roy,
1977), or lauryl mercaptan (Berck, 1975). The excess reagent is then back-
titrated with the appropriate titrant in the presence of an indicator or by the
potentiometric method.
A method based on the interaction of acrylonitrile with alkaline KMnO^
solution, which produces a change in the permanganate color, has been used to
determine the concentration of acrylonitrile. In this method, the concentration
4-24
-------
was determined by comparison of the color from a calibration curve (Gisclard
et al., 1958).
In the thin layer ehromatography (TLC) procedure, the mercuric acetate
adduct of acrylonitrile was separated by TLC on silica gel. The solvent system
used was described by Braun and Vorendohre (1963). In a recent method
(Plieninger and Shartna, 1978), the indole adduct of acrylonitrile was separated
by TLC procedure.
The instrumental techniques for identification and quantification have used
gas ehromatography, polarography, infrared and u.v. spectroscopy, and mass
spectrometry. Of these methods, mass spectrometric is the most unambiguous
method for the identification of acrylonitrile..
The vast majority of GC procedures have relied on flame ionization detection
(FID) . Electron capture detection resulted in one-fifth the sensitivity of FID
(Barrett, 197^, cited in Going eit al.. 1978). The use of nitrogen/phosphorous
detectors (NPD) has resulted in a dramatic increase in the sensitivity of detec-
tion. The insensitivity of this detector toward compounds that do not contain
nitrogen and phosphorous eliminates interferences from many compounds
(DiPasquale et al., 1978). Neither carbon disulfide nor N-containing solvents
are suitable for NPD.
The selection of column packing materials for the separation of acrylo-
nitrile from interferences depends on the source of the sample. Numerous packing
•
materials have been used in the past. Based on their chemical characteristics,
the stationary liquid or solid phases that have been used for the analysis of
acrylonitrile can be summarized as follows:
Polyglycol: (a) Carbowax-1500 (Babina, 1979; DiPasquale et al., 1978;
Steinchen, 1976; Gawell, 1979) (b) Carbowax-1510 (Chudy and Crosby, 1977)
(c) Carbowax-400 (LysyJ, I960; DiLorenzo and Russo, 1969) (d) Carbowax-20M
(Chudy and Crosby, 1977; DiPasquale et al., 1978; Gawell, 1979) (e) SP-1000
(Marano et al.. 1978; Gagnon and Posner, 1979) and (f) Di-glycerol
(Ustinovskaya et al., 1977)
4-25
-------
Hydrocarbon: (a) Apiezon (Babina, 1979) (b) Tween-80 (Chopra et al., 1978)
Esters and Polyesters: (a) Polyethylene glycol adipate (Panova et al.,
1969; Kleshcheva et al.., 1971; Markelov and Semenenko, 1976) (b) Neopentyl
glycol succinate (Kleshcheva et al., 1971; Korzhova et al., 197*0 (c) Tri-
ethylene glycol butyrate (Pokrovskaya and Frolova, 1969) (d) Polyethelene
glycol succinate (Lysyj, I960) (e) Dioctyl phthalate (Nestler and Berger,
1965)
beta-beta'-oxydipoprionitrile (Pokrovskaya and Frolova, 1969; Reichle and
Tengler, 1968; DiLorenzo and Russo, 1969)
Silica gel ASK .(Ivanenko and Lukashevaskaya, 1976)
Porous Polymers: (a) Porapak Q (Balak et al., 1977; Campbell and Moore,
1979) (b) Porapak QS (Brown et al., 1978) (c) Porapak N (Russell, 1975;
Tanaka e_t al.., 1975) (d) Porapak S (Brown et al_., 1978) (e) Chromosorb 101
(McNeal et al., 1979; Brown et al., 1978) (f) Chromosorb 102 (Marano et al.,
1978; DiPasquale et al., 1978) (g) Chromosorb 101 (Going et al., 1979) (h)
Chromosorb 108 (McNeal et al., 1979)
Methylsilicone: (a) SE 30 (Berck, 1965) (b) DC-200 (Beaumont and Garrido,
1979)
Aminoalcohol: THEED (Hughes and Horn, 1977)
Tetracyanoethyl: Pentaerytritol (Ustinovskaya e£ al.. 1977; Deur-Siftar
and Svob, 1976; Marrs et al., 1978)
The polarographic method was used most extensively in the past for the
determination of acrylonitrile. Tetramethyl ammonium iodide (Sevest'yanova and
Tomilov, 1963; Gorokhovskaya and Geller, 1962; Uhde and Koehler, 1967; Chao and
Ch'en, 1966; Deus and Hamner, 1957; Lezovic and Singliar, 1977; Mekhtiev et al.,
1968; Klyaev et^ al., 1966; Rogaczewska, 196H), tetramethyl ammonium hydroxide
(Berck, 1962; Sevast'yanova et al_., 1966), and LiCl (Sevest'yanova and Tomilov,
1963; Bogaczek and Joworski, 1970) have been used as the supporting electrolyte
for the dropping mercury electrode. Both standard calomel and silver were used
as the reference electrode. A continuous polarographic method has been used for
the determination of acrylonitrile in industrial streams (Bogaczek and Jaworski,
1970).
Spectroscopic methods using both infrared and ultraviolet techniques have
been used for the detection and quantification of acrylonitrile. The sensitivity
4-26
-------
of acrylonitrile determination by the earlier IR methods was rather poor
(Scheddel, 1958; Karaenev et al., 197 *0 . In recent years, however, the use of
multiple reflection, which, in essence, has the effect of increasing the cell
path-length, has increased the sensitivity significantly (Kurapov £t al^., 1977;
AIHA, 1970; Beaumont and Garrido, 1979). The same principle has been applied for
the continuous monitoring of acrylonitrile at 10.5 run with a portable infrared
analyzer (Jacobs and SyrJala, 1978).
An infrared laser technique, called laser stark spectroscopy, that applies
electric fields to perturb the molecular rotational energy levels has been used
to enhance and modulate the absorption of acrylonitrile. Using the P(28) line
from a CO. laser and a MO cm IR cell, the method was shown to detect 0.03 ppm of
acrylonitrile in air (Sweger and Travis, 1979).
Acrylonitrile in solution has been determined by u.v. absorption at 195 run
(Petrova e£ al., 1978) and at 210 nm (Brieger £t al., 1952).
Mass spectrometry alone is rarely employed for the quantification of
acrylonitrile. It is usually used as a confirmatory technique. In one study,
however, it was used for monitoring acrylonitrile concentration in process
streams (Thomson, 197*0 . In combination with GC, mass spectrometry was used for
confirmatory identification (Grote et al.. 1978; Marano et al.. 1978; Tanaka
et al., 1975; Going et al., 1979; McNeal et al., 1979). The use of multiple ion
monitoring mode increases the detection limit five-fold over full mass scan mode
(McNeal et al., 1979).
Finally, detector tubes for area monitoring (Kobayashi, 1956) and gas
badges for personnel monitoring (Silverstein, 1977) of acrylonitrile have been
proposed.
4-27
-------
5. SOURCES IN THE ENVIRONMENT
5.1 PRODUCTION PROCESSES
Acrylonitrile can be produced by the following methods:
(a) Oxidation of propylene in the presence of ammonia (ammoxidation of
propylene) using either a bismuth phosphomolybdate or uranium-base
catalyst;
(b) Addition of hydrogen cyanide to acetylene using a cuprous chloride
catalyst;
(c) Catalytic reaction of propylene with nitrous oxide;
(d) Reaction of ethylene oxide with hydrogen cyanide, followed by cata-
lytic dehydrogenation of ethylene cyanohydrin; and
(e) Ammoxidation of propane.
Processes (a) through (d) have been used for the commercial production of
acrylonitrile, and process (e) has been studied on a pilot scale (Hughes and
Horn, 1977). Since 1971, however, process (a) (the ammoxidation of propylene) is
the only process that has been used commercially in the United States. The
process is patented by the Standard Oil Company (SOHIO) and is known as the SOHIO
process.
5.2 ACPYLONITRILE PRODUCERS
The producers of acrylonitrile monomer in the United States are given in
Table 5-1, while distribution for 1980 and growth projections through 1981
figures are-given in Table 5-2.
5-1
-------
Table 5-1. Producers of Acrylonitrile in the United States (Grume, 1982)
Producer Capacity, Mg x 10^
American Cyanimid
Westwego, LA 123
DuPont
Beaumont, TX 159
DuPont
Memphis, TNa 136
Monsanto
Alvin, TX 113
Monsanto
Texas City, TX 191
Vistron
Lima, OH 136
Vistron
Victoria, TX 270
TOTAL 1,128
aDupont has announced plans to close this plant (C and EN Oct. 11, 1982)
Table 5-2. Distribution of Acrylonitrile in 1980 and Projected Growth
Through 1984 (Chemical Marketing Reporter, 1980)
Acrylonitrile Projected Annual
Distribution ( Mg x 103)a Growth
Consumption
Imports Negligible
Exports 198 Decline
Production at Vistron, Victoria, Texas, is not included
5.3 ACRYLONITRILE USES
Acrylonitrile is used primarily as a raw material in the synthesis of
acrylic and modacrylic fibers, ABS and SAN resins, nitrile rubbers, adipon-
itrile, acrylamide, and barrier resins. Other miscellaneous uses include the
production of fatty amines and their derivatives, cyanoethylation of various
5-2
-------
Table 5-3. Primary Uses for Compounds Synthesized from Arylonitrile-Containing
Compounds (Suta, 1979)
Compound
Uses
Acrylic and Modacrylic Fibers
ABS Resin
SAN Resin
Nitrile Elastomers
Adiponitrile
Acrylamide
Nitrile Barrier Resins
More than 60$ of these fibers is used
in apparel. Carpeting is the second
largest use. Home furnishing uses
include blankets, draperies, and upholstery.
Industrial uses include sandbags, filter
cloths, tents, and tarpaulins. The
fibers are also used in synthetic hair
wigs.
Its major markets are pipes and pipe
fittings, and automotive components.
Other important markets are large appliances,
housing for business machines and telephones,
recreational vehicle components, toys,
sporting goods, sheeting material
for luggage, and food containers.
Its primary uses are for drinking tumblers
and other houseware items, for automobile
instrument panels, instrument lenses,
and food containers.
Its major uses are in rubber hose,
seals, gaskets, latex, adhesives, polyvinyl
chloride blending, paper coatings,
and pigment binders.
It is hydrogenated to hexamethylenediamine,
which is used to produce nylon.
Its largest use is in the production
of polyacrylamides for waste and water
treatment flocculants. Other acrylamide
products are used to aid sewage dewatering,
and for papermaking strengtheners and
retention aids.
They are used in the manufacture of
non-beverage containers for glue, nail
polish, correction fluid, air freshener,
contact lenses, tooth brushes, and
combs (Miller and Villaume, 1978).
5-3
-------
alcohols and amines, fumigant formulations, as an absorbent, and as an anti-stall
automotive additive (Miller and Villaume, 1978). A flow diagram summarizing
direct and indirect uses of acrylonitrile is given in Figure 5-1. The primary
uses of the compounds that are synthesized from acrylonitrile-containing
compounds are presented in Table 5-3-
5.H CONSUMPTION OF ACRYLONITRILE BY PRODUCT
A breakdown of acrylonitrile consumption and projected growth of products
using acrylonitrile in their manufacture is given in Table 5-H.
Table 5-U. Acrylonitrile Consumption and Project Growth of Products
Using Acrylonitrile
Product
Acrylic and
modacrylic fibers
ABS and SAN resins
Nitrile elastomers
Adiponitrile
Acrylamide
Nitrile barrier
resins
Other
1977
Consumption
(Mg x 103)a
331
142°
24
73
24
9
33
Projected
Annual
Growth 1 980
Through Consumption
1982($)a (Mg x 103)b
4.5-5.5 345
7.5-9.5 172
2.0-3.0
10.5-12.5
8.0-10.0
12.0
4.0-6.0 147
Projected
Annual
Growth
Through
1984
None
Some Growth
—
—
—
__
Decline
aSuta, 1979
bCMR, 1980
C126 x 103 Mg for ABS and 16 x 103 Mg for SAN resins
The U.S. manufacturing plants that use acrylonitrile in production (except
for the plant producing adiponitrile) are listed in Tables 5-5 through 5-8. The
only U.S. producer of adiponitrile from acrylonitrile is Monsanto. Their
5-4
-------
ACRYLONITRILE-
FIBERS•
NITRILE RUBBER
& LATEXES
ACRYLAMIDE
ADIPONITRILE-
GLUTAMIC ACID
ABS & SAN RESINS
BARRIER RESINS
CYANOETHYLATION
FATTY AMINES
ACRYLIC
MODACRYLIC
POLYACRYLAMIDE -
ADHESIVES & FILMS
DYES
PHOTOGRAPHIC EMULSIONS
INTERNAL PLASTICIZERS
NYLON
FLOCCULENT
SIZING PAPER
PLASTICS
THICKENING AGENT
-IMONOSODIUM GLUTAMATE
Figure 5-1. Flow Diagram for Acrylonitrile Usage
(NIOSH, 1977c)
5-5
-------
Table 5-5. Producers of SAN and ABS Resins (Crume, 1982)
Producer Capacity (Mg x
ABTEC (Mobay)
Louisville, KY 31.8
Borg-Warner
Ottawa, IL 105
Washington, WV 136
Port Bienville, MS NA
Dow
Allyns Point, CT 29.5
Midland, MI 65.9
Pevley, MO 29.5
Torrance, CA 3*4.1
Irontown, OH 29.5
Monsanto
Addyston, OH 159
Muscatine, IA 56.8
Springfield, MA 13.6
USS Chemical
Scotts Bluff, LA 90.9
NA = Not Available
Table 5-6. Producers of Acrylic and Modacrylic Fibers (Crume, 1982)
Producer Capacity (Mg x 10^)
American Cyanamid
Milton, FL 59.1
Badische
Williamsburg, VA 36.8
DuPont
Camden, SC 77.3
Waynesboro, VA 61.1*
Tennessee, Eastman
Kingsport, TN 16.U
Monsanto
Decatur, AL 1H5
5-6
-------
Table 5-7. Producers of Nitrile Rubbers and Elastomers (Crume, 1982)
Producer Capacity (Mg x l(r)
Copolymer Rubber
Baton Rouge, LA 6.2
B.F. Goodrich
Akron, OH 1H.1
Louisville, KY 28.6
Goodyear
Akron, OH 7.5
Houston, TX 18.2
Uniroyal
Painesville, OH 16.4
Reichold
Cheswold, DE 10.14
Table 5-8. Producers of Acrylamide (SRI, 1978)
Producers Capacity (Mg x 10 )
American Cyanamid
Linden, NJ 36a
Westwego, LA
Dow
Midland, MI 23
Nalco
Garysville, LA U.5
aCombined total for both plants
5-7
-------
facility at Decatur, Alabama, produced 67 x 10^ Mg of adiponitrile in 1977 (SRI,
1978).
5.5 SOURCES OF EMISSIONS
Acrylonitrile emissions occur during: (1) monomer and polymer production;
(2) transportation; and (3) end product usage. These emission sources are
discussed below.
5.5.1 Monomer and Polymer Production
Acrylonitrile emission estimates for monomer, ABS-SAN resin, acrylic fiber,
nitrile elastomer, and adiponitrile production are given in Tables 5-9 through
5-13i respectively. (Other acrylonitrile emission sources are relatively minor
and are not included). These estimates are based on information provided by both
industry and EPA, and are included in a report prepared by Suta (1982). The
estimates assume full capacity; however, in recent years actual production
levels have been somewhat below full capacity. Actual production levels were not
used in the estimates because production levels vary considerably from year to
year. Also, some plants consider their production levels proprietary
information. The estimates of public exposure potential presented elsewhere in
this report are based on the emission estimates contained in the Suta (1982)
report.
EPA has independently calculated the emission estimates contained in the
Suta (1982) report (Crume, 1982). The current EPA estimates agree closely with
the Suta estimates for most acrylonitrile production plants. The current EPA
estimate for total acrylonitrile emissions from the entire industry differs from
the Suta total by less than +2$.
5-8
-------
Table 5-9. Estimated Atmospheric Emissions of Acrylonitrile from
Monomer Production Facilities (Suta, 1982)
Acrylonitrile Emissions
Producer (Mg/year)
American Cyanamid
Westwego, LA 9H.7
DuPont
Beaumont, TX 78.8
Memphis, TN 216
Monsanto
Alvin, TX 95.1
Texas City, TX 162
Vistron
Lima, OH 132
Victoria, TX 23-3
TOTAL 802
5-9
-------
Table 5-10. Estimated Acrylonitrile Emission Rates from ABS-SAN
Resin Production (Suta, 1982)
Acrylonitrile Emissions
Producer (Mg/year)
ABTEC (Mobay)
Louisville, KY 5.0
Borg Warner
Ottawa, IL 96.6
Washington, WV 657
Port Bienville, MS 1.3
Dow
Allyns Point, CT 7.9
Midland, MI 17.3
Pevley, MO 1.1
Torrance, CA 9.1
Irontown, OH 6.2
Monsanto
Addyston, OH 89.6
Muscatine, IA 365
Springfield, MA 10.0
USS Chemical
Scotts Bluff, LA 158
TOTAL 1,l|2U
Table 5-11. Estimated Acrylonitrile Emission Rates from Acrylic
Fiber Production (Suta, 1982)
Producer Acrylonitrile Emissions
(Mg/year)
American Cyanamid
Milton, FL 1H2
Badisohe
Williamsburg, VA 357
DuPont
Camden, SC 352
Waynesboro, VA 309
Tennessee Eastman
Kingsport, TN 25.0
Monsanto
Decatur, AL 91.0
TOTAL 1,276
5-10
-------
Table 5-12. Estimated Acrylonitrile Emission Rates from Nitrile
Elastomer Production (Suta, 1982)
Producer Acrylonitrile Emissions
(Mg/year)
Copolymer Rubber
Baton Rouge, LA H.3
B.F. Goodrich
Akron, OH 123
Louisville, KY 63.H
Goodyear
Akron, OH 55.2
Houston, TX 0.2
Uniroyal
Plainsville, OH HO.O
Reichold Cheswold, DE 8.1
TOTAL 295
Table 5-13. Estimated Acrylonitrile Emission Rate from Adiponitrile
Production (Suta, 1982)
Producer Acrylonitrile Emissions
(Mg/year)
Monsanto
Decatur, AL 59.0
5-11
-------
5.5.2. Emissions During Transportation
Estimates concerning the importance of potential acrylonitrile spills
during transportation-related accidents are presented in Table 5-11. These
estimates were originally developed to determine the relative costs of different
modes of transportation and were not intended to indicate the magnitude of
potential spills for the entire acrylonitrile industry. The estimates do suggest
the magnitude of potential spills from several modes of transportation, but under
highly idealized circumstances. (See, for example, the footnotes to
Table 5-14).
Some data on actual acrylonitrile spills have been reported to the Oil and
Hazardous Materials Spill Information Retrieval System (OHM-SIRS) of the EPA.
From August 1970 to July 1975, 12 acrylonitrile spills were reported to OHM-SIRS,
10 of which occurred during transport. Of these 10 spills, 7 occurred from tank
cars, 2 from barges, and 1 from a tank truck. However, EPA cautions that only 10
to 20$ of all spills are ever reported (Miller and Villaume, 1978). The Inter-
governmental Maritime Consultative Organization estimated that 11 tons of
acrylonitrile were discharged into the sea from transport and handling in 1970
(NAS, 1975, cited in Miller and Villaume, 1978).
5.5.3 Emissions from End-Product Usage
Another source of environmental contamination is from residual monomer
release during end-product usage. Monomer residues occurring in end-products
•
are given in Table 5-15. The level of acrylonitrile in fibers is so low that
handling of the fibers is not a likely source of acrylonitrile exposure. This
conclusion is supported experimentally by Finkel jet^ al. (1979). Even if the
product were heated, this would not result in a significant release of acrylo-
nitrile (Federal Register, 1978a). Acrylonitrile may possibly be leached from
fabrics during laundering; however, research has not been conducted in this area.
5-12
-------
Table 5-1^. Hazards of Acrylonitrile Transportation
(A.D. Little, Inc., 197H as cited in
Miller and Villaume, 1978)
Hazard Parameter
Barge
Truck
Rail
Spill Pool
Radius (meter)
Hazard Radius
(meter)
Hazard Area
(m2)
Relative Exposure ($)
Urban/Rural
Expected Number
of Annual Spills
Probability of Ignition
Following Spill
Expected Annual Number of
People Exposed
Urban/Rural
Expected Annual
Property Damage ($)r
61.0
122
H6,700C
5,U60C
8/92
17.1
38. «l
U,U50C
23/77
31.7
68.3
0.0117
0.30
0.063
0.25
13,1*00°
27/73
0.17
O.UO
0.008/0.OOU 0.010/0.002 0.16/0.016
Urban/Rural
f
Recurrence Interval
( Years)
129/55
85.5
160/20
15.8
2H23/252
5.8
Calculations are based upon the assumption that each mode of transportation
handles 100 percent of the quantity shipped, and that a total of 73,000 Mg
per year of acrylonitrile are shipped between two points.
Area affected by spills into water which ignite. Assumes entire spill
quantity contributes to burning pool.
c Area affected by spills on land which ignite. If no ignition occurs the
exposed land area is equivalent to the pool spill area (irR spill).
For spills into water which do not ignite. The water toxicity hazard dis
tance (meters) measured downstream from spill location for a 152 meter wide,
3.05 meter deep river flowing at 0.70 meters per second. Assumes vertical
dispersion rate at 0.30 meters per minute until uniform mixing is achieved
throughout the entire depth of the river. Thereafter, plug flow is assumed
with no synergistic or antagonistic reaction between the pollutant and the
receiving body of water. For this situation the entire spill quantity contributes
to water.
Expected number of people exposed annually and property damage is based
upon ignition of the flammable pool for both land and water based spills.
f
Average number of years between accidents.
5-13
-------
Table 5-15. Monomer Residue in End-Products of Acrylonitrile
Product Name
Usage
Monomer Residue
(ppm)
Reference
Acrylic and Modacrylic
Fiber
Hycar
Kralastic and Paracril
UCAR-380
UCAR-1358
Acrylamide Monomer
Polyacrylamide
ABS Resin
SAN Resin
SAN Resin
Fabric
Rubber
Resin
Latex
Latex
See Figure 5-1
See Figure 5-1
Packaging
Containers
Containers
<1
0-100
50
250
750
50-100
1
2*
3-7
2-5
Miller and Villaume,
1978
Miller and Villaume,
1978
Miller and Villaume,
1978
Miller and Villaume,
1978
Miller and Villaume,
1978
Miller and Villaume,
1978
Kearney, 1978
Brown e_t al. , 1978
McNeal et al . , 1979
Gawell, 1979
5-14
-------
Although no study has been made, it is possible that some acrylonitrile will
be released from automobile tires, since the rubber polymers in the tires contain
large amounts of monomer residue. Research by A.T. Kearney, Inc. (Kearney,
1978) indicates that in non-food contact, ABS/SAN containers do not release any
acrylonitrile under normal use; in contact with foods, however, these ABS/SAN
containers may release acrylonitrile into the foods. Brown et al.. (1978)
determined that SAN bottles (7 ppm residual monomer content) in contact with 3%
acetic acid at H9°C for 1 month releases 0.013 ppm of acrylonitrile to the acetic
acid. ABS resin (2H ppm residual monomer content) under the same conditions
released 0.283 ppm of acrylonitrile (Brown et al.. 1978) .
Acrylonitrile may be present as an impurity in products made from acrylamide
(Miller and Villaume, 1978). However, the extent to which these products release
acrylonitrile to the environment is not reported. Fumigant formulations
containing acrylonitrile were once used as pest control for residential
buildings, tobacco, grains, and nuts (Davis et al., 1973). However, today
fumigants containing acrylonitrile are no longer in use.
5.5.JJ Conclusions
The major sources of acrylonitrile emissions in the U.S. are monomer and
polymer production facilities. The estimated acrylonitrile emissions from these
facilities are shown below:
Estimated Acrylonitrile
Emissions (Suta, 1982)
Production facility (Mg/yr)
Monomer 802
ABS-SAN resin 1U2H
Acrylic and modacrylic fiber 1276
Nitrile elastomer 295
Adiponitrile 59.0
3856
5-15
-------
Although the relative importance of other potential sources of acrylonitrile
emissions is difficult to assess, these emissions are believed to be small
relative to monomer and polymer production.
5-16
-------
6. ENVIRONMENTAL FATE, TRANSPORT, AND DISTRIBUTION
the environmental fate of acrylonitrile in air, water, and soil, is dis-
cussed in the following sections. The discussion is based on only a few studies
that have been conducted in the field.
6.1 ATMOSPHERIC FATE, PERSISTENCE, AND TRANSPORT
Very few studies have been conducted to investigate the fate of acryloni-
trile under atmospheric conditions. Based on the similarity of the physical and
chemical properties of acrylonitrile and the olefins, however, it is possible to
predict the atmospheric fate of acrylonitrile from what is known about that class
of compounds. Like other olefins, acrylonitrile is expected to undergo both
chemical and photochemical reactions in the atmosphere. These reactions are
discussed individually in the following sections.
6.1.1 Atmospheric Chemical Reactions
Although no specific references are available, atmospheric oxidation reac-
tions typical of olefins may take place with acrylonitrile. For example, oxygen
atoms formed as a result of the photolysis of nitrogen dioxide in the atmosphere
usually add to the olefinic double bond. Oxygen atoms react with olefins more
rapidly than with other unsaturated aromatic and acetylinic hydrocarbons. This
addition reaction forms an excited epoxide that subsequently decomposes to allcyl
and acyl radicals (U.S. EPA, 1979).
Hydroxyl radicals, formed as a result of atmospheric photolysis of nitrous
acid and degradation of other free radicals, add to the double bond of the
olefins. The rate constant for this addition reaction is about 10 times greater
than for the atomic oxygen olefin reaction (Morris and Niki, 1971).
Atmospheric ozone is formed in significant quantities when nitrogen dioxide
levels in the atmosphere are about 25 times greater than nitrogen monoxide
6-1
-------
levels. Ozone, while not as strong an oxidizing agent as 0» or »OH radicals,
reacts with olefins at appreciable rates when ozone concentrations reach or
exceed 0.25 ppm. Ozone adds to the olefinic double bond forming an aldehyde and
a diradical. The diradical may further decompose or may participate in reactions
with 02, N02, and NO.
6.1.2 Photochemical Reactions
The photochemistry of acrylonitrile vapor at 213.9 nm was studied by Gandini
and Hackett (1978). The photolysis was shown to proceed via two molecular
elimination pathways, one yielding acetylene and hydrogen cyanide and the other
yielding cyanoacetylene and hydrogen. The quantum yields for the two processes
were determined to be 0.50 and 0.31, respectively. In the presence of such
photosensitizers as xanthene, triphenylene, benzophenone, acetophenone, fluore-
none, and dibromoanthracene, the major product of photolysis of acrylonitrile in
solution was shown to be 1,2-dicyanocyclobutane (Gale, 1971; Hosaka and
Wakamatsu, 1968) . The dicyanocyclobutane is not very stable, however, and it is
unlikely that the reaction will proceed in the gas phase.
6.1.3 Atmospheric Persistence and Transport
Only one study that experimentally investigated the atmospheric persistence
of acrylonitrile is available. Joshi (1977, cited in Suta, 1979) estimated the
atmospheric half-life of acrylonitrile to be 9 to 10 hours. An atmospheric half-
life of 9-10 hours is sufficiently long for aerial transport to play a signifi-
•
cant role in the distribution of acrylonitrile in the neighborhood of emission
sources. It has been calculated by Suta (1979) that when the average wind speed
is H meters/second, 86$ of the emitted acrylonitrile will survive at 30 km
downwind from the source, and 78$ will survive at 50 km downwind.
6-2
-------
6.2 FATE, PERSISTENCE, TRANSPORT, AND BIOACCUMULATION IN AQUEOUS MEDIA
6.2.1 Chemical Reactivity in Water
The chemical stability of acrylonitrile in water at different pH's was
studied by Going e_t al. (1979) . He spiked distilled water and Mississippi River
water with 10 ppm acrylonitrile. Prior to spiking, the pH was adjusted to 4 or 10
or left unadjusted. All samples were stored at room temperature in Teflon-capped
vials for 1, 6, and 23 days. There was no indication of sample decomposition in
distilled water even after 23 days at any of the tested pH values. The samples in
river water showed decomposition on storage. The sample with unaltered pH showed
complete decomposition after 6 days. The sample stored at pH 10 showed little
decomposition after 6 days but completely decomposed after 23 days. The pH *J
sample showed even less decomposition after 6 days and only 23$ decomposition
after 23 days. It is not certain whether the decomposition of acrylonitrile is
at least partly due to microbial effect. If so, the extremes of pH may have an
inhibitory effect on the microorganisms in the river water. Spiking sterilized
water with acrylonitrile and monitoring sample decomposition may provide an
answer.
Acrylonitrile, if present in surface waters that are used as sources of
drinking water, may react with chlorine or hypochlorite during the chlorination
step of the treatment process. It has been suggested that the reaction products
could be a mixture of OHCH2CHC1CN and C1CH2CHOHCN (Kondratenko et al., 1971).
The presence of detectable levels of acrylonitrile in previously aerated surface
water (used as a source of drinking water), however, is not very likely because
of rapid volatilization.
6-3
-------
6.2.2 Photochemical Reaction in Water
Another mode of acrylonitrile degradation may be photochemical reaction in
water; however, little is known regarding the photochemistry of acrylonitrile in
water in the concentration region likely to be present in natural water bodies.
6.2.3 Degradation of Acrylonitrile by Microorganisms
Limited data suggest that loss of acrylonitrile from water systems via
biological degradation can be expected. Both aerobic and anaerobic microo-
rganisms are capable of degrading acrylonitrile, especially acclimated micro-
organisms. The breakdown products of aerobic microorganisms may include ammonia
and acrylic acid (Mills and Stack, 1955), followed by nitrification of ammonia
(Chekhovskaya et al., 1966). The latter authors found that acrylonitrile at
concentrations of 50 ppm or higher may inhibit nitrification.
Mills and Stack (1955) suggested a mechanism for the biological oxidation of
acrylonitrile. Using microorganisms from the Kanawha River (WV) that had been
acclimated with acrylonitrile for 27 days, the Biological Oxygen Demand (BOD) of
acrylonitrile was measured. The rate of aerobic oxidation was quite rapid and
reached completion in five days. As shown in Figure 6-1A, about 70$ of the
acrylonitrile was degraded. From the nitrogen balance data, the authors sug-
gested that the biological oxidation of acrylonitrile proceeds by an enzyme-
catalyzed hydrolysis of the nitrile group to acrylic acid and ammonia.
The microbial fate of acrylonitrile in natural water was studied by Cherry
et al. (1956). Acrylonitrile (10 ppm) was added to filtered aerated water from
the Hackensack River (NJ). Nitrogen and phosphorus nutrients were added to the
water. The complete disappearance of acrylonitrile from water took about
20 days. Subsequent redosing with acrylonitrile reduced the degradation time.
These results are shown in Figure 6-1B. Similar results were obtained at 25 and
6-4
-------
496
DAYS OF INCUBATION
9 10
- SO
a.
a.
2*°
o
o
x
o
o
10
Figure 6-IB
REDOSE
REDOSE REDOSE
REDOSE
10
20 30 40
ELAPSED TIME (DAYS)
50
60
80
2Q°C
5°C
o
u uj
W.P 4O
•*
w 5 20
^5 o
° <"20i
Figure 6-1.
10
I
Figure 6-1C
I I
20 30
60
70 60
DAYS
90 100 110
Biological Oxidation of Acrylonitrile in Aqueous Systems
A (Mills and Stack, 1955)
B (Cherry £t al., 1956)
C (Ludzack £t al., 1958)
6-5
-------
50 ppm acrylonitrile; that is, the acclimated microorganisms degraded the
acrylonitrile more rapidly than did the unacclimated microorganisms.
Ludzack et al. (1958) also found similar results. These authors spiked Ohio
River water with 10 ppm acrylonitrile at 20°C. As shown in Figure 6-1C, there
was a lag period of about a week, followed by several days of rapid degradation
after which a plateau was reached. By day 22, another period of activity
occurred. Redosing this water with acrylonitrile produced no lag period and
plateau but produced rapid degradation of acrylonitrile by the already accli-
mated microorganisms. Evidently the degradation rate was temperature dependent;
when a sample was redosed at 5°C, the degradation rate was found to be slower
than that seen in the sample redosed at 20°C.
Ludzack e_t al. (1958) noted that acrylonitrile was degraded more rapidly by
microorganisms in Ohio River water than by microorganisms in aged sewage. They
also found that acrylonitrile was more resistant to biological degradation than
aceto-, adipo-, benzo-, and lacto-nitriles.
The aerobic degradation of acrylonitrile in water can also proceed via
activated sludge. Experiments conducted by Dow Chemical Company (NAS, 1975,
cited in Miller and Villaume, 1978) indicated almost complete degradation of
acrylonitrile to ammonia in 20 days. The effectiveness of acclimated activated
sludge for the rapid degradation of acrylonitrile in water was also shown by
Ludzack et al. (1961) .
The effectiveness of activated sludge systems for almost complete
biodegradation of acrylonitrile in industrial wastewater was also shown by
Kincannon and Stover, 1981 and Freeman e_t al. (1981) . The latter authors
demonstrated that the residue level of acrylonitrile in wastewater treatment
basins after aerated activated sludge treatment was below the detection limit
(<0.1 ppm).
6-6
-------
Kato and Yamamura (1976) discovered that aerobic microorganisms of the
genus Nocardia were capable of degradation of cyanides and nitriles. More than
90J of the acrylonitrile was degraded by these microorganisms.
The proceeding studies show that acrylonitrile can be degraded aerobically.
Ludzack et al. (1961), Lank (1969, cited in Miller and Villaume, 1978), and
Hovious et al. (1973, cited in Miller and Villaume, 1978) studied acrylonitrile
degradation under anaerobic conditions. Lank (1969, cited in Miller and
Villaume, 1978) found that acrylonitrile at a concentration of 10 ppm could be
treated by anaerobic digestion. Hovious et al. (1973, cited in Miller and
Villaume, 1978), however, determined that, even at a concentration of 50 ppm,
acrylonitrile was inhibitory to some anaerobes. The inhibition was not complete,
so some residual activity remained. Acrylonitrile1s inhibition of anaerobic
digestion by microorganisms was also confirmed by Ludzack e_t al. (1961). These
authors recommended that the anaerobic digestion should not be used for treatment
of acrylonitrile-containing water.
6.2.1! Bioaccumulation in Water
A bioconcentration factor (BCF) relates the concentration of a chemical in
water to the concentration in aquatic organisms. It is important to determine
the BCF for acrylonitrile in aquatic organisms in order to evaluate the levels of
human intake of acrylonitrile from this source and also to assess ecological
effects. There are a number of theoretical correlation equations that have been
established to relate BCF to either the octanol-water partition coefficient or
the water solubility of the chemical. These equations are given below:
log BCF = 0.76 log KQW - 0.23 (Veith et al., 1979)
log BCF = 0.512 log KQW + 0.12H (Neely et al., 197U)
log BCF = -0.508 log S + 3.^1 (Chiou et al., 1977)
6-7
-------
where K = partition coefficient of the chemical between octanol and water, and
S = water solubility of the chemical expressed in unol/1.
The above equation of Veith et al. (1979) is applicable for the whole fish;
whereas the equations of Neely et al. (1974) and Chiou et al. (1977) are appli-
cable for fish muscle only.
If the values for KQW and S for acrylonitrile are assumed to be 0.12 (Leo
et al., 1971) and 1.385 x 10 umol/1 (Groet, 1978), respectively, the theoretical
values for BCF can be calculated to be 0.1 (equation of Veith et al.. 1979) for
the whole fish and O.U (equation of Neely e_t al., 1974) and 2.0 (equation of
Chiou et al., 1977) for fish muscle.
U.S. EPA (1978a) experimentally measured the steady-state BCF for acrylo-
nitrile in bluegills containing about 4.8$ lipids. The experimental value was
48. The BCF for lipid-soluble compounds is proportional to percent lipids (U.S.
EPA, 1979). The weighted average lipid content in all the aquatic foods consumed
by an individual in the United States was calculated to be 3.0$ (U.S. EPA, 1979) .
An adjustment factor of 3.0/4.8 = 0.625 was used to adjust the measured BCF from
the 4.8$ lipid of the bluegill to the 3.0$ lipids that is the weighted average
for consumed fish and shellfish. Thus, the weighted average bioconcentration
factor for acrylonitrile in the edible portion of all aquatic organisms consumed
by an individual in the United States was calculated to be 48 x 0.625 = 30 (U.S.
EPA, 1980).
It can be concluded from the above discussions that the experimental BCF for
acrylonitrile in consumable aquatic foods is about two orders of magnitude higher
than the calculated value.
6.2.5 Transport in Water
Few studies are available that investigate the transport of acrylonitrile
in water other than the river water studies discussed previously. The partial
6-8
-------
vapor pressure of acrylonitrile in its water azeotrope is 80 mm Hg at 20°C
(Miller and Villaume, 1978); this pressure is significant enough to cause
evaporation of acrylonitrile from water. Using the method of Billing (1977) , the
half-life of evaporation of acrylonitrile from water with an assumed depth of
1 meter can be calculated to be 795 minutes. It should be mentioned, however,
that no experimental data are available to demonstrate this transport possi-
bility and as discussed in Section 3.1.8 the calculated evaporative half-life may
not accurate for acrylonitrile in natural aquatic systems.
The removal of acrylonitrile from an aquatic system will take place through
various chemical (photochemical, oxidative etc.), biologicaland physical
(evaporative, adsorptive etc.) processes as discussed in subsections 6.2.1,
6.2.2, 6.2.3 and this subsection. The percent of acrylonitrile removal through
each of the processes will depend on the characteristics of the aquatic system.
For example, in biologically active systems <0.1$ of the acrylonitrile was found
to be removed from the aquatic media due to volatilization (Freeman et al., 1980;
Kincannon and Stover, 1981). The calculated half-life value given above is only
for the fraction of acrylonitrile that is removed from the water by the evapora-
tive route.
The observations from the two studies of accidental spills of acrylonitrile
can be used to provide further insight into the transport of acrylonitrile from
other medium into water (see Section 7.3.1). Both these spill incidents show
that it is possible to transport acrylonitrile from contaminated land to surface
waters (since the medium of land to water transfer was percolated water). In the
absence of evaporative effect, acrylonitrile can be expected to have a long
persistence in water. The bacterial decomposition of acrylonitrile in soil will
probably be of little importance in cases of spills because the toxic effect of
6-9
-------
the large spills on the bacteria. If the spill occurs during winter, the low
temperature of soil will further decrease the influence of biodegradation.
6.3 FATE, PERSISTENCE, AND TRANSPORT IN SOIL
Few data are available on this subject. Acrylonitrile can be degraded by
soil fungi (Giacin et al., 1973). Fungi capable of acrylonitrile biodegradation
included Penicillium, Aspergillus. and Cladosporium species (Giacin et al..
1973)• The products of decomposition were probably carbon dioxide and ammonia.
Although other microorganisms slowly degraded acrylonitrile, best results were
obtained with soil fungi. The microbe Nocardia rhodochrous LL100-21 slowly
degraded acrylonitrile, but the rate of degradation increased with added acetate
(DiGeronimo and Antoine, 1976).
6-10
-------
7. ENVIRONMENTAL LEVELS AND EXPOSURE
7.1 ENVIRONMENTAL LEVELS
Acrylonitrile levels in occupational atmospheres (Marrs et al., 1978;
Sakurai et aK, 1978; Martin, 1978) and in industrial point sources (Hughes and
Horn, 1977; Hollingsed, 1978, cited in Miller and Villaume, 1978; Suta, 1979)
have been determined. One published study conducted by Midwest Research
Institute (Going et al.. 1979) has determined environmental levels of
acrylonitrile. This study was designed to determine the levels of acrylonitrile
in ambient air samples, surface waters, and soils and sediments around
acrylonitrile and acrylonitrile polymer manufacturing facilities.
PEDCo Environmental has recently completed a study monitoring levels of AN
near U acrylonitrile producing or consuming plants. This study is yet
unpublished. Preliminary results indicate acrylonitrile to have been measured
at all four locations and, in general, levels around user plants were greater
than around producing plants.
7.1.1 Atmospheric Levels of Acrylonitrile Around Its Major Production and
Usage Facilities
The atmospheric levels of acrylonitrile around manufacturing facilities are
given in Table 7-1. The samples were collected by passing the air through
activated carbon over a period of approximately 21 hours for all stations, except
Monsanto in Decatur, Alabama, where two 21-hour samples were collected. The
acrylonitrile from the activated carbon was desorbed by carbon disulfide and
analyzed by gas chromatography with a flame ionization detector. The average
recovery of acrylonitrile determined with spiked samples was 63^ (Going et al.,
1979) .
The values for maximal average concentrations given in Table 7-1 have been
derived in the following manner: The sum of all the determined concentrations
7-1
-------
Table 7-1. Atmospheric Monitoring Data for Acrylonltrile (Going et al_., 1979)
Site
American Cyanamld,
New Orleans, LA
American Cyanamld,
Linden, NJ
Monsanto,
Texas City, TX
Monsanto,
Decatur, AL
DuPont,
Lugoff, SC
DuPont,
Waynesboro, VA
Borg -Warner,
Washington, WV
Ooodrich,
Louisville, KT
Monsanto,
Addyston, OH
Unlroyal,
Paneavllle, OH
Vlstron,
Lima. OH
Product
Acrylonitrlle
Acrylonitrlle
Acrylonltrlle
Aoryllo and
Modacryllc Fibers
Acrylic Fibers
Acrylic and
Modacryllc Fibers
ABS/SAN Resins
Nltrlle Elastomers
> ABS/SAN Resin
ABS/SAN Resins
Nltrlle Elastomers
Acrylonltrlle
> Acrylamlde
Distance (km)
of Collection
from Plant3
0.6-1.8
0.5-1.8
0.7-2.6
1.3-5.0
0.7-2.2
0.3-0.9
0.5-1.3
0.4-3.0
0.3-1.0
0.3-0.9
0.2-0.6
No. of Samples
Collected0
5
6
7
5
8
6
4
4
5
5
5
High
13.6
15.9
8.9
4.2
1.1
7.0
325.0
4.3
1.1
3.1
148.0
Concentration of Acrylonltrlle( iig/n^)
Low Maximal Average6
<0.1 2.6
<0.1 3.2
<0.3 3.2
<0.1 1.2
<0.1 0.3
<0.2 1.3
<0.2 84.0
<0.2 1.2
<0.2 0.5
<0.1 1.0
<0.2 3«.«»
*Tne distances of sample collection points are obtained from Suta, 1979.
Soaetlnes replicate samples were collected from the same sampling point.
C3ee text for explanation.
-------
has been divided by the number of determinations. When the determined concentra-
tions were less than the detection limit, the data used for averaging are the
values at the detection limit.
The highest individual concentration from this monitoring data was
325 ug/nr and the lowest was <0.1 ug/m^ (Going et al., 1979). The recorded
concentrations depend greatly on the meteorological conditions, the production
stage within the plant at the time of sampling, and the presence of emission
control devices in the plant. This is reflected in the high maximal average
acrylonitrile level (8H ug/nr) in one plant and a low level (0.5 ug/nr) in
another plant, even though both produced ABS/SAN resins. Such high and low
atmospheric levels of acrylonitrile are also reflected in the data for other
plant emissions, all manufacturing the same acrylonitrile (see Table 7-1). A
comparison of the experimental monitoring data (Going e_t al., 1979) with the
dispersion modeling data of Suta (1979) is given in Table 7-2.
The chemical analysis technique used for generation of the data presented in
Table 7-2 have since been improved. The extraction of the collected
acrylonitrile was performed for the monitoring data using carbon disulfide. The
present accepted extraction procedure uses a mixture of carbon disulfide and
acetone (19:1) which improves recovery (95$) (Gagmon, 1979) over that of carbon
disulfide alone (60-75$) (Silverstein, 1977).
A comparison of the dispersion modeling data (Suta, 1979) with the actual
monitoring data (Going et aU, 1979) shows that, although the difference between
the experimental concentrations and the concentrations derived from dispersion
modeling (Suta, 1979), on the average, was about 20$, 90$ of the individual
values had much higher variations. In many instances, the agreement between the
two were poor. Therefore, the need for more experimental monitoring data cannot
be overemphasized.
7-3
-------
Table 7-2. Comparison of Monitoring and Dispersion Modeling Data
(Suta, 1979)
Plant/Location Distances
(km)
Acrylonitrile Concentration (mg/m^)
Monitoring
Dispersion
Modeling0
American Cyanamid
New Orleans, LA
American Cyanamid
Linden, NJ
Monsanto
Texas City, TX
Monsanto
Decatur, AL
DuPont-May
Cam den, NJ
DuPont
Waynesboro, VA
Borg-Warner
Washington, WV
B£FJl Goodrich
Louisville, KY
Monsanto
Addyston, OH
Uni royal
Plainsville, OH
Vistron
Lima, OH
0.50-0.99
1.00-1. 49
1.50-1.99
0.50-0.99
1.00-1.49
1.50-1.99
0.50-0.99
1.00-1.49
1.50-1.99
2.00-2.49
2.50-2.99
1.00-1.49
1.50-1.99
2.00-2.50
2.50-4.99
5.00-5.49
0.50-0.99
1.00-1.49
1.50-1.99
2.00-1.50
0.30-0.49
0.50-0.99
0.50-0.99
1.00-1.49
0.30-0.49
0.50-1.99
2.00-2.49
2.50-2.99
3.00-3.49
0.30-0.49
0.50-0.99
1.00-1.49
0.30-0.49
0.50-0.99
0.30-0.49
0.50-0.99
1.3
0.1
0.1
0.5
6.0
0.7
2.4
nd
3.8
0.9
5.2
1.2
2.3
0.8
nd
0.2
0.7
0.3
0.1
0.2
3.6
0.2
157.6
0.3
2.3
nd
0.2
nd
0.2
0.2
O.U
1.1
1.3
0.7
43.4
0.2
H.3
2.5
1.9
d
d
d
6.5
-
2.5
1.8
1.3
21.0
9.3
7.2
-
3.1
7.4
2.3
1.3
1.1
5.2
3.4
42.1
30.4
7.2
-
1.7
-
1.0
5.9
3.9
2.8
1.2
0.9
10.6
5.2
nd = no data
Estimated distance from the acrylonitrile production within the plant.
Average of all monitoring stations within the indicated distances.
cEstimated concentrations at the midpoint of the distances.
TJispersion modeling estimates were not made for acrylamide plants.
7-4
-------
It is interesting to note that the atmospheric acrylonitrile level near the
American Cyanamid Plant in Linden, New Jersey, which produces only acrylamide, is
comparable to that near plants that manufacture other products derived from
acrylonitrile. In deriving the sources of emissions (see Section 5.5.2), Suta
(1979) made the assumption that acrylamide production is a negligible source of
acrylonitrile emission.
7.1.2 Acrylonitrile Levels in Surface Waters
The acrylonitrile monitoring data for surface waters is given in Table 7-3.
These data were obtained by Going £t al. (1979). Whenever possible, grab samples
were collected upstream and downstream of the plant discharge points. In some
instances, discharged wastewater from the plants were collected for analysis.
Two complementary techniques, azeotropic distillation and purge-trap, were
used for reducing interference and concentrating acrylonitrile in the samples.
The method used for quantification was GC with a Hall nitrogen-selective detec-
tor. Samples that appeared to contain acrylonitrile were confirmed by GC-MS
analysis. Analytical quality assurance was done by spiking and analyzing field
samples (Going et^ al., 1979)•
The two highest levels of acrylonitrile shown in Table 7-3 were obtained
from the Monsanto plant in Decatur, Alabama, and the Uniroyal plant in
Painesville, Ohio. These values were high, however, because these samples repre-
sent discharged wastewater prior to adequate dilution in surface water. The high
acrylonitrile content in the discharged wastewaters is an indication that an
effective control of these wastewaters is necessary to minimize the pollution of
surface waters.
7-5
-------
Table 7-3- Acrylonitrlle Monitoring Data for Surface Waters (Going e£ al., 1979)
~o
CTi
Site
Product
City, TX
Tennessee River near Monsanto, Decatur, AL
Wateree River near DuPont, Lugoff, SC
South River and Discharge Point at DuPont,
Vaynesboro, VA
Ohio River near Borg-Warner, Washington, WV
Ohio River near BIPl Ooodrlch, Louisville, KY
Ohio River near Monsanto, Addyaton, OH
Influent and effluent from treatment facilities
from wastewater at Onlroyal, Palnesvllle, OH
Discharge point at Ottawa River at Tlstron,
Lima, OH
Acrylic and modacryllc fibers
Acrylic fibers
Acrylic and modacyllc fibers
ABS/SAN resins
Hltrlle elastomers and
ABS/SAN resins
ABS/SAN resins
Nltrlle elastomers
Acrylonitrile and acrylaolde
Concentration in water, ug/1
High
3500
19.7
n.d.
1.9
2.0
8.0
1300
n.d.
Low
Mississippi River near American Cyanamld,
New Orleans, LA
Arthur Hill near American Cyananld,
Linden, NJ
Texaa City Ship Channel near Mom ant 0, Texas
Acrylonitrile
Acrylamide
Acrylonitrile
n.d . <0.1
O.B n.d.
0.1 <0.1
0.1
9.3
*n.d. t not determined
-------
7.2 ACRYLONITRILE LEVELS IN SOILS AND SEDIMENTS
The environmental levels of acrylonitrile in a few soil and sediment samples
are shown in Tables 7-1* and 7-5. These data were obtained from the investiga-
tions of Going et al. (1979).
The soil samples were collected from the air sampling locations. The
collection of sediment samples was restricted by the accessibility of the sedi-
ments from the waterbody. The methods of analyses were the same for both soil
and sediment samples. The sediments free from excess water and the soil samples
were ultrasonically agitated with water. In most analyses, the water extracts
were directly injected into the gas chromatograph equipped with a Hall nitrogen-
selective detector. One sample each from soil and sediment was further purified
and concentrated by the purge-trap technique before GC injection. The recoveries
of acrylonitrile from the samples were not determined.
It is obvious from Tables 7-^ and 7-5 that, with the exception of one soil
sample, the level of acrylonitrile in all other samples was below the detection
limit of the method used. This may be expected in view of the relatively high
water solubility and high volatility of acrylonitrile. The detection limit,
however, could have been lowered either with the purge-trap or azeotropic distil-
lation of the water extract.
7.3 ENVIRONMENTAL EXPOSURE
Population exposure from environmental acrylonitrile emissions can take
place through four principal sources: (1) industrial emissions in air;
(2) drinking water; (3) consumed foods; and (1) spillage during transportation.
The exposure from each of these sources is discussed below.
7-7
-------
Table 7-4. Acrylonitrile Monitoring Data for Sediments (Going .et al., 1979)
Site Concentration (ug/kg)'
Mississippi River near American Cyanamid, <0.5
New Orleans, LA
Tennessee River near Monsanto, Decatur, AL <50
Wateree River near DuPont, Lugoff, SC <50
South River near DuPont, Waynesboro, WV <50
figures were the lowest detection limit
Table 7-5. Acrylonitrile Monitoring Data for Soils (Going et al., 1979)
Site Concentration (ug/kg)a
American Cyanamid, New Orleans, LA =0.5
American Cyanamid, Linden, NJ <50
Monsanto, Texas City, TX <100
Monsanto, Decatur, AL <50
DuPont, Lugoff, SC <50
DuPont, Waynesboro, VA <50
Borg-Werner, Washington, WV <50
B£F£ Goodrich, Louisville, KY
-------
7.3.1 Exposure From Air Polluted by Industrial Sources
The total number of people expected to be exposed to different levels of
acrylonitrile concentrations from different industrial sources was calculated by
Suta (1979) and is given in Table 7-6.
