United States
             Environmental Protection
             Agency
            Office of Health and
            Environmental Assessment
            Washington DC 20460
EPA-600/8-82-007F
October 1983
Final Report
              Research and Development
&EPA
Health Assessment  Final
Document for         Report
Acrylonitrile

-------

-------
                             EPA-600/8-82-Q07F
                                   October 1983
                                    Final Report
     Health Assessment
Document for Acrylonitrile
     U.S. ENVIRONMENTAL PROTECTION AGENCY
        Office of Research and Development
     Office of Health and Environmental Assessment
     Environmental Criteria and Assessment Office
        Research Triangle Park, NC 27711

-------
                      NOTICE

This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication.  Mention of trade names
or commercial products does not constitute endorse-
ment 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 Agency-wide use.   The  health




assessment document was originally developed  at  the request of the Office of Air



Quality Planning and Standards; however, the scope of  the assessment has since




been expanded  to address multimedia aspects.  This assessment will  help  ensure




consistency in the Agency's  consideration of the relevant scientific  health data




associated with acrylonitrile.




     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  environ-




mental levels.
                                      ii!

-------

-------
                               TABLE OF CONTENTS
PREFACE	     iii

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 WEIGHT	   3-1
     3.3  BOND ANGLES AND BOND DISTANCES			   3-1
     3 .4  PHYSICAL PROPERTIES .	.,		..		   3-1

        3.4.1   Description ...	   3-1
        3.4.2   Boiling Point	   3-1
        3 .4.3   Melting Point	   3-1
        3 .4.4   Density	. ^................   3-2
        3.4.5   Refractive Index	,,,.   3-2
        3.4.6   S pec troscopic Index	, 4.   3-2
        3.4.7   Solubility ....		'...   3-2
        3.4.8   Volatility in Water	,.,.,...,   3-2
        3.4.9   Volatility	 •...........   3-3
        3.4.10  Stability			   3-3
        3.4.11  Octanol-Water Partition Coefficient	   3-3
        3.4.12  Conversion Factor ,	   3-3

     3.5  CHEMICAL PROPERTIES	   3-3

        3.5.1   Reactivity	   3-3
        3.5.2   Polymerization	   3-3
        3.5.3   Reaction at the Nitrile Groups	,..	   3-4
        3.5.4   Reactions at the Double Bond	   3-5
        3.5.5   Cyanoethylation Reaction	   3-5

     3.6  CHARACTERISTICS OF THE CHEMICAL PRODUCT	   3-6

     3.7 .CONCLUSION	,	   3-6

-------
                      TABLE OF CONTENTS (cont.)
SAMPLING AND ANALYTICAL METHODS	  4-1

4.1  AIR	  4-1

   4.1.1   Sampling	  4-1
   4.1.2   Analysis	  4-6
   4.1.3   Conclusions	 4-10

4.2  WATER	 4-11

   4.2.1   Sampling	 4-11
   4.2.2   Analysis 	„	 4-12
   4.2.3   Conclusions	 4-15

4.3  WASTEWATER	„	 4-15

   4.3.2   Analysis	 4-15
   4.3.3   Conclusions 	.,	 4-17

4.4  SOIL AND SEDIMENT	 4-17

   4.4.1   Sampling	 4-17
   4.4.2   Analysis	 4-17
   4.4.3   Conclusions
4.5  RESIDUE IN POLYMERS AND THE EXTENT OF MONOMER MIGRATIO
         4-19
IN
     FOOD-SIMULATING SOLVENTS	  4-19

   4.5.1   Analysis	  4-20
   4.5.2   Conclusions	  4-25

4.6  OTHER MEDIA	  4-25

4.7  GENERAL METHODS FOR THE ANALYSIS OF ACRYLONITRILE 	  4-25

SOURCE 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

-------
                           TABLE OF CONTENTS  (con-t. )
     5.5  SOURCES OF EMISSIONS	,	  5-8

        5.5.1   Monomer and Polymer Production	  5-8
        5.5.2   Emissions During Transportation	  5-8
        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 Acrylonitrile  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

     7 .4  CONCLUSIONS	 7-16

8.   BIOLOGICAL EFFECTS ON MICROORGANISMS	  8-1
                                      VI I

-------
                           TABLE OF CONTENTS (cont.)

                                                                           Page

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	    12-5
        12.1.3  Freshwater Invertebrates 	    12-6
        12.1.1}  Marine Invertebrates	    12-6

     12.2 SUBCHRONIC TOXICITY	    12-7

        12.2.1  Freshwater Fish	,	    12-7
        12.2.2  Freshwater Invertebrates	    12-8

     12.3 SUMMARY AND CONCLUSIONS	    12-9

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
        13.1-3  Summary and Conclusions	  13-16

     13.2 ACUTE, SUBCHRONIC,  AND CHRONIC TOXICITY	  13-18

        13.2.1  Acute Toxicity	  13-18
        13.2.2  Subchronic Toxicity in Non-Human Mammals  	  13-42
        13.2.3  Chronic Toxicity in Non-Human Mammals	  13-45
        13.2.4  Summary and Conclusions  	.  13-50

     13.3 TERATOGENICITY AND  REPRODUCTIVE TQXICITY  	  13-52

        13.3.1  Summary and Conclusions  	  13-59
                                     vii

-------
                           TABLE OF CONTENTS (cont.,)
     13.4 MUTAGENICITY	  13-61

        13.4.1  Gene Mutation Studies 	.	  13-61
        13.4.2  Chromosomal Aberration Studies	  13-70
        13.4.3  Other Tests Indicative of Genetic Damage 	  13-73
        13-4.4  Summary and Conclusions	  13-77

     13 .5 CARCINOGENICITY 	..,..,.		.  13-89

        13.5.1  Anjmal Studies .	...;...,	.•.-..........,......,*..  13-89
        13.5.2  Epidemiologic Studies 	13-118
        13.5.3  Quantitative Estimation	 13-138
        13.5.4  Summary	 13-169
        13.5.5  Conclusions	 13-176
        13.5.6  Appendix—Comparison of Results by Various
                Extrapolation Models	 13-178

REFERENCES	    R-1
                                       ix

-------
                                LIST OF TABLES
No.

4-1


4-2

4-3

4-4

4_5

4-6


5-1

5-2


5-3
5-5

5-6

5-7

5-8

5-9


5-10


5-11


5-12
Title

Various Sorbents and Trapping Media for Collection of
  Acrylonitrile in Air	     4-3

Recovery of Acrylonitrile from Various Solvents 	     4-8

Direct Analysis of Acrylonitrile 	     4-9

Analyses of Acrylonitrile in Water 	    4-16

Analysis of Acrylonitrile in Wastewaters 	    4-18

Analysis of Acrylonitrile Residue in Polymers and
  Food-Simulating Solvents 	    4-21

Producer of Acrylonitrile in the United States 	     5-2

Distribution of Acrylonitrile in 1980 and Projected Growth
  Through 1984	     5-2

Primary Uses for Compounds Synthesized from AeryIonitrile-
Containing Compounds	     5-3

Acrylonitrile Consumption and Project Growth of Products
  Using Acrylonitrile	     5-4

Producers of SAN and ABS Resins 	     5-6

Producers of Acrylic and Modacrylic Fibers 	     5-6

Producers of Nitrile Rubbers and Elastomers 	     5-7

Producers of Aerylamide (SRI, 1978) 	     5-7

Estimated Atmospheric Emissions of Acrylonitrile from
  Monomer Production Facilities	     5-9

Estimated Acrylonitrile Emission Rates from ABS-SAN
  Resin Production	    5-10

Estimated Acrylonitrile Emission Rates from Acrylic
  Fiber Production	    5-10

Estimated Acrylonitrile Emission Rates from Nitrile
  Elastomer Production	    5-11

-------
                                LIST OF TABLES
No.

5-13


5-14

5-15

7-1

7-2

7-3

7-4

7-5

7-6



7-7


7-8


11-1


12-1


13-1


13-2


13-3


13-4


13-5
Title

Estimated Acrylonitrile Emission Rate from Adiponitrile
  Production	    5-11

Hazard of Acrylonitrile Transportation 	    5-13

Monomer Residue in End-Products of Acrylonitrile 	    5-14

Atmospheric Monitoring Data for Acrylonitrile	     7-2

Comparison of Monitoring and Dispersion Modeling Da.ta	     7-4

Acrylonitrile Monitoring Data for Surface Waters 	     7-6

Acrylonitrile Monitoring Data for Sediments 	     7-8

Acrylonitrile Monitoring Data for Soils 	     7-8

Estimates of Population Exposures to Atmospheric
  Acrylonitrile from Specific Emission Source
  Categories	    7-10

Acrylonitrile Migration Under Different Storage
  Conditions	    7-12

Amounts of Various Acrylonitrile Copolymers Used in
  Food-Contact Applications	    7-13

Lethal Dose Values for Insects Exposed to Acrylonitrile
  Fumigation 	    11-2

Median Lethal Concentration (LC  ) Values for Fish
  Exposed to Acrylonitrile	    12-2

Recovery of Radioactivity from Rats Given Single Oral
  Doses of 0.1 or 10 mg/kg   C-Acrylonitrile 	      13-2

Recovery of Radioactivity from Rats Exposed by Inhalation
  to 5 or 100 ppm   C-Acrylonitrile for 6 hours 	      13-3

Distribution of Radioactivity in Selected Tissue of Rats
  Given   C-Acrylonitrile	      13-5

Urinary Metabolites Following the Oral Administration of
  C-1 (Cyano) Labeled Acrylonitrile	   13-14

               14
Metabolites of   C-Acrylonitrile Separated from Various
  Fluids of Rats by High Pressure Liquid Chromatography
  (HPLC)	   13-17
                                      XI

-------
No.

13-6


13-7


13-8


13-9


13-10



13-11


13-12



13-13



13-1*1


13-15


13-16

13-17


13-18

13-19


13-20
                LIST OF TABLES (oont.)

 Title                                                        page

 Summary of  Results  of Exposures of Rats to
   AeryIonitrlie	.	  13-22

 Minimal Lethal Concentration  of Acrylonitrile During
   Four-Hour Exposure	  13-23

 Comparison  of the Effects of  Acrylonitrile and of Hydro-
   cyanic Acid on Various Species of  Animals  	  13-26

 Cyanide and Thiocyanate in Blood of  Animals Exposed
   to  Acrylonitrile	  13-28

 Effect of Methemoglobinemia on Mortality Ratios in Albino
   Rats Poisoned with Acrylonitrile,  Potassium Cyanide, and
   Acetone Cyanohydrin 	  13-33

 Therapeutic Effect  of SH and  S-S Compounds on Acute, '
   Acrylonitrile Poisoning	,	  13-37

 Concentration of Protein (PBSH) and  Nonprotein (NPSH) SH
   Groups in Normal  and Acrylonitrile-Intoxicated Animals
   ([jmoles SH/100 g  wet tissue)	  13-39

 Cumulative Mortality of Male  and Female Rats Maintained
   for Two Years on  Drinking Water Containing
   Acrylonitrile	  13-49

 Significant Changes Considered to be Secondary to Ingestion
   of  Acrylonitrile	  13-51

 Incidence of Fetal  Malformations Among Litters of Rats
   Given Acrylonitrile	  13-55

 Pup Weight on Days  4 and 21 of Lactation	.:	  13-58

Tissues Examined for Histopathologic Changes in the
   F3b Litter	„	  13-59

Mutagenicity Tests  of Acrylonitrile  	  13-80

 Cumulative Mortality Data of Male Rats Maintained for
 2  Years on Drinking Water Containing Acrylonitrile	  13-91

 Cumulative Mortality Data of Female Rats Maintained           : •"•••
 for 2  Years on Drinking Water Containing Acrylonitrile .....  13-92
                                      XI I

-------
LIST OF TABLES (cont.)
Nov
13-21


13-22


13-23

13-24

13-25

13-26

13-27

13-28

13-29
13-30

13-31

13-32

13-33

13-34.
13-35


Title
Histopathologic Diagnoses and Tumor Incidences in Male
Rats Maintained for 2 Years on Drinking Water Containing

Histopathologic Diagnoses and Tumor Incidences in Female
Rats Maintained for 2 Years on Drinking Water Containing

Tumor Incidences .in Sprague-Dawley Rats Fed Acrylonitrile

Tumor Incidences in Fischer 344 Rats Fed Acrylonitrile

Incidence of Tumors Observed in Rats During a Three-
Generation Reproductive Study 	 •. 	 	 	 	 	
Tumor Incidence in Rats Fed Acrylonitrile Orally by

Tumor Incidence in Rats Following Inhalation of

Stability and Trace Impurity Analysis of the Acrylonitrile
Liquid Test Material 	 	 	

Cumulative Mortality Data of Male Rats Exposed by

Cumulative Mortality Data of Female Rats Exposed by

Tumor Incidence in Sprague-Dawley Rats Exposed to

Enhancement of SA7 Transformation by Treatment of
HEC with ACN 	 	 	
Transformation of HEC by ACN 	 	 	 	 	
Number (Percentage) of Rats Developing Tumors in at
Least One Target Organ: Three Acrylonitrile Drinking

Page


13-93


13-94

13-98

13-100

13-102

13-105

13-106

13-108
13-109

13-110

13-111

13-112

13-115
13-117


13-157
         XI 1 I

-------
                            LIST OF TABLES (cont.)
No.

13-36


13-37



13-38

13-39


13-40
13-41
13-42
13-43
Title
Estimates of 95/6 Upper-Limit Slopes for Three Drinking
Water Studies, By Sex	   13-160

Relative Carcinogenic Potencies Among 54 Chemicals
Evaluated by the Carcinogen Assessment Group as  Suspect
Human Carcinogens	   13-165

Carcinogenicity of Acrylonitrile in Rats	.   13-170

Epidemiologic Studies Reviewed in Acrylonitrile
Risk Assessment	   13-172

Estimates of Low-Dose Risk from Both Male and Female
Sprague-Dawley Rats from the Dow Chemical Company (Quast
et al., 1980b) Inhalation Study Derived from Four
Different Models	   13-181

Estimates of Low-Dose Risk from Both Male and Female
Sprague-Dawley Rats from the Biodynamics, Inc. (1980a)
Inhalation Study Derived from Four Different Models	   13-182

Estimates of Low-Dose Risk from Both Male and Female
Sprague-Dawley Rats from the Dow Chemical Company (Quast
et al., 1980b) Drinking Water Study Derived from Four
Different Models	   13-183
Estimates of Low-Dose Risk from Both Male and Female
Sprague-Dawley Rats from the Biodynamics, Inc. (1980a)
Drinking Water Study Derived from Four Different Models.
                                                                      ,.  13-184
                                      XI V

-------
                                LIST OF FIGURES

No.          Title                                                         Page

5-1          Flow Diagram for Acrylonitrile Usage 	     5-5

6-1A,B,C     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  ..   13-30

13-4         Effect of Sodium Thiosulfate on the Distribution of
               Acrylonitrile, Cyanide,  and Thiosulfate 	   13-31

13-5         Distribution of Acrylonitrile, Cyanide,  and Thiocyanate
               in the Blood after a Single Injection  of Acrylonitrile
               (Rabbit)	   13-35

13-6         Effect of L-Cysteine on the Blood Concentrations of
               Acrylonitrile, Cyanide,  and Thiocyanate (Rabbit)  	    13-36

13-7         Histogram Representing Frequency Distribution of the
               Potency Indices of 54 Suspect Carcinogens Evaluated
               by the Carcinogen Assessment Group 	   13-164
                                      xv

-------
     The EPA Office of Health arid Environmental Assessment  (OHEA) is responsible

for the preparation of  this  health assessment  document.  The OHEA Environmental

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 physical 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

                    Dr.  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 Assessment  Group (GAG) was responsible for preparation

of the sections on carcinogenicity.  Participating members  of  the GAG are listed

below (principal authors of present  carcinogenicity materials are designated by

*).
                                      XVI

-------
                   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.
                   Dhann 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  author  of present mutagenicity  section  is  indicated by

*).
                    Vicki Vaughan-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.
     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.
                     ffice of Health and Environmental Assessment
                     vironmental Criteria and Assessment Office
                                     XVI I

-------
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 Toxic Substances
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

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
                 XV! I I

-------
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 Reviews

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 the Holy Cross
Worcester, Massachusetts

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
                  XIX

-------
Gary Williams, Ph.D.
Associate Director
Naylor Dana Institute for Disease Prevention
American Health Foundation
Valhalla, New York
                  xx

-------
                          1.  SUMMARY AND CONCLUSIONS


     Acrylonitrile is  a clear, colorless, and highly flammable liquid that has an


unpleasant and irritating odor.  The boiling point of acrylonitrile is  77.3°C;
                 4

the melting point is -83.55°C,  and  the density  of the  liquid  at  20°C is  0.8050.


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 M% by  volume in  air  at 25°C.


Synonyms for  acrylonitrile  include 2-propenenitrile,  cyanoethylerie,  and vinyl


cyanide.  Aerylonitrile has a molecular weight  of 53-06  and a molecular  formula


of C H_N.  The structural formula is  given below.



                                 5^>C=C—C=N
                                  H    I
                                       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 11% (664,000 Mg)  will  be used domestically;  the


remainder is exported.  Acrylonitrile is  used primarily  as a raw material in  the


synthesis  of acrylic  and  modacrylic  fibers,   acrylonitrile-butadiene-styrene


(ABS)  and  styrene-acrylonitrile  (SAN)  resins,  adiponitrile,  acrylamide,   and


barrier resins.  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  acrylic


fiber production facilities emit 802 Mg, 1424 Mg, and  1276 Mg, respectively, of


acrylonitrile in the atmosphere per year. The atmospheric half-life of acryloni-
                                      1-1

-------
 trile 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,
                                                             V

 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.  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 ACN  production  and  polymer


 manufacturing plants were  investigated by  Hughes and Horn (1977),  Going  et al.

 (1979),  and Howie (Pedco)  (1982).


      In  general, acrylonitrile 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  4300  \ig/I


 detected in some samples  near wastewater discharge points.   Acrylonitrile has


 also  been detected  in  drinking water, although  the levels  were  not quantified.


 The inhalation exposure of acrylonitrile in the vicinity of  a plant site esti-


mated  by dispersion modelling does not  agree with the experimental monitoring


data  obtained  from the same site.   There are  insufficient data  with which to


determine the human intake of acrylonitrile through food  and  drinking  water.
                                      1-2

-------
     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;  nitrification  of




ammonia would follow, producing nitrogen.



     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 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.   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
                                       1-3

-------
 larger doses, 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 thiocyanate and  by  cyanoethylation of  sulf-



 hydryl 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,



 nausea,  and  convulsions.   The TLV  of acrylonitrile is 4.5 mg/m3  (2 ppm) for



 humans.  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 LD50  for acrylonitrile



 is  between 80 and 113 mg/kg.  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  non-



glandular portion  of the stomach,  proliferation of glial cells in the brain, and



mammary gland hyperplasia in females.
                                      1-4

-------
     AcryIonitrlie adversely affected pip survival following exposure of preg-



nant 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.   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 acrylonitrile for 6 hours a day  on days 6 to  15 of



gestation had  no statistically  significant changes in  reproductive  success or




fetal development.   Only the pups of rats  administered acrylonitrile  per  os




(65 mg/kg) for days 6 to 15 of gestation  had  an  increase  in malformations.  This




increase was in total malformations, with no  statistically significant increase



occurring in any single malformation.  It was concluded that these fetal abnor-




malities 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,  embryotoxic,   and  reproductive



effects, the limitations of the available data do not allow for a full assessment




of  these effects.



      There is  evidence that acrylonitrile and an epoxide metabolite causes point



mutations  in  bacterial  test  systems, and  there  is suggestive  evidence  that



acrylonitrile  may produce a positive response in the  sex-linked recessive lethal



mutation assay in Pro sop hi la  melanogaster (Benes  and Sram, 1969 ).  Chromosomal



damage  was not detected in plants or whole animals  when treated with  acrylo-



nitrile.   In  vitro DNA  binding   studies  indicate  that acrylonitrile  in  the



presence of a  rat liver activation system  as  well  as its epoxide metabolite,



2-cyanoethylene  oxide binds DNA.    Acrylonitrile  also induces  sister chromatid



exchange in  Chinese hamster ovary (CHO) cells but  requires metabolic activation.
                                       1-5

-------
 From the data base available,  it appears that acrylonitrile may have  the poten-
 tial to bind to DNA and cause genetic  damage.  If  the pharmacokinetics of this
 chemical substance in humans results in  metabolic products that can  interact with
 DNA, it also may cause  somatic mutations in  humans.
      Acrylonitrile is not a direct acting  carcinogen; 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 bio-
 assays  in  rats:  four in drinking water,  one by gastric intubation, and  two by
 inhalation.  In addition,  10 epidemiologic studies of cancer incidence have been
 reported.   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 rats  was
                                      1-6

-------
further confirmed in four other studies:   a three-generation  reproduction  study
performed  at Litton-Bionetics  by Beliles et  al.  (1980);  three  studies by
Biodynamics Inc.  (1980a, b,c)  in which acrylonitrile was administered in drinking
water  and  via gastric  intubation;  and  an inhalation  study by  Quast e_t  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/day,  5  days/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 exposure
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 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  methodology,
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.,  1980; Werner and Carter,  1981),
the  problems were sufficient  to  cast doubt  on   the  finding of significantly
                                      1-7

-------
 elevated risks of lung cancer  reported  in each  study.   In the  fourth  study by
 O'Berg, the problems were insufficient to obscure the significant finding of lung;
 cancer.  After adjusting for latent factors and  evaluating the contribution due
 to smoking,  the  finding of a  statistically significant 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  signifi-
 cantly elevated risk of lymph  system cancer  found  in the Thiess e_t al.  (1981)
 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 carcinogenicity  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."
      To provide a rough estimate  of  the potency  of acrylonitrile  relative to
other chemicals and a crude  indication of population  risks associated with known
exposure, unit risk estimates have been  calculated.  The unit cancer  risk for air
 is defined as the  lifetime cancer  risks  occurring if an individual is exposed to
air containing  1  [ig/m3 continuously for a lifetime.  The  linear non-threshold
dose extrapolation model has been  used to give a rough but plausible upper-bound
                                      1-8

-------
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  limitations 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,  1980a,b) and two based on rat cancer bioassays (Quast




e_t  al.,  1980a,b).   The  upper-bound  lifetime  risk of cancer  associated  with




lifetime  inhalation  exposure of 1 ng/irP 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




study is  1 .5 x 10"  (or  2.6  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 relatively



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 \ig/i  of acrylonitrile in drinking water



is 1.5  x  10, based  on the mean value of three drinking water studies in rats.



     There is evidence that acrylonitrile is a human carcinogen.  This conclusion



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 chromatid  exchange  tests;  4) in vitro



evidence  of  interaction of acrylonitrile and/or its metabolites with  DNA; and 5)



aeryIonitrile's  structural similarity  to  vinyl  chloride,  a  known  animal and




human carcinogen.
                                       1-9

-------
     The carcinogenic potency of acrylonitrile is in the third quartile among 54



suspected carcinogens evaluated by the Carcinogen Assessment  Group.



     Using the IARC classification 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.




1.2  RESEARCH NEEDS



     The present data base  from human and toxicologic studies provides  enough



evidence such that  the  International  Agency  for  Research on Cancer  (IARC) has



characterized acrylonitrile as an animal carcinogen and a likely  human carcino-



gen.  Unlike the animal bioassay data,  the human data base does not unequivocally



demonstrate a causal association.   In  addition to the human data,  there are also



limitations in the available animal bioassay data within  the  areas of reproduc-



tion and genotoxicity, which  have an  important bearing on  both  the qualitative



and quantitative aspects of carcinogenicity.



     The highest priority  with respect to future  recommendations  for research



should be placed on well-designed epidemiology studies, (i.e., case referent or



more vigorously designed historic and  prospective  studies  that will adequately



sort out and control the effects of smoking).  Careful workplace  exposure moni-



toring should be routinely conducted on members of  the  study population in order



to adequately determine acrylonitrile dosage.   In the opinion  of  some  expert



epidemiologists, the available  cohort  identified  in  the  O'Berg  study involving



DuPont acrylonitrile workers in Camden, S.C.  should be  followed  up  as long as



possible  to determine  if  a  significant  elevated risk  remains  for  all  site-



specific cancers as well as  for  the  higher frequency of  lung and  intestinal



(colon) cancer.  In connection  with this  follow-up study,  it would be advanta-



geous  to collect  information  on  histopathology  of  lung and brain  tumors to
                                      1-10

-------
correlate  with  the  reported absence of lung  tumors  and the presence  of brain



tumors  (glial cell) in animals.




     Although  several  studies have been  conducted  to evaluate  the  ability  of



acrylonitrile   to  cause  adverse  teratogenic,  embryotoxic  and  reproductive




effects, the limitations of the available  data do not allow for a full assessment




of  these  effects.   Any  new studies that are conducted should be designed  to



demonstrate the possibility of a dose-response relationship.




     Additional  carcinogenesis testing is  recommended using  mice  exposed  to



acrylonitrile by the oral and  inhalation  routes  of  administration,  because the




available  animal  bioassay data are  limited  to  rats.  Acrylonitrile  carcino-



genesis could also be evaluated using in vitro cell transformation systems,  such




as BALB/c  3TC cell  lines,  Syrian  hamster embryo cell  cultures,  rat  hepatocyte




cell strains, and retrovirus-infected rat  embryo cells.  An understanding of the




mechanism  of acrylonitrile carcinogenesis  (i.e.,,  genetic versus  epigenetic)



would be enhanced by pharmacokinetic and macromolecular binding studies in  mice



and rats.  Adequate pharmacokinetic data are presently available only in the  rat.



In  addition,  the selection  of a  model  animal species for comparison  pf  the



effects of acrylonitrile to humans would be aided by comparative in vitro metabo-



lism studies  using target  organ  homogenates  from  several species,  including



humans.




     With  respect to acrylonitrile mutagenicity,  further testing is needed  in



eucaryotic  organisms,  other  than Drosophila melanogaster,  to  confirm  that



acrylonitrile is genotoxic  as indicated  by  the  positive  response observed  in



bacteria.  An assessment of genetic risk with respect to germ cell mutagenicity



cannot presently be made because  of the lack  of appropriate data.
                                     1-11

-------

-------
                               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 acrylonitrile under



Section 112  of  the  Clean Air Act.  However,  based on the expressed interest of




other agency offices,  the scope of this document was expanded to address acrylo-



nitrile in relation to  sectors of the environment outside of air.  It is fully




expected that this document will serve the information needs of many government




agencies and private groups  that may  be  involved  in decision-making activities




related to acrylonitrile.



     In the development of the assessment document, existing scientific litera-




ture has been surveyed  in detail.  Key studies  have been evaluated and summary



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 effect  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 appropriate, 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; 2824




     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-54  are  no




longer manufactured in the United States.




3.2  STRUCTURAL AND MOLECULAR FORMULAS AND MOLECULAR WEIGHT





          H""C=C-C=N

          H'  H




          C H-N     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,  1954):




C-H:  1.09  A; C-C:  1.46 A;  C=C: 1.38 A; and C =  N: 1.16 A.




3.4  PHYSICAL PROPERTIES




3-4.1  Description




     Acrylonitrile is a clear, colorless, and highly flammable liquid  that has an
                                                                   V



unpleasant and irritating characteristic odor (Fassett,  1963).




3.4.2  Boiling Point




     77.3°C at 1 atmosphere (Groet, 1978).