The estimated values were derived on the basis of a dispersion modeling from
the emitted acrylonitrile concentration values and estimated population density
around the plants. The estimated exposure values were determined for people
residing within 10 concentric rings (of various radii ranging from 0 to 30 km)
about each plant.
The estimated exposures in Table 7-6 are somewhat underestimated for two
reasons. First, the exposures beyond 30 km were not included in all the calcula-
tions. Second, the exposure from acrylamide plant emissions was ignored, even
though the experimental data from Going e_t al. (1979) indicated such exposures
might be significant.
7.3.2 Exposure From Drinking Water
Trace amounts of acrylonitrile have been detected in drinking water
(Kopfler et a^., 1976), although the amount was not quantitated. In the absence
of such data, it is impossible to evaluate the human intake from this source.
7.3-3 Exposure From Foods
The three possible sources of acrylonitrile exposure from foods are:
(a) fish and shellfish; (b) food containers and packaging materials; and
(c) foods fumigated with acrylonitrile-containing fumigants.
Edible aquatic organisms may bioconcentrate acrylonitrile from contaminated
waters. The weighted average bioconcentration factor for acrylonitrile in the
edible portion of all aquatic organisms consumed by Americans has been calculated
to be 30 (U.S. EPA, 1980).
7-9
-------
Table 7-6. Estimates of Population Exposures to Atmospheric Acrylonitrile from Specific
Emission Source Categories (Suta, 1982)
Average Annual AN
Concentration
( ug/m)
15.0
10.0
5.0 -
1.0
0.50 -
0.10 -
0.050 -
0.010 -
0.005 -
0.001 -
19.9
114.9
9.9
U.9
0.99
O.U9
0.099
0.0149
0.009
0.004
AN
Monomer
U.900
36,000
260,000
720,000
1,602,000
8,000b
28,000b
ABS/SAN
Resins
2,700
T4.000
53,000
238,000
382,000
1,1407,000
1456, 000b
1,701,000b
Acrylic
and
Modacrylic
Fibers
390
5,000
mo, ooo
UUO.OOO
320,000
26,000b
15,000b
Nitrile
Elastomer
18,000
146,000
1421,000
1425,000
1,25*4,000
265,000b
292,000b
Adiponitrile
22,000
20,000
78,000
ob
ob
Total People
Exposed
2,659,000'
14,254,OOO1
9146,OOO1
2,721,000
aTo convert for ng/nr to ppb, multiply by O.i<6.
Exposures in these ranges are underestimates because calculations were only made for
exposures within 30 km of each plant.
120,OOO1
-------
The polymers and copolymers containing residual acrylonitrile monomer could
migrate from the food-contact items to the food itself. The amount of migration
depends on the residual monomer content in the polymer or copolymer, the time of
storage, and the temperature of storage. The effects of these factors on acrylo-
nitrile migration are shown in Table 7-7.
Since the monomer migration is substantial from ABS/SAN resins to ethanol,
the FDA currently does not permit the use of these containers for alcohol and
carbonated beverages. Gawell (1979), using SAN bottles containing 3 to 5 ppm
residual monomer, showed, however, that the migration in some samples of beer and
soft drinks amounted to < 0.005 mg/kg. The author did not specify the storage
conditions. FDA has determined that the migration of monomer from SAN resin
containers (3-3 ppm residual acrylonitrile) to the beverages could be as high as
11 ppb (Flood, 1980) after 96 days contact at 120°F.
Under FDA regulations, copolymers of acrylonitrile listed in Table 7-8 are
permitted in food-contact applications including food packaging, such as for
luncheon meats, peanut butter, margarines, fruit Juices, and vegetable oils.
Some acrylonitrile exposure may result from these sources, although it is diffi-
cult to quantify the amount. FDA has determined that the migration of the
monomer from the containers to vegetable oil and margarine could be as high as
37 ppb (Flood, 1980).
Foods that have been fumigated with acrylonitrile-containing fumigants pre-
sent a risk of acrylonitrile exposure. Fumigants containing acrylonitrile for
grain pest control have been voluntarily withdrawn from the market. Other foods,
such as walnuts, are no longer fumigated with acrylonitrile-containing fumi-
gants. The residue level of acrylonitrile in walnuts that have been fumigated
may range from 7.5-17.5 ppm after 2 days to 0-8.5 ppm after 38 days (Berck,
I960). Finally, acrylonitrile has been used as a fumigant for stored tobacco.
7-11
-------
Table 7-7. Acrylonitrile Migration Under Different Storage Conditions
Migrating Nature
Solvent of
Plastic
Residual
Monomer
Content
of
Plastic
(ppm)
Time of
Storage
(months)
Temp, of
Storage
(°C)
Acrylo-
nitrile
Found
(ppm)
Reference
3$ acetic acid SAN
3% acetic acid SAN
3$ acetic acid ABS
8% ethanol SAN
7
3
24
10
25
1
2
3
4
5
2
2
1
2
3
5
?
?
49
49
49
49
49
49
66
49
49
49
49
49
49
0.013
0.017
0.022
0.028
0.038
0.012
0.023
0.283
0.557
0.723
0.920
0.078
0.126
Brown _et al. ,
Brown ^t al. ,
Brown jjt al. ,
McNeal et al.
Brown et al. ,
McNeal et al.
McNeal et al.
Brown e^ al. ,
Brown e^ al. ,
Brown et^ al. ,
Brown e^ al. ,
McNeal et al.
McNeal et al.
1978
1978
1978
, 1979
1978
, 1979
, 1979
1978
1978
1978
1978
, 1979
, 1979
7-12
-------
Table 7-8. Amounts of Various Acrylonitrile Copolymers Used in
Food-Contact Applications (Troxell, 1980)
Copolymer
ABS Resins
ABS Resins
ABS Resins
ABS Resins
SAN Resins
Nitrile Elastomers/
Latexes
High Nitrile Resins
Polyvinylidine
chloride -
Acrylonitrile Resin
Use£
Amount used in 1977
(millions of pounds)
Refrigerator/freezer linings
Small appliances (motor housings,
bases for blenders and can
openers, etc.)
Packaging (sheet and film for
blister packs)
Margarine tubs
Drinking tumblers, blenders, Jars,
components of appliances, etc.
Hoses and paper coating
Mostly vegetable oil bottles
Cellophane > paperboard
125
10-15
15
<1
=50
=20
Some of the applications may result in incidental food contact.
7-13
-------
Guerin e_t al^ (1971*) have qualitatively determined the presence of acrylonitrile
in tobacco smoke, and, the amount has been determined to be 1-2 mg per U.S.
cigarette smoked (IARC, 1979). It is not clear whether the source of acrylo-
nitrile is from fumigation of the tobacco or from the combustion process itself.
7.3.1* Exposure From Spillage during Transportation
The expected numbers of people in urban/rural areas exposed annually from
spills during transportation of acrylonitrile by three modes of transportation
were calculated as follows: barge, 0.008/0.004; truck, 0.010/0.002; and rail,
0.16/0.016 (Miller and Villaume, 1978). These figures represent the cases in
which the entire commodity is transported by one mode. It can be concluded that
transportation of acrylonitrile by rail (10.4$ of overall shipment) poses the
greatest hazard and by barge (1.7$ of the overall shipment) the least hazard.
The preceding estimate was based on exposure due to ignition of the spilled
area and toxic exposure from contaminated surface water due to spill (see
Section 5.5.3)• However, contamination of the ground water and the subsequent
exposure from usage of the water as a source of drinking water was not considered
in the calculation.
Two recent cases of accidental spills leading to ground water contamination
show the possibility of population exposure from this source. A spill of
36,000 gallons of acrylonitrile onto farmland (Gilford, Inc., 2/22/77) caused
contamination of the nearby groundwater and a creek (Miller and Villaume, 1978).
Evidently, the acrylonitrile percolated through the soil into the groundwater
and the creek. For several months after the spill, the concentration of acrylo-
nitrile in the groundwater increased after it rained (Miller and Villaume, 1978).
Another spill of 20,000 gallons of acrylonitrile (near Mapleton, IL,
12/23/77) caused similar contamination of the groundwater and creeks located
near the spill (Miller and Villaume, 1978) . Monitoring data from five wells, all
7-14
-------
located within 100 feet from the spill site, showed high levels of acrylonitrile
in the water (16 mg/1 to 3520 mg/1) at the end of 108 days (Miller and Villaume,
1978). Acrylonitrile levels started decreasing about 170 days after the spill,
but acrylonitrile had not completely disappeared from the well water even after
351 days. Nine additional wells, located at an average distance of about 1000
feet from the spill site, showed no trace of acrylonitrile. Tap water at six
nearby residences (an average distance of 1150 feet from the spill site) con-
tained no acrylonitrile. A little Marsh Creek located about 750 feet from the
site of the spill showed 32 mg/1 of acrylonitrile at the end of 58 days after the
spill, but acrylonitrile finally disappeared after about 108 days.
7.3.5 Exposure From Thermal Degradation
Thermal decomposition of polymers containing acrylonitrile is another
source of acrylonitrile in the atmosphere. Pyrolysis of the following polymers—
polyacrylonitrile (Tsuchiya and Sumi, 1977; Guyot et al^., 1978), ABS/SAN resins
(Chaigneau and LeMoan, 1974), acrylonitrile-methacrylate copolymers (Guyot
et_ a^., 1978), vinyl chloride-acrylonitrile copolymers (Tanaka et al., 1975)—
produces hydrogen cyanide and acrylonitrile.
The composition of the gases evolved during pyrolysis and combustion of
polymers and copolymers of acrylonitrile depends on a number of factors, includ-
ing the nature of the polymer, gas composition, gas flow rate, and heating rate
of the flame (Sarofim e_t al., 1973) • Pyrolysis prevails at lower temperature and
decomposition requires higher temperature (Sarofim et^ al., 1973) • The pyrolysis
of SAN bottles with He and air at a flow rate of 1 to 3 1/min and heated at a rate
of 5°C/min to 18000°C/min produced the following major nitrogen components
(Sarofim et al., 1973) .
7-15
-------
HNC: 5-1* to U.3JE of total nitrogen in SAN
Acrylonitrile: 7.5 to 38.0$ of total nitrogen in SAN
The combustion of SAN bottles under a variety of conditions was studied in
detail by Kaiser and Bergman (1973). Their results are summarized below:
The combustion of SAN bottles alone with B3% excess air produced smoke,
odor, and toxic gases in appreciable amounts. The concentrations of two major
nitrogen components and hydrogen cyanide in the flue gases were the following:
Acrylonitrile: 270 ppm of total gases
Methacrylonitrile: 34 ppn of total gases
HCN: 1 ppm of total gases
From these results, the authors concluded that the burning of SAN bottles in
campfires, fireplaces, and outdoor trash burners could be hazardous.
Mixed with household refuse or the usual commercial or industrial plant
wastes, a few percent (2 to 4$) of SAN containers will burn in large incinerators
(about 1 ton/hr) without producing any detectable level of HCN or acrylonitrile.
As a matter of fact, other than a slight increase in NO production, the SAN
addition to household and commercial refuse will produce less harmful gases than
the burning of refuse without SAN. This is probably due to the fact that SAN in
refuse aids in the completion of combustion of the refuse.
As the size of the incinerator decreases, combustion of SAN with other
refuse will increase production of smoke, particulate matter, acrolein, acid
gases, HCN, NO , and total hydrocarbons.
Preliminary estimates of ambient concentrations and exposure levels of
acrylonitrile may be obtained from the modeling of the dispersion of acrylo-
nitrile as a result of pyrolysis and combustion. However, in the absence of any
such dispersion modeling data, it is difficult to estimate the exposure level
from this source.
7-16
-------
7.4 CONCLUSIONS
Most acrylonitrile exposures result from ABS/SAN resin and nitrile elas-
tomer production. The risk for population exposure from the acrylonitrile
sources is dependent on a number of factors, including the height of the release
point source; that is, the higher release point will result in greater dilution
of the pollutants at ground level and also spreads it over a larger area. In
general, for the same amount of emission and topography, acrylonitrile emission
points with lower elevation will result in higher ground level concentrations
than elevated acrylonitrile emission points. On the basis of the elevation and
other relevant factors, monomer and ABS/SAN resin production results in the
highest estimated total risk in terms of exposed population and exposure concen-
tration.
7-17
-------
8. BIOLOGICAL EFFECTS ON MICROORGANISMS
Loveless et al. (195*0 studied the effect of acrylonitrile on growth and
cell division of yeast (Saccharomyces cerevisiae) and bacteria (Escherichia
coli), as measured by dry weights and cell counts during the logarithmic growth
phase. Treatment with 1000 mg/1 reduced the growth of E. coli but had no effect
on cell size; this concentration inhibited growth and division in S. cerevisiae.
Treated cells were 170$ larger than control cells and weighed 52$ more.
Acrylonitrile was not toxic to the bacterium Nocardia rhodochrous at a
concentration of 10,000 mg/1 in both (DiGeronimo and Antoine, 1976). This
concentration of acrylonitrile supported growth of these bacteria as a sole
source of nitrogen, but not as a source of carbon.
Acrylonitrile has been shown to be inhibitory to anaerobic bacteria.
Ludzack et al. (1961) reported that acrylonitrile inhibited gas production by
anaerobic digester cultures which were dosed repeatedly with 10 to 10 mg/1
acrylonitrile. Hovious et al. (1973, cited in Miller and Villaume, 1979) found
that 50 to 100 mg/1 acrylonitrile inhibited gas production by anaerobic methano-
genic bacteria in proportion to dose.
Acrylonitrile was used as a fumigant to control mold growth on packaged
papads, an Indian bread (Narasimhan ejt al., 1972). Papads with 18 or 20$
moisture content were sealed in polyethylene bags, fumigated for ^8 hours with 32
or 6M mg/1 acrylonitrile, and checked for mold growth after one month. The
higher dose prevented mold growth at both moisture levels, whereas the lower dose
prevented mold growth only in the 18$ moisture papads. The species of molds were
not identified.
Some limited information concerning the effects of acrylonitrile on aquatic
microorganisms was provided by Cherry et al. (1956). Nutrient-enriched and
aerated river water was dosed with 10, 25, or 50 mg/1 acrylonitrile. Balanced
8-1
-------
populations of bacteria, diatoms, algae, protozoa, and rotifers developed at 10
and 25 mg/1, whereas fungal species predominated at 50 mg/1.
8-2
-------
9. BIOLOGICAL EFFECTS ON PLANTS
There is limited information concerning the effects of acrylonitrile on
plants.
Garrison (1978) studied the effects of acrylonitrile on cultured seagrass
(Ruppia maritima). Acrylonitrile was added to the water column to give concen-
trations ranging from 10 ng/1 to 10 g/1. Concentrations greater than 100 mg/1
totally inhibited photosynthesis and respiration, as measured by dissolved
oxygen changes. Lower concentrations had no effect on these processes. Although
all concentrations reduced the growth rate of shoots, the growth rate of roots
was stimulated at concentrations below 1 mg/1 acrylonitrile.
The effect of acrylonitrile on pea seedlings (Pisum sativum) was studied by
Burg and Burg (1967), who reported that 0.17 mM (ca. 9 ppm) acrylonitrile was
"toxic" (undefined effect) and that lower levels showed no effect on seedling
elongation.
Fumigant mixtures of 1:1 acrylonitrile:carbon tetrachloride had no adverse
effects on seed germination of beans, beets, corn, peas, lettuce, onions,
tomatoes, wheat, and oats when seeds were fumigated for 2^ to M8 hours at
concentrations ranging from 1 to 25 pounds per 1000 cubic feet (Glass and
Crosier, 19^9, cited in Miller and Villaume, 1978).
When acrylonitrile was added to aerated, nutrient-enriched river water,
balanced growth of bacteria, diatoms, algae, protozoa, and rotifers occurred at
10 to 25 mg/1 acrylonitrile (Cherry ejt al., 1956). At 50 mg/1, however, fungal
growth predominated.
Kihlman (1961) reported that 1 mM (53 PPm) acrylonitrile was not mutagenic
to broad bean root tips (Vicia faba). The details of this study are given in
Section 13.4.
9-1
-------
10. BIOLOGICAL EFFECTS ON DOMESTIC ANIMALS
No information was found concerning the effects of acrylonitrile on domes-
tic animals other than dogs and cats. The effects of acrylonitrile on these
animals are discussed in Section 13.
10-1
-------
11. BIOLOGICAL EFFECTS ON WILDLIFE
No information was found on the toxicity of acrylonitrile to wildlife other
than insects.
11.1 INSECTS
Judson et, al. (1962) studied the ovicidal effects of acrylonitrile and other
chemicals on the eggs of the yellow-fever mosquito (Aedes aegypti). Mature eggs
were exposed for 24 hours to the vapor of 5 or 10 ul added acrylonitrile in sealed
one quart jars (21 to 32°C, 100$ humidity) and then placed in deoxygenated water
to determine hatchability. The percent mortality at the two treatment levels
(4.2 or 8.4 mg/1, by calculation) was 60 and 92$, respectively.
Bond (1963) exposed adult granary weevils (Sitophilus granarius) and
cadelle larvae (Tenebroides mauritanicus) to a series of acrylonitrile fumigant
concentrations for an unspecified period. The exposed insects were then divided
into three groups, which were kept for 48 hours in an atmosphere of nitrogen,
oxygen, or air and then placed in air for 5 days. The dosage (expressed as the
product of concentration and exposure time) required to kill 50$ of the insects
kept in air was 23-0 for T. mauritanicus and 5.4 for S. granarius. The median
lethal concentration cannot be calculated from these values because Bond did not
specify the duration of exposure to the fumigant. The results did indicate,
however, that oxygen enhanced the toxicity of acrylonitrile and most of the other
chemicals tested.
Lindgren et^ al. (1954) fumigated eight species of insects with a series of
acrylonitrile concentrations for 2 or 6 hours. Mortalities were counted 4 days
after fumigation. The LD50 and LD95 values are given in Table 11-1.
Similar toxicity studies were conducted with acrylonitrile by Bond and
Buckland (1976, 1978) with several insect species. The duration of exposure,
11-1
-------
Table 11-1. Lethal Dose Values for Insects Exposed to Acrylonitrlle Fumigation
Species
Length of
Exposure
(hours)
Length of
Test
(days)
LD50
(mg/1)
LD95
(n.g/1)
LD99
(mg/1)
Reference
Granary Weevil
(Sltophllua granarlua)
Rice Weevil
(Sitophllua oryra)
Mexican Bean Weevil
(Zabrotea pectoralla)
8
6
5
2
6
2
24
6
2
Bean Weevil 6
(Acanthosceelldea obtectus) 2
Drug-store Beetle
(Stegoblum panlceum)
Confused Flour Beetle
(Trlbollum confusum)
Pulse Beetle
(Calloaobruchus chlnenala)
Red Flour Beetle
(Trlbollun caataneura)
5
24
Saw-toothed Grain Beetle 6
(Oryzacphllus surlnamenala) 2
24
Lesser Grain Borer
(Rhyzopertha domlnlca)
Cadelle
(Tenebroldes maurltanlcus)
6
2
24
8
5
0.7
2.0
1.4
4.5
1.0
2.5
0.40
1.4
2.0
1.1
3.0
1.7
3.0
1.9
3.0
6.5
0.43
2.6
1.05
1.1
3.5
0.76
0.8
2.5
0.76
2.8
3.8
2.9
8.0
1.8
6.5
0.78
2.1
4.0
2.0
5.5
2.5
7.0
4.9
11.0
0.56
1.32
2.0
6.5
1.27
1.4
4.0
1.08
1.2 Bond 6 Buckland, 1976
Llndgren et^ aj^., 1954
2.2 Bond & Buckland, 1978
Llndgren e^ aJL., 1954
Llndgren et_ al., 1954
Llndgren et^ al. , 1954
Rajendran & Mutlm, 1976
Llndgren et_ a_l., 1954
Llndgren et^ al.., 1954
Llndgren ejc al., 1954
Llndgren et^ nl., 1954
Llndgren et^ a^., 1954
Llndgren et_ a_l., 1954
2.5 Bond & Buckland, 1976
Llndgren et^ al., 1954
Llndgren e_t a_l., 1954
Rajendran & Huthu. 1976
4.3 Bond & Buckland, 1978
Rajendran & Hulhu, 1976
Llndgren et^ tU., 1954
Llndgren et^ al., 1954
Rajendran & Huthu, 1976
Llndgren et^ al., 1954
Llndgren et_ al_., 1954
Raj end ran & Huthu, 1976
6.0 Bond & Buckland, 1976
8.4 nond & Buckland. 1976
-------
time at which mortalities were counted, and the LD50 and LD99 values are given in
Table 11-1. Bond and Buckland (1976) found that acrylonitrile alone was more
toxic than methyl bromide alone or mixtures of both compounds. Although acrylo-
nitrile is too flammable for use as a fumigant alone, it does enhance the
toxicity of non-flammable methyl bromide, especially at low temperatures. Bond
and Buckland (1978) showed that fumigation with acrylonitrile and methyl
bromide:acrylonitrile mixtures was more effective in atmospheres of 20 to 50$
carbon dioxide than in air.
Rajendran and Muthu (1976) also conducted fumigation bioassays with six
species of stored product insects. As shown in Table 11-1, the longer exposure
period (24 hours) resulted in LD50 and LD95 values that were lower than those
reported for the same species by the previously cited workers.
The results presented in Table 11-1 indicate that the concentration of
acrylonitrile required to kill 95$ or more of test groups of insects is between
about 0.5 and 10 mg/1, depending on species and exposure time. This concentra-
tion is between 1.3 and 2.6 times the LD50 concentration.
The only other information found concerning effects of acrylonitrile in
insects was by Benes and Sram (1969) , who found that acrylonitrile was not
mutagenic in fruit flies (Drosophila melanogaster).
11-3
-------
. 12. BIOLOGICAL EFFECTS ON AQUATIC ORGANISMS
12.1 ACUTE TOXICITY
The acute toxicity of acrylonitrile has been determined for several species
of marine and freshwater fish and invertebrates.
12.1.1 Freshwater Fish
The majority of information concerning the acute toxicity of acrylonitrile
to aquatic organisms has been developed with freshwater fish. The most compre-
hensive study is that of Henderson jjt al. (1961), who reported median lethal
concentration (LC50) values for fathead minnows (Pimephales promelas), bluegill
sunfish (Lepomis macrochirus) and guppies (Poecilia reticulata = Lebistes
reticulatus). Each bioassay utilized 5 acrylonitrile concentrations in a geo-
metric series and 10 fish per concentration. The test solutions were not renewed
during the 96-hour exposure period. LC50 values were calculated by graphical
interpolation from mortality data at 21!, U8, and 96 hours of exposure. These and
other values are presented in Table 12-1.
The 96-hour LC50 values ranged between 11.8 mg/1 for the most sensitive
species (bluegills) and 33.5 mg/1 for the least sensitive species (guppies). The
LC50 value decreased by a factor of about 2 between 2H and 96 hours in most tests,
indicating that toxicity increased with exposure time. In contrast, toxicity
tests with other organic nitriles (lactonitrile, benzonitrile, acetonitrile,
adiponitrile, oxydipropionitrile) showed relatively little or no increase in the
toxicity of these compounds with longer exposure time. Acrylonitrile LC50 values
for fathead minnows were slightly lower in hard water (320 mg/1 hardness) than in
soft water (20 mg/1 hardness), which indicates that acrylonitrile toxicity may
increase with higher water hardness. Because confidence intervals for these LC50
values cannot be calculated from the reported data, it is unknown whether the
12-1
-------
Table 12-1. Median Lethal Concentration (LC50) Values
for Fish Exposed to Acrylonitrile
LC50 (mg/1)
Species
Temp. Type Dilution 24 hr 48 hr 72 hr 96 hr Reference
(1*C) Teat Water
Fathead Minnows* 25 S FV(hard) 32.7 16.7
(Piaephales
promelas)
Fathead Minnows* 25 S FW(soft)8 34.3 21.5 20.5
(Pimephales
promelas)
Fathead Minnows* 25 F FW(soft)g 33.5 14.8 11.1
(Pinephale3
promelas)
Bluegill Sunfishb 25 S Fw(soft)8 25.5 14.3
(LepomlB
macrochirus)
Cuppies
(Poecilia
reticulata)
25 S FW(soft)8 44.6 33.5
Goldfish
(Carasslua
auratus)
Zebrafish
(Brachydanlo
rerio)
Minnow
(Phoxinus
phoxinus)
NR S FW
20 F FW
NR NX FW
KR NR FW
40
Carp
(Cyprinus
carp io)
Rainbow Trout NR NR FW(hard)
(Salmo
gairdnerl)
d
Golden Ide
(Leueiscus idus
melanotus)
Pinfish
Lagodon
rhomboides)
20
13.7- S SW
20.4
15
38.2 17.6
37.4 24.0
70
13.28
24.5
14.3 Henderaon «t al..
1961
18.1 Henderson et al..
1961
10.1 Henderson et al.,
1961
11.8 Henderson et al.,
1961
33.5 Henderson et al.,
1961
Paulet and Vidal,
1975
Slooff, 1978
Marconi and
lonescu, 1974
Marconi and
lonescu, 1974
Jackson and
Brown. 1970
Juhnke and
Luedemann, 1978
Daugherty and
Garrett. 1951
"length 50.8-63.5 am (2-2.5 inches); weight "1.5 g
blength 38.1-50.8 mm (1.5-2 inches); weight »2 g
Clength 25.4 mm (1 inch); weight «0.1 g
length 50-70 mm; weight 1.5 ± 0.3 g
estandard length 57-113 mm
fhardwater - pU 8.2, alkalinity 320 mg/1; acidity 0 mg/1; hardness 380 mg/1
8soft water - pH 7.4, alkalinity 16 mg/1; acidity 2 mg/1; hardness 20 mg/1
hpH 7-8; degree of water hardness 15 t 3' (German "Hartegrad")
S • static exposure
F • flow through exposure
FW- freshwater
SW- seawater
NR- not reported
12-2
-------
difference is statistically significant. It can be concluded, however, that
water hardness has little effect on acrylonitrile toxicity.
Henderson et al. (1961) also conducted continuous-flow acrylonitrile bio-
assays with fathead minnows in soft water. The exposure conditions are described
in Section 12.2 and the 21, 18, 72, and 96 hour LC50 values are given in
Table 12-1. Comparison of static and continuous-flow LC50 values shows that
toxicity is equal at 2U hours, but is greater under continuous-flow conditions
after M8 hours. Lower toxicity under static conditions may indicate that acrylo-
nitrile was lost from water through adsorption, volatility, chemical change,
fish uptake, or biodegradation.
Renn (1955) exposed bluegill sunfish under static and continuous-flow expo-
sure conditions to 0.38-3.79 mg/1 acrylonitrile and found no mortality during a
214-hour exposure period. White crappies (Pomoxis annularis) exposed to H acrylo-
nitrile concentrations under continuous-flow conditions began dying after about
2 hours in 90.9 mg/1 and after about 8 hours in 68.2 mg/1. No mortality occurred
during 2U-hour exposure to 37-9 or 22.7 mg/1. The concentrations given here were
calculated from Renn's concentration data, which were reported in mg/1 nitrogen
as acrylonitrile.
The US-hour LC50 of acrylonitrile to zebrafish (Brachydanio rerio) was
determined to be 15 mg/1 by Slooff (1978) . This bioassay was conducted with 10
fish per concentration in closed 10-liter aquaria under flow-through (6 1/hr)
conditions.
Paulet and Vidal (1975) determined a 72-hour LC50 of MO mg/1 for goldfish
(Carassius auratus). This bioassay was conducted under static conditions in
12-liter aquaria. No other information was provided.
Bandt (1953) provided some limited information concerning the static
toxicity of acrylonitrile to two freshwater fish species, bleak (Ablurnus
12-3
-------
alburnus) and roach (Rutilus rutilus). Bleak and roach were exposed in groups of
one or two fish to 20 to 100 mg/1 acrylonitrile for up to 20 days. Although
concentrations of 25 mg/1 and greater were eventually lethal, the only bleak
exposed to 20 mg/1 showed no effect after 20 days of exposure. Renn concluded
from this limited information that the threshold concentration for prolonged
exposure would be about 20 to 25 mg/1. As seen in Section 12.2, this value is
much too high.
Juhnke and Luedemann (1978) reported acute toxicity levels of acrylonitrile
to golden ide (Leuciscus idus melanotus) that had been determined by using the
identical protocol in two different laboratories. The 48-hour LCD, LC50, and
LC100 (0, 50, 100$ mortality) values determined at each laboratory are given
below:
48 Hour Lethal Concentration Values (mg/1)
LCO LC50 LC100
Laboratory 1 16 28 48
Laboratory 2 8 13 20
Although these tests presumably were conducted under identical conditions (see
Table 12-1), the values reported by laboratory 1 are about twice as high as those
reported by laboratory 2. The authors did not discuss the reasons for the
different results.
12.1.2 Marine Fish
The only report of acrylonitrile toxicity to marine fish is by Daugherty and
Garrett (1951). Groups of eight pinfish (Lagodon rhomboides) were acclimated for
22 to 24 hours in 30 liters of aerated seawater at 13.7 to 20.4°C. Acrylonitrile
was then added to give 16 concentrations ranging from 0.25 to 60 mg/1. No deaths
occurred at 20 mg/1 or less, or in controls. All fish died at 30 mg/1, which was
the next higher concentration tested. The 24-hour LC50, as determined by
12-4
-------
graphical interpolation, was 21.5 mg/1. This value is about the same as the 24-
hour LC50 values reported for freshwater fish (Table 12-1).
12.1.3 Freshwater Invertebrates
Bandt (1953) provided some limited information concerning the static acute
toxicity of acrylonitrile to scuds (Gammarua sp., a freshwater crustacean).
Groups of ten scuds were exposed to 25, 50, or 100 mg/1 acrylonitrile for up to 72
hours. All animals exposed to 50 and 100 mg/1 were dead by 48 hours. There was
no mortality in the group exposed to 25 mg/1 for 3 days.
Acrylonitrile may be more toxic to Daphnia magna, another species of fresh-
water Crustacea (LeBlanc, 1980). The 24 and 48-hour LC50 values (and 95? confi-
dence interval values) for this species tested under static conditions were 13
(11-15) and 7.6 (6.2-9.2) mg/1, respectively. The "no discernible effect concen-
tration" was 0.78 mg/1.
Randall and Knopp (1980) determined the static 48-hour EC50 (median effec-
tive concentration) of acrylonitrile for D. magna. This value (and its 95$
confidence interval) was 10.95 (9.54-12.56) mg/1, which is in reasonably good
agreement with the 48-hour LC50 value reported by LeBlanc (1980). Both of these
studies utilized young (< 24 hours old) D. magna. and water of similar mean
hardness (155-173 mg/1 as CaCOj and temperature (22°C).
12.1.4 Marine Invertebrates
Portmann and Wilson (1971, cited in Miller and Villaume, 1978) exposed
groups of 8 to 25 brown shrimp (Crangon crangon) to serial dilutions of acrylo-
nitrile in aquaria containing 10 gallons (37.85 1) seawater at 15°C. The
reported LC50 was 10-33 mg/1.
12-5
-------
12.2 SUBCHRONIC TOXICITY
There is information concerning the subchronic toxicity of acrylonitrile to
several species of freshwater fish and one species of freshwater invertebrate.
No information was found for marine fish or invertebrates.
12.2.1 Freshwater Fish
Henderson £t al. (1961) conducted five replicate subchronic bioassays with
fathead minnows under continuous-flow exposure conditions. Groups of 50 fish
were exposed to each of 7 acrylonitrile concentrations or to control water
(20 mg/1 hardness, 25°C) for up to 30 days. Controlled amounts of test solutions
were pumped into glass bottles containing 10 fish and 10 liters of water so that
a renewal time of 100 minutes was obtained. The fish were fed daily. The mean
LC50 values determined for the five replicate bioassays are given below:
Exposure Time (Days) 1 2 3 1 5 10 15 20 25 30
LC50 (mg/1) 33.5 1H.8 11.1 10.1 8.1 6.9 5.2 1.2 3.5 2.6
The LC50 decreased linearly with exposure time between 15 and 30 days, indicating
that the lethal threshold concentration had not been determined at the 30 day
mark and that mortalities would have continued to occur at concentrations well
below 2.6 mg/1 after 30 days.
Cumulative subchronic toxicity was also found by Jackson and Brown (1970,
cited in Miller and Villaume, 1978), who exposed rainbow trout to 2 to 200 mg/1
acrylonitrile for time periods as long as 100 days. Although the LC50 after
48-hour exposure was 70 mg/1, exposure to 2.2 mg/1 for 100 days resulted in 50$
mortality.
Very little is known about the mechanisms of acrylonitrile toxicity in fish.
Henderson et al. (1961) noted that the first sign of acrylonitrile toxicity in
fathead minnows was extreme darkening of the skin, followed in 1 to 3 days by
death. They found no cyanide, formed from acrylonitrile, in the exposure water.
12-6
-------
In contrast, the more rapid, non-cumulative toxicity of lactonitrile was attri-
buted to the formation of cyanide, which was measured in the exposure water. It
is not known whether metabolic formation of cyanide from acrylonitrile occurs in
fish, although cyanide or thiocyanate formation has been reported in mammals
(Section 13). Toxic action in mammals is attributed primarily to direct effects
of acrylonitrile and secondarily to cyanide toxicity. Slooff (1978) studied the
effect of acrylonitrile and other chemicals on respiratory activity of rainbow
trout. The frequency of respiratory movements was significantly increased after
24-hour exposure to 5 mg/1 acrylonitrile under flow-through conditions. This
effect was preceded by a temporary, but significant, decrease in breathing rate.
The concentration at which this sublethal effect occurred (5 mg/1) was three
times lower than the 48-hour LC50 for zebrafish (15 mg/1), determined under
similar flow-through exposure conditions.
12.2.2 Freshwater Invertebrates
The only information concerning subchronic toxicity of acrylonitrile to
aquatic invertebrates was developed with Daphnia magna (U.S. EPA, 1978). No
adverse effects were found when this invertebrate was exposed over its entire
life cycle to as much as 3.6 mg/1 acrylonitrile. As mentioned previously
(12.1.3), the 48-hour EC50 reported for this species was 7.55 mg/1, which is only
about twice as high as the chronic no-observed-effect concentration. In
contrast, the 48-hour LC50 was about 32 times higher than the 100-day LC50 for
rainbow trout (Jackson and Brown, 1970, cited in Miller and Villaume, 1978) and
about six times higher than the 30-day LC50 for fathead minnows (Henderson
et al., 1961). Although the chronic no-observed-effect concentration has not
been determined with any fish species, it is probable that it would be consider-
ably lower than 2 mg/1.
12-7
-------
12.3 SUMMARY AND CONCLUSIONS
Acrylonitrile has been shown to affect some terrestrial and aquatic plants
at exposure concentrations of 9 to 100 mg/1. Acrylonitrile is toxic to aquatic
animals at exposure concentrations in the low milligrams per liter range. The
reported acute LC50 values for fish ranged between 10.1 and 70 mg/1. Subchronic
exposure of fish for 30 to 100 days resulted in LC50 values of about 2 mg/1,
with no evidence that a threshold concentration had been reached. Although
the only tested invertebrate, Daphnia magna, had the lowest acute LC50 value
(7.6 mg/1), this species was not adversely affected by chronic exposure to 3.6
mg/1 throughout its whole life cycle.
12-8
-------
13. BIOLOGICAL EFFECTS IN MAN AND EXPERIMENTAL ANIMALS
13.1 PHARMACOKINETICS
13.1.1 Absorption and Distribution
Young et al. (1977) studied extensively the pharmacokinetic and metabolic
14
fate of acrylonitrile in male Sprague-Dawley rats by giving C-acrylonitrile
with different routes and dose levels. When acrylonitrile was orally admin-
istered to rats, essentially all (95$) of the dose was absorbed. After 72 hours,
the percentage (^ the standard deviation) of recovered radioactivity accounted
for 82 + 9.61$ and 101 + 14.4$ of the administered dose at 0.1 and 10 mg/kg,
respectively (Table 13-D• The present recovery of administered radioactivity
in the urine was much higher in the animals administered the higher dose, while
both doses resulted in a 5% recovery of radioactivity in the feces.
The absorption of acrylonitrile through inhalation was also determined by
Young et al. (1977). Animals were exposed to acrylonitrile vapor in a "nose
only" chamber at concentrations of 5 ppm and 100 ppm acrylonitrile. Following
exposure at the higher dose, a significantly higher recovery was found in urine,
but a smaller percentage was recovered in the expired air as CO- and in the body
(Table 13-2).
Rogaczowska (1975) studied dermal absorption of acrylonitrile vapor in rab-
bits and found that the penetration rate of acrylonitrile vapor through the skin
is about 1$ in relation to the quantity absorbed through the lungs.
The plasma concentration of acrylonitrile (ug Eq/ml plasma) as a function of
time, routes of administration (per os and intravenous), and dose levels (0.1, 1,
and 30 mg/kg) has been studied (Young e± al., 1977). Following oral and intra-
14
venous administration of C-acrylonitrile, typical plasma concentration versus
time curves were observed. The biphasic disappearance of radioactivity indi-
cated a pharmacokinetic two-compartment open model for elimination. The half-
13-1
-------
Table 13-1. Recovery of Radioactivity from Rats
Given Single Oral Doses of 0.1 or
10 mg/kg IuC-Acrylonitrilea-(Young etal., 1977)
Radioactivity in Urine
Radioactivity in Feces
Expired Air
Organics in carbon
Organics in solvent
H1ACNe
14c02
Body
Carcass
Skin
Cage Wash8
h
Total Recovery
Percentage of
0.1 mg/kg
Dose, Mean + S.D.
10 mg/kg
34.22 ± 6.26b
5.36 + 1.43
0.09 + 0.09
0.19 + 0.19
0.07 + 0.05
4.56 + 1.82
37.02 + 6.09b
24.24 + 5.02b
12.78 + 1.17
0.86 + 0.37
V,
82.37 + 9.64°
66.68 + 10.60b
5.22 + 1.17
0.11 + 0.06
0.20 ± 0.16
0.08 ± 0.03
3.93 + 1.79
26.61 + 5.91b
16.04 + 1.87b
10.57 + 4.55
1.22 + 0.41
K
104.04 + 14.40°
Two groups of rats housed individually in glass metabolism cages were
given 0.1 mg/kg (4 rats) or 10 mg/kg (5 rats); excreta were collected
at 8 hour intervals for 72 hours.
The mean values for the two dose levels are different at the P=0.05
level of significance.
Q
Pittsburg activated coconut charcoal, 12 x 30 mesh.
trap « 2-methoxyethanol
"trap =• 0.02 M Ag^O^, in 0.1 N ^SO^
'trap - 5 M ethanolamlne in 2-methoxyethanol
8,
'water-acetone
mean and S.D. of the recovery were calculated from the total
recovery of the dose In individual rats.
13-2
-------
Table 13-2. Recovery of Radioactivity from Rats Exposed by Inhalation to
5 or 100 ppm ^C-Acrylonitrile for 6 hours3 (Young e£ al., 1977)
Urine
Feces
14co2
Body
Cage Wash
Total dose
in pg Equivalents
Percentage of
5 ppm
Recovered Dose
100 ppm
68.50 + 9.38C
3.94 + 0.97
6.07 + 1.58°
18.53 ± 4.68C
2.95 + 3.95
172.92 + 28.35
82.17 + 4.21C
3.15 + 0.82
2.60 + 0.83°
11.24 + 2.85C
0.85 + 0.58
2556.65 + 672.10
Rats were exposed in a "nose only" chamber under dynamic air flow
conditions for 6 hours. After exposure, rats were housed in glass
metabolism cages and excreta were collected for 220 hours.
Values are the mean ;+ S.D. for 4 rats per exposure level.
CThe mean values for the two exposure levels are different at the
P-0.05 level of significance.
13-3
-------
life values of alpha and beta phases, calculated by linear regression analysis,
ranged from 3-5 to 5.8 hours and 50 to 77 hours, respectively.
Freshour et al. (I960) investigated the phannacokinetic profile of intact
acrylonitrile after intravenous or per os administration of acrylonitrile to
male Fischer F3W rats. The plasma concentration of acrylonitrile versus time
obtained after intravenous administration was characteristic of a one-
compartment model with first-order elimination, but a Diphasic elimination was
observed following a 30 mg/kg oral administration. The half-life of first-order
elimination ranged from 7.8 to 13.9 minutes after 30 mg/kg intravenous and per os
doses, and the half-life for terminal phase after 30 mg/kg oral administration
was 85 to 120 minutes.
Hashimoto and Kanai (1965) have studied the blood concentrations of acrylo-
nitrile and cyanide as a function of time in relation to toxicity. When acrylo-
nitrile was given to rabbits intravenously at a sublethal dose (30 mg/kg, LD50 =
50 mg/kg), acrylonitrile was biphasically eliminated with 1 ug/ml of acrylo-
nitrile remaining four hours after dosing. The cyanide concentration rose to
about 1.5 ng/ml at 1.5 hours after dosing, and then gradually returned to zero at
four hours. After injection of a lethal dose of acrylonitrile (75 mg/kg), the
hydrogen cyanide concentration rose steadily until the death of the animal.
Tissue distribution of radioactivity (acrylonitrile and its metabolites)
in
was determined in rats given a single oral or intravenous dose of C-
acrylonitrile at doses of 0.1 and 10 mg/kg (Young et al., 1977). Acrylonitrile
and its metabolites were distributed to all tissues examined (lung, kidney,
liver, stomach, skin, blood, etc.); notably, high levels of radioactivity were
observed in stomach, skin, and red blood cells regardless of route and dose level
(Table 13-3). The high accumulation of radioactivity in the stomach wall
following intravenous acrylonitrile administration indicated that the high
13-4
-------
Table 13-3. Distribution of Radioactivity in Selected
Tissues of Rats Given 1<4C-Acrylonitrile
(Young e£ al., 1977)
Lungs
Liver
Kidneys
Stomach Wall and Contents
Intestines
Duodenum
Je j unum
Ileum
Cecum
Colon
Skeletal Muscle
Heart
Spleen
Brain
Thymus
Testes
Skin
Carcass
Packed Blood Cells
Plasma Concentration
(ug Eq/ml)
Tissue
i.
2 hours
1 mg/kg
14
to Plasma Ratios of C-Activity
a
V.
24 hours
1 mg/kg
0.96
0.68
1.33
8.11
0.72
0.85
0.92
1.01
0.37
0.51
-
0.64
0.98
0.65
0.74
0.67
1.91
0.54
3.33
0.727
1.07
1.20
3.15
7.36
0.31
0.68
0.57
0.44
0.07
0.07
0.50
0.78
-
0.76
-
-
2.85
0.73
7.16
0.220
b
p.o.
72 hours
0.1 mg/kg 10
mg/kg
0.87
0.57
0.83
14.28
1.06
c
-
-
-
-
0.35
-
-
-
-
-
2.10
0.71,
2.26d
0.022
0.96
0.84
1.08
11.26
0.99
-
-
-
-
0.41
-
-
-
-
-
2.70
0.55
5.73e
1.437
One rat per time. (i.v. = intravenous.)
Average of 2 rats per dose. (p.o. = per os.)
"A dash indicates no analysis was performed.
For all 4 rats in this group, the ratio was 2.18 + 0.20 (Mean + S.D.)
EFor all 5 rats in this group, the ratio was 5.16 + 1.19 (Mean + S.D.)
13-5
-------
concentration in the stomach wall following acrylonitrile oral dosing was not due
to poor absorption. Analysis of the stomach following intravenous administra-
tion showed that the radioactivity had increased from 30.33 ug Eq at 5 minutes to
68.64 pig Eq at 21 hours. The particular retention of acrylonitrile and its
metabolites in the stomach seems in part due to enterogastric circulation (Young
et al., 1977) . The accumulation of radioactivity in blood was mainly due to
covalent binding of acrylonitrile to macromolecules and lipids in the red blood
cells (Ahmed and Patel, 1979; Ahmed e£ al., 1982). In other tissues, the amount
of radioactivity declined rapidly with time because of excretion (Young et al.,
1977; Ahmed and Patel, 1979; Ahmed e_t al., 1982).
Similar distribution patterns were obtained by Sandberg and Slanina (I960)
using the technique of whole-body autoradiography following administration of
TU
labeled acrylonitrile to rats and monkeys. Male and female rats received C-
acrylonitrile by intravenous injection (13mg/kg) and were killed 1 minute,
20 minutes, 1 hour, H hours, and 7 days after treatment. Two monkeys were given
1 or 6 mg/kg acrylonitrile orally and killed 1 and 6 hours later, respectively.
There was no difference in the distribution patterns of acrylonitrile between the
two routes of administration in rats, except for the slower absorption time
following oral dosing. High levels of activity were present in the blood and
excretory pathways (bile, intestinal contents, and urine) with the liver,
kidney, lung, and adrenal cortex also accumulating appreciable label. The
stomach and hair follicles showed constant uptake of label throughout this study.
Radioactivity was still present in animals killed seven days after administra-
tion of a_.'ylonitrile. In fetuses exposed in utero, only the eye lens accumu-
lated label at a higher concentration than that observed in maternal blood. The
labeling pattern in monkeys was similar to that in rats except for a more
pronounced activity in the liver.
13-6
-------
13.1.2 Metabolism
The metabolism of acrylonitrile has been studied by many researchers.
Dudley and Neal (1912) and Brieger e£ al. (1952) suggested that acrylonitrile is
metabolized to cyanide, which is transformed to thiocyanate and eliminated in
urine; however, less than one third of administered dose has been accounted for
by this metabolic route (Gut et al., 1975) . Acrylonitrile reacts with sulfhydryl
groups through cyanoethylation, which prevents further metabolism to cyanide and
thiocyanate (Hashimoto and Kanai, 1965). Other metabolic pathways have been
suggested including formation of mercapturic acid (U.S. EPA, 1979), oxidation to
CO via cyanate ion (Boxer and Richards, 1952) , and conjugation with D-glucuronic
acid (Hoffman e£ al.., 1976) . Evidence from Gut et al. (1975) , Wright (1977) , and
Young ^t al. (1977) suggested that the metabolism of acrylonitrile in mammals was
related to dose level, route of administration, and species.
13.1.2.1 Metabolism to Cyanide and Thiocyanate
Acrylonitrile is metabolized in laboratory animals to cyanide (CN~), and
then converted to thiocyanate (SCN~). Evidence of this pathway has been pre-
sented by Dahm (1977). The excreted urine collected from orally-dosed C-
acrylonitrile rats was analyzed by high pressure liquid chromatography (HPLC).
The suspected thiocyanate peak matched the retention time of a thiocyanate stan-
dard. Another experiment by Dahm (1977) showed direct evidence that thiocyanate
comes from the cyano group carbon of acrylonitrile. When rats were orally
m
administered C-acrylonitrile, labeled on either the olefin or cyano group, the
thiocyanate peak by HPLC was observed only when the cyano group was labeled.
Ahmed and Alreu (1982) have shown that acrylonitrile is converted to cynide in
the brain and liver of rats by enzymes associated with the microsomal fraction.
Urinary excretion of thiocyanate following acrylonitrile administration at
different dose levels and by various routes accounted for from 2 to 33$ of dose
(Gut et a^., 1975; Hashimato and Kanai, 1965). Gut et al. (1975) emphasized that
13-7
-------
the extent of conversion of acrylonitrile to cyanide and thiocyanate is both
route- and species-dependent.
Using female Wistar rats, albino mice, and Chinese hamsters, Gut et al.
(1975) studied the extent of the conversion of acrylonitrile to thiocyanate. In
rats, there was higher thiocyanate excretion after oral administration (I1*.6 to
33.1$) than after intraperitoneal (2.2 to 5.7$), subcutaneous (1.6$), or intra-
venous (1.2$) injection. After oral dosing, thiocyanate excretion showed a
distinct lag period (l| hours), which suggested that acrylonitrile was not appre-
ciably metabolized shortly after oral dosing (Gut e_t al^., 1975). In mice, there
was also a higher percentage of thiocyanate excreted after oral (35$) than after
intraperitoneal (8 to 10$) and intravenous (11$) injections. The portion of
acrylonitrile converted to thiocyanate was approximately the same for mice and
for rats. The higher total recovery of thiocyanate in mice compared to rats
indicated that mice may have a higher metabolic capacity and a lower binding
capacity for acrylonitrile (Gut et al., 1975). Hamsters were similar to rats and
mice in that more thiocyanate was excreted in urine after oral than after intra-
peritoneal administration of acrylonitrile. In contrast to rats, elimination of
thiocyanate in mice and hamsters after oral dosing showed no lag period. Gut
et a^. (1975) indicated that body size may be an influencing factor.
Pretreatment of rats with phenobarbital (a microsomal inducer), SKF525-A (a
microsomal inhibitor), cysteine, or dimercaprol did not significantly influence
the excretion of thiocyanate in the urine after the administration of acrylo-
nitrile. Simultaneous intraperitoneal administration of acrylonitrile and thio-
sulfate to rats and mice increased thiocyanate elimination twofold in rats and
threefold in mice, but no effect of thiosulfate was observed after oral adminis-
tration of acrylonitrile (Gut et al.. 1975). The lack of effect of microsomal
enzyme induction or inhibition on acrylonitrile-thiocyanate balance indicated
13-8
-------
that the route-dependent differences in thiocyanate elimination observed within
one strain of animals are caused by factors affecting the tissue distribution of
acrylonitrile (Gut et al., 1975) .
13.1.2.2 Reaction with Sulfhydryl Groups
Acrylonitrile reacts with sulfhydryl compounds in laboratory animals
through cyanoethylation (Hasimoto and Kanai, 1965; Gut et al., 1975; Dahm, 1977
and Langvardt et al., 1979); the reaction products are not further metabolized by
animals but rather are excreted unchanged (Gut et al., 1975). It has also been
demonstrated in hamsters following interperitoneal injection, that acrylonitrile
treatment resulted in a rapid decrease (within 1 hours) in the levels of
glutathione in both the liver and brain (Zitting et al., 1981). Hasimoto and
Kanai (1965) showed that acrylonitrile rapidly forms stable conjugates in vitro
with materials that have active hydrogen atoms, such as L-cysteine and L-
glutathione. That acrylonitrile forms conjugates with cysteine in vivo was also
demonstrated and the chemical structures of the metabolites were identified as S-
(2-cyanoethyl)cysteine (Gut et al., 1975; Dahm, 1977) and N-acetyl-S-( 2-
cyanoethyl) cysteine(cyanoethylated mercapturic acid) (Dahm, 1977; Langvardt
et al., 1979).
It has been suggested that acrylonitrile reacts in vitro with glutathione
(GSH) via GSH transferase, but the conjugation was determined indirectly by
measuring the disappearance of GSH substrates (Boyland and Classeaud, 1967).
Recently, the presence of cyanoethylated mercapturic acid in rat urine was con-
•
firmed (U.S. EPA, 1979). A proposed scheme for the various metabolic pathways of
acrylonitrile biotransformation is presented in Figure 13-1.
Langvardt et al. (1980) detected 7 urinary metabolites of acrylonitrile, by
high pressure liquid chromatography, in rats following administration of acrylo-
nitrile (30 mg/kg) by oral intubation. Of the 7 metabolites, the 3 major metabo-
lites (metabolites 1, 6, and 7) were further investigated to identify chemical
13-9
-------
I OIIDATIVE PATHWAYS
KM CKZWIC CMJUEAUON
i. KKUIC KIM
z. pwniw
1. IIIXOCICAi MUMTMKSxmOU: l) MRCWU.1XC *«D ITS
«MLOCS; 6) SEROTONIH; c) T-MIKMinT'IC ACID;
D) NISTMIKE
4.
-------
structures. Metabolite 7 was confirmed to be N-acetyl-S-( 2-cyanoethyl) cystein,
and metabolite 1 was thiocyanate, as indicated by i.n vivo labeling patterns when
m iu
animals were given 2,3- C acrylonitrile or 1- C acrylonitrile, and retention
times and fragmentation patterns on GC-MS. Metabolite 6 was tentatively identi-
fied as N-acetyl-3-carboxy-5-cyanotetrahydro-l,H-2H-thiazine since it had an
apparent molecular weight 2 units less than metabolite 7 and formed artifacts
similar to metabolite 7 on being reduced to dryness under N~ gas. No authentic
standard to metabolite 6 could be synthesized. By using analogues to known
metabolic pathways for other halogenated vinyl compounds, Langvardt £t al. pro-
posed a possible metabolic pathway of acrylonitrile which proceeds through the
formation of an epoxide intermediate (Figure 13-2) . The authors suggest that the
epoxide metabolite may be an important factor in the toxicity of acrylonitrile.
Abreu and Ahmed (I960) also demonstrated that an in vitro system, employing
a rat liver microsomal fraction and NADPH, was capable of forming cyanide from
acrylonitrile. Using microsomal fractions from rats induced with phenobarbital
or aroclor 1251* increased the production of cyanide, while addition of SKF525-A
to the reaction mixture inhibited the formation of cyanide, These results were
consistent with a metabolic pathway employing an epoxide as an intermediate.
In a similar study using [1- C] - or [2,3- C] -acrylonitrile incubated with
rat liver microsomes, Guengerich et^ al,. (1981) isolated labeled compounds which
co-chromatographed on high performance liquid chromatography (HPLC) with authen-
tic 2-cyanoethylene oxide. The formation of the epoxide was enhanced by the
inhibition of epoxide hydrolase. Addition of epoxide hydrolase to a solution of
0.5 mM 2-cyanoethylene oxide resulted in the destruction of the epoxide at a rate
of 1.7 mmol/min, and the formation of HCN. Both acrylonitrile and
2-cyanoethylene oxide when mixed with GSH (reduced glutathione) reacted non-
enzymatically, and HCN was released in the reaction with the epoxide. The
epoxide reacted less rapidly than did acrylonitrile in this system, however, when
13-11
-------
H/
[GS-CH2-CHj-CN]
+Ac
-Glu, Gly
0 0
II II
HO-C-CH-NH-C-CH,
I 3
^*"
-------
a cytosol preparation from rat liver was used, the rate of GSH conjugate forma-
tion was 309 and 31 mmol/mg/min, respectively, for 2-cyanoethylene oxide and
aer»ylonitrile. The cytosol from rat brain and human liver was inactive in
forming GSH conjugates with acrylonitrile. These preparations were active with
2-cyanoethylene oxide, although the activity was diminished by more than an order
of magnitude when compared to that in the rat liver preparation. A metabolite
with a mass spectrum similar to that which Langvardt e_t al. (1980) ascribed to
l|-acetyl-3-carboxy-5-cyanotetrahydro-1,i»-2H-thiazine was obtained by non-
enzymatic reaction of 2-cyanoethylene oxide with N-acetylcysteine. This
evidence supports a metabolic pathway in which acrylonitrile is metabolized by
microsomal enzymes through an epoxide intermediate to HCN. It was demonstrated
that 2-cyanoethylene oxide non-enzymatically bound to both protein and DNA, and
that acrylonitrile did not, indicating that the 2-cyanoethylene oxide may be an
important metabolite in the toxicity and carcinogenicity of acrylonitrile.
Wright (1977) studied the dose-related and species strain-related meta-
bolism of acrylonitrile. Spartan rats, a Charles River rat, and rhesus monkeys
were orally dosed with cyano-labeled acrylonitrile and the urinary metabolites
were determined (Table 13-1*) . In rats, a lower dose (0.1 mg/kg) of acrylonitrile
resulted in about 90$ of the administered dose being metabolized to cysteine
conjugates. At a higher dose (30 mg/kg), a great portion of administered acrylo-
nitrile was excreted as thiocyanate and unidentified metabolite "C." Wright
(1977) suggested that conjugation with cysteine is the preferred metabolic path-
way for acrylonitrile but has limited capacity. No significant difference
between the two strains of rats given a 30 mg/kg oral dose of acrylonitrile was
observed. As shown in Table 13-^, there is no significant difference in the
proportion of metabolites excreted in urine by rhesus monkeys following 0.1 and
30 mg/kg of acrylonitrile. This may indicate that the rhesus monkey has a
13-13
-------
Table 13-4. Urinary Metabolites Following the Oral Administration
of C-l (Cyano) Labeled Acrylonltrlle (Wright, 1977)
Animal
Spartan Rat
Spartan Rat
Charles River Rat
Rhesus Monkey
Rhesus Monkey
Number
1
1
1
1
1
Dose
(rag/kg)
0.1
30.0
30.0
0.1
30.0
Percent of Radioactivity In
Urinary Metabolites
Cystelne-
Conjugates
88.6
37.3
47.5
91.
84.
"C"
(unidentified)
3.5
20.5
13.8
6.
14.
Thio-
cyanate
8.0
42.1
38.8
3.
2.
CO
I
0-20 hours In rats and 0-24 hours In monkeys.
-------
greater metabolic capacity in conversion of acrylonitrile to cysteine-conjugates
than does the rat.
13.1.2.3 Minor Metabolites
In addition to the cyanide and thiocyanate formed by acrylonitrile's meta-
bolism, acrylonitrile-glucuronide was formed and excreted in rat urine. Hoffman
et al. (1976) suggested that acrylonitrile is initially hydrolyzed, and then
undergoes a condensation reaction with UDP-glucuronic acid.
Young et^ al. (1977) identified C02 as a metabolite of acrylonitrile. Carbon
dioxide was expired in the breath and accounted for 5 to 6% of the administered
dose of acrylonitrile. Boxer and Richards (1952) demonstrated that cyanide and
thiocyanate are in a dynamic equilibrium in the rats and that, to a large extent,
cyanide and thiocyanate carbon are oxidized directly to CO.. Thus, C0? may arise
as a product of cyanate metabolism.
The majority of the administered dose of acrylonitrile was metabolized and
excreted; however, Hashimoto and Kanai (1965) and Hoffmann et al. (1976) esti-
mated that 15$ of an administered acrylonitrile dose is eliminated unchanged in
the urine and breath.
13.1.2.^ Route and Dose Dependence of Acrylonitrile Metabolism
The metabolic fate of acrylonitrile and its dependence on dose level and
route of administration have been studied by many researchers (Gut Q al., 1975;
Wright, 1977; Young et al., 1977). Gut et al,. (1975) indicated that the forma-
tion of both cyanide and thiocyanate are dose- and route-related. Wright (1977)
found that the reaction of acrylonitrile with sulfhydryl groups is dose-
dependent. Recently, Young et a],. (1977) quantitatively isolated three metabo-
lites designated as "A," "C," and "E" as well as C02 in male Sprague-Dawley rats
13-15
-------
14
given doses of C-acrylonitrile by several routes. Metabolites A, C, and E,
which accounted for more than 95% of the total radioactivity, were excreted
primarily in the urine, while CCL was excreted in the breath. The chemical
structures of metabolites A, C, and E were cited by Keresztesty et al. (1977)
as N-acetylated conjugates of cysteine, thiocyanate, and thiocyanate, respectively
(apparently there was an error in the paper; thus, the reader should not
assume that the structure of C and E are the same), but there was no evidence
for structure identification in the original report (Young et al., 1977).
The proportion of the metabolites A, C, and E in various rat fluids was
route- and dose-dependent (see Table 13-5). The 0-72 hour urine sample showed
the highest amount of metabolite C (73%) after an oral dose of 0.1 mg/kg, but
the maximum amount of metabolite A (61%) was observed after a higher oral dose
of 10 mg/kg (Young et al., 1977).
When rats were exposed to acrylonitrile vapor at a concentration of 5 ppm
for 6 hours, 61% of metabolite E was excreted in the urine, while a dose of
100 ppm resulted in equal proportions of the three metabolites.
The predominant metabolite excreted in bile after a 1 mg/kg intravenous
dose was metabolite C (91%). For stomach, red blood cell, and plasma, metabolite
E accounted for 93, 83, and 42% of total radioactivity, respectively.
13.1.2.3. Summary and Conclusions—Acrylonitrile is readily absorbed in
animals following ingestion or inhalation, while dermal absorption is poor and
occurs at about 1% of that of the lungs. Following absorption of radiolabeled
acrylonitrile, the radioactivity disappears in a biphasic manner, with a
half-life for the first phase of 3.5 to 3.8 hours and the second phase of 50
to 77 hours (Young et al., 1977). The predominant route of elimination is
through the urine. The routes of elimination are dose-related with the percent
eliminated through the urine less for small doses as compared to larger doses,
13-16
-------
while the relative amount retained by the carcass is greater for the small
dose as compared to a larger dose. Acrylonitrile is metabolized to cyanide,
which is transformed to thiocyanic and by cyanoethylation of sulfhydryl groups
to S-(2-cyanoethyl)cysteine, followed by elimination of these metabolites in
the urine. Other minor metabolites are formed from acrylonitrile. The toxicity
of acrylonitrile is caused by both the acrylonitrile molecule itself and its
metabolites.
13.2 ACUTE, SUBCHRONIC, AND CHRONIC TOXICITY
13.2.1. Acute Toxicity
The initial investigation of the acute toxic effects of acrylonitrile
began in the 1940's when industrial use of acrylonitrile increased dramatically
due to the production of oil-resistant synthetic rubber. These studies included
both laboratory investigations in animals and compilations of case histories
of human intoxication that had occurred in the workplace. Since these early
cases of intoxication, occupational exposure to acrylonitrile has been greatly
limited by
13-17
-------
14
Table 13-5. Metabolites of C-Acrylonitrlle Separated from Various
Fluids of Rats by High Pressure Liquid Chromatography (UPLC)
(Young £t_ al., 1977)
Sample
Drine, 0-72 hr.
0.1 mg/kg, p.o.
10.0 mg/kg, p.o.
5 ppm
100 ppm
Bileb, 1 hr.
In vitro0
Scomachb>d, 24 hr.
RJBCb>dt, 24 hr.
Plasma , 24 hr.
14 a
Percentage of Total C
A
12
61
9
32
2
17
4
7
28
C
73
8
30
33
91
78
3
10
28
E
15
32
61
35
1
1
93
83
42
aMetabolites A, C, and E were the major metabolites separated by HPLC.
Values are expressed as percentages of the total radioactivity in
each sample applied to the column. The total percentages of A, C,
and E are less than 100% when minor metabolites B and D are present.
A dose of 1 mg C-acrylonitrile/kg was given intravenously.
Metabolites extracted from reaction mixtures with ethanol. The source
of enzyme was rat liver 9000 x g supernatant fluid.
Ethanol extracts of stomach or red blood cells (RBC). About 80% of
the radioactivity was extracted.
13-18
-------
engineering controls, and no recent cases of acute human intoxication have
occurred.
13.2.1.1 Acute Systemic Human Toxicity
Non-fatal intoxication by acrylonitrile has been reported in workers who
clean polymerizers in plants that manufacture synthetic rubber (Wilson, 194H;
Wilson e_£ jil., 1948). During these operations, workers were exposed to between
16 and 100 ppm of acrylonitrile for periods of 20 to 45 minutes. All workers
complained of nasal irritation and an oppressive feeling in the upper respiratory
passages. Dull headaches, nausea, and subjective feelings of apprehension and
nervous irritability were also common complaints following mild intoxication.
In more severe cases of intoxication, a low grade anemia and mild jaundice
occurred. All exposed workers appeared to recover fully, with the exception of
one who had severe Jaundice. This worker complained of lassitude and fatigue a
year later, although no pathological signs could be found. Similar symptoms of
headache, vertigo, nausea, and vomiting were reported by a chemist engaged in
distilling acrylonitrile (Sartorelli, 1966). Tremors, uncoordinated movement,
and convulsions were also experienced. As with the other cases of intoxication,
complete recovery occurred. Baxter (1979) has recently summarized the symptoms
of acrylonitrile poisoning in man, in order of occurrence, as follows: irrita-
tion of eyes and nose, limb weakness, labored breathing, dizziness and impaired
Judgement, cyanosis and nausea, collapse, irregular breathing, and convulsions,
possibly followed by cardiac arrest. A few additional case histories of occupa-
tional exposure to acrylonitrile were presented in an IARC Monograph (1979).
Five cases of acute lethal human intoxication have been reported (Davis
et al., 1973) following fumigation of domiciles with acrylonitrile mixed with
ether 66$ carbon tetrachloride or 70$ methylene chloride (v/v). It is not clear
whether acrylonitrile or the other component of the fumigant was the causitive
13-19
-------
agent in these deaths; however, one patient had measurable levels (0.05 mgt) of
cyanide in his blood at autopsy, and, in another case, the odor of cyanide was
noted in the tissue. The presence of cyanide only indicates that acrylonitrile
was metabolized prior to death. The only consistent symptom in these victims was
nausea as an initial indication of intoxication. On autopsy, no gross abnormali-
ties that could be attributed to the fumigant were noted.
In a report of four case histories of human disease following fumigation of
domiciles with acrylonitrile and carbon tetrachloride mixtures, Radimer et^ al.
(1971) described the appearance of toxic epidermal necrolysis. The first symp-
toms appeared between 11 and 21 days following rehabitation of the house (all
houses were declared safe by the exterminators). Symptoms observed in victims on
entering the hospital resembled those of extensive second-degree burns. Three of
the patients died 2 to 9 days after hospitalization and one recovered and was
discharged in 10 days. Biopsy samples showed blisters at the junction of the
dermis with a necrotic epidermis. It was believed that exposure occurred through
inhalation rather than skin absorption. Although a participating role for carbon
tetrachloride could not be overruled, the lack of reports of epidermal disease
following carbon tetrachloride exposure made the authors suspect that acrylo-
nitrile was the toxic agent.
13.2.1.2 Acute Systemic Animal Toxicity
Acrylonitrile is a highly toxic compound to laboratory animals, regardless
of the route of administration. In rats, the oral LD50 of acrylonitrile was
determined to be 113 mg/kg by Smyth et al. ( 1969) and the minimal fatal oral dose
in rats was reported to be 150 mg/kg (Wilson et al., 1918). The LD50 in Wistar
rats following subcutaneous or intraperitoneal injection of acrylonitrile was
80 mg/kg and 100 mg/kg, respectively (Knoblock et al., 1971). Benes and Cerna
( 1959) had reported the oral LD50 of acrylonitrile in both rats and mice as
13-20
-------
78 mg/kg and 27 mg/kg, respectively. The approximate lethal doses in mice have
been determined for acrylonitrile administered intragastrically and intraperi-
toneally, and were greater than 20 mg/kg and 15 mg/kg, respectively (McOmie,
19^3). The symptoms of intoxication in rats were briefly described by Wilson
et al^ (19^8) as respiratory changes, cyanosis, and convulsions followed by
death. For the investigation of acute toxic effects of acrylonitrile, exposure
by inhalation has been studied at length since this is the route of greatest
concern in industrial exposure.
Dudley and Neal (19^2) exposed Osborne-Mendel rats to atmospheres of 90 to
2^5 ppm acrylonitrile for 30 minutes to 8 hours. The subjective signs of
toxicity and the lethality of acrylonitrile were strictly dependent on the length
of exposure and the concentration of acrylonitrile. Short periods of exposure
(665 ppm for 30 minutes) produced only mild effects, while a similar dose given
over a long exposure period (635 ppm for H hours) was fatal to all the animals.
Smyth and Carpenter (19^8) observed fatalities in all 6 Sherman strain rats
following a 4-hour exposure to 1000 ppm acrylonitrile, and no deaths from a
500 ppm exposure. In a second study, however, Carpenter et^ al. (19^9) reported
that 2 to H rats (Sherman strain) of a group of 6 animals succumbed following a
500 ppm exposure for 1 hours. Brieger et al. (1952) exposed Wistar rats to
100 ppm acrylonitrile for 7 hours, with lJ of 20 animals dying. A similar dose
(130 ppm for H hours) was a non-lethal dose in the study of Dudley and Neal
(1912) (Table 13-6). Due to insufficient data, it is not possible to determine
whether the observed differences in susceptibility of rats in these experiments
were a result of mere experimental uncertainty, or were related to variations in
strains of rats used, sex of the animals (this was not reported in any of the
studies), or age of the animals (all animals were between 100 and 150 g, except
the Osborne-Mendel rats, which were 295 g).
13-21
-------
Table 13-6. Summary of Results of Exposures of Rats to Acrylonitrile*
(adapted from Dudley and Neal, 1942)
Length of
Exposure
(hours)
1/2
1
2
4
8
Concentration
(ppm)
2445
1490
1270
665
2445
1490
1270
665
1260
595
305
635
315
130
320
270
210
135
90
Dying
During
Exposure
0
0
0
0
0
0
0
0
0
0
0
50
25
0
94
44
6
0
0
Total
Deaths
(%)
0
0
0
0
81
25
0
0
100
6
0
100
31
0
94
44
6
0
0
Severity of Effects and Remarks
Marked. Slight residual effects to
24 hrs.
Marked. No residual effects in 24 hrs.
Marked. No residual effects in 24 hrs.
Moderate, transitory.
Deaths in 4 hrs. Show some slight
effects 24 hrs after exposure.
Deaths in 4 hrs. Show some slight
effects 24 hrs after exposure.
Marked. Showed some effects 24 hrs
after exposure. Normal in 48 hrs.
Marked, transitory.
Fatal. Dead in 4 hrs after exposure.
Marked, transitory.
Slight, transitory.
Fatal. Dead in 4 hrs after exposure
Marked. Survivors show no effects
in 24 hrs.
Slight transitory effects.
Fatal
Marked. Survivors show no effects
in 24 hrs.
Marked, transitory.
Moderate, transitory.
Slight discomfort only.
16 rats were exposed at each concentration for length of time shown.
Rats used for these studies were adult albinos, pure Osborne-Mendel;
average weight, 295 grams.
13-22
-------
There was considerable species variation in the susceptibility of animals
to the lethal effects of acrylonitrile, with the guinea pig being the most
resistant species and the dog the most sensitive (Table 13-7). Many of the
animals did not die during the exposure period but rather succumbed to the toxic
effects later. The number of fatalities given in the table indicates the total
number of animals that died during the exposure and observation period. The
concentrations given in Table 13-7 should not be interpreted as the minimum
lethal doses in the cases where a substantial number of challenged animals died;
these were the lowest concentrations reported but are presumably higher than
would be the minimum lethal doses.
The symptoms of intoxication were similar in all species except dogs and
guinea pigs. As reported by Dudley and Neal (1912), the symptoms following
lethal exposures included gasping (particularly of the abdominal type), spasm-
like convulsive movements of the abdominal wall, general convulsions, and coma
followed by death. With less severe exposure, hyperactivity and flushing (parti-
cularly in the Rhesus monkeys) were noted along with indication of slight irrita-
tion of the mucous membranes. Animals that were removed from the acrylonitrile
atmosphere prior to the gasping stage recovered completely with no therapeutic
attention (McOmie, 19^3). Dogs showed all the same symptoms as other species
upon exposure to a lethal concentration of acrylonitrile; however, for the dogs
that did recover, the residual effects of poisoning lasted for a longer period of
time (Dudley and Neal, 19^2). The slow recovery included refusing to eat for
10 days following exposure in 1 animal, and, for another, 3 days of partial
paralysis of the hind legs. In the most resistant species, the guinea pig,
symptoms included watering of the eyes, nasal discharge, and coughing. When
death did occur, it followed exposure by 3 to 6 days. Dudley and Neal (19^2)
13-23
-------
Table 13-7. Minimal Lethal Concentration of Acrylonltrile
During Four-Hour Exposure
Animal Species
Guinea Pig
Cat
Rat
Micea
Rabbit
a
Rhesus Monkey
Dog
PPM of Acrylonitrile
575
600
315
415
260
75
65
Fatalities/No.
Challenged
10/16
16/16
5/16
3/16
16/16
1/3
1/2
Reference
Dudley & Neal, 1942
Dudley & Neal, 1942
Dudley & Neal, 1942
McOmie, 1943
Dudley & Neal, 1942
Brieger et al., 1952
Dudley & Neal, 1942
Exposure time was three hours instead of four hours.
The great sensitivity of dogs was confirmed in the study of Brieger et al.,
1952.
13-24
-------
concluded that death of guinea pigs was due to mucous membrane and pulmonary
irritation and lung edema.
13.2.1.3 Mechanisms of Acute Toxicity
Early observation of animal experiments and human intoxication suggested
that acrylonitrile toxicity was due to the liberation of cyanide following meta-
bolism. Dudley el^ al. (19^2) compiled data comparing the toxic levels of hydro-
cyanic acid and of acrylonitrile in a number of species of experimental animals
(Table 13-8). Both in the quantities needed for intoxication and in the symptoms
observed, acrylonitrile appeared to act in a manner similar to that of hydro-
cyanic acid. Furthermore, in humans, the gross signs and symptoms of acrylo-
nitrile and cyanide poisoning were similar (Brieger ^t al., 1952).
With human exposure, the main biochemical evidence in support of the theory
that acrylonitrile toxicity occurred through the action of cyanide was the pre-
sence of thiocyanate in the blood and urine of acrylonitrile-exposed indivi-
duals. Thiocyanate is the end product of the body's detoxification of cyanide.
Following occupational exposure to acrylonitrile, maximum serum thiocyanate
levels occurred immediately after exposure, while urinary maxima were reached
between 24 and 48 hours (Mallette, 19*43). The persistence of thiocyanate in the
serum was proportional to the level of exposure, with blood levels normal at
2.5 hours after an exposure of 22 ppm acrylonitrile for 30 minutes and blood
levels not returning to normal values by 12 hours after an exposure of 50 ppm for
30 minutes (Wilson £t al., 1948). Due to the quantitative correlation between
acrylonitrile exposure and urinary and blood thiocyanate concentrations, mea-
surement of levels of thiocyanate was suggested as a method of monitoring acrylo-
nitrile absorption in individuals (Mallette, 1943).
To determine whether acrylonitrile acts by the same mechanism as cyanide,
animal models have been used to identify and compare cyanide metabolites in
13-25
-------
Table 13-8. Comparison of the Effects of Acrylonitrile and of Hydrocyanic Acid on Various Species
of Animals (Dudley and Neal, 1942)
Species
Rats
n
Guinea Pigs
n n
Rabbits
It
M
Cats
M
Hogs
tt
M
N
Monkey •
Acrylonitrllc
Concentration
•g/i
0.26
0.21
1.25
0.58
0.56
0.29
0.21
0.60
0.33
0.24
0.213
0.12
0.063
0.33
Calculated
as HCN
Bg/1
0.14
0.10
0.62
0.29
0.28
0.15
0.10
0.30
0.17
0.12
0.11
0.06
0.03
0.17
pp»
130
100
575
265
260
135
100
275
153
110
100
56
30
153
SyaptOM
Slight transitory effect
Slight transitory effect
Fatal during or after
exposure
Slight transitory effect
Fatal during or after
exposure
Harked transitory effect
Slight transitory effect
Markedly toxic
Markedly toxic, sovetlaes
fatal
Fatal to 3/4 of the doga
Convulsions + coma, 0 death
Transitory paralysis.
1 dog died
Very alight effects
Definite toxic effects
Hydrocyanic Acid
Concentration
•8/1
0.12
0.35
0.23
0.35
0.13
0.35
0.14
0.125
0.1
0.07-0.04
0.035
0.14
PP"
110
315
200
315
120
315
125
115
90
63-35
30
125
Svxptosu
Fatal In 1 1/2 hours
Fatal concentration
Tolerated for 1 1/2 hours without
ayoptosja
Fatal concentration
No narked toxic sysipto«a
lo 2 Bin respiratory paralysis, In
5 to 10 »ln death
Markedly toxic in 6 to 7 "In
Fatal to dogs
May be tolerated for houra, death
after exposure
Voaltlng, convulsions, recovery.
•ay be fatal
May be tolerated
Distinctly toxic after 12 mln
CO
I
ro
-------
animals exposed to acrylonitrile. Cyanide acts by complexing with heavy metal
ions, alone or as part of organic molecules, and the ultimate toxic effect is due
to the complex of cyanide with the ferric ion in the cytochromes. This causes
disruption in cytochrome function and cellular anoxia. During cyanide intoxica-
tion, two less toxic metabolites—cyanmethemoglobin and thiocyanate—are formed
which competitively inhibit the toxic action of cyanide. The ability of cyanide
antidotes, which facilitate the formation of competitive metabolites of the
cyanide ion, to protect against acrylonitrile poisoning has also been investi-
gated.
Cyanide has been found at various levels in animals after exposure to
acrylonitrile. Dudley and Neal (19^2) examined the liver, kidney, and gluteus
muscle of rats and guinea pigs immediately after death due to inhalation of
acrylonitrile and could find no detectable levels of cyanide. Hashimoto and
Kanai (1965), however, detected cyanide at 2.5 to 3.5 ug/ml in blood of rabbits
following intravenous administration of acrylonitrile (75 mg/kg) and in rats and
guinea pigs after intraperitoneal administrations of 100 and 125 mg/kg, respec-
tively. Cyanide was also detectable in the blood of both dogs and Rhesus monkeys
following inhalation exposure to acrylonitrile, but not in the blood of rats
(Table 13-9). In dogs, the levels of cyanide in the blood were proportional to
the extent of exposure to acrylonitrile. This was the only study that found a
positive correlation between the sensitivity of the species exposed to acrylo-
nitrile (Table 13-7) and the level of cyanide in the blood.
Both cyanide metabolites, thiocyanate and cyanmethemoglobin, have been
detected in the blood (and urine, in the case of thiocyanate) of animals after
exposure to acrylonitrile. Breiger ejt aU (1952) compared the levels of thio-
cyanate in the blood of dogs with the levels in the blood of rats, a species which
was less sensitive to the toxic effects of acrylonitrile (Table 13-9). An
13-27
-------
Table 13-9. Cyanide and Thlocyanate in Blood of Animals Exposed
to Acrylonitrile (adapted from Breiger et al., 1952)
Species
Rats
Dogs
Dogs
Rats
Dogs
Dogs
Rats
Dogs
Dogs
Rhesus monkey
Rhesus monkey
Acrylonitrile
in Air (ppm)
100
100
100
75
75
75
50
50
50
75
75
Length of
Exposure
7
1
7
7
1
7
7
1
7
3
7
Cyanide in
Blood
(vimol/ 100ml)
1
2.93
10.10
0
1.85
6.65
0
trace
0.51
3.70
8.72
Thiocyanate mg/100 ml
Blood
6.63b
0.45
0.99
4.21b
0
0.39
1.74b
0.55
0.80
1.67
7.98
Urine
2.2
3.6
1.1
1.93
0.25
1.47
i
4.12
CO
I
ro
00
Zero exposure levels are subtracted from the values presented.
Determined in plasma.
-------
inverse relationship was demonstrated between the levels of thiocyanate, levels
of cyanide, and degree of sensitivity, with the species less sensitive to acrylo-
nitrile poisoning having lower blood cyanide and higher thiocyanate levels. The
authors were surprised at the high thiocyanate levels as compared to the cyanide
levels in the blood from monkeys. Cyanmethemoglobin was also detected in the
blood of all exposed animals, with the highest concentrations found in the Rhesus
monkey (H.8 umol/100 ml) and the dog (2.7 umol/100 ml) following a 7-hour expo-
sure to 75 ppm acrylonitrile. There was no detectable cyanmethemoglobin in the
blood of rats at this exposure level; however, traces were detected at the
100 ppm exposure level. Breiger et al. (1952) concluded that acrylonitrile must
be metabolized to cyanide and they suggested the possibility that a species'
ability to detoxify cyanide was related to its resistance to acrylonitrile
poisoning.
Several antidotes to cyanide poisoning have been evaluated for their effec-
tiveness following acrylonitrile exposure, and as an indication of whether
acrylonitrile has a mechanism of action similar to that of cyanide. Two types of
antidotes have been used. The first, sodium thiosulfate, facilitates the detoxi-
fication of cyanide to thiocyanate by the following reaction:
Thiosulfate + cyanide rhodanese sulfite + thiocyanate
thiocyanate oxidase
The second type consists of chemicals (methylene blue and nitrites) that cause
the formation of methemoglobin, which competes with the cytochromes for the
cyanide ion. The use of both types of antidotes has indicated that acrylonitrile
toxicity cannot be attributed solely to the release of cyanide, and that either
acrylonitrile or another metabolite of acrylonitrile has significantly toxic
properties.
Hashimoto and Kanai (1965) examined blood levels of acrylonitrile, thio-
cyanate, and cyanide in rats, guinea pigs, and rabbits after intravenous
13-29
-------
(rabbits) or intraperitoneal (rats and guinea pigs) injection of acrylonitrile
or of sodium thiosulfate followed in three minutes by acrylonitrile. As illus-
trated in Figures 13-3 and 13-1*, in the rabbit, thiosulfate pretreatment did not
affect the levels of acrylonitrile in the blood, but thiosulfate did cause a
dramatic decrease in the cyanide levels and an increase in thiocyanate levels.
The use of thiosulfate prevented the death of three of the four rabbits exposed
to 75 mg/kg acrylonitrile and reduced the severity of the symptoms of intoxica-
tion. At 100 mg/kg, death was delayed by thiosulfate, but all animals eventually
succumbed to the toxic effects of acrylonitrile, even though the cyanide levels
present in the blood were lower than those in animals that survived a 30 mg/kg
exposure to acrylonitrile alone. Similar results were obtained using rats and
guinea pigs, with the animals dying in spite of reduced blood cyanide levels
(2.62 to 0.18 ug/ml in guinea pigs and 3-01 to 0.51 ng/ml in rats) following
treatment with thiosulfate. It was apparent that even with extensive detoxifica-
tion of liberated cyanide, acrylonitrile lost little of its toxicity.
Magos (1962) attempted to prevent the toxic effects of subcutaneously
injected acrylonitrile by pretreating rats with an intraperitoneal injection of
sodium nitrite. The results are shown in Table 13-10. Sodium nitrite effec-
tively protected rats from the lethal effects of cyanide, but had no effect on
either survival rate or survival time of animals treated with acrylonitrile. An
estimation of the percentage inhibition of cytochrome oxidase in rats receiving
•
cyanide and in rats receiving acrylonitrile revealed greater inhibition (8H$) in
rats that survived cyanide poisoning than in rats (71.5$) that were killed by
acrylonitrile. From these results, it appeared that cyanide did not play a
significant role in the toxic effects of acrylonitrile in rats unless the forma-
tion of cyanide from acrylonitrile in the tissue of exposed animals allows
toxicity at a lower cyanide concentration.
13-30
-------
£30-
OACRYLONITRILE
• THIOCYANATE
ACYANIDE
0 60 120 180 240
TIME AFTER INJECTION OF ACRYLONITRILE (min.)
a. ACRYLONITRILE 30mg/kg
| I I I I I I I I Tl I
0 60 120 160
TIME AFTER INJECTION OF ACRVLONITRILE (min.)
b. ACRYLONITRILE 7Sm«/kg
Figure 13-3.
Distribution of Acrylonitrile, Cyanide, and Thiocyanate
in the Blood after a Single Injection of Acrylonitrile
(Rabbit) (Hashimoto and Kanai, 1965)
13-31
-------
TOXIC ACTION OF ACRYLONITRILE
O ACRVLONITRILE
• THIOCYANATE
ACYANIDE
BO-
DEATH
0 60 120 180 240
TIME AFTER INJECTION OF ACRVLONITRILE (min.)
• . ACRVLONITRILE 75mg/kg 3 MINUTES
AFTER SODIUM THIOSULFATE 320mg/kj
1 I "'" I
0 60 120 180
TIME AFTER INJECTION OF ACRYLONITRILE (min I
b. ACRYLONITRILE lOOmg/kg 3 MINUTES
AFTER SODIUM THIOSULFATE 430m«/kg
Figure 13-4.
Effect of Sodium Thiosulfate on the Distribution
of Acrylonitrile, Cyanide, and Thiosulfate
(Rabbit) (Hashimoto and Kanai, 1965)
13-32
-------
Table 13-10. Effect of Methemoglobinemia on Mortality
Ratios In Albino Rats Poisoned with
Aerylonitrile, Potassium Cyanide, and
Acetone Cyanohydrin (Magos, 1962)
Compound
Acrylonitrile
Potassium cyanide
Potassium cyanide
Acetone cyanohydrin
Dose
(millimole/kg)
2.8
2.4 x 10"1
3.7 x ICf1
1.6 x 10"1
No. That Died /No. Given
Cyanide Compound
Without Sodium
Nitrite
5/5
5/5
—
5/5
With Sodium
Nitrite
5/5
1/5
4/4
0/5
13-33
-------
Benes and Cerna (1959) treated mice and rats orally with both sodium nitrite
and thiosulfate and observed changes in the lethality of a subsequent oral dose
of acrylonitrile. In rats, the cyanide antidotes protected 30f of the animals
from a LD100 dose of acrylonitrile. In mice, the effect was more dramatic, with
protection afforded at up to three times the LD50 dose. It was concluded that
toxicity in mice might be due to cyanide, while in rats the toxicity is due
mainly to the acrylonitrile molecule itself. It was shown that by 3 days post
exposure, rats had only metabolized 19.'I? of the acrylonitrile to cyanide (the
data for mice were not given). Benes and Cerna (1959) suggested that species
variation in susceptibility to acrylonitrile intoxication might be due to the
ability of animals to metabolize acrylonitrile to cyanide. Other investigators
(Magos, 1962; Gut et al_., 1975), however, suggested that the difference was due
to differences in ability to detoxify the liberated cyanide. The exact role of
liberated cyanide both in the toxicity of acrylonitrile and in the almost tenfold
difference in susceptibility of experimental animals to acrylonitrile1s toxic
effects is still unclear.
In further studies of the mechanism of acrylonitrile's toxicity, Hashimoto
and Kanai (1965) pretreated rabbits intravenously with L-cysteine 3 minutes
prior to dosing with acrylonitrile. It was shown in a test tube reaction that
L-cysteine forms an addition product with acrylonitrile by cyanoethylation, and
Gut e_t al. (1975) demonstrated that S-(2-cyanoethyl)-cysteine was non-toxic and
not metabolized to thiocyanate when administered orally to rats. Hashimoto and
Kanai (1965) analyzed the blood of rabbits treated with acrylonitrile and with
acrylonitrile and cysteine, and found a large decrease in the levels of acrylo-
nitrile and cyanide in animals treated with cysteine (Figures 13-5 and 13-6).
Unlike the antidotes to cyanide poisoning, cysteine not only relieved the symp-
toms of poisoning, but rabbits receiving 100 mg/kg acrylonitrile also survived
13-34
-------
OACRYLONITRILE
• THIOCYANATE
A CYANIDE
120
180
240
TIME AFTER INJECTION OF ACRYLONITRILE (men.)
I. ACRVLONITRILE 30mfl/kg
0 60 120 180
TIME AFTER INJECTION OF ACRYLONITRILE (min.l
b. ACRYLONITRILE TSmg/kg
Figure 13-5-
Distribution of Acrylonitrile, Cyanide, and Thiocyanate
in the Blood after a Single Injection of Acrylonitrile
(Rabbit) (Hashimoto and Kanai, 1965)
13-35
-------
OACRYLONITRILE
• THIOCYANATE
ACYANlOE
r
0 60 120 180 240
TIME AFTER INJECTION OF ACRYLONITRILE (mm.I
». ACRYLONITRILE 75ing/kt 3 MINUTES
AFTER L-CYSTEINE-HCI 220mg/kg
10-
o
8
a
ut
Z
»-
Z
8
s-
0 80 120 180 240
TIME AFTER INJECTION OF ACRYLONITRILE I mm. I
b. ACRYLONITRILE lOOmg/kg 3 MINUTES
AFTER L-CYSTEINE-HCl300m«/kg
Figure 13-6.
Effect of L-Cysteine on the Blood Concentrations
of-Acrylonltrile, Cyanide, and Thiocyanate
(Rabbit) (Hashimoto and Kanai, 1965)
13-36
-------
and completely recovered from the treatment. Similar results were noted with
guinea pigs, rats, and mice (Table 13-11). Positive effects were noted with
other sulfur-containing compounds. Because cysteine could protect animals from
the toxic effects of acrylonitrile by reacting directly with the acrylonitrile
molecule, and cyanide antidotes were not effective in preventing death,
Hashimoto and Kanai (1965) speculated that acrylonitrile itself was a toxic agent
and may act in conjunction with cyanide to produce the ultimately toxic effects.
Pilotti (1975) found no acrylonitrile toxicity in vitro, however, as indicated by
inhibition of growth of ascites sarcoma BP8 cells in culture (1.0 mraol acrylo-
nitrile was assayed) and attributed this to the lack of formation of toxic
metabolites. It is uncertain whether the toxic activity ascribed to acrylo-
nitrile is due to the acrylonitrile molecule per se or to a non-cyanide metabo-
lite of acrylonitrile.
Acrylonitrile also reacts with endogenous glutathione by either one or both
of two possible mechanisms. Direct interaction may occur by S-cyanoethylation of
glutathione, or alternatively by enzymatic conjugation with glutathione during
the detoxification of acrylonitrile. Dinu (1975b) demonstrated dramatic
decreases in non-protein tissue thiol concentration (protein thiol levels were
unchanged) in rats following subcutaneous injection of two times the LD50 dose of
acrylonitrile (Table 13-12). Similar results were observed by Szabo et^ al.
(1977a, 1977b), who used direct measurements of tissue glutathione. The possi-
bility that glutathione levels were important in mediating the toxicity of
acrylonitrile was indicated by the greater sensitivity of rats to the toxic
effects of acrylonitrile at night when liver glutathione levels were low due to
normal diurnal variation (Jaeger, 1979). Also, Szabo ejt al. (1977a) showed a
correlation between the doses of acrylonitrile that caused a precipitous decline
in glutathione concentrations in the brain of rats and the doses that caused
13-37
-------
Table 13-11. Therapeutic Effect of SH .and S-S Compounds on Acute
Acrylonltrlle Poisoning3 (Hashimoto and Kanai, 1965)
Animal
Amount of
Acrylonitrile
Injection
Site of
Injection
Rabbit (5)b
100 rag/kg
iv
Guinea-pig (5)
125 rag/kg
IP
Rat (5)
100 mg/kg
IP
Mouse (5)
100 mg/kg
IP
Control
survival survival
rate time
(min)
0/5 98(5)
0/5 86(5)
0/5 85(5)
0/5 101(5)
L-Cysteine HC1
300=370 mg/kg
survival survival
rate time
(min)
5/5 -C(5)
2/5 132(3)
2/5 254(3)
5/5 -C(5)
L-Glutathione
570 mg/kg
survival survival
rate time
(min)
5/5 -c(5)
d
— — — _— _
... ...
a-Mercapto-
propionylglyclne
350 mg/kg
survival survival
rate time
(min)
4/5 100(1)
"
L-Methionine
425 mg/kg
survival survival
rate time
(min)
5/5 (5)
— — — — —
i
CO
I
OJ
00
Amount of AN = 2 x LD50, amount of SH and S-S compounds = AN x 1 = 1.5 equivalent
All numbers in parentheses represent number of animals.
CAnimals survived treatment.
•» Animals Not Treated.
-------
Table 13-12. Concentration of Protein (PBSH) and Nonprotein (NPSU)
SH Groups In Normal and Acrylonitrile-Intoxicated
Animals (ymoles SH/100 g vet tissue) (Dlnu, 1975b)
Liver
Kidney
Brains
Lung
Testis
Adrenal
PBSH
NPSH
PBSH
NPSH
PBSH
NPSH
PBSH
NPSH
PBSH
NPSH
PBSH
NPSH
Normal Animals
Mean + Standard Error
1604 +28.3
412 + 16.8
952 + 40
250 + 9.63
968 + 27.4
103 + 3.55
989 + 30
82 + 3
1070 + 41
237 + 3.6
528 + 20.7
127 + 4
Intoxicated Animals
Mean + Standard Error
1524 +46.3
42.4 + 1
900 + 26.3
24.6 + 0.6
836 + 26
34 + 1.48
864 + 24.4
28 + 1.48
1000 + 39
86 + 4.9
538 + 11
0 +
13-39
-------
mortality; however, they noted no tissue damage by light microscopy in the liver,
lung, or kidney of rats 60 minutes after an intravenously administered dose
(15 mg/100 g body weight) of acrylonitrile, even though the glutathione levels
were reduced by 80 to 90$.
Dinu (1975a) presented evidence that low glutathione levels caused by
acrylonitrile lead to excess formation of lipid peroxides and subsequent disrup-
tion of membrane function. Following treatment of rats with acrylonitrile (two
times the LD50 dose administered subcutaneously), increases in malonaldehyde
were observed in the liver. According to Dahle et al. (1962), the formation of
malonaldehyde parallels the formation of lipid peroxides. The increase in lipid
peroxides occurred even though the treatment with acrylonitrile stimulated the
activity of glutathione peroxidase. The low levels of glutathione in intoxicated
animals made the protective effects of glutathione peroxidase ineffective.
Although comment was made on the disruptive properties of lipid peroxides on
membrane function, no measurements of membrane function were made.
Hashimoto and Kanai (1965) examined respiration in guinea pig brain and
liver slices incubated with acrylonitrile (10" M) . The oxygen consumption of
the brain slices was measured under both potassium stimulating and non-potassium
stimulating conditions. Acrylonitrile caused a 20$ decrease in the oxygen usage
of the potassium stimulated preparation, while no inhibition was noted in the
non-potassium stimulated preparation. This inhibition of respiration appeared
to be specific to the brain slice. Inhibition of respiration in liver slices was
not observed until 5 x 10~ M acrylonitrile was used, at which point precipita-
tion of protein occurred in the liver. There was no protection offered by the
simultaneous administration of sodium thiosulfate with acrylonitrile. It was
also apparent that inhibition of respiration was not due to the liberation of
_H
cyanide, since cyanide at 10 M inhibited both potassium stimulated and non-
•13-40
-------
stimulated brain slices, and sodium thiosulfate protected the brain slices from
the detrimental effects of cyanide. Effects on the peripheral nerves were
demonstrated by applying acrylonitrile to the nerve trunk in an isolated Sciatic-
gas trocnemius preparation from frog and observing a rapid anesthetic action.
This anesthetic effect rapidly disappeared following removal of acrylonitrile.
It was concluded that acrylonitrile produces abnormal function of both the peri-
pheral and central nervous systems.
Acute adrenal apoplexy has been described in rats following the administra-
tion of acrylonitrile (Szabo et al., 1976b). Following intravenous administra-
tion of 15 mg acrylonitrile/100 g body weight, the lesions appeared bilaterally
between 1 and 2 hours after treatment. The intravenous administration of this
dose produced hemorrhage in 90 to 100$ of the animals, while administration of
the same dose of acrylonitrile by gavage produced hemorrhage in only 20$ of the
animals. As reported by Szabo and Selye (1971), mortality in both cases was
100$. Szabo e_t a^. (1977b) observed early hematological changes consistent with
a longer blood clotting time in rats following the intravenous administration of
acrylonitrile, but it was not clear if these hematological changes were suffi-
cient in themselves to cause the hemorrhage. Protection from adrenal apoplexy
and, to some degree, mortality was offered by pretreating the animals for 1 days
with phenobarbital and some steroids. It was believed that this protection was
offered by the ability of these xenobiotics to induce drug-metabolizing enzymes
and hence alter the metabolism of acrylonitrile (Szabo and Selye, 1972). Since
mortality was not always associated with adrenal apoplexy, it was not apparent
how this hemorrhage effect related to the systemic toxicity of acrylonitrile.
13.2.1.1 Acute Topical Irritation and Toxicity
McOmie (1913) applied acrylonitrile to the shaved abdomens of rabbits at
doses of 1.0, 2.0, and 3.0 ml/kg to areas of 100, 200, and 200 cm2, respectively.
13-41
-------
No attempt was made to prevent evaporation of the test compound. Slight vaso-
dilatation was observed at the low dose, while the two higher doses caused slight
erythema. Only at the highest dose was systemic toxicity noted, as indicated by
depression and increased respiration rate. The animal showing toxic signs
recovered without any therapeutic treatment. When acrylonitrile was kept in
contact with the abraded skin of rabbits by absorption into a cellulose pad, the
LD50 was 0.25 ml/kg (Roudabush et^ al., 1965). When guinea pigs were used, the
LD50 for animals with abraded skin was 0.86 ml/kg, while the LD50 for animals
with intact skin was O.H6 ml/kg. It is interesting to note that in the guinea
pig, toxicity was greater for acrylonitrile when applied to intact skin.
Acrylonitrile caused only mild irritation to the eye of a rabbit when
applied in a single 0.05 ml drop (McOmie, 19^3) • One hour following treatment,
mild conjunctivitis was observed with no clouding of the cornea or pupillary
damage. The animal had completely recovered by 2H hours.
Skin irritation as a result of contact with acrylonitrile has also been
observed in humans (Wilson £t al_., 19^8). When the skin was exposed to acrylo-
nitrile vapors only, workers complained of itching although there was no
observable dermatitis. When the skin was in contact with the liquid, the symp-
toms included irritation, erythema followed by the formation of blebs, desquama-
tion, and slow healing. Dudley and Neal (19^2) reported on a laboratory accident
in which a small quantity of acrylonitrile was spilled on the hands of an
•
individual. The symptoms of diffuse erythema occurred by 24 hours, with blisters
on the finger tips and slight swelling. Erythematous and painful itching
occurred by the third day. Ten days after exposure, the skin was cracked and
peeled and large areas of tender new skin were present. No cases of human
fatality have been directly associated with absorption of acrylonitrile through
the skin.
13-42
-------
13.2.2 Subchronic Toxicity in Non-Human Mammals
In a preliminary study, Dudley et al. (1912) exposed 4 rhesus monkeys and
2 dogs to an atmosphere containing 56 ppm acrylonitrile for M hours a day, 5 days
a week, over a 1-week period. The rhesus monkeys showed no overt signs of
toxicity. Of the two dogs treated, one died during the first exposure period and
the second had intermittent paralysis of the hind legs following three of the
exposure periods (the fifth, thirteenth, and fourteenth). In subsequent experi-
ments, 16 rats, 16 guinea pigs, 3 rabbits, and 1 cats were exposed to 100 ppm
acrylonitrile, and 16 rats, 16 guinea pigs, 1 rabbits, I* cats, and 2 rhesus
monkeys were exposed to 153 ppn acrylonitrile. The exposure conditions were
similar to those described above except the period was increased to 8 weeks. At
the 100 ppm level, rats and guinea pigs showed only slight lethargy. Three
female rats gave birth to normal pups during the study. The rabbits did not gain
weight during the experiment, and the cats suffered the most severe symptoms of
vomiting and loss of weight. At the highest exposure levels, severe toxicity was
noted in all animals, with many dying prior to the completion of the study;
symptoms were similar to those noted after acute exposure to acrylonitrile. In
general, species sensitivity to subchronic acrylonitrile intoxication was simi-
lar to that observed following acute exposure, with the dog being the most
sensitive and rats and guinea pigs the least sensitive to acrylonitrile.
Dudley e_t al. (1912) examined the spleen, kidneys, liver, lungs, heart,
pancreas, lymph nodes, stomach, and small and large intestines of 18 rats, 6
rabbits, 6 cats, 16 guinea pigs, and 1 monkey for pathological changes (there was
no explanation of how the animals were selected for examination). The spleen of
rats showed hemosiderosis, which is indicative of blood destruction. The kidneys
of most animals showed renal irritation, with hyaline casts in animals in the two
higher exposure groups (the exceptions were 1 cat and 1 rabbit exposed at the
13-43
-------
100 ppm level and 1 monkey exposed at the 153 ppm level). Interstitial nephri-
tis, although not very extensive, was also noted in many animals (the exception
being the monkey and rats exposed at the 100 ppm level). Subacute broncho-
pneumonia was present in at least some members of all the species studied except
the cat; the pneumonia appeared as congestion and edema in the alveoli. Cats
were the only species that showed signs of liver damage, although the exact
lesions were not described.
Hematological determinations in rats and rabbits were made by Dudley and
coworkers (19^2) using animals that had been exposed to 153 ppm acrylonitrile.
These determinations revealed no changes in red and white blood cell counts or
hemoglobin levels. At the end of the first week of exposure, an increase in
eosinophils was noted using differential cell counts, with the maximum for
rabbits and rats being U2$ and 21$ above control, respectively. Minami e_t al.
(1973) claimed that changes in the partial pressures of oxygen and carbon dioxide
in the blood occurred, along with increases in hemoglobin and hematocrit values
following exposure of 8 rabbits once a week for 8 hours to 20 ppm acrylonitrile
for a period of 8 weeks. Due to the small differences observed and the rela-
tively large standard deviations, however, the statistical significance of these
conclusions is in doubt.
Some cumulative effects were observed in rats that received 50 mg/kg acrylo-
nitrile by daily intraperitoneal injection for a 3-week period (Knoblock e^ aj..,
1971)• The animals lost body weight, but the liver, kidneys, and heart increased
in weight. Signs of functional disorders of both liver and kidneys were noted,
and microscopic examination showed slight damage in neuronal cells of the brain
cortex, brain stem, and in the parenchymal cells of the liver. The only hemato-
logical change noted was leukocytosis.
13-44
-------
In an abstract, Szabo and coworkers (1976a) reported subacute effects of
acrylonitrile in female Sprague-Dawley rats (200 g) exposed to the compound at
2000 ppm in the drinking water for 21 days. At this concentration of acrylo-
nitrile, water intake was reduced in the experimental animals to one-third of
that of the control animals. Limited pathological studies indicated atrophy of
the zona fasciculata and increases in the zona glomerulosa of the adrenals.
Elevated plasma Na* concentration of 9 mEq/1 (K* was not changed) and reduced
corticosterone concentrations of 2H ug (the denominator was not indicated) were
noted. The authors concluded that acrylonitrile affected the mineral-corticoid
and glucocorticoid cells of the adrenal cortex. Using the same strain of rats
(Sprague-Dawley), Szabo £it al. (1977b) demonstrated a dose-dependent increase in
glutathione levels following a 21-day exposure to water containing 20, 100, or
500 ppm acrylonitrile. It was suggested that this increase in glutathione was
due to a rebound effect following the utilization of glutathione in the detoxifi-
cation of acrylonitrile.
Because of the limited number of studies on subchronic exposure to acrylo-
nitrile, it is difficult to formulate a unified pattern of toxic action, both in
terms of the mechanism of toxicity and the target organs. Although a number of
biological parameters have been reported to change following subacute exposure
to acrylonitrile, it is not clear how these effects relate to the primary toxic
action of acrylonitrile.
13«2.3 Chronic Toxicity in Non-Human Mammals
Two chronic toxicity studies have been conducted by the Dow Chemical Toxi-
cology Research Laboratory for the Chemical Manufacturers Association. In both
studies, the animals were exposed by ingesting drinking water containing acrylo-
nitrile. The animals were frequently evaluated during the studies for food and
water consumption, weight gain, and blood and urine chemistry. On death or at
13-45
-------
necropsy at the end of the study, autopsies were performed in order to detect
gross or microscopic pathological conditions.