3.4.3  Melting Point




     -83.55°C (Groet, 1978).
                                      3-1

-------
3.4.4  Density
     dj:0(liquid) :  0.8060 g/ml;  vapor density:  1.83 (air=l)  (Groet,  1978).
3.4.5  Refractive Index
     nfj° : 1.3911 (Weast, 1976).
3.4.6  Spectroscopic Data
     The  X    is at 203  run with  a molar extinction  coefficient  of 6100;  a
compilation of  infrared, Raman, NMR,  and mass spectral  data is available  in
Grasselli and Ritchey (1975).
3.4.7  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?
               40°C:     7.9  weight?
               60°C:     9.1  weight?
3.4.8  Volatility in Water
     Henry's law constant:  0.063 at 25°C (Bocek,  1976).  Partial vapor pressure
(water azeotrope):   log P  =  7.518 - ^r^'1\   i.e., 80 ram  at  20°C  (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 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.
                                      3-2

-------
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


               45.5°C:  249.8


               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): -1°C


(Patterson et al., 1976), -4.4°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).


3.4.12  Conversion Factor

                             •3
     1 ppm in air = 2.17 mg/mj  at 25°C.


3.5  CHEMICAL PROPERTIES


3.5.1  Reactivity


     Undergoes reactions at both the  nitrile group  and  the double  bond  (Maltoni


et al.,  1977).   Some  of these  reactions  are used  for  the quantification of


acrylonitrile and have been discussed in Section 4.  Acrylonitrile also undergoes


the following reactions (Miller and Villaume, 1978), many of  which  are important


pommercially.


3.5.2  Polymerization


     Polymerization, forming high molecular weight  products, is  the most  impor-


tant commercial reaction of acrylonitrile.   The polymerization usually  requires
                                      3-3

-------
the presence of free  radical initiators, such as peroxydisulfate.  Heat or light

          o
(X < 2900 A) can also initiate polymerization reactions.  Oxygen and methylhydro-


quinone 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 methacrylonitrile


are the so-called barrier resins.


3«5.3  Reaction at the Nitrile Groups


     Acrylonitrile when  reacted  with 84.5/6 HpSOj. at  100°C produces acrylamide


sulfate, which yields acrylamide upon neutralization:
          CH^CHCN + H20
CH2=CHCONH2
     CH2=CHCONH2
CH2=CHCONH2
                    [2S04 + 2NaOH


     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  acrylonitrile  is


heated with less concentrated HJSOj. 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 10~3 min"1 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 HJ3CK produces


esters of acrylic acid.   With olefins, it forms N-substituted acrylamides in the


presence of concentrated
                                      3-4

-------
3.5.4  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:
                                            •CH,
     CH2=CHCN + CH2=CH-CH=CH2
                        HC'

                         II
                        HC
•CHCN
                                             CH;
                                        A-3-tetrahydrobenzonitrile


     In the presence of  catalyst,  acrylonitrile can be hydrogenated to propio-


nitrile, which can be further hydrogenated to n-propylamine:
                         H,
                           H,
               CH2=CHCN
2CH2=CHCN
     It can also be  reduced,  in the presence of magnesium and methanol, in the


following manner to produce adiponitrile:


                                H + Mg - >  2CH2CH2CN  H- (OB^CO^g


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


and their salts  (Miller 'and  Villaume, 1978).  The generalized  reaction can be


written:
                    CH2=CHCN + AH
                           ACH2CH2CN
                                      3-5

-------
     The cyanoethylation of pseudouridine,  inosine,  and 4-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

-------
                      4.  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;  (4)  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.



4.1  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
                                      4-1

-------
satisfactory  for ambient air samples,  even  with the most  sensitive  method of
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  4-1  lists the  different  sorbents and trapping media  used  for
collection of acrylonitrile in air.  Activated carbon, silica,  and porous poly-
mers 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

-------














^

•
e
•H
CD
•H
•H
C
O
£,

|
.
O
c
o
11)
**";


o

5
cd
CD
s
s

a
2
H

T3
S
to
43
I
5
CO
eg.
o
•*H
c3
>



|
J3-
_j
3
cd
EH















Reference






§
•H
$43
0) -H
43 a
CD -H
« J



§
•H

2 O*
E iH
•H C
^ O
5 co
O EH






CD
m 3
•H O1
So
ra o
C CD
•=t EH







C
0
•iH
43 T3
0 5
1*






c
bO 0
C -H
•H 43
rH 'H
O, T3
18
CO 0







43
C
J8
5
CO

CD
f
co


0
t—

^~
Krynska,










i
i
i






CD

Z




O
•(H


01
a

0
•3
o










5
43
3!




i.
O
HH
(0
a
a
rH rH
52









OJ
0
•H
CO

i


•=*•

o^
Korzhova
et al., 1










i
i






0)
c
o
z










o
ft.
8




O
O
CM
ij
CO

1
1
EH











1





43
(0
CM

§
f u
•H in
co i

Plastic
producing
plant


•a

43
Sakurai e
1978







s
o,
o.

•^
o







o» *>o
a oj
»- o
V V






CD CD
c a
O 0
z z










Q O
1 1
CJ O
Q a




0
§

HP
13

cd cd
g g
CD d)
EH EH






0 rH
*"* in H
a *— in
0. ' 1
*- rH 0 CO
^in 4&
• i .0
O CO O IH







Z Z
a a
cd cd
cS fe
o-, OK

Simulated air
Simulated air




•a c—
m
5 c"
B £0








bO
C
0







 "^
ox a o






CD CD
C B
O 0
z z




o
•H
.c
o,
cd
L
a
Q O
HH t*
Cu S
1 rH
O OK










Bm !
5° !

rH
0
1
o


Q


a a
o. E
T— T—





03
rH
CD O
a i
1 «
s S^
O cr\

•H
a
Js i
O CO


-------


















^
l\
B
O
fj


f,
•H
C
•H
CD
rH
*J^
43
ti
O
1
o
s
o
3
0
to
rH
•3
U
i.
«
3
CO
a:
to
B
Q.
ca
L.
EH
•o
m
S
B
S
g
co
CO
1
co
•^"
•
J

^
0)
£
m
£•«




























i
s
a>
«







B
O
^
O fl

a j


B
o
43 CD
cd 3
0 S
0 EH


0}
til
B CO
«S EH






B

43 "O
O. Q
£. £

I*




bflg
•iH 43

l§
CO CJ











43
B

.a
O
CO







CO
cl
1
CO






CM
UD

CO
t,
CO
cq



£

in
I
0
^
CO
o

«—






a)
§
z


o
•H
•s
cd
5
Polaro










|
!



a
rH
CO
CO
^
J
m
o
o








Id


13
rH
rH
O


43
B

•H
§
 a\

a
0.
0.

o
o

o

CO
a

K






1
z




3










|
!

rH
0

^
Cw

rH
^








15


T3
CO
rH

CJ




rH
1
43
cd
D.
3
85
o cd







•a

CO
(D
SjPal
£C -








XI
a,
0.
o
in






CO
§




a
M
CC4










|
!







rH
r—






SH
CD
Id


•a
CO
rH

CJ



CO
rH
-H

0
rH 43
<3 rH
o! D.




CO
[^

*~

S
CO
9
CB







°°a
"fa^
in

o






CD
1



0
43
CD
Colorm










|
!







i













t_
tD
I








,
1

4-4






•o in

i}
B-H
rH
O OQ







i
O
o
T
o
CM






CO
§
z



o
t
•8
Titrim










1
i







rH
r—








•S
•O CO

cd %

O m
CO Cd
CMH
SB M








1
1
1


t—

a\

O CM »
VO t~ O O
CT\ CJ^ C~- *O rH

•* B •* *O
> a n - « t,
CD & CO 43 CO
B O CO B -t> -H
Cd 43 43 *H -r-t O
.Q Cd 43 ^ CO CO
(0 £4 *H Cd CO rH
QQ *J! DQ PQ Cz4 J







rH CO
^ "5b
a
in
oo i in i
• I • I
O 1 O 1






CO CO CD CD
B B B B
O O O O
z z z z



o o
•s ti
•8 ti
ga 00
« H M
O O |vt jv,
o o 08










II II
II II
II II



S
a
rH
CO
O
1 1 -M
o




-
cd

rH
0


O O 43
CO CO CO
CM CM B

MM. 2 in in
«- «- « CM i-




43
g
"o. B

tt rH
CO PL,
J2 | |
•S ' z i
3 | -rf |
P3 CO

















































£
o
43
8
43
CO
"0
B
O
vH
43
a
g
o
•iH
CO
a
,_i



-------
      For air nearly saturated with moisture, the breakthrough volume for acrylo-




nitrile  in Porapax N  (4" x 1/4"  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 the 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 0.404 g/g and 116.0




min." , respectively.  With acrylonitrile that contained 50% relative humidity,



Nelson and Harder (1974) determined  the  adsorption capacity to be 0.174 g/g.  The




breakthrough  times  for 1%,  10%,  and 99$  passage from  a 58 g  charcoal  in  a



respirator cartridge were determined  to be  48.5  min.,  61.1 min., and 168 min.,



respectively  (Nelson and Harder, 1974).  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 et 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 ± 9.4/5 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 were 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
                                      4-5

-------
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 et  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% HJSOj,,  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 et al.
(1979)f who demonstrated that acrylonitrile in carbon disulfide remained stable
for over 12 days when stored at 3°C.
4.1.2  Analysis
4.1.2.1  Pretreatment
     Acrylonitrile collected  on  solid  sorbents requires a desorption procedure
before  identification  and quantification.    Thermal  desorption and  solvent
                                      4-6

-------
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  chroma to graph.   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.

-------
     The  recovery efficiencies  of  acrylonitrile  from  activated  carbon  with




various solvents are given in Table 4-2.







         Table 4-2.  Recovery of Acrylonitrile from Various Solvents
Solvent
% Recovery
Reference

Methanol
Acetone
2% acetone in CSp
2% acetone in CS?
CS2 (2 ml)
CS2 (4 ml)
ca. 50?
73.5 ± 5.3?
95.5 + 7.9%
94?
58?
75?
Going et al., 1979
Marano £t ail . , 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 |ig to 200  |o.g acrylo-



nitrile, Going et al. (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.



4.1.2.2  Identification and Quantification



     The methods  utilized  for the  identification and quantification  of acrylo--



nitrile collected by sorbent or trapping techniques are shown in Table 4-1 (see



4.1.1).  The methods for the analysis of acrylonitrile collected without precon--



centration appear in Table  4-3.  The infrared techniques are used exclusively for



samples that need  no preconcentration.  Although a number of methods including
                                      4-8

-------
1















0
rH
•H
3
°
<*
0
"^
CM
0
W
"I
*5
S-J

4-5
m
S-,
*Q
•
i*
'
~^

Q/
*H
rrt
*"























1

co
0
o
§
i



G
O
•H
-P -P
0 -H
0 a
.p -H
0 J
o

c
0
go-
•H G
s
*~3 CO




a
a
a
CM
o



0

o





0
-P
1

Q
O
o
&
1
W
H


Q«
3
3
a
•o

Co
0)
,Q
o

0)
, — 1
a

CD
CO



1
CO
§•
o
o
o

•O -H
0 cO
"31 G
CO 4J


O
t—

S3
M





a
a
a
in
o



0
G
0











K
M




5_,
3 O
f. . j^
O -H rH
C bO CD
o cd o
•rH jQ
O W 0

O 3
§o
cd
G SH >
M 
0 CO
CO H



a

a
en
o
•
0



0
G
0
&










lasers
K
M


j_
i>
o
cO "**
oo
o

** "cd
G
0 rH
•H rH
% §
3
•o a
O 3
t. 3
-P O
G cd
M >




C
•H
0
rH
•H
SL,
4J
•H
§
£
S_
0 CM
«a! £5


^
•i-i

§a
w a



0
G
0







5
3

tector
0
o





o>
4J
0

.p
O
i.
T3
§

Q*
s
3
DH





£H
3

•o
0
•s
•3
5
co

„
^1

[ "^
0
4-5 CO
0 t-
XH O>
0 -






i
i





^?
o
o







&
3
4->
tector
0
Q











•H
CO
0
o
C.
0


o
>>-H
fQ i ^
•o cd
0 rH
•8 a
SH CO
0 PQ
G <
0
bOg,
8'S
co cd
co 0
0 43
C—
'-
•H
0
-P
GO
SH
£
r-j
CO

1
co
•
O

aa
ro a
K a



0
G
O




s

ry*|
Q
«a)
PQ
9
o
0
J3

a »-




>o
a
a
o
0
*~





i










Q
CL,
O
O





G
0
•H
"o
0
•o
G
•H
4->
O
0
S^!
•H
Q





•rH
CO

•o
0
to
•3
•3
CO


o
C—
CT\
M
•at




a
a
a
in
•
o



0
G
a










Q
M
1
o
C3





G
0
•H
O
0
"r"3
C
•H
4J
0
0
S-4
•H
«




•3
r-H
cd
G
o
4J
CO
a
3
o
o
0




























CO
o
4J
o
0
.P
0
•o
§
£_,
>> o
SH .^"i
• i QI
0 10
S ^
O JM
t, a
0 ^
0 0
a bo
co o
to -P
CO *iH
a V
i >>
fcj O<
D, cd
cd £-1
EH bO
bO O
O 4-5
-P cd
cd S
Q O
I'll
X! O
W S.
cd &0
bO
H
"o
g §
o o
o o






























co
S-,
detecto

G
O
•H

crt
CN
1
•H

0
co
rH
"V
>.
04
cd
£.
^a S)
t, o
1 ^ j i ^
0 cO
a a
o o
ft
a co
CO CO
bO
CO
S "
a Q
M
1
CO O
S O
4-9

-------
colorimetrio,  titrimetric,  and  polarographio were  used for  the analysis  of
aorylonitrile 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/phosphorus 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 ek al.,  1979), SP-2100  (Grote  et  al.,  1978)  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 4.7.
4.1.3  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  24 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
                                     4-10

-------
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 et  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 e_t 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 e_t al., 1979).  In a few
instances, multiple  grab  samples were composited  for analysis  (Going _et al.,
1979).  Proposed EPA methods require that the  samples be adjusted- to pH 7*0 ±  0.5
and be collected  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,'
1979 b;  U.S.  EPA,  1980a).   The  ASTM  (1980) procedure allows  headspace,  but
                                      4-11

-------
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  <_ 4 (Going est al.,  1979).  The  samples
should be maintained at 0 to 4°C by ice or an ice substitute during transporta-
tion (Going et al., 1979; Kopf ler 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.
4.2.2  Analysis
4.2.2.1  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
(Kopf ler et al.,  1976;  Going  et  al., 1979), distillation (Going e_t  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 Kopf ler e_t al.  (1976),.
The system used by Kopfler e_t al. (1976) appears to be preferable to  the system
used by Going et al. (1979 ), because the former allows  purging  of  140 ml of water
                                      4-12

-------
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 al.  (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 azeo-
tropic  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 organic of  interest is collected in
the  distillate.    Peters   (1980)  has  used  this  technique  to  concentrate
acrylonitrile 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 18N HpSOj. and the content was distilled  at a  rate of 1  ml/min.
The first 10 ml  of the distillate was collected for subsequent analysis.  The
                                      4-13

-------
percent recovery was about 90% 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 this  system  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 past  had faced  problems  owing to the diffi-



culty 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  con-



ditions will not produce  the same  equilibrium vapor pressure.  This problem  of a



calibration curve has been largely 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.



4.2.2.2  Detection and Quantification



     With  the exception  of  one  case  in which a  titrimetric method  was  used



(Wronski and  Zbigniew,  1974),  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 e_t al.,  1978) were used,  although Going _et 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  et al.,  1979;  Kopfler et  al., 1976;



Federal Register, 1979).   Other GC columns including carbowax  1500  and Tenax GC



have been  used for  the  separation  of acrylonitrile   (Markelov  eit al., 1981;
                                     4-14

-------
Peters, 1980).  An  aliquot  of the distillate from  azeotropic  distillation  was
injected directly into the GC column.  In the purge-trap technique, the acryloni-
trile from the trapping column was thermally desorbed onto the separating column.
A  summary  of water  analysis  techniques   is  given  in  Table  4-4.   Although
nitrogen/phosphorus detectors (including the Hall detector) and flame ionization
detectors  were used  for  quantification, the former detectors (nitrogen/phos-
phorous)  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.
4.2.3  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.
4.3 WASTEWATER
4.3.1  Sampling
     No  details  regarding  the sampling of  wastewaters  are available.   Grab
samples  may be suitable in certain cases.   To  monitor  the discharges that are
dependent on  process  operation  stages,  a 24-hour composite sample  is preferable.
4.3.2  Analysis
4.3.2.1   Pre trea tment
     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,
                                      4-15

-------













io
3:
a
•H
CD
•H
ti
o
^
1
3
a
3
C
<"!
*
j=r
,=r
0)
2!
m















0)
o
§
§5
S




c
o
O -H
CO J
Q



0 (D
cd o*
g -d
il


CO (D
•H 3
to o*
>> tH
CO (£4
c o
•*£
1
O O
Q) *C
O


0)
•cL
CO



CO
e-

<-
73
J ^
Q)
o
§
S




ft
a.
o







§

g~
•H
4-3
O
a>
"2
•H
•8S
a) S
S- 1
•H O
Q 0

1
1


t,
CO
i§
»
T3
JJ
•H
a,
CO
VO
t^.
o>

..
i
•s
c3
rH






Q)
g

X
cd
§
E-i
D.CQ
cd S
So
1 0
?-
3 O
O-i O

1
0

i.
co
.p
cd
bo
C
^
C
ti
o


cr»
t—
cn
•0 ^ Hi
a col
3 -
•H S "co
•^ -H
gs s1
SS -d
& ISI O
a
cL

M<
O
° &
LO O.
o »-





CO
c to
o s
A
J3 £3*
-P cH O
•r) O -H
* § "0
o «  I 40
Q) Q) 1 Q)
1 1 1
O C5 O
A

QJ O C >>
a o o o
& -H C
x> on -p co
0, • T3 D. -H
rv t- Q) t4 0
1 S O -H
0 0 CO "O ?
                                   co cd -P cd a
                                   §1 S§|
4-16

-------
ether, and isobutyl acetate were used (Ponomarev et_ al., 1974).   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  4-5.  Table 4-5 also lists the different detection




and  quantification methods and  their  detection  limits  where available.   The




principles of the detection methods are discussed in subsections 4.2.2.2 and 4.7.




4.3.3  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.



4.4  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 jet al. (1979 ).



4.4.1  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.



4.4.2  Analysis



4.4.2.1  Pre treatment



     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
                                     4-17

-------


CO
JL,
0
Id
vu
3
4j
co
3:
C
•H
0
rH
•H
i,
s of Acrylonil
•H
P>
"nj
•a!

0
ITk

zy

Q)

»o
cd
H








1









; Reference
•4-»
§
•H
•8
0
-p
0
(-1



4J
g
vy
4J
Analysis trea













1
1
cd
0
M
f
Ok










0
rH
0.
m
CO

Sevest'yanova
et al., 1966

pt
ex
o









Polarographic















1
'











I
1

Jl
Ponomarev et
1974

D*
0,
in
on








Polarographic







c
o
•H
4J
O
cd
0
4J
C
0)
1
co











1
1
1
0
a
33
TO
§
CO
0 C--
2 ^
(o CTi
Q '-
S
D,
O<
^
O
V







Polarographic



§
•H
4-5
cd
rH
rH
•H

CO
•H
•o
0
•d,
2
4-3
o
0
IS]











1
1
1
•o
Stefanescu an
U.rsu, 1973

f^f
cu
CM
v









Titrimetric



g
•H
4->
cd
rH
rH

j_3
co
•H
•o
O
•d,
o
4J
0
0
IS]











1
1
1
t-
t-
Lawniczak, 19

f^_
D,
in
o
V







Colorimetric
















0
S











I
|
1

Covic-Horvat
jt al., 1970



i









Titrimetric



g
•H

cd
rH
, — |
•H

CO
•H
•a
o
•H
o.
o
4^
o
N











1
1

•a
•P ^.
•H r-
co
1
c, o
3 o
0 >
« co

5
o.
r—
O








Q
H
O
O
















0
g
a


c
o
•H
S
•H
§3
H
r-l
a CO
co
8 0
5 o
^ i,
fV0 fSj
















o
D
0
3
i
5
H

gj
3
,
~t
3
H
D
H
1
d
&
o
,3
zi
3
^
o

1
60
|
3
H

i
i
4-18

-------
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 4.2.2.1.  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.

4.4.2.2  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 400 jig/kg.   For the purge-trap technique, the detection

limit was ca. 0.5 |ig/kg.

4.4.3  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.

4.5  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.
                                      4-19

-------
4.5.1  Analysis
4.5.1.1  Pretreatment
4.5.1.1.1  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 4-6.   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  (Steichen,  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
(Steichen,  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 concentration
must be related  to  the original concentration in the polymer either by assuming
100?    diffusion   of  the  monomer   into   the   head-space   or   through
                                      4-20

-------
n
-p


I
rH
O
co
•H

(0
T3

8
6-.
S
n
•3
CH
due


8
c
0)
S
CM
IV
PS



o
S«
0 -rH
ss
0> J
a
g
•H a)
•P 3
(0 cr
g -H

**-4 C)
C 0)



CD
3&
Rti
4l
C 0)
«< t-1


4^
S
B
g
4J
OJ
t,

H
                                                                    o
                                                                        o
                     1
                         g
                         4S
                         8-3


                                              i
                                                    8-3


                                                                              1*-
-------
-7

0
o
^^
n
43
c
>
*rJ
co
to
c
*r1
43
id
CO
1
8
b
•a
§
m
i,
a>
S
•3
cu
c
•H


•o
**"*
S
K
0)
43


o


M
Jj*

o
CO
1-t

^»

5
0)
a)
EH













o
o
c
> c
H -C
id o
C Q)
< EH




43
C
0)
43
id
0)
£,
43

ti
CU








(U
f
CO
VO
C—


•»
g
S
o
•S
43
CO






g.
O.
in
0







0)
c
0
2










Q
M

8




1
isc
C  iH Id J3
01 -H 'H
m -a -a 3
•H i C O*
a o id ID








3
H
l-l
0
a.


S

^~

rH
3
(d
C3






1
0.
•
O







>T3
&s
Q. 0)
C1d §
•ri § oH
T3 ^3 O 18
a> ^ id s-i
> 3 a. .a
rH O CO .H
01 0) T3 -rl
01 C Id 3
•H a) a) o1
Q r-l J3 0)








o
•H
43
1-1
CM


o\
t-
o>


jj
s
s
C3






0,
o.
in








01 >
C> (1)
ca o

































m
i,
0
43
O
0)
43

•a
ra
S3
43 t,
O O
0) 01
T3 0

^1
a) a
N O
•d ^
3 43
5ti
ii,
•a -a
id id
M fe
o o
43 43
id id
S S
II
0 ^
23 CQ
SIM

II It
Q O

u o
o Q
U-22

-------
determination of equilibrium concentration using Henry's law and the appropriate

partition coefficient.

     The solution head-space approach has been used with  a much wider range of

samples (Steichen, 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

(Steichen,  1976).  The advantages of this  approach are:   (a)  head-space equili-

bration 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 Tangier,  1968 ).

4.5.1.1.2  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  e_t 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 al.,  1978;  Hartshorn,  1975).
                                     4-23

-------
     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 Q% 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 et al., 1979).  When heptane was used as the food-simula-



ting solvent, it was extracted  with water and the  water extract was distilled



with methanol (McNeal _et al., 1979).  The  average  recovery of acrylonitrile 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 ETD.
                                      4-24

-------
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; Steichen,. 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
et al., 1979) and Porapak-QS and -S (Brown et al.,  1978) have been used.
4.6  OTHER MEDIA
     Aery Ion itrile  levels  have  been  determined in many  other media such as
tobacco, production  streams,  and foods and  grains  consumed by humans.   Since
acrylonitrile-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.
4.7  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.   The chemical
techniques that have proved useful are based on (a) hydrolysis of the nitrile
                                      4-25

-------
group and (b) additions to the double bond.  In  the  hydrolysis  method,  acrylo-
nitrile  is  hydrolyzed  to ammonia  and aorylate  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. radiation.  ,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 411 nm (Hall and Stevens, 1977).
     The titrimetric procedures have been  used following the reaction of acrylo-
nitrile with excess Na^SO- (Burkart  et  al.,  1961; Terent'ev and Obtemperanskaya,
1956; Taubinger,  1969), thioglycolic  acid  (Stefonescu  and Ursu,  1973;  Covic-
Horvat  et  al., 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  KMnOjj
solution, which produces  a change  in the permanganate color, has  been used to
determine the concentration of acrylonitrile.  In this method, the concentration
                                      4-26

-------
was determined by  comparison  of the color  from a calibration curve  (Gisclard

et al., 1958).

     In  the  thin  layer  chromatography (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 Sharma, 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,  1974,  cited  in Going _et,al., 1979).   The use  of nitrogen/phosphorus

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 phosphorus 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;
     Steichen,  1976;  Gawell,  1979) (b) Carbowax-1540 (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)
                                      4-27

-------
     Hydrocarbon:   (a) Apiezon (Babina, 1979) (b) Tween-80 (Chopra jet al., 1978)

     Esters and Polyesters:   (a) Polyethylene  glycol  adipate  (Panova et al.,
     1969;  Klesheheva jet al.,  1971; Markelov and Semenenko, 1976) (b) Neopentyl
     glycol succinate (Kleshcheva et  al.,  1971;  Korzhova  et al., 1974)  (c) Tri-
     ethylene glycol butyrate (Pokrovskaya and Frolova,  1969T  (d) Polyethylene
     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 et 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 104  (Going et al.,  1979) (h)
     Chromosorb 108 (McNeal jet al., 1979)

     Methylsilicone:  (a) SE  30  (Berck, 1965) (b) DC-200  (Beaumont and Garrido,
     1979)

     Aminoalcohol:  THEED (Hughes and Horn, 1977)

     Tetracyanoethyl:   Pentaerytritol  (Ustinovskaya  et  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.   Tetramethylammonium iodide  (Sevest'yanova and

Tomilov, 1963; Gorokhovskaya and Geller, 1962;  Uhde and Koehler, 1967;  Chao and

Ch'en, 1966;  Daues and Hamner, 1957; Lezovic and Singliar, 1977;  Mekhtiev et al.,

1968; Klyaev et  al.,  1966;  Rogaczewska,  1964),  tetramethylammonium  hydroxide

(Berck, 1962; Sevast'yanova jet al., 1966), and LiCl (Sevest'yanova and Tomilov,

1963; Bogaczek and Jaworski, 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
                                      4-28

-------
of  acrylonitrile determination  by  the earlier  IR  methods was  rather  poor
(Scheddel, 1958; Karaenev _et al., 1974).   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 et al., 1977;
AIHA, 1970; Beaumont and Garrido, 1979).  The same principle has been applied for
the continuous monitoring of acrylonitrile  at  10.5  nm 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 COp laser and a  40 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 nm
(Petrova _et al., 1978) and at 210 nm (Brieger et 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, 1974).   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.
                                     4-29

-------

-------
                        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 of Ohio  (SOHIO) and is known as

the SOHIO process.

5.2  ACRYLONITRILE PRODUCERS

     The producers  of  acrylonitrile  monomer  in the United States are given in

Table  5-1,  while  distribution  for  1980 and growth  projections  through  1984

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

DuPont
  Beaumont, TX

DuPont
  Memphis, TNa

Monsanto
  Alvin, TX
Monsanto
  Texas City, TX
Vistron
  Lima, OH
Vistron
  Victoria, TX
TOTAL
                               123


                               159


                               136


                               113


                               191


                               136


                               270

                             1,128
 Dupont 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)
Distribution
Acrylonitrile
(Mg x 10^)a
Projected Annual
     Growth
Consumption
Imports

Exports
     664

  Negligible

     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  modaorylic  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

alcohols and amines, fumigant formulations, as an absorbent, and as  an anti-stall
                                      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

-------
 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.