In the first study (Quast et al., 1975), beagle dogs (four dogs of each sex
and exposure level) were exposed to 300, 200, and 100 ppm acrylonitrile for a
6-month period starting when the dogs were 8 months old. By using water consump-
tion data, the mean doses of acrylonitrile ingested at the three exposure levels
were for male dogs 17 + 4, 16 + 2, and 10 + 1 mg/kg/day, respectively, and for
female dogs 18 + 5, 17 +. 2, and 8+1 mg/kg/day, respectively. The two highest
exposure levels (300 and 200 ppm) were highly toxic to the dogs; 3 of the *4 males
and 2 of the 4 females at the 300 ppm level and 2 of the H males and 3 of the 4
females at the 200 ppm levels either died spontaneously or were euthanized
because of their debilitated state prior to three months (the ingested dose of
acrylonitrile at these exposure levels was nearly identical). Two male dogs at
the 300 ppm exposure level were removed from the study at 57 days because of
signs of overt toxicity and were placed on water containing no acrylonitrile.
These dogs appeared to recover completely (they regained their pretest weight)
and, after removal from acrylonitrile exposure for approximately 1 month, they
were returned to the study. Subsequently, the dogs again showed signs of
toxicity and died within 24 and 15 days. All dogs that died showed the following
progression of signs: terminal depression, lethargy, weakness, emaciation, and
respiratory distress. Dogs that ingested water containing 100 ppm acrylonitrile
and dogs that survived treatment at the higher doses appeared normal during the
period of study.
The presence of acrylonitrile in the drinking water caused a significant
decrease in water and food consumption in both male and female dogs at the
300 ppm level. At the other exposure levels, male dogs showed normal water and
food consumption, while female dogs had sporadically lower water consumption and
13-46
-------
consistently lower food consumption. A supplemental study, which used 8 female
dogs that ingested water containing 100 ppm acrylonitrile, failed to demonstrate
any decrease in either food or water consumption. With regard to general weight
gains (only surviving animals were considered since emaciation was a symptom
ensued by morbidity), all male dogs gained weight, although less than that gained
by control animals, while female dogs lost weight, except for the animals at the
100 ppm exposure levels. It was not clear whether the unpalatability of water
containing acrylonitrile may have affected these results.
Chemical studies on blood and urine samples revealed no significant abnor-
malities that were associated directly with acrylonitrile treatment. Some hema-
tologic changes, however, did occur at the higher exposure levels. These changes
were noted and were consistent with bronchopneumonia, which was chronically
present (and the reason for death) in the high dosed animals. Analysis of the
liver and kidneys for non-protein sulfhydryl content revealed no statistical
differences in the animals at the 100 ppm exposure levels and no apparent effect
at other doses, although these data could not be analyzed statistically because
of the small group size. There was also no difference in the immunologic status
of treated animals as assessed by percent of albumin, alpha 1, alpha 2, beta, and
gamma globulin in the serum.
On autopsy, some minor or indirectly-caused pathological findings were
noted. In male dogs at the 100 and 200 ppm exposure levels, there was an increase
in the relative kidney to body weight ratio (there were no changes in the single
male dog exposed to 300 ppm acrylonitrile). The kidney appeared normal on
histological examination and the weight gain may have been compensatory. There
was also a decrease in the relative brain weight of the 2 surviving male dogs at
the 200 ppm exposure level and an increase in the relative brain weight of the 2
surviving female dogs at the 300 ppm exposure level. No mention was made of
13-47
-------
results from histological examination of these brains. The lungs of dogs at the
higher exposure levels showed typical pathologies of foreign body broncho-
pneumonia, the cause of which appeared to be aspirated food particles. The only
treatment-related pathology was dilation, thinning of the walls, and focal
erosions or ulcerations of the esophagus in male and female dogs in the 300 and
200 ppm exposure groups. In some animals, although not consistently, there was a
thickening of the epithelium lining of the dorsal surface of the tongue. It was
believed that acrylonitrile may have caused irritation of the membrane of the
throat, resulting in the aspiration of food particles.
In the second study conducted by Dow Chemical Company, Quast et. aK (1980a)
exposed Sprague-Dawley rats to drinking water containing 35, 85, and 210 ppm
acrylonitrile. After the first 21 days the concentrations of acrylonitrile were
changed to 35, 100, and 300 ppm. The calculated amount of acrylonitrile ingested
was 3.12, 8.53, and 21.18 mg/kg/day for male rats and H.36, 10.76, and
24.97 mg/kg/day for female rats. By 9 months of treatment, the animals at the
two highest dose levels had rough hair coats and an unthrifty appearance. There
was a treatment-related early mortality in female rats at all exposure levels and
in male rats at the highest exposure levels (Table 13-13). The differences in
males and females may be partly related to the larger doses of acrylonitrile
ingested by female rats.
There was a significant dose-related decrease in daily water consumption of
•
both male and female rats exposed to acrylonitrile. This decrease in water
consumption was probably due to voluntary restriction of intake as a result of an
unpalatable taste imparted by acrylonitrile. The lower water consumption in
treated rats was the probably cause of the concomitant decrease in food
consumption and body weight. Chronic renal disease, which was common in older of
this strain, occurred less frequently in treated rats probably as a result of the
13-48
-------
Table 13-13. Cumulative Mortality of Male and Female Rats Maintained for Two Tears on Drinking Water Containing Acrylonltrlle (Quast e_t al.. 1980a)
Percent Dead
Days on Test
0-30
31-60
61-90
91-120
121-150
151-180
181-210
211-240
211-270
271-300
301-330
331-360
361-390
391-120
M21-U50
151-180
181-510
511-510
5H-570
571-600
601-630
631-660
661-690
691-720
721-715
Total Number
of Rats
Control
Male Female
0
0
0
0
0
1.3
1.3
1-3
2.5
3-8
7.5
8.8
8.8
10.0
13.8
18.8
26.3
36.3
«1.3
50.0
60.0
67.5
83-8
87.5
91.3
80
0
0
0
0
0
0
1.3
1.3
1.3
1.3
1.3
1.3
1.3
3-8
3.8
7.5
11.3
13.8
22.5
?7.5
12.5
16.3
56-3
67.5
75.0
80
35 ppra
Male Female
0
0
0
0
0
0
2.1
2.1
2.1
2.1
4.3
".3
6.1
10.6
1U.9
17.0
27.7
29.8
12.6
18.9
57.1
70.2
83.0
87.2
89.1
17
0
0
0
0
0
0
0
0
0
0
2.1
2.1
6.3
6.3
10.1
11.6
20.8
25.0
11.7°
50.0
56.3.
68.8"
79. 2a
87. 5a
91.7
18
100 ppm
Male Female
0
0
0
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
0.2
10.4
11.6
22.9
33-3
37.5
17.9
51.2
68.8
72.9
83.3
89.6
18
0
0
0
0
0
0
2.1
2.1
1.2
6.3
6.3
6.3
10.1s
12.5
12.5
22.9s
27. 1B
35.1"
56.3"
70.8s
79.2s
87. 5a
91.7s
95.8s
97.9s
IB
300 ppa
Male Female
0
0
0
0
0
0
0
0
4.2
4.2
4.2
8.3
11.6
16.7
22.9
31.3
15.8s
60.1s
66.7S
70. 8a
77.1s
83.3s
100.0
100.0S
100.0S
48
0
0
0
0
2.1
2.1
2.1
2.1
6.3
8.3.
18.8s
29.2s
33.3s
41.7s
50.0°
64.6s
72.9S
77.1s
87.5S
93.8s
95.8s
95. 8a
100. 0B
100.0s
100.0°
48
u>
I
aSlgnlflcantly different from controls by Fisher's Eitact Probability Test, p < 0.05.
-------
lower consumption of food and water. With this decrease in renal disease
there was a decrease in both gross and microscopic pathologies and clinical
findings associated with renal failure. Decreased food and water consumption
resulted in emaciation of the animals, which caused other clinical signs and
pathologies to change.
The majority of clinical chemistry and pathological changes that occurred
at a statistically significant level (P < 0.05) were attributed to secondary
effects of treatment and were not considered to be caused by a toxic action of
acrylonitrile. A summary of both the clinical chemistry and pathological
changes observed is presented in Table 13-14. Although it was proposed that
these changes were, for the most part, a result of restricted food and water
consumption, it was impossible to state this unequivocally because of the lack
of pair-fed controls. There were no severe or irreversible non-tumorigenic
pathological findings. The only non-tumorigenic changes that were considered
to be treatment-related were of a possible pre-neoplastic type; these include
hyperplasia and hyperkeratosis of the squamous epithelium of the nonglandular
portion of the stomach (100 and 300 ppm males and all treatment levels in
females), proliferation of the glial cells in the brain (300 ppm males and 35
and 100 ppm females), and mammary gland hyperplasia (300 ppm females). To
fully understand the chronic toxicity of acrylonitrile, studies must be performed
with pair-fed controls that are in the same nutritional state as the experimental
animals, and evaluation of clinical signs must be made in the absence of gross
neoplasisas.
13.2.3.A Summary and Conclusions
Acrylonitrile intoxication in humans results in irritation of the eyes
and nose, weakness, labored breathing, dizziness, impaired judgement, cyanosis,
3
nausea, and convulsions. The TLV of acrylonitrile is 45 mg/m for humans.
13-50
-------
Acrylonitrile also causes severe burns to the skin. In experimental animals,
there is considerable species variation in susceptibility to acrylonitrile
intoxication; the guinea pig is the most resistant and the dog is the most
sensitive. In animals, effects of intoxication include respiratory changes,
cyanosis, convulsions, and death. In rats, the LD,.n for acrylonitrile is
between 80 and 113 mg/kg (Knoblock et al., 1971; Smyth et al., 1969). There
is some evidence that acrylonitrile produces abnormal function of both the
peripheral and central nervous systems and that acrylonitrile causes damage to •
the adrenals. With subchronic exposure of animals to acrylonitrile, some
signs of functional disorders of the liver and kidney are observed. Chronic
exposure of dogs and rats results in unthrifty appearance, weight loss, and
early death. Some of these signs may be related to low food and water consumption
resulting from the unpleasant taste of acrylonitrile in the water. Pathological
changes in the rats believed to be treatment-related included hyperplasia and
hyperkeratosis of the squamous epithelium of the nonglandular portion of the
stomach, proliferation of glial cells in the brain, and mammary gland hyperplasia
in females.
13.3 TERATOGENICITY AND REPRODUCTIVE TOXICITY
The teratogenicity of acrylonitrile has been investigated using Sprague-
Dawley rats exposed to the compound by inhalation or by gavage on days 6 to 15 of
gestation (the results of both the inhalation and gavage treatments have been
reported by Murray et al., 1978, while a detailed report of the results of the
13-51
-------
Table 13-1U. Significant Changes Considered to be Secondary to Ingestion of Acrylonitrile
(Quast et al., 1980a)
Type of Analyses
Significant Findings
Suspected Cause of Findings
Hematology
Urinalysis
Clinical Chemistry
Pathology
Decreased white blood cell count
Decreased packed cell volume
Decreased hemoglobin
Increased urine specific gravity
Increased blood urea nitrogen
Decreased serum glutamic pyruvic
transaminase
Emaciation
Decreased adipose tissue
Dental malocclusion
Increased blood and bile-like fluid
Decreased chronic renal disease
Increased hepatocellular atrophy
Decreased signs of altered nervous
system function
Changes in organ weight
Blood loss from ulcerated tumors
Restricted water consumption
Restricted water and food consumption
Within normal limits
Restricted water and food consumption
Ear canal tumors deviating the lower Jaw
Restricted food consumption and ulcerating
tumors
Restricted food and water consumption and
early death
Poor nutritional state
Decrease in chronic renal disease
Poor nutrition
-------
gavage treatment was presented by Murray e_t al., 1976). Acrylonitrile was
administered at doses of 10, 25, and 65 mg/kg/day per os or by exposing the
animals to an atmosphere of 10 or 80 ppm acrylonitrile for 6 hours per day. It
was estimated by measuring blood levels of acrylonitrile and its metabolites that
an exposure to 80 ppm acrylonitrile for 6 hours was equivalent to a single
23 mg/kg dose of acrylonitrile. On day 21 of gestation, the dams were sacrificed
and examined for implants and fetal abnormalities.
Rats exposed to 65 mg/kg/day showed moderate signs of toxicity with
decreased body weight, thickening of the non-glandular portion of the stomach,
and increased liver weight. The only adverse effect on animals exposed to
25 mg/kg/day was a slight thickening of the non-glandular portion of the stomach,
while animals exposed to the lowest dose or animals exposed by inhalation showed
no major treatment-related toxic effects. In all cases, except the animals
treated with 10 mg/kg/day, food consumption initially declined during the first
3-day interval following treatment, but returned to normal during the two subse-
quent 3-day intervals. Neither the stress on the dams due to the mild toxic
effects of acrylonitrile nor any direct toxic effects of acrylonitrile on the
fetus significantly altered the number of litters, implants per dam, or live
fetuses per litter. A slight but significant decrease in fetal body weight and
crown-rump length was noted in the 65 mg/kg/day group, but not in any of the
other groups.
Fetal malformations were observed primarily in the offspring of animals
that received acrylonitrile by gavage (Table 13-15). Following inhalation of
acrylonitrile, there was no significant single major malformation observed; how-
ever, when malformations were considered collectively, there was a significant
(P = 0.06) but slight increase in the 80 ppm exposure group. At the high dose
level in the gavage-treated animals, there was a significant increase in acaudate
13-53
-------
Table 13-15. Incidence or Petal Malformations Among Litters of Rats Given Acrylonltrlle
(Murray et a_l., 1978)
Type of
Malformation
External and skeletal malformations
Visceral malformations
External ma 1 Populations
Short tall
Short trunk
Iraperforate anus
Omphalocele
Visceral malformations
Right-sided aortic arch
Missing kidney, unilateral
Anteriorly-displaced ovaries
Skeletal malformations
Missing vertebrae (associated with
short tall)"
Missing two vertebrae and a pair
of ribs
Healvertebra
Total malformed
Acrylonltrlle
0
1113/38
151/38
1(1)
0(0)
0(0)
0(0)
0(0)
1(1)
0(0)
1(1)
7(1)
0(0)
8(2)
given by gavage (mg/kg body weight/day) Acrylonltrlle
10
388/35
135/35
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
25
No.
312/29
111/29
No.
2(2)
0(0)
0(0)
0(0)
1(1)
0(0)
1(1)
2(2)
7(2)
0(0)
10(1)
65
fetuses/No.
212/17
71/17
of fetuses
8(6)a
3(3)a
2(2)
0(0)
1(1)
1(1)
1(1)
8(6)a
0(0)
0(0)
8(6)a
0
litters examined
121/33
110/33
(litters) affected
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
8(1)
0(0)
8(1)
given by inhalation (ppn!
10
111/36
1H8/36
0(0)
0(0)
0(0)
1(1)
0(0)
0(0)
0(0)
0(0)
2(1)
0(0)
3(2)
80
106/35
136/35
2(2)
1(1)
0(0)
1(1)
0(0)
0(0)
1(1)
2(2)
7(2)
1(1)
11(6)°
00
I
in
Values differ significantly from the control value (P < 0.05) using the Wllcoxon test as modified by Baseman and Hoel (1971).
This ralfomatlon was observed only In fetuses with short tall.
cValue differs significantly from the control value (P - 0.06) using the Wllcoxon test as modified by Haseman and Hoel (1971).
-------
or short-tailed fetuses. The majority of other abnormalities including short
trunk, anteriorly displaced ovaries, missing ribs, and imperforate anus were
observed in the acaudate and short-tailed fetuses whether these animals were from
the control or experimental group. The only anomaly that occurred solely in
treated animals was a right-sided aortic arch, which appeared in one fetus from
the 25 mg/kg/day group and one fetus from the 65 mg/kg/day group. This anomaly
had never been noted in over 1000 litters examined by this laboratory.
The authors concluded that the malformations that occurred in the
65 mg/kg/day group were caused by effects of acrylonitrile directly on the fetus,
and not by effects on the dam. This conclusion was based on historical data
indicating that mild stress, as caused by acrylonitrile treatment, did not lead
to these types of malformations and on the lack of correlation between gross
signs of toxicity in the dams (e.g., poor weight gain, and decrease in food
consumption) and litters with malformed fetuses. There also appeared to be a
dose response relationship because the 25 mg/kg/day group and the 80 ppm exposure
group also showed increased incidences of malformed fetuses. This was only
suggestive, however, since these increases were not statistically significant
when compared with control animals.
The teratogenicity of acrylonitrile was also evaluated in chick embryos
(Kankaanpaa e_t al^, 1979). Acrylonitrile was injected into either the air space
or yolk sac of 3-day old White Leghorn eggs. The acrylonitrile was dissolved in
saline at concentrations of 0.001, 0.01, 0.1, 1.0, and 10 umol, and a total of
25 ul was used for injection. The only observable adverse effect of treatment
was embryo toxicity (dead embryos and empty eggs) of approximately 100$ and Wf>
at the two highest concentrations when acrylonitrile had been injected into the
air space. When acrylonitrile was injected into the yolk sac, only the 10 uraol
13-55
-------
dose was toxic (60$) . There was no evidence of a teratogenic effect of acrylo-
nitrile treatment in this study.
Beliles et al. (1980) performed an extensive three-generation reproductive
study on Charles River rats that ingested either 100 or 500 ppm acrylonitrile in
drinking water. The rats were initially exposed following weaning, and mated
100 days after the start of the experiment. Two litters were produced from each
female and members of the second litter were used to produce the next generation
(rats from the first litter were used only when an insufficient number of pups
was available from the second litter). The parents of the first generation
showed some adverse effects of treatment in the 500 ppm group, with food and
water consumption and body weights significantly lower than those of control
animals (other generations were not monitored for these parameters). Reproduc-
tive toxicity was observed in the two matings of the first generation, with an
increased number of deaths during the lactation period among pups of animals that
had been treated at the 500 ppm level. These deaths may have been a result of
acrylonitrile's effect on the dams, since pups fostered by untreated dams had
normal survival. In the other generations, reproductive capacity and pup sur-
vival were within the normal range. The only adverse effect observed in pups
that survived treatment was a decrease in body weight in the 500 ppm group
(Table 13-16).
Poor weight gain in the pups may have been caused by poor lactation in the
dams due to the decreased water consumption. Ten weanlings of each sex from the
control and high dose groups of the F3b litter were sacrificed and examined for
histopathological changes. No adverse findings were noted in the tissues listed
in Table 13-17. It appeared that acrylonitrile had little direct effect on the
development of the embryo and pup up to the time of weaning.
13-56
-------
13.3.1 Summary and Conclusion
Acrylonitrile adversely affected pup survival following exposure of
pregnant rats and, in one study, produced teratogenic events. In a three-
generation study in which rats were exposed to 500 ppm acrylonitrile in the
drinking water, there was reduced pup survival in the first generation (Beliles
et al., 1980). This was a maternal effect inasmuch as fostering the pups on
untreated dams eliminated the poor survival. Reproductive capacity was unchanged
in the other generations, and the offspring showed no adverse effects on
development. Similarly, rats exposed by inhalation to 40 or 80 ppm of acrylo-
nitrile for 6 hours a day on days 6 to 15 of gestation had no satistically
significant changes in reproductive success or fetal development (Murray et
al., 1978). Only the pups of rats administered acrylonitrile per pjs (65 mg/kg)
for days 6 to 15 of gestation had an increase in malformations (Murray et al.,
1976). This increase was in total malformations, with no statistically signi-
ficant increase occurring in any single malformation. It was concluded that
these fetal abnormalities were the result of acrylonitrile and not the result
of toxicity in the dams. Although several studies have been conducted to
evaluate the ability of acrylonitrile to cause adverse teratogenic, embryo-
toxic, and reproductive effects, the limitations of the available data do not
allow for a full assessment of these effects. However, there are data that
indicate that acrylonitrile has the potential to cause fetal malformations in
rats, in a non dose-related manner.
13-57
-------
Table 13-16. Pup Weight on Days H and 21 of Lactation
(Beliles et al., 1980)
Generation
Dose Level
(ppm)
Mean Pup
Day H
Weight (g)
Day 21
Fla
Fib
F2a
F2b
F3a
F3b
0
100
500
0
100
500
0
100
500
0
100
500
0
100
500
0
100
500
11
10
9a
10
9
10
11
10
9a
11
10
9
10
9a
8a
10
10
8a
i»2
J40
28a
38
35
3Ua
39
39a
30a
51
16
30a
13
43a
30a
U9
46
32a
ap < 0.05 using Students t-test.
13-58
-------
Table 13-17. Tissues Examined for Histopathologic Changes in
the F3b Litter (Beliles et al., 1980)
Adrenal gland
Aorta
Bladder, urinary
Bone
Bone marrow
Cerebellum
Cerebrum
Colon
Esophagus
Eye
Heart
Ileum
Jejunum
Kidney
Liver
Lung
Lymph node
Mammary gland
Muscle, skeletal
Nerve, sciatic
Ovary
Pancreas
Parathyroid
Pituitary gland
Prostate
Salivary gland
Seminal vesicle
Skin
Spinal cord
Spleen
Stomach
Testis
Thymus
Thyroid
Tongue
Trachea
13-59
-------
13.4 MUTAGENICITY
The objective of this mutagenicity evaluation is to determine whether or
not acrylonitrile has the potential to cause mutations in humans. This
qualitative assessment is based on information derived from several tests
measuring different types of genetic damage (e.g., gene mutation, chromosome
damage, DNA strand breakage). These tests include bacterial, Drosophila,
plant, cultured mammalian cell, and whole rodent assays. The results of these
studies on acrylonitrile are discussed below and each study is summarized in
Table 13-18.
13.4.1 gene mutation studies
bacteria
Acrylonitrile has been shown to cause point mutations in bacteria.
Venitt et al. (1977) evaluated the mutagenicity of acrylonitrile (purity >
99%, impurities not reported) in a plate incorporation assay with the
tryptophan-dependent Escherichia col i strains WP2 (repair proficient), WP2
uvrA (lacks excision repair), WP2 uvrApolA (lacks excision repair and DNA
polymerase I), and WP2 lexA (deficient in an error-prone pathway). In the
absence of rat liver enzyme activation, acrylonitrile produced a weak
dose-related mutagenic response in all strains except WP2 lexA when examined
at concentrations of 0, 75, and 150 umoles/plate. It should be pointed out
that acrylonitrile is volatile and in this test no precautions to prevent
evaporation were reported. If evaporation occurred under the standard test
conditions, the weak responses observed would not reflect the mutagenic
potency of acrylonitrile in this system. These reported weak responses also
are difficult to interpret because no positive control data were given and
thus, it is unclear how well the test system was responding to mutagen
13-60
-------
treatment. The responses reported were as follows: WP2 was slightly more
sensitive to the mutagenic effect of acrylonitrile than the other tester
strains; a fourfold increase in the spontaneous level was produced by 150
umole/plate compared with a threefold increase for WP2 urvA and a twofold
increase for WP2 urvApolA. The authors reported that doses above 150 umole
per plate caused a decline in the mutagenic response, which was explained by
increasing toxicity (reduction in bacterial lawn observed).
Because of the toxic effects and weak mutagenicity of acrylonitrile in the
plate test, Venitt et al. (1977) conducted a fluctuation test (which is a
sensitive assay for detecting low levels of mutagens) at concentrations that
ranged from 4 x 10"4 M to 2 x 10"3 M. The results of the fluctuation
tests confirmed the mutagenicity detected in WP2, WP2 uvrA (results based on
one experiment and data were not presented), and WP2 urvApolA by the plate
assay. In contrast to the results derived from the plate tests, WP2 uvrApolA
was more sensitive to the mutagenic effects of acrylonitrile than WP2; the
responses of WP2 and WP2 uvrA were reported as similar. Acrylonitrile again
tested negative in WP2 lexA. The lack of a mutagenic effect in WP2 lexA
suggests that acrylonitrile may produce mutations by mis-repair DNA damage,
which is believed to be associated with the generation of DNA strand breaks
(Green and Muriel 1976). When the resistance factor pKM 101 was transferred
to WP2 (strain designated WP2P) to increase the sensitivity of this strain,
WP2P (pKM 101) was reported to be more sensitive to the mutagenic effects of
acrylonitrile than WP2. A dose-related increase in mutagenic activity was
observed at a concentration range as low as 4 to 40 umoles per plate in the
plate test. These results found in £. coli strain WP2P further support the
notion that acrylonitrile is producing mutations via mis-repair DNA damage
because Salmonella strains containing the pKM 101 plasmid have been shown to
13-61
-------
be more sensitive to the mutagenic effects of chemical mutagens which cause
mutations by a error-prone DNA repair-dependent mechanism (McCann et al.
1975).
Venitt (1977) indicated in his report that activation by an exogenous S-9
system (S-9 mix prepared from livers of Aroclor 1254-induced rats) did not
enhance the mutagenic activity of acrylonitrile in £. coli; thus
acrylonitrile, according to the authors, is primarly detected as a
direct-acting mutagen in these bacteria. It should be pointed out that
because data were not presented in this report, it is uncertain whether Venitt
included an activation-dependent mutagen in his study to demonstrate that the
S-9 fraction used was functional. Thus, his negative results with the
activation system may be an experimental error.
Venitt et al. (1977) also studied acrylonitrile in Salmonella typhimurium
and reported that acrylonitrile was not detected as mutagenic for tester
strains his D3052, TA 1535, TA 100, TA 1538, and TA 98 using either plate
incorporation assays or fluctuation tests. Although mutagenic activity was
reported to be detected in strain his G46 using a fluctuation test, the
results were reported as "erratic and statistically nonsignificant". The
authors did not present the protocol; thus it is not known if mammalian liver
activation was employed in the Salmonella assays or if precautions to prevent
evaporation were taken. In addition, they did not present data to support
their negative conclusions for Salmonella. Therefore, Venitt's conclusion of
nonmutagenicity in Salmonella cannot be evaluated.
McMahon et al. (1979) reported that acrylonitrile (purity not reported)
was mutagenic in both Escherichia coli and Salmonella typhimurium strains when
screened in a qualitative gradient plate assay, but they do not indicate if
acrylonitrile required metabolic activation for activity, the concentration(s)
13-62
-------
at which activity was observed, and which bacterial strains were reverted.
Therefore, the conclusions of the authors cannot be evaluated.
Milvey and Wolff (1977) reported positive results for acrylonitrile (99%
purity, impurities not reported) in a modified Salmonella/microsome assay when
tested as a vapor. The authors emphasized that the mutagenic effects of the
acrylonitrile in Salmonella were observed only in the presence of metabolic
activation (S-9 mix prepared from livers of Aroclor 1254-induced male
Swiss-Webster mice). Data were not shown in the absence of S-9 activation.
There are major deficiencies in the reporting of the data that make
interpretation of this study difficult. In addition, Venitt (1978) pointed
out errors in calculation found in this report. The studies by DeMeester and
coworkers (discussed below), however, provide more convincing evidence for the
mutagenicity of acrylonitrile in Salmonella typhimurium.
DeMeester et al. (1978) and other investigators (Venitt 1977) have noted
that acrylonitrile is not detected as mutagenic in Salmonella when tested in a
"classical" plate incorporation assay.t DeMeester et al. (1978) were able to
detect mutagenic activity in Salmonella when acrylonitrile exposure was as a
vapor in a desiccator. Unlike the results for Escherichia coli, DeMeester
et al. (1978) found that the mutagenicity of acrylonitrile in Salmonella was
detected only in the presence of an in vitro liver microsome-activation
system. This difference is not necessarily contradictory because these are
two different bacterial test systems, which may have different metabolizing
capabilities and sensitivities. Also, as stated earlier, Venitt does not
tin the plate incorporation test, the chemical is mixed with the bacteria,
S-9 mix, and then melted top agar is immediately added. This mixture is
poured on petri plates, which are then incubated 2 to 3 days at 37°C.
13-63
-------
provide data in his report to support his conclusion that the mutagenic
activity of acrylonitrile 1s not enhanced by a S-9 activation system.
In the study by DeMeester et al. (1978), 0.15 liters of gaseous
acrylonitrile (99% purity, impurities not reported) was injected into a
desiccator and the cells were exposed for one hour. Using a gas
chromatograph, they measured the concentration of acrylonitrile to be about
0.2% in the atmosphere. The concentration of acrylonitrile on the test plates
was also measured by freezing the plate agar and analyzing the concentration
of acrylonitrile by gas chromatography. This was found to be about 200
ug/plate. In Salmonella, the toxicity of 0.15 1 of gaseous acrylonitrile was
reported as weak (cell survival 80% to 100%) after a one hour exposure, and
the toxicity was reported to be high when the exposure was either to 0.15 1 of
gaseous acrylonitrile for 2 hours or to 0.24 1 for one hour (the method used
to measure toxicity was not described). Reversion to histidine prototrophy
was reported for the base-pair substitution-sensitive strains TA 1538, TA
1535, TA 1530, TA 100, TA 1950, and for the frameshift-sensitive strains TA
1978, TA 98, in the presence of microsomal activation (Aroclor 1254-induced
rats, 300 ul S-9/ml mix). Acrylonitrile had the most pronounced effect on the
base-pair substitution-sensitive strains TA 1530, TA 1535, and TA 1950, with
TA 1530 exhibiting the highest number of revertants over the spontaneous
number (up to approximately a 15-fold increase over the negative control).
Activity in base-pair substitution sensitive strains would be consistent with
the properties of an alkylating agent. The frameshift-sensitive strains TA 98
and TA 1978 and the base-pair substitution-sensitive strain TA 100 were only
weakly reverted (approximately a twofold or less increase over spontaneous
level). Negative results were found with the strains TA 1975, TA 1532, TA
1537 and his G46. A fluctuation test was conducted to confirm the sensitivity
13-64
-------
of TA 1530 to the mutagenic effects of acrylonitrile. Mutagenic activity was
detected at a concentration as low as 2.5 ug/ml (P < 0.001). Although
DeMeester and coworkers only evaluated one dose and thus do not demonstrate a
dose-related mutagenic response, their studies do indicate a clear increase in
the number of revertants over the background number which appears to be a
reproducible increase.
When the amount of acrylonitrile needed to induce a mutagenic response and
the magnitude of the response are considered, acrylonitrile appears to be a
weak/moderate mutagen in Salmonella. However, it is difficult to
unequivocally classify the mutagenic activity of acrylonitrile from DeMeester
et al. (1978) studies because no concurrent positive control chemicals were
evaluated and only one dose was tested. Because of this, Monsanto (Dr. L.
Kier 1982) conducted a modified liquid suspension assay using Salmonella
exposed to acrylonitrile and several standard mutagens
(4-nitroquinoline-N-oxide, 2-acetylaminofluorene, 2-aminoanthracene,
benzo[a]pyrene) to determine the relative activity of acrylonitrile to these
known mutagens. In this test, the bacteria, chemical, and S-9 mix (Aroclor
1254-induced rat liver) were mixed and incubated for two hours at 30°C. After
a two hour exposure to concentrations of acrylonitrile that ranged from 0.02
mg/ml to 20 mg/ml, the cells were washed, mixed with top agar, plated in petri
dishes and incubate at 37°C for 2 days before scoring. Viability was not
quantitatively determined in these experiments (only the observation of
bacterial lawn growth or presence of small colonies). It was concluded that
acrylonitrile was weakly mutagenic in tester strain TA 100 with S-9 activation
but not towards TA 98 and TA 1535. No increased responses were observed
without S-9 mix. In TA 100, average revertant counts at the highest dose (6
mg/ml) tested was about 1.5-fold over the solvent control. Therefore, under
13-65
-------
the liquid suspension treatment conditions used by Monsanto, acrylonitrile
caused only a marginal Increase in the number of revertants.
The bacterial studies by DeMeester et al. (1978) and Monsanto (1982)
indicate that acrylonitrile was detected as mutagenic only when metabolized j_n
vitro by using homogenates of mammalian liver. In addition to acrylonitrile
being activated by in vitro S-9 systems, Lambotte-Vandepaer et al. (1980)
examined the ability of the whole mammal to metabolize acrylonitrile. These
authors reported that urine (0.1 ml per plate) from both rats (adult male
Wistar) and mice (NMRI) treated with a single intraperitoneal dose of
acrylonitrile (30 mg/kg; purity 99%, impurities not reported) was mutagenic in
Salmonella strain TA 1530 (9-fold increase in the spontaneous level of
revertants when exposed to mouse urine and 13-fold increase for exposure to
rat urine). S-9 mix (from Arcolor-1254 treated mice) added to the test plates
caused a fivefold reduction in the mutagenicity of urine from
acrylonitrile-treated rats. Only a slight reduction (1.3-fold) was seen with
the urine of acrylonitrile-treated mice. When animals were pretreated with
phenobarbital and exposed to acrylonitrile, the urine from rats produced no
detectable mutagenicity and the mutagenicity of urine from mice was reduced.
The addition of B-glucuronidase to the test plates only slightly enhanced
(approximately 1.4-fold) the mutagenic effects of urine collected from
acrylonitrile-treated rats and mice (B-glucuronidase is added to cleave
possible conjugates). However, 3-glucuronidase had a marked effect on the
mutagenicity of urine from rats treated with phenobarbital and exposed to
acrylonitrile (an eightfold increase in mutagenic activity observed). The
authors indicated that XAD-2 resin concentrates of urine from all treatment
groups were not detected as mutagenic (data not presented), suggesting that
the mutagenic metabolite(s) was very hydrophilic. Therefore, the urine from
13-66
-------
acrylonitrile-exposed rats and mice was mutagenic and glucouronoconjugation
appeared to play a minor role in the deactivation of the acrylonitrile
derivative(s) (except in the case of phenobarbital/acrylonitrile-treated rats,
where a greater effect was found).
Conner et al. (1979) also examined whole mammal activation of
acrylonitrile and found that bile obtained from rats treated with
acrylonitrile (i.p. injection of 45 mg/kg body weight) was not mutagenic in
Salmonella. However, these negative findings cannot be interpreted because
data were not presented in the report and several known mutagens (e.g.,
dimethylnitrosamine, 3-methyl-cholanthrene), which require metabolic
activation, and the direct-acting mutagen methyl methanesulfonate were also
found as negative in this study.
The Salmonella studies indicate that metabolic activation of acrylonitrile
is necessary to detect mutagenic activity. A potential reactive metabolite of
acrylonitrile is an epoxide. Monsanto (Dr. L. Kier, 1982)) evaluated the
acrylonitrile epoxide, 2,3-epoxypropionitrile, (purity 97.4%) in a modified
liquid suspension Salmonella assay and found it to be mutagenic in tester
strains TA 100, TA 1535, and TA 98 in the absence or presence of Aroclor-1254
induced rat S-9 mix. This epoxide was most active in the base-pair
substitution sensitive strains. For example, in TA 100 with S-9 mix, a
fourfold increase in revertant counts over background counts was seen at 0.4
mg/ml. This epoxide does not appear to be a strong-acting mutagen under the
liquid suspension treatment conditions used in this study compared to the
positive control mutagens (4-nitroquinoline-N-oxide, 2-acetylaminofluorene,
2-aminoanthracene, benzo[a]pyrene).
13-67
-------
Drosophila
Although acrylonitrile has been shown to cause point mutations in
bacteria, its ability to act at the level of the gene in a eucaryotic organism
has not been adequately examined. Only one eucaryotic gene mutation study was
available. This was a study by Benes and Sram (1969) in which the occurrence
of sex-linked recessive lethal mutations in Drosophila melanogaster was
examined. A 0.1% solution of acrylonitrile (purity not reported) was injected
into the abdomen of male flies. (The authors do not give the percent survival
but indicate that chemical concentrations given to male flies were the highest
possible being slightly below a lethal or sterile dose.) Although the authors
report the results as insignificant, the data can be considered suggestive of
an increase in the mutation frequency over that of the spontaneous frequency
by a factor of three. The compound, however, has not been properly evaluated
in this system and retesting is necessary to permit a final judgment on the
mutagenic activity of acrylonitrile in Drosophila. Several deficiencies in
this report were found:
1. A small sample size of flies was tested (1297 chromosomes-in order to
detect a doubling in mutation frequency 7000 chromosomes would have to
be tested, Lee et al. in press).
2. No information was given on whether clustering of mutations occurred.
3. The purity of the test material was not described.
4. Post-mating days were not given.
5. No information was reported for concurrent positive or historical
negative controls.
13.4.2 chromosomal aberration studies
The ability of acrylonitrile to induce chromosomal aberrations in vivo has
13-68
-------
been investigated in both rats and mice. Acrylonitrile was not detected as
clastogenic in two independent bone marrow assays (Rabello-Gay and Ahmed 1980,
Leonard et al. 1981). Rabello-Gay and Ahmed (1980) tested acrylonitrile
(purity reported as 99.5%, impurities not identified) for chromosomal effects
in mice (Swiss albino) and rats (Sprague-Dawley). No significant increase (at
a confidence level of 0.05) in the incidence of chromosomal aberrations (gaps,
breaks, fragments, Robertsonian translocations) were found in mouse bone
marrow cells. At least 1200 cells were examined per treatment group and male
mice were given acrylonitrile by gavage for 4, 15, and 30 days each at doses
of 7, 14, and 21 mg/kg/day. In a second experiment mice received 10, 15, and
20 mg/kg/day for the same time intervals. The toxicity of 7, 14, and 20
mg/kg/day given orally was reported to correspond to 0.25, 0.50, and 0.75 of
the LDsQ, respectively. Mortality for the acrylonitrile-treated animals was
reported (3 of the 72 animals died). The chemical also was found to be
negative when rats were exposed orally to 16 doses at 40 mg/kg/day (reported
to represent one-half of the LO^g). Because current evidence (see Abreu and
Ahmed 1980, and Guengerich et al. 1981) indicates that acrylonitrile can be
biotransformed to cyanide these authors also evaluated potassium cyanide (KCN)
and found no increase of chromosomal aberrations in the bone marrow cells of
rats treated with 16 oral doses of KCN 5 mg/kg/day (0.5 LD5g).
Leonard et al. (1981) evaluated the clastogenicity of acrylonitrile
(purity not given) in mouse bone marrow cells in vivo. These authors
conducted both chromosomal aberration and micronuclei analyses. The
percentage of chromosomal aberrations or micronuclei formation did not differ
between acrylonitrile-treated animals and negative control animals. Male mice
(NMRI) were injected intraperitoneally with a single acute dose of 20 or 30 mg
acrylonitrile/kg of body weight. The authors indicated that 30 mg/kg was the
13-69
-------
maximum dose that allows survival of mice for several weeks. For chromosomal
aberration analysis (gaps, breaks, fragments, rearrangements), 200 cells (4
animals and 50 cells/animal) were examined 6, 18, 24, 48, and 72 hours after
each treatment. The background control frequency was 0.5% cells with
chromosomal anomalies (1 gap/200 cells) and the acrylonitrile-treated animals
did not exceed 1.5% cells with chromosomal anomalies (3 gaps/200 cells). It
should be noted that gaps are not regarded as true chromosome aberrations at
this time. For micronucleus analysis, polychromatic erythrocytes were sampled
24, 48, or 72 hours after initial treatment. The background frequency was
1.8% cells with micronuclei and acrylonitrile-treated cells did not exceed
2.5% with micronuclei. No dose-related effects were seen in these studies.
Leonard et al. (1981) also conducted a mouse dominant lethal assay with
acrylonitrile. This test evaluates the DMA damaging effects of chemicals on
germinal tissue (i.e., is indicative of the potential of a chemical to
contribute to the genetic disease burden). Male mice (NMRI) were given a
single injection of 30 mg of acrylonitrile/kg of body weight followed by a
weekly mating (1 to 5 weeks). The authors concluded that there was no
difference in the percent of postimplantation loss between the background
control group and the acrylonitrile-treated groups. The findings in this
test, however, are considered inconclusive because of the high frequency of
dominant lethals observed in the negative control group (33.6%
postimplantation loss and only 41% pregnant females).
Thiess and Fleig (1978) examined the clastogenic effects in lymphocytes of
18 workers who had been exposed to acrylonitrile. The workers had been
exposed for an average of 15.3 years, and were compared with 18 age-matched
workers who had no known exposure to acrylonitrile or any other compounds that
were suspected of causing chromosomal damage. Although no information is
13-70
-------
available concerning the exact levels of exposure in the past, atmospheric
monitoring data betweeen 1963 and 1974 indicated a typical exposure level of 5
ppm for acrylonitrile, and atmospheric monitoring data between 1975 and 1977
indicated an exposure level on an average of 1.5 ppm. The authors point out
that individual workers may have been exposed to higher concentrations during
specific operations. Also, the workers could have been exposed on the job to
styrene, butadiene, ethylbenzene, butylacrylate, and diphyl (a mixture of
diphenyl and diphenyl oxide ether). For each subject, 100 metaphases were
examined. The type of chromosomal aberrations scored was not indicated in the
report except for gaps and isogaps. One hundred cells were analyzed per
person. Chromosomal aberrations in the exposed group were 1.8 +_ 1.3%
(excluding gaps) and 5.5 +_ 2.5% (including gaps and isogaps), while chromosome
aberrations in the control group were 2.0 _+ 1.6% (excluding gaps) and 5.1 +_
2.4% (including gaps and isogaps). Therefore, apparent chromosomal damage was
not detected in these workers.
Loveless (1951) and Kihlman (1961) reported negative findings for
chromosomal effects in plants (Vivia faba). However, data were not reported
in these articles and details of the protocol were not presented. Thus, the
negative conclusions of the authors cannot be evaluated.
13.4.3 other tests indicative of genetic damage
Other tests have been conducted which do not measure mutation pe_f_ If. ^ut
may indicate that acrylonitrile has the potential to cause mutations.
Chemical-adduct formation in DNA is a critical event in certain types of
mutagenesis. DNA binding studies have been conducted using radiolabeled
acrylonitrile to determine its ability to react with DNA. Guengerich et al.
13-71
-------
(1981) provided evidence that acrylonitrile can alkylate DNA ui_ vitro.t These
authors found a low level of ^C-labeled acrylonitrile binding to calf
thymus DNA when a metabolizing system (uninduced-rat liver microsomes plus
NADPH) was included in the reaction mixture. A twofold increase in DNA
binding over the response observed with uninduced rat microsomes was seen when
liver microsomes were derived from rats pretreated with phenobarbital, an
inducer of cytochrome P-450. There was a much higher level of binding to
microsomal protein than to DNA (protein to DNA ratio was approximately 20 to
1). It should be pointed out that several known mutagens (e.g. ethylene
oxide, methyl methanesulfonate) have been shown to alkylate protein to a
greater extent than DNA (Ehrenberg and Ostermar-Golkar 1980, Ehrenberg 1979).
When Guengerich and coworkers (1981) used rat brain or human liver microsomes
plus NADPH, the binding to DNA was much lower than when rat liver microsomes
were used for metabolic activation. Apparently, the in vitro rat brain and
human liver microsomes activation systems do not metabolize acrylonitrile as
effectively. These authors showed that the rat liver microsomes (or a
reconstituted cytochrome P-450 system) metabolize acrylonitrile to
2-cyanoethylene oxide, an epoxide. This epoxide was shown to be stable and to
also bind calf thymus DNA and protein.
Glutathione S-transferase appeared to play a role in the deactivation of
acrylonitrile (and/or its metabolites) in these in vitro studies by Guengerich
et al. (1981). A rat liver cytosol preparation of glutathione conjugated
2-cyanoethylene oxide at a greater rate than it conjugated acrylonitrile.
Human liver and rat brain cytosol preparations reacted with 2-cyanoethylene
oxide at a much lower rate than did rat liver, and not at a detectable rate
tAcrylonitrile has been shown to react with certain minor tRNA
nucleosides, and at a slower rate, ribothymidine and thymidine (Ofengard 1967
and 1971.
13-72
-------
with acrylonitrile. Therefore, if acrylonitrile was activated to
2-cyanoethylene oxide and reached the brain via blood circulation, this
epoxide may not be effectively inactivated. It should be cautioned, however,
that these are in vitro studies and whole mammal tests are needed to elucidate
the events in vivo. These results, however, are consistent with the results
obtained mutagenicity studies in Salmonella and in Chinese hamster ovary cells
(discussed below) in that metabolic activation was required for a detectable
response.
Monsanto has sponsored preliminary experiments to study alkylation of DNA
in various tissues of rats (Dr. Paul Wright, personal communication). During
a EPA Science Advisory Board Meeting (August 2, 1982) Monsanto indicated that
no binding of acrylonitrile to DNA was detected in these experiments.
Monsanto made the preliminary data available to EPA, however, without details
of the protocol and additional data, a definitive judgment cannot be made
regarding these whole mammal tests. It appears, however, that the specific
activity of the label was too low to detect binding in these studies. In
addition, only a nontoxic dose (approximately 10 mg/kg) was evaluated.
Therefore, further studies are needed using radiolabeled acrylonitrile with a
higher specific activity and using several doses (including a toxic dose).
If a chemical does react with DNA, it may ultimately cause DNA damage.
Parent and Casto (1979) reported that acrylonitrile caused single strand DNA
breaks in primary Syrian golden hamster embryo cells in vitro as detected by
alkaline sucrose gradient sedimentation. When cells were treated for 18
hours at concentrations of 200 or 400 ug of acrylonitrile/ml, a shift in
sedimentation patterns was observed. This did not occur at 50 ug/ml or 100
ug/ml. The authors stated that these shifts are comparable to those produced
by known carcinogens and are not seen with non-carcinogenic chemicals. They
13-73
-------
concluded that the results are suggestive of carcinogenicity. However,
because the toxicity of the concentrations tested was not given, it is
uncertain if the DNA damage observed at 200 or 400 ug/ml is simply a
reflection of nonspecific toxicity. It should be noted that these data were
generated in the absence of an exogenous metabolic activation system.
Dr. G. Williams (personal communication, 1982) evaluated the DNA damaging
potential of acrylonitrile by measuring DNA repair synthesis in a rat
hepatocyte primary culture. A major advantage of the intact hepatocyte test
system is that it provides a better approximation than an exogenous S-9 system
of the activation/deactivation of a chemical substance as it would occur in
the rat liver in vivo. Dr. Williams indicated that acrylonitrile was negative
in this system; however, he emphasized that the test agent was intensely toxic
to the hepatocytes. He found that the hepatocytes were biotransforming
acrylonitrile to cyanide, which is very toxic to the liver. Thus, these
negative findings do not necessarily indicate the inability of acrylonitrile
to damage DNA. This conclusion is supported by other studies by Dr. S. Ved
Brat and Dr. G. Williams (unpublished, 1982) that have shown a significant
increase in sister chromatid exchange formation in Chinese hamster ovary (CHO)
cells when metabolic activation is mediated by rat hepatocytes.
S. Ved Brat and G. Williams (1982, unpublished) studied the induction of
sister chromatid exchange (SCE) in CHO cells by acrylonitrile. The evaluation
of SCE formation, which is thought to involve DNA breakage, is a very
sensitive method that can detect genotoxic effects at lower concentrations
than those needed to detect an increase in chromosome aberrations (Wolff 1977,
Latt 1974). In two separate experiments, exposure to acrylonitrile did not
produce SCE in CHO cells without a metabolic activating system when evaluated
at five concentrations from 10~7M up to IQ-^M, but did cause a progressive
13-74
-------
increase in the SCE incidence when CHO cells were cocultured with rat
hepatocytes for metabolic activation. The highest response (approximately
twofold increase over background levels) was observed at 10"4M
acrylonitrile. It should be pointed out that coculturing hepatocytes with the
CHO cells resulted in a elevated background level of SCE in CHO (twice than
what is normally observed). Nevertheless, in two separate experiments an
dose-related response (with a twofold increase at the highest dose) was
demonstrated. These results indicate that the rat liver cells can mediate the
activation of acrylonitrile and that the reactive metabolite(s) apparently is
reaching the CHO cells to cause genetic damage.
13.4.4 summary and conclusions
Acrylonitrile has been evaluated for its ability to cause gene mutations
in bacteria and Drosophila and for its ability to cause chromosome damage in
plants and rodents (acrylonitrile-exposed workers have also been screened for
induced chromosome damage). Other endpoints indicative of mutagenicity have
been examined, such as DNA repair synthesis and sister chromatid exchange
formation in cultured mammalian cells and chemical-DNA adduct formation j_n_
vitro.
There is evidence that acrylonitrile can cause point mutations in both
Salmonella tyhimurium and Escherichia coli. A difference in the requirement
of acrylonitrile for exogenous hepatic metabolic activation was reported for
the Salmonella and Escherichia strains, i.e., liver S-9 activation was
required for acrylonitrile mutagenicity in Salmonella but not in Escherichia.
Further investigation would be necessary to determine the reason for this
reported difference. The epoxide of acrylonitrile (a potential metabolite)
has been shown to be mutagenic in Salmonella. Because acrylonitrile and the
13-75
-------
metabolite can cause point mutations in bacteria it may cause point/gene
mutations in other organisms as well. Results from a sex-linked recessive
lethal test in Drosophila, in which the data are considered suggestive of an
increase in mutation frequency over that of the spontaneous frequency by a
factor of three, provide some support for this expectation. Further testing,
however, is needed in other organisms to confirm the ability of acrylonitrile
to cause point/gene mutations.
Acrylonitrile has not been detected as causing chromosome damage. Studies
in plants have been reported to be negative. In rats and mice, acrylonitrile,
(when evaluated up to toxic doses) was not found to be clastogenic in the bone
marrow cells of these mammals. In addition, no apparent chromosomal damage
was detected in peripheral blood lymphocytes from workers exposed to
acrylonitrile. However, it did induce an increase in sister chromatid
exchange formation in Chinese hamster ovary (CHO) cells in vitro when
metabolic activation was mediated by intact rat hepatocytes. The induction of
SCE is thought to involve DNA breakage and this type of genetic damage can be
detected at concentrations lower than those needed to detect an increase in
chromosome aberrations. Acrylonitrile was negative in a DNA repair test using
primary rat hepatocytes. In this test, acrylonitrile was intensely toxic to
the liver cells because of the production of cyanide. Thus, the compound
cannot be adequately evaluated in the rat hepatocyte-DNA repair test system.
DNA binding studies indicate the potential of acrylonitrile to react with
DNA, a critical event in certain types of mutagenesis. A low level of binding
of radiolabeled acrylonitrile to calf thymus DNA in vitro was detected in the
presence of exogenous rat liver microsomal activation. This requirement of
metabolism is consistent with the Salmonella mutagenicity studies and SCE
studies in CHO cells. A metabolite of acrylonitrile, the epoxide
13-76
-------
2-cyanoethylene oxide, also binds calf thymus DMA in vitro. However,
additional studies are necessary to determine if DNA binding occurs in intact
cells In vitro and in whole mammals after acrylonitrile exposure.
The ability of acrylonitrile and/or an active metabolite to reach
mammalian gonads and cause germ cell mutations has not been adequately
studied. Only one study was available for review. This was a mouse dominant
lethal assay in which the test results were not significant. However, the
high background of dominant lethals in this study precluded a negative
interpretation of the results.
In conclusion, the weight-of-evidence indicates that acrylonitrile has
the potential to bind to DNA (as demonstrated by the calf thymus DNA binding
studies in vitro), and cause genetic damage (as shown in the point mutation
studies in bacteria and in sister chromatid exchange studies in mammalian
cells in vitro and as suggested in the Drosophila studies). The available
evidence indicates that conversion of acrylonitrile to a metabolite(s), such
as an epoxide, may be required to produce the ultimate mutagenic form(s).
Thus, if the metabolism and pharmacokinetics of this chemical substance in
humans results in metabolic products that can interact with DNA, as is the
case for the test systems employed in the reported studies, it may cause
somatic mutations in humans as well. An assessment of genetic risk with
respect to germ cell mutagenicity cannot be made because of the lack of
appropriate data. Additional studies are needed, particularly in eukaryotic
organisms to confirm that acrylonitrile is mutagenic as indicated by the
positive responses observed in bacteria. It would also be appropriate to
conduct tests using radiolabeled acrylonitrile to measure alkylation of DNA in
various tissues of whole mammals.
13-77
-------
A. GENE MUTATION TESTS: BACTERIA
fable 13-18. MUTAGEN1CITY TESTS OF ACRYLONITRILE
—i
00
Reference
Connor et al .
1979
DeMeester
et al. 1978
Test System
Body fluid
analysis: bile
from rats.
Salmonella
typhlmurium as
indicator organ-
ism; plate incor-
poration assay
Salmonel la
test: plate
test, fluctu-
ation assay
Strain
TA 1535
Activation
System
In vivo male Sprague-
DTwley rats
Concentration
45 mg/kg acrylo-
nitrl le in sal ine
Injected i.v. Into
femoral vein, 200 ul
of bile tested/plate
Reported
Result*
Negative 1.
2.
3.
Source: Aldrich Chemical
TA 1530
TA 1535
TA 1537
TA 1538
TA 98
TA 100
TA 1950
his G46
TA 1978
TA 1532
TA 1975
S-9 mix prepared from
1 Ivers of styrene,
Aroclor 1254, acrylo-
nitrlle, methyl -3-chol -
anthrene, phenobarbital
-induced male rats
(Wlstar); Aroclor
1254- and phenobarbital
-Induced female mice
(NMRI); uninduced male
beagle dog
Company
Tested in atmos-
phere at + 0.2%
for 1 hour, purity
reported as 99%
(Impurities not
described).
Source: Aldrich,
Europe
4.
5.
Positive 1n the 1.
presence of meta-
bolic activation
in both framesnlft
-sensitive and
base-pair sub-
stitution sen-
sitive strains, 2.
with most pro-
nounced effects In 3.
base-pair substi-
tution sensitive 4.
strains. Negative
in the absence of
metabolic activation
Comment
Negative findings are
inconclusive.
One bacterial strain
tested.
Number of animals used not
reported.
Number of revertants not
reported.
Purity of chemical not
given.
Responses are considered to
Indicate mutagenic activity
because there were clear
Increases in revertant colonies
over the background frequencies
and these increases were
reproducible.
No concurrent positive controls
reported.
Single dose tested; thus, no
dose-response curve.
Method to determine toxlcity
not described.
•Authors conclusions.
(continued on following page)
-------
Table 13-1B. (continued)
A. GENE MUTATION TEST: BACTERIA
GO
I
in
Reference
Kier
1982
Monsanto
Report
K1er
1982
Monsanto
Report
Test System
Salmonella/
micros ome
assay:
modi fied
1 iquld
suspension
test
Salmonella/
micros ome
assay:
modi fied
liquid
suspension
test
Activation
Strain System
TA 98 S-9 mix from
TA 100 livers of
TA 1535 Aroclor 1254-
induced rats
TA 98 S-9 mix from
TA 100 livers of
TA 1535 Aroclor 1254-
induced rats
Concentration
Acrylonitrile
0.02 mg/ml to
20 mg/ml . 1 iquld
material diluted
in distilled H20
Purity: > 991
2,3-epoxypropionitri le
(acrylonitrile epoxlde)
0.2 mg/ml to 20 mg/ml
diluted in distilled
H20
Purity: 97.4%
Reported
Result
Weak mutageniclty 1.
in strain TA 100
only after metabolic 2.
activation
Mutagenic in strains 1.
TA 100 and TA 1535,
TA 100 as the most
sensitive strain.
Mutagenic activity
expressed in absence
of S-9 mix and a si ight
Increase was found in
the presence of S-9 mix.
Comments
Results considered
border! ine at best.
A much higher concen-
tration (43 ug/ml )
was used for the
TA 1535 positive control
2-amlnoanthracene
than is generally '
employed (2.5
ug/plate). The
response of this
control was weak.
Therefore, results In
tester strain TA 1535
are difficult to
Interpret.
These data demonstrate
the mutageniclty of
acrylonitrile epoxlde.
(continued on the next page)
-------
A. GENE MUTATION TESTS: BACTERIA
Table 13-18. (continued)
00
i
oo
o
Reference
Lambotte-
Vandepaer
et al. 1980
McMahon
et al. 1979
Test System Strain
Body fluid TA 1530
analysis:
urine from rats
and mice,
Salroonel la
typhimurium
as indicator
organism; plate
incorporation
assay
Reversion In
Salmonel la
typhimurium
and Escherichia
coll; gradient
plate test
Activation
System
In vivo adult male
FTstar rats and
adult male NMRI
mice; in vitro,
S9 mixTrom livers
of Aroclor 1254-
induced mice
(strain not
reported)
Concentration
30 mg/kg (purity >
99%, impurities
not described; in
saline injected
1 . p . ; urine
collected 24 hours
Source: Aldrich,
Europe
Reported
Result Comments
Positive 1. Single dose tested, thus
no dose-response curve
determined, but the data
are still considered to
Indicate mutagenic activity
2. No concurrent positive
control data reported.
Reported as 1. Positive conclusions cannot be
positive for both evaluated because of lack of
Salmonella and data.
Escherichia 2. Parity and source of chemical
not given.
3. Authors did not Indicate 1f
mutagenic activity was
dependent on exogenous
metabolic activation.
4. Concentratlon(s) where activity
was observed not reported.
5. Bacteria strains, which were
reverted, were not given.
6. Information on toxlcity not
reported.
(continued on following page)
-------
A. GENE MUTATION TESTS: BACTERIA
Table 13-18. (continued)
CO
oo
Reference
HI Ivy and
Wolff 1977
Test System Strain
Salmonella TA 1535
test: spot TA 1538
assay, liquid TA 1978
suspension.
plate test (In
vapor phase)
Activation
System
S9 mix from
1 i vers of
Aroclor 1254-
induced male
Swiss-Webster
nice
Concentration
Spot test: 5 and
10, ul , liquid
suspension: 5, 10,
20 ul for 30 min.
to 4 hrs
Source: Aldrich
Chemical
Company
Purity: Reported
as 99+J
Reported
Result
Liquid suspension 1.
and vapor: positive
only In the pre-
sence of metabol 1c
activation in both 2.
frameshlft and base-
pair substitution 3.
sensitive strains;.
negative results using
spot test and testing
In the absence of
metabolic activation
Comments
Positive conclusions are
difficult to interpret
because of study design
problems and lack of data
No clear dose-response observed
with vapor exposure.
Neither the actual revertant
counts per plate not the
standard deviation per dose
was given; thus the extent of
variation Is not known.
(continued on following page)
-------
Table 13-18. (continued)
A. GENE MUTATION TESTS: BACTERIA
u>
i
oo
ro
Reference
Venltt
et al. 1977
Test System
Escherichia
col i test:
plate incor-
poration ,
fluctuation
assay
Salmonella
test: plate
Incorporation,
fluctuation
assay
Strain
WP2
WP2 urvA
WP2 urvApolA
WP2 lexA
WP2P (pKM 101)
TA 100
TA 1538
TA 98
TA 1535
his G46
his 03052
Activation
System
S9 mix from 1 ivers
of Aroclor 1254-
induced CB hooded
male rats, 0.5 ml/
plate
Concentration
0, 75, 150 umole/
plate for plate
incorporation
assay 4xlO'4M to
2xlO-3M for
fluctuation assay.
Dissolved in DMSO,
(purity > 99t as
determined by GLC) ;
Source: Kochlight
Lab
Reported
Result
Weak positive in
the absence of
metabol ic activa-
tion for al 1
strains except
UP2 lexA
Reported as
negative
Comments
1. Clear dose-related response.
2. No concurrent positive control
reported.
3. Data in the presence of
metabolic activation not
reported.
4. No precautions to prevent
evaporation were reported.
1. Negative conclusion cannot
be evaluated because data
were not presented and
protocol was not described.
(continued on following page)
-------
Table 13-18. (continued)
B. GENE MUTATION TESTS: DROSOPHILA
Reference
Test System
Dose
Result
Brood I Brood II
(postmeiotic) (melotlc and premeiotic)
Total
Comments
Benes and
Sram 1969
oo
I
oo
CO
Sex-linked
recessive
lethal test
In Drosophlla
melanogaster
Injection of a
0.1% solution
Into abdomen
spontaneous
frequency
2/572 (0.351)
4/725 (0.55%)
0.46X
58/41.921 (0.14%)
1. Results considered suggestive of
three-fold Increase in mutation
frequency.
2. Toxlcity not given.
3. Sufficiently large sample size not
analyzed (7000 chromosomes are
needed to detect a doubling 1n
mutation frequency).
4. Purity and source of chemical
was not given.
5. No concurrent positive controls
and information concerning
negative controls not given;
appears to be historical.
6. Post-mating times not given.
(continued on following page)
-------
Table 13-18. (continued)
C. HAMMALIAN IN VIVO CYTOGENETICS
Reference
Test System Strain
Reported
Concentration Result
Comments
Leonard.
et al.
1981
Bone marrow
assay
(micronucleus
and chromosomal
aberration
analysis)
Dominant lethal
assay
NMRI male mice
NMRI male mice
Mice give single i.p. Injection Negative
of 20 or 30 mg/kg and cells
examined 6, 18, 24, 48, and 72
hours later for chromosome analysis
and 24, 30, and 48 hours later for
micronucleus analysis.
Mice given I.p. treatment of Negative
30 mg/kg followed by weekly
matings (1 to 5 weeks).
Source: AldMch Europe
(purity not given)
1. Aberrations scored were gaps,
breaks, and rearrangements.
1. Results considered Inconclusive
because background control
frequency was very high (33.6%
postimplantation loss).
OJ
i
CO
Rabello-
Gay and
Ahmed 1980
Bone marrow
assay
Albino Swiss
mice. Sprague-
Dawley rats
Mice given at 7, 14, and 20
mg/kg/day by gavage
for 4, 15, 30 days ;
rats exposed orally
to 16 da1ly doses of
40 mg/kg/day (purity
99.5i).
Source: AldMch Chem. Co.
Negative 1. Aberrations scored were gaps,
Isogaps. breaks, fragments, and
translocatlons.
Thelss and
Flelg 1978
Examination of
lymphocytes from
workers with
known exposure
to acrylonltrlle
Atmospheric monitoring
data indicated a
typical exposure level
of 5 ppm between
1963 and 1974, and a
level of 1.5 ppm
between 1975 and 1977
Negative
Cohort was exposed to
acrylonltrlle for an average of
15.3 years.
Aberrations scored included gaps
and Isogaps; other aberrations
scored not defined in the report.
(continued on following page)
-------
Table 13-18. (continued)
0. OTHER TESTS INDICATIVE GENETIC DAMAGE
Reference
Guengerlch
et a). 1981
Test System Strain
DNA adduct
formation
in vltro-
cFlf thymus
DNA
Activation
System
Rat and
human liver
microsomes.
and puri fied
rat liver
P-450
Concentration
ImM 14C labeled
acrylonltrile (purity
> 99%) for 1 hour.
14C labeled
2-cyanoethylene oxide,
0.25 mM at different
times up to 4 hours
Reported
Result Connents
Low level of DNA binding
detected with rat liver
activation for acrylonitrlle,
2-cyanoethylene also
binds DNA In vitro.
I
CO
Parent and
Casto 1979
Detection of
single strand
DNA breaks by
alkaline sucrose
gradient
Primary Syrian None Cells Incubated for
Golden hamster 18 hours at 50, 100,
embryo cells 200, and 400 ug of
In vitro acrylonitrile/ml
(>_ 99% pure); and
contains 30-45 ppm
l-hydroxy-4-methoxybenzene
0.3% water, < 0.5% acetonitrile.
Source: Aldrlch Chemical Co.
DNA breakage at
200 and 400 ug/ml
but not at 50 or
100 ug/ml.
1. Toxlclty data not given;
thus, it is unclear
whether the results are
due to nonspecific toxlclty.
(continued on the next page}
-------
Table 13-18. (continued)
D. OTHER TESTS INDICATIVE OF GENETIC DAMAGE
Activation
Reference Test System Strain System Concentration
Ved Brat and Sister chromatid Intact rat Cells were exposed
Williams exchange (SCE) hepatocytes 3 hours to
1982 in CHO eel Is jji acrylonitri le
Unpublished vitro coculture dissolved in DMSO.
with rat hepatocytes Dose range from
10-7 H to 10-3M.
Purity: Not given
OJ
i
co
CT>
Concentration
Positive Control DMBA 5 x lO"5
Acrylonitri le 10"3
lo-^
10-5
ID"6
10-7
DMSO (1%)
Reported
Result
Negative results
for CHO cells
without hepatocytes.
Positive results for
CHO with hepatocytes
(M) Experiment
33.5 + 1.8
55.7 +_ 3.6
28.2 +_ 1.2
24.9 +; 1.4
23.7 jf 1.2
21.7 _* 1.7
Comments
1. Response was dose-
related In two
separate
experiments.
2. Background level of
SCE was very high.
SCE/CELL
1 Experiment 2
54.5 + 2
Toxic
45.6 +_ 1.5
36.4 + 1.0
30.9 ± 1.9
24.9 + 1.9
-------
13.5 CARCINOGENICITY
The purpose of this section is to provide an evaluation of the likelihood
that acrylonitrile is a human carcinogen and, on the assumption that it is a
human carcinogen, to provide a basis for estimating its public health impact
including a potency evaluation in relation to other carcinogens. The evaluation
of carcinogenicity depends heavily on animal bioassays and epidemiologic
evidence. However, information on mutagenicity and metabolism, particularly in
relation to interaction with DNA, as well as the pharmacokinetic behavior, have
an important bearing on both the qualitative and quantitative assessment of
carcinogenicity. The available information on these subjects is reviewed in
other sections of this document. This section presents an evaluation of the
animal bioassays, the human epidemiologic evidence, the quantitative aspects of
assessment, and finally, a summary and conclusions dealing with all of the
relevant aspects of carcinogenicity.
13.5.1 ANIMAL STUDIES
Seven studies in which acrylonitrile was administered to rats will be
discussed. In four studies (three cancer bioassays and one three-generation
reproductive study), the route of administration was via drinking water, one
study was via gavage, and two (cancer bioassays) were via inhalation.
Drinking Water Studies
Dow Chemical Company (Quast et al. 1980a)--
Quast et al. (1980a) of the Dow Chemical Company performed a 2-year chronic
study under the auspices of the Chemical Manufacturers Association. The
acrylonitrile used in this study was produced by E.I. du Pont de Nemours and
13-87
-------
Company, Inc. Its purity was greater than 99fc. In this study 6-to 8-week-old
male and female Sprague-Dawley rats (48 animals of each sex at each exposure
level and 80 animals of each sex in the control group) were given acrylonitrile
in drinking water. For the first 21 days, the concentrations given were 35, 85,
and 210 ppm; however, the two higher concentrations were subsequently raised to
100 and 300 ppm. After 9 months of treatment, the animals at the two higher
doses showed signs of toxicity as indicated by decreased self-grooming and an
unthrifty appearance. There was also a dose-related decrease in food
consumption (except in male rats at the lowest exposure level), and concomitant
dose-related decrease in water consumption. Using water consumption data and
the weight of the rats, the calculated mean amount of acrylonitrile ingested for
males was 3.42, 8.53, and 21.18 mg/kg/day, and for females, 4.36, 10.76, and
24.97 mg/kg/day for the 35, 100, and 300 ppm exposure levels, respectively.
Cumulative mortality data for male and female rats at all exposure levels
are presented in Tables 13-19 and 13-20. The results indicate that early
mortality was observed at all dose levels in females.
On death or at necropsy at 24 months, histopathologic examination was
performed on complete sets of tissues from the control and high dose animals and
on selected target tissues and tissues with grossly recognized tumorous changes
from the other exposure groups.
Statistically significant increases in tumor incidence at multiple sites in
male and female rats exposed to acrylonitrile were observed as described in Tables
13-21 and 13-22. Statistically significant increases in the incidence of
specific tumor types with respect to individual treatment groups are summarized
as follows: central nervous system tumors in males and females in all treatment
groups; Zymbal gland tumors in females in all treatment groups and in males in
the high-dose group; tumors in the nonglandular portion of the stomach in males
13-88
-------
TABLE 13-19.
CUMULATIVE MORTALITY DATA OF MALE RATS MAINTAINED FOR 2 YEARS ON
DRINKING WATER CONTAINING ACRYLONITRILE
(Quast et al. 1980a)
Days on test
0-30
31-60
61-90
91-120
121-150
151-180
181-210
211-240
241-270
271-300
301-330
331-360
361-390
391-420
421-450
451-480
481-510
511-540
541-570
571-600
601-630
631-660
661-690
691-720
721-745
Total number
of rats
Control
No. Dead
(% dead)
0
0
0
0
0
1(1.3)
1(1.3)
1(1.3)
2(2.5)
3(3.8)
6(7.5)
7(8.8)
7(8.8)
8(10.0)
11(13.8)
15(18.8)
21(26.3)
29(36.3)
33(41.3)
40(50.0)
48(60.0)
54(67.5)
67(83.8)
70(87.5)
73(91.3)
80(10056)
35 ppm
No. Dead
(% dead)
0
0
0
0
0
0
1(2.1)
1(2.1)
1(2.1)
1(2.1)
2(4.3)
2(4.3)
3(6.4)
5(10.6)
7(14.9
8(17.0)
13(27.7)
14 29.8
20(42.6)
23(48.9)
27(57.4)
33(70.2)
39(83.0)
41(87.2)
42(89.4)
47(100%)
100 ppm
No. Dead
(% dead)
0
0
0
1(2.1)
1(2.1)
1(2.1)
1(2.1)
1(2.1)
1(2.1)
1(2.1)
1(2.1)
1(2.1)
1(2.1)
2(4.2)
5 10.4)
7(14.6)
11(22.9)
16(33.3)
18(37.5)
23(47.9)
26(54.2)
33(68.8)
35(72.9)
40(83.3
43(89.6)
48(100%)
300 ppm
No. Dead
(% dead)
0
0
0
0
0
0
0
0
2(4.2)
2(4.2)
2(4.2)
4(8.3)
7(14.6)
8(16.7)
11(22.9
15(31.3)
22(45.8)*
29(60.4)*
32(66.7)*
34(70.8)*
37(77.1)*
40(83.3)*
48(100)*
48(100)*
48(100)*
48(100%)
*Sign1f1cantly different from controls by Fisher s Exact Probability Test,
P < 0.05.
13-89
-------
TABLE 13-20.
CUMULATIVE MORTALITY DATA OF FEMALE RATS MAINTAINED FOR 2 YEARS ON
DRINKING WATER CONTAINING ACRYLONITRILE
(Quast et al. 1980a)
Days on test
0-30
31-60
61-90
91-120
121-150
151-180
181-210
211-240
241-270
271-300
301-330
331-360
361-390
391-420
421-450
451-480
481-510
511-540
541-570
571-600
601-630
631-660
661-690
691-720
721-745
Total number
of rats
Control
No. dead
(% dead)
0
0
0
0
0
0
1(1.3)
1(1.3)
1(1.3)
1(1.3)
1(1.3)
1(1.3)
1(1.3)
3(3.8)
3(3.8)
6(7.5)
9(11.3)
11(13.8)
18(22.5)
22(27.5)
34(42.5)
37(46.3)
45(56.3)
54(67.5)
60(75.0)
80(100%)
35 ppm
No. dead
(% dead)
0
0
0
0
0
0
0
0
0
0
1(2.1)
1(2.1)
3(6.3)
3(6.3)
5(10.4)
7(14.6
10(20.8)
12(25.0)
20(41.7)*
24(50.0)*
27(56.3)
33(68.8)*
38(79.2)*
42(87.5)*
44(91.7)*
48(100%)
100 ppm
No. dead
(% dead)
0
0
0
0
0
0
1(2.1)
1(2.1)
2(4.2)
3(6.3)
3(6.3)
3(6.3)
5(10.4)*
6(12.5)
6(12.5)
11(22.9)*
13(27.1)*
17(35.4)*
27(56.3)*
34(70.8)*
38(79.2)*
42(87.5)*
44(91.7)*
46(95.8)*
47(97.9)*
48(100%)
300 ppm
No. dead
(% dead)
0
0
0
0
1(2.1)
1(2.1)
1(2.1
1(2.1)
3(6.3)
4(8.3)
9(18.8)*
14(29.2)*
16(33.3)*
20(41.7)*
24(50.0)*
31(64.6)*
35(72.9)*
37(77.1)*
42(87.5)*
45(93.8)*
46(95.8)*
46(95.8)*
48(100)*
48(100)*
48(100)*
48(100%)
P < 0.05.
13-90
-------
TABLE 13-21. HISTOPATHOLOCIC DIAGNOSES AND TUMOR INCIDENCES IN MALE RATS
MAINTAINED FOR 2 YEARS ON DRINKING WATER CONTAINING ACRYLONITRILE
(Quast et al. 1980a)
OJ
i
Diagnoses
Ear Canal Gland (Zymbal gland)
carcinoma
Tongue
squamous cell tumors
(papHloma and/or carcinoma)
Stomach - Nonglandular portion
papllloma, squamous cell
Stomach - Nonglandular portion
squamous cell tumors
papllloma and/or carcinoma
Brain and/or Spinal Cord
focal or multlfocal gllal
cell proliferation suggestive
of early tumor
Brain and/or Spinal Cord
gllal cell tumor
(astrocytoma)
Brain and/or Spinal Cord
gllal cell tumor
(benign and/or malignant)
0 ppm (%)
3/80 (3.8)
1/75 (1.3)
0/80 (0.0)
0/80 (0.0)
0/80 (0.0)
1/80 (1.3)
1/80 (1.3)
Tumor Incidences
35 ppm (%)
4/47 (8.5)
2/7§
2/46 (4.3)
3/46 (6.5)
P = 0.047t
4/47 (8.5)
P = 0.017t
8/47 (17.0)
P = 1.5 x 10~3t
12/47 (25.5)
P = 2.27 x 10'5t
at Various Dose Levels*
100 ppm (%)
3/48 (6.3)
4/9§
P = 3.10 x 10"4t
20/48 (41.7)
P = 1.40 x 10-10t
23/48 (47.9)
P = 2.24 x 10'12t
4/48 (8.3)
P = 0.0181
19/48 (39.6)
P = 7.86 x 10"9t
22/48 (45.8)
P = 1.61 x 10-lot
300 ppm (X)
15/48 (31.3)
P = 2.57 x 10-5t
5/40§
P = 0.019f
34/47 (72.3)
P = 1.59 x 10-20t
39/47 (83.0)
P = 3.95 x 10-25t
8/48 (16.7)
P = Z.64 x 10-"
23/48 (47.9)
P = 4.16 x lO"111
30/48 (62.5)
P = 1.21 x 10-15t
•Tumor Incidence expressed as number of rats bearing certain tumor types over the number of rats examined
microscopically for that particular organ.
tp-values calculated using the one-tailed Fisher Exact Test.
§A11 samples were not taken from all animals.
-------
TABLE 13-22. HISTOPATHOLOGIC DIAGNOSES AND TUMOR INCIDENCES IN FEMALE RATS
MAINTAINED FOR 2 YEARS ON DRINKING WATER CONTAINING ACRYLONITR1LE
(Quast et al. 1980a)
<£>
ro
Diagnoses
Mammary Gland
f Ibroadenoma/
adenof Ibroma
Mammary Gland
benign and/or malignant
mammary gland tumors
Ear Canal Gland (Zymbal gland)
adenoma
Ear Canal Gland (Zymbal gland)
carcinoma
Ear Canal Gland (Zymbal gland)
benign and/or malignant tumors
Tongue
carcinoma,
squamous cell
Tongue
papllloma and/or carcinoma,
squamous cell
Stomach - Nonglandular
papllloma, squamous cell
0 ppm (%)
56/80 (70.0)
57/80 (71.3)
0/80 (0.0)
1/80 (1.3)
1/80 (1.3)
0/78 (0.0)
0/78 (0.0)
1/80 (1.3)
Tumor Incidences
35 ppm (%)
41/48 (85.4)
P = 0.037t
42/48 (87.5)
P = 0.026f
1/48 (2.0)
4/48 (8.3)
5/48 (10.4)
P = 0.028t
1/5
1/5
1/47 (2.1)
at Various Dose Levels*
100 ppm (%)
39/48 (81.3)
42/48 (87.5)
P = 0.026t
3/48 (6.3)
P = 0.051*
6/48 (12.5)
P = 0.011*
8/48 (16.7)
P = 1.67 x 10-3t
0/3 (0.0)
2/3 (66.7)
P = 9.26 x 10~4t
12/46 (25.0)
P = 2.72 x 10-5t
300 ppm (%)
25/48 (52.1)
35/48 (72.9)
4/48 (8.3)
P = 0.018f
14/48 (29.2)
P = 3.00 x 10-6t
18/48 (37.5)
P = 2.72 x 10'8t
10/45 (22.2)
P = 2.13 x 10"5t
12/45 (26.7)
P = 2.00 x 10~6T
29/48 (60.4)
P = 6.03 x 10'15t
•Tumor Incidence expressed as number or rats bearing certain
microscopically for that particular organ.
tp-values calculated using the one-tailed Fisher Exact Test.
§A11 samples were not taken from all animals.
tumor types over the number of rats examined
(continued on the following page)
-------
TABLE 13-22. (continued)
h-»
OJ
1
10
CO
Diagnoses
Stomach - Nonglandular portion
carcinoma, squamous cell
Stomach - Nonglandular portion
papllloma and/or
carlnoma, squamous cell
Small Intestine
cystadenocardnoma, muclnous
Brain and/or Spinal Cord
focal or multlfocal glial
cell proliferation suggestive
of early tumor
Brain and/or Spinal Cord
focal or multifocal glial
cell tumor (astrocytoma)
Brain and/or Spinal Cord
focal or multlfocal glial
eel 1 tumors
(benign and/or malignant)
0 ppm (%)
0/80 (0.0)
1/80 (1.3)
0/80 (0.0)
1/80 (1.3)
0/80 (0.0)
1/80 (1.3)
Tumor Incidences at
35 ppm (%)
0/47 (0.0)
1/47 (2.1)
l/7§
5/48 (10.4)
P = 0.028f
17/48 (35.4)
P = 6.90 x 10-9t
20/48 (41.7)
P = 2.21 x 10-9*
Various Dose Levels*
100 ppm (%)
0/48 (0.0)
12/48 (25.0)
P = 2.72 x 10'5t
4/ll§
P = 1.23 x 10-4t
4/48 (8.3)
22/48 (45.8)
P = 9.15 x 10-12t
25/48 (52.1)
P = 2.52 x 10'12t
300 ppm (%)
12/48 (25.0)
P = 2.94 x 10~6t
30/48 (62.5)
P = 1.21 x 10-15t
4/48 (8.3)
P = 0.018t
7/48 (14.6)
P = 4.38 x 10'3t
24/48 (50.0)
P = 5.34 x lO'13*
31/48 (64.6)
P = 2.31 x 10'16t
*Tumor Incidence expressed as number of rats bearing certain tumor types over the number of rats examined
microscopically for that particular organ.
tp-values calculated using the one-tailed Fisher Exact Test.
§A11 samples were not taken from all animals.
-------
in all treatment groups and in females in the mid- and high-dose groups; tongue
tumors in males in all treatment groups and in females in the mid- and high-dose
groups; mammary gland tumors in females in the low- and mid-dose groups; tumors
in the small intestine in females in the mid- and high-dose groups.
The central nervous system tumors included a statistically significant
increase in the incidence of astrocytomas. There was also a statistically
significant increase in the incidence of glial cell proliferations, suggestive
of early tumors, which were observed most frequently in the cerebral cortex.
Zymbal gland tumors were observed in the ear canal, and these tumors were
usually ulcerated and, in some animals, caused displacement of the lower jaw
that resulted in the inhibition of food consumption. In the nonglandular
portion of the stomach, both papillomas and carcinomas were found with a
progression from hyperplasia and hyperkeratosis to papilloma and, finally, to
carcinoma. Tongue tumors were diagnosed as squamous cell papillomas and
carcinomas, and tumors in the small intestine were identified as
cystadenocarcinomas.
Biodynamics, Inc. Study in Sprague-Dawley Rats (1980a)--
A report entitled "A Twenty-four Month Oral Toxicity/Carcinogenicity Study
of Acrylonitrile Administered to Spartan Sprague-Dawley Rats in the Drinking
Water," dated June 30, 1980, was submitted to the U.S. Environmental Protection
Agency by the Monsanto Company, St. Louis, Missouri. This study, conducted by
Biodynamics, Inc. for the Monsanto Company, was designed to evaluate the
toxicity/carcinogenicity of acrylonitrile. Acrylonitrile (100% pure, supplied
by the Monsanto Company) was administered in the drinking water to 100
Sprague-Dawley rats of each sex at dose levels of 0, 1, and 100 ppm. Interim
necropsies were performed at 6, 12, and 18 months (10/sex/group). The study was
13-94
-------
terminated early due to low survival rates; females were sacrificed at 19 months
and males were sacrificed at 22 months.
Body weights for the high-dose males and females were consistently lower
than the weights of controls; body weight differences between controls and
treated male rats were less than 10%, while the body weight differences between
controls and treated female rats were less than 8%. Body weights for the
low-dose males and females were generally comparable to the controls throughout
the study. Reduced water intake (test substance) was consistently noted in the
high-dose group. Water intake for the low-dose group was generally comparable
to the controls throughout the study. Slight, but consistent decreases in
hemoglobin concentration, hematocrit, and erythrocyte counts were noted for the
high-dose males and females.
Histopathology evaluation revealed an increased incidence of astrocytomas of
the brain and spinal cord, carcinomas and adenomas of the Zymbal gland or ear
canal, and squamous cell carcinomas and papillomas of the forestomach in the
high-dose males and females (Table 13-23). Increased incidences of the
aforementioned tumor types were observed predominantly in animals dying,
sacrificed in a moribund condition, or at sacrifice intervals after the 12th
month of the study, although the increased incidences of astrocytomas of the
brain and carcinomas of the Zymbal gland or ear canal were noted earlier in the
high-dose female after the sixth month of the study.
In conclusion, the carcinogenic effect of acrylonitrile administered to rats
in drinking water further reconfirmed the earlier findings of Dow and Litton-
Bionetics study discussed latter in this section.
13-95
-------
TABLE 13-23.
TUMOR INCIDENCES IN SPRAGUE-DAWLEY RATS FED
ACRYLONITRILE IN DRINKING WATER*
(Biodynamics, Inc. 1980a)
Dose
Level
(ppm)
0
1
100
Sex
M
F
M
F
M
F
Brain
Astrocytoma
2/98 (2%)
0/99 (0%)
3/95 (3%)
1/100(1%)
23/97(24%)t
32/97(33%)t
Spinal Cord
Astrocytoma
— §
0/96(0%)
— §
0/99(0%)
— §
7/98(7%)t
Zymbal Gland/
Ear Canal
Carcinomas
1/100(1%)
0/99 (0%)
0/91 (0%)
0/95 (0%)
14/93(15%)t
7/98(7%)t
Stomach
Papil loma/
Carcinoma
3/98 (3%)
1/100(1%)
3/98 (3%)
4/99 (4%)
12/97(12%)t
7/99(7%)t
*Animals intentionally sacrificed at 6, 12, and 18 months after
acrylonitrile administration are included in the denominator.
tStatistically significant at P < 0.05.
§Tissues not analyzed.
13-96
-------
Biodynamics, Inc. Study in Fischer 344 Rats (19806)--
A report entitled "A Twenty-four month Oral Toxicity/Carcinogenicity Study
of Acrylonitrile Administered to Fischer 344 Rats in Drinking Water," dated
December 12, 1980, was submitted to the U.S. Environmental Protection Agency by
the Monsanto Company, St. Louis, Missouri. This study, conducted by Biodynamics
Inc. for the Monsanto Company, was designed to evaluate the toxicity/
carcinogenicity of acrylonitrile. Acrylonitrile (100% pure supplied by the
Monsanto Company) was administered in the drinking water to 100 Fischer 344 rats
of each sex at dose levels of 1, 3, 10, 30, and 100 ppm, and the control group
contained 200 animals/sex. Interim necropsies were performed at 6, 12, and 18
months (20/sex from the control group and 10/sex from each treatment group each
time interval). This study was originally designed to be 24 months in duration;
however, to ensure at least 10 animals/sex/group for histopathological
evaluation at termination, all females were sacrificed at 23 months due to low
survival. The males were continued on test until the 26th month when similar
survival levels were reached. Mortality in the males and females receiving 100
ppm was markedly greater than controls, while mortality in the 10 ppm males and
females receiving 3 and 30 ppm was also somewhat greater than controls. Food
consumption was comparable for all groups on a g/kg/day basis. Liquid
consumption for the females receiving 100 ppm was slightly lower than controls
on a ml/kg/day basis, while values for the males in this group were comparable
to or greater than controls.
Slight, but generally consistent reductions in hemogoblin, hematocrit, and
erythrocyte counts were noted for the females receiving 100 ppm throughout the
study.
Histopathological evaluation presented in Table 13-24 revealed an increased
incidence of malignant tumor-bearing animals in the groups receiving 10, 30, and
13-97
-------
TABLE 13-24. TUMOR INCIDENCE IN FISCHER 344 RATS FED ACRrLONITRlLE
IN DRINKING WATER*
(Biodynamics, Inc. 19805)
Dose Level
Groups (ppm) Sex
Brain Spinal Cord
Astrocytomat Astrocytomat
Ear Canal
(Zymbal Gland) Stomach-Squamous
Papil Ionia/Adenoma/ Cell Papilloma/
Carcinoma Carcinoma of
(Squamous Cell )t Forestomacht
IA & IB Control H
F
II
III
IV
V
VI
1 M
F
3 M
F
10 M
F
30 M
F
100 M
F
2/200 (1.0)
1/199 (0.5)
2/100 (2.0)
1/100 (1.0)
1/100 (1.0)
2/101 (2.0)
2/100 (2.0)
4/95 (4.2)§
10/99(10. 1)§
6/100 (6.0)§
21/99(21. 2)§
23/98(23. 4)§
1/196
1/197
0/99
0/97
0/92
0/99
0/98
l/92§
0/99
0/96
4/93
1/91
(0.5)
(0.5)
(0.0)
(0.0)
(0.0)
(0.0)
(0.0)
(1.0)
(0.0)
(0.0)
(4.3)§1I
(1.1)
2/189(1.0)
0/193(0.0)
1/97 (1.0)
0/94 (0.0)
0/93 (0.0)
2/92 (2.2)
2/88 (2.3)
4/90 (4.4)§
7/94 (7.4)§
5/94 (5.3)§
16/93(17. 2)§#
10/86(11.6)§
0/199(0.0)
1/199(0.5)
1/100(1.0)
1/100(1.0)
4/97 (4.1)§
2/100(2.0)
4/100(4. 0)§
2/97 (2.1)
4/100(4. 0)§
4/100(4. 0)§
1/100(1.0)
2/97 (2.1)
*AnimaTs sacrificed intentionally ate, 12, ana 18 months after acrylomtrile
administration are included in the denominator.
tNumbers in the parentheses are percentages.
§Statist1cally significant at P < 0.05.
HThese rats had astrocytoma in both brain and spinal cord.
fOne rat had unilateral papilloma/carcinoma.
13-98
-------
100 ppm. The observed tumors were: astrocytomas of the central nervous system
(brain and/or spinal cord) and squamous cell carcinomas of the ear canal, as
well as mammary gland carcinomas in the females receiving 100 ppm.
In summary, the ingestion of acrylonitrile via drinking water at doses of
10, 30, and 100 resulted in an increased incidence of certain tumors. The
target organ specificity (central nervous system, ear canals) confirms the
similar earlier findings of the drinking water study in rats by the Dow Chemical
Company.
Litton-Bionetics Study (Beliles et al. 1980)--
Beliles et al. (1980) of Litton-Bionetics, Inc., sponsored by the Chemical
Manufacturers Association, performed a three-generation reproductive study in
Charles River rats [CRL:COBS CD (SD) BR]. These rats and their offspring
ingested water containing 100 or 500 ppm acrylonitrile starting approximately 15
days post weaning and were mated after 100 days. Female rats were maintained on
water containing acrylonitrile for 20 weeks; following delivery of the second
litter, the animals were exposed to acrylonitrile for approximately 45 weeks.
Following exposure, the animals in the three generations FQ, F^b, and p£b
were sacrificed and observable masses were evaluated histologically. Results of
histologic evaluations are presented in Table 13-25. The tumor incidence was.low;
only rats of the second generation at the high-dose level showed a significant
increase in the number of tumors. The low tumor incidence was probably due to
the relatively short exposure and observation period (approximately 45 weeks).
This study was suggestive of tumorigenic action of acrylonitrile since the type
of tumors observed were the same as the statistically highly significant tumor
incidence of Quast et al. (1980). This study provides further confirmation of
the incidence of astrocytoma and Zymbal gland tumors observed by Quast et al.
(1980).
13-99
-------
TABLE 13-25. INCIDENCE OF TUMORS OBSERVED IN RATS DURING
A THREE-GENERATION REPRODUCTIVE STUDY
(Bellies et al. 1980)
Generation 0
F0 0/19 (0%)
Fjb 0/20 (0%)
F2b 0/20 (0%)
Total 0/59 (0%)
Generation 0
F0 0/19 (0%)
Fjb 0/20 (0%)
F2b 0/20 (0%)
Total 0/59 (0%)
Astrocytoma Incidence
Dose (ppm acrylonitrile in water)
100 500
1/20 (0%)
1/19 (5.2%)
1/20 (5%)
3/59 (5%)
Zymbal Gland Tumor
Dose (ppm acrylonitri
100
0/20 (0%)
2/19 (10.5%)
0/20 (0%)
2/59 (3.4%)
2/25 (8%)
4/17 (23.5%)
P = 0.036*
1/20 (5%)
7/62 (11.2%)
P = 7.8 x ID'3*
Incidence
le in water)
500
1/25 (4%)
4/17 (23.5%)
P = 0.036*
3/20 (15%)
8/62 (12.9%) o
P = 3.7 x ID'3*
*P-values calculated using the Fisher Exact Probabiity Test
13-100
-------
Gavage Studies
Maltonl et al. 1977--
Maltoni et al. (1977) performed a cancer bioassay of acrylonitrile in which
40 Sprague-Dawley rats of each sex in both the treated and control groups were
exposed to a single dose of 5 mg/kg acrylonitrile by gavage dissolved in olive
oi1, 3 times a week, for 52 weeks. On spontaneous death, a moderate increase in
tumors of the mammary gland region and forestomach of female rats was described.
Although this study was very limited, with only a single exposure level and a
relatively short observation period (52 weeks), the results present further
evidence for the carcinogenicity of acrylonitrile.
Biodynamics, Inc. Gavage Study in Sprague-Dawley Rats (1980c)--
A report entitled "A Twenty-four Month Oral Toxicity/Carcinogenicity Study
of Acrylonitrile Administered by Intubation to Sprague-Dawley (Spartan) Rats"
was submitted to the U.S. Environmental Protection Agency, June 30, 1980, by the
Monsanto Company, St. Louis, Missouri. This study was conducted by Biodynamics,
Inc. for the Monsanto Company. In this study, acrylonitrile (100% pure*) was
administered by intubation to Sprague-Dawley (Spartan) rats (100/sex/group) at
three dose levels of 0, 0.10, and 10.0 mg/kg/day, 5 days/week. Interim
necropsies were performed at 6, 12, and 18 months (10/sex/group). This study
was originally designed to terminate at 24 months; however, because only 10 and
13 high-dose males and females, respectively, were alive by 20 months, all
surviving animals in all groups were terminated during the 20th month to ensure
at least 10 animals/sex for histopathological evaluation. The body weights of
high-dose group males were consistently slightly lower than control.
*Supplied by Monsanto Company, Texas City, Texas.
13-101
-------
Histopathological evaluations presented in Table 13-26 show that there were
statistically significant increased incidences in tumors of the brain and ear
canal (Zymbal gland) in both high-dose males and females. Stomach and
intestinal tumors were observed only in high-dose males and mammary gland tumors
were observed in high-dose females. Statistically significant tumor incidences
were not observed in low-dose groups either in males or females.
In summary, acrylonitrile administered orally via intubation produces a
carcinogenic effect in Sprague-Dawley rats at the following tumor sites: brain,
Zymbal gland, mammary gland, stomach, and intestine.
Inhalation Studies
Maltoni et al. (1977)--
Maltoni et al. (1977) exposed Sprague-Dawley rats to atmospheres containing
5, 10, 20, and 40 ppm acrylonitrile 4 hours/day, 5 days/week, for 12 months.
The rats were maintained for their entire lifetime. Histological examination of
the selected tissues were made. The incidence of tumors observed in this study is
shown in Table 13-27. Slight increases-were observed in mammary gland tumors of
males and females, the forestomach of males, and the skin of females. Maltoni
et al. (1977) claimed that these results indicated a "border-line carcinogenic
effect." The detection power (sensitivity) of this assay was low because of the
low concentration of acrylonitrile used and the relatively short exposure period
(12 months).
Dow Chemical Company (Quast et al. 1980b) —
A second inhalation study of acrylonitrile was conducted by Dow Chemical
Company, sponsored by the Chemical Manufacturers Association. In this study,
100 male and female Spraque Dawley rats (Spartan substrain) were exposed to 0,
13-102
-------
TABLE 13-26. TUMOR INCIDENCE IN RATS FED ACRYLONITRILE ORALLY BY INTUBATION
(Biodynamlcs, Inc. 1980c)
1— •
CO
1— '
CD
Dose Level
mg/kg/day Sex
0 M
F
0.10 H
F
10.0 M
F
Brain
Astro-
cytoma
2/100 (2%)
1/99 (1%)
0/97 (0%)
2/100 (2%)
16/98 (16%)*
17/100(17%)*
Spinal Cord
Astrocytotna
0/94 (0%)
0/100(0%)
0/93 (0%)
0/95 (0%)
1/97 (1%)
1/99 (1%)
Ear Canal
(Zymbal Gland)
Squamous Cel 1
Carcinomas
1/96 (1%)
0/85 (0%)
0/93 (0%)
0/94 (0%)
10/96(10%)*
9/94 (10%)*
Stomach
Carcinoma/
Papilloma
2/99 (2%)
2/99 (2%)
6/97 (6%)
4/99 (4%)
40/99(40%)*
17/99(17%)
Intestine
0/100 (0%)
— t
1/100 (1%)
— t
6/100 (6%)*
---t
Mammary Gland
---t
7/101 (7%)
— t
6/100 (6%)
---t
22/101(22%)*
*Statistically significant at P < 0.05.
tTlssues not analyzed.
-------
TABLE 13-27. TUMOR INCIDENCE IN RATS FOLLOWING INHALATION OF ACRYLONITRILE
(Maltoni et al. 1977)
Tumor Type (sex) Number of Animals with Tumors (%)
Exposure Concentration of Acrylonitrile (ppm)
0 5 10 20 40
Number of animals of each
sex
Mammary tumors (female)
Mammary tumors (male)
Zymbal gland (female)
Zymbal gland (male)
Encephalic tumors (female)
Encephalic tumors (male)
Forestomach papillomas
(female)
Forestomach papillomas
(male)
Skin carcinomas (female)
Skin carcinomas (male)
Uterine carcinomas
(female)
Total tumors (female)
Total tumors (male)
30
5(16)
1(3)
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
0(0)
1(3)
9(30)
4(13)
30
10(33)
0(0)
0(0)
0(0)
0(0)
0(0)
1(3)
1(3)
4(13)
P = 0.056*
0(0)
3(10)
17(57)
1(3)
30
7(23)
1(3)
1(3)
1(3)
0(0)
0(0)
2(7)
2(7)
1(3)
0(0)
1(3)
11(36)
10(33)
30
10(33)
4(13)
P = 0.17*
1(3)
0(0)
0(0)
1(3)
1(3)
0(0)
1(3)
0(0)
2(7)
14(47)
13(43)
30
7(23)
4(13)
P = 0.17*
0(0)
0(0)
0(0)
2(7)
0(0)
3(10)
1(3)
1(3)
1(3)
8(27)
12(40)
*P values calculated using the one-tailed Fisher Exact Test.
13-104
-------
20, or 80 ppm of acrylonitrile 6 hours per day, 5 days per week, for 2 years,
except during weekends or holidays. The acrylonitrile used for this study was
produced by E.I. du Pont de Nemours and Company, Inc. Its stability and
impurities are summarized in Table 13-28. The nominal concentration of
acrylonitrile in the exposure chamber was calculated from the rate at which the
liquid test material was dispensed and the rate of airflow through the chamber.
The target concentrations and chamber concentrations are shown in Table 13-29.
During the course of the 2-year study, hematology, urinalysis, and clinical
chemistry determinations were performed at periodic intervals. The results of
these determinations indicated that acrylonitrile did not have an adverse effect
on bone marrow, kidney, or liver functions in either male or female rats.
During the first 6 months of the study, the exposed rats drank more water and
appeared to excrete a lower specific gravity urine than control rats.
In lifetime observations of male and female rats exposed to acrylonitrile
vapors, toxic effects, characterized by decrease in body weight and early
mortality, were observed. The cumulative mortality data are shown in Tables 13-30
and 13-31.
Microscopic examination of tissues revealed a treatment-related
statistically significant incidence of tumors in the central nervous system, ear
canal gland (Zymbal gland), tongue, small intestine, and mammary gland. In male
and female rats, statistically significant increased incidences were observed
only at the 80 ppm dose levels with the exception of glial cell tumors of the
central nervous system which were also increased in female rats at 20 ppm. Detailed
results are presented in Table 13-32. however, these results are based on
data presented in the Quast et al. (1980b) inhalation study and do not reflect
separation of individual animals having one or more tumor types. The CAG has
requested these individual data from Dr. Quast and the Chemical Manufacturers
Association.
13-105
-------
TABLE 13-28. STABILITY AND TRACE IMPURITY ANALYSIS OF THE ACRYLONITRILE LIQUID TEST MATERIAL
(Quasi et al. 1980b)
Refractive Index (0), 25.00°C
P-Methoxy phenol (MEHQ) Inhibitor, +1 ppm
Methacrylonltrile by GLC, +2 ppm
Other Compounds by GLC
Acetone by GLC, ppm
Volatiles by Mass Spec.
Peroxides as ppm \\2®2 (Polargraphic Data)
Metals by Atomic Emission with their Lower
Detection Limit in Parentheses, ppmll
Sample #1
Dated
1/27/76
1.3908
34
74.5
None
None
Water Onlyt
5
Sample 12
Dated
5/24/76
1 . 3908
32
75
None
None
Water Onlyt
1
Sample 13
Dated
11/19/76
1 . 3908
35.5
74.8
None
None
Hater Onlyt
4.5
Sample |4
Dated
3/21/77
1.3901
34.5
47
None
88+5
Water Onlyt
3.8
Sample, 15
Dated
9/7/77
1.3904
38.8
33-
None
195+5
Water Onlyt
0.2§
Sample 16
Dated
1/13/78
1 . 3909
41.7
56.5+1.5
•
131+7
AN Unly
N.O.
Metal
Fe
Ni
Cu
Al
Ca
Si
Mg
Mn
Pb
Cr
Sn
Zn
Ti
Data 1s given In ppm on a
0.5a
N.D.(0.2)
O.la
N.D.(0.2)
0.7
b
N.D.(0.2)
N.D. (0.1)
N.D. (0.5)
N.D. (0.2)
N.O. (0.2)
N.D. (0.5)
N.D. (0.2)
volume/volume basis for the liquid
0.8a 0.8(0.2)
N.D.(0.2) N.D. (0.2)
O.la 0.1(0.1)
N.D.(0.2) 0.4(0.2)
1.0a 0.8(0.2)
b c(0.2)
N.D.(0.2) N.D.(0.2)
N.D (0.1
N.O. (0.5
N.D. (0.2
N.D. (0.2
N.D. (0.5
N.D. (0.2
0.1(0.1)
N.O. (0.5)
N.D. (0.1)
N.O. (0.2)
c(0.5)
N.D. (0.2)
test material.
la
N.D.(l)
N.D. (0.2)
b
b
b
N.D. (0.1)
N.D. (0.5)
N.O. (3)
N.D. (0.5)
b
b
N.D.(l)
N.D.(l)
N.D.(l)
N.D. (0.2)
N.D.(l)
N.D.(l)
b
N.D. (0.1)
N.D. (0.5)
N.D. (3)
N.D.(l)
N.D.(l)
N.D. (3)
N.D.(l)
KD
N.D.(l)
N.D. (0.2)
N.D.(l)
Kl)
b
N.D. (0.1)
N.D. (0.5)
N.O. (3)
N.D.(l)
N.D.(l)
N.D. (3)
N.D.(l)
•Small unidentified peak ahead of acetone estimated to be about 25 ppm.
tSens1t1v1ty 1s limited to about 100 ppm.
§Detection limit 1s 0.2.
HThe analysis of the sample analyzed 1s accurate to +_ 10-15% of amount present.
*Lower detection limit was not specified.
bNo analysis was made for this metal.
cMetal was detected, but unmeasurable.
N.D. - Not Detectable.
-------
TABLE 13-29. CHAMBER CONCENTRATIONS OF ACRYLONITRILE VAPORS
(Quast et al. 19806)
Target Concentration (ppm)
Analytical Concentration* (ppm)
X +_ S.D.
Range of Daily Analytical Concentration
(ppm)
Nominal Concentrationt (ppm)
X +_ S.D.
Range of Daily Nominal Concentration
Average Analytical Concentration/
Average Nominal Concentration
Number of Exposure Days Within +10%
of Target Concentration
Number of Exposure Days Within +_25%
of Target Concentration
Total Exposure Days
Total Number of Exposure Analyses
Exposure
20 ppm
20.0+1.9
7.3-35.0
20.7+2.6
11.00-37.8
0.97
433
64
507§
1484
Group
80 ppm
80.0+5.9
45.2-106.0
90.1+7.3
59.5-138.2
0.89
470
33
508
1494
*Data represents mean (X) +_ standard deviation (S.D) of daily 6-hour
time-weighted average for the total number of exposure days.
tData represents mean (X) _+ standard deviation (S.D.) of daily 6-hour
nominal concentration for the total number of exposure days.
§The 20 ppm exposure group had one less exposure day than the 80 ppm
exposure group as a result of mechanical failure in the ventilation system of
the 20 ppm exposure chamber.
13-107
-------
TABLE 13-30. CUMULATIVE MORTALITY DATA OF MALE RATS
EXPOSED BY INHALATION FOR 2 YEARS TO ACRYLONITRILE VAPORS*
(Quast et al. 1980b)
Exposure Concentration
Days on Test 0 ppm 20 ppm 80 ppm
0-30
31-60
61-90
91-120
121-150
151-180
181=210
211-240
241-270
271-300
301-330
331-360
361-390
391-420
421-450
451-480
481-510
511-540
541-570
571-600
601-630
631-660
661-690
691-720
721-735
0
0
0
0
1
2
2
2
2
2
3
3
4
6
11
14
19
23
27
35
43
62
71
78
82
1 0
1 0
1 1
2 1
2 1
3 2
3 6
4 12t
5 13t
5 14t
6 16t
6 18t
8 19t
9 22t
12 24t
15 28t
26 39t
34 47t
38 56°
47 63t
59t 76t
68 83t
72 85t
81 94t
86 96t
Total Number of Rats 100 100 100
*Data listed as number dead which is equal to percent "cfead^
tSignificantly different from control by Fisher's Exact Probability Test,
P < 0.05.