 5.4  CONSUMPTION OF ACRYLONITRILE BY PRODUCT

     A breakdowi  of acrylonitrile consumption and  projected growth  of  products

 using acrylonitrile in their manufacture is given in Table 5-4.


       Table 5-4.  Acrylonitrile Consumption and Project Growth of Products
                   Using Acrylonitrile
Product
Acrylic and
mod aery lie 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
Pro jec ted
Annual
Growth
Through
1982(!?)a
4.5-5.5
7 .5-9 .5
2.0-3.0
10.5-12.5
8.0-10.0
12.0
4 .0-6 .0
1980
Consumption
(Mg x 10d)D
345
172
—
—
—
•M^
147
Projected
Annual
Growth
Through
1984b
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

facility at Decatur,  Alabama,  produced 67 x 103 Mg of adiponitrile  in  1977 (SRI,

1978).
                                      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
I*.

MONOSODIUM GLUTAMATE
FLOCCULENT
SIZING PAPER
PLASTICS
THICKENING AGENT
                Figure 5-1.   Flow Diagram for Acrylonitrile Usage
                             (NIOSH,  1977c)
                                        5-5

-------
        Table 5-5.  Producers of SAN and ABS  Resins  (Grume,  1982)
     Producer
ABTEC (Mobay)
     Louisville, KY

Borg-Warner
     Ottawa, IL
     Washington, WV
     Port Bienville, MS

Dow
     Allyns Point, CT
     Midland, MI
     Pevley, MO
     Torrance, CA
     Iron town, OH

Monsanto
     Addyston, OH
     Muscatine, IA
     Springfield, MA

USS Chemical
     Scotts Bluff, LA
Capacity (Mg x 103)
          31.8


         105
         136
          NA


          29.5
          65.9
          29.5
          34.1
          29.5


         159
          56.8
          13.6


          90.9
NA = Not Available
   Table 5-6.  Producers of Acrylic and Modacrylic Fibers (Grume,  1982)
     Producer
Capacity (Mg x 103)
American Cyanamid
     Milton, PL

Badische
     Williams burg, VA

DuPont
     Camden, SC
     Waynesboro, VA

Tennessee Eastman
     Kingsport, TN

Monsanto
     Decatur, AL
          59.1

          36.8


          77.3
          61.4

          16.4

         145
                                      5-6

-------
  Table 5-7.  Producers of Nitrile Rubbers and Elastomers (Grume, 1982)
     Producer
Capacity (Mg x 103)
Copolymer Rubber
     Baton Rouge, LA

B.F. Goodrich
     Akron, OH
     Louisville, KY

Goodyear
     Akron, OH
     Houston, TX

Uniroyal
     Painesville, OH

Reichold
     Cheswold, DE
       6.2


      14.1
      28.6
       7.5
      18.2
      16.4


      10.4
             Table 5-8.  Producers of Acrylamide (SRI, 1978)
     Producers
                 •3
Capacity (Mg x 10°)
American Cyanamid
     Linden, NJ
     Westwego, LA
Dow
 Combined total for both plants
       36'
Midland, MI
Nalco
Garysville , LA
23
4.5
                                        5-7

-------
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
5-13, respectively.  (Other aery Ion itrile 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  informa-
tion.  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 (Grume, 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%.
5.5.2.    Emissions  During Transportation
     Estimates  concerning  the importance  of potential  acrylonitrile spills
during  transportation-related  accidents are presented  in  Table 5-14.   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
                                      5-8

-------
       Table 5-9.  Estimated Atmospheric Emissions of Acrylonitrile from
                   Monomer Production Facilities (Suta,  1982)
Producer
Acrylonitrile Emissions
      (Mg/year)
American Cyanamid
  Westwego, LA

DuPont
  Beaumont, TX
  Memphis, TN

Monsanto
  Alvin, TX
  Texas City, TX

Vistron
  Lima, OH
  Victoria, TX

  TOTAL
           94.7


           78.8
          216
           95.
          162
          132
           23.3

          802
                                          5-9

-------
     Table 5-10.   Estimated Acrylonitrile Emission Rates from ABS-SAN
                 Resin Production (Suta,  1982)
Producer
Acrylonitrile Emissions
            (Mg/year)
ABTEC (Mobay)
     Louisville, KY
Borg Warner
     Ottawa, IL
     Washington, WV
     Port Bienville, MS
Dow
     Allyns Point, CT
     Midland, MI
     Pevley, MO
     Torrance, CA
     Irontown, OH
Monsanto
     Addyston, OH
     Muscatine, IA
     Springfield, MA
USS Chemical
     Scotts Bluff, LA
     TOTAL
            4.5

           96.6
          657
            1.3

            7.9
           17.3
            1.1
            9.1
            6.2

           89.6
          365
           10.0

          158
        1,424
     Table 5-11.  Estimated Acrylonitrile Emission Rates from Acrylic
                 Fiber Production (Suta,  1982)
Producer
Acrylonitrile Emissions
       (Mg/year)
American1 Cyanamid
     Milton, PL
Badische
     Williamsburg, VA
DuPont
     Camden, SC
     Waynesboro, VA
Tennessee Eastman
     Kingsport, TN
Monsanto
     Decatur, AL

     TOTAL
          142

          357

          352
          309

           25.0

           91.0

        1,276
                                       5-10

-------
     Table 5-12.   Estimated Aerylonitrile Emission Rates from Nitrile
                    Elastomer Production (Suta,  1982)
Producer
                                             Aerylonitrile Emissions
                                                    (Mg/year)
Copolymer Rubber
     Baton Rouge, LA
B.F. Goodrich
     Akron, OH
     Louisville, KY
Goodyear
     Akron, OH
     Houston, TX
Uniroyal
     Painesville, OH
     Reichold Cheswold, DE

     TOTAL
  4.3

112
 63.4

 55.2
  0.2

 40.0
  8.4

295
   Table  5-13.  Estimated Aerylonitrile Emission Rate from Adiponitrile
                 Production (Suta,  1982)
Producer
                                             Aerylonitrile Emissions
                                                    (Mg/year)
Monsanto
     Decatur, AL
 59.0
                                  5-11

-------
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 (DIM-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  41  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 e_t 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.
     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 49 °C for  1  month  releases
0.013  ppm of acrylonitrile  to  the  acetic  acid.   ABS  resin (24 ppm  residual
                                      5-12

-------
             Table 5-14.  Hazards of Acrylonitrile Transportation'
                       (A.D. Little, Inc.,  1974 as cited in  .
                       Miller and Villaurae, 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 Exposed6
Urban/Rural

Expected Annual
Property Damage ($)
                                     61.0
                                    122
                                 46,700C
                                  5,460C
                                   8/92
                                      0.0117

                                      0.30
                 17.1

                 38.4

               4,450°



               23/77


                  0.063


                  0.25
    31.7
    68.3
13,400C
   27/73
    0.17
    0.40
                                0.008/0.004    0.010/0.002     0.16/0.016
Urban/Rural
f
Recurrence Interval
(Years )
129/55

85.5
160/20

15.8
2423/252

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 (rrR  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.
0
  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 Reference
(ppm)

Acrylic and Modacrylic
Fiber
Hycar
Kralastic and Paracril
UCAR-380
UCAR-4358
Acrylamide Monomer
Polyacrylamide
ABS Resin
SAN Resin
SAN Resin
Fabric <1
Rubber 0-100
Resin 50
Latex 250
Latex 750
See Figure 5-1 50-100
See Figure 5-1 1
Packaging 24
, Containers 3-7
Containers 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 et al. , 1978
McNeal et al. , 1979
Gawell, 1979

-------
monomer content)  under  the  same  conditions  released  0.283  ppm of acrylonitrile

 (Brown et al., 19 78 ).

     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  formula tiotts

containing  acrylonitrile  were   once  used  as  pest  control  for  residential

buildings, tobacco, grains,  and  nuts (Davis  et al., 1973).  However, today fumi-

gants containing  acrylonitrile are  no  longer in use.

5.5.4  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:
          Production facility

          Monomer
          ABS-SAN resin
          Acrylic and modacrylic fiber
          Nitrile elastomer
          Adiponitrile
Estimated Acrylonitrile
Emissions (Suta, 1982)
	(Mg/yr)	

          802
         1424
         1276
          295
           59 .0
                                                          3856
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.
                                     5-15

-------

-------
              6 .  ENVIRONMENTAL FATE ,  TRANSPORT ,  AND DISTRIBUTION



     The environmental fate of acrylonitrile in air, water, and soil is discussed



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 hydrocarbons.   This  addition



reaction forms an excited epoxide  that subsequently decomposes to alkyl 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  TS% will  survive  at 50 km  downwind.
                                     6-2

-------
6.2  FATE, PERSISTENCE, TRANSPORT ,  AND  BIO ACCUMULATION 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 et 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 4




sample  showed even  less decomposition  after 6 days and only 23/5 decomposition



after 23  days.  It  is hot 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 OHCHgCHClCN and  CICHgCHOHCN  (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 micro-



organisms are  capable  of degrading  acrylonitrile,  especially acclimated micro-



organisms.  The breakdown products of aerobic microorganisms may include ammonia



and  acrylic acid (Mills  and  Stack,  3955),  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



aery Ion itrile.  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 7056 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



e_b 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

-------
                                             Figure 6-1A
                             456
                             DAYS OP  INCUBATION
                                                               10
   — 30
   a.
   a.
   o
   <
     20
   o
   X
   o
   <
   o
   Ui
   o
      10
                   Figure 6-1B
                                         REDOSE
               REDOSE     REDOSE
                                          REOOSE
                 10
                20        30       40
                  ELAPSED  TIME (DAYS)
                                                       50
                                 6O
      
-------
 50  ppm aery Ion itrile;  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  eb al. (1968) 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  (MAS,  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  biodegra-
dation of acrylonitrile in industrial wastewater was also shown by  Kincannon and
Stover  (1981) and Freeman et 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




90$ of the acrylonitrile was degraded by these microorganisms.



     The preceeding 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.   Aery Ion itrile's  inhibition of  anaerobic



digestion by microorganisms was also confirmed by Ludzack et al.  (1961).   These



authors recommended that the anaerobic digestion should not be used for treatment




of  acrylonitrile-con tain ing water.



6.2.4  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  several  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  K  - 0.23  (Veith et al., 1979)
                              ow
           log BCF = 0.542 log  K    +  0.124  (Neely  et al., 1974)
                               ow
           log BCF  = -0.508  logS +  3.41  (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 nmol/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 j3t al.  (1977) are appli-



cable for fish muscle only.



     If the values for  K   and S for acrylonitrile  are  assumed  to be 0.12 (Leo
                        ow


et al., 1971)  and 1.385 x 10  jimol/1 (Groet, 1978), respectively,  the theoretical



values for BCF can be calculated  to be 0.1 (equation of  Veith  et al., 19 79 ) for



the whole  fish  and 0.4 (equation of Neely et.  al.,  1974)  and 2.0  (equation of



Chiou jet al., 1977) for fish muscle.



     U.S. EPA  (1978a)  experimentally measured  the steady-state BCF for acrylo-



nitrile in bluegills containing about 4.8/5 lip ids.  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 43  x 0.625 = 30  (U.S.



EPA, 1980b).



     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 3.4.8 the calculated evaporative half-life may
not be accurate for acrylonitrile in natural aquatic systems.
      The removal of acrylonitrile from an aquatic  system will take place  through
various  chemical  (photochemical,  oxidative  etc.),  biological  and 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 media into water  (see Section 7.3-4).  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  of the toxic effect
                                       6-9

-------
of the large spills an  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  e_t al.,  1978;

Sakurai e_t al.,  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 Insti-

tute (Going _et a_l., 1979) has determined environmental levels of aery Ion itrile.

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 4 acrylonitrile producing or consuming plants.  During this study, quanti-

fiable amounts of ACN (>2.5  ppb) were  found in the vicinity of each of the plants

that  were tested.   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 24 hours for all  stations, except

Monsanto  in  Decatur, Alabama,  where  two  24-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).
                                      7-1

-------














gf
'"'I
•gl
1
bO
5
«H*
e>
a

|
o

t4
o
e«
5
S
|
5
r*

£
O
1
i
•a
<
:
H











| Oa>
*o S
rH O
•H >
t. <
1 I
M ig
O 52
vW
O
|
4i
3
C
a> ?E
o o
c «-i
•s
S



^


n
#3
Cj O
O 0
• rH




311
^"o "e
O rH rH

<0 O
SO . O



4^
O
1







o
S
W








vO CM CM CM
CM rn en rH





rH rH CO rH
o o o o
V V V V.
VD CA O"» CM
• • • •
rn uj co •=»•







in VD t— Ln








CO CO VO O
rH rH CM IT!
1 1 1 1
\o in c~ m
O O O rH


n
c.
Q> Q> CO Q>
rH rH rH XI
•H iH iH "C1 -H
C« C< ti C {x«
*> *J 43 (0
T) -H -rH 0
I I I rig
& & & &s
o o o o S3
«< < •$ «< -g



a """ a EH

§n S j
C W >"D " ^
&s &= s ;
rH - -O ~ t.
C t, C B o 03
tao ea rt S
(US 0) J CEH CO
55^^






m PO C3 CM
CO





rH C\J CM CM
O O O O

rH 00 CO
• • • •
rH c- JQ a-
rn















CM OJ CO CJ
CM O rH CO
till

O O O O


n
co E* n n
i, . £" "^ ™
O O ro 83 rH en
g A


< g g

w 2 • tT o I-H

«*4H <^ffi) (d*H On
4^0 4Jfi sx: -H -H
CM c'>> To5 t, s
o •* o cd DO ro *o o

3 3 O O
Q Q (6 O
7-2





in o =r
0 M m





C\J »H CM
O O O

rH H O
H m eo
rH















o aj vo
i-H C3 0
i r i

o o o



CO CO
c t,
•H 0)
co e a>
0) O rH fl)
CC 4i "*H *CJ
CO £, -H
Z nl 43 S
S rH g, 0
s


O

f-<
•* o •* > o
S-P rH 00 n
CO ffJ M C **
C >> >> c o cd
(D *O p *H SH S
C ^! 'H (Li g _-3
S CD >




























• a
r^ 4J
K 5
S g.
^ M
S 5
co a.
I S
<+-* Q)

"S S
c
•H 0)
is -^
° §
2 .&
01
n |
q o
•H 0)
0 rH
ex *-*
o
I °
rH 0)
O Q)
1-1 S
(rt QJ
CO 4^
cd
CM o
O jrj
CO D,

S J;
4-> (D
•H -H
d>
(1) S

CD X>












































O
•H
a
n)

t

t,
.2

0)
0)
o


-------
     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



has been divided  by the number of determinations.   When the determined concentra-



tions were less than the detection limit, the data used  for averaging were the




values at the detection limit.



     The  highest  individual  concentration  from  this  monitoring data   was




325 ng/m3- and  the  lowest was <0.1  |ig/m3 (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  (84 ng/m3)  in  one plant and a  low level  (0.5 ng/m3)  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  et  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 has 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




(49:1) which improves recovery (95$) (Gagnon, 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 al., 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
                                      7-3

-------
    Table 7-2.  Comparison of Monitoring and Dispersion Modeling Data
                (Suta, 1979)
Plant/Location Distances
(km)
0
Acrylonitrile Concentration ((ig/nr)
Monitoring
Dispersion
Modeling0

American Cyanamid
New Orleans, LA

American Cyanamid
Linden, NJ

Monsanto
Texas City, TX



Monsanto
Decatur, AL



DuPont-May
Camden, NJ


DuPont
Waynes boro, VA
Borg-Warner
Washington, WV
B.F. Goodrich
Louisville, KY



Monsanto
Addyston, OH

Uniroyal
Painesville, 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
4.3
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
0.4
1.1
1.3
0.7
43.4
0.2
4.3
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,
     °Estimated concentrations at the midpoint of the distances.
      Dispersion modeling estimates were not made for acrylamide  plants,
                                      7-4

-------
two was poor.  Therefore,  the  need  for  more  experimental monitoring data cannot




be overemphasized.



     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 are given in Table 7-3.




These data were obtained by Going _et al. (1979).   Whenever possible,  grab  samples




were collected upstream and downstream of the plant discharge points.   In some



instances,  discharged wastewater from the plants was 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

-------








en
t—
rH



"c5l

431
,
!


rn
1
t-
0)
5











* g
** *J
t>
43
cd
*
B
B
O
3
cd
L,
43
§ •§>
O tH
g *
U

























O
S
•a
2
cu













CD
43
iH
CO





rH T) rH
• • •
0 B 0
V V










cd
•o co a-

BOO

























CD CD
•H -H
43 TJ 43
•H -H  rH >>X
O. rH  4> EH
no fl) o •*
tO L, TJ >>
11 13 as

S «S EH

O
V











o
o
in
en











0}

9
£>
•H
CM

O
•H
rH

L.
O



•o
E
cd
0
tH
ft
|






j


•b
onsanto, Decatur
33
Li
cd
CD


>
•5
05
B
g
EH
rH
O
V











t—
•
en

























2
•H
CM
O
•H
ft











O
co
CM
I
if
O
%
o


CO
0)
B
03
CD
CD
fe

*
on
rH
V











•o

B












m
L.

a
•H


C)
i-l

O


o
e
•a
§
0
•H
ft
|






^

E
0
Q
1o
B
CM
1

•H
O
<
r.
If
05 2
.C &
43 (d
i*
co

rH












en

rH

























g
SS
*t
co
CO
§






&


B
arner, Hashingto
1


to

Li
fe
•H
O

0

rH
V











O

C\J


















•0
g

CO

CD

Is
S^ 1
ra
rH CD
CD Li
CD 2
rH *t
•H CO
43 r/^





>H
^

rH
rH
oodrich, Louisvi
o
•
Cu

(0

S
"
•rl
BE
O

O

rH
V











O

CO

























g
1
z
•^
CO
s










g
g"
43
1
•S
cj"
43
g
ra
B
o
a:

i
B
•H
05
O

O

OA












0
o
en
a-





















ra

a!

43
S
CD
rH
•H
Li
43
•H
Z .




CO
H35
rH
•H "
O CD
Cd rH
from treatment f
royal, Painesvil
43 C
B a


rH Cd

CM Li
0) O
43
•O (d
g?
43
43 ra
§ S
3
rH 0
CM S
M CM
cn
rH
V











•O

B












CD
•O
•H
B
9
rH

O
cd

•o


CD
S
43
•rl
B
O
ft







B
O
Li
43
ra
•rl
«
I
05
cd
43
43
o


CO

13
•H
a
K
L, O
o cd
m 8
•rl -H
0 J
























































•o
CD
B
ID
43
CD
•o


B
II
•o



7-6

-------
7.2  ACRYLONITRILE LEVELS IN SOILS AND SEDIMENTS
     The environmental levels of acrylonitrile in a few soil and sediment samples
are shown in Tables 7-4 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 7-4 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 (4) spillage during transportation.
The exposure  from each of these sources is discussed below.
                                      7-7

-------
  Table 7-1.   Acrylonitrile Monitoring Data for Sediments  (Going ^ al., 1979)
Site
Concentration (|ag/kg)'
Mississippi River near American Cyanamid,
  New Orleans, LA
Tennessee River near Monsanto, Decatur, AL
Wateree River near DuPont, Lugoff, SC
South River near DuPont, Waynesboro, WV
         <0.5

        <50
        <50
        <50
 These figures were the lowest detection limit
    Table 7-5.   Acrylonitrile Monitoring Data  for Soils (Going et al., 1979)
Site
Concentration (|ag/kg)'
American Cyanamid, New Orleans, LA
American Cyanamid, Linden, NJ
Monsanto, Texas City, TX
Monsanto, Decatur, AL
DuPont, Lugoff, SC
DuPont, Waynesboro,- VA
Borg-Werner, Washington, WV
B,F,  Goodrich, Louisville, KY
Monsanto, Addystori, OH
Uniroyal, Painesville, OH
Vistron, Lima, OH
          aO.5
         <50
        <100
         <50
         <50
         <50
         <50
        <400
        <400
        <400
        <100
 zhese figures were the lowest detection limit
                                        7-8

-------
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 et 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 al., 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, 198Ob).
                                       7-9

-------









o
•H
*w
•H
8
0.
CO

£

0)
rH
Li
•H ^^
CCM
ooo
^~» ON
o"
4J rt
4-5
O 3
TjS
CD*""'
JS CO
8*'rl
S't.
0
-P bO
«3j CO
"in
•P O
CO CO
go
e.
>"*
3 3
eo o
0 CO
0.
X fi
Cd O
•H
g !
Id w
3
8-
PL,
 ra
•rH CO
w





o

o t>^ w
3,-s si
M CO T3 ;H
O Ob
<3j 23









B ra
^i .3
i/j *n
^. ra
|K







t.
i
*— fS
^•g W
«gj Cj
S







%,
Oi
ca o
3 iH
G -P
C aJ ^
4^-5

bO CD 31
S C ^
^cS
«a!











O
o
f"^
92








0
R












0 0
0 0
C- 0
CM ^r
"







o
o
cr>
*









O^ ON ON
* • • ON
ON ~r ON *


i i i



o o o o
• • • •
in o in t-
*~ *~
7-1

o
o
o
CM





O O
O O
O CD
*t *s
* 5








0 0
o o
o o
in o
.=!•










O O
o o
0 O
OO CO
^ K?







o o
o o
o o
vo O
CM








en en
en a-
o o


1 1



0 0
in <—
• •
0 0

0

o
0

£]





o
o
o
s








o
0
o
0
.=}•
^f









o
o
o
CM
ft







0
o
o
ft
C*—







CTv
s
o


1


o
in
o
•
0



o
o

"JO





0
0
o
•=3-"
ft
-







0
0
o
•s
fp
cQ
OO









0
o
0
t-—
o
*~






0
o
0
CM
o
^D
»»
r—





CT>
0
0


1


o
o
•
o



0







_Q
0
o
o
in







^
0
o
o
^D
CM









XI
o
o
0

in







XI
o
0
o
CO









o
0
o


1


in
o
o
•
o


XI
0







Xi
0
0
o
c\r
R?







XI
o
o
o
in
t—









Xi
o
o
o
,_
0
*~






XI
o
o
o
«








=f
o
o
o


1


t—
o
o
*
o


Xi
0
o
o
o
CM




XI
o
o
o
CVJ
C--
c\r






XI
o
o
o
^o
zzf
en








XI
o
o

_jT
in
CM
.=r






Xi
o
o
o
R
vX1
•*
CM








0)
rH
a
o -a
CU CD
Q-i ra
PL, g
rH Q.
4J W
O
H









































VD

O
XI
>>
rH
cx

4-5
rH
i
xT
D.
0.

A
|S
OO
.S

3!
i,
(2

_j ^
SL,
CO
rj
5
o
o
cd J


4-4
O

•^
o
XI
4J
•H
S
ra
fl)
JL,
W
o
D,
S
O
c!

CO

§
>,
rH
S
QJ
£
CO
ts
ra

o
•ri
4-3
, — j
3
0
rH
a
o
CO
3
8
X5
ra

nj
^

CO
•s
3
2
CS

w
CO
Krt
C
fi

CO
ra
£
_
•H •

ra c
£3

ra
O X!
iv] <£>
d"^


-------
     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




14 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
7
7
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 jit al. ,
Brown et al. ,
Brown jt al. ,
McNeal et al.
Brown et al. ,
McNeal _et al.
McNeal et al.
Brown et al . ,
Brown et al. ,
Brown et al. ,
Brown et 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
          Usec
     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

              7-14
            =20
       Some of the applications may result in incidental food contact.
                                      7-13

-------
Guerin_et al. (1974) 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  olear whether the source of acrylo-



nitrile is from fumigation of the tobacco or from the combustion process itself.




7.3.4  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 (40.455  of  overall  shipment)  poses the



greatest hazard and by barge  (1.7/5 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 5.5.2).  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  (46 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 etal., 1978),  ABS/SAN resins



 (Chaigneau  and  LeMoan,  197^),  acrylonitrile-methacrylate  copolymers  (Guyot



jstal., 1978), vinyl chloride-aery Ion itrile  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 oh a number of factors,  includ-



 ing the nature of the polymer,  gas composition, gas  flow rate,  and  heating  rate



of the flame (Sarofim et 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  480°C/min  produced  the following  major  nitrogen  components



 (Sarofim et al.,  1973).
                                     7-15

-------
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 results  in  greater  dilution of
the pollutants  at ground level and also  spreads  them 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
concentration.
                                      7-16

-------
                   8 .  BIOLOGICAL EFFECTS ON MICROORGANISMS




     Loveless gt al.  (1954)  studied the effect of  acrylonitrile  on growth and




cell  division  of  yeast  (Saccharomyces  oerevisiae)  and  bacteria  (Escherichia




coll), 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 £3. cerevisiae.




Treated cells were 170/6 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  40 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  e_t al.,  1972).   Papads with  18  or 20%




moisture content were sealed  in polyethylene bags, fumigated for 48  hours with 32




or 64  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
                                      8-1

-------
aerated river water was dosed with  10,  25,  or 50 mg/1 acrylonitrile.   Balanced



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-




, tra.tions ranging from 10 |ig/l 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  ppni)  acrylonitrile was



 "toxic"  (undefined  effect) and that  lower  levels  showed no effect  on  seedling




 elongation.