13-108
-------
TABLE 13-31. CUMULATIVE MORTALITY DATA OF FEMALE RATS
EXPOSED BY INHALATION FOR 2 YEARS TO ACRYLONITRILE VAPORS*
(Quast et al. 1980b)
Days on Test
Exposure Concentration
0 ppm 20 ppm 80 ppm
0-30
31-60
61-90
91-120
121-150
151-180
181-210
211-240
241-270
271-300
301-330
331-360
361-390
391-420
421-450
451-480
481-510
511-540
541-570
571-600
601-630
631-660
661-690
691-720
721-735
1
1
1
1
1
1
1
1
3
5
5
7
9
11
14
14
19
26
34
36
50
63
66
71
78
0
0
0
0
0
0
1
1
1
1
2
2
3
5
10
14
22
31
36
43
54
70
81t
88t
91t
0
0
0
0
0
0
1
2
4
6
9
11
19t
27t
33t
41t
57t
71t
80t
88t
94t
98t
98t
99t
99t
Total Number of Rats
100
100
100
*Data listed as number dead which is equal to percent dead.
tSignificantly different from control by Fisher's Exact Probability Test,
P < 0.05.
13-109
-------
TABLE 13-32. TUMOR INCIDENCE IN SPRAGUE-DAWLEY RATS
EXPOSED TO ACRYLONITRILE BY INHALATION
(Quast et al. 1980b)
Diagnoses
0 ppm
20 ppm
80 ppm
Males
Zymbal gland tumors
of external ear canal
(benign and malignant)
Small intestine tumors
(benign and malignant)
Brain and/or spinal cord
glial cell tumors
(benign and malignant)
Tongue-squamous cell tumors
(benign and malignant)
Females
Zymbal gland tumors
of external ear canal
(benign and malignant)
Mammary gland
fi broadenoma/adenofi broma
Mammary gland
adenocarcinoma
Mammary gland tumors
(benign and malignant)
Brain and/or spinal cord
glial cell tumors
(benign and malignant)
2/100
2/99
4/100
2/20
(P
(P
11/100
= 0.009)*
15/98
= 7.03 x.lO'4)*
0/100
1/96
0/100
79/100
9/100
88/100
0/100
4/99 22/99
(P = 5.71 x lO'8)*
0/14 7/89
(P = 0.0251)*
1/100 11/100
(P = 3.65 xlO'4)*
96/100 75/100
(P = 2.06 x ID'4)*
8/100
96/100
8/100
(P = 0.003)
20/100
(P = 0.022)*
85/100
21/100
(P = 3.32 x ID'7)*
*P values calculated by Fistier" Exact P>obabiTfty~Test.
13-110
-------
In summary, in the Dow study acrylonitrile induced a statistically
Significant increased incidence of tumors in male and female rats following
exposure by inhalation.
13-111
-------
CELL TRANSFORMATION STUDY
Parent and Casto (1979) reported the effect of acrylonitrile on Syrian
golden hamster embryo cells (HEC), and found that acrylonitrile (ACN)
transforms cells in culture and enhances the transformation of cells
previously affected with simian adenovirus SA7, a colony transforming
oncogenic virus.
ACN from Aldrich Chemical Co. (Milwaukee, Wis.) was used in this
experiment. The purity was stated to be greater than or equal to 99%. The
impurities included about 0.3% water, less than 0.5% acetonitrile, and 30 to
45 ppm l-hydroxy-4-methoxybenzene. In this experiment cultures of primary
Syrian golden HEC were prepared by trypsinization of decapitated and
eviscerated embryos after 13-14 days of gestation. Cells were plated into
60-mm-diameter Lux plastic dishes at a density of 5 x 106 cells/dish with
modified Dulbecco's medium and 10% fetal bovine serum (Reheis Chemical Co.,
Kankakee, 111.) and incubated at 37°C for 3 days in 5% C02. ACN was
dissolved in 100 mg acetone/ml and diluted in complete medium to give the
final concentrations.
In this viral transformation enhancement assay, HEC were exposed to ACN in
concentrations of 0, 25, 50, 100, and 200 ug/ml. Treatment of HEC with ACN
for 18 hours before SA7 inoculation resulted in only slight but significant
enhancement to 1.8-fold (Table 13-33). When cells were treated with ACN 5 hours
after they were inoculated with virus (Table 13-33), a significant enhancement of
8.9- and 8.4-fold was observed at 200 and 100 ug ACN/ml, respectively.
Treatment with 200 ug ACN/ml reduced the cloning efficiency to less than 10%,
but the number of SA7 foci only decreased from 41 in control to 26 in treated
cells. The enhancement found when cells were chemically treated after virus
inoculation was observed with several other chemicals.
13-112
-------
TABLE 13-33.
ENHANCEMENT OF SA7 TRANSFORMATION BY TREATMENT OF HEC WITH ACN*
(Parent and Casto 1979)
Time of ACN
treatment
ACN
ug/ml
Surviving SA7 Enhancement
fractiont foci§ ration
18 hr before SA7
18 hr before SA7
5 hr after SA7
200
100
50
25
0
200
100
50
25
0
200
100
50
25
0
0.08
0.34
0.62
0.75
1.00
0.18
0.46
1.07
1.02
1.00
0.07
0.21
0.60
0.69
1.00
1
32
41
45
50
6
19
31
37
20
26
75
37
41
41
0.3
1.8
"O
1.2
1.0
1.6
2.1
1.5
1.8
TTfJ
8.9
F^f
TT5
1.4
1.0
*Chemical dilutions were added to mass cultures of HEC 18 hr before or 5 hr
after treatment with SA7. Virus was absorbed 3 hr, and the cells were
transferred for survival (500-700 cells/dish) and for transformation assays
(200,000-300,000 cells/dish).
tDetermined from plates receiving 500-700 cells. Number of colonies from
virus-treated and chemically-treated cells was divided by the number of
colonies from virus-inoculated control cells to give the surviving fraction.
Cloning efficiency of control cells was 10-15%.
§Number of foci from 106 plated cells.
^Enhancement ratio was determined by dividing the TF of untreated cells (TF
= SA7 foci x reciprocal of the surviving fraction) by that obtained from
control cells. Underlined values are statistically significant at the 5%
level.
13-113
-------
When HEC were treated for 6 days with ACN (chemical transformation)
without added virus, foci of morphologically transformed cells were observed
that were similar to those described previously with known chemical
carcinogens. At 100 ug ACN/ml, three foci were observed on nine dishes and
two foci on six dishes at 50 ug ACN/ml; BP treatment resulted in three foci on
four dishes at 1.25 ug/ml and two foci on ten dishes at 0.62 ug/ml (Table 13-34)
No foci were observed on medium or solvent control dishes.
The observation that acrylonitrile transforms cells adds support to the
animal and human evidence that acrylonitrile may be carcinogenic.
13-114
-------
TABLE 13-34. TRANSFORMATION OF HEC BY ACN
(Parent and Casto 1979)
Treatment*
ACN
BP
Control
ACN
ug/ml
100
50
25
12
1.25
0.62
—
Surviving
fractiont
0.06
0.76
0.84
1.06
0.78
0.94
1.00
Foci /dishes
3/9
2/6
0/6
0/5
3/4
2/10
0/7
*Chemicals were added to tertiary HETT plated 24 hr earlier with 50,000
(transformation) or 1,000 (survival) cells/dish. Fresh medium with chemical
was added after 3 days and removed after 6 days. Colonies were fixed and
stained at 9 days for survival assays; focus assays for transformation were
done 25 days after treatment was indicated.
tDetermined from dishes receiving 1,000 cells. Number of colonies from
treated plates was divided by the number from control dishes.
13-115
-------
13.5.2 EPIDEMJOLOGIC STUDIES
O'Berg (1980)
There are no community epidemiologic studies available that attempt to
demonstrate an association between exposure to ambient levels of acrylonitrile
and the development of disease. However, an occupational epidemiologic study
that involved workers exposed to acrylonitrile at a Du Pont textile fibers
plant in Camden, South Carolina was conducted by Maureen T. O'Berg of E.I. du
Pont de Nemours and Co., Inc. (O'Berg 1980). The study cohort was 1,345 male
employees "identified as having had potential exposure to acrylonitrile at
some time between start-up in 1950 and 1966." The 1966 cutoff date allowed
for a minimum 10-year follow-up through the end of 1976. Expected numbers
were stated to be based both on company and national rates. The analyses
presented in the paper, however, derive expected number of cases based only on
company rates, ignoring the possible effects of other chemicals on this
"control" cohort.* The analyses consider calendar time, payroll
classification, occupation, duration of exposure, latency, and severity of
exposure. The severity of exposure levels were designated as high, medium,
and low. Du Pont representatives agreed that 20 ppm, 10 ppm, and 5 ppm might
be used to represent the designated classification of high, medium, and low
exposure levels. This was documented in a trip report by Jane Brown,
Industrial Hygienist, National Institute for Occupational Safety and Health
(NIOSH 1978).
Overall, 25 cases of cancer occurred, with 20.5 expected based on company
*While this paper also attempts to compare observed and expected deaths,
there is an inconsistent result. According to O'Berg, 4.4 respiratory cancer
cases were expected in this cohort using the Du Pont controls. However, using
the same controls, O'Berg predicts 6.1 respiratory cancer deaths. A telephone
conversation with O'Berg confirmed that this inconsistency was due to
different methods of following the cohort for cases and deaths. Cases were
described only for active employees, while deaths included retirees.
13-116
-------
rates. Of these 25 cases, eight were respiratory cancer cases versus 4.4
expected. Excesses were found primarily among wage roll employees who had
worked during plant start up, 1950 to 1952, and had been exposed for at least
6 months. For these employees there were 8 cases of respiratory cancer versus
2.6 expected (P < 0.01). Furthermore, most of this above excess occurred
during the latest follow-up period, 1970 to 1976, when there were six cases of
respiratory cancer versus 1.5 expected (P < 0.01). Total cancer cases in this
latest follow-up period for this group were also significant, 17 versus 5.6
(P < 0.01).
A trend toward increased risks was seen not only with increased follow-up
time but also with severity of exposure. In wage roll workers with at least
moderate exposure and probable latent period of at least 15 years, the
observed and expected numbers of cancer cases was 13 and 5.5, respectively.
Furthermore, half of this excess cancer was due to respiratory cancer, 5
versus 1.4 (P < 0.05). Thus, this study provides some evidence that acryloni-
trile is carcinogenic to humans.
However, because of the known relationship between smoking and lung
cancer, further analysis is attempted concerning the role of smoking behavior
in these findings. (Seven of the eight lung cancer cases were reported to be
smokers by their supervisors or associates; the eighth was unknown.) Dr.
Bruce Karrh of Du Pont stated that there were pathology slides for five of the
eight respiratory cancer cases. He further stated that these were identified
as four squamous cell carcinomas and one oat cell carcinoma. Of the three
remaining respiratory cancers cases for whom slides were unavailable, he
stated that two were bronchogenic and the other unknown. These cell types are
generally believed to be associated with both chemicals and smoking by most
pathologists.
13-117
-------
In an attempt to investigate the impact of smoking on the risk of
developing lung cancer in this population, the Du Pont Company provided
additional data (letter from Sidney Pell, Du Pont, July 23, 1980) to the
Carcinogen Assessment Group (CAG) regarding the smoking habits of 32 of the 36
cancer cases reported on in this plant (some of these were not in the study
cohort), as well as data on the smoking habits of a matched group of non-cases
in the plant. Of the 32 cancer cases, 22 were cancers other than lung, and 16
or 73% of the nonlung cancer cases were smokers. The smoking habits of the
matched group of 36 noncancer controls from the same plant were also provided.
They were matched on a three to one basis to certain selected cancers*
occurring in the study group on the basis of age, payroll classification, date
of first exposure, and date of termination. It was found that 25 or 69% of
the 36 cases were smokers in this group.
Based on this information, we can estimate that 70% of the plant
population were smokers, and 30% nonsmokers. Of the 70% who were smokers, we
will assume that 50% were "moderate" smokers while 20% were "heavy" smokers
based on figures by Axelson (1978). We will assume also that the relative
risk of lung cancer is 1, 10, and 20, respectively, for the nonsmokers,
moderate smokers, and heavy smokers (Doll and Hill 1952, Hammond 1975). Using
the method of Axelson (1978) to adjust for smoking differences in cohorts, the
*TFe author selected only cancer cases that have a high correlation with
cigarette smoking for his case control study. The purpose was to test whether
cigarette smoking was different in the matched non-cases selected from the
same plant compared to the cases. These cases were eight lung cancers, one
esophageal, one nasopharygeal, and two bladder cancers. The resulting test
statistic (X^) was nonsignificant, indicating that there was no reason to
suspect that there was an uneven distribution of cigarette smoking in the
cases compared to the matched non-cases.
13-118
-------
relative incidence of lung cancer in the plant population is
(1) Ip = 1(0.30)I0 + 10(0.50)I0 + 20(0.20)I0 = 9.3 I0
where I0 equals the risk of lung cancer in a nonsmoking population.
Based on a nationwide survey (U.S. Health, Education, and Welfare 1973)
which found that 62% of the male blue collar workers smoked in 1966, we assume
that the overall company population had 40% nonsmokers. In addition, we
assume that the distribution of moderate and heavy smokers is the same as
above, so that moderate and heavy smokers constitute 43% and 17% of the
population respectively. The computed incidence of lung cancer in this
comparison population is then
(2) Ig = 1(0.40)I0 + 10(0.43)I0 + 20(0.17)I0 = 8.1 IQ
where Iq is the incidence in the company population. Hence, the relative
contribution of smoking to the risk of lung cancer in the study cohort is the
ratio of the two incidence rates:
(3) Ip/lg = 9.3 I0/8.1 I0 = 1.15
Therefore, because of the slightly higher proportion of smokers in the study
cohort relative to the reference company population, the number of respiratory
cancer cases would be about 15% higher than the 1.4 cases expected without
considering smoking differences, or 1.4 x 1.15 = 1.61 cases. Assuming a
Poisson distribution of cases, the probability of seeing 5 cases or more when
only 1.61 are expected is only 0.024. Therefore, after the adjustment for
13-119
-------
smoking differences, the respiratory cancer rate in the study cohort exposed
to acrylonitrile for over 6 months is significantly higher than that of the
reference population.
In an attempt to assess whether the excess respiratory cancer rates could
be ascribed to a higher ratio of heavy to moderate smokers in the
acrylonitrile-exposed cohort than that in the reference population, the
following calculation was done. In equation (1) for Ip the ratio of heavy
to moderate smokers was increased from 0.2/0.5 = 1/2.5, keeping the total
fraction of smokers at 0.7, until the point was reached where the probability
of observing five cases or more giving the expected number, 1.4 x (Ip/Ig),
was increased from 0.024, as above, to 0.05. The result is that in order for
a higher ratio of heavy to moderate smokers in the cohort to account (at the
0.05 level of significance) for the observed excess respiratory cancer rate,
that ratio would have to be 1.6/1. Our judgment is that such a marked excess
of heavy versus moderate smokers would not likely occur in the acrylonitrile
workers, since for blue collar workers generally there are about 2.5 times
more moderate than heavy smokers.
In conclusion, the observations by O'Berg of a statistically significant
excess of respiratory cancer in workers exposed to acrylonitrile and followed
up for more than 10 years constitute significant evidence that acrylonitrile
is likely to be a human carcinogen, although smoking, at least as a
contributing factor, cannot be completely ruled out at this time.
Delzell and Monson (1982)
Delzell and Monson (1982) conducted a cohort mortality study of 327 white
male employees of a rubber manufacturing plant in Akron, Ohio who were
employed for at least 2 or more years between January 1, 1940 and July 7, 1971
13-120
-------
in two areas with potential exposure to acrylonitrile. The first area was a
nitMle rubber manufacturing operation. Butadiene and styrene were present in
this area between 1938 and 1949 while vinyl pyridine has been present since
1949. The latex that was produced in closed vessels was "blown down" into
tanks after excess monomers were removed. In the second area, latex was
coagulated and dried into solid rubber. From 1938 to 1960, the nitrile rubber
was washed on an open mill and dried in vacuum ovens. After 1960, this job
was done by extruder drying. Delzell and Monson did not report the levels of
acrylonitrile that these workers were exposed to while on the job.
Follow-up was continued until July 1, 1978, according to the authors, but
the extent of success is not indicated. The authors state that the follow-up
was "identical" to that of a recently published study by the same authors. The
authors do indicate that anyone without a death record prior to July 1, 1978
was assumed alive on that date. Expected deaths were generated based on U.S.
age and calendar time specific white male mortality. Overall, 74 deaths
occurred in this group compared to 89.5 expected, a deficit readily
attributable to the healthy worker effect. On the other hand, the risk of
death due to cancer was somewhat elevated with 22 observed versus 17.9
expected. Most of this excess was due to lung cancer deaths which were
nonsignificantly higher with 9 observed versus 5.9 expected. Only in workers
employed for 5 to 14 years and followed for at least 15 years, did a
significantly elevated risk of lung cancer appear (4 observed versus 0.8
expected P <0.01). However, if workers who were employed for more than 14
years and followed for at least 15 years are added to the latter group, not an
unreasonable consideration since presumably they are most exposed, the risk of
lung cancer was actually reduced but still significant (4 observed versus 1.4
expected, P <0.05).
13-121
-------
The excessive risk of lung cancer in this group may not be entirely due to
exposure to acrylonitrile, but may be partially reflective of exposures to
other substances within the same environment, i.e., the author reports that
one lung cancer victim had worked for 17 years in the curing division of the
company prior to employment in the acrylonitrile area, but the authors did not
elaborate further. Additionally, no discussion of the effects of smoking on
lung cancer is included.
However, the possibility that the excess risk of lung cancer is due to
acrylonitrile cannot be dismissed.
Thiess et al. (1980)
A cohort mortality study of 1,469 workers from 12 factories located in the
Federal Republic of Germany was conducted by Thiess et al. (1980). These 12
factories are part of the BASF company. The BASF company produces no
acrylonitrile but buys acrylonitrile in order to produce styrene-acrylonitrile
and acrylonitrile butadiene-styrene polymers as well as organic intermediate
products. The processing methods differ from factory to factory. No historic
exposure data for acrylonitrile exists according to the authors.
The population at risk was defined as all workers who were employed for
over 6 months in acrylonitrile processing presumably from time of first use of
acrylonitrile (around 1956) until the cut-off date of May 15, 1978. Included
were 1,081 German workers and 338 "foreigners" (nationality not given).
Follow-up was 98% complete on the German workers but only 56% complete on the
foreign segment. This left about 170 lost to follow-up or about 12% of the
work force.
Expected deaths were generated based upon mortality in three areas of
Germany, the city of Ludwigshafen, the state of Kheinhessen-Pfalz, and the
13-122
-------
Federal Republic of Germany (FRG). A total of 89 deaths were observed
compared to 92.3, 96.4, and 99.0 expected deaths, respectively, in the three
geographic entities cited. Twenty-seven deaths from malignant tumors were
observed compared to 20.5 expected based upon rates of the FRG. The most
striking finding was a statistically significant higher risk of cancer of the
lung (11 observed versus 5.65 expected based on FRG rates, P < 0.05; or 5.92
expected based on rates of Rheinhessen-Pfalz, P < 0.05). The authors
recalculated their results for the FRG only, with the 78 members of one
factory excluded from the group of 12 factories whom they said had "contact
with substances since proven to be carcinogenic." Not only did an excessive
significant risk remain with respect to lung cancer (9 observed versus 4.37
expected based on FRG rates, P < 0.05), but also a significant excess risk of
cancer of the lymphatic system was seen (4 observed versus 1.38 expected,
P < 0.05).
These results are questionable. First, the members of this cohort were
apparently exposed to a number of different carcinogens, i.e., vinyl chloride,
distillation residues containing polycyclic hydrocarbons, cadmium,
B-napthylamine, dimethylsulfate, and epichlorohydrin. Several of these have
been associated with a higher risk of lung cancer. Second, all lung cancer
victims were found to be smokers, while not all lung cancer victims were
proven to be exposed to acrylonitrile, according to the author.
On the other hand, the lung cancer risk estimated from this study may
actually be an underestimate of the true risk for the following reasons:
1. Lumping together workers from 12 different factories who may be
subjected to differing levels of exposures to acrylonitrile could have led to
an underestimate of risk by the inclusion of minimally exposed or unexposed
members.
13-123
-------
2. The healthy worker effect would contribute to an underestimate of
risk.
3. Insufficient follow-up on a relatively youthful cohort did not allow
enough latency in the segments of the cohort most likely to exhibit a true
estimate of risk. Only 447.1 person-years were accumulated in members over 64
years of age.
4. Underascertainment of vital status may have resulted in an undercount
of observed deaths. Twelve percent were lost to follow-up. Person-years have
been accumulated for each of the lost individuals until the last date known to
be alive.
Although, there are many limitations to this study, it is possible that
exposure to acrylonitrile may indeed be related to an excessive risk of lung
cancer and cancer of the lymph system as well as to other carcinogens present
in the workplace.
Werner and Carter (1981)
Werner and Carter (1981) conducted a cohort mortality study of 934 men who
worked on the polymerization of acrylonitrile and spinning of acrylic fiber at
six different factories sometime between 1950 and the end of 1968 for a
minimum of one year. Two of the six factories were located in Scotland and
Northern Ireland; the remainder were located in England and Wales. The cohort
was followed through the end of 1978, which resulted in a vital status
ascertainment near 100%. Expected deaths were calculated based upon standard
age-specific mortality rates for England and Wales combined.
Overall, only 68 deaths had occurred through 1978 where 72.4 were
expected. Twenty-one were due to malignant neoplasms versus 18.6 expected.
Deaths from cancer of the stomach were statistically significantly elevated
13-124
-------
over all age groups (5 observed versus 1.9 expected P < 0.05) with deaths to
persons in the 55-64 age group contributing the largest portion (3 observed
versus 0.7 expected, P < 0.05). On the other hand, a statistically significant
elevated risk of cancer of the lung, trachea, and bronchus appeared only in
the age group 15-44 (3 observed versus 0.7 expected, P < 0.05) but not in any
other age group.
In a nested (within the larger study) small case control study, the three
lung cancer cases were compared with controls (born the same year and living
beyond the date of death of the case) with respect to duration of employment.
In one case the duration of exposure was significantly greater than the
average for the matched controls. In the other two cases, the differences
were not significant. The result are equivocal at best.
There are many difficulties with this study, not the least of which is the
lack of data regarding actual measured levels of exposure to acrylonitrile
during the period 1950 thru 1968, according to the author. This lack of
quantitative data makes it difficult to distinguish between plants in terms of
their relative exposure levels. Implicit in the assumption of lumping the
employed populations of six different factories together is that they are all
relatively similar with respect to levels of exposure. This is usually not
the case, however, for some factories tend to be better controlled than
others.
Other problems with this study include the relatively short follow-up in
the small subgroup of the cohort where one would expect to have the greatest
risk, i.e., the 158 men who had the earliest exposure in 1950-58. This group
was followed for a minimum of 20 years accumulating a total of 3,241
person-years of which only 780 person-years were to individuals age 55 and
13-125
-------
over. Like several other studies, this is a relatively youthful cohort and
the number of expected deaths are too small to provide confidence in the
results. The findings of this study are only suggestive due to small numbers,
youthfulness of the cohort, insufficient follow-up, and lack of consideration
of effects of smoking. However, this study should be followed several
additional years to see if the significant excesses of lung and stomach cancer
found in this study remain high.
Monson (1978)
An epidemiologic study was conducted by Richard R. Monson for the B.F.
Goodrich Company and the United Rubber, Cork, Linoleum, and Plastic Workers of
America (Federal Register 1978). This retrospective cancer morbidity and
mortality study included some workers with exposure to acrylonitrile. Among
workers with potential exposure to acrylonitrile (extent of exposure not
known), there was a slight overall excess in the number of deaths from cancer
observed compared to the number expected (the data used for determining the
expected number were not specified): lung cancer deaths (7 observed versus
4.4 expected); genitourinary cancer deaths (2 observed versus 1.6 expected
deaths, and 6 observed versus 3.1 expected incidences); and Hodgkin's disease
deaths (2 observed versus 0.3 expected). Deaths from all causes (cancer and
noncancer) were not significantly elevated in the worker study group. Because
the workers in this study had potential for exposure to other carcinogens, the
Occupational Safety and Health Administration concluded that the study could
not be used to support an association between acrylonitrile exposure and human
cancer development (Federal Register 1978).
13-126
-------
Zack (1980. unpublished)
Judith A. Zack (unpublished report 1980) of the Monsanto Chemical Company
investigated the mortality experience of 76 white males at the Texas City
facility and 276 white males at the Decatur facility who were exposed to
acrylonitrile for a minimum of 6 months prior to January 1, 1968 and followed
until December 31, 1977. A total of 15 deaths, all causes, were found to have
occurred during that interval of time, 9 at the Texas City facility and 6 at
the Decatur facility. Of the 15 deaths, 3 were attributable to cancer, at all
sites. Only one was due to respiratory cancer. The author expected 18.11
deaths, all causes, based upon the Monson life table technique with 2.80
malignant neoplasms expected. The author concluded that although the number
of deaths is small, the observed numbers do not differ significantly from the
expected numbers.
This study is relatively insensitive in its ability to assess a cancer
risk in this cohort of acrylonitrile workers for two reasons. First, because
of the limited cohort size, the small numbers of expected deaths (18.11 due to
all causes, 2.8 due to malignant neoplasms, 0.84 due to respiratory
malignancy) do not provide enough information to place any confidence in the
statement of nonsignificance made by the author. Second, the usual latent
period for cancer of most sites is in excess of 15 years. Although the study
design guaranteed at least 10 years of lapsed time since onset of employment
to the cutoff date, still 52% or 183 of 352 members of the cohort were
observed for no longer than 15 years since initial employment, a time frame
which is probably insufficient for a carcinogenic potential to manifest
itself, if, in fact, one is present. This report cannot be considered
supportive of a "negative" risk assessment of acrylonitrile with respect to
its cancer-causing potential in humans.
13-127
-------
Gaffey and Strauss (1981, unpublished)
The Gaffey and Strauss study (1981, unpublished) is a cohort mortality
study of 326 white males who were potentially exposed to acrylonitrile in a
chemical plant in Decatur, Alabama. This plant was also included in the
earlier study by Zack (unpublished). The cohort was defined to be all white
males who had achieved 6 months of employment at the plant between April 1952
and December 1953. Follow-up for vital status ascertainment was continued to
the end of 1977 and was 95% complete according to the author. However, it was
noted that 73 who were found "alive" by retail credit check were subsequently
assumed to be "alive" by virtue of the fact that their names appeared in a
current telephone directory. Such an assumption may be unwarranted without
actual confirmation with the study member. Many widows keep their telephone
listing in their husbands' names for security reasons. Another 15 remained
with an unknown vital status.
Only 26 deaths were observed during this time frame resulting in an
unusually low standard mortality ratio (SMR) overall of 47, while for
malignant neoplasms the SMR was only 37. Only 4 of the 26 observed deaths
were attributable to cancer. And of these four only two were cancer of the
lung (expected = 3.74), while the remaining two were cancer of the kidney
(expected = 0.3). Workers in this study could not be classified by level of
exposure according to the author. However, the cohort was first dichotomized
into hourly versus salaried employees and then into maintenance versus
everyone else. No significant excessive risks were evident from the data
presented in any of the groups mentioned not even after a 20-year follow-up.
The author notes that the small number of malignancies observed make any
conclusions at best tentative.
13-128
-------
Furthermore, the definition of the cohort was not restrictive enough to
Identify only an "exposed" population for study. The author points out that
21 of the members of his present cohort were also members of the earlier
cohort study by Zack. Zack was careful to include only white male employees
who "worked in an acrylonitrile manufacturing or processing area" for a
minimum of 6 months prior to January 1, 1968). In fact, during the period
1952-53 Zack noted that only 17 employees of the Decatur plant could qualify
for inclusion into her cohort of 352 males, while Gaffey counted everyone who
was there at start-up (326 white males) including 95 salaried employees.
Hence, there is a distinct likelihood that Gaffey's cohort consists of large
numbers of unexposed employees. Thus, in addition to the healthy worker
effect and incomplete follow-up, it is likely that the relative insensitivity
of this analysis to detect a significant risk is due to 1) the relatively
youthful nature of the study group (average age at the end of follow-up was
only 54.9 years), 2) the inclusion of large numbers of unexposed or minimally
exposed employees in the cohort, and 3) the possibility that large numbers of
workers were classified erroneously as "live" via the telephone directory
check. These problems make this study questionable with respect to the
detection of a cancer risk.
Kiesselbach et al. (1980. unpublished)
This study appears to be a cohort mortality study of some 884 male workers
who were presumably exposed to acrylonitrile for a minimum of 1 year during
the period from 1950 to August 1, 1977 in 16 different plants (pilot,
laboratory, and production) of Bayer's Leverkusen Division in West Germany.
Workers selected for inclusion into the cohort were involved in the production
and processing of acrylonitrile. Follow-up was complete on 93.2% of the
13-129
-------
cohort. Only 60 persons were without a vital status ascertainment by the
close of the study. The male population of the German state of
Nordrhein-Westfalen was used as a reference population, although no indication
is given that any of the 16 plants are located in that state. A total of 58
persons died in this cohort compared to 104.3 expected. By contrast, deaths
due to malignant neoplasms ("Tumors" as they are referred to in the
translation) were very close to expected (20 observed versus 20.4 expected).
Similarly, 6 respiratory "tumors" were observed versus 6.9 expected. No
excessive risks appeared even considering "risk time" after 10 years and 15
years. With respect to lung tumors after a 15 years "risk time," there were 2
observed versus 5.6 expected (assuming "risk time" is the translators word for
follow-up). The authors noted that there were no existing data on
acrylonitrile concentration in the air in any work areas over the period of
the study.
There are several problems with this study. One of the more important
concerns was the selection criteria for inclusion of male workers in the
cohort. Originally the authors had in their possession lists containing 1973
names of former and active employees of the company. These lists were
submitted to present and former plant managers, supervisors, and foremen to
determine the degree to which each person on the list was exposed to
acrylonitrile and the periods of time covered by this exposure. To be
included each worker must have worked more than 12 months in one of the 16
plants and either worked at least 30% of his time with acrylonitrile, or if
less than 30%, he had to have been under a heavy acrylonitrile burden such as
would be found in test plants, according to the authors. Over a thousand
persons were eventually excluded from the cohort through this review process.
13-130
-------
Additionally, a justification is needed for the choice of the
Nordrheim-Westfalen state as a comparison population. The location of the 16
factories with respect to the comparison population is not given.
Furthermore, without exposure data it is impossible to tell how similar each
of these plants are with respect to the extent of exposure to acrylonitrile.
Also, the study cohort was a relatively youthful group even after the close of
the study. Only 13% of the total person-years accumulated were in the age
category 55 and over, the category in which the greatest increase in lung
cancer mortality could be expected to occur. In individuals under age 55,
lung cancer is a disease which appears infrequently and a rather large number
of person-years accumulating to such persons would be necessary to allow the
detection of a significant risk if in fact one exists, and only after a
sufficiently lengthy latent period has passed.
Furthermore, inconsistencies in the tabular data presented by the authors
cast doubt on the validity of the presented findings. Two examples are : 1)
the calculation of expected deaths in calendar-time periods where none should
appear based on the tabular heading description, and 2) dates of onset of
cancer in individuals who exhibit more exposure to acrylonitrile than their
ages warrant (one individual was exposed 12 years before he was born). In
short, it appears that this study suffers from a general lack of critical
review by its author.
Therefore, because of the problems and inconsistencies in this study, the
results do not provide evidence to support a lack of a cancer risk to
acrylonitrile-exposed employees and cannot be used to refute the results of
the O'Berg study as was suggested by the author.
13-132
-------
This selection can be affected by recall bias on the part of the
reviewers. They are more likely to better remember details regarding the
working status and working conditions of actively employed persons than of
inactive and former employees. Managers, supervisors, and foremen frequently
cannot recall even the names of former employees. Active employees, who have
better survivorship, are likely to be overrepresented as a result. The
combined total effect of such a biased selection, the influence of the healthy
worker effect, and the lack of a complete follow-up, such as occurred in this
study, could have led to a substantial underestimation of risk.
Another major concern involves the computation of expected deaths in this
study, i.e., the choice of an appropriate comparison population as well as the
proper allocation of person-years into the appropriate risk categories as the
members of the cohort pass through those categories in time. The greatest
deficit of lung cancer deaths compared to expected appeared in members of the
cohort who attained an age of 60 and over. With respect to all causes
combined, the greatest deficit occurred in persons 45 and over. Only 937
person-years were generated in the age group 60+ in this cohort (out of 13,375
person-years altogether), but the authors calculate that 3.708 lung cancer
deaths could have been expected to occur versus only 1 observed death in this
age group. This deficit of deaths in persons age 60 and above was reversed in
persons under age 60 (5 observed versus 3.23 expected). Such a sudden
reversal of risk ratios leads one to suspect that a methodologic problem
exists in the analysis. Since the greatest risk of deaths occurs to older
persons and is reflected in the death rates based on that age group, either an
improper allocation of person-years to older age categories or overestimates
of the death rates in the comparison population would have the effect of
adding additional expected deaths to that cause and would result in an
underestimate of the true risk.
13-131
-------
compared to 92 expected based on the above method. Eleven of these were
"neoplasms" versus 16.13 expected. Only 3 lung cancers were found versus 5.34
expected. This rather striking shortfall of deaths compared to expected can
be attributed only partially to the healthy worker effect and to the fact that
no cause of death was found for 4 of the 59 deaths. More importantly, at the
close of the study on December 31, 1977, few members of the cohort had reached
age 60 and beyond (less than 9%), the age category in which the greatest
mortality is likely to occur. In addition, if latent factors are also
considered, then it is unlikely that a carcinogenic effect would be seen in
this relatively youthful group. Time since onset of employment was not
considered, although the author himself suggested another 10 years of
surveillance would be required before any exposure-related deaths would occur.
No exposure data are provided. This study provides little evidence to support
the premise that exposure to acrylonitrile does not increase the risk of
cancer.
Stallard (1982, unpublished)
Stallard (1982, unpublished) conducted a cohort mortality study of 419
white male employees of an oil company facility in Lima, Ohio. The production
of acrylonitrile began at this plant March 14, 1960. Members of the cohort
had to have been employed at the facility at some time during the period March
14, 1960 to March 14, 1980 in a job that was assessed by plant industrial
hygiene personnel as having potential exposure to acrylonitrile. Follow-up
through March 14, 1980 was complete on 92% of the cohort (33 remained lost to
follow-up) expected deaths were generated through the utilization of 1970
age-cause specific death rates in white males for the United States and
separately for the State of Ohio.
13-134
-------
Herman (1981, unpublished)
A cohort mortality study of 989 male employees in two rubber plants at
Baton Rouge and Scotts Bluff, Louisiana was conducted by D. R. Herman (1981,
unpublished). There were 799 wage and 190 salaried employees having a minimum
of 1 year service in the plants during the period 1951 through 1977. The
author reported that race was known on 85.5% of this cohort of 989 males; 162
(24.8%) were black, while 654 (60.7%) were white. Vital status ascertainment
was accomplished on a cohort that included 88 females. In this larger cohort
97% were successfully followed to the end of 1977. Only 28 remained with an
unknown vital status. The 88 females were excluded from subsequent analyses
because of the small number.
Expected deaths were generated separately for wage versus salaried
employees as follows: for salaried workers, because almost all were white and
a few were unknown with respect to their race, expected deaths were generated
utilizing U.S. white male age and calendar-time-specific mortality rates only.
However, race was not known for 40% of the wage workers. A separate
calculation of expected deaths was done for known whites and blacks by the
method described above. In the remaining 40% with unknown race, two
calculations of expected deaths were accomplished, first, assuming they were
all white, and second, assuming they were all black. The author indicated
that these represented high and low estimates. Since the author also knew
that about 75% of the total cohort was white, a weighted average of these two
separate estimates was obtained for this group as follows:
Expected deaths (weighted) = 3/4 x (white expected deaths)
+ 1/4 x (black expected deaths)
No further explanation is given.
Overall, 59 observed deaths occurred to the 989 males of the cohort
13-133
-------
Only 7 deaths occurred during this time span versus 25.5 expected based on
U.S. rates, another surprising shortage of deaths overall. Of the seven
deaths, four were due to malignant neoplasms versus 4.66 expected based on
U.S. rates and 4.99 expected based on Ohio rates. Of these four malignant
neoplasms, two were due to lung cancer versus 1.56 expected based on U.S.
rates. Unfortunately, the expected deaths calculated for lung cancer based on
the rates for Ohio are in error (in the age group 45-64, the author shows a
figure 15.4 expected deaths; the correct figure is 1.54 and the resulting
corrected total number of expected lung cancer deaths all ages combined based
on Ohio death rates is 1.78).
Aside from the biases introduced by the incomplete follow-up of all
employees (former and present) and the healthy worker effect, a major problem
with this study is the relative youthfulness of the cohort even at the end of
the observation period (only one person in the successfully followed portion
of the cohort of some 386 persons reached age 65). In fact the author himself
reported that 87% of the cohort was under age 45 in 1970. Thus, the relative
youthfulness of the cohort and a follow-up of insufficient duration did not
allow an accurate assessment of the latent effects of the risk of cancer in
this cohort. The author himself states the study is "inconclusive" and cannot
be used to "support or refute whether or not acrylonitrile is carcinogenic in
humans."
13-135
-------
13.5.3 QUANTITATIVE ESTIMATION
Introduction
This quantitative section deals with the unit risk for acrylonitrile in air
and water and the potency of acrylonltrile relative to other carcinogens that
the CAG has evaluated. The unit risk estimate for an air pollutant is defined
as the lifetime cancer risk occurring in a hypothetical population in which all
individuals are exposed continuously from birth throughout their lifetimes to a
concentration of 1 ug/m^ of the agent in the air which they breathe. This
calculation is done to estimate in quantitative terms the impact of the agent as
a carcinogen. Unit risk estimates are used for two purposes: 1) to compare the
carcinogenic potency of several agents with each other, and 2) to give a crude
indication of the population risk which might be associated with air or water
exposure to these agents, if the actual exposures are known.
13.5.3.1 Procedures for Determination of Unit Risk
The data used for the quantitative estimate is one or both of two types:
1) lifetime animal studies, and 2) human studies where excess cancer risk has
been associated with exposure to the agent. In animal studies it is assumed,
unless evidence exists to the contrary, that if a carcinogenic response occurs
at the dose levels used in the study, then responses will also occur at all
lower doses with an incidence determined by the extrapolation model.
There is no solid scientific basis for any mathematical extrapolation model
that relates carcinogen exposure to cancer risks at the extremely low
concentrations that must be dealt with in evaluating environmental hazards. For
practical reasons such low levels of risk cannot be measured directly either by
animal experiments or by epidemiologic studies. We must, therefore, depend on
our current understanding of the mechanisms of carcinogenesis for guidance as to
13-136
-------
which risk model to use. At the present time the dominant view of the
carcinogenic process involves the concept that most agents that cause cancer
also cause irreversible damage to DNA. This position is reflected by the fact
that a very large proportion of agents that cause cancer are also mutagenic.
There is reason to expect the quanta! type of biological response that is
characteristic of mutagenesis is associated with a linear non-threshold
dose-response relationship. Indeed, there is substantial evidence from
mutagenicity studies with both ionizing radiation and a wide variety of
chemicals that this type of dose-response model is the appropriate one to use.
This is particularly true at the lower end of the dose-response curve; at higher
doses, there can be an upward curvature probably reflecting the effects of
multistage processes on the mutagenic response. The linear non-threshold
dose-response relationship is also consistent with the relatively few
epidemiologic studies of cancer responses to specific agents that contain enough
information to make the evaluation possible (e.g., radiation-induced leukemia,
breast and thyroid cancer, skin cancer induced by arsenic in drinking water,
liver cancer induced by aflatoxin in the diet). There is also some evidence
from animal experiments that is consistent with the linear non-threshold model
(e.g., liver tumors induced in mice by 2-acetylaminofluorene in the large scale
EDgi study at the National Center for Toxicological Research and the
initiation stage of the two-stage carcinogenesis model in rat liver and mouse
skin).
Because it has the best, albeit limited, scientific basis of any of the
current mathematical extrapolation models, the linear non-threshold model has
been adopted as the primary basis for risk extrapolation to low levels of the
dose-response relationship. The risk estimates made with this model should be
regarded as conservative, representing the most plausible upper-limit for the
13-137
-------
risk, i.e., the true risk is not likely to be higher than the estimate, but it
could be lower.
The mathematical formulation chosen to describe the linear non-threshold
dose-response relationship at low doses is the linearized multistage model.
This model employs enough arbitrary constants to be able to fit almost any
monotonically increasing dose-response data and it incorporates a procedure for
estimating the largest possible linear slope (in the 95% confidence limit sense)
at low extrapolated doses that is consistent with the data at all dose levels of
the experiment.
Description of the Low Dose Extrapolation Model--
Let P(d) represent the lifetime risk (probability) of cancer at dose d. The
multistage model has the form
P(d) = 1 - exp [-(q0 + q^ 4 q2d2 + ... 4 q
where
q. > 0, i = 0, 1, 2, .... k
I ™"™'
Equivalently,
A(d) = 1 - exp [(qjd + q2d2 + ... + qRdk)]
where
A(d) =
_ P(d) - P(o)
- P(o)
is the extra risk over background rate at dose d.
The point estimate of the coefficents qi, i = 0, 1, 2 k, and
consequently the extra risk function A(d) at any given dose d, is calculated by
maximizing the likelihood function of the data.
13-138
-------
The point estimate and the 95% upper confidence limit of the extra risk A(d)
are calculated by using the computer program GLOBAL 79 developed by Crump and
Watson (1979). At low doses, upper 95% confidence limits on the extra risk and
lower 95% confidence limits on the dose producing a given risk are determined
from a 95% upper confidence limit, qt, on parameter q.. Whenever
q, > 0, at low doses the extra risk A(d) has approximately the form A(d) =
q. x d. Therefore, q. x d is a 95% upper confidence limit on the extra
risk and R/qt is a 95% lower confidence limit on the dose producing an
extra risk of R. Let LQ be the maximum value of the log-likelihood function.
The upper limit q1 is calculated by increasing q. to a value qt
such that when the log-likelihood is remaximized subject to this fixed value
qf for the linear coefficient, the resulting maximum value of the
log-likelihood LI satisfies the equation
2 (L0 - L\) = 2.70554
where 2.70554 is the cumulative 90% point of the chi-square distribution with
one degree of freedom, which corresponds to a 95% upper-limit (one-sided). This
approach of computing the upper confidence limit for the extra risk A(d) is an
improvement on the Crump et al. (1977) model. The upper confidence limit for
the extra risk calculated at low doses is always linear. This is conceptually
consistent with the linear non-threshold concept discussed earlier. The slope,
q|, is taken as an upper bound of the potency of the chemical in inducing
cancer at low doses.
In fitting the dose-response model, the number of terms in the polynomial is
chosen equal to (h-1), where h is the number of dose groups in the experiment
including the control group.
Whenever the multistage model does not fit the data sufficiently well, data
13-139
-------
at the highest dose is deleted and the model is refitted to the rest of the
data. This is continued until an acceptable fit to the data is obtained. To
determine whether or not a fit is acceptable, the chi-square statistic
h
2 (Xi - N-iP-^2
A ~ I. p* 7 ^ "i
1 = 1
is calculated where Ni is the number of animals in the ith dose group, XT
is the number of animals in the i^1 dose group with a tumor response, P^ is
the probability of a response in the itn dose group estimated by fitting the
multistage model to the data, and h is the number of remaining groups. The fit
is determined to be unacceptable whenever X^ is larger than the cumulative 99%
point of the chi-square distribution with f degrees of freedom, where f equals
the number of dose groups minus the number of non-zero multistage coefficients.
Selection of Data--
For some chemicals, several studies in different animals species, strains,
and sexes, each run at several doses and different routes of exposure, are
available. A choice must be made as to which of the data sets from several
studies to use in the model. It may also be appropriate to correct for
metabolism differences between species and absorption factors via different
routes of administration. The procedures used in evaluating these data are
consistent with the approach of making a maximum-likely risk estimate. They are
1isted below.
1. The tumor incidence data are separated according to organ sites or tumor
types. The set of data (i.e., dose and tumor incidence) used in the model is
the set where the incidence is statistically significantly higher than the
13-140
-------
control for at least one test dose level and/or where the tumor incidence rate
shows a statistically significant trend with respect to dose level. The data
set which gives the highest estimate of the lifetime carcinogenic risk, q*,
is selected in most cases. However, efforts are made to exclude data sets which
produce spuriously high risk estimates because of a small number of animals.
That is, if two sets of data show a similar dose-response relationship, and one
has a very small sample size, the set of data which has larger sample size is
selected for calculating the carcinogenic potency.
2. If there are two or more data sets of comparable size which are
identical with respect to species, strain, sex, and tumor sites, the geometric
mean of q* estimated from each of these data sets, is used for risk
assessment. The geometric mean of numbers A}, A2, ..., A,,, is defined as
m
x A2 x ... x
3. If two or more significant tumor sites are observed in the same study,
and if the data are available, the number of animals with at least one of the
specific tumor sites under consideration is used as incidence data in the model.
4. Following the suggestion of Mantel and Schneiderman (1977), we assume
that mg/surface area/day is an equivalent dose between species. Since, to a
close approximation, the surface area is proportional to the 2/3rds power of the
weight as would be the case for a perfect sphere, the exposure in mg/day per
2/3rds power of the weight is also considered to be equivalent exposure. In an
animal experiment this equivalent dose is computed in the following manner.
Let
Le = duration of experiment
le = duration of exposure
m = average dose per day in mg during administration of the agent (i.e.,
during le), and
W = average weight of the experimental animal
13-141
-------
Then, the lifetime average exposure is
1 x m
d=
I. x W2/3
Often exposures are not given in units of ing/day and it becomes necessary
to convert the given exposures into nig/day. For example, in most feeding
studies exposure is in terms of ppm in the diet. In this case the exposure in
ing/day is
m = ppm x F x r
where ppm is parts per million in the diet of the carcinogenic agent, F is the
weight of the food consumed per day in kg, and r is the absorption fraction.
In the absence of any data to the contrary, r is assumed to be equal to one.
For a uniform diet, the weight of the food consumed is proportional to the
calories required, which in turn is proportional to the surface area or 2/3rds
power of the weight, so that
2/3
m a ppm x W x r
or
a ppm
rW2/3
As a result, ppm in the diet is often assumed to be an equivalent exposure
between species. However, we feel that this is not justified since the
calories/kg of food is very different in the diet of man compared to laboratory
animals primarily due to moisture content differences. Instead we use an
empirically-derived food factor, f = F/W, which is the fraction of a species
13-142
-------
body weight that is consumed per day as food. We use the following rates:
Species W f
Man 70 0.028
Rats 0.35 0.15
Mice 0.03 0.13
Thus, when the exposure is given as a certain dietary concentration in ppm, the
exposure in mg/w2/3 is
m = £E1xF = ppm x f x U = x f
w2/3
When exposure is given in terms of mg/kg/day = m/Wr = s, the conversion is
simply
- s x
W1/3
When exposure is via inhalation, the calculation of dose can be considered
for two cases where 1) the carcinogenic agent is either a completely water-
soluble gas or an aerosol, and is absorbed proportionally to the amount of air
breathed in, and 2) where the carcinogen is a poorly water-soluble gas which
reaches an equilibrium between the air breathed and the body compartments.
After equilibrium is reached, the rate of absorption of these agents is expected
to be proportional to the metabolic rate, which in turn is proportional to the
rate of oxygen consumption, which in turn is a function of surface area.
Case 1—
Agents that are in the form of particulate matter or virtually completely
absorbed gases, such as S02, can reasonably be expected to be absorbed
proportional to the breathing rate. In this case the exposure in mg/day may be
expressed as
m = I x v x r
13-143
-------
where I = inhalation rate per day in m3, v = mg/m3 of the agent 1n air, and
r = the absorption fraction.
The inhalation rates, I, for various species can be calculated from the
observations (FASEB 1974) that 25 g mice breathe 34.5 liters/day and 113 g rats
breathe 105 liters/day. For mice and rats of other weights, W (in kilograms),
the surface area proportionality can be used to find breathing rates in m3/day
as follows:
For mi
For rats
ice, I = 0.0345 (W/0.025)2/3 m3/day
ats, I = 0.105 (W/0.113)2/3 m3/day
For humans, the values of 20 m3/day* is adopted as a standard breathing rate
(ICRP 1977).
The equivalent exposure in mg/W2/3 for these agents can be derived from
the air intake data in a way analogous to the food intake data. The empirical
factors for the air intake per kg per day, i = I/W, based upon the previous
stated relationships are tabulated as follows:
Species W i = I/W
Man 70 0.29
Rats 0.35 0.64
Mice 0.03 1.3
Therefore, for particulates or completely absorbed gases, the equivalent
exposure in mg/W2/3 is
. m_ _ Ivr
w2/3 " W2/3
*From "Recommendation of the International Commission on Radiological
Protection," page 9. The average breathing rate is 10? cm3 per 8-hour
workday and 2 x 107 cm3 in 24 hours.
13-144
-------
In the absence of experimental information or a sound theoretical argument to
the contrary, the fraction absorbed, r, is assumed to be the same for all
species.
Case 2—
The dose in mg/day of partially soluble vapors is proportional to the Q£
consumption, which in turn is proportional to W^/3 and is also proportional to
the solubility of the gas in body fluids, which can be expressed as an
absorption coefficient, r, for the gas. Therefore, expressing the 02
consumption as 02 = k W^/3> where k is a constant independent of species, it
follows that
m = k W^/3 x v x r
or
d = —JJ-TT = kvr
W2/3
As with Case 1, in the absence of experimental information or a sound
theoretical argument to the contrary, the absorption fraction, r, is assumed to
be the same for all species. Therefore, for these substances a certain
concentration in ppm or ug/m^ in experimental animals is equivalent to the
same concentration in humans. This is supported by the observation that the
minimum alveolar concentration necessary to produce a given "stage" of
anesthesia is similar in man and animals (Dripps et al. 1977). When the animals
are exposed via the oral route and human exposure is via inhalation or
vice-versa, the assumption is made, unless there is pharmacokinetic evidence to
the contrary, that absorption is equal by either exposure route.
13-145
-------
5. If the duration of experiment (Le) is less than the natural lifespan
of the test animal (L), the slope q*, or more generally the exponent g(d),
is increased by multiplying a factor (L/Le)3. We assume that if the average
dose d, is continued, the age-specific rate of cancer will continue to increase
as a constant function of the background rate. The age-specific rates for
humans increase at least by the 2nd power of the age and often by a considerably
higher power as demonstrated by Doll (1971). Thus, we would expect the
cumulative tumor rate to increase by at least the 3rd power of age. Using this
fact, we assume that the slope qt, or more generally the exponent g(d),
would also increase by at least the 3rd power of age. As a result, if the slope
q* [or g(d)] is calculated at age Le> we would expect that if the
experiment had been continued for the full lifespan, L, at the given average
exposure, the slope q| [or g(d)] would have been increased by at least
(L/Le)3.
This adjustment is conceptually consistent with the proportional hazard
model proposed by Cox (1972) and the time-to-tumor model considered by Crump
(1979) where the probability of cancer by age t and at dose d is given by
P(d,t) = 1 - exp [-f(t) x g(d)]
Calculation of the Unit Risk
The risk associated with d mg/kg2/3/day is obtained from GLOBAL 79 and,
for most cases of interest to risk assessment, can be adequately approximated by
P(d) = 1 - exp (-q?d). A "unit risk" in units X is simply the risk
corresponding to an exposure of X = 1. To estimate this value we simply find
the number of mg/kg^/3/day corresponding to one unit of X and substitute this
value into the above relationship. Thus, for example, if X is in units of
13-146
-------
ug/m3 in the air, we have that for case (1) d = 0.29 x 701/3 x 10~3
mg/kg2/3/day and for case (2) d = 1, when ug/m3 1s the unit used to compute
parameters in animal experiments.
If exposures are given in terms of ppm in air, we may simply use the fact
that
1 ppm = 1.2 x molecular weight (gas) mg/m3
molecular weight
(gas)
TaTr)
Note, an equivalent method of calculating unit risk would be to use mg/kg for
the animal exposures and then increase the jth polynomial coefficient by an
amount
(wh/wa)J/3 ; j = i, 2, ..., k
and use mg/kg equivalents for the unit risk values.
Estimation of Unit Risk Based on Human Data
If human epidemiologic studies and sufficiently valid exposure information
are available for the compound, they are always used in some way. If they show
a carcinogenic effect, the data are analyzed to give an estimate of the linear
dependence of cancer rates on lifetime average dose, which is equivalent to the
factor BH» If they show no carcinogenic effect when positive animal evidence
is available, then it is assumed that a risk does exist, but it is smaller than
could have been observed in the epidemiologic study, and an upper-limit to the
cancer incidence is calculated assuming hypothetically that the true incidence
is just below the level of detection in the cohort studied, which is determined
largely by the cohort size. Whenever possible, human data are used in
preference to animal bioassay data.
Very little information exists that can be utilized to extrapolate from high
exposure occupational studies to low environmental levels. However, if a number
13-147
-------
of simplifying assumptions are made, it is possible to construct a crude dose-
response model whose parameters can be estimated using vital statistics,
epidemiologic studies, and estimates of worker exposures.
In human studies, the response is measured in terms of the relative risk of
the exposed cohort of individuals compared to the control group. The
mathematical model employed assumes that for low exposures the lifetime
probability of death from lung cancer (or any cancer), PQ, may be represented
by the 1inear equation
P0 = A + BHx
where A is the lifetime probability in the absence of the agent, and x is the
average lifetime exposure to environmental levels in some units, say ppm. The
factor, BH, is the increased probability of cancer associated with each unit
increase of r, the agent in air.
If we make the assumption that R, the relative risk of lung cancer for
exposed workers, compared to the general population, is independent of the
length or age of exposure but depends only upon the average lifetime exposure,
it follows that
R - JL - A + _B_n (x, + x?)
- P0 " A + BH X!
or
RP0 = A + BH (K! + x2)
where Xj = lifetime average daily exposure to the agent for the general
population, X£ = lifetime average daily exposure to the agent in the
occupational setting, and PQ = lifetime probability of dying of cancer with no
or negligible acrylonitri le exposure.
13-148
-------
Substituting P« = A + BH x, and rearranging gives
BH = P0 (R - l)/x2
To use this model, estimates of R and x? must be obtained from the epidemic-
logic studies. The value PQ is derived from the age-cause-specific death
rates for combined males found in 1976 U.S. Vital Statistics tables using the
life table methodology. For lung cancer the estimate of PQ is 0.036. This
methodology is used in the section on unit risk based on human studies.
Interpretation of Quantitative Estimates
For several reasons, the unit risk estimate is only an approximate indication
of the upper-bound on absolute risk in populations exposed to known carcinogen
concentrations. First, there are important species differences in uptake,
metabolism, and organ distribution of carcinogens, as well as species differences
in target site susceptibility, immunological responses, hormone function,
dietary factors, and disease. Second, the concept of equivalent doses for
humans compared to animals on a mg/surface area basis is virtually without
experimental verification regarding carcinogenic response. Finally, human
populations are variable with respect to genetic constitution and diet, living
environment, activity patterns, and other cultural factors.
The unit risk estimate can give a rough indication of the relative potency
of a given agent compared with other carcinogens. The comparative potency of
different agents is more reliable when the comparison is based on studies in
the same test species, strain, and sex, and by the same route of exposure,
preferably by inhalation.
The quantitative aspect of the carcinogen risk assessment is included
here because it may be of use in the regulatory decision-making process, e.g.,
13-149
-------
setting regulatory priorities, evaluating the adequacy of technology-based
controls, etc. However, it should be recognized that the estimation of cancer
risks to humans at low levels of exposure is uncertain. At best, the linear
extrapolation model used here provides a rough, but plausible estimate of the
upper-limit of risk; i.e., it is not likely that the true risk would be much
more than the estimated risk, but it could very well be considerably lower. The
risk estimates presented in subsequent sections should not be regarded as an
accurate representation of the true cancer risks even when the exposures are
accurately defined. The estimates presented may be factored into regulatory
decisions to the extent that the concept of upper risk limits is found to be
useful.
Alternative Methodological Approaches
The methods used by the CAG for quantitative assessment are consistently
conservative, i.e., tending toward high estimates of risk. The most important
part of the methodology contributing to this conservatism in this respect is the
linear non-threshold extrapolation model. There are a variety of other
extrapolation models that could be used, all of which would give lower risk
estimates. These alternative models have not been used by the CAG because they
have no biological basis, whereas the linear model has at least some biological
plausibility.
The position is taken by the CAG that the risk estimates obtained by use of
the linear non-threshold model are upper-limits and the true risk could be
1 owe r.
Another alternative method involves the choice of animal bioassay as the
basis for extrapolation. The present approach is to use the most sensitive
responder. Alternatively, the average responses of all of the adequately tested
13-150
-------
bioassay animals could be used. In the case of acrylonitrile, only rats were
tested in the various studies with little evidence for substantial variation in
tumorigenic susceptibility.
Extrapolations from animals to humans could also be done on the basis of
relative weights rather than surface areas. The latter approach, used here, has
more basis in human pharmacological responses; it is not clear which of the two
approaches are more appropriate for carcinogens. In the absence of information
on this point, it seems appropriate to use the most generally obtained method,
which also is more conservative.
13.5.3.2 Unit Risk Estimate Based on Human Studies
Of the ten epidemiologic studies reviewed in this document, the study of
workers at the Du Pont May Plant, Camden, South Carolina (O'Berg 1980) .clearly
presented the most significant evidence of acrylonitrile as a human lung
carcinogen. Furthermore, it is the only plant for which we have been able to
estimate exposure levels. Therefore, this study is used to estimate a unit risk
for human inhalation. Furthermore, the lack of exposure levels for the other
studies precludes estimating an upper-limit risk based on negative studies.
The carcinogenic effectiveness (potency) of acrylonitrile, BH, is
calculated as follows:
where PQ = 0.036, the background lifetime probability of death due to
respiratory cancer. This factor is derived from the 1976 U.S. Vital Statistics
tables via a life table calculation (CAG 1978).
13-151
-------
The Du Pont Chemical Company follow-up study by O'Berg (1980) of 1,345
workers exposed to acrylonitrile between I960 and 1966 found the observed and
expected number of cancer cases to be 13 versus 5.5, respectively, in workers
with at least moderate exposure and with a probable latency period of at least
15 years. This excess incidence was found to be statistically significant
(P < 0.05). Half of this excess was due to respiratory cancer, 5 versus 1.4,
which is also statistically significant (P < 0.05). If we adjust for the
effects of smoking as described earlier (p 34), we have 5 observed vs. 1.6
expected, which is statistically significant. Therefore, the relative risk, R,
is 5.0/1.6 = 3.1. For this analysis we will equate respiratory cancer incidence
with mortality.
Estimates of exposure at the plant are those suggested by J. Brown, an
industrial hygienist at N10SH who visited the plant in 1978. She states, "While
no statistical data were used to establish the system used (to rank the included
cohort according to exposure), the Du Pont representatives agreed that 8-hour
time-weighted averages might be used to represent possible past exposure levels,
based principally on past recall of process levels taken some years past" (NIOSH
1978). Also in the Brown report were some measurements of potential ACN
exposure which averaged between 1 and 3 ppm for the first three quarters of 1977
and less than 1 ppm for the last quarter.
Since the workers were exposed to at least a moderate level (10 ppm =
moderate level, 20 ppm = high level), we will assume 15 ppm to be the 8-hour
time-weighted average (TWA). We estimate the uncertainty attached to this
estimate to be as high as a factor of 5.
To convert the 8-hour TWA exposure to a lifetime average, we use
8 240 9
Xl = 15 PPm x 24 X 365 X 60
= 0.5 ppm continuous equivalent lifetime exposure
13-152
-------
where 9 years is estimated to be the average exposure duration, and 60 years is
estimated to be the maximum possible age at the end of the observation period.
The value of BH [in (ppb)-1] is derived using the above estimates as
follows:
R 0.036 (3.1 - 1)
H 560 ppb
=1.5 x 10-4 (ppb)-l
Therefore, the lifetime risk of cancer for people continuously exposed to 1 ppb
is 1.5 x 10-4.
To express risk in terms of ug/m3 concentration, the conversion factor for
acrylonitrile is
1 ppb . 1.2 x * ffy""""-"' . M"-°6) . 2.21 ug/m3
or
1 ug/m3 = 0.45 ppb
Therefore the upper-bound risk associated with a lifetime exposure of 1
in air is
P = 1.5 x 10'4 x 0.45 = 6.8 x 10-5
Unit Risk Estimate Based on Animal Studies
For comparison, a unit risk assessment is calculated based on both
inhalation and drinking water studies (Quast et al. 1980a, b).
13-153
-------
Drinking Water Study--
The three rat bioassays of acrylonitrile in drinking water have all shown
significant increases in brain and/or spinal cord astrocytomas, Zymbal gland
carcinomas, and stomach papillomas/carcinomas (Tables 13-21 through 13-24).
Of all three studies, the Dow Chemical Company rat study (Quast et al.
1980a) showed the highest incidence of tumors per unit of administered dose. On
the assumption that humans are as sensitive as the most sensitive of the test
animals, this study was chosen as the basis of the risk estimate. Since several
tumor sites are affected, the overall risk of tumors is determined from the
number of animals having tumors at any of the sites. This is shown in Table 13-35.
Also presented is the percentage of animals by sex with the tumors discussed
above. In addition, the small intestine and mammary gland, which are target
sites for rats, are included. For the mammary gland, however, because the
control group incidence is so high, the percentages for the females are
tabulated separately.
To convert the animal doses into a human equivalent dose, the standard
approach is to equate the doses on the basis of mg per body surface area.
Estimating the weight of the rats to be about 350 grams, the human dosage
equivalent to 3.42 mg/kg/day for the rat (Table 13-35) is
3.42 mg/kg/day x (70/0.350)1/3 = 0.6 mg/kg/day
When these human dose equivalents are used with the animal response data in
fable 13-35, the multistage model yields a value of qf = 1.7 x 10
(mg/kg/day)~l for the males, qt = 1.6 x 10~1 (mg/kg/day)-l for the
females excluding mammary gland tumors, and qf = 4.2 x 10"1
(mg/kg/day)-l for the females including mammary gland tumors. The latter
value is chosen as the basis for the unit risk estimate, since it is the
13-154
-------
TABLE 13-35. NUMBER (PERCENTAGE) OF RATS DEVELOPING TUMORS IN AT LEAST ONE OF THE
FOLLOWING TARGET ORGANS: ZYMBAL GLAND, TONGUE, STOMACH, BRAIN, AND SPINAL
CORD BY DOSE AND BY SEX
(adapted from Quast et al. 1980a)
Males Females*
Animal Dose
(mg/kg/day)
0
3.42
8.53
21.18
Number/Total
(Percent)
4/80 (5.0%)
18/47 (38.3%)t
36/48 (75.9%)t
45/48 (93.8»)t
Animal Dose
(mg/kg/day)
0
4.36
10.76
24.97
Including Mammary
Gland Tumors
Number/Total
(Percent)
59/80 (73.8%)
47/48 (97.9%)§
46/48 (95.8%)§
48/48 (100%)§
Excluding Mammary
Gland Tumors
Number/Total
(Percent)
3/80 (3.8%)
24/48 (50.0%)t
37/48 (77.1%)t
45/48 (93.8%)t
*Includes those with small intestine tumors.
tP < 0.00001.
§P < 0.001.
13-155
-------
highest value, and since it includes all of the organ sites at risk to tumor
induction. If the human dose had been assumed equivalent on a mg/kg/day basis,
the value of these parameters would be smaller by a factor of
3Nf(70/0.35f = 5.8
In order to estimate the unit risk for 1 ug/1 of acrylonitrile in drinking
water, we assume that the average 70 kg person drinks 2 liters of water per day.
This corresponds to a daily dose of 2 ug/day x 10-3 mg/ug x 1/70 kg = 2.86 x
10"5 mg/kg/day. The upper limit unit risk corresponding to 1 ug/1
acrylonitrile concentration in water is then
P = 1 . e~4'2 x I0"1 x 2.86 x lO'5
= 1.2 x 10-5
The drinking water study could also be used to estimate the inhalation risk,
although such an estimate is expected to be unreliable because of the difference
in exposure routes. The dose rate, d(mg/kg/day), resulting from breathing 20
m3/day of air containing a concentration of 1 ug/m3 can be determined if one
assumes that 100% of the inhaled acrylonitrile is absorbed into the body. This
was shown to be the case by Young et al. (1977). With this assumption, the dose
rate is
d = 1 ug/m3 x 20 m3/day x 10~3 mg/ug x 1/70 kg
= 2.86 x 10-4 mg/kg/day
The upper-limit estimate of the air unit risk, P, can be found using this value
of d and the value of q* estimated above as follows:
P = 1 _ e-4-2 x l°~l x 2*86 x 10~4
= 1.2 x 10-4
13-156
-------
This is a factor of 2 greater than that derived from the human epidemiologic
studies and one order of magnitude greater than that derived from the animal
inhalation studies (see below).
Inhalation Study--
Of the two inhalation bioassays previously discussed, the Quast et al.
(1980b) study produced a clear carcinogenic effect, while the Maltoni et al.
(1977) study produced no statistically significant increases in tumors in rats
exposed to low dosages for 12 months. Therefore, the Quast study is used for
determining a unit risk estimate via the inhalation route.
The tumor incidence data on individual tumor types have been presented pre-
viously (Table 13-32). For the males the response occurs at the following four
target organs: Zymbal gland, small intestine, brain, and/or spinal cord. For
males the percentages of rats with tumors at one or more of the target organ
sites are 5% (5/100), .9% (9/100), and 47% (47/100) for the 0 ppm, 20 ppm, and 80
ppm groups, respectively. For the females the corresponding percentages are 0%,
(0/100), 9% (9/100), and 31% (31/100). In analyzing these data, both sets of
responses were fit by the model. The estimate of carcinogenic potency for the
females was slightly higher and is presented below.
In this study, animals were exposed to 0, 20, or 80 ppm of acrylonitrile 6
hours per day, 5 days per week, for 2 years. Therefore, the lifetime average
concentration for the 20 ppm group is
20 x "2T * 7 = 3*57 ppri
Similarly, the lifetime equivalent dosage for the 80 ppm group is 14.29 ppm.
Using the multistage model with incidence data for females, the carcinogenic
potency is qt = 3.35 x 10~2 (ppm)-l, and the upper-bound estimate of
the lifetime risk of cancer associated with 1 ug/m^ = 4.53 x 10~4 ppm of
13-157
-------
acrylonitrile is
p = 1 - exp (-3.35 x ID'2 x 4.53 x 10'4)
= 1.5 x 10-5
In summary, the upper-limit unit risk estimates for humans breathing
1 ug/m3 of acrylonitrile in ambient air (equivalent to 0.45 ppb) are
6.8 x 10-5 based on the occupational study, 1.2 x 10-4 based on the rat
drinking water study, and 1.5 x 10~5 based on the rat inhalation study.
Parenthetically, it should be noted that if the human equivalent dose assumption
were changed to dose per body weight, the unit risk for inhalation based on the
rat drinking water study would be 1.2 x 10~4 x 1/5.8 = 2.1 x 10~5, a value
which is close to the other two estimates. Although this estimate is considered
unreliable because of the inappropriate route of administration, it is included
here as a matter of interest. The upper-limit unit risk for 1 ug/1 of
acrylonitrile in drinking water is estimated to be 1.2 x 10"^.
Relative Potency
One of the uses of unit risk is to compare the potency of carcinogens. To
estimate the relative potency, the unit risk slope factor is divided by the
molecular weights and the resulting number expressed in terms of
(mMol/kg/day)-l. This is called the relative potency index.
Figure 13-7 is a histogram representing the frequency distribution of potency
indices of 54 suspect carcinogens evaluated by the CAG. The actual data summar-
ized by the histogram is presented in Table 13-36. When human data are
available for a compound, they have been used to calculate the index. When no
human data are available, animal oral studies and animal inhalation studies have
been used in that order.
13-158
-------
The potency index for acrylonitrile based on the O'Berg study of Dupont
workers is 4.5 x 10~3 (mMol /kg/day)'1. This is derived as follows: the
slope estimate from the O'Berg study [6.8 x 10~5(ug/m3)-l], is first
converted to units of (mg/kg/day)"*, assuming a breathing rate of ZOm3 of
air per day and a 70 kg person.
6.8 x 10-5(ug/ra3)-l x 1^- x — l^L x 70 kg
v 3 ' 20 m3 10-3 mg
-1
= 0.24(mg/kg/day)
Dividing by the molecular weight of 53.1 gives a potency index of 4.5 x
10~3. Rounding off to the nearest order of magnitude gives a value of 10~3
which is the scale presented on the horizontal axis of Figure 13-7. The index of
4.5 x ID"3 lies in about the middle of the third quartile of the 54 suspect
carcinogens.
13-159
-------
12-
10-
u
c
5xlO"4
and ^6x10-3
Index>6xlO"3
and <6xlO'2
1st quartile: Index>6xlO
-6 -5 -4 -3 -2 -1 0 +1 +2 +3 +4
10 10 10 10 10 10 10 10 10 10 10
Potency Index
Figure 13-7. Histogram representing frequency distribution of the potency
indices of 54 suspect carcinogens evaluated by the Carcinogen
Assessment Group.
13-160
-------
TABLE 13-36. RELATIVE CARCINOGENIC POTENCIES AMONG SUSPECT CARCINOGENS
EVALUATED BY THE CARCINOGEN ASSESSMENT GROUP
Compounds
Acrylonitrile
Aflatoxin Bj
Allyl Chloride
Aldrin
Arsenic
B[a]P
Benzene
Benzidine
Beryllium
Cadmium
Carbon Tetrachloride
Chlordane
Hexachlorobenzene
1 ,2-dichloroethane
1 ,1 ,2-trichl oroethane
1 ,1 ,2,2-tetrachloroethane
Hexachl oroethane
2,4 ,6-trichlorophenol
Bis(2-chloroethyl )ether
Bis(chloromethyl )ether
Chloroform
Chromium
Dichlorobenzidine
DDT
Slope
(mg/kg/day)
0.24
2924
1.19x10-2
11.4
14(H)
11.5
5.2xlO-2
234(W)
4.86
6.65(1)
8.28xlO-2
1.61
1.67
3.70x10-2
5.73x10-2
0.20
1.42x10-2
1.99x10-2
1.14
9300(1)
0.11
63(W)
1.69
8.42
Molecular Potency
Weight Index
53.1
312.3
76.5
369.4
149.8
252.3
78
184.2
9
112.4
153.8
409.8
284.4
98.9
133.4
167.9
236.7
197.4
143
115
119.4
104
253.1
354.5
4.5xlO-3
9xlO+0
2xlO-4
3x10-2
9x10-2
5x10-2
7xlO-4
lxlO+0
SxlO'1
6x10-2
5x10-4
4xlO-3
6xlO-3
4xlO-4
4x10-4
IxlO-3
6x10-5
1x10-4
8x10-3
8xlO+l
9x10-4
6X10-1
7xlO-3
2x10-2
(continued on the
Order of
Magnitude
(exponent of
base 10)
-3
0
-4
-2
-2
-2
-4
0
-1
-2
-4
-3
-3
-4
-4
-3
-5
-4
-3
1
-4
-1
-3
-2
following page)
13-161
-------
TABLE 13-36. (continued)
Compounds
1,1-dichloroethylene
Dieldrin
Dinitrotoluene
Tetrachlorodioxin
Diphenylhydrazine
Epichlorohydrin
Ethyl ene Di bromide (EOB)
Ethylene Dichloride (EDC)
Ethyl ene Oxide
Formaldehyde
Heptachlor
Hexachlorobutadiene
HexacM orocycl ohexane
technical grade
alpha Isomer
beta isomer
gamma isomer
Nickel
Nitrosamines
Di methyl nitrosamine
Diethylnitrosamine
Oibutylnitrosamine
N-n1trosopyrrol Idine
N-nitroso-N-ethylurea
N-nitroso-N-methyl urea
N-ni troso-di phenyl ami ne
PCBs
Slope
(mg/kg/day)
1.04
30.4
0.31
4.25x10$
0.77
7.69xlO-4(I)
8.51
1.44x10-2
1.86x10-2(1)
2.14x10-2(1)
3.37
7.75x10-2
4.75
11.12
1.84
1.33
6.30(W)
25.9(not by q*)
43.5(not by q{)
5.43 '
2.13
32.9
302.6
4.92xlO-3
4.34
Molecular
Weight
97
380.9
182
322
180
92.5
187.9
99.0
44.0
30
373.3
261
290.9
290.9
290.9
290.9
58.7
74.1
102.1
158.2
100.2
117.1
103.1
198
324
Potency
Index
1x10-2
8x10-2
2xlO-3
lxlO+3
4xlO-3
8xlO"6
5x10-2
1x10-1
4xlO-4
7xlO-4
9x10-3
3xlO'4
2x10-2
4x10-2
6xlO-3
5x10-3
IxlO'1
4x10-*
4x10-1
3x10-2
2x10-2
3x10-1
3x1 0+0
2x10-5 -
IxlO-2
Order of
Magnitude
(exponent of
base 10)
-2
-2
-3
3
-3
-6
-2
-4
-4
-4
-3
-4
-2
-2
-3
-3
-1
-1
-1
-2
-2
-1
0
-5
-2
(continued on the following page]
13-162
-------
TABLE 13-36. (continued)
Compounds
Tetrachloroethylene
Toxaphene
Trichloroethylene
Vinyl Chloride
Vinyl idene Chloride
Slope
(mg/kg/day)
5.31x10-2
1.13
1.26x10-2
1.75x10-2(1)
0.13(1)
Molecular
Weight
165.8
414
131.4
62.5
97
Potency
Index
3xlO-4
3xlO-3
IxlO-4
3xlO-4
ixin-3
Order of
Magnitude
(exponent of
base 10)
-4
-3
-4
-4
-3
Remarks:
1. Slopes (q?) in (mg/kg/day)-1 are calculated based on animal oral studies,
except for those indicated by I (Animal inhalation), W (human occupational exposure),
and H (human drinking water exposure).
2. The potency index Is a rounded-off slope 1n (mMol/kg/day)'1 and is calculated by
dividing the slopes in (mg/kg/day)-l by the molecular weight of the compound.
3. Not all the carcinogenic potencies presented in this table are final. Some are
subject to change as the CAG is getting the individual risk assessment documents
approved.
13-163
-------
13.5.4 SUMMARY
Qualitative Assessment
Acrylonitrile is not a direct acting carcinogen and hence the localization
and nature of the effects depend on its metabolism. It is probable that the
proximal carcinogen is 2-cyanoethylene oxide, since it has been demonstrated
as a reaction product with calf thymus DNA. However, the metabolite has not
been tested directly for its carcinogenicity. It has been shown to be
produced in the liver and possibly circulates to other organs. However,
studies have not been done to determine where else in the body this metabolite
is produced. There appears to be a clear difference between animals and
humans in their tumorigenic response to acrylonitrile: no lung tumors have
been produced in animals and no brain tumors have been observed in humans.
There are no human studies on the metabolism of acrylonitrile and there are no
pharmacokinetic studies that would be relevant to the characterization of
dose-response relationships at low levels of exposure.
The carcinogenicity of acrylonitrile has been studied in seven cancer
bioassays in rats: four in drinking water, one by gastric intubation, and two
by inhalation. These are summarized in Table 13-37. In addition, ten
epidemiologic studies of cancer incidence have been reported. These are
summarized in Table 20. A short description of these studies is presented
below.
Quast et al. (1980a) administered acrylonitrile in drinking water to
Sprague-Dawley rats for 2 years at dose levels of 35, 100, and 300 ppm. A
statistically significant incidence of tumors was observed in the central
nervous system, Zymbal gland, stomach, tongue, and small intestine in both
male and female rats, as well as in the mammary gland of female rats. The
occurrence of central nervous system and Zymbal gland tumors in Sprague-Dawley
13-164
-------
TABLE 13-37. CARCINOGEN1CITY OF ACRYLONITRILE IN RATS
Reference
Strain of Rat
Route of
Administration
Dose Level*
Treatment
Schedule
Site of Tumors
CO
i
Quast et al. (1980a)
Sprague-Dawley Ingestlon 1n 35-300 ppm 24 mo.
drinking water
Bellies et al. (1980) Charles River Ingestlon 1n 500 ppm
drinking water
Blodynamlcs, Inc. (1980a) Spartan
Btodynamics, Inc. (1980c) Fischer 341
Blodynamlcs, Inc. (1980b) Spartan
Ingestlon 1n 100 ppm
drinking water
45 wk.t
19-20 mo.
Ingestlon 1n 10-100 ppm 23-26 mo.
drinking water
gavage
10 mg/kg/day 19-20 mo.
Brain (astrocytomas)
Ear canal (Zymbal gland)
Stomach (nonglandular
portion)
Mammary gland (females only)
Tongue
Pituitary gland
Pancreas (males only)
Uterus (females only)
Brain (astrocytomas)
Ear canal (Zymbal gland)
Brain (astrocytomas)
Ear canal (Zymbal gland)
Stomach
Brain (astrocytomas)
Ear canal (Zymbal gland)
Stomach
Mammary gland (females only)
Brain (astrocytomas)
Ear canal (Zymbal gland)
Stomach
Mammary gland (females only)
(continued on the following page)
*Not all the tumors Indicated for each study were observed at the lower dose.
tThis was a three-generation reproductive study; the tumors were observed in the second generation.
-------
TABLE 13-37. (continued)
oo
i
en
CTl
Reference
Maltonl et al. (1977)
Quast et al. (1980b)
Route of Treatment
Strain of Rat Administration Dose Level* Schedule
Sprague-Dawley Inhalation 5-40 ppm 4 hr/day
5 days/wk.
for 12 mo.
Sprague-Dawley Inhalation 20-80 ppm 6 hr/day
5 days/wk
for 24 mo.
Site of Tumors
Stomach (males only)
Mammary gland (females only)
Skin (females only)
Central nervous system
Ear canal (Zymbal gland)
Gastrointestinal tract
(males only); mammary gland
(females only); tongue
(males only)
"Not all the tumors IndTcatea for eacn study were observed at the 1 ower dose.
tThls was a three-generation reproductive study; the tumors were observed in the second generation.
-------
rats was further confirmed in four other studies: a three-generation repro-
duction study performed at Litton-Bionetics by Beliles et al. (1980); three
studies by Biodynamics, Inc. (1980a, b, c) in which acrylonitrile was admin-
istered in drinking water and via gastric intubation; and an inhalation study
by Quast et al. (1980b).
A second inhalation study by Maltoni et al. (1977) exposed rats to atmos-
phere containing 5, 10, 20, and 40 ppm acrylonitrile 4 hours per day, 5 days
per week, for 12 months. Marginal increases in tumors of the mammary gland in
females and the forestomach in males were observed, although the sensitivity
of this test was limited by the relatively low dose levels and the short
duration of exposure.
Ten epidemiologic studies of acrylonitrile and cancer have been reported:
five published [Monson (1981), O'Berg (1980), Thiess et al. (1980), Werner and
Carter (1981), Delzell and Monson (1982)] and five unpublished [Gaffey and
Strauss (1981), Herman (1981), Kiesselbach et al. (1980), Stallard (1982), and
Zack (1980)]. These are summarized in Table 13-38. Six of these studies present
no evidence of carcinogenic risk from exposure to acrylonitrile. However, all
suffer from problems in the design or methodology, i.e., small cohort size,
insufficient characterization of exposure, short follow-up, and relatively
youthful cohort. Because of these problems none of these studies can be cited
as adequate evidence that acrylonitrile is not carcinogenic.
Data presented in the remaining four epidemiologic studies consistently
demonstrate a statistically significant risk of lung cancer in various sub-
groups of the populations studied. All four have problems with the method-
ology, definition, and/or size of the population, whether or not exposure to
other carcinogens occurred, and short follow-up intervals. In three of the
four studies [Delzell and Monson (1982), Thiess et al. (1981), Werner and
Carter (1981)], the problems were sufficient to cast doubt on the finding of
13-167
-------
TABLE 13-38. EPIDEMIOLOGIC STUDIES REVIEWED IN ACRYLONITRILE RISK ASSESSMENT
Study
Results
Observed/Expected Highlights and Deficiencies
O'Berg (1980)
cr>
oo
Delzell and Monson
(1982)
Thiess et al.
(1980)
Werner and Carter
(1981)
Monson (1978)
Zack (1980,
unpubli shed)
Gaffey and Strauss
(1981, unpublished)
Positive
lung cancer
Positive
lung cancer
Positive
lung cancer
Lymph cancer
Positive
lung cancer
stomach cancer
Inconclusive
Inconclusive
Inconclusive
5/1.4 (P <0.05)
4/1.4 (P <0.05)
9/4.4 (P <0.05)
4/1.4 (P <0.05)
3/0.7 (P <0.05)
5/1.9 (P <0.05)
Significant excess occurred in group
with highest exposure followed for the
longest time. Excess remains after
adjustment for smoking.
Potential exposure to other carcinogens,
Incomplete follow-up on foreign workers,
Mixed exposures to beta-naphthalamine,
polycyclic hydrocarbons, and vinyl
chloride.
Relatively short follow-up. Combined
Population from 6 polymerization
factories.
Potential exposure to other carcinogens.
Small cohort size. Short follow-up on
most members of cohort.
Small cohort size. Includes large numbers
of minimally or unexposed employees.
(continued on the following page)
-------
oo
I
TABLE 13-38. (continued)
Study
Results
Observed/Expected Highlights and Deficiencies
Kiesselbach et al.
(1981, unpublished)
Herman (1981,
unpublished)
Stallard (1982,
unpublished)
Inconclusive
Inconclusive
Inconclusive
Markedly reduced mortality due to 1)
healthy worker effect, and 2) possible
selection bias in determination of study
population. Lack of critical editorial
review.
Relatively short duration of employment.
Insufficient follow-up.
Small cohort. Short follow-up period.
VO
-------
significantly elevated risks of lung cancer reported in each study. In the
fourth study by O'Berg, the proglems were insufficient to obscure the signif-
icant finding of lung cancer. After adjusting for latent factors and evaluat-
ing the contribution due to smoking, the finding of a statistically signifi-
cantly elevated risk of lung cancer remained. Thus, one study appears adequate
and three are suggestive, while the remaining six are inadequate to address
the issue of a risk of lung cancer.
In addition to lung cancer, two other findings of concern are the sig-
nificantly elevated risk of lymph system cancer found in the Thiess et al.
study (4 observed versus 1.38 expected, P < 0.05) and the significantly ele-
vated risk of stomach cancer found in the Werner and Carter study (5 observed
versus 1.9 expected, P < 0.05). These findings provide additional suggestive
evidence of the carcinogenicity of acrylonitrile.
This level of animal evidence would be regarded as "sufficient" evidence
of carcinogenicity according to the International Agency for Research on
Cancer (IARC) classification scheme. The human evidence for the carcino-
genicity of acrylonitrile would be regarded as somewhere between "sufficient"
and "limited," using the IARC classification. Therefore, in combining the
human and animal evidence, acrylonitrile would be .placed in group 2A, which
IARC characterizes as "probably carcinogenic in humans, where the evidence for
human carcinogenicity is almost sufficient."
Quantitative Assessment
Three unit risk estimates for air are calculated; one based on a human
occupational study (O'Berg, 1980a, b) and two based on rat cancer bioassays
(Quast et al., 1980a,b). The upper-bound lifetime risk of cancer associated
with a lifetime inhalation exposure of 1 ug/m is 6.8 x 10 from the human
study and 1.5 x 10 from the rat inhalation study. The value based on the
13-170
-------
-4 -5
rat drinking water study is 1.2 x 10 (or 2 x 10 if the equivalent human
dose is assumed to be mg/kg/day rather than surface area) but this study is
less reliable because of the inappropriate route of exposure.
The estimate based on the human study is uncertain because of the relati-
vely weak documentation of the available exposure estimates of the acrylonitrile
workers. The air concentration had not been measured when the workers experienced
their heaviest exposure and was estimated 12 years after the end of the exposure
period. However, in spite of these difficulties, the estimates are consistent
with those of the animal studies.
The upper-bound risk estimate for 1 ug/1 of acrylonitrile in drinking
water is 1.2 x 10 , based on the Quast et al. (1980a) drinking water study in
rats.
These values are regarded as rough but plausible estimates of an upper-bound
of risk; i.e. it is not likely that the true risk would be much more than
these values, but it could very well be considerably lower. In using these
quantitative estimates, the assumptions made in their derivation and the
limitations of their interpretation should be kept clearly in mind. This is
discussed on pages 13-136 through 13-151 of the document.
13.5.5 CONCLUSIONS
There is evidence that acrylonitrile is a human carcinogen. This conclu-
sion is based on: 1) findings of three positive drinking water rat bioassays
and one positive rat gastric intubation study; 2) statistically significant
positive findings of respiratory cancer in four epidemiologic studies; 3) the
positive mutagenic evidence in bacteria and sister chromated exchange tests;
4) jjn vitro evidence of interaction of acrylonitrile and/or its metabolites
with DNA; and 5) acrylonitrile's structural similarity to vinyl chloride, a
known animal and human carcinogen.
13-171
-------
The carcinogenic potency of acrylonitrile is in the third quartile among
54 suspected carcinogens evaluated by the Carcinogen Assessment Group.
Using the International Agency for Research on Cancer (IARC) classifi-
cation scheme, this level of evidence in animals and humans would be considered
sufficient for concluding that acrylonitrile is likely to be a human carcinogen
with rank of 2A.
13-172
-------
REFERENCES
Aarato, S.E., and Bittera, E. (1972). Determination of acrylonitrile in the air
in the presence of dimethylformamide and ammonia. Hunkavedelen 18:50-51.
(In Hung.) Taken from: Chem. Abst. l£:57280b, 1973.
Abreu, M., and Ahmed, A.E. (1980). Metabolism of acrylonitrile to cyanide: In
vitro studies. In: Abstracts of Papers: Society of Toxicology. 19th
Annual Meeting, Washington, DC, March 9-13, 1980. Abstract No. 393,
p. A131.
A.D. Little, Inc. (1971). A Model Economic and Safety Analysis of the Trans-
portation of Hazardous Substances in Bulk, Final Report (NTIS COM-71-11271)
261 pp. (Cited in Miller and Villaume, 1978.)
Ahmed, A.E., and Patel, K. (1979). Pharmacokinetics, distribution and binding of
l-1lC-acrylonitrile in rats. Pharmacologist 21( 3):A131. (Abstract.)
Ahmed, A.E., and Abreu, M.E. (1982). Microsomal metabolism of acrylonitrile in
liver and brain. Adv. Exp. Med. Biol. 1366:1229-1238. Taken from: Chem.
Abst. 57:l8598p.
Ahmed, A.E.; Farooqui, M.Y.H.; Upreti, R.K.; and El-Shabrawy, 0. (1982).
Distribution and covalent interactions of [1-11C] acrylonitrile in the rat.
Toxicology 23^:159-175. Taken from: Chem. Abst. 27:67136t.
AIHA (American Industrial Hygiene Association) (1970). Acrylonitrile. Amer.
Ind. Hyg. Assoc. J., July-August 1970, pp. 529-531.
ASTM (American Society for Testing and Materials) (1980). Nitriles in aqueous
solution by gas-liquid chromatography, 1980. Annual Book of ASTM
Standards, Part 31, Water, Method No. ASTM D3371-79, ASTM, Philadelphia,
PA.
Axel son, 0. 1978. Aspects on confounding In occupational health and
ep1den1ology. Scan. J. Work Environ, and Health 4:85-89.
Babanov, G.P. (I960). Problems of industrial hygiene in the production of
synthetic butadiene-acrylonitrile rubbers. Gigiena Truda i Professional.
Zabolevaniya JK12):7-12. (In Russ.) Taken from: Chem. Abst. 55:23883a,
1961.
Babina, M.D. (1979). Gas-chromatographic determination of trace impurities
released into the air in the production of some styrene copolymers. Gig.
Tr. Prof. Zabal. 2:55-7. (In Russ.) Taken from: Chem. Abst. 90;19l833q.
1979.
Balak, J.; Cavojcova, E.; and Polievka, M. (1977). Problems of acrylonitrile
analysis in cyanoethylene reaction mixtures using gas chromatography.
Petrochemia lJ(l-2):37-16. (InSlo.) Taken from: Chem. Abst. 88:1l531v,
1978.
Bandt, H.J. (1953). [Acrylonitrile as an effluent.] Z. Fisch. 2:157-161. (In
Ger.)
R-l
-------
Barrett, V.J. (1971). Sampling and Analysis of Four Organic Compounds Using
Solid Sorbents. Southern Research Institute Report, SORI-EAS-71-301, U.S.
Department of Health, Education and Welfare. (Cited in Going et al., 1978.)
Baxter, R.A. (1979)* Evaluation and control of Industrial exposure to acrylo-
nltrile. Ann. Occup. Hyg. 22:129-135.
Baxter, R.A. (1980). Written communication from R.A. Baxter, Monsanto Europe,
Brussels, Belgium, to W.R. Gaffey, Monsanto World Headquarters, St. Louis,
MO, dated June 21, 1980, including minutes of the meeting held to Review
Recently Completed Studies on the Epidemiology of Acrylonitrlle, held in
Brussels, April 23 and 21, 1980.
Beaumont, G.P., and Garrido, C.H. (1979). Respirator cartridge test system and
test-results for benzene and acrylonitrile. Aoer. Ind. Hyg. Assoc. J.
JKK 10): 883-87.
Bellies, R.P.; Paulin, R.J.; Makris, N.G.; and Weir, R.J. (1980). Three-
Generation Reproductive Study of Rats Receiving Acrylonitrile In Drinking
Water. Prepared by Litton Bionetlcs, Inc., Kensington, MD, under LBI
Project No. 2660 for the Chemical Manufacturers Association, Washington,
DC. Available from: CMA, Washington, DC.
Benes, V., and Cerna, V. (1959). [Acrylonitrile: Acute toxicity and mechanism
of action.] J. Hyg. Epidemlol. Mlcrobiol. Immunol. 3_: 106-116. (In Ger.)
Benes, V., and Sram, R. (1969). Mutagenic activity of some pesticides in
Drosophila melanogaster. Ind. Med. Surg. 38(12) :112-111.
Berck, B. (I960). Fumlgant residues: Retention of acrylonitrile and carbon
tetrachloride by shelled walnuts fumigated with Acrylon. J. Agr. Food Chem.
8(2): 128-31.
Berck, B. (1962). Polarographlc determination of methyl bromide, ethylene
dibromide, acrylonitrile, chloropicrln, and carbon tetrachloride in air.
J. Agr. Food Chem. 10:158-62.
Berck, B. (1965). Determination of fumlgant gases by gas chromatography.
J. Agr. Food Chem. 113( D;373-77.
Berck, B. (1975). Analysis of fumlgants and fumigant residues. J. Chromatogr.
Sci. 13.(6): 256-67. .
Bio/dynamics Inc. (1980a). A Twenty-Four Month Oral Toxlcity/Carcinogenicity
Study of Acrylonitrile Administered to Spartan Rats in the Drinking Water.
Vols. 1 and 2 of Final Report. Prepared by Bio/dynamics Inc., Division of
Biology and Safety Evaluation, East Millstone, NJ, under Project No. 77-
17U5 for Monsanto Company, St. Louis, MO.
Bio/dynamics Inc. (1980b). A Twenty-Four Month Oral Toxlcity/Carcinogenicity
Study of Acrylonitrlle Administered by Intubation to Spartan Rats. Vols. 1
and 2 of Final Report. Prepared by Bio/dynamics Inc., Division of Biology
and Safety Evaluation, East Millstone, NJ, under Project No. 77-1716 for
Monsanto Company, St. Louis, MO.
R-2
-------
Bio/dynamics Inc. (1980c). A Twenty-Four Month Oral Toxicity/Carclnogenicity
Study of Acrylonitrile Administered In the Drinking Water to Fischer 311
Rats, Vols. 1-1 of Final Report. Prepared by Bio/dynamics Inc., Division of
Biology and Safety Evaluation, East Millstone, NJ, under Project No.
77-1744 (BDN-77-27) for Monsanto Company, St. Louis, MO.
Bocek, K. (1976). Relationships among activity coefficients, partition coeffi-
cients and solubilities. Experientia, Suppl. 23:231-10.
Bogaczek, J., and Jaworski, M. (1970). Continuous polarographic determination
of acrylonltrile. Fresenius1 Z. Anal. Chem. 251(3): 171-76. (InGer.) Taken
from: Chem. Abst. £l:91406u, 1970.
Bond, E.J. (1963). The action of fumigants on insects. IV. The effects of
oxygen on the toxicity of fumigants to insects. Can. J. Biochem. Physiol.
11:993-1004.
Bond, E.J., and Buckland, C.T. (1976). Control of Insects with fumigants at low
temperatures: Toxicity of mixtures of methyl bromide and acrylonitrile to 3
species of insects. J. Econ. Entomol. 69(6);72527»
Bond, E.J., and Buckland, C.T. (1978). Control of Insects with fumigants at low
temperatures: Toxicity of fumigants in atmospheres of carbon dioxide.
J. Econ. Entomol. 71(2);307-9*
Boxer, G.E., and Rickards, J.C. (1952). Studies on the metabolism of the carbon
of cyanide and thlocyanate. Arch. Biochem. 39:7-26.
Boyland, E., and Chasseaud, L.F. (1967). Enzyme-catalyzed conjugations of
glutathione with unsaturated compounds. Biochem. J. 101:95-102.
Braun, D., and Vorendohre, G. (1963)• Thin-layer chromatography of some
unsaturated polymerizable compounds. Z. Anal. Chem. 199(1);37-11. (In
Ger.) Taken from: Chem. Abst. 60:l253d, 1964.
Brieger, H.; Rleders, F.; and Modes, W.A. (1952). Acrylonitrile: Spectophoto-
metric determination, acute toxicity, and mechanism of action. Ind. Hyg.
Occup. Med. 6:128-40.
Brown, M.E.; Breder, C.V.; and McNeal, T.P. (1978). Gas-solid chromatographic
procedures for determining acrylonitrile monomer in acrylonitrile-
containlng polymers and food simulating solvents. J. Assoc. Off. Anal.
Chem. 6K 6): 1383-388.
Burg, S.P., and Burg, E.A. (1967). Molecular requirements for the biological
activity of ethylene. Plant Physiol. 12_:111-152.
Burkhart, P.; Bartels, D.; and Hoyme, H. (1961). Analytical chemistry of poly-
acrylonltrlle fibers. I. Analytical studies on the separation of vinyl
monomer mixtures. Faaerforsch. Textiltech. l£:581-87.. (In Ger.) Taken
from: Chem. Abst. 5_7_:1107i, 1962.
R-3
-------
Campbell, D.N., and Moore, R.H. (1979). The quantitative determination of
acrylonitrile, acrolein, acetonitrile and acetone in workplace air.
Amer. Ind. Hyg. Assoc. J. 40(10):904-09.
Carcinogen Assessment Group, U.S. Environmental Protection Agency. 1978.
Carcinogen Assessment Group's final report on population risk to ambient
benzene exposures. 41 pp.
Carpenter, C.P.; Smyth, H.F., Jr.; and Pozzani, U.C. (1949). The assay of
acute vapor toxicity, and the grading and interpretation of results on
ninety-six chemical compounds. J. Ind. Hyg. Toxicol. 3_1:343-346.
Chaigneau, M., and LeMoan, G. (1974). Pyrolysis of plastic materials. VIII.
Polyacrylonitrile and copolymers. Ann. Pharm. Fr. 3^(9-10):485-90. (In
Fr.) Taken from: Chem. Abst. 82:73614f, 1975.
Chao, L.-S., and Ch'en, S.-S. (1966). Polarographic determination of acrylo-
nitrile-vinyl acetate copolymer systems. Hua Hsuen Tung Pao (l):47-9.
(In Ch.) Taken from: Chem. Abst. 65:10669g, 1966.
Chekhovskaya, E.V.; Gorban, N.S.; Yakovleva, L.I.; Nazarenko, G.P.; and Necha-
eva, A.V. (1966). Data for experimental studies of toxicity of wash
waters from acrylonitrile production. vodosnabzh. Kanaliz. Gidrotekh.
Sooruzh. Mezhued. Resp. Nauch. Sb. No. l_:83-88. (In Russ.) Taken from:
Chem. Abst. 66:88487k, 1967.
Chemical Marketing Reporter (1980). Chemical profile: Acrylonitrile. Febru-
ary 4, 1980, p. 9.
Cherry, A. B.; Gabaccia, A.J.; and Senn, H.W. (1956). The assimilation be-
havior of certain toxic organic compounds in natural water. Sewage Ind.
Wastes 28(9):1137-46.
Chiou, C.T.; Freed, V.H.; Schmedding, D.W.; and Kohnert, R.L. (1977). Parti-
tion coefficient and bioaccumulation of selected organic chemicals.
Environ. Sci. Technol. ll.:475-478.
Chopra, S.K.; Vishnoi, S.C.; Kapoor, V.B.; and Gupta, P.L. (1978). A gas
chromatographic method for the estimation of residual monomers in sty-
reneacrylonitrile copolymer. Res. Ind. 23(1):27-30. Taken from: Chem.
Abst. 89-.164097a, 1978.
Chudy, J.C., and Crosby, N.T.(1977). Some observations on the determination
of monomer residues in foods. Food Cosmet. Toxicol. 15(6):S47-51.
Conner, T.H.; Forti, G.C.; Fitra, P.; and Legator, M.S. (1979). Bile as a
source of mutagenic metabolites produced in vivo and detected by Sal-
monella typhimurium. Environ. Mutag. 1^:269-276.
Covic-Horvat, S.; Vrzina, J.; and Sateva, M. (1970). Determination of acrylo-
nitrile in waste waters. Nafta 21_(2) :79-81. (In Croatian.) Taken from:
Chem. Abst. 73:28605g, 1970.
Cox, C.R. 1972. Regression model and life tables. J. Roy. Stat. Soc. B
34:187-220.
R-4
-------
Crume, R.V. (1982). Summary of acrylonitrile producers, emissions, and plant
capacities. Referenced memo available from the Chemicals and Petroleum
Branch, Emission Standards and Engineering Division, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina. October 29,
1982.
Crump, K.S., H.A. Guess, and L.L. Deal. 1977. Confidence intervals and test
of hypotheses concerning dose-response relations inferred from animal
carcinogenicity data. Biometrics 33:437-451.
Crump, K.S. 1979. Dose-response problems in carcinogenesis. Biometrics
35:157-167.
Crump, K.S., and W.W. Watson. 1979. GLOBAL 79. A fortran program to extra-
polate dichotomous animal carcinogenicity data to low dose. Natl. Inst.
of Environ. Health Science. Contract No. l-ES-2123.
Dahle, L.,- Hill, and Holman, R. (1962). The thiobarbituric acid reaction and
the autoxidations of polyunsaturated fatty methyl esters. Arch. Biochem.
Biophys. 98:253.
Dahm, K.J. (1977). Indentification of metabolites of acrylonitrile. Testi-
mony before FDA. (Cited in Miller and Villaume, 1978.)
Daues, G.W., and Hamner, W.F. (1957). Determination of small amounts of
acrylonitrile in aqueous industrial streams. Anal. Chem. 29:1035-37.
Daugherty, F.M., Jr., and Garrett, J.T. (1951). Toxicity levels of hydro-
cyanic acid and some industrial by-products. Texas J. Sci. 3:391-396.
Davis, J.H.; Davies, J.E.; Raffonelli, A.; and Reich, G. (1973). The investi-
gation of fatal acrylonitrile intoxications. In-. Pesticides and the
Environment: A Continuing Controversy. Intercontinental Medical Book
Corp., Vol. 2, pp. 547-556.
Delzell, E., and R.R. Monson. 1982. Mortality among rubber workers. VI. Men
with exposure to acrylonitrile. J. Occup. Med. 24(10):767-769.
DeMeester, C.; Duverger-VanBogaert, M. ,- Lambotte-Vandepaer, M. ; Roberfroid,
M.; Poncelet, F.; and Mercier, M. (1979). Liver extract mediated muta-
genicity of acrylonitrile. Toxicology 13(1):7-15.
DeMeester, C.; Poncelet, F.; Roberfroid, M.; and Mercier, M. (1978). Muta-
genicity of acrylonitrile. Toxicology 11:19-27.
Deur-Siftar, D., and Svob, V. (1976). Procedure for gas chromatographic
determination of acrylonitrile as a residual monomer in polymerization
process waste waters and in polymers. Kern. Ind. 2_5( 12)-.703-08. (In
Croatian.) Taken from: Chem. Abst. 87:58072n, 1977.
DiGeronimo, M.J., and Antoine, A.D. (1976). Metabolism of acetonitrile and
propionitrile by Nocardia rhodochrous LL100-21. Appl. Environ. Micro-
biol. 31(6):900-906.
R-5
-------
Dilling, W.L. (1,977). Interphase transfer processes. II. Evaporation rates
of chloromethanes, ethanes, ethylenes, propanes, and propylenes from
dilute aqueous solutions. Comparison with theoretical predictions.
Environ. Sci. Technol. n(4);405-09.
DiLorenzo, A., and Russo, G. (1969). Gas chromatographic analysis of mixtures
of acrylonitrile with acetonitrile and water. Chim. Ind. 51^(2): 170-71.
(In Ital.) Taken from: Chem. Abst. 70:120896p, 1969.
Dinu, V. (1975a). Activity of glutathione peroxidase and catalase and the
concentration of lipid peroxides in acute intoxication with acryloni-
trile. Rev. Roum. Biochim. 12(1):11-14.
Dinu, V. (1975b). Intracellular thiol concentration in acrylonitrile intoxi-
cation. Rev. Roum. Biochim. 12(3);155-58.
DiPasquale, G.; Dilorio, G.; Capaccioli, T.; Gagliardi, F.; and Verga, G.R.
(1978). Gas chromatographic head-space determination of residual acrylo-
nitrile in acrylonitrile-butadiene-styrene resins and migration into a
simulated fatty foodstuffs liquid. J. Chromatogr. 160(1):133-40.
Doll, R., and A.B. Hill. 1952. A study of the aetiology of carcinoma of the
lung. Br. Med. J. 2:1271-1286.
Doll, R. 1971. Weibull distribution of cancer. Implications for models of
carcinogenesis. J. Roy. Statistical So. A13:133-166.
Dripps, Robert D., J.E. Eckenhoff, and L.D. Vandam. 1977. Introduction to
anesthesia, the principles of safe practice. 5th Ed. W.B. Saunders
Company, Phil. Pa. pp. 121-123.
Dudley, H.C., and Neal, P.A. (1942). Toxicology of acrylonitrile. I. A
study of the acute toxicity. J. Ind. Hyg. Toxicol. 24:27-36.
Dudley, H.C.; Sweeney, T.R.; and Miller, J.W. (1942). Toxicology of acrylo-
nitrile (vinyl cyanide). II. Studies of effects of daily inhalation.
J. Ind. Hyg. Toxicol. 24:255-258.
Durkin, P.; Sugatt, R. ; Colman, J.; and Ryan, E. (1981). Ambient Water Qual-
ity Criteria Document for Ammonia. Prepared for the Environmental Cri-
teria and Assessment Office, U.S. Environmental Protection Agency, Cin-
cinnati, OH.
Ehrenberg, L. and S. Ostermar-Golkar. 1980. Alkylation of marcomolecules for
detecting mutagenic agents. Terato. Carcinog. and Mutag. 1:105-127.