     Fumigant  mixtures   of  acrylonitrile:carbon   tetrachloride  (1:1)  had  no



 adverse  effects on seed germination of beans, beets, corn, peas, lettuce, onions,



 tomatoes,  wheat,  and oats  when seeds  were fumigated  for 24  to 48  hours  at



 concentrations  ranging  from 1  to 25  pounds  per  1000  cubic  feet  (Glass  and



 Crosier, 1949, 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  et 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 e_t _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 |il 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  9255, 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 £3. 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, LD95,  and LD99  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

-------
 c
 o
 cd
 60
rS
 o
 U


 o
 (JJ
 U)
 O
 U)
 4J
 (J

 0)
 (0
 Q)


r-4
 Q)
 CO
 O
a
 ra
 m
 ,
C H -0,
U-l
O OJ /-*
U CO
x: 3 P
*J CO 3
00 O O
c cux;
OJ X »-*
rJ U











D
O
•H

O
C.
en




\o  rH
V -H
ai x:
S D.
»s
U Tt
•H CO
CZ N-'

VD
rH in
CM
« rH
3
ZJ •
3 rH
E: ca
co e
P 01
C 00
a; -D
Ctf ,J

i i
00
r- rH
O CS
o
-3- -a-
O rH








-3- vO
CS







rH
•H

01
01
JZ
CO
01
ca

co
CJ
-H
01
2


•3-
L-l
CM
rH


rH
wl
01
00
^

1

o


o
CM








CM






^f^
CO
•H
rH
CO
P
o

U
01
a.
CO
0)
o
p
CO
Nl
N— '


•3"
in
ON
rH

t
rH
CO
4J
01
C
01
00
•o
rJ

1

0
CM

-H
rH








vO














rH
1H
>
G
B
CO
01



•3-
m
ON
rH

B
rH
CO
U
01
C
01
00
.•3,
C
^

1

in
m

o
CO








CM

^s
to
3
4J
o
OJ
4-1
r>
o

CO
OJ
•a
•H
rH
01
01
CJ
CO
o
4-1
e.
CO

"— '


-3-
m
CM
rH

t
rH
CO
01
C
01
00
•§
J



m
CM

r~
-H








vO










01
rH
4J
01
0>
CQ
01
P
O


-a-
m
ON
rH

t
rH
CO
4J
01
01
00
•§
^

1

o


o
CO








CM







^— ^

3
QJ
O
•H
C
CO
O.
3
•H
U XI
a> o
1 00
00 01
p
a
4J
C/3
^^


r*. in
CM CM
rH rH

*a •
CO COJ
rH
8*1
pq C
01
t^J J_|
00
•§•§
O -H
PQ _3
in
• I
CM

ON
•a-

ON O
rH CO








OO vO






0)


01 3
01 CO
pa 3
u-l
P C
3 O
O U
fe 3
•O i-l
0) rH
co o
S XI
U-- 1H
C P
0 H
O — '

vD
-a- CM
in rH
ON
rH *
3
• 4J
rH| 3
col "z.
Lindgren et
Rajendran &

i i
vO
o m
rH O
rH
CO
in *3-
vO O








CM <•
CM

"co
-H
CO
c
01
c
•H
x:
o

co
3
oi Ti
rH 3
4J p
OI -Q
to O
P9 CO
O
01 rH
CO rH
r-t CO
s S


vO
00 CM
CM
rH ••
3
*^J ^J
C 3
CO S
rH
.* >«
O
s n
CQ CO
p
cO T3
C
•3 01
O CO
PS as
CO
• i
•3-
CM
CO
i •
rH

vO O
CM rH








in -a-
CM





y— ^
e
3
0)
£2
0) CO
rH 4-1
4-1 CO
01 CO
01 U
0 e
P 3
3 -H
O rH
rH O
•H
T3 P
01 EH
ty± *_x


-3-
m
ON
rH
^
.
' ^
Lindgren et

i

o


rH
rH








VD



01
rH

01
01


fj
•H
a
p
o
•a
0)
x:
jj
o
o
4-1
1
CO
en


-3-
m
ON
rH
^
.
CO
Lindgren et

i

in
VO

m
ro








CM

/"••»
CO
-H
CO
C
01
E
CO

•rt
p
n
co

CO
3
rH
•H
xi
c.
01
CO
N
|


vO
ON -.
CO XI
0! Crf
*J ^-*

vO
r-
ON vO VO
rH r- 1 —
ON ON
" rH rH -
3
1J "3 "3
5 C! C
S CO ct
rH "~^
co co ca
*3 *^3 *&
O T3 'O
•n e c
CO O O
Oi CO PS
0 <3-
1 • •
^iD OO
03
o


r»- co co
O CN f^i


p*. r-s






-------
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 evaluated the mutagenic potential of
acrylonitrile in fruit flies  (Drosophila  melanogaster).   Details of this study
are presented in Section 13.4.1.2.
                                      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  £t 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  24, 48, 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 24 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

-------




0
r-t
•d
"*"^
*o
5
"
£
<
o
4J
•o
0
co
0.
fa
S3
co
•H

J^
o
CM
CO
0
3
1
O
§
**^
j-«
O
Is
i.
40
C
1
o
o

"«)
jo
0
•-1
C
•H
•O
2
•
T
CM
T—
2
cd
E-t










0
O
c
0
su
0
0
K


£-,
vo
O1


H J3
*i> CM
8 t-
^-^
O
ID
0 £<

CO
if

S3
J^-
CM




C
O
•H J-,
40 0
3 43
•H |3
«


0 40
P> $
H E-i


a o
^3








8
•d
0
0.
CO
VO
cn
<£
o->
vo
CJ\
r—
vo
?
VO
C^v
T^


• • • • • rH
•ai
0
o
CO

0
T3
03


oo
=£
*~



1



^_

"£

c—
CM
oo



CH
"•o
cd
^
fo



CO



in
CM


cd
CO

o

1 8
s l—j '
Fathead
(Pimeph
cd 1 cdt cd I cd | 'O
•H
4J
a
0
CO

0
•o
§
03


t —
CO
«—


in
o
CM


in

CM

oo
•
=j-




faO
^ — V
o
CO
X — *

fa



00



in
CM


cd
CO
S
O
c
•H 0

10 73 S3
H Cd O.
0 00
a •£! e
O 4-5 *rt
i, Cd PL,
p>t pTj \**S
0
&
O
co

0
T3
s
ffi


^
CD
'~


i-
,_
»~


CO

r—

LO
•
OO
oo



bO
CO
J3
b



Cn



in
CM


cd
CO
S
O

co
•H 0
— ^ S r-j '
-1 cd D,
0 00
§J3 a
4J -H
M TO CU
40
C
o
CO
JH
0
•a
s
03


CO
r^
'



1



OO

^

•
in
CM



bO
CO
^3
fe<



co



in
CM

&
CO
•H
C|-1
C '•
3
CO
-^
CO iH CO •
CO rH -H ;
0 'bO O
a g> n
O 3 0 <
^ £ J
Qt pQ v^'
0
G
o
CO
£4
0
CO
•o c-
Cu *""

4J) **
Q) Q_j
•o r-i in o
C 3 t- 0
0 cd cr> H
03


in
oo
00



0^4 r- CO



1 t




1 0 1



in

oo
oo

vo
•
j^i-
•=3"



bO
CM
O
CO
SB
fc,



co



in
CM





-^
o ^^
3 Cd
M Cd 4-3
M 0 -H Cd
3 CO rH iH
•J 0 -H 3
D -H O 0
M Q. 0 -H
•J D, O 4J
d 3 PL, 0
3 O ^^ t-i
=J-




i in
r—


1 1






Is ;s
fa fc,



CO b



05 o
a CM






23
43 -H ^> CO
CO CO CO -H
•H CO 3 CM
CM CO 4-3 Cd
"o ^ cd (M
rM TO t, £1
O O 3 0
O ^ cd ISI
                                             o  o
                                             cd  -H
                                             t,  SH
                                             PQ  CD
12-2

-------




-P

o

CD ®
vu 0
"~\ G
d ^
•^ o>
'P <4-l
0 °>
^ PH
*>•>
t,
O
•^ £H
0 ^
,j>
•o <*
0)
00
0
a
M -" £

ft I1 ™
co S fr-
vH
Pr.
•*> O
L ^
o o t,

MH
s s
3
"cd

*•— "n *G
o .n-
i f\ """*'
U 1 ,-v I
o °°
j
S 	 '


Q
•H
"*~J £*
2 o
!l -H SH
•g 40 CD
1 31
o ^
o
rH
M ® *»
1 £s
EH E~*
•—3
G
• I—I
•o '~
0) >-*< t_)
a St.
P . N-_^

,_
1
v—

CD
2
cd
H

00
•H
O
CD
a
CO

3 3
O O
CO CO C
CD CD S
G G 0
O O J-,
M M fP
"^3 'O 'O
C3 C C
CD TO (O

•H -rH G
G G O
O O CO
O =f O =f X O
SH C~- SH t— O C—
cd o^ cd o^ cd CP\
^4 ^~ ^4 1P~ *~O ^~




i i i




i i i






vo O
• •
t- =r o
«— CM t~



CM ^r
CO C~^ 1
oo ro




^^
.-o
cd

pr [ pf^ P-TJ


g g g





g § g







40
2 -
CO ^- CO EH -r
3 CO 3 t
G 3 G ^ S Q
S -H G -HO O O E
S^'d ^,^'ri -9 • a x
cox aaa GC-HJ.
C JG O i-t I^a £-j 'H Cd «r
•HCUX! cdoro cdCOCl
2 ^^ a o » 	 o K — ' b
G
1

Q)
CD
3 "0 T-
J G in
cd cr>
•a »-
G >•>
cd 40 •»
Si 40
CD CD 40
£4 SS CO
G 00 bO i4
c"! ^— J3 ^
3 o> cd cd
"-3 <- Q O




I I




i i





oo
CM
w
00 1



in
I =f
CM







•S *
fa co


co co



i
t- ^r
o oo o
CM '—CM






CO
3
•a
•o -H • ^-x
•* CD ' — * CO
H "O CO CO CD
< M 3 3 Q) G "O
) CO 40 JG o -H
5 G -H 0 DO -0 O
J CD O C -H O '.a
, -0 3 cd Q-, bO g
1 rH CD rH G Cd O
1 O J CD -H J JG
0 cj ^ e PL, ^^ S-.










•a
CD
t.
O
a
CD
su
-P
o

II
s;

L"
CD
rH Id
•jv. rH !3

S bo CD
S co
o
CO O H
00 CM

00 CO CO
co co
CD 0)
C G t,
*o T3 ^" "^ (U
Sj J-i S 40
cd cd T3 cd
^3 ,G cd S
J-, J3
.~ ... bO CO
rH rH CD CD
\ Vj, 40 t,
bO bO L CM
S S CD
Ixl it
0 CM =
bO S
>> >> G fa
in bO 40 40 cd
• -iH -H S •••
»— CM T3 T3 SH CD
21 21 .H -H 0) J-,
O O O 3
40 40 Cd OJ — m
X! X! 0
bO hO .- .- o a
•H -H rH rH CO X
CD CD bO \ ^ CD
S S bO bO +1
<- B S *3
.^ .^ . bO in bO
'-^ '^ O O VO «~ 3
CO CO 21 OO CM *— O
CD CD • OO CO S-.
X! X! 40 O >> CO X!
OCX! >-> 40 CD 40
G C bO +| 40 -H G
•H -H -H -H C 'O S
0) in C -H S-i O
^^^ *«'Hd Sr1
• 1 *— S rH (tl 43 <4-l
CM in s cd ^i
1 « '-^ 40 i! rH SH II
s 	 ' v — ^ c_) GO ' — ccj 40 p*"!
G -H »- • Cd
SB ^ t — CM • CD
«— in • t- £> in B bo a CD a
II 0 G | EH X
CO *~ •=!" t— Q) | bO CD
• • • 1 rH S-. Q)
o co in • o SH
-------
difference is  statistically significant.   It  can be concluded,  however,  that
water hardness has little effect on acrylonitrile toxicity.
     Henderson &t 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 24,  48,  72,  and 96  hour  LC50  values  are  given  in
Tlble 12-1.   Comparison of static and  continuous-flow LG50 values  shows  that
toxicity is equal at  24  hours,  but is greater under  continuous-flow conditions
after 48 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
24-hour exposure period. White crappies (Pomoxis annularis) exposed to 4 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 24-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  48-hour LC50  of  acrylonitrile  to  zebrafish (Brachydanio  rerio)  was
determined to be 15 mg/1 by Sloof  (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 40  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
                                      12-4'

-------
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  LCO, LC50, and
LC100  (0, 50,  10055 mortality) values  determined  at each laboratory are given
below:
                               4B Hour Lethal Concentration Values (mg/1)
                               	LCO            LC50           LC100
          Laboratory 1
          Laboratory 2
16
 8
13
48
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
                                      12-5

-------
graphical interpolation, was 24.5 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 (Gammarus  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 95/6 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 3D.  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) ]). magna,  and water of  similar mean
hardness  (155-173 mg/1 as CaCO_) 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.
                                      12-6

-------
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 et 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 rag/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     4     5    10   . 15    20    25   30
LC50  (mg/1)            33.5  14.8  11.1  10.1  8.1   6.9   5.2   4.2   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 ail.  (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.
                                      12-7

-------
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.  Sloof (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  US-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.
                                      12-8

-------
12.3 SUMMARY



     Acrylonitrile has been shown to affect some terrestrial and aquatic plants



at exposure concentrations of 9 to 100 mg/Jl.  Acrylonitrile is toxic to aquatic


animals at exposure concentrations in  the low milligrams per liter range.  The



reported acute LC,-n values for fish ranged between  10.1 and 70 mg/£.  Subchronic
                 ou


exposure of fish for 30 to 100 days resulted in LC   values of =2 mg/£, with no



evidence that  a  threshold concentration  had  been  reached.   Although  the only



tested invertebrate, Daphnia magna, has the lowest acute LC,-.. value (7.6 mg/&),



this species was not adversely affected by chronic  exposure to 3.6 mg/5, through^



out its whole life cycle.
                                      12-9

-------

-------
            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 + 9.64? and 104 +_ 14.4? of the administered dose at  0.1  and 10 mg/kg,



respectively  (Table  13-1).   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 and 100 ppm (11  and 217 mg/m  ) acryloni-


trile.   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



C02 and  in  the body  (Table 13-2).


      Rogaczewska (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 ((ig 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 et al., 1977).  Following oral and intra-


 venous administration of l4C-acrylonitrile, typical plasma concentration versus



 time  curves  was  observed.    The  biphasic  disappearance  of  radioactivity
                                       13-1

-------
         Table 13-1.  Recovery of Radioactivity from Rats
                      Given Single Oral Doses of 0.1 or
                      10 mg/kg 1^C-Acrylonitrile*-(Young et al., 1977)


Radioactivity in Urine
Radioactivity in Feces
Expired Air
Organics in carbon0
Organics in solvent
H14CNe
14 f
C02
Body
Carcass
Skin
Cage Wash2
Total Recovery
Percentage
0.1 mg/kg
of 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
82.37 + 9.64b
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
, 104.04 + 14.40b
 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.
n?ittsburg activated coconut charcoal, 12 x 30 mesh.
 trap - 2-methoxyethanol
trap - 0.02 M
                      in 0.1 N
 trap « 5 M ethanolamine in 2-methoxyethanol
^water-acetone
 The 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 l>* C-Acrylonitrile for 6 hoursa (Young _et al.,  1977)
                                             'Percentage of Recovered Dose

                                                5 ppm            100 ppm
     Urine

     Feces



     Body

     Cage Wash
68.50 + 9.38C

 3.94 + 0.97

 6.07 ± 1.58C

18.53 + 4.68C

 2.95 + 3.95
82.17 + 4.21°

 3.15 + 0.82

 2.60 + 0.83C

11.24 + 2.85C

 0.85 + 0.58
     'Hats 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.

      The mean values for the two exposure levels are different at the P=0.05
level of significance.
                                        13-3

-------
indicated  a  pharmacokinetic two-compartment open  model for elimination.   The



half-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.  (1980)  investigated the pharmacokinetic profile of intact




acrylonitrile after  intravenous or £er  £s  administration of  acrylonitrile to




male Fischer F344 rats.   The plasma concentration of acrylonitrile versus time



obtained after intravenous administration was  characteristic of a one-compart-



ment model with first-order elimination,  but  a  biphasic  elimination was observed



following a 30 mg/kg oral administration. The half-life of first-order elimina-




tion 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  |j.g/ml  of  acrylo-



nitrile remaining four  hours after dosing.  The  cyanide  concentration  rose to



about 1.5 [ig/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)


                                                                  14
was determined in rats given a single oral or intravenous dose of   C-acryloni-



trile at doses of 0.1 and 10 mg/kg (Young &b 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
                                      13-4

-------
       Table 13-3.  Distribution  of Radioactivity  in Selected
                    Tissues of Rats Given l^C-Acrylonitrile
                    (Young et 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
(pg Eq/ml)
Tissue
14
to Plasma Ratios of C-Activity
. a
i.v.
2 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
24 hours
1 mg/kg
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
0.87
0.57
0.83
14.28
1.06
c

-
-
-
0.35
-
-
-
-
-
2.10
0.71,
2.26d

,0.022




mg/kg
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.)
BFor all 5 rats in  this group, the ratio was 5.16 + 1.19 (Mean + S.D.)
                           13-5

-------
 intravenous acrylonitrile administration indicated that the  high  concentration



 in the  stomach wall following  acrylonitrile  oral dosing was  not due to  poor




 absorption. Analysis of the stomach following intravenous  administration  showed



 that  the radioactivity had increased from 30.33 [ig Eq at 5 minutes to 68.64 (ig Eq




 at 24 hours.   The particular retention of acrylonitrile and  its metabolites in



 the stomach seems in part due to enterogastric circulation (Young ^t 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 et al., 1982).  In other tissues,  the amount  of  radioactivity



 declined rapidly with  time because of excretion (Young  est al., 1977; Ahmed and



 Patel, 1979; Ahmed et al., 1982).




      Similar distribution patterns were obtained by Sandberg  and Slanina  (1980)



 using the  technique  of whole-body autoradiography following  administration of



 labeled  acrylonitrile to  rats  and  monkeys.  Male and female  rats received   C-



 acrylonitrile  by intravenous  injection  (13  mg/kg) and  were  killed  1  minute,



 20 minutes, 1 hour,  4 hours, and 7 days after treatment. Two  monkeys were given



 4 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 acrylonitrile.   In fetuses exposed d.n  utero,  only  the eye  lens  accumu-



lated label at a higher concentration than that observed in maternal blood.  The
                                      13-6

-------
labeling  pattern  in monkeys  was  similar  to  that  in  rats  except  for  a  more



pronounced activity in the liver.



13.1.2  Metabolism



     The  metabolism of  acrylonitrile  has  been  studied  by many researchers.



Dudley and Neal (1942)  and Brieger £t  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 j2t 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  etal., 1976).  Evidence  from Gut ^t  al. (1975), Wright (1977), and



Young et 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


             14
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 Abreu (1982)  have shown that acrylonitrile is converted to cyanide in



the brain and  liver of rats by enzymes associated with the  microsomal  fraction.
                                      13-7

-------
     Urinary excretion of thiocyanate following acrylonitrile administration at



different dose levels and by various routes accounted for from 2 to 33% of dose




(Gut jit al., 1975; Hashimoto and Kanai,  1965).  Gut et al. (1975) emphasized that



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  (14.6 to




33.1$) than after intraperitoneal (2.2  to 5.7?), subcutaneous (4.6?), or intra-



venous  (1.2?)  injection.   After oral  dosing,  thiocyanate excretion  showed a



distinct lag period (4 hours),  which suggested that acrylonitrile was not appre-



ciably metabolized shortly after oral dosing (Gut _et_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 j2t 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



jet ^1. (1975) indicated  that body size may  be an influencing factor.



     Pretreatment of rats with phenobarbital (a  microsomal inducer), SKF525-A (a



microsoraal 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
                                      13-8

-------
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



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 ^t al., 1975).



13.1.2.2  Reaction with Sulfhydryl Groups



     Acrylonitrile  reacts  with  sulfhydryl  compounds  in laboratory  animals




through cyanoethylation (Hashimoto 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 intraperitoneal injection,  that acrylonitrile




treatment resulted in a rapid decrease  (within 4 hours)  in the levels of gluta-



thione in both the liver  and brain (Zitting et al., 1981).  Hashimoto 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-cyano-



ethyl)cysteine  (Gut jet  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 Chasseaud, 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.
                                       13-9

-------
                                 NON EHZYMIC CONJUGATION
                                                          R-CH2-CH2CN
NOK 3XIDATIVE PATBUAYS
   CH^CH-CN
OXIOATIVE PATHWAYS
1.  NUCLEIC ACIDS
2.  PROTEINS
3.  BIOLOGICAL NEUROTRANSMITTERS:   l) ADRENALINE AND ITS
   ANALOGS; b) SEROTONIN;  c) Y-AMINOBUTYRIC ACID;
   d) HISTAHINE
4.  OTHER NUCLEOPHILE COHPONEHTS OF TISSUES
                                                                   SLUTATHIONASES
                  /COW
         »   CHZ-CH-NHCOCH3

              S-CH2-CH2-CN

      CYANOETHYLATED HOICAPTURIC ACID
                                                                  HCN +
                                                                       CHO

                                                                  GLYCOLALDEHYDE
                                                   CO,
                                                                   HYDRIDE TRANSFER
                                                                                      COOH

                                                                                  GLYCOLIC ACID


                                                                                         «
                       GLYOXALIC ACID  OXALIC ACID
                                                                                        .   - ».CH3COH*HCN
                                                                   CYANIDE TRANSFER    ^Vf
                                                                                     CM OH
                                                                                  CYANOACETIC ACID

                                                                                  OOH
                                                            GLUTATHIONASES
                                                                            CH,
                                                GS-CH--C + HCN
                                                  •CH2.
                                                                                              N-ACETYLATICN
                                                                                                                  MERCAPTURIC ACID
                 Figure  13-1.    Proposed  Pathways  for  Acrylonitrile  Biotransformation
                                      (U.S.  EPA,  [1979])
                                                           13-10

-------
     Langvardt et al. (1979)  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




structures.    Metabolite  7  was  confirmed  to  be  N-acetyl-S-(2-cyanoethyl)




cysteine, and metabolite 1 was thiocyanate,  as  indicated by i.n  vivo labeling




patterns when animals were given  2,3-  C  acrylonitrile or 1-  C acrylonitrile,




and  retention  times and fragmentation  patterns  on GC-MS.   Metabolite 6  was



tentatively  identified  as  N-acetyl-3-carboxy-5-cyanotetrahydro-l,4-2H-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 N2



gas.   No authentic  standard to  metabolite 6 could be synthesized.   By  using



analogues  to known  metabolic  pathways  for other halogenated  vinyl compounds,




Langvardt j2t al.  proposed  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 (1980) 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 1254 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 [1-1  C]- or  [2,3-  C]-acrylonitrile  incubated with



rat  liver microsomes, Guengerich ^t ail. (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
                                      13-11

-------
                                     CN
                                 c=c
                                /
   [GS-CH2-CH*-CN]
         -Glu, Gly
    0       0

 HO-C-CH-^IBE-C-CH
Metabolite 7*
                 0
                 II
                                        GSH
                     HV/°v/CN
                                              H'
                                          OH

                                   GS-CH2-CH-CN
           +Ac
-Glu, Gly
             0      0
             II      li
          HO-CCH-NH-CCH,
                                                 ?H
                                       /CH2-S-CH2-CH-CN
          0      C-CH3
          C-^  N
             Additional
         Catabolic ProductsNu-
                                                              Additional
                                                           Catabolic  Products
               L
CN
                           [HCN]
         Metabolite 6**
                              S2°3=
                                                     SCN
                                                 Metabolite 1*
         *Confirmed Assignment
        **Tentative Assignment
       [ ]Hypothetical Assignment
  Figure 13-2.  Proposed Scheme for Metabolism of Acrylonitrile by the
               Rat  (Langvardt et^ al., 1980)
                                 13-12

-------
0.5 mM 2-oyanoethylene 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-cyano-




ethylene  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




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



acrylonitrile.   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 et al. (1980) ascribed to




4-acetyl-3-carboxy-5-cyanotetrahydro-1,4-2H-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




(Langvardt et al.,  1980).



     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 13-4).  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
                                      13-13

-------
 cd r*»
 M o\
 4J tH
 ca
 -H   -
 C  4J
 •H 4=
 S  oo
 -a -H
 cd
 1-1  a)
O i-H
    •H
 01  )-t
J2 4J
4J -rl

 60  O
 C ^
•H  >,
 &  M
 O  O
 o -a
te  cu
    1-1
 co  cu
 01 42
4J  Cd
 o /-x
43  O
 cd  C
•u  cd
 o>  ^
S o
 cd  I
 C U
•H
en
 0)
i-H


•§




tw*
ioactivit-<
tabolites
•C 0)
a s

t[ ] tj
o cd
C
•U -H
S M
01 £>
u
M
01










01
4J
I cd
0 C
•H cd
/— \
-a
01
•H
<4-l
•r)

n
•o
•H

^
NM^




CO
1 0)
d) -u
C cd
•H 60
01 3
U 1-)
co a
>> O
u u
60
CD .y
CO --^
&£
n
01
"i
3
ts





r-)
cd
B
•H
41


0 <-l 00

oo CN oo en eN


m m oo
en o en vo  ^
at 01 cu
4J U > ^fi ^
5 5 3 S S
(% (35 PS O O
C C en
cd cd 01 co co
4J 4J i-l 3 3
M l-i M CO CO
cd cd cd oi oi
(X O, 42 42 J=
OT CO O Oi CA
 co

 3





CN

O




 cd

 CO
4J
 cd
                                                                                   o
                                                                                   CM

                                                                                   O
                                                                                 cd

-------
(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  13-4,  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



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-glucuroriic acid.



     Young et al. (1977) identified C00 as a metabolite of acrylonitrile.  Carbon
               1                       eL


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 C0?.  Thus, C0y 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.



13.1.2.4  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 jet ' al., 1975;



Wright, 1977; Young est al.,  1977).  Gut jet ail. (1975) indicated that the forma-
                                     13-15

-------
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 ^t al. (1977) quantitatively isolated three metabo-


lites designated as "A," "C," and "E"  as  well as C02 in male Sprague-Dawley rats

               ill
given doses  of   C-acrylonitrile  by several  routes.   Metabolites A,  C,  and E,


which accounted  for more  than 9555 of  the total radioactivity,  were excreted


primarily  in the urine, while  C0? was excreted in  the breath.  The chemical


structures of metabolites A, C, and E were cited by Keresztesy 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  structures of C and E  are the same), but  there was  no evidence for


structure identification in the original report (Young ^t 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


(11 mg/m  ) for 6 hours, 61/5 of metabolite E was excreted  in the urine, while a


dose of  100 ppm  (217 mg/nr)  resulted in equal proportions  of the  three meta-


bolites .


     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.


13.1.3    Summary and Conclusions


     Acrylonitrile  is readily absorbed in animals following  ingestion  or inhala-


tion, while  dermal  absorption is  poor and occurs  at =1? of  that  of the lungs.
                                     13-16

-------
                            14
Table 13-5.  Metabolites of   C-Acrylonitrile Separated from Various
             Fluids of Rats by High Pressure Liquid Chromatography (HPLC)
             (Young et al., 1977)
Sample

Urine, 0-72 hr.
0.1 mg/kg, p.o.
10.0 mg/kg, p.o.
5 ppm
100 ppm
Bileb, 1 hr.
In vitro0
Stomachb>d, 24 hr.
RBCb'd', 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
  Metabolites 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.
 b               14
  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.
                               13-17

-------
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, 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  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.



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 1940s 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



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,  1944;



Wilson j3t al.,  1948).   During these operations, workers were  exposed to between



16 and  100 ppm  of  acrylonitrile for periods of 20  to 45 minutes.  All workers
                                      13-18

-------
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



etal.,  1973)  following fumigation of  domiciles with  acrylonitrile mixed with



either 66% carbon tetrachloride or 70% methylene chloride  (v/v).   It is not clear



whether acrylonitrile or the  other  component of the fumigant was the causative



agent in these deaths; however,  one patient had measurable levels (0.05 mg$) 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.
                                     13-19

-------
     In a report of four case histories of human disease following fumigation of



domiciles with  acrylonitrile  and carbon tetrachloride mixtures, Radimer j3t al.



(197*0 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 _e_t  al.  (1969) and the minimal fatal  oral dose



in rats was reported to be  150 mg/kg (Wilson jit _al., 1948).  The LD50  in Wistar



rats following  subcutaneous or  intraperitoneal  injection  of acrylonitrile was



80 mg/kg and  100 mg/kg, respectively  (Knoblock ^t  al.,  1971).  Benes  and Cerna



(1959) had  reported the  oral  LD50  of  acrylonitrile  in both rats  and mice as



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



Jit _al.  (19*18)  as respiratory  changes, cyanosis,  and  convulsions  followed by



death.  For the investigation of acute toxic effects of acrylonitrile, exposure
                                      13-20

-------
by inhalation has  been  studied at length since  this  is the route  of greatest

concern in industrial exposure.


     Dudley and Neal (1942) exposed Osborne-Mendel rats to atmospheres of 90 to


2445 ppm acrylonitrile for 30 minutes  to 8 hours.  The onset of symptoms and the


lethality of acrylonitrile were strictly dependent on the length of exposure and


the  concentration  of acrylonitrile  (Table  13-6).   Short periods  of exposure

                   o
[665 ppm (1443 mg/nr) for 30  minutes] produced only mild effects, while a similar


dose given over a  long  exposure period [635 ppm  (1378  mg/m3)  for 4 hours] was


fatal to all the animals.


     Smyth and Carpenter (1948) observed fatalities in all 6 Sherman strain rats


following a 4-hour  exposure to  1000 ppm (2170 mg/m3) acrylonitrile, and no deaths


from a 500 ppm (1085 mg/m3) exposure.  In a second study, however, Carpenter et_


al.  (1949) reported  that 2 to 4 rats  (Sherman strain)  of a  group of 6 animals

                                        o
succumbed following a 500 ppm  (1085 mg/m )  exposure for  4  hours.  Brieger et al.

                                              •a
(1952) exposed Wistar rats to 100 ppm (217 mg/m )  acrylonitrile  for 7 hours, with

                                                          o
4 of 20  animals dying.   A similar dose [130 ppm  (282 rng/nr) for 4 hours] was a


non-lethal dose in the study  of Dudley and Neal (1942)  (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).


     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
                                      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 Remarksb


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.
   Symptoms  included respiratory distress and convulsions in cases of fatal
   exposure;
                                   13-22

-------
            Table  13-7.  Minimal Lethal Concentration of Acrylonitrile
                         During Four-Hour  Exposure
Animal
Species
Guinea pig
Cat
Rat
Micea
Rabbit
Rhesus
Monkeya
Dogb
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 and
Neal, 1942
Dudley and
Neal, 1942
Dudley and
Neal, 1942
McOmie, 1943
Dudley and
Neal, 1942
Brieger et al.,
1952
Dudley and
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.
                                   13-23

-------
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  (1942),  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 raucous membranes.  Animals that were removed from the acrylonitrile



atmosphere prior to the gasping stage recovered completely with no therapeutic



attention  (McOmie,  1943).   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,  1942).  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 (1942)



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-
                                      13-24

-------
bolism.  Dudley £t al.  (1942) 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 .et 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,  1943). The persistence of  thiocyanate in the

serum  was proportional  to  the level of  exposure,  with blood levels normal  at
                                             •2
2.5  hours after an exposure of  22  ppm (48 mg/m ) acrylonitrile for 30 minutes and

blood  levels  not  returning  to normal  values by 12 hours  after an exposure  of

50 ppm (108 mg/m3) for 30 minutes  (Wilson  et al., 1948).  Due to the quantitative

correlation between acrylonitrile  exposure and urinary and  blood  thiocyanate

concentrations, measurement of levels of thiocyanate was suggested as a method  of

monitoring acrylonitrile 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

 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 eytochrome function and cellular  anoxia.  During cyanide intoxica-

 tion, two less toxic metabolites—cyanmethemoglobin and thiocyanate—are formed
                                      13-25

-------
 •a
 c
 a
 CO

 n

 o
 e
 o
•o
ft
 o
 I
 O

•O
 C
 n

 0)
 §

•H
rH
CO


O
en





^
o



c?
o



CO

n!
e





















































43
O
CD
CM
CM
CD
^
L,
O
43
•H
CO
I
43

1
•H
rH
CO


0
O




O

o



CM
0



























c
0

43
CO


c
CD
O
c
0
0

rH
CO

h



in
en










in
en
o





CD
43
CO


0
bO
I
•O
3
43
CO



in
in




CM
to
o



in
CM

•H
O.

CO
CD
C
•H
3
a

m
43
0.
E
CO
43
3
O
S3
43
•rl
3
CO
L,
3
O
S3

CM




L,
O


•a
CD
43
CO
Li
CD

EH



0
8










£
0

43
0
>
Li
O
43
•H
CO
!
CD 43
3 43
CO S3
O bO
O.-H
X rH
CD CO


in
«





CM
0



8
0



























§
•H
43
CO


C
CD
O
E
0
O

co
43
CO




en










in
en
o





CD
43
^3

L.
0
bO
C
iH
Li
D
•a
rH
ea
H



o





C\J
0



s
0

CO
43
•H
JO
JO
CO
K














CO


43
O.
I
CO

o


o
43

•a
CD

Li
§
o




T-










en
0

43 43
O O
CD CD


CD CD
>> >>

O O
43 43
tH -H
CO CO
CO CO
Cl) 43 43
3 -0 43
CO (U S3
o jaj bo
CD S CO


in o
en o





«- «-
0 0



8f CM
O O








C
o
o
43
in

iH


CO
•H

>>
^

co
0.


^
43
ct!
L,
•H
a.




c

B

rt
C CD
M T3



en










in
en
0










o
TJ
o
43
rH
•a
CD
1
E


in
t~
CM




O
en
o



o
o




CO
43
CO
0











B
•H
°


o
43

U3
C


o
iH

O
43

^,


.§
Li
CO




CM










^r
o
CO
CD
a
•H
43
CD

CO


O
•*•
o
43
rH
•o
33
L, 43
cO CO



en
in




t—

o



en
en
o








                                                                                       2
  'as

    SCM


    fe_?

   e*
    co  n)

   A
   43   K
    CO  >>
    CD  Li
   •a  CD
       >

    CO  O
    Li  CD
    3  Li

   5   -
       CO

    f3§
   CM -H
       CO
   •O i-H
    CD  3
   43  >
 CO  CO  C
 bO  L,  o

•83  °


SS"

rH WO 43
 CO    >H
43  >.  S
 CO  CO  O
fc. S >
                                                                                       en
                                                                                 in o  i
                                                                                 »— o> en
                                                                                  -
      =r
      o
       •
      o
in     i     in
cvj    t-    en
<— i- o    o
   43 O
   co -a
                                                                                                   a
                                                                                                   •H
        0)
        43

        CM
        CO





  1     •§
  CO     O
  Li     43
  CD
  rH     >>
  O     rH
  43     43

        O
  CD     B


        43

  >>    CO
  CO     
. I Li Iff o T) O - CO CD -fH JS CO CO 43 a >> O H CM O CO O Li 4* CO v, CO en E >> o5fe 43 CO 43 rH 3 CO CO >• C 1 43 C CO I CO O Li *i d. O E-< i O O vo >- o in CM t- ya T- t- O =r <— c\j v£> CM CM i- O CO 43 o CO O 43 CM 0) Q O Q § 13-26
-------
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 |ig/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.  Brieger ^t  al.  (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
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
                                     13-27

-------




0
rH
•H
Is
•H
S
rH
i,
O


s

•o
0
co
o
o.
w *-+
CM
to in
ti O\
cd «-

•3 •»
c •
0 .p
0
•O
0 t,
O 0
H bO
« 0
*f!
C S-.
•H m
0) 3
-P O

C *M
w^ T3
o 55
0 4J
•H a,
J3 cd
£3

s
CO
0
•a
•H

cd
o
v
CTl
m
""
0
H
cd









cd
"a
o
0
^~
bO
g
0
4-3
g
cs
o
o
•H
X!
E-*










S3
•H

0 T3
•a o
•35
cd co
>,
o





0







0
5
SH
&





•o
o
0
rH
PQ




^
s

o

t —
rH
O
(3
3"





2
3
hO O
C
0
, "]



0
r-H
•H
SL,

•H
g
•H
t
O














0.

py|




^.^





en in c~- CM
CM 1 >J3 •<- 1 CT> CM I •=*" 1 r-
• i • • I • • 1 * 1 *
CM | CO v- | t- o 1 »- 1 •=!•







£1 £l X)
coincr>T~ 05 •=)" LO o c— co
VO -^" O^ ^^ C*") l>- Ln OO ^O CTv
• ••« ••••••
vo O C^ ~J" O O ^ O O *" C








coo inLn 0'<-oCM
CT,'- OOVD OLnt~t—
«• «• cd***
T- CM O O »- vO O S-iOCOoo









t-t-f-c-T-c-.c-T-C'-cnt-








s i
p,
a,

^
•r)
c
•H







co
0
•H

0
o,
CO




oooc~-c— t-inuiLnt— c—







>> >>
0 0

o o






























0
a
co
0

D.
co
0
3
rH
cd

0
•£3
•p
3
o
t-1

•o
0
•p
o
cd
C-i
-p
1

£
cd
CO
"0
>
0
rH
£
3
co
o
a
S S ?s

CO W
3 3
cococococococococococo
•P bOM-P bObO-P bO&00 0
cdoocdoocdooxixi
0

2
0
C**3
nJ

























































cd
S
CO
rH
0.
•H

•s
C!
•H


0
0
13-28

-------
blood of all exposed animals, with the highest concentrations found in the Rhesus



monkey (4.8 [j.mol/100 ml) and the dog (2.7 |imol/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. Brieger ^t_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:




                                  rhodanese     „
      Thiosulfate + cyanide
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




(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 13-4, 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.
                                      13-29

-------
                OACRYLONITRILE
                • THIOCYANATE
                ACYANIDE
        I I I  I I I  I I I  I Illl

   0       60     120      180      240
 TIME AFTER INJECTION OF ACRYLONITRILE (tnin.)

         a. ACRYLONITRILE 30mg/kg
                                                                  DEATH
           ni I  i i
       0       60      120     180
TIME AFTER INJECTION OF ACRYLONITRILE (min.

         b. ACRYLONITRILE 75mg/kg
Figure  13-3.  Distribution of Acrylonitrile,  Cyanide,  and Thiocyanate
               in the Blood after a Single Injection  of Acrylonitrile
               (Rabbit)  (Hashimoto and Kanai,  1965)
                                    13-30

-------
                         TOXIC ACTION OF ACRYLONITRILE
                  OACRYLONITRILE
                  • THIOCYANATE
                  ACY AMIDE
                                              60 —
                                                  DEATH
   0      60      120      180      240
TIME AFTER INJECTION OF ACRYLONITR1LE (min.

    a. ACRYLONITRILE 75mg/kg 3 MINUTES
     AFTER SODIUM THIOSULFATE 320mg/kg
                                  0       60      120      180
                          TIME AFTER INJECTION OF ACRYLONITRILE (min.

                             b. ACRYLONITRILE 100mg/kg 3 MINUTES
                               AFTER SODIUM THIOSULFATE 430mg/kg
   Figure  13-4'-.
Effect of Sodium Thiosulfate on the Distribution
of Acrylonitrile,  Cyanide,  and Thiosulfate
(Rabbit) (Hashimoto  and Kanai, 1965)
                                      13-31

-------
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  0.48 ug/ml in guinea  pigs and 3.01 to  0.51 (ig/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  le.thal 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  (84/&) 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.
     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  3056 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
                                      13-32

-------
   Table 13-1O.   Effect  of Methemoglobinemia on Mortality Ratios in Albino Rats
                 Poisoned with Acrylonitrile, Potassium Cyanide, and
                 Acetone Cyanohydrin  (Magos, 1962)
Compound
     Dose
(millimole/kg)
                                            No. That Died/No. Given
                                                 Cyanide Compound
                                         Without Sodium
                                            Nitrite
                 With Sodium
                    Nitrite
Acrylonitrile

Potassium cyanide

Potassium cyanide

Acetone cyanohydrin
  2.8

  2.4 x 10"1

         ,-1
  3.7 x 10
  1.6 x 10
          -1
5/5

5/5



5/5
5/5

1/5

4/4

0/5
                                   13-33

-------
 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 metabolized only 19.4? 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 acrylonitrile«s  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 _et _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



 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,
                                     13-34

-------
I 30 —
                     OACRYLONITRILE
                     • THIOCYANATE
                     ACYANIDE
   0-
              I F  I I I I T I1 I InITT
        0       60     120     180      240
      TIME AFTER INJECTION OF ACRYLONITRILE (min.)

              i. ACRYLONITRILE 30mg/kg
                                            i i n i   n i  PI
                                   0       60      120     180
                            TIME AFTER INJECTION OF ACRYLONITRILE {min.

                                    b. ACRYLONITRILE 75mg/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

-------
U
                      OACRYLONITRILE

                      • THIOCYANATE

                      ACYANIDE
                 I i i  i i i  I i  i i i  i | i  r r ri

        0       60     120      180      240

      TIME AFTER INJECTION OF ACRYLONITRILE (min.l

          a. ACRYLONITRILE 75mg/kg 3 MINUTES

            AFTER L-CYSTEINE-HCI220mg/kg
                                                   10 —
                     Q

                     8

                     CO
                     lit
                     X
                                                 O


                                                 <
                                                 cc
                                                 I-

                                                 ui
                                                 o

                                                 O
                                                 U
                                                    C r^^j
                                                      \
          I
  0       60       120      180      240

TIME AFTER INJECTION OF ACRYLONITRILE (min.)

   b. ACRYLONITRILE lOOmg/kg 3 MINUTES
      AFTER L-CYSTEINE-HCI300mg/kg
            Figure 13-6.
Effect of L-Cysteine on the Blood  Concentrations

of Acrylonitrile, Cyanide,  and Thiocyanate

(Rabbit) (Hashimoto  and Kaiiai, 1965)
                                             13-36

-------










in
vO
3
c
a

•o
S
0
s:
a
co
rn
^°"'
id
bO
•H
c
0
m
o
D-,
,
£-,
Q
«!
>
111
1 -H O
a ain
O CO
t,
o.

rH
CD
> —
"> 1 -H
p

rH
>
•H <1>
1!








ii>
c
0 hO
•rt ^
^ "h?l
m a
40
3 0
rH t—
o in
1
J


rH
g ~
°> S -H
i 23
to



rH
S.
> 40
CH Cd
3 b
to




rH

CO "5l
c a
•H
CO O
40 C-
ra on
S** cl
CJ O
1 O
j on
rH
> co --~
•H a c
> nH -H
t 40 ^B

to


^


23
3-g
CO




rH
2
a
o
0




c-l
g
•iH CO ^
> e c
i. 'H -H
S ^ a
n ^-^




td
•H 0)
> 40
L co
3 t,
n




rH
•H

40 C C
<+•< *H O O
O C -H -H
O 40 fc-4 40
rH 40 i-l O O O
CO g >,  9CMQ, n) o a. o o a.
P3*-'H O'-'H P3 «~ *H Sl*™iH



































40
C
CO

>
cr
CO
in
,J
21
^_

X
2
«rj

ll
to
T3

1
O
co
CO
•a
§

EC
CO
O
40


O
cl
fc
O
in
Q

CM

n




o
40
J
13-37
















































m
I-l
i
•H
S
CM
O
t.
CO
Q
40
to
£
o.
co
i.
m
CO
5
!
fri
Q.

C
•H

to
C,

C
rH
rH





























































40
C
CO
S
id
I
40
•o
CO

•H
>

3
to
CO
rH
a
1
o5






























































•a
CO
CO
i.
0
c

n

CO
a

ii
o


-------
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  mmol 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  jet  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 jet  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



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
                                     13-38

-------
Table 13-12.
Concentration of Protein (PBSH) and Nonprotein (NPSH)  SH  Groups
 in Normal and Acrylonitrile-Intoxicated Animals
    (limoles SH/100 g wet tissue)  (Dinu,  1975b)
                        Normal Animals
                      Mean + Standard Error
                                       Intoxicated Animals
                                      Mean + Standard Error
Liver
PBSH
NPSH
Kidney
PBSH
NPSH
Brains
PBSH
NPSH
Lung
PBSH
NPSH
Testis
PBSH
NPSH
Adrenal
PBSH
NPSH
1604 + 28.3
412 + 16.8
952 + 40
250 + 40
968 + 27.4
103 ± 3.55
989 + 30
82 + 3
1070 + 41
237 ±3.6
528 + 20.7
127 + 4
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

-------
peroxides and subsequent disruption of membrane  function.   Following treatment

of  rats with  acrylonitrile  (two  times  the  LD50  dose administered  subcuta-

neously), increases in malonaldehyde were observed  in  the  liver.   According to

Dahle ^t 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 disrup-

tive  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  (10~3  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
                          o
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

cyanide, since cyanide  at  10~  M inhibited both potassium stimulated and non-

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-

gastrocnemius  preparation  from  frog and  observing a  rapid anesthetic action.

This  anesthetic  effect rapidly disappeared following removal of  acrylonitrile.
                                      13-40

-------
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 £t 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 et 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 4 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.


13.2.1.4  Acute Topical Irritation and Toxicity


     McOmie  (1943) applied acrylonitrile to the shaved  abdomens  of rabbits at

                                                               2
doses of 1.0, 2.0,  and 3.0 ml/kg to areas of  100, 200, and 200 cm , respectively.


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
                                      13-41

-------
contact with the abraded skin of rabbits by absorption into a cellulose pad, the
LD50 was 0.25 ml/kg  (Roudabush jit  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 0.46 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,  1943).  One hour following treatment,
mild conjunctivitis  was observed with  no  clouding of  the  cornea or pupillary
damage.  The animal had completely recovered by 24 hours.
     Skin irritation  as a result  of contact with acrylonitrile  has also  been
observed in humans (Wilson ^t _al., 1948).   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  (1942) 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.
13.2.2  Subchronic Toxicity in Non-Human Mammals
     In a preliminary study,  Dudley  _et  al. (1942) exposed  4 rhesus monkeys and
2 dogs to an atmosphere  containing 56 ppm (122 mg/m ) acrylonitrile for  4 hours a
day, 5  days a  week,  over a  4-week period.   The rhesus  monkeys showed  no overt
signs of toxicity.  Of the two dogs  treated,  one died during the first exposure
                                     13-42

-------
period and the second had intermittent paralysis of the hind legs following three



of the exposure periods (the fifth, thirteenth, and fourteenth).  In subsequent




experiments, 16 rats, 16 guinea pigs, 3 rabbits, and 4 cats were exposed to 100



ppm (21? mg/m3) acrylonitrile, and  16  rats,  16 guinea pigs,  4 rabbits, 4 cats,




and 2  rhesus  monkeys  were exposed  to  153  ppm (332 mg/m ) acrylonitrile.   The




exposure conditions were similar to those described above except the period was




increased to 8 weeks.   At the  100 ppm (21? mg/m ) 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 similar to that observed following acute expo-




sure,  with the dog being the most sensitive and rats and guinea pigs  the least



sensitive to acrylonitrile.



     Dudley et al.  (1942) 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



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
                                     13-43

-------
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
                                                                             o
ooworkers  (1942)  using animals  that  had been exposed to  153 ppm  (332 mg/m )

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 42? and  21%  above  control, respectively.

Minami _et 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

ppra  (43 mg/m  )  acrylonitrile for  a  period  of  8 weeks.   Due to  the   small

differences observed and  the relatively  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 ^t al.,

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 parenchyma! cells of the liver.  The  only hemato-

logical change noted was  leukocytosis.

      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
                                      13-44

-------
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 24 ng (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_et 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




necropsy -at  the end of  the  study,  autopsies were performed in order to  detect



gross or microscopic pathological conditions.



     In the  first study (Quast jit 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-
                                      13-45

-------
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 4 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  45 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



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
                                     13-46

-------
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



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
                                     13-47

-------
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 al. (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  3.42,   8.53, and  21.18 mg/kg/day   for  male  rats   and  4.36,  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 probable  cause of the concomitant decrease in food consump-

tion and body weight.  Chronic renal disease, which was common in older rats of

this strain, occurred less frequently in  treated rats probably as a result of the

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.
                                     13-48

-------
 CO-
o
CO
o>
4J
 to
 en
           E

          Cb
                                                          rtcdcOfljcdcOrtrtcortrtcdtOrtcD
 C

 CIS


1
cj


 0>

 n>


 if
•H
X.
 C
 CO

X
I
•o
 g
•o
CO
CD
Q
                 O
                 O
                 m
        e  v
        O.EL.
        o.
                    V


                    CO
                                                                            CO

                             rHojmsrirxiot^.t—coo-io^cSooo
                                                                 rH r-t »H












                       ojojojcnvot—cr\nncorrt~eorHfnooo
                         •   •••••.••••••••••    CO





















                      ic\jmrnrr)^irtiricy»rH3-rococ\jine--cocri
                                                                            CO









                liHrHrHr-ii-lrHrHC\Ja-VCO>f>inO\C\JCOe*mvO


                                         i-HrHc\jmm.a-irivDt—coco


















                                                 i O C^ O COCO CM IT* C—
                                                                            CO










                                                                            c—
oooooocvc\jc\jcy^-a-vDo^t-c—oojcot-pr^c^-c^    :3'
                                                                                                                        C3

                                                                                                                        o
                                                                                                               5
                                                                                                                        4J
                                                                                                                        o
                                                                                                                        CO
                                                                                                                        i.
                                                                                                                        CD
 •o
 c
 CO

 V
        §
        o
                                                mmmmmmmcoco iTirocoinminpntnino
                                                 •  •«*••••••***••••••    o
                              OOOOOOrHrHrHrHrHrHrHmmC*-rHPr»rat-C\JVD*Ob-m    CO
                                             mmmmco IAOOGO  oooco mmmo o inoo inro
                                              •  ••(,«••••••••*••••*•    o
                                            ,,H(-4,-)C\jrnf- cocoorneovD\orHoot*-rnt».rH    co
                                                                     rHrHrHCJCn^T IH^OvOCOCO C^
                                                                                                                        I
 m

 *T
              CD
              H
     -f
                                       oooooooooooooooooooootn
                                                    ---
                              en VD i
                               I   i
                              O rH I
                                                   o
                                                   rn
                                                                             &.
                                                                             CD  n
                                                                             xi  -^
                                                                                §co
                                                                                cc
                                                                                                           rH O
                                                                                                           CO
                                                                    13-49

-------
     Most of the 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 acryloni-



trile.   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 consump-



tion, 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), prolifera-



tion of 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 neoplasias.



13.2.4    Summary and Conclusions - Acrylonitrile intoxication in humans results



in irritation of the  eyes and nose,  weakness, labored breathing,  dizziness,



impaired judgement,  cyanosis,  nausea,  and convulsions.  The TLV  for acryloni-



trile is 4.5 mg/m  (2 ppm) for  humans. 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 resis-



tant and  the dog is  the most sensitive.  In animals,  effects  of intoxication



include respiratory changes, cyanosis, convulsions, and death.  In rats, the LD



for acrylonitrile is  between  80  and  113 mg/kg (Knoblock ^t al., 1971; Smyth et



al.,  1969 ). There is some evidence that  acrylonitrile produces abnormal function
                                     13-50

-------










r-«
0
4J
W
bn
c-t


2
£>
cd
G «:•>.
o o
£S
0 *~
.Q -
-S cd
•a -P
0 CD
0 4->

•rH Cd
CO 3
G O1
O **••"•''
O
CD
CO rH
CD -H
&Q S-,
G -P
cd *H
X3 G
0 0
rH
-p >>
C S-i
cd o
O =Ti
CM CM
•H 0
bO
•H
CO
.
•~*
-
OO
r—
0
cd




















co
W>
G
•H
•a
G
•H
fa
CM
O

0
W
cd
u
•o
0
-P
8
o.
CO









co
bO
C
•H
•s
•H
fa
4-5
cd
0
fat)
•H
CO






CO
0
CO
K*>
H
cd
•§
CM
O
0
£»}
H





CO
su
o
s

0
.p
s
0
o
rH
3

S

CM
CO
CO
o
rH
•a
o
o
S







-P
G
o
0 0
rH 3
rH rH
CD O
0 >
•0 rH
8rH G
CD -H
rH 0 XJ
X) O
0 0 bO
-P .id 0
•H O 8
X! Cd 0
5 O.XJ
000
co co co
cd cd cd
0 0 0
o o o
0 0 0
« « Q








O
r-l
S
a
0
33








g
•H
1
CO
g
0

£4
0
4J
cd
•o
0
o
•H
W
0







>>
4-5
'>
£
bO
o
•H
•H
S
8-
urine
•g
CO
cd
o
£j
M








.3
co
>>
H
cd
.s
t,
£3
G
O
•H
-P


3
to
O

•8
o
CM
•o
cd

J_t
0
4-5
cd
•o
0
O
•H
co
0
K









a
0
bO
g
4J
•H
G
cd
s
3
•o
O
5
XI
•8
w
ns
0)
o
S-3
M


£
co
•rH
S
0
si
rH
cd
o
G
rH
O












co
•H
gj!
•rH
f-\

rH
O
C
C
4J
g







O
•H
3
fe
D.
O
•H
4J
3
rH
bO
0 0
co co
„§
0 -H
CO S
cd cd
8cd
s«^
Q -P














c
o
•H
4J
D.
B
3
CO
O

O
o
CM
•a
cd

5^,
0
4->
g
•o
0
-P
O
•H
4->
CO
0
K











.B
co
CO
0
CO
o
•H
G^
l"g
-P co
cd cd
•H 0
cd o
S 3)
W Q








13
0
H
O
£3
4->
cd
(X,







0
5
bO
•rH
4J
•H
&
•a

co

o

rM
S^
cd 0
£H rH
cd
W












0
•H
CO
3
O
0
O
rH
cd
a
rH
cd
G
QJ
Q




















•o
G
G
5
4J
o.

co
§ra
s.,

g
•O J3
O 4-5
O
CMg,
T3 'r-\
0 4J
-p cd
0 i,
•H 0
CO 3
0
«




•o
•H
3
r-H
CM
0
^
rH
CD
rH
B
cd
•s
o
X)
to
cd
£
o
G
M














G
O
•H
4-5
O.
a
3
CO
O

0
4-5
s

cd <~{
J \


p -o
>
T3 rH
0 S-,
4-5 Cd
O 0
•H
M cd
0


.-





0
co
to
•H
T3
rH
o
•H
I
o
•a
0
to
cd
o

o

























4J
cd
4J
co

^
cd
G
1
-p
•rH
£_
+5
a
o
o







>>

a
4-5
cd
cd
rH
rH
0
O
3
cd
o.
0
.C
to
cd
o

M


















0
co
cd
cS
X)

a

o
•H



•G G
° S
C 4^
•H -H
0 4J
CO 3
cd C
0 O
f™\ ^1





CO
3
O
>
£-,
e
TD
2
0 4-5
cd -H
° I
co G cd
GO bO
bO-H t,
•H 4-5 O
CO O
ID 
-------
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 basic viewpoints and definitions of the terms "teratogenic" and "feto-



toxio" were  summarized by the Office  of Pesticides  and Toxic Substances  (U.S.




EPA, 1980c) as follows:



     "Generally, the  term 'teratogenic'  is defined  as  the  tendency to produce




physical and/or functional defects in offspring _in utero.  The term "fetotoxic"



has traditionally been used to describe a wide variety of  embryonic and/or fetal



divergences from the  normal  which cannot be  classified  as  gross  terata (birth



defects) — or which are  of unknown or doubtful significance.  Types of effects



which fall under the very broad category of fetotoxic effects  are death, reduc-



tions in fetal weight, enlarged renal pelvis, edema, and  increased incidence of



supernumerary ribs.   It should be emphasized,  however,  that  the phenomena of



terata and  fetal toxicity as currently defined  are  not  separable into precise



categories.    Rather,  the  spectrum  of  adverse  embryonic/fetal  effects  is



continuous, and all deviations from the normal must be considered  as examples of



the developmental toxicity.  Gross morphological  terata represent  but one aspect



of this  spectrum, and while the significance  of such structural changes is more
                                      13-52

-------
readily evaluated, such  effects  are not necessarily more  serious than certain



effects which are ordinarily classified as  fetotoxic—fetal death  being the most




obvious example.



     In view  of  the  spectrum of effects at  issue,  the Agency suggests that it




might  be  useful to  consider developmental  toxicity  in terms  of  three  basic



subcategories.  The first subcategory would be embryo or fetal lethality.  This




is, of course, an irreversible effect  and  may  occur witsh or  without the occur-




rence of gross terata.  The second subcategory would be teratogenesis and would



encompass those  changes  (structural and/or functional) which are induced pre-



natally, and which are irreversible.  Teratogenesis includes structural defects




apparent in the fetus, functional deficits which may become apparent only after



birth,  and  any  other long-term  effects  (such  as  carcinogenicity) which  are




attributable to in utero exposure.  The third category would be embryo or fetal



toxicity as comprised  of those effects which are potentially reversible.  This




subcategory would therefore include such effects  as weight reductions, reduction



in the degree of skeletal ossification, and delays in organ maturation.