Ehrenberg, L. 1979. Risk assessment of ethylene oxide and other compounds.
Banbury Report 1: Assessing Chemical Mutagenesis-The Risk to Humans.
(Eds., V.K. McElheny and S. Abrahamson), Cold Spring Harbor Laboratory.
FASEB. 1974. Biological Data Books. 2nd ed. Vol. III. Edited by Philip L.
Altroan and Dorothy S. DHtmen. Federation of American Societies for
Experimental Biology. Bethesda, MD. Library of Congress No. 72-87738.
R-6
-------
Fasitta, V.A., and Licciardello, G. (1977). Determination of acetonitrile and
acrylonitrile in the air by gas chronatography. Ann. 1st. Super. Sanita.
13(1-2) :245-48. (In Ital.) Taken from-. Chem. Abst. 89:11151p, 1978.
ITassett, D.W. (1963). Cyanides and nitriles. In: Toxicology. Passett, D.W.,
and Irish, P.O., editors. Vol. 2 of Industrial Hygiene and Toxicology, 2nd
ed. Mew York: John Viley and Sons, Inc., pp. 2009*2012.
Federal Register (197Ba). Emergency temporary standard for occupational expo-
sure to acrylonitrile (vinyl cyanide): Notice of bearing.
11) :2586-2607.
Federal Register (1978b). Occupational exposure to acrylonitrile (vinyl
cyanide) . l£( 192) :U5762-U58l9.
Federal Register (1979). Method 603. acroleln and acrylonitrile.
jyj.{233):69l79-81.
Finkel, J.M.; James, R.H.; and Miller, B.C. (1979). Residual Monomers in Acrylic
and Modacrylic Fibers and Fabrics. Draft Final Report. Prepared by
Southern Research Institute, Birmingham, AL, under Contract No. 68-01-4716.
U.S. Environmental Protection Agency, Office of Toxic Substances,
Washington, DC.
Flood, M. (1980). Memorandum from M. Flood, Div. of Chemistry and Physics., Food
and Drug Administration, Department of Health, Education, and Welfare,
Washington, DC, to R.M. Bruce, U.S. Environmental Protection Agency,
Research Triangle Park, NC, dated July 17, 1980.
Freeman, R.A.; Schroy, J.M.; Klieve, J.R.; and Archer, S.A. (1980). Experimental
studies on the rate of air stripping of hazardous chemicals from waste
treatment systems. APCA Meeting, Montreal, Canada, June 1980.
Freshour, K.L.; Ramsey, J.C.; and Braun, W.H. (1980). Acrylonitrile: Pharmaco-
kinetic profile in rat plasma after single oral or intravenous administra-
tion. In: Abstracts of Papers: Society of Toxicology. 19th Annual
Meeting, Washington, DC, March 9-13, 1980. Abst. No. 179, p. A60.
Caffey, W.R.. and M.E. Strauss. 1981. A mortality study of workers potentially
exposed to acrylonltHle during startup. Monsanto Decatur plant.
Unpublished report. *
Gagnon, Y.T. , and Posner, J.C. (1979). Recovery of acrylonitrile from charcoal
tubes at low levels. Amer. Ind. Hyg. Assoc. J. JKX 10) :923-5.
Gale, D.M. (1970). Mechanistic aspects of the photochemistry of unsaturated
nitriles. J. Org. Chem. 3JK H) : 970-72.
R-7
-------
Orote, A. A.; Kim, V.S.; and Kupel, E.R. (1978). Establishing • protocol from
laboratory studies to be used in field sampling operations. Amer. Ind. Hyg.
Assoc. J. ££( 11): 680-6)4.
Guengerich, F.P.; Oeiger, L.E.; Hogy, L.L.; and Wright, P.L. (1981). In vitro
metabolism of acrylonitrile to 2-cyanoethylcne oxide, reaction with gluta-
thione, and Irreversible binding to proteins and nucleic acida. Cancer Res.
Guerin, M.R.; Olerich, G.; and Norton, A.D. (197*0. Routine gas chromatographic
component profiling of cigarette smoke for the identification of biologi-
cally significant constituents. J. Chromatogr. Sci. 1_2( 7) : 385-91.
Gunther, F., and Blinn, R. (1955). Analysis of Insecticides and Acarlcldes. New
York: Interscienoe Publishers.
Gut, I.; Nerudova, J.; Kopecky, J.; and Holecek, V. (1975). Acrylonitrile
biotransformation in rats, nice, and Chinese hamsters as influenced by the
route of administration and by phenobarbital, SKF 525 -A, cystelne, dimer-
caprol, or thlosulfate. Arch. Toxicol. 33(2) : 151-161.
Guyot, A.; Bert, M.; Hamoudi, A.; McNeill, I.; and Grassie, N. (1978). Pyrolysis
of acrylonitrile-methacrylic and copolymers and derivatives. Eur. Polymer
J. 1^(2): 101-07.
Hall, M.E., and Stevens, J.W., Jr. (1977). Spectrophotometric determination of
acrylonitrile. Anal. Chem. «<9(U) :2277-80.
Harmon, E.G. 1975. Tobacco, 1n: V.F. Fraumenl (ed.). Persons at high risk of
cancer: an approach to cancer etiology and control. Academic Press, New
York, San Francisco, and London.
Hartshorn, J.H. (1975). Detection of acrylonitrile In FDA "food simulating
solvents." In: Abstracts of Papers of the American Chemical Society
1975(169): 121.
Baseman, J.K., and Hoel, D.G. (197^) . Table of Gehan's generalized Wilcoxon test
with fixed point sensoring. J. Statist. Comput. Simul. 3_:117.
Hashimoto, K., and Kanai, R. (1965). Studies on the toxicology of acrylonitrile:
Metabolism, mode of action, and therapy. Ind. Health (Kawasaki, Japan)
Henderson, C.; Pickering, Q.H.; and Lemke, A.E. (1961). The effect of some
organic cyanides (nitriles) on fish. In: Proceedings of the 15th
Industrial Waste Conference, Purdue University J<5( 2) : 120-130.
Herman, D.R. 1981. Cohort mortality study of the Scotts Bluff/Baton Rouge
Uniroyal Plant. Unpublished report.
Hoffman, P.; Kleine, D.; and Franzen, E. (1976). [On the proof of acrylonitrile
exposure: Detoxification of acrylonitrile by coupling with D-glucuronic
acid.] A Gesampte Hyg. Ihre Grenzgeb. 22( 5) : 310-312. (In Ger.)
Hollingseed, D.W. (1978). Letter from D.W. Hollingsed of Vitron to F.
Letkiewicz of OTS/EPA dated 5/3/78; personal communication from D.W.
Hollingsed to P. Hilgard of OTS/EPA on 5/10/78. (Cited in Miller and
Villaurae, 1978.)
Hosaka, S., and Wakamatsu, S. (1968). Photodimerization of acrylonitrile.
Tetrahedron Lett. 2:219-220.
R-8
-------
Hovious, J.C.; Haggy, G.T.; and Conway, R.A. (1973). Identification and control
of petrochemical pollutants inhibitory to anaerobic processes. Available
from: National Technical Informa tion Service, Springfield, ?A (RT1S PB-
222-287). (Cited in Miller and Villaume, 1976.)
Hughes, T.W., and Horn, D.A. (1977). Source Assessment: Acrylonitrile Manufac-
ture (Air Emissions), Publication No. EPA-600/2-77-107J. Prepared by
Monsanto Research Co., Dayton, OH, under Contract No. 68-02-1871. U.S.
Environmental Protection Agency, Office of Research and Development, Indus-
trial Environmental Research Laboratory, Research Triangle Park, NC.
Available from: National Technical Information Service, Springfield, VA
(NTIS PB-271-969).
IAFC (International Agency for Research on Cancer) (1979). Acrylonitrile,
acrylic and modacrylic fibers, and acrylonitrile-butadiene-styrene and
styrene-acrylonitrile copolymers. In: Some Monomers, Plastics and Synthe-
tic Elastomers, and Acrolein. IARC Monographs on the Evaluation of Carcino-
genic Risk of Chemicals to Man. Lyon, France: World Health Organization,
IAPC, Vol. 19, PP. 73-113.
International Commission on Radiological Protection. 1977. Recommendation of
the International Commission on Radiological Protection, Pub. No. 26.
adopted Jan. 17, 1977. Per gammon press, Oxford, England.
Ivanenko, C.S., and Lukashevskaya , A.V. (1976). Chromatographic method of
determining bound acrylonitrile in butadiene-nitrile rubbers. Metody Anal.
Kontrolya Proizvod. Khlm. Prom-sti. 2:19-20. (InRuss.) Taken from: Chem.
Abst. 8j4_:l8m03», 1976.
Jackson, S., and Brown, V.M, (1970). Effect of toxic wastes on treatment pro-
cesses and watercourses. Water Pollut. Contr. 6£( 3) :292-313. (Cited in
Miller and Villaume, 1978.)
Jacobs, H.W., and Syrjala, R.H. (1978). The use of infrared analyzers for
monitoring acrylonitrile. Amer. Ind. Hyg. Assoc. J. 3JK 2) ; 161-65.
Jaeger, R.J. (1979). Time-related variation of non-protein sulfhydryl concen-
trations in rat tissues and human blood. Int. Arch. Occup. Environ. Health
Joshi, S.B., Northrop Services Corp. (1977). Letter to J. Cervello, U.S. EPA
regarding photochemical reactivity of acrylonitrile, dated July 10, 1977.
(Cited in Suta, 1979.)
Judson, C.L.; Hokama, Y.; and Bray, A.D. (1962). The effects of various chemi-
cals on eggs of the yellow-fever mosquito, Aedes acgypti. J. Econ.
Entomol. 5_5( 5) : 805-07.
Junke, I., and Ludemann, D. (1978). (Results of the study of 200 chemical
compounds on acute fish toxicity using the Golden Orfe test.] Z. Wasser
Abwasser Forsch. IK 5) : 161-61. (In Ger.)
Kaiser, E.R., and Bergman, F. (1973). Combustion Characteristics of Lopac Con-
tainers. In: Proceedings of the Symposium on Environmental Impact of
Nitrile Barrier Containers, Hartford, CT, July 19, 1973. Sponsored by and
available from Monsanto Co., St Louis, MO.
R-9
-------
Kankaanpaa, J.; Elovaara, E.; Hemminki, K.; and Vainio, H. (1979). Embryo-
toxicity of acrolein, acrylonitrile and acrylamide in developing chick
embryos. Toxicol. Lett. 4(2):93-96.
Karaenev, S.S.; Kostov, G.; Milina, R.; and Mikhailov, M.Kh. (1974). Deter-
mination of the composition of ternary copolymers by IR spectroscopy and
pyrolytic gas chromatography. Vysokomol. Soedin., Ser. A., 1£(9): 2162-
165. (In Russ.) Taken from-. Chem. Abst. 82-.32109b, 1975.
Kato, A., and Yamamura, K. (1976). Treating waste water containing nitriles
and cyanides. U.S. Patent 3,940,332. February 24, 1976. Taken from:
Chem. Abst. 85:67783f, 1976.
Kearney, A.T. (1978). Assessment of Acrylonitrile Contained in Consumer Pro-
ducts. Final Report. Prepared by A.T. Kearney, Inc., Washington, DC,
under Contract No. CPSC-C77-0009. U.S. Consumer Product Safety Commis-
sion, Washington, DC.
Keresztesy, J.; Taylor, T.J.; Katz, A.; Haddock, J.T.; Byer, W.L.; and Mai-
seed, R.T. (1977). Acrylonitrile. Prepared by Auerbach Associates,
Inc., for the U.S. Consumer Product Safety Commission, Washington, DC.
Available from: National Technical Information Service, Springfield, VA
(NTIS PB-280-478).
Kier, L.D. 1982. Ames/Salmonella mutagenicity assay of acrylonitrile. Mon-
santo Company, St. Louis. Report No: MSL-2063, Job/Project No: 800282-
ML-80-441.
Kier, L.D. 1982. Ames/Salmonella mutagenicity assay of acrylonitrile epoxide.
Monsanto Company, St. Louis. Report No: MSL-2064, Job/Project No:
800283-ML-80-442.
Kiesselbach, N., U. Korallis, H.J. Lange, A. Neiss, and T. Zwingers. 1980.
Bayer-ACN-Studie 1977. Zentralblatt fur arbeitsmedizin, arbeitsschutz,
prophylaxe und ergonome, Band 7.
Kihlman, B.A. (1961). Cytological effects of phenylnitrosaraines. I. The
production of structural chromosome changes in the presence of light and
acridine orange. Radiat. Bot. 1^35-42.
Kincannon, D.F., and Stover, E.L. (1981). Stripping characteristics of prior-
ity pollutants during biological treatment. 74th Annual AICHE Meeting,
New Orleans, LA, November 1981.
Kleshcheva, M.S.; Usacheva, V.T.; and Pozharova. V.N. (1971). Determination
of residual monomers and non-polymerizing impurities in polystyrene
plastics by a gas chromatographic method. Plast. Massy 7:57-58. (In
Russ.) Taken from: Chem. Abst. 75:130411a, 1971.
Klyaev, V.I.; Slisarenko, F.A.; and Finkel'shtein, A.V. (1966). Polarographic
determination of acrylonitrile. Fiz.-Khim. Issled. Prir. Sorbentov Ryada
Anal. Sist. 1^27-33. (In Russ.) Taken from: Chem. Abst. 69:64465n,
1968.
Knoblock, K.; Szendzikowski, S.; Czajkowska, T.; and Krysiak, B. (1971). Acute
and subacute toxicty of acrylonitrile. Med. Pracy 22(3):257-69.
R-10
-------
Kobayashi, Y. (1956). Determination of acrylonitrile vapor by means of de-
tector tubes. Yuki Gosei Kagaku Kyokai Shi JL4:673-75, 1956. (In Japan.)
Taken from: Chem. Abst. 5J.:7240i, 1957.
Kondratenko, V.I.; Geller, B.A.; and Khaskin, I.G. (1971). Chloro derivatives
of aliphatic nitriles. IV. Hypochlorination of acrylic and methacrylic
acid nitriles. Zh. Orgn. Khim. 7(7):1343-47. Taken from: Chem. Abst.
7_5:129033r, 1971.
Kopfler, F.C.; Melton, R.G.; Lingg, R.E.; and Coleman, W.E. (1976). GC/MS
determination of volatiles for the National Organics Reconnaissance
Survey (NORS) of drinking water. In: Identification and Analysis of
Organic Pollutants in Water. Keith, L.H., ed. Ann Arbor, MI: Ann Arbor
Science Publishers, Inc., p. 87.
Korzhova, I.T.; Kleshcheva, M.S.; Porodina, M.N.; Korkhova, R.D.; Balandina,
V.A.; and Khokhlov, V.A. (1974). Gas chromatographic determination of
toxic concentrations of styrene and acrylonitrile vapors during the
synthesis of polystyrene plastics. Plast. Massy 5:67-9. (In Russ.)
Taken from-. Chem. Abst. 81.: 110759J , 1974.
Kostin, L.D., and Vidanova, V.A. (1957). Quantitative determination of acrylic
acid nitrile. USSR Patent 104,643. January 25, 1957. Taken from:
Chem. Abst. 5_l:7246e, 1957.
Krynska, A. (1970). Method of determining acrylonitrile in air. Pr. Cent.
Inst. Ochr. Pr. 20(67):303-10. (In Pol.) Taken from: Chem. Abst. 74:
90798w, 1971.
Kuparov, P.B.; Klyuev, N.A. ,- Mal'tseva, T.A.; and Palagushkina, L.A. (1977).
Quantitative determination of acrylonitrile in SKN vulcanizates by an IR
spectroscopic method. Izv. Vyssh. Uchebn. Zaved., Khim. Khim. Techknol.
20(11):1638-640. (In Russ.) Taken from: Chem. Abst. 88: 63091t, 1978.
Lambotte-Vandepaer, M. ; Duverger-Bogaert, M.; deMeester, C.; Poncelet, F. ,- and
Mercier, M. (1980). Mutagenicity of urine from rats and mice treated
with acrylonitrile. Toxicology 16:67-71.
Langvardt, P.W.; Putzing, C.L.; Young, J.D.; and Braun, W. H. (9179). Isola-
tion and identification of urinary metabolites of vinyl-type compounds:
Application to metabolites of acrylonitrile and acrylamide. Toxicol.
Appl. Pharmacol. 48(1):161.
Lank, J.C., Jr. (1970). Effect of Acrylonitrile on Anaerobic Digestion of
Domestic Sludge. Thesis, University of Idaho Graduate School, March
1970, 76 pp. (Cited in Miller and Villaume, 1978).
Latt, S.A. 1974. Sister chromatid exchanges, indices of human chromosomes
damage and repair: detection of fluorescence and induction by mitomycin
C. Proc. Natl. Acad. Sci. (U.S.A.) 71:3162-3166.
Lawniczak, H. (1977). Quantitative determination of vinyl cynaide in indus-
trial effluents. Barwniki-Srodki Promocnicze 2_1(4) :87-91. (In Pol.)
Taken from: Chem. Abst. 89:220318n, 1978.
R-ll
-------
LeBlanc, G.A. (1980). Acute toxicity of priority pollutants to water flea
(Daphnia aagna). Bull. Environ. Contain. Toxicol. 24:684-91.
Lee, W.R., S. Abrahamson, R. Valencia, E.S. Von Halle, F.E. Wurgler, and S.
Zimmering. In press. The sex-linked recessive lethal test for Buta-
genesis in Drosophila melanoqaster. Mutat. Res.
Leo, A.; Hansch, C. ,- and Elkins, D. (1971). Partition coefficients and their
uses. Chem. Rev. 71_(6) -.525-616.
Leonard, A., V. Garny, F. Poncelet, and M. Mercier. 1981. Hutagenicity of
Acrylonitrile in Mouse. Toxicol. Lett. 7:329-334.
Lezovic, A., and Singliar, M. (1977). Polarographic determination of acryla-
nide and acrylonitrile. Petrochemia 17(5-6) :128-32. (In Slo.) Taken
front: Chen. Abst. 8916338q, 1978.
Lindgren, D.L.4, fincent, L.E.; and Krohne, B.E. (1954). Relative effectiveness
of ten fuaigants to adults of tight apeoiaa of etored-produet inaeets.
J. Econ. Entoool. *7_: 923-26.
Linetskii, V.A., and Serebryakov, B.R. (1965). The alkaline hydrolysia of
nltriles. Khia. Proo. JM(7) :H98-99.
Loveless, A. (1951). Qualitative aspects of the chemistry and biology of radio-
mimetic (mutagenic) aubstances. Mature (London) 16?:33B-I>2.
Loveless, L.E.; Spoerl, E.; and Weisman, T.H. (195^). A aurvey of effects of
chemicals on division and growth of yeast and Eseherichia coli.
J. Bacteriol. 68:637-W.
Ludzack, F.J.; Schaffer, R.B.; Bloomhuff, R.N.; and Ettinger, M.B. (1958).
Biochemical oxidation of some commercially important organic cyanides. I.
River water oxidations. In: Proceedings of the Industrial Vaste
Conference, pp. 297-312.
Ludzack, F.J.; Schaffer, R.B.; and Bloomhuff, R.N. (1961). Experimental treat-
Bent of organic cyanides by conventional processes. J. Water Pollut.
Contr. Fed.
I. (I960). Gas-chromatographic analysis of nltriles. Anal. Chem.
Magos, L. (1962).. A etutfy of acrylonitrile poisoning in relation to »ethaeao-
globin-CK oomplex formation. Brit. J. Ind. Med. I9_:283-286.
Hal let te, F.S. (19^3) . Industrial hygiene in synthetic rubber manufacture. Ind.
Med.
Maltonl, C.; Cillberti, A.; and DlWaio, V. (1977). Carclnogeniclty bloassays on
rats of acrylonitrile administered by inhalation and by Ingestion. Med.
Lavoro 68(6) :U01-«11.
R-12
-------
Mtntel, *.. and M.A. Schntldenun. 1977. CfttlMtlng •§•!«• Uvtll. • Mzardous
undertaking. Cancer fees. 35:1379-1586.
Marino, R.S.; levine, 8. P.; and Banrey, T.M. (1976). Tract determination of
aubnanograa amounts of acrylonitrile in complex aatrioas by gas ehromato-
graphy with a nitrogen-ael active detector. Anal. Cheo. 50(13) : 194 8-50.
Marooni, S., and lonescu, M. (1974). Toxioation of acrylonitrile on fiah. Stud.
Prot. Calitatii Apelor 16:7-22. (Cited in Miller and Villaume, 1976.)
Markelov, M.A., and Semenenko, E.I. (1976). Determination of •icroeoncentra-
tions of substances migrating froo polymer materials to water. Plast.
Massy 1:57-59. (In Buss.) Taken from: Chea. Abst. 84:155747g, 1976.
Markelov, M.E.; McMaster, A.; and Meyer, N. (1981). Analysis and Identification
of organic impurities in water by a headspace gas chrooatographic
technique, presented at the 6th Annual Meeting of the Federation of
Analytical Chemistry and Applied Spectroscopy Societies, Philadelphia, PA,
Septemer 20-25, 1981.
Marrs, D.B.; Campbell, R.L.; Boyko, J.M.; Parrish, A.L., Jr.; and Standish, N.W.
( 1976) . Measurement of acrylonitrile in ambient air in the area of plastic
processing equipment. Paper presented at the 1976 National Technical Con-
ference of the Society of Plastics Engineers, Inc., Chicago, IL, November
14, 1978. Available from: The Standard Oil Company, Cleveland, OH.
Martin, G.M. (1976). Acrylonitrile. Job Safety Health 6(3) :*-13.
McCann, J.; Choi, £.; Yamasaki, £.; and Ames, B.N. (1975a). Detection of
carcinogens as mutagens in the Salmonel la/micros ome test: Assay of 300
chemicals. Proc. Nat. Acad. Sci. USA 21:5135-5139.
McCann, J.; Splngarn, N.E.; Kobori, J.; and Ames, B.N. (1975b). Detection of
carcinogens as mutagens: Bacterial tester strains with R factor plasmids.
Proc. Nat. Acad. Sci. J2( 3) :979-83.
McMahon, B.E.; Cllne, J.C.; and Thompson, C.Z. (1979). Assay of 855 test chemi-
cals in ten tester strains using a new modification of the Ames test for
bacterial mutagens. Cancer Res. 39( 3) ; 682-93.
McNeal, T.; Brumley, V.C.; Breder, C.; and Sphon, J.A. (1979). Gas-solid chrooa-
tographic-mass spectrooetrlc confirmation of low levels of acrylonitrile
after distillation from food-simulating solvents. J. Assoc. Off. Anal.
Chem. 62(l):4l-46.
McNeil, T., and Broader, C.V. (1981). The determination of sub-ppb quantities of
acrylonitrile in 3? acetic acid using manual headspace gas solid
chromatography. U.S. Food and Drug Administration, draft procedure, June,
1981.
McOmie, W.A. (19^3). Comparative toxlcity of methacrylonitrile and acrylo-
nitrile. J. Ind. Hyg. Toxicol. 3J.:113-H6.
R-13
-------
Mekhtiev, S.I.; Mamedova, V.M.; Guseinov, A.G.; Abaahln, V.C.; and Agaeva, I.E.
(1968). Polarographic determination of acrylonitrile and Bethacryloni-
trile. Azerb. Heft. Khoz. ^[(6):33-3^. (In Ruas.) Taken froo: Chem.
Ab3t. 68:113370p, 196B.
Miller, L.M., and Villaume, J.E. (1978). Investigation of Selected Potential
Environmental Contaminants: Acrylonitrile. Publication Ho. EPA-560/2-78-
003. Prepared by the Franklin Institute Research Laboratories,
.Philadelphia, PA, under Contract No. 68-01-3893. U.S. Environmental
Protection Agency, Office of Toxic Substances, Washington, DC. Available
froo: National Technical Information Service, Springfield, VA (NTIS PB-
285-881).
Mills, E.J., Jr., and Stack, V.T., Jr. (1955). Acclimation of microorganisms for
the oxidation of pure organic chemicals. In: Proceedings of the Ninth
Industrial Waste Conference, Purdue University, IN, pp. 4U9-6II.
Milvy, P., and Wolff, M. (1977). Mutagenlc studies with aorylonitrile. Mutat.
Res. jlSCS-1*): 271-278.
Minami, M.; Sato, M.; Suzuki, A.; Tanamura, K.; and lahizu, 5. (1973)* Studies
on acrylonitrile poisoning. 1. On the result of exposure to acrylonitrile
vapor. Tokyo Joshi Ika Diagaku Zasshi j£(10):8U9-S55. (In Jap.; Eng.
aummary.)
Monson, R.R. 1978. Mortality and cancer morbidity among chemical workers with
potential exposure to acrylonltrlle.
Morris, E.D., Jr., and Niki, B. (1971). Reactivity of nydroryl radicals with
olefins. J. Phys. Chem. 7£(23):3640-41.
Murray, F.J.; Nltschke, J.A.; John, P.A.; Smith, F.A.; Quast, J.F.; Blogg, C.D.;
and Sehwetz, B.A. (1976). Teratologlc Evaluation of Acrylonitrile Monomer
Given to Rats by Gavage. Prepared by Toxicology Research Laboratory, Health
and Environmental Research, Dow Chemical USA, Midland, MI, for Chemical
Manufacturers Association, Washington, DC. Available from: CHA,
Washington, DC.
Murray, F.J.; Schwetz, B.A.; Nltschke, K.D.; John, J.A.; Norris, J.M.; and
Gehring, P.J. (1978). Teratogenicity of acrylonitrile given to rats by
gavage or by inhalation. Food Cosmet. Toxicol. 16( 6) :5*<7-552.
Narasiahan, K.S.; Shurpalekar, S.R.; and Venkatesh, K.V.L. (1972). Use of fumi-
gants in prevention of insect infestation and Bold growth in papads.
J. Food Scl. Technol. 9_( 3): 13^-37.
National Academy of Sciences (1975). Acrylonitrile Assessing Potential Ocean
Pollutants, A Report of the Study Panel on Assessing Potential Ocean Pol-
lutants to the Ocean Affairs Board Committee on Natural Resources, National
Research Council, pp. 209-227. (Cited in Miller and Villaume, 1978..)
National Institute for Occupational Safety and Health. 1978. Trip report to
review environmental monitoring and personal work history records at Dupont
May plant. Camden, South Carolina, by Jane Brown. N10SH (April 17).
Submitted to Branch Chief of Industrial Workers Safety Board, MOSH.
R-14
-------
Nazarova, V.I., and Nakrap, L.I. (1978). Determination of acrylonitrile in air.
Melody Anal. Kontralya Kach. Prod. Khim. Prom-sti. £:28-29. (In Buss.)
Taken from: Chem. Abst. 2p_:2826lt, 1979.
Neely, V.B.; Branson, D.R.; and Blau, G.E. (1971). Partition ooeffioient to
•easure bioconcentration potential of organic chemicals in fish. Environ.
Sci. Technol. 8(13) : 1113-15.
Nelson, C.O., and Harder, C.A. (197M). Respirator cartridge efficiency studies.
V. Effect of solvent vapor. Aaer. Xnd. Byg. Aasoc. J. 35J 7) : 391-410.
Nestler, H., and Berger, V. (1965). Gas chromatographic analysis of water
containing solutions of acrylonitrile and various esters in dimethyl-
fonnamide. Chem. Tech. 17( 3) :169-70. (InGer.) Taken from: Chem. Abst.
63_:1213d, 1965.
NIOSH (National Institute for Occupational Safety and Health) (1977a). Acrylo-
nitrile Method No. S156. NIOSH Manual of Analytical Methods, 2nd ed.
Cincinnati, OH: U.S. Dept. of Health, Education, and Welfare, Public Health
Service, Center for Disease Control, NIOSH, Vol. 3.
NIOSH (National Institute for Occupational Safety and Health) (1977b). Method
No. P and CAM 202. NIOSH Manual of Analytical Methods, 2nd ed. Cincinnati,
OH: U.S. Dept. of Health, Education, and Welfare, Public Health Service,
Center for Disease Control, NIOSH, Vol. 1.
NIOSH (National Institute for Occupational Safety and Health) (1977c). Current
Intelligence Bulletin 18: Acrylonitrile. Current Intelligence Bulletin:
Acrylonitrile. July 1, 1977. U.S. Dept. of Health, Education, and Welfare,
Public Health Service, Center for Disease Control, NIOSH. Available from:
NIOSH, Rockville, MD.
Norris, M.B. (1962). Acrylonitrile. Encycl. Ind. Chem. Anal. £:368-8l.
O'Berg, M. (1980). Epldemlologic study of workers exposed to acrylonitrile.
J. Occup. Med. 22:2^5-252.
Ofengand, J. (1967). The function of pseudouridylic acid in transfer ribonucleic
acid. I. The specific cyanoethylatlon of pseudourldine, inosine, and 4-
thiouridine by acrylonitrile. J. Biol. Chem. 2H2(21) t
Ofengand, J. (1971). Cyanoethylation of nucleotides and tRNA by acrylonitrile.
In: Methods of Enzymology. New York: Academic Press, Vol. 20, pp. 150-68.
Panova, R.V.; Belova, R.A.; and Lavrushina, T.S. (1969). Chromatographic deter-
mination of residual monomers in latexes. Prom. Sin. Kauch., Nauch.-Tekh.
Sb. .1:21-23. (In Russ.) Taken from: Chem. Abst. 7JUl3238d, 1971.
Parent, R.A., and Casto, B.C. (1979). Effect of acrylonitrile on primary Syrian
golden hamster embryo cells in culture: Transformation and DNA fragmenta-
tion. J. Nat. Cancer Inst. 62( «4) : 1025-29.
Patterson, R.M.; Bornstein, M.I.; and Carshick, E. (1976). Assessment of Acrylo-
nitrile as a Potential Air Pollution Problem, Publication No. GCA-TR-75-
32-0(6). Prepared by CCA Corporation, CCA/Technology Division, Bedford,
MA, under Contract No. 68-02-1337. U.S. Environmental Protection Agency,
Research Triangle Park, NC.
R-15
-------
Paulet, C.t and Vidal, Mrs. (1975). [Toxicity of aone acrylic and •ethaerylie
esters of acrylamide and polyacrylaaides .] Arch. Mai. Prof. Med. Trav.
Secur. Soc. 26(1-2) : 58-60. (In Fr.) Taken from: Chem. Abst. 83j9l832a,
1975 •
Pell, $., C.I. duPont de Nemours and Co. 1980. Letter to the Carcinogen
Assessment Group. U.S. Environmental Protection Agency (July 23).
Perry, F.H., and Chi It on, C.R., editors (1973). Chemical Engineers* Handbook,
5th ed. New York: McGraw-Hill Book Co., pp. 3-61.
Peters, T.L. (1980). Steam distillation apparatus for concentration of trace
water soluble organics. Anal. Chem. J52_:21 1-213.
Petrova, L.I.; Maksimova, N.S.; and Boikova, Z.K. (1978). Determination of
traces of atyrene, acrylonltrile, and nethacrylate monomers in binary mix-
tures. Gig. Sanit. 14:814-86. (In Buss.) Taken from: Chem. Abst.
2P_:6935P, 1979.
Pilotti, A.; Ancker, K.; Arrhenius, E.; and Enzell, C. (1975). Effects of
tobacco and tobacco smoke constituents on cell multiplication in vitro.
Toxicology ^(1): 19-62.
Plieninger, H., and Sharma, S. (1978). Mote on a simple detection of acrylo-
nitrile in solvents. Justus Liebigs Ann. Chem. 3j535-36. (In Cer.) Taken
from: Chem. Abst. 8_9_:52913g, 1978.
Pokrovskaya, L.A., and Prolova, C.S. (1969)* Chromatographic method in produc-
tion control of butadiene-acrylonitrile rubbers. Piz. Khim. Khim.
Tekhnol., pp. 201-06. (In Puss.) Taken from: Chem. Abst. 15_:9913**e, 1971.
Ponomarev, Yu P.; Anufrieva, T.L.; and Shkorbatova, T.L. (19710. Polarographic
determination of acrylonltrile In waste waters using extraction.
Vodosnabzh., Kanallz., Cldrotekh. Sooruzh. n.:67-70. (In Russ.) Taken
from: Chem. Abst. tte^SOSx, 1975.
Portmann, J.E., and Wilson, K.E. (1971). The toxiclty of 1*40 substances to the
brown shrimp and other marine animals. Shellfish Inf. Leafl. No. 22, 12 pp.
Minist. Agric., Fish, and Food, Fish. Lab., Burnham-on-Crouch, Essex,
England.
Quast, J.F.; Enriquez, R.M.; Wade, C.E.; Humiston, C.G.; and Schwetz, B.A.
(1977). Toxicity of Drinking Water Containing Acrylonitrile (AN) in Rats:
Results after 12 Months. Prepared by the Toxicology Research Laboratory,
Health and Environmental Research, Dow Chemical USA, Midland, MI, for the
Chemical Manufacturers Association, Washington, DC. Available from: CMA,
Washington, DC.
Quast, J.F.; Humiston, C.G.; Schwetz, B.A.; Frauson, L.A.; Wade, C.E.; and
Norris, J.M. (1975) . A Six-Month Oral Toxicity Study Incorporating Acrylo-
nitrile in the Drinking Water of Purebred Beagle Dogs. Prepared by the
Toxicology Research Laboratory, Health and Environmental Research, Dow
Chemical, USA, Midland, MI, for the Chemical Manufacturers Association,
Washington, DC. Available from: CMA, Washington, DC.
R-16
-------
Quasi, J.F.; Wade, C.E.; Humiston, C.G.; Carreon, R.M.; Hermann, E.A.; Park,
C.N.; and Schwetz, B.A. (1980a). A Two-Year Toxicity and Oneogenlcity Study
with Acrylonitrile Incorporated in the Drinking Water of Rats. Prepared by
the Toxicology Research Laboratory, Health and Environmental Sciences, Dow
Chemical USA, Midland, MI, for the Chemical Manufacturers Association,
Washington, DC. Available from: CMA, Washington, DC.
Quasi, J.F.; Schuetz, D.J.; Balmer, M.F.; Gushow, T.S.; Park, C.N.; and McKenna,
M.J. (198Ob). A Two-Year Toxicity and Oncogenicity Study with Acrylo-
nitrile Following Inhalation Exposure of flats. "repared by the Toxicology
Research Laboratory, Health and Environmental Sciences, Dow Chemical, USA,
Midland, MI, for the Chemical Manufacturing Associalion, Washinglon, DC.
Available from: CMA, Washington, DC.
Rabello-Gay, M.N., and Ahmed, A.E. (1980). Acrylonitrile: In vivo cytogenic
studies in Bice and rats. Mutat. Res. 79_:2l9-255«
Radimer, G.F.; Davis, J.H.; and Ackernan, A.B. (1971). Fumigant-induced toxic
epidermal necrolysis. Arch. Dennatol. 110(1);103-01*
Rajendran, S., and Muthu, M. (1976). Toxicity of acrylonitrile to the adulta of
five apecies of atored producls insects. Bull. Grain Technol.
11( 3): 179-181.
Randall, T.L., and Knopp, P.V. (1980). Detoxification of specific organic aub-
slances by wel oxidation. J. Water Pollut. Contr. Fed. £2:2117-2130.
Reichle, A., and Tengler, H. (1968). Gas-chrooatographic procedures for deler-
nining residual monomers in the polymer. Delerminalion of acrylonitrile in
polyacrylonitrlle. Plastverarbeiter ^9_( 12):921-21. (InGer.) Taken from:
Chen. Abst. £0:97206c, 1969.
Renn, C.E. (1955). Biological properties and behaviors of cyanogenic wastes.
Sewage Ind. Wasles 27( 3):297-310. (Cited in Miller and Villaume, 1978.)
Roberfroid, M.; Poncelet, F.; Lambotte-Vandepaer, M.; Duverger-VanBogaert, M.;
DeMeester, C.; and Mercier, M. (1978). Acute biotoxic effect of atyrene on
rat liver. Correlation with enzyme-mediated nutagenicily of benzypyrene
and acrylonitrile. Scand. J. Work Environ. Heallh 1(suppl. 2):163-68.
Rogaczewska, T. (1961). Polarographic determinallon of acrylonitrile in air.
Chem. Anal. $ 3):117-23. (In Pol.) Taken from: Chem. Abst. 62:995d, 1965.
Rogaczewska, T. (1975). [Percutaneous absorption of acrylonitrile vapors in
animals.l Med. Pracy 26:159-65. (In Pol.)
Rogaczewska, T. (1976). Acrylonitrile determination in air. Med. Pracy
27(2):115-26. (In Pol.) Taken from: Chem. Absl. 86:21137r, 1977.
Roudabush, R.L.; Terhaar, C.J.; Fassett, D.W.; and Dziuba, S.P. (1965). Compara-
tive acute effects of some chemicals on the skin of rabbits and guinea pigs.
Toxicol. Appl. Phannacol. l(1):559-65.
Roy, S.S. (1977). Titrimetric delerminalion of residual monomers in olyrene-
acrylonitrile copolymers. Analyst 102( 1213);302-05.
R-17
-------
Russell, J.V. (1975). Analysis of air pollutants using sampling tubes and gas
chromatography. Environ. Sci. Technol. £(13) :1175-78.
Russkikh, A. A. (1971). Determination of acetone oyanohydrin and acrylonltrile
in air. Piz.-Rhim. Metody Anal. Kontr. Proizvod., Mater. Konf. Rab. Vuzov
Zavod. Lab. Yugo-Vostoka SSSR, 4th 1:100 (Pub. 1972). (In Russ.) Taken
from: Chen. Abst. tU:l6301g, 1974.
Russkikh, A. A. (1973). Photometric determination of acrylonitrile and neth-
acrylonitrile in air. Zavod. Lab. 29_(l):5-6. (In Russ.) Taken from:
Chem. Abst. J_9_:101566e, 1973.
Sakurai, H.; Onodera, M.; Utsunoaiya, T.; Minakuchi, H.; Iwai, H.; and Matsumura,
H. (1978). Health effects of acrylonitrile in acrylic fiber factories.
Brit. J. Ind. Med. 25:219-22.
Sandberg, E. Ch., and Slanina, P. (1980). Distribution of [I-^CJ -acrylonitrile
in rat and monkey. Toxicology Lett. 6:187-191.
Sana one, E.B.; Tewari, I.E.; and Jonas, L.A. (1979). Prediction of removal of
vapors from air by adsorption on activated carbon. Environ. Sci. Technol.
12(12): 1511-13.
Santodonato, J.; Basu, D.; and Howard, P.H. [19791 . Multimedia human exposure
and carcinogenic risk assessment for environmental PAH. In: Chemistry and
Biology of Pol/nuclear Aromatic Hydrocarbons. (Proceedings of the lith
International Battelle Symposium on Polynuclear Aromatic Hydrocarbons,
Columbus, OH, October 2-1, 1979.) Ann Arbor, MI: Ann Arbor Science
Publishers, Inc. (Book in press.)
Sarofim, A. P.; Wagner, J.P.; and Tewarson, A. (1973). Gas evolution during the
pyrolysis and combustion of acrylonitrile/styrene polymers. In: Proceed-
ings of the Symposium on Environmental Impact of Nitrile Barrier Cont-
ainers, Hartford, CT, July 19, 1973. Sponsored by and available from
Monsanto Co., St. Louis, MO.
Sartorelli, E. (1966). [Acute poisoning due to acrylonitrile.) Med. Lavoro
52:184-87. (In Ital.)
Scheddel, R.T. (1958). Infrared quantitative analysis data: Analysis of acrylic
acid, ethylbenzene, methyl methacrylate, °-methylstyrene, styrene, and
acrylonitrile mixtures. Anal. Chem. 2
Sevast'yanova, I.G., and Tomilov, A. P. (1963). Reduction of nitriles on a
dropping mercury cathode. Zh. Obshch. Khim. ^1( 9) : 2815-19. (In Russ.)
Taken from: Chem. Abst. 60:1583e, 19&U.
Sevast'yanova, I.G.; Tomilov, A. P.; and Yanaleev, I.Ya. (1966). Polarographic
determination of acrylonitrile in industrial solutions. Zavod. Lab.
22(10): 1210. (In Russ.) Taken from: Chem. Abst. 66:31651r, 1967.
Silverstein, L.G. (1977). Validation of abcor GASBADGE for acrylonitrile and
improved desorption efficiency. Amer. Ind. Hyg. Assoc. J. 58(8) : 112-13.
Sloof, W. (1978). Biological monitoring based on fish respiration for continuous
water quality control. In: Aquatic Pollutants: Transformation and Bio-
logical Effects. Hutzlnger, 0.; VanLelyveld, L.H.; and Zoeteman, B.C.J.,
editors. Elmsford, NY: Pergamon Press, pp. 501-06.
R-18
-------
Smith, C.J. (1975). Solvent vapour phototoxicity. Lab. Pract. 2^:9-11.
Smyth, H.F., Jr., and Carpenter, C.*. (1918) . Further experience with the range
finding test in the industrial toxicology laboratory. J. Ind. Hyg.
Toxicol. 20:63-68.
Smyth, H.F., Jr.; Well, C.S.; West, J.S.; and Carpenter, C.P. (1969). An
exploration of joint toxic action: Twenty-seven industrial chemicals intu
bated in rats In all possible pairs. Toxicol. Appl. Pharmacol.
1JU 2): 3^0-07.
Stallard, C. 1982. AcrylonHMle epidemiology. Standard 011 Company of Ohio
Unpublished report.
Stanford Research Institute (1978). 1978 Chemical Economics Handbook. Menlo
Park, California, 1978. p. 607.5032J.
Steere, N.V. (1968). Safety in the chemical laboratory. XLV1. Hazardous
chemicals data. J. Chem. Ed. ££( 3) :A199-A205.
Stefanescu, T., and Ursu, Oh. (1973)' Method for acrylonitrile and acetonitrile
determination in residual waters. Mater. Plast. 10(6) : 330 -3^. (In Rom.)
Taken from: Chen. Abst. 60: 10770 J, 1971.
Steichen, R.J. (1976). Modified solution approach for the gas ehrooatographic
determination of residual monomers by head-space analysis. Anal. Chem.
18(9): 1398-1102.
Suta, 6.E. (1979). Assessment of Human Exposures to Atmospheric Acrylonitrile.
Prepared by Stanford Research Institute, Menlo Park, CA, under Contract No.
68-02-2835. U.S. Environmental Protection Agency, Office of Air Duality
Planning and Standards, Research Triangle Park, NC.
Suta, B.E. (1982). Letter to J.E. Hadley, Jr. (LaRoe, Vinn, and Moennan,
Attorneys at Lav, Washington, D.C.) transmitting enclosed report entitled:
"Revised Assessment of Human Exposure to Atmospheric. Acrylonitrile Using
Industry Supplied Emission Estimates". June 21, 1982.
Sweger, D.M., and Travis, J.C. (1979). An application of infrared lasers to the
selective detection of trace organic gases. Appl. Spectrosc. jg( 1) ;16-51.
Szabo, S., and Selye, H. (1971). Adrenal apoplexy and necrosis produced by
acrylonitrile. Endokrinologie 5J( 3) : 105-08.
Szabo, S., and Selye, H. (1972). Effect of phenobarbital and steroids on the
adrenal apoplexy produced by acrylonitrile in rats. Endocrinol. Exp.
6(3): 111-16.
Szabo, S.; Reynolds, E.S.; Komanicky, P.; Moslen, M.T.; and Melby, J.C. ( 1976a) .
Effect of chronic acrylonitrile ingestion on rat adrenal. Toxicol. Appl.
Pharmacol. J7_:133. (Abstract.)
Szabo, S.; Reynolds, E.S.; and Kovacs, K. (1976b). Animal model of human
disease. Waterhouse-Friderichsen syndrome. Animal model: Acrylonitrile-
induced adrenal apoplexy. Amer. J. Pathol. 82( 3) :653-56.
Szabo, S.; Bailey, K.A.; Boor, P.J.; and Jaeger, R.J. ( 1977a) . Acrylonitrile and
tissue glutathione: Differential effect of acute and chronic interactions.
Biochem. Biophys. Res. Commun. 7_9_( 1) : 32-37.
R-19
-------
Szabo, S.: Boor, P.J.; Oelwang, M.C.; and Bailey, K.A. (1977b). Pathogenesis of
acrylonitrile poisoning: Role of glutathione and coagulation changes. In:
International Congress on Toxicology. Abstracts. Burford, P.O., editor.
Toronto, Ontario, Canada: International Congress on Toxicology.
Tanaka, M.; Nishinura, P.; and Shono, T. (1975). Pyrolysis-gas ohromatography of
vinyl chloride-methyl nethacrylate and vinyl chloride-acrylonitrile copoly-
mers. Anal. Chim. Acta ]±( 1) : 119-21.
Taubinger, R.P. (1969). Direct determination of free aerylonitrile in aqueous
copolyaer latexes. Analyst 9JK 1121) : 628-33.
Terent'ev, A. P., and Obtenperanskaya , S.I. (1956). Quantitative determination
of acrylonitrile with sodium sulfite. Zhur. Anal. Khim 11: 638-39. (In
Buss.) Taken from: Chem. Abst. £l:11935h, 1957.
Thiess, A.M., and Plelg, I. (1978). Analysis of chromosomes of workers exposed
to aerylonitrile. Arch. Toxiool. |U(2) :ll9-52.
Thiess, A.M., R. Frentzel-Beyme, R. Link, and H. Wild. 1980.
Mortal itatsstudle Be1 ChemlefacharbeUern VerscMedener Produktlonsbetriebe
Mit Exposition Auch Gegenuber Acrylonitrile Zentralblatt Arbeltsmedizln
30:259-267.
Thompson, K.R. (1971). Process BBSS ipeetrometry . DuPont Innovation
£(3): 18 -20.
Troxell, T.C. (1980). Letter from T.C. Troxell, Division of Pood and Color
Additives, Food and Drug Administration, Department of Health, Education,
and Welfare, Washington, DC, to J. Santodonato, Syracuse Research Corpora-
tion, Syracuse, NY, dated August 20, 1980.
Tsuchiya, T., and Sumi, K. (1977). Thermal decomposition products of poly-
acrylonitrile. J. Appl. Polymer Sci. 21(1) : 975-80.
Uhde, W.J., and Koehler, 0. (1967). Testing of plastic articles. Determination
of monomerlc compounds in the volatile fraction of etyrene polymers. Z.
Lebensm.-Unters. Forsch 135(3) : 135-10. (InGer.) Taken from: Chem. Abst.
67:117599b, 1967.
U.S. EPA (Environmental Protection Agency) (1976b). EPA Method 621, Purgeables,
11 Fed. Feg. 69532-69593, Dec. 3, 1979.
U.S. EPA (U.S. Environmental Protection Agency) ( 1978a) . Data submitted to U.S.
EPA, Washington, DC, under Contract No. 68-01-1616 by E.G. and G. Bionomics,
Wareham, MA. Contract title: In-Depth Studies on Health and Environmental
Impacts of Selected Water Pollutants.
U.S. EPA (Environmental Protection Agency) (1979a). EPA Method 603, Acrolein and
Acrylonitrile, 11 Fed. Reg. 69179-69181, Dec. 3, 1979.
U.S. EPA (U.S. Environmental Protection Agency) (1980). OAQPS, Emission
Standards and Engineering Division, "Organic Chemical Manufacturing -
Vol. 10: Selected Processes, Chapter 2: Acrylonitrile, "
EPA-l50/3-80-028e, Research Triangle Park, NC, December 1980.
R-20
-------
U.S. EPA (U.S. Environmental Protection Agency) (1980). Ambient Water Quality
Criteria Document: Acrylonitrile. U.S. EPA. EPA U10/5-80-017, Oct., 1980.
Available from: National Technical Information Service, Springfield, VA.
U.S. EPA (Environmental Protection Agency) (1980c). EPA Method 1621. Volatile
Organic Compounds by Purge and Trap Isotope Dilution GC-MS, U.S. EPA
Effluent Guidelines Division, June, 1980.
U.S. EPA (U.S. Environmental Protection Agency) (1981). The Carcinogen Assess-
ment Group's Carcinogen Assessment of Acrylonitrile. Available from: U.S.
EPA, Office of Research and Development, Washington, DC.
U.S. Environmental Protection Agency. Science Advisory Board Environmental
Health Committee. Transcripts from August 2. 1982.
U.S. Health, Education, and Welfare, National Clearinghouse for Smoking and
Health. 1973. Adult use of tobacco, 1970. U.S. HEW 1973.
Ustinovskaya, I.A.; Klzllova, L.I.; Oavrllina, L.Ta.; Malakhov, V.V.; and
Yashin, Ya.I. (1977)* Gas-chromatographlo analysis of products from the
oxidative amaonolysis of propylene. Izv. Sib. Otd. Akad. Hauk SSSR, Ser.
Khim. Nauk 2:115-18. (In Fuss.) Taken from: Chem. Abst. 87_:77980y, 1977.
Ved Brat, S. and G.M. Williams. 1982. Hepatocyte-mediated production of sister
chromatid exchange 1n cocultured cells by acrylonitrile: evidence for
extra-cellular transport of a stable reactive Intermediate. Unpublished.
Veith, G.D.; DeFoe, D.L.; and Bergstedt, B.V. (1979). Measuring and estimating
the bioconcentration factor of chemicals In fish. J. Pish. Res. Board Can.
26:1010-18.
Venitt, S. (1978). Mutagenlc studies with acrylonitrile. Mutat. Res.
5_7_(1): 107-09.
Venitt, S.; Bushell, C.T.; and Osborne, M. (1977). Mutagenicity of acrylonitrile
(cyanoethylene) In Eseherichia coli. Mutat. Res. 45_(2) :283-88.
Werner, J.B., and Carter, J.T. (1981). Mortality of United Kindom acrylonitrile
polymersation workers. British Journal of Industrial Medicine 38:217-253.
Wilcox, W.S., and Goldstein, J.H. (1951). The microwave spectrum of vinyl
cyanide. J. Chem. Phys. 22(3):516-18.
Wilson, R.H. (1911). Health hazards encountered in the manufacture of synthetic
rubber. J. Amer. Med. Assoc. 121:701-03.
Wilson, R.H.; Hough, G.V.; and McCormlck, W.E. (1918). Medical problems encoun-
tered in the manufacture of American-made rubber. Ind. Med. 17(6):199-207.
Wolff, S. 1977. Sister chromatid exchange. Annu. Rev. Genet. 11:183-201.
R-21
-------
Wright, P.L. (1977). Studies on the metabolism of acrylonitrile. Testimony
before the FDA. (Cited in Miller and Villaume, 1978.)
Wronski, M., and Zbigniew, S. (1971). Thlotnercurinetric determination of
acrylonltrile In water. Chem. Anal. 1£(3):633-38. (In Pol.) Taken from:
Chem. Abst. 8ll75988c, 197H.
Young, J.D.; Slauter, R.V.; and Karbovstcl, R.J. (1977). The Phannacoklnetic and
Metabolic Profile of UlC-Acrylonitrile Given to Rats by Three Routes. Pre-
pared by the Toxicology Research Laboratory, Health and Environmental
Research, Dow Chemical USA, Midland, MI, for the Chemical Manufacturers
Association, Washington, DC. Available from: CMA, Washington, DC.
Zack, J.A. (I960). Written communication from J.A. Zack, Monsanto, St. Louis,
MO, to W.R. Gaffey, Monsanto, St. Louis, MO, dated June 26, 1980, regarding
Acrylonitrile Epidemiology Study: Latent Analysis.
Zitting, A.; Tenhanen, R.; and Savolainen, H. (1981). Effects of
intraperitoneally injected acrylonltrile on liver, kidney and brain. Acta
Pharmacol. Toxicol. Jr9:Ul2-ll15. Taken from : Chem. Abst. 56:117119u.
R-22
------- |