     Two major problems with a definitional scheme of this nature must be pointed



out, however.  The first  is that  the reversibility of any phenomenon is extremely



difficult to prove.  An organ such as the kidney, for example, may be delayed in



development and  then  appear to  'catch  up1.  Unless a series of specific kidney



function tests is performed on  the neonates,  however, no conclusion may be drawn



concerning  permanent  organ function  changes.    This  same  uncertainty  as  to



possible long-lasting after-effects from developmental deviations is  true for



all examples of  fetotoxicity.  The second  problem is that the reversible nature



of embryonic/fetal effects  in one  species  might, under a given agent,  react in



another species  in a more serious and  irreversible manner."
                                      13-53

-------
     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 ^t ^1., 1978, while a  detailed report of the results of the



gavage  treatment was presented by Murray  _et  al.,  1976).   Acrylonitrile  was



administered at doses of 10,  25,  and 65 mg/kg/day  per JDS or  by  exposing  the



animals to an atmosphere of  40 or  80  ppm (87 or 147 mg/nr) acrylonitrile for 6



hours per day.  It was estimated by measuring blood levels of acrylonitrile and



its metabolites that an exposure to 80 ppm (174  mg/m  ) acrylonitrile for 6 hours



was equivalent  to a single 23 mg/kg dose of acrylonitrile.  On day 21 of gesta-



tion, 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
                                     13-54

-------
CO
CTv


*
cd
4J
CD

cd

52
CD
•rH


4-3
'c
o
rH
£-,
0
"^
C
CD
•H
O
CO
cd
K

CM
O
CO
%
4J
•H

(f
o
i
g
0
!o
1
o

rH
Cd

rH
4J
CD

H-.
0
CD
O
g

•H
O
a
H
B
in

i
CO




ft
cd
E-i


8
D.
O,
^
O
•H
4-1
cd
"cti
•g
•H
J3

C
CD

•H

CD
rH
•H
i,
-p
•H
a
o

L
O








S,
cd
crt
"1
4J
S -S
CD bO
> -H
•H CD
CO 3
CD >*
rH *O
'S S
•H W
C ^
*KT
>, e
« ^^ *
^


























0
CD







O





O






in
wo




in
OJ






2




















Type of
Malformation







in in
co in
wo wo
O CO



WO WO
CO CO
*x. ' —

=r i-


CO CO
_, CO CO T3
T3 %- -s^ CD
fi ^0 S
'd J=T «- CD
S CM
iS 



CO &4


O 4J O
Visceral malformations
Right-sided aortic ar
Missing kidney, unila
Anteriorly-displaced








f 	 ^
OJ
OJ




^_^
o

o



,-,
o
0





cd
wo
OO




CM
OJ




^_^
0
o





^.^
9





•o
CD
-U
cd
•H
o
o
CO
Skeletal malformations
Missing vertebrae (as
with short tail)








	 ^ ^^
CM «-
C— »-




^-^ ^-N
«- O

CM O



^-,^^
*— o
OO O





o o
o o




OJ O
C^ CD




,-^,-x
o o
o o





^^ ^^
*- o
c— o



•H
cd
0.

cd

•a
c
cd

Missing two vertebrae
of ribs
Hemivertebra








^^
wo
-




^^^
OJ

CO



^
*—
oo





cd
CO




o




r~.
o
o





, 	 ^
OJ
CO













Total malformed




^^
t-
2^
'*•''
i-H
CD
O
EC
13
r«
§
c
cd

CD
CO
cd
K
J3
T3
•rH
•rl
Q
a
CO
cd

4J

C
g
0
o
rH

CD
bO
C
•H
CO

in
o
o
fc
—
CD
3
cd


'Q
i,
a.
o
a
1
a

&4



rH

C
cd
Values differ signific





































rH
cd

•g
0
CO

4J
•rH
9
CO
CD
9
CD
CM

C
•H
0

•o
CD

Li
CD


O
This malformation was


^^

t—

^
rH
(jj
o
33
1
§
a
CD

cd
CC

J?
T3
CD
•rH
CM
TH
1
a
S


CO
CD
4J
1
8
rH
•iH
CD

JJ
bO
B
•H
§

S

o
M
P-.
	
<3>
d
r-H
cd
>

Q
1
o
o
CD
a

M
CM

!»
rH


cd
O
•H
CM
•H
S
•rH
CO
CO
L.
CD
CM
CM
•H
•o
3
3
cT
13-55

-------
acrylonitrile, there was no  significant single major malformation observed; how-


ever, when malformations were  considered  collectively,  there was a significant

                                                     o
(P = 0.06) but slight increase  in the 80 ppm  (174 mg/nr) exposure group.  At the

high dose level in the gavage-treated animals, there was a significant increase


in  aoaudate  or  short-tailed  fetuses.    The majority  of other  abnormalities


including short  trunk,  anteriorly displaced ovaries, missing ribs,  and imper-


forate 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 (174


mg/m )  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.


     Acrylonitrile  was  also evaluated  in  chick  embryos  (Kankaanpaa ^et 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 concentra-


tions of 0.001, 0.01,  0.1,  1.0,  and 10 jimol.   A  total of  25  |il  was  used for
                                      13-56

-------
injection.  The only observable adverse effect of treatment was embryo toxicity ,
(dead embryos and empty eggs)  of  approximately 100?  and 44? 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 \rnol dose was toxic
(60?).  There was no evidence of a teratogenic  effect of acrylonitrile 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 ini:
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
                                       13-57

-------
            Table 13-16,
Pup Weight on Days 4 and 21 of Lactation
  (Beliles et al., 1980)
Generation
Fla


Fib


F2a


F2b


F3a



F3b


Dose Level
(ppm)
0
100
500
0
100
500
0
100
500
0
100
500
0
100

500
0
100
500
Mean Pup
Day 4
11
10
9a
10
9
10
11
10
9a
11
10 •
9
10
9
o
8
10
10
8a
Weight (g)
Day 21
42
40
28a
38
35
34a
39
39
30a
51
46
30a
43
43
a
30a
49
46
32E
p<0.05 using Students t-test.
                                 13-58

-------
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.


         Table 13-17.  Tissues Examined for Histopathologic Changes in
                       the F3b Litter (Bellies 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
13-3.1.   Summary and Conclusions

     Acrylonitrile adversely affected pup  survival  following exposure of preg-

nant 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  (Bellies  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 acrylonitrile for  6

hours/day on days 6 to  15 of gestation had no statistically significant changes

in reproductive  success  or fetal development (Murray ^t al.,  1978).   Only the

pups of rats  administered  acrylonitrile  per £s (65 mg/kg) for days  6 to 15 of

gestation had an increase in malformations  (Murray jst 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
                                     13-59

-------
several studies have been conducted to evaluate the ability of acrylonitrile to



cause adverse  teratogenic,  embryotoxic,  and reproductive  effects,  the limita-



tions of the available data do not allow for a full assessment of these effects.
                                      13-60

-------
13.4  MUTAGEEICITY




     The objective of this mutagenicity evaluation is to determine whether




acrylonitrile has the potential to cause mutations in humans.  This




qualitative assessment is based on available information derived from several




tests that measure different types of genetic damage (e.g., gene mutation,




chromosome damage, DNA strand breakage).  These tests include assays in




bacteria, Drosophila, plants, cultured mammalian cells, and whole rodents.




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




13.4.1.1  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 coli 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 at concentrations of 0,




75 umoles (4000 ug)/plate, and 150 umoles (8000 ug)/plate produced a weak




dose-related mutagenic response in all strains except WP2 lexA.  It should be




pointed out that acrylonitrile is volatile and that no precautions to prevent




evaporation were reported.  If evaporation occurred under the test conditions,




the responses observed may represent an underestimation of the mutagenicity of




acrylonitrile in this system.  The magnitude of the responses are also




difficult to interpret because no positive control data were given, and it is




therefore unclear how well the test system was responding to mutagen treatment
                                     13-61

-------
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 umble/plate




compared with  a threefold increase for WP2 urvA and a twofold increase for WP2




uvrApolA.  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 toxicity of  acrylonitrile in the plate test, Venitt et al.




(1977) performed a fluctuation  test, which is reputed to be a sensitive assay




for detecting  low levels  of mutagens, at concentrations that ranged from 4 x




10""^ 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; data




not presented), and WP2 uvrApolA in the plate test.  In contrast to the




results in the plate tests, WP2 uvrApolA was more sensitive to the mutagenic




effects of acrylonitrile  than WP2 in the fluctuation test; 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 misrepair of DNA damage, which is




believed to be associated with  the generation of DNA strand breaks (Green and




Muriel, 1976).  When R-factor plasmid pKMlOl was transferred to WP2 (strain




designated WP2P) to increase the sensitivity of this strain, WP2P 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 umoles  (212 ug)/plate to 40 umoles (2122 ug)/plate in the




plate test.  These results in E_. coli strain WP2P further support the notion




that acrylonitrile is producing mutations via misrepair of DNA damage because




Salmonella typhimurium strains  containing the pKMlOl plasmid have been shown
                                      13-62

-------
to be more sensitive to the mutagenic effects of chemicals whose mechanism of




action depends on error-prone DNA repair (McCann et_ &L., 1975).




    Venitt et_ al^. (1977) indicated that acrylonitrile in the presence of an




exogenous activation system (S-9 mix prepared from livers of Aroclor




1254-induced rats) did not cause an increase of revertants of E_. coli; thus,




according to the authors, acrylonitrile is primarly detected as a




direct-acting mutagen in these bacteria.  Because data were not presented in




this report, however, it is uncertain whether Venitt et al. included an




activation-dependent control mutagen in the study to demonstrate that the S-9




fraction used was functional.  However, Dr. D.E. Levin of Bruce Ames*




laboratory also found that acrylonitrile caused a dose-related mutagenic




response [10-fold increase above the background number of revertants at 800 ug




(1 ul)/plate; personal communication, December 1982] in the absence of rodent




liver S-9 mix when evaluated in the Salmonella plate test using a recently




developed tester strain called TA102.  This new J3_. typhimurium strain is much




like the E_. coli WP2 strains in that TA102 contains a A:T base-pair at the




site of mutation (Levin et_ a^L., 1982).  This is in contrast to all previously




developed Salmonella tester strains (e.g., TA1535, TA98, TA100, etc.) that




have been used in mutagen evaluation; they detect mutagens that damage G:C




base-pairs at the site of reversion.  The results from the "G:C" strains are




discussed below.




     Venitt et al. (1977) reported that acrylonitrile was not detected as




mutagenic in Salmonella tester strains hisD3052, TA1535, TA100, TA1538, and




TA98 using either plate incorporation assays or fluctuation tests.  Although




mutagenic activity was reported to be detected in strain hisG46 using a




fluctuation test, the results were reported as "erratic and statistically




nonsignificant."  Because no protocol was presented, it is not known whether
                                      13-63

-------
mammalian  liver activation was  employed  in  the  Salmonella assays or if




precautions to prevent evaporation were  taken.  In addition, no data were




presented  to support  the negative results reported in Salmonella.  Therefore,




Venitt's conclusion of nonmutagenicity in the Salmonella strains cannot be




evaluated.




    McMahon ^t_ ail^. (1979) reported that  acrylonitrile (purity not reported)




was mutagenic in both E^. coli and J3_. typhimurium when screened in a




qualitative gradient  plate assay.  However, they did not indicate if




acrylonitrile required metabolic activation in  these strains, the




concentration^ ) at which activity was observed, and which bacterial strains




were reverted.  Therefore, the  conclusions  cannot be evaluated.




    Milvy and Wolff (1977) reported positive results for acrylonitrile (99%




purity, impurities not reported) when tested as a vapor in a modified




Salmonella/microsome  assay using standard tester strains.  The mutagenic




effects of acrylonitrile were observed only in  the presence of metabolic




activation (S-9 mix prepared from livers of Aroclor 1254-induced male




Swiss-Webster mice);  data from  tests in  the absence of S-9 activation were not




presented.  Acrylonitrile caused about a doubling in background revertants for




tester strain TA1535 when cells were exposed to vapor or when cells were




preincubated.  Although the data suggests a positive result, the




interpretation of this study is difficult because of deficiencies in the




reporting of the data:  positive control data were not reported and the




revertants for each plate were  not reported (thus the variation in revertant




counts is not known).  In addition, Venitt (1978) pointed out errors in




calculation found in this report.  The studies by DeMeester and coworkers




(discussed in the following paragraph), however, provide more convincing




evidence for the mutagenicity of acrylonitrile in S^. typhimurium (containing
                                     13-64

-------
G:C base—pairs at the site of reversion).

    DeMeester et_ al. (1978) and other investigators (Venitt et_ al., 1977) have

noted that acrylonitrile is not detected as mutagenic in Salmonella tester

strains (which have G:C base-pairs at the site of mutation) when assayed in a

"classical" plate incorporation assay with or without liver activation.*  But

DeMeester et_ al. (1978) were able to detect mutagenic activity in Salmonella

when the bacteria were exposed.to acrylonitrile as a vapor in a desiccator

only in the presence of a S-9 mix.  The requirement of mammalian liver

activation for a positive response in this case and the positive results with

the new Salmonella tester strain TA102 (Levin, personal communication) and 15.

coli WP2 (Venitt et_ 3Li_., 1977) in the absence of exogenous liver activation

may be ascribed to the ability of acrylonitrile to react directly with the A:T

sites in the DNA to form an adduct that reverts E_. coli WP2 and Salmonella

TA102 but requires biotransformation to revert the Salmonella "G:C" strains.

    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 bacteria were exposed for one hour.  Using a gas

chromatograph, the concentration of acrylonitrile was determined 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.  The cellular toxicity of 0.15 liters 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 0.15 liters

of gaseous acrylonitrile for 2 hours or 0.24 liters for one hour (the method
    *In 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.
                                      13-65

-------
used to measure toxicity was not described).  Reversion to histidine




prototrophy was reported for the base-pair substitution-sensitive strains




TA1535, TA1530, TA100, and TA1950 and for the frameshift-sensitive strains




TA1538, TA1978, and TA98 in the presence of microsomal activation (Aroclor




1254-induced rats, 300 ul S-9/ml mix).  Acrylonitrile had the most pronounced




effect on TA1530, TA1535, and TA1950, with TA1530 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.  Strains TA98, TA1978, and TA100 were only weakly reverted




(approximately a twofold or less increase over spontaneous level).  Negative



results were found with the strains TA1975, TA1532, TA1537, and hisG46.  A




fluctuation test was conducted to confirm the sensitivity of TA1530 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 evaluated only 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, and the increase appears to be




reproducible.



    Available studies show acrylonitrile to be a mutagen in bacteria.




Although the data suggest its mutagenicity to be weak to moderate, there is




not yet adequate basis for categorizing the potency of acrylonitrile.  It is




difficult to quantify the mutagenic activity of acrylonitrile from the studies




of DeMeester et_al. (1978) or Venitt ej^ al. (1977), because no concurrent




positive control chemicals were evaluated, and in the DeMeester eit_ aJL. (1978)




study only one dose was tested.  (The data from Dr. Levin is not yet




available.)  Because of this, Monsanto (Kier, 1982a) conducted a modified
                                      13-66

-------
Liquid suspension, assay using standard tester strains of S_. typhimurium




exposed to several concentrations of acrylonitrile and concurrently tested




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 2-hour exposure to concentrations of acrylonitrile




that ranged from 20 ug/ml to 20,000 ug/ml, the cells were washed, mixed with




top agar, plated in petri dishes, and incubated 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




TA100 with S-9 activation but not in TA98 and TA1535.  No increased responses




were observed without S-9 mix.  In TA100, average revertant counts at the




highest dose (6000 ug/ml) tested was about 1.5-fold over the solvent control.




Therefore, under the liquid suspension treatment conditions used by Monsanto,




acrylonitrile caused, at most, a marginal increase in the number of




revertants.  However, these studies alone do not allow for a quantitation of




the mutagenic activity of acrylonitrile.




    The bacterial studies by DeMeester et_ al. (1978) and Kier (1982a) indicate




that acrylonitrile was detected as mutagenic in the Salmonella "G:C" strains




only when metabolized ±n vitro by 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 by  testing urine from rodents exposed to acrylonitrile in




Salmonella.  These authors reported that urine (0.1 ml per plate)  from both




rats (adult male Wistar) and mice (NMRI) treated with a single intraperitoneal
                                      13-67

-------
 dose  of  acrylonitrile  (30 mg/kg; purity 99%,  impurities not reported) was




 mutagenic  in  Salmonella  tester  strain TA1530  (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 Aroclor-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, B-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




 acrylonitrile-exposed  rats and  mice was mutagenic in Salmonella, and




glucuronoconjugation 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 (intraperitoneal  injection of 45 mg/kg body weight)  was not




mutagenic in Salmonella TA1535.  However,  these negative findings are




questionable because data were not presented in the report and several known
                                     13-68

-------
mutagens (e.g., dimethylnitrosamine, 3-methylcholanttirene), which require




metabolic activation, and the direct-acting mutagen methyl methanesulfonate,




were also negative in this study.




    Studies using Salmonella tester strains that contain G:C base-pairs at the




site of mutation indicate that metabolic activation of acrylonitrile produces




a mutagenic form(s).  A potential reactive metabolite of acrylonitrile is an




epoxide.  Monsanto (Kier, 1982b) evaluated the acrylonitrile epoxide




2,3-epoxypropionitrile (purity 97.4%), in a modified liquid suspension




Salmonella assay (described previously) and found it to be mutagenic in the




standard tester strains TA100, TA1535, and TA98 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 TA100 with S-9 mix,




a fourfold increase in revertant counts over background counts was seen at 400




ug/ml.  This epoxide does not appear to be a strong-acting mutagen under the




liquid suspension treatment conditions used in this study when compared to the




positive control mutagens 4-nitroquinoline-N-oxide, 2-acetylaminofluorene,




2—aminoanthracene, and benzo[a]pyrene.









13.4.1.2  Drosophila




    Although acrylonitrile has been shown to cause point mutations in




bacteria, its mutagenicity in a eucaryqtic organism has not been adequately




examined.  The only available eucaryotic gene mutation report is 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
                                      13-69

-------
 slightly  below a lethal or sterile dose.)   Although the  authors  report  the




 results as nonsignificant,  the  data can be  considered  as suggestive  of  an




 increase  in  the mutation frequency by  a factor  of  three  over  that  of the




 spontaneous  frequency.   Several deficiencies were  found  in  this  report.




     1.  A small sample  size of  flies was tested (1297  chromosomes).   In order




        to preclude a doubling  in  mutation  frequency,  7000  chromosomes  would




        have to be tested (Lee  ^t^  aJL.,  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.




Acrylonitrile  has, therefore, not  been  properly evaluated in  this  system, and




retesting is necessary  to permit a final judgement  on  the mutagenic  activity




of acrylonitrile  in Drosophila.








 13.4.2  Chromosomal Aberration  Studies




     The ability of acrylonitrile to induce  chromosomal aberrations in vivo has




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 je_t 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)




was  found in mouse bone marrow  cells.  Male mice were  given acrylonitrile by




gavage for 4,  15, and 30  days each  at doses of  7, 14,  and 21 mg/kg/day, and at
                                      13-70

-------
 least 1200 cells  were examined per treatment group.   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 LD50,  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  LD50).   Because




 current  evidence  (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 U^Q).   It should be noted that concurrent  or historical




 positive controls were  not reported in the study by  Rabello-Gay and Ahmed




 (1980).




     Leonard jit_ _al.  (1981)  evaluated the  clastogenicity  of  acrylonitrile




 (purity  not  given)  in mouse  bone marrow  cells  in vivo.   These authors




 conducted  both  chromosomal aberration and  mlcronucleus  analyses.   The




 percentage of chromosomal  aberrations or micronuclel  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




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
                                     13-71

-------
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 ^t_ aJ^. (1981) also conducted a mouse-dominant lethal assay with




acrylonitrile.  This test evaluates the DNA-damaging effects of chemicals on




germinal tissue.  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) conducted a cytogenetic analysis of lymphocytes




from 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




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 average exposure level 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
                                      13-72

-------
examined.  Types of chromosomal aberrations were not classified in the report

except for gaps and isogaps.  Frequencies of 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,

chromosomal damage was not detected in these workers.

    Loveless (1951) and Kihlman (1961) reported negative findings for

chromosomal effects in plants (Vicia 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 per se but

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.

(1981) provided evidence that acrylonitrile can covalently bind with DNA in

vitro.*  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


      *Acrylonitrile has been shown to react with certain minor tRNA
 nucleosides, and,  at  a  slower  rate, ribothymidine and  thymidine (Ofengard,
 1967  and 1971.
                                       13-73

-------
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




tat liver microsomes were used for  metabolic activation.  Apparently, the in




vitro rat brain microsomes do not metabolize acrylonitrile as effectively




under the conditions tested.  Because  information  was not given concerning the




freshness of the human autopsy liver species used  to prepare the microsomes,




the ages of the human subjects that the livers came from, and the cause of




death of the human subjects, it is  difficult to conclude that human liver



raicrosomes do not effectively metabolize acrylonitrile.  These authors also




showed that the rat  liver microsomes (or a reconstituted cytochrome P-450




system) metabolize acrylonitrile to 2,3-epoxypropionitrile, an epoxide.  This




epoxide was shown to be stable and  also to 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




2,3-epoxypropionitrile at a  greater rate than  it conjugated acrylonitrile.




Human liver and rat  brain cytosol preparations reacted with 2,3-epoxy-




propionitrile at a much lower rate  than did rat liver and not at: a detectable




rate with acrylonitrile.  Therefore, if acrylonitrile was activated to




2,3-epoxypropionitrile and reached  the brain via blood circulation, this




epoxide may not be effectively inactivated.  It should be cautioned,  however,




that these are in vitro tests, and  studies using radiolabeled acrylonitrile of



high specific activity are needed in whole mammals to elucidate the events in




vivo.  Also,  because of the  low level  of DNA binding detected in these in
                                      13-74

-------
vitro studies, it would be more convincing and informative if specific DNA




adducts were characterized instead of simply measuring covalent DNA binding.




These results, however, are consistent with the results obtained from




mutagenicity studies using the Salmonella "G:C" strains and in Chinese hamster




ovary cells (to be discussed) 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




an EPA Science Advisory Board Meeting (August 2, 1982), Monsanto indicated




that no binding of acrylonitrile to DNA was detected in these experiments.




Although Monsanto made the preliminary data available to EPA, without details




of the protocol and additional data, a definitive judgement cannot be made




regarding these whole mammal tests.  It appears, however, that the specific




activity of the radiolabel was too low to preclude the possibility of DNA




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 reacts with DNA, it may ultimately cause genetic 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, shifts in




sedimentation patterns were,observed.  This effect 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 noncarcinogenic chemicals.




They concluded that the results are suggestive of carcinogenicity.  However,
                                      13-75

-------
because  the  toxicity of acrylonitrile at the concentrations tested was not




given, it is uncertain whether the DNA damage observed at 200 or 400 ug/ml is




a reflection of premutational damage or of nonspecific toxicity.  It should be




noted that these data were generated in the absence of an exogenous metabolic




activation system.




    Dr.  G. Williams evaluated the potential of acrylonitrile to damage DNA by




measuring DNA repair synthesis in a rat hepatocyte primary culture (personal




communication, 1982).  A major advantage of the intact hepatocyte test system




is that  it provides a better approximation of the activation/deactivation of a




chemical substance as it would occur in the rat liver in vivo than does an




exogenous S-9 system.  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 biotransform 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 (SCE) 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




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 at five concentrations, from 10~'M up




to 10~^M, did not produce SCE in CHO cells  without an exogenous metabolic




activation system.   In contrast,  acrylonitrile caused a progressive
                                     13-76

-------
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" Tl




acrylonitrile.  It should be pointed out that coculturing hepatocytes with the




CHO cells resulted in an elevated background level of SCE in CHO (twice that




normally observed).  Nevertheless, in two separate experiments a dose-related




response was demonstrated.  These results indicate that the rat liver cells




can mediate the activation of acrylonitrile and that the reactive




metabolite(s) apparently reaches the CHO cells where it causes 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 in




cultured mammalian cells and DNA binding in vitro.




    There is evidence that acrylonitrile can cause point mutations in both




Salmonella typhimurium and Escherichia coli.  When A:T base-pairs (i.e.,




Salmonella TA102 and JE. coli WP2) are at the site of reversion, acrylonitrile




is mutagenic without an exogenous liver activation system; however, when




acrylonitrile is tested in Salmonella tester strains with G:C base-pairs at




the site of mutation, the presence of an S-9 activation system is required to




detect mutagenic activity.  The epoxide of acrylonitrile, 2,3-epoxy-




propionitrile  (a potential metabolite), has been  shown to be mutagenic in




Salmonella.  Because acrylonitrile (and the metabolite) can cause point




mutations  in  bacteria, it may  cause  point/gene mutations in other
                                      13-77

-------
organisms as well.  Results of acrylonitrile from a sex-linked recessive




lethal test in Drosophila provide some support for this expectation in that




the data are considered suggestive of an increase in mutation frequency over




that of the spontaneous frequency by a factor of three.  Further testing,




however, is needed in other eucaryotic organisms to confirm the ability of




acrylonitrile to cause point/gene mutations.




    Acrylonitrile has not been shown to cause chromosome aberrations.  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 bone




marrow cells.  In addition, no apparent chromosomal damage was detected in




peripheral blood lymphocytes from workers exposed to acrylonitrile.  However,




acrylonitrile did induce an increase in sister chromatid exchange (SCE)




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 effect can be detected at




concentrations lower than those needed to detect an increase in chromosome




aberrations.  (These studies strongly reinforce the need to further examine




the ability of acrylonitrile to cause gene mutations in eucaryotic organisms.)




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




covalent binding of radiolabeled acrylonitrile to calf thymus DNA in vitro was




detected in the presence of exogenous rat liver microsomal activation.  A




metabolite of acrylonitrile, 2,3-epoxypropionitrile, also binds calf thymus
                                      13-78

-------
UNA in vitro.  However, additional studies are necessary to determine if MA.




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.  The only study available for review was a mouse dominant lethal




assay.  Although a significant increase in dominant lethals was not detected,




the high background of dominant lethals in this study precludes interpreting




that acrylonitrile is nonmutagenic in this test.




     In conclusion, the weight-of-evidence indicates that acrylonitrile has




the potential to cause genetic damage (as shown'by point mutation studies in




bacteria and SCE studies in cultural mammalian cells and as suggested by




Drosophila studies).  Some available evidence indicates that conversion of




acrylonitrile to a metabolite(s), such as an epoxlde, may result in a




mutagenic form(s).  Thus, acrylonitrile may cause somatic mutations in humans




if its pharmacokinetics in humans result in metabolic products that can




interact with DNA, as is the case in several test systems.   Additional




mutagenicity studies are needed, in eucaryotic organisms,  to confirm that




acrylonitrile is mutagenic.  Additional tests using radiolabeled acrylonitrile




to measure alkylation of DNA in various tissues of whole mammals are also




needed.  An assessment of genetic risk with respect to germ cell mutagenicity




cannot be made yet because of the lack of appropriate data.
                                      13-79

-------































u
M

r- 1

O
s


tH

CO
to

p
M


W
>
JJ CO
o


c-
•H
cO

JJ
CO



CU
jj
CO
CO
JJ
CO
E-t





CU
o
c
CU
Q)
4-t

•H
•g
•H
Ifcj
QJ
>
•H
JJ
CO
00
CU
2
-<






QJ

•H
jj
cO
So
!3






^
t-H
£?
o
M
DO

CO
e

in


1



CO
CU
CO
Q)
rH
CO
E
O
i-t

£
in

in
.—4
H



-O
T-t

rH
o
pa

•
^JJ
CO


CU
("4
g
e
0
o




•



e
•H

M
CO

. 1-1
0) CO
s T!




u
M
C)
CU
CO
3

CO
-t
=1
rt
C
CO
CO 0) MH
3 JJ O
r-l O .
CJ CO T3
e .0 o)
0 w
O Ol CO

0)
£]
a
e e CD a
•H o JJ 2:
CM <
















,_(
3
o
CU 4-> O
ceo
iH -H CNJ '
T-H
ra > e"
' iH
•H >
•0
Q) a) i-l
rH JJ CO
i-l O M
K 0) O
JJ "-> B
*•>
















(U

CO





4_l
O
C JJ
O
CD C
JJ
C i-H
cO (0
4-1 0
14 -rH
0) B
01 J=
(H O
. LM • m
CU CU
U p 4-) >^ •
M CU P -U C
O .0 O -H 0>
o. B o. (H >
01 3 01 3 -H
IJ ES U P-t GO
-* in



















rH T-H
P..C « >,
— . ej CJ C
T3
•rH *r4 CO
oi <-> e a.
u -a 0) e
CO i-H X O
CU
4->

CU



•H 13 0) >M
C -H 14H O









CO
JJ
CO

a)
i-H
3
a
a






01
rH
43
• CO
• • CO i-{
CO t-t rH
•H CO 0>
co r< e
r-l B B
CO O rH
co MH in








a\
r^
a\























CO
co
e
3
iH

I
O-
JJ












< ° °

CU
a
u
3
o
CO
























4J
tJ •« CO
0 6 CU
AJ CO 4J
CO *^H
o e o)
•H CO 4-1
Tj t>0 CO
C ^ rH
1-1 o eu















o
^ .
T^
0> •
t-t
01

CO

3

2
cO
CO
0)
CO

o
a.
*
•^




CU

JJ




CU
>
•H
JJ

CO







>•*
j
rH •
>
H
U
O
CO
J
rH
j;
30
j
i
0)
u

3
*H
H





u

g




CU
o
c
01

CU




1
CO 6-S
o
C"-J
JJ O
e
>r4
•rj

JJ
CO
H





U-l
•a
g
DJ
CU

CU
•H
s
CTi
"
o
CO


H



CO
rH
r-1
QJ
C
O
r?
CO
en






0)
CO
01
a)
S
(U
o
+ 1
JJ
CO


CU
CO
CD O
Q) J-i O >•>
 QJ i-t > JJ JJ
0) O *H «H •
U O CU JJ •« QJ QJ
-JCOCCO «*rH 13>C

HPID-JJJ3CX fOOJ-t
O QJ *H "U QJ •
:OOU-ie3C COQJCU
-1 T3 OH CO O f^
CUOJOOJCUOCU Q) CD

3 H 00 J-J O M ' — 1 1 O 'C
o>OcoJ-t P.CCOJJJJ
Sg.S-ggSSS.SS!
CN CO J>JJ r»

jJCUC^b CCOOi^JJ OOO'U



4-ii4-it>S-iq:jJJJ(DMQJ JD U
O -HnHOCDCO t-HCO •Oj*r-(
rt X •>-* «0 ^ OTdaJ co T-t
CJOCOl 3> Ol O-HXr^
-r-trO 0 QJJJ*i-)X C H COJJ CO
rH QJcO-HJJJJ3CDJJj-t JJ
JO-H 1 X COCDS tf-ClJJ COi-t E



CO
QJ X
13 «H O >t^5 Q) CU

o *H cr\ H
j_r o *-i -I-1 3
X O cvj 3 co O O
CXUH v^ o, co e en





M
c
QJ
J-t
w
CO

U-4
O
CO
QJ
>


ro


H



QJ
CO
rH
ex
jj
CO
CD
JJ


CO

CT»


**
rM
CO

JJ
QJ
| rH i-H
rH CC CO 0)
| O JJ JJ rH
O X *"~f *H QJ CO
rH O X X O S
X I M CO i-l -H
J-i CO CO JJ P CO E T3
tO^OprHOCUO
X C O C rH 3
 t— 1 i— t CO I*-H <-H rH
H H H H H S|H H H


1
3
JJ >-.
O CO
3 CO
rH CO
4H CO
JJ O
CO iH
CU JJ
JJ CO












OJ
00
CO
ex
00
3
•H
3
o
rH
rH
3

CU
-

3
•a
CU
3
2
•H
J
3
CJ






















































CO
e
o
CO
3
I-t
O
e
o

CO
IH
O

JJ
3
*




13-80

-------
















































/•"s

0)

d
•H

d
O
o



.
CO
T-H
t
CO
•-H

i
EH


















^
E
w



EH

W
H
Z
0
t-H
EJ.
EH
O
S



' e



 en

4J CO
O

d
•H


4J
CO

0)
JJ
CO

CO

CO
&





01
0
E
2
CD

0)
pa




•a
0)
H
01
•0
iH
CO

O
CJ

CO
4-1
i-H
3
CO
a)




rH






4J
•H
O

B
a)
60
to
4J
•3



to
a)
3









a)
I-H
•H

4J

B
O
t-H
£*,

O




e
o
p



•H
e


i
en


oo
cr>

H

"tO
rH


E
O
B
CO








fc

a)
•H

to
1 to E
B Is iH in
• o) m co
4-io^-Ncn*** com

o o E I-H co E i-H en J*: I-H -i CO MH OJ


Q) CJ iH 4-1 iH O i— 1 O P 0) 0) 13 P
13 3 4-1 'O iH B B i — 1 p Pi 4-) p 4-1 Q)
P E CO OJ CO CO tO p.--- CO E 01 CO CU 4J

iO <£ 4J 3 Pi CN1 4-)CU3pCJEH4-ltO iH



CM


o

•-H
o

O CO
O 4-1
rH CU
<3 E
H •
r< E
E 0) O

tO UH 4J
LJ co CO
4J >
en >, iH
iH 4-1
B B eJ
•H O to
i-H
B
•C. T3
00 0)
e I-H
rH
O -H
CM rH 4-1
CO to B*S
O iH iH 0\
4J M 13 a\
a)
i-H 4J E /\
E CO iH

bO irj ••

iH 4J 4-1
CM 3 3 iH
O CT rH O P
• iH iH CM 3
O i-H 13 Ed P^


1
 o -a
•H H _E

in
O CO
o in

< <
H H

B
0) O
B T3 iH
O Q) to
eo •• IH 13 E
O t-,UH iH 0)
r* CO iH 3 P-
O eo 13 er en
•H en o IH 3
E to B rH co






o
4-1
E 4-1
tO to P
CM eo o
CO E P.
CT* O Q)
-H S {*

0)
4-1
tO
u

CO
B
O
S
a)
•a

to
4J
CO
13

0)
eo
0)





-H

en
E
iH
to
l-i
4-1
CO

E
iH

O
•H
B
CO
60
to
4-1

S
cu
i-H
•H

4J

E
O
iH
P.
O
IH
P.

^
O
P<
0)
1
en

CM




o
t-l



^H
B


1
CO


oo


H

CO


0)
E
O
B
rH
CO
CO








n

a

Sxt


0)
•a
UH iH
O X
0
r>*> Pi
4-1 a)
O 0)
iH i-H
B iH
01 P
00 4J
tO IH
4-1 E
3 O
B rH

OJ VJ
4= CJ
4J CO



U X
X iH
00 E B
4J eU iH iH
eo o I-H a^
O >-. B CO T3 1
•* e • 4J o) B co
in E -H eo to 3
en a) IH > .c o UH

rH 4-1 P 4J B
<; 4-1 o B to to o)
H 00 CO CO iH CO O
E X S B
T) iH 0) O 13 iH 0)
Bol>iHOEQ)eo
co ,a -ri E en en o)
4J o) eo c^ co i-i
o o iH 60 oi I eu o.
o o to to P en v*
rH rH E 4J P. O QJ

HH cos; o)oiH4j
0)
13 rH
•H B 13

O 00 rH
P. B i-H
eu IH
O 4-1
OJ CM CO S*S
rH in -a-
•H O 13 •

4J E a*
iH rH iH
B S
O ^ 13
i-H 60 Cl) X
!*, E 4-1 4J
H 3 iH
CJ CM rH O rJ
CO • iH CM 3
*— ' o 13 m P-*


i
 0 T)
iH l-i B
rH •> UH -H cu
P co iH 3 p. 4J
cJ en 13 cr en en
•H CO O iH 3 OJ
S tO B rH tO 4-1






o
4-1
E 4J
f> CO H
CM eo o
CO B P.
ON O 0)
rH g OS
^
0)
00
CO
P.

.S

Q!
r-J
T-H
O
M-J

a)

4_J

d
o

13
0)

d
d
o
o






























































CO
d
o

to

rH
O
d
o
o
w
}J
o
rd
JJ
3

*




13-81

-------



































•o
CJ
3
B
•^
4J
B
0
U



CO
«— 1



to
"3
^
E~*
















<
S
td
g
«S
ea



CO
H
CO
Cd

55
O
E-4
£5
2
•^

j2

O



**










CO
4J
B
01

o
CJ







•u
01 *
4J 4J

O 3
CX CO
Ol 0)
Cd Cd











B
O
•H
id
p

B
CU
u
B
o
0








B
o
M e
4-1 Ol
Cd 4J
> CO
~r> >,
4J CO
4J
B
•H
cd
p
CO




Q
CU

CO
CO


CO
s




CU
o
B
0)
p
CU
M-l
01
03






cr






i






d



f
4-
«T-




s
J
[








J 4-1 O i-l
0) Id 4J 4J 0) •
4J > 13 0 > -0
p -a cd i-
» 3  o
4-

CO CU -O 6
O P CU iH
•a

I e
CU T
' rH
4. 1-1
a) co B 4-1
f-f O P CO
CO T3 CU
B 4-1 CU
•H
CO

i— 1






01

•H
4J
•H
CO
O
CU


/N

^

•H

3
O CU P
B "0 id


















C
T-

to •
Ol ^-
*H t;
4-1 OJ
a. -H ^

cm

-^
CO
a

o
fl

01
'm
S

4J

3
•0
Id

0
3


5
o
en
m
—i
E"






•H
3
t-l



O
M




1
0>
4J
4J
0
,O
S
cd

3 P
CX CJ
e co
•H Ol
•a

5-S 4J
CT\ O
CTv B



•a i-

cd X

CO
cd i-
p It

id 4J
4J 1-1

i-l -O
3 id






CO
id
p



CO P
•H IH
CO



g ij


o
CO
p 2
CU
cd -
o. •
01 iH
•a id
B
id 4-1
> a)





















^ 13
• 01

O
_cu
B
i-t

CU
B
1-4
r-1
id
CO



„
O
p

•H

B

• A
CU
O













01
o
•H
e

•a
B
cd













0
4-
E
CU
id

p t3
3
CJ

i-j
SB 0
0 P
•H CJ 4J
T3 B
•5 ° °

























01
B
t-

-

• «

CX

•H
a]

OJ
•H
i-l

e
o
p

X
•g


CO










id
iH
i-H
01
B
O


cd
CO














CS
















CO

3
O


^.
C-4

O>
4J
O
CU
rH
r-4
O
O



-=f
in
CN


P
o
rH
O
O
p


M-l
O









e
3
•H
p

•rt
4=
CX
4&
































*

O Ol
i-l CX
P O
•a p
 4-


•H Id
4J 3 .
•H i-l CO
CO cd 4J
O > O
CU 01 t

— '



i-
11
O
•H
E
0
C.
C
01
u
p
g
CO

T3
C
CtJ

>•
J-)
•H
M
3
P-i

CN













C
a,
p.


O
>4-I p 4J
•H 0) O

Q) tf
-p .a «
Cfl CO CO O > 4J
U CO 3 • -H -H »H
•H SO C ,d 60 O
^ COS .H
C >^ (D *H 4-> ><
•H 4J 00 -M *> O O
•HO CO CO d 4J
O -H CD *H -H (DC
o c o P a)
T3 CO O CtJ 4J S C
•H CO , O
H3 O *J CJ *vH


•H M QJT3 OUJJSOJ
60 O bOc^rOO Mt-i M
C
C


5 1*3 iffflMM J

co 
4-1 iH
P 4J
O -H
CX CO
0) O
13
B
CC



iH
01
B
O
g

id
e£ cxco









































B

o
B t-l
0 r-4.
1-4 0)
CO B
P 0
01 e
> rH
cu id
& CO


cr*
r^
2

B «
o •
S -H
co id


H cu










































£3
3
•H
p
§
i-f

O.
4J
















CO
•H

O
•H
I-l
CU

CJ
w
w







































C8
•H
O
•H
t-l
(D

CJ
CO


*a
C
cd



































































c
OJ
•H

ct)
)-i
00
* Vl
•H

"o
CJ






































































4-)
CO
a)

a>

rt
i— i
cu













^
W)

CU

(jQ
C
•H
S
o
rH
o

(U
4_J

c
o


-------







































•o
01
c
j_j

0
o


00
«— i
I
en
Ed
•J
PQ
H
















„.
3
la
B
ca


en
•X
CO
w
H

o
H
^
g3


w
w
o


 to
t"* t*i
J-J CO


c
•H
2
4-J
CO



S
01

CO
en
4J
CO
01
E-i




01
o
c
01
01
(4-1
S,
•0 \^~.
 e
P O
01 CO tH
• to 4J 4J

0) 4J O 3 (3 -H
*-l C CO O Ol M
M4JOO-OOI O>ffl
01 t-l M t-l >
tOMtOM-tC »4JoO
OMT3 CX1-IC04JO
i-l ^O 0)COMC!4J
3 B t3 CO ^J O 01 C

O 4J 01M01H4J.

°4JiHe°'1> Sois
4-J O O CO 0) 01 C
>301HtOto'Sto£'i!
•HOCOOtH 4J
4-J t-l 3 iH t— ( COCL»«O

tog-ioo 4J01U3
O *H 01 tJ O t-l _£ 01 JJ to


• • • 00
•— ' CN ro C W) •
I t-i e c
CU 0) CO t-l O
C-H t-14-JcOCcQ COM-I4J
O4J|t-IO^3OCtO(UOcO
COCOI-IJ2 t34JcOrH olt-l
oicx •U'HcflSijco'ed^j
a. ni a -tHtooitooico
CO.. J3SC4-J4J i-l CO
3M4J OUHCOQ] 4J^JO
COO Lnt-lt-l^3>o)tOtOt-l
CXC04J.C 3-H> SS
T3ctJt-l tOCOCOOJ-HA-IGJO
3 >^CJt-lgtJtoc04J4JcO
111 I a£1.g f !.g §
fl
•» E
O rH *t3
C n o o C 4J +
t*J " *H ^-( CO t-l &\

m ^ o 13 at E a.
3 •* M-I tH.noa)to
••crc m H 3 3
QJ CO O .-
4J -* C O X (1) X
rH  o *o -H o
1 T-l f-I C S *H
U3 rH < *H CO E
m co co

m m a\
<; -, a) a) t-i
S 4J nj o- 4J o
tH CO O) CO C1J CX
03 a) co 3 rH nj
en 4-J nJ co ex. >


r-^.


e .-H
« <
* •*
P tM
rH rH
•H O
00

ex

60

•H
O
rH
o


0)
X
4-J

£
o
Q)

rj

4J
C
O
CJ
V"'















































^
CO

5
^
3
H

3
o

0
D
S

IT
S



13-83

-------





































•o
CD
C
JJ
d
0
u



CO
"•*
en
•— '
a

CQ
• Ed
H
O
CQ



CO
H
H

2
O
M
1
[d
2
g



 CO
•H >,
JJ 05
U
"*




d
•H
CO
M
jj
CO




E
a)
jj
CO

CO
CO
cu
E-I

0)
u
d
cu
CU
MH
CU
«
t-H
o
• M
0) JJ
co d •

o o jj d cu

CO d) O C > P
0) > CU O
p TH cu d P a.
JJ O O O. CU
*O -H d *H P
CU CO CU JJ O
JJ O CO CO JJ CU

rH p nH CO 0)
CU JJ CX JJ d &
p d cjo
1 CU CU CO iH d
Q) P .r" JJ Q
d JJ >H O JJ CU P
POP J=> p p O
cu cx jj jj cx co
rHOCUCOCUCUO>
»— i OJ co 

cu o jj a.
> O rH 0)
•rt CU CO rH O
JJ O CO X
T-I d o o
CO CU -H P -£
O CO rH O CO >?

J^ co d CO
• « ^N JJ
*^- >-. O U X
CU O CO CO CO ,J 60
O JJ 03 Q rH
B CO C E CO ^S >^J=
3 rH O>iiH..
•< D, CO 3 rH JJ E CU
in cu p >\\ jj o -H p o


•* rH d CO K . 1 -, 1 Q, nj o
O EX^H COCN)UHP^-^T3CO

CO
 1 cu E

rH in o m
CN O •
8 rH JT O
o
P P (O -
4-t O O CO
X CJ -O CO
•rt O CU P
E P O CU
 O
j: 4-1 a) *H
U CO QJ J-l
•H Q) OJ «
^ 4J D
O i-4 10 3 CO
CO O t-i i-( CO
W O CX IW 03

B

M
•U •
•U M
•H rt
C
 a)
                       o     d  o
                          cu  co  co
                       d  co     cu
                       O  3 T3 T3
                      «H  CO  CU
                       CO  O JJ JJ

                      rH JO  CU  d
                       CJ     0)
                       d nj  CU  co
                       > rH  O  O
                      •H  CO  d  O
                      JJ  >     O
                       co  a)  cu jj
                                            cu
                                            00
                                            CO
•a
 cu
 3
 d
iH
JJ
 d
 o
 o
                      0)0)
                      pdC
13-84

-------


















TJ
CD
3
d

4-1
O
O


.
00
I
ro


0)
rH
-Q
rt
tH






















<1
M
33
P-I
0
en
o
: -H «H CO d CO
•H 4-1 CO «H CJ
4J CO CO rH -H
M 4J 01 CD .0 E
CD 3 iH B 3 01
60 B CX O O JZ
60 B CO -O O
3d to O
CD i-l CO S CO • "4-1
. O ^ O
•a 01 e CD >-< 01 ^
01CO 0)60X1300)
UcO >UO3CCJ
01 CD -H CO r-t 01 t-t
T3P 60rHOCJ33
•H cJ o o) cr o
eo c! 4-i >,0 U 01 co
C 1-1 o t-l r- o. ij
0 . C 4J v-' <4-l t3
o -0 >-, C o C
iHOXOlTJJ-lC CO
CO O C! 4-1 -H CD O
WU-lOl-HCJNTJ-rf:^
iH013Oi-l>>014J4-l
3 0) CTi-IU-liHT3 CO*H
COUCDXt-l C0014-1IJ
01XVJO3CCD33
pi4-i>HE-icn co C BCM
• • • B
r-i o-j c^ -w o e
O i-l 01 to

C 3 O O >,
O rH Tj 01 O
•H O J3 C C
4J CO CO tO 01
O 4-13
01 fr^ o c! cr
•n _i 4J O CD
C • d ex n
H o -H co in





co
iH U
4J i-l 01
Tj CD ,£ 4J
CD CD 0) O- CO
J 4J O CO
C i-l CO 60
i-l CO rH O O
r-l tO CO ^4 CJ
I 01 x o to
*: o 4J ,H
CD 4) 01 CJ 01
en M iH i-l B





C cr,
CO .-I
: tO •>
CD S
C CO
CD U
m en

«
CO
iH
o
1-1 ...
4-1 C C
d 60 01 0)
o e > • >
O i-l iH iH i-l
C 60 10 60
0) rJ O
> 01 4J 1-1 4-1
i-l CJ O P O
4-1 d c: o c
•HO 4-1
CO O CO CO CO
O iH i-l 01
ex e o j= B
OH 1-1
d 4-1 1-1 4-1 01 4J
01 d 4J C J3
> 0) CO O 60
•H u S cJ o d
60 M p 4J i-l
3 O 0) 4-1
4J O «-t > CO CO
o c d 1-1 f s
d O -H 4J CO 1
O CO CD 4-1
CO T3 60 CX CD
co o d 01 ex o
S Z CO C co cu
in ^o














































































13-85

-------














•o
TABLE 13-18. (continue









en
CJ
H
VIVO CYTOGENt
z
1-1
1
;S

s
s



CJ

Comments

•0
CU *
p i-H
0 3
D. CO
at a>
E; a:






Concentration






e
•i-l
n

4J
en


QJ
CO
en
j-i
CO


QJ
u
c
£

Q)
IU
u
t*
QJ
Aberrations scored were gaps,
breaks, and rearrangements.
. Results considered inconclusj
because background control
frequency was very high (33.6
postimplantation loss).

CU CU
t* >
U 4J
CO cO
60 60
QJ QJ
Z Z
10
CO O
C 5-.U-!
0 i-l
•H CN CO ^1
4J P^ C Q)
Mice given single i.p. injec
of 20 or 30 mg/kg and cells
examined 6, 18, 24, 48, and
hours later for chromosome a
and 24, 30, and 48 hours lat
micronucleus analysis.
Mice given i.p. treatment of
30 mg/kg followed by weekly
matings (1 to 5 weeks).
Source: Aldrich Europe
(purity not given)
CU CU
0 0
e -g

QJ QJ
iH rH
§ B
H M
0? g
I I

.-< r-l
cd co
CO B f.
3O 4J
3 CU CO (U
o r-( o c: r-i
u a e Os~.
K 3 O -H CO 4-1
CO C M JJ ^ C
S O .fi CO CO CO
X Ij O U >-, C >,
QJ CO O I-l i-< ^ CO
oeo'ec^cs o™
00



•o -

Srj
CO -H
O CO
CU -Ulcrs
i_] CU -M
•o
• Aberrations scored were gaps,
isogaps, breaks, fragments, an
translocations.
No concurrent or historical
positive controls reported.
-H IN
QJ

T-t
JJ
CO
00
CU





« T1
Mice treated at 7, 14, and
20 mg/kg/day by gavage for 4
15, and 30 days; rats exposei
orally to 16 daily doses of
40 mg/kg/day (purity 99.5%).
Source: Aldrich Chemical
Company
I
QJ
CO 3
CO 00 CO
i-< CO -U
5 >j co
en p. M
o >>
e « QJ
•H OJ r-H
J= 0 S
i-l -H CO
< B O


U
QJ co
C CO
O CO
pa co
o
CO

O 13
i-H C -
i-( CO -O
Q)  S*, X i-l
CO O CO QJ >,
e o h f
•H X Q) C 0
B CLJ^ S co
CO S H 0
x >, o c o


•a r--

CO •— <
CO *•
CO 60
QJ *H

E- E!
aj
(continued on the following pagi


























0
0










13-86

-------






















































TJ

3
C
•tH
4J
C
o
o
~
CO
rH
1

rH

w
m
EH















Id




a
0
rH
H
Id
O


O


M
EH


M
Q
Z
HH

CO
EH
CO
EH
at


H
O


O












to
4-1
e

g

O
0
























13

JJ LJ
}~t rH
0 3
Cu to
tu co
cJ pi












c
0
T-l
4-1
to
l-l
4-1
C
to
o
1



c
o
•H S
4-1 0)
10 4-1
> to

4-1 CO
<








£3
•H
to
U
4-1
CO





e
CO
4-1
to

CO

4J
CO
9)
EH





0
O
C
CO
CO

CO
p£




























00
C P
•H CO
•a >


jO
4J
 a
CO 4-1
rH O
CO
E5 4-
O CO
iJ •a




>•
4-1
•H
U
3
a

cu
rH CO
CU rH
.0 T-l
tO M
rH 4-1
1 T-l
o c
•* o
i—4 rH
>-
1= O


P
to

•H
•0 rH
a
to e

4-1 S
S.S




















0 CJ
3 O
*O *H
T3 4-)
cd oj
<; M

§ 'S




CO
CJ «— (
•r-t
t-l •>
Q) »
CO r~)
C 
(_j
3
O
X
r— 1
O



&*?

ON



to
tu
B
o
to
0

o
e




















i
o

JJ
•H

C
•H













































O
CO e
rH O
tO l-l
4-J
CO -H
C >
CO
rH C

•C«
4-1 <
Q) Z
0 0
C
CO CO
>,T3
c; c
1 -H




^
CO
rQ 4J
•H C to
X CO SJ
0^3
tu o
tu *w ^
C >w
tU -H -d-
rH TJ
•0 >N 0
CO £ 4J 4J
rH JJ tO
CO CO O.
£* O SI 3
to c B
rH tO CO
! j>^ m cu
o o o-i B
— 1 c-3 O 4-1

•a
tu
T-l

iH tO
rJ >
3 T-l
CXrH O
in
•a 4-1  T-l
*H 14-1
00 l-l to T-I
CO 4J O
4-1 CD rH CO
0 rH 3 0-
C! O CO CO
C! CU C
CO 3 rJ O
•U C
tO tO Q>
t3 -H X O
4-1 4J •

4J T-l V4 tu 4-1
•H CO 3 -H
o «j= -a o
•H CO -U T-l
X 3 CO tU X
0 JC X r-l 0
EH 4J S to J-l


v— 4 '






rH
e ti
4J -^. O
to bo
3 O
to in
ooo
CO O 4J .
X. •& co i—t
CO S
CU TJ 4J ^
r-l C O bO
& CO C 3 •
tu
, tU
0 0 UJ B X B-S X
u-irHO *oc.omo
I-H e a. ,c •
•O "bOgto 4-lOX>>
tuos — . into oc
4-j m tu --^ s v T-I to
CO O rH x~- 1 1 l-l O.
j34JOfi CO < U
C COT3T-4 CXtO K4J
•HUCC COtO-
3tOOS^T-lr^5CU
CO O rH ON tO 13 O
rH J2 « >-,O> 4J >^S>S rJ
rH 0 U P J3 CO 3
to co o o /M o 1 • o
O — 10^ CO ^^ O — * O tri







CO
c
o
2:

C n
co cu
T-l 4J tO
rJ tO rH
>% 0 rH
CO W CO
J2 O O
>s M
P C O 4J
tO tl) >-. T-l
E -a iJ p-
•H rH rd
U 0 S C
PU O CO -H
4J
C
tu
•a >, -H
«-l (3 ,0 TJ
o to to
l-l tO t4
c: 4J ^ oo
o co to co
•H tU C CO
4J to V -H tO
O rH .a rH O
CO 60 CO l-l
4-1 q; <; O o
CO T-l Z rH 3
Q to P CO to




a.
c S\
tO -H

4-1 «•
c o
CO 4-1
S-i to
tO to

tu
oo
to
a

00
c
•H


rH
rH
o


CO

4-1

c
o



3

4-1
c
o
o







































































•
to
g
•H
CO

r-H
o

o
o

^.
W
o
5


c




13-87

-------









































^^
•o
0)
3
e
•H
4J
c
o
o



00
•— i
.— t
w
J
cc

H









u


Q
CJ
I-H
H
U
s:
td
U

fc,
O
td

I— I

^

M
O
z

H
cn
Ed
H
M
Cd


O



a











CO
4J
C
cu

g
o









•a
cu *

P rH
O 3
O. CO
CU 0)
a: c£








e
0

jj
CO
p
JJ
e
0)
o
c
S




a
o
•H S
JJ CU
CO JJ
> CO

jj cn
u
c
CO
M
CO











S
4J
CO

CO

JJ
CO
cu





CU
o
c
01
P
CU

CU


cu
CO
o
•a o
S
CO JJ •
CO CO
3 e JJ
•H C
cu a) cu
co -a jj 6
B cu cd -H
O JJ P P
a. cd cd cu
CD I-H o. a
CU 0) 01 X
p£ P CO 01
rH
t
CO
CU
cn >>
JJ O
rH O
3 CO JJ
CO rH CO
CU rH O.
P 01 0)
0 X
CU
> O JJ
•H X 3
JJ O O
cd X
00 P JJ
CU O iH
£5 '1 1 S
•a
cu o
cn co
0 S

X cu
CU rH C

cu o p
P JJ JJ TJ
01 iH 0)
3 cn c >
P 0 rH
CO 3 rH O
rH O >, CO
rH X P CD
CU O iH
o c*°i to ™



CO
JJ CU
cd JJ
p >*
o
JJ O
tl 4J
CO CO
JJ B.
B CU
rH X








O
P
73 JJ
•H iH
JJ ^ >
CO td
S CJ CO C
O CO i-H iH
p ^ rH
J= 01 1)
0 CU 0 P
00 3
P C 0 JJ
0) CO -C rH
JJ _r* O 3
cn U o
•H X C 0
CO CU iH CJ


•a
e -a
to cu
4J CO CO
CO B iH
p CO rH

i-l c>T 3
•O rH CO O.
0) iH a* C!
> S rH »
U-J
o •

rH 00
ttl iH
> J2
01
rH ^
p
•0 01
C >
3
O CO
P cd
oo 3

O Cd
CO O
CO CO
CN
p CO
o cu
4H JJ
CO O
JJ O
rH JJ
3 cd
cn a.
cu cu


CU J=

iH iH
JJ 3
iH
CO O
O X
a. o
C!
CU

B S iH
Of! 00
p 1
CH o JJ
—> o
cu z:
00 O
e JJ
cd
p S ><
jj
cut-- -H
CO 1 p
003
O rH CU



















CO
01
JJ
>,
0
o
JJ
cd
0.
JS

4_t
tO
p


JJ
•H
&













     o
    CJ

    w
    CJ
    CO
c

!
p
cu
p.
K
u
i
iH
p
CU
&
w
              cs

              +1
              10
              +1   +1   +1
          + 1
                                          + 1
B
cu
o
B
o
CJ
              o
              rH

              X
              ca
              S
              a
              g
              o
             ft

              'o
vD

 'o
                                 o
                                 CO
                                 g
13-88

-------
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, other factors, including mutagenicity, metabolism




(particularly in relation to interaction with DNA), and 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.









13.5.1.1  Drinking Water  Studies




 13.5.1.1.1  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
                                      13-89

-------
 Company, Inc.  Its purity was greater than 99%.  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




 j.00 and 300 ppm.   After 9 months of treatment, the animals at the two higher




 uoses 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




 •is  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
                                     13-90

-------
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(100%)
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%)
    *Significantly different from controls by Fisher's Exact Probability Test,
P < 0.05.
                                    13-91

-------
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%)
    *Slgnificantly different from controls by Fisher's Exact Probability Test,
P < 0.05.
                                     13-92

-------













CO

53 .
•H £_x
o
0 0
M ex
CX

o m
3 "





Q
^
6
ex
ex
o












gnoses
w
•H
P
•4-
tO
,«*-x |
CO O
• 1—1 *i —
i— t o*s
co X —i
<~s O

oo m o o
^— CN ~-. ||
in m
-H 11 P-i
PJ





^T
1
o
, — i
CO X

xD O
^^ i — I
•
OO V CO
-d- ON
"" — *^> II
CO  — 'X  	 s x-x ^^s \ 	 ^ s 	 ' cfl
Q> l/*i «^J- ^Q , 	 | k^
r^ in i"— ON co vo oo t—* co CM cu
"5-. i-H -21 f*~) -^, CNJ ~? -rf -i. ^H rn
~d" ON CO CO O 4-J
co II co II II CM || co u cO
P-I P^ pL( p t p [
u-<
0
OJ
&
£
•)- 4- ^
O CM -1— O CU
1 — ' ' — ' ON. i — i ,0

r^- o ON o /-^ vD o oo o
• f-H • i — 1 CO • T — i • , — ! Vl
•— ' r~~ • ON m cu
<(• X 
N-^ **-S N^^ CO N^ ^_ S Q
O --,
CM II CM 11 II -H || CM || 4-J
p., p_j p | Q_^ n, LJ
o
CJ
|

1
•)— m 0 .
CO | -H 4-J

s~~* ^*\ ^^ o o in t— i 4_> m
co m m »^-i »vtH
. .-i— ^i— t^, m X cu
*<1" vD r**- OO 1 — • »-H X CM O 4-J
o o m CM oo co
^J. J- J—) ^_». f~) ,-L. ^ 	 . _.. -^j J~i 'Jj
->^_ ^. 	 ^^^ ^^_^ .^^ .^ ^
CM co II -a- u co u CM u ct) vi
—t CU O)
CO •
CD -H CO
4-* pr, , 	 1
cO cfl
,-*. vi -rt e
O O O CO COlH^-10
O O O -H .— i 0 cO
_ Q) 00 1 i-l
O O O O O^QViCUcO
CO OO OO OO 000O0
O O O »-H >— I 0 Vl O
cfl CU Vi
CO i— 1 ,0 >4-l
CO 3 4-1
CU 00
O O -r-1 cU 4-> 0 ^i
•H *H cfl 4-) CD Vl -H CO
4-1 rH 4-1 @ iHCD ^^CDCOCD4-)
Vi i— 1 Vi O cOCU 4JCUCX3
OCUO0 -HOC 0S-I4-)
CXO eX'iHTlJi — 100 TJ T3 cfl CX4->TOO
VlCD VlCOVl O CD O O OOCU^P
cfl3cflVicflOiH C_) U-H 4-)cOCU
1-HOrHOO C00 1-HCUrHVl
3B 3E i-lOO iH ,-H m OS-J3CU
'SS "S3^ COO-H cflVi cOViB 0OO&
03 04->O 04-14-) 0O 0O CU>4-(rH
cfl cr1 cO ~-^ -H-HcOVi -r-)6 -HBVi T3 cflco
i-ICO iHrHTS CX4-JVIO CX3^^ CX3O -H>NOCO CO4-J~-~~ O>H rH
Ocfl OO VI 3 -iH 4-1 Vl rH O VliH0 -HCflCUE
i3E 53 CO O l-l OrH4J Or- 1 Cfl O3CO
O COB -^VlO!^ ~-~.0)>N --~CU VI -H rH CO
li-l I3O TSOVirH T3OO T3O0 OCXcO
i"H O ! — 1 0 CX Vi 0 O 0 OO E O f> i — 1
,0'H ,0g,-H cfl,-H cfl cfltHVi cfliH-H 3O|rH
OCX Ocflv-l cflrHcU cfl4-i c00 HCDP-iQ^-s co OON_X o
4J 4-1 VI Vl Vl -H
co co M po pa E
13-93

-------




X"N
5*5
^
s
Cu
Cu

O
o
CO
CO
H
2
t*3 »J
hj M -JC
*5 ?5 ^
^i £j r""^
pK4 J2J ^
O Q)
*5 >J t-3
«* ^J
W CJ 10 B-S
o pa '"^
is o s
U !S CO PL,
OH 3 Cu
W 2 0
O M iH O
2 
O O x-> 4J
«£• ca ca
y—i DcJ O
H cd co to
H C\ QJ
Q •< r-t O
•5 S- P*
*C •* Q) x*^y
O • 13 B-5
CO JS rH iH «-^
W M R) CJ
co ^ pi s
O 2 4J M Cu
25 W Q) CU
0 OS M
•< d AJ o in
M 0) g CO
R 5= ta s
O 3 H
o o-
*rj? jj-j
o 3
o ><
S
^J ^J ^"N
Si o£ $5
o o ^
£"* f^4 0
CO CX
MQ o-
53 ^D
•Tj
r>j iS
Ixt S-
l^ ^
rv\ .*»*
^^ ^S
*™* 5S

j
^3
^s
6™*




CO
tu
CO
o
c
60
ca
•rl
O



^"^
^-1
•
CNJ
in
"*~S
OO
•>^
"*"X^
m
CN









/""\
CO
1 — 1
oo
oo
CT\
CO





^-^
^1*
•
inn—
CO P^
N_X CO
o
oo •
si" CO
r-l II

PM




O
•
O
r-»

0
co
VD
m
















ammary Gland
f ibroadenoma/
adenofibroma
S



x— \
CTl
•
CN
*~s
OO
si-
••v^
m
CO









S***t:
in
OO vD
v-x CN
0
oo •
sf 0
CN II
*sJ"
PM




,— ^
m
•
00 VD
v-/ CN
O
CO •!
st- O
CN II

PM




CO
•
!-M
!•-•

O
CO
in








4J
d
ca
d co
60 te
•H O
M E
cd 3
K 4-1

SM 13
o d
d 13 i-H
cd d 60
i-( ca
CJ >-,
>-. 60 CO
K -H 6
ca d S
e cu ca
ca
S
••— -t— H— -i— in cu
vo oo in vo * — i 60
It 1 II ca
x™ ^ C3 x* v G> x— \ O x~v CO x— N O "tj Cu
^"^ CNi — 1 ini — I CNi — 1 | — .1 — 1 st* i— I CU
CO • • « »«d60
M- ONXr-~X CNX VDXOX-H d
OOOOCN CO CM CN vD B-H
v-'r-i V-XO ^XCM V-XCO V-XO ^^OO Cd B
oor^ i— i 00X0
co«co»oo« m» m • co • oiiH
sfOsfcOs|-CN sfcN sf CM si- vO i— I
^"^ *^^> -^^. *^^— ^"*». ^*^- cfl o
Si" || St- || OO II O II CN || ON II 4J l(H
•-1 >-l i-l — 1 CN Cfl
PMPMPM PM PMPMS-ICU
o
M O
CU
XI T3
E cu -
d s
d d
H— ^— -1— -H
co sf in cu 4-)
I i i xi d
X-N X-\ O O X-N O 4-1 O
^"> m p~ i— i /— \ ^M o i— i o
CO • • x"^. I*^. • $M \^s
•4— CM-)— vox o «XmXcu
VD i — Ii — If — li — I • ^D CM t>
^^ m ^— ' i— 1 N-X | — O VD vD -^ CN O
o o VD v-' s_x CM r-.
oo«co«oo» «oo«co
Sf O Sf O Sf i— 1 CO COCTi-,
i™~l 4— )
P-t (^ PM PM PL,
Q
S
_s
4-J

^**N* t-j *
/""N , ^^ ->.t" x™\ «H 4J
c^ co • i — i £.3 cfl
• • O-t— • 4J cD
tN CO f-< CO CN] M H
^wx ^— / s-x CM s^x  o)4-3dri<;
g d d3 Cd COM'Hcd
cd cd cd4J E cocdt04j
r- It— 1 t— 1 O i-MCUCuS
6060604-) d I-IJM4J
d -H 11 Cu 4-> 13 O
I-H i— ( iHcd o cJXcdcud
cd CO cfld JM JM ni rf 4-J
-S -2 -2°° <" "i"1 4Jcacu
E E E-H CJ t-1;3cUrH(-i
l>i >. >-,r-l SOOJ^dcU
NI M tSJnj 5-j i^EdocJIs
v-' v-x ^-s G O dCdCUM-JrH
~-^ cd S 13 cd Cfl
•U 'O '2'-1 "-< rOt-Hr-lcr.H>-,Ocu
P- d dO i-H diH60COaiHi-l
CO cd cd^~. (U cdcud PliHCDCu
i-H tH i-li3 -O CJO«-rlcdcuE
O cjcood co cd K cd a^cd
_.._,§ Cd ECO ECO 0 SM.rHrHcO
d2 ^2 d_ 23 °=) l° ocuto
c8E cdd cdd P!O iHO iH 6 O > rH.
PjO d-fj d60 -r^S -HE XI rH SCJIrH
cOd cdO cd-H cuocd cu-Hcd a-H HCDPM<;
OCU USM Ud 3SMS dCu3 cdCu-K 0-K^ cS
13 cd (UMcOcfoOcdcrEcO !M
i-jcO >JO SMXI dOCO dCuco OCu O
co cd cd o o 4-i «rH
W W w H H CO e '
13-94

-------





































/— s
cu
3
d
•H

d
o
0



»
CM
CM
|
CO
i — 1

w
1
H














































/— X
B-S


s
P,


O
o
CO





•K
CO
' rH
CU

CU

B-S
CU ^-^
CO
ai
&
CO
3 0
O O
•H rH

cfl
4-J
cfl

CO
V
0
d
cu /-^.

•iH ^
o

M P.
P.
SH
o m
S co
rj
H



/— x
5-S


&
P.
P.

O
























CO
cu
CO
o
a
cd
•H
Q
_,_
xO
1
/-°v o
O r-l

in >i
CM
V-^ xxf
O*\
OO •
-xf CM

CM ||
i — l
P-l










/^-s
o

o

OO
•
Is cd cd
E B
1 O O
T*"1 £3
,0 rH T-l
O -H O
Cfl CX S-l
B cfl cd
O P- O
4-J
CO




/ — *
CO
•H—
00 00
**^ 1 — 1
o
oo •
•
1
o
r-H

X

CO
c^J CM
i-H •
T— J 1— t
•*- 	
<3" ||

P-I













c^a
p^
»^_
i — t





^^
0

0


0
00
«• —
o




CD
3
O
d
•H
O
3

*,
cd
g
o
d
•rH
CU O
C V4
•iH CO
4-J O
CO O
^
I-- II

P-l










s~\
m
•
00

00
*^"








/~*\

•iH
4J
rH CO
cd cu
•H 60
t3 rH 60
SH 60 3
O CO
0 rH
cd d
rH O O
cd O -H
d 4H 4J
•H -iH cd M
P. 4-J SJ O
CO rH CU B
3 4H 3
S-l B -H 4-J
O rH

13 O S-l rH
d ft !-i
Cd r-l Cd
Cd rH CU
•H O 0) 
C-3 rH t>%
cfl O
rH 0 0
cd o s-i
d HH 4->
•H -iH CO
p. 4-> cd
CO rH v-^
3
S-i B J-i
0 0

T3 O 3

Cd rH
Cd rH
d 0 rH
•rH O CU
Cd 4H 0
•M
P3
•f—
vD
1 — 1
1
^~* o
VO r-l
•
.
3 CO Tj
s-i S P d
o o cd

• T3 0 3 d
d -U 60
Cd rH -H
cd <-i d
d CJ rH 0)
•H O CU ,£>
cd HH O -^
SH
PQ


T3
CU
d
•H
6
cd
X
cu

CO
4-J
cd
SH

<4H
O
SH
CU
^O
ti
3
d

CU
,d
4_)

SH
0)
O

CO
CU
P,

4-J
O
B
3
4J

d
•rl 4-J
cd CD
4J CU
!H H
CU
0 4-J
O
60 cfl
d X
•rH Cd
SH
cd rH
CU CU
CO •
CO T-l CO
cd cd
CU T-l
<4H i-H d
O • T-l CO
d cd
SH cd 4-J rH
CU 60 1 rH
f> SH Q)- cd
S o d
3 06
d SH 0
CO CU SH
CO rH ,r! MH
Cfl 3 4->
o d
13 -H 60 CU
0) 4-> d ^
CO SH .H Cfl
CO Cfl CD 4J
CU P. 3
t j ] 1
C"s _t 4J ""^ O
X cd cu d
CU r!3 4-)
4J Cfl CU
CU rH SH
O SH 3 CU
d 0 0 S
CU MH rH
TJ Cfl CO
•H >~, O CU
O rH r-l
d rH CO P.
•rH Cfl Q) S
0 3 cfl
SH T-I i-l CO
0 P, cfl
B 0 > rH
3 0 1 rH
H CD PH <;
4! O-r- c«o
SH
0
•H
E
13-95

-------
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.  In some animals, they caused displacement of the lower jaw,




which 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.








13.5.1.1.2  Bio/Dynamics 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




Bio/Dynamics 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
                                       13-96

-------
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 than18%.  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 or




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 later in this section.
                                    13-97

-------
           TABLE  13-23.  TUMOR INCIDENCES IN SPRAGUE-DAWLEY RATS FED
                        ACRYLONITRILE IN DRINKING WATER*
                            (Bio/Dynamics 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
Papilloma/
Carcinoma
3/98 (3%)
1/100(1%)
3/98 (3%)
4/99 (4%)
12/97(12%)t
7/99(7%)t
    *Aniraals 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.
                                    13-98

-------
 13.5.1.1.3  Bio/Dynamics Inc. Study in Fischer 344 Rats (1980b)




     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




 Bio/Dynamics 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 in 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 hemoglobin,  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
                                      13-99

-------
             TABLE  13-24.  TUMOR  INCIDENCE  IN FISCHER 344 RATS FED ACRYLONITRILE
                                     IN DRINKING WATER*
                                   (Bio/Dynamics Inc., 1980b)
                                                      Ear Canal
                                                      (Zymbal Gland)
                                                 S tomach—Squamous
        Dose Level
Groups     (ppm)    Sex
   Brain      Spinal Cord
Astrocytomat  Astrocytomat
Papilloma/Adenoma/  Cell Papilloma/
Carcinoma           Carcinoma of
(Squamous Cell)t    Forestomacht
IA & IB Control M
F
II
III
IV
V
VI
1 M
F
3 M
F
10 M
F
30 M
F
100 M
F
2/200 (1
1/199 (0
2/100 (2
1/100 (1
1/100 (1
2/101 (2
2/100 (2
4/95 (4
10/99(10
6/100 (6
21/99(21
23/98(23
.0)
.5)
.0)
.0)
.0)
.0)
.0)
.2)§
.!)§
.0)§
.2)§
.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.
(0.
(0.
(0.
(or.
(0.
(0.
(1.
(0.
(0.
(4.
(1.
5)
5)
0)
0)
0)
0)
0)
0)
0)
0)
3) §11
1)
2/189(1
0/193(0
1/97
0/94
0/93
2/92
2/88
4/90
7/94
5/94
(1
(0
(0
(2
(2
(4
(7
(5
.0)
.0)
.0)
.0)
.0)
.2)
.3)
.4)§
.4)§
.3)§
16/93(17. 2)§#
10/86(11. 6)§
0/199(0.
1/199(0.
1/100(1.
1/100(1.
4/97 (4.
2/100(2.
4/100(4.
2/97 (2.
4/100(4.
4/100(4.
1/100(1
2/97 (2
0)
5)
0)
0)
D§
0)
0)§ ,
1)
0)§
0)§
.0)
.1)
    *Animals sacrificed intentionally at 6, 12, and 18 months after acrylonitrile
administration are included in the denominator.

    tNumbers in the parentheses are percentages.

    §Statistically significant at P < 0.05.

    HThese rats had astrocytoma in both brain and spinal cord.

    $0ne rat had unilateral papilloma/carcinoma.
                                            13-100

-------
100 ppra.  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 .
13.5.1.1.4  Litton-Bionetics Study (Bellies et^ al. , 1980)




    Beliles j|t_ 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 F2b




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 types of tumors  observed were the same as the




statistically highly significant tumor incidence of Quast et_ a^L. (1980a).  This




study provides further confirmation of the incidence of astrocytoma and  Zymbal




gland tumors observed by Quast et_ ajU  (1980a).
                                     13-101

-------
        TABLE 13-25.  INCIDENCE OF TUMORS OBSERVED IN RATS DURING
                   A THREE-GENERATION REPRODUCTIVE STUDY
                          (Bellies et al., 1980)
Generation
                                        Astrocytoma Incidence
                                Dose (ppm acrylonitrile in water)
                                          100               500
FO
Fib
F2b
Total
0/19 (0%)
0/20 (0%)
0/20 (0%)
0/59 (0%)
1/20 (0%)
1/19 (5.2%)
1/20 (5%)
3/59 (5%)
2/25 (8%)
4/17 (23.5%)
P = 0.036*
1/20 (5%)
7/62 (11.2%)
P = 7.8 x 10~3*
Generation
                         0
  Zymbal Gland Tumor Incidence
Dose (ppm acrylonitrile in water)
            100                500
F2b
                        0/19 (0%)

                        0/20 (0%)


                        0/20 (0%)
         0/20 (0%)

         2/19 (10.5%)


         0/20 (0%)
                                     1/25 (4%)

                                     4/17 (23.5%)
                                     P = 0.036*

                                     3/20 (15%)
Total
0/59 (0%)
                                          2/59 (3.4%)
                            8/62 (12.9%)
                            P = 3.7 x 10~3*
 *P-values calculated using the Fisher Exact Probability Test.
                                  13-102

-------
13.5.1.2  Gavage Studies




13.5.1.2.1  Maltoni 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




oil, 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.
13.5.1.2.2  Bio/Dynamics 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 Bio/Dynamics




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.
                                     13-103

-------
    Histopathological evaluations presented in Table 13-26 show that there were




statistically significant increased incidences in tumors of the brain and ear




canal (Zyrabal 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.









13.5.1.3  Inhalation Studies




13.5.1.3.1  Maltoni at 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 &t_ _a!L. (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).









13.5.1.3.2  Dow Chemical Company (Quast et_ a!L., 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 Sprague-Dawley rats (Spartan substrain) were exposed  to 0,
                                      13-104

-------











o
M
M
H

1 	 |

^_i
PQ

j,..
»J
O

i — |
2
£-1 X"N
M O
£3 O
O OO
tt •"•*
O •>
o
Q a
Cd M

CO
CO CJ
r ' [ >r_i
<5 d
Pi cd
C
i-l Q
W O
O -r-l
Q ^
1-1

^
M

py^
O
K
H

.
CN
1
co
t — !
W
pf}
H





















"ti
cd
tH
O

t>-»
^
cd
1
1


a)
a
•H

CD
(1)

0
M


^^
d 1
,C O O
0 0 rH
cd *H i — !
0 O -H
0 VJ CX
4-1 cd cd
CO O CM

x^,
TZJ l~!
a M
cd a>
.H O CD
tH O cd
cd co @
C rH 3 0
cd cd o c
O ,0 d -H
0 cd 0
J-J >". 3 !-<
cd N cr1 cd
pd •-' CO O

*o cd
^ 0
o o
U 4-1
t>N
iH CJ
cd o

P. CO
co -,
pa •< o





0)'
CO




j 	 (
CD J>N
> cd
dJ 'O

~60
cu .^
CO "~^
O 60
Q 0

/_^
5^2
^ —
*f~— ^»x
|

1 0
1 — 1

r-


Q
0

^ 	
0 1
0 1
r-l 1

O



QQ
CN CN
^ ^
o^ o^
o*\ o^
CN CN



x-s x~s
B-S B~S
r-H 0
\^S **~*S
ON OO
7^ CD



X~N ^~N
B^ 5s?
0 0
^-X ^^
O
-* o
ON ^-H
0 0



Q Q"
CS1 r-l
***• X ^-^

O
O ON
r-H ON
CN -—I





S PK













O

^^
g^g
vO
•1— ^^
1
1 0
1 0
1 — 1
•^^
vD


Q
r— 1
x 	 '
-1—
0 1
0 1
—1 1

I— <



s Q
vO ^
rH
cd
a
cd
4-J
O
C

CD
0)
3
CD
CO
•H
H
•t—




13-105

-------
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)

Number of animals of each
sex
Mammary tumors (female)
Mammary tumors (male)
Zymbal gland (female)
Zymbal gland (male)
Encephalic tumors (female)
Encephalic tumors (male)
Fores toraach papillomas
(female)
Forestomach papillomas
(male)
Skin carcinomas (female)
Skin carcinomas (male)
Uterine carcinomas
(female)
Total tumors (female)
Total tumors (male)
0
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)
5
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)
10
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)
20
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)
40
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.
                                   13-106

-------
 20, or 80 ppra 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




 GAG has  requested these individual data from Dr. Quast and the Chemical




Manufacturers Association.
                                     13-107

-------





t 1
M
bJ

<
H
us
td
E-

Jj
S
Cf


ti3
.J
OS
H



K
U
"*•
U
H o
u. eo
o o\
Ui "*
t-4 M
cn «
< **'
5 ***!
<.  p^-

^



00
o
tn CN m
en en p"*-
^





oo m
o •
en en p^
1—1




E
o.

.— i

^
J--
0 E
•u o.
•H CX
O &
o|¥
# M •*
m o
CM ^x J
o-o
yl-^ fjj ^
o s: -°
0)
X i-H iH
0) O i-l
C 0) -U

P- C
a> o

JJ O f-i
nl *S nl
f~ QJ 2
IW S AJ
0) I (U


rH*
] i'° °
1 CO g Z

G to*
g ?,0 CM
o in %j o
Z CT\ QJ
-H U
3


+—
^

o
a) tn
C +|K oo
O OO (U •
Z 00 4J CO
cd
3


•I—
>p
C
OA 0 O m
O O U -*
Z Z Q)
01
S

^
(U (U O
c e -H
0 0 V.
21 2
M
^

-t—
4?
a) a) o
c c m
o o n
Z E  O
CO S B-
4J o e
CO >-3 CX
O CX
CJ T-l *
•H (U to
j= .c CD
ex AJ to
CO 0)
j-i jc: x
^ -H C
O C cO
» s_x iH
ej o co c
_3 Q) CM CO -H
O CXO -H

>^ cx ni ta *H
jo ex co E

en •* to ex •»-) )_j
T3 CJ S CX E
C ,-J O C

O J"> cO "^3 'rl
CX >> 4-1

O QJ tH T3 4J
- Q Tj 7J Jj Q
CU AJ CO O CO
JS (U iH V< U
AJ O O (0 CU
O • P^ !£

CM — t m
^_0_Hx-x^oocn-.-.en-i
^ o a o ^ G  ^^ ^j1 ^.^ J^ ^"^ ^ ^
C-g .r-fcMCM/^N >-H • • *x-\ •
oSoo o^SoSSS^S
*^s . x.^ ^/ >^x o • >— ' • • • O •

ozooo ssozzs z

rH
CM CM, cMi-HincMCSmCM i-l
COOcOOcO OOOOOOO <1>
OCSOQ-^ QQOOOOO E
2 Z ZZZIZZZIS AJ
to
AJ

•H
CM CM CM i— I in CM CM in CM CJ*
COOCOO OOOOOOO «H
OQOCaO QPQOOOO Ji
z z zzzzzzz ^
o

tn
•H
tn
CO

0)
s
»H
o
>

00

CO

rt
4J
CO
Q






























e
a.
m
CM
u
o
*rt

CU
Q

•a
CU
j_>
CO
B
•H
4J
cn
01

cu •
C S
o a.
jj a.
cu
o o
(0 O
•"*
o w
3
13 0
J-3
cO O
4J
CO T3
CD QJ
CX AJ
-a *E
•H rH
IM

d ^

*o AJ
"c *^

J-l
rH en

E Q)
•K 4-





























J
C
en
OJ
ex
4-1
£
3
o
1
 w-co


















































•

rH rH
CO J3
AJ CO
E 3
cn
cn cO
3 i
•u C
3
O AJ
M-l 3
(U
'S T3"
AJ
to o

CU

•H
>» CO

cO
CO CO

2. JJ
ja O




13-108

-------
          TABLE 13-29.  CHAMBER CONCENTRATIONS OF ACRYLONITRILE VAPORS
                             (Quast et al., 1980b)
Target Concentration (ppm)
       Exposure Group
20 ppm               80 ppm
Analytical Concentration* (ppm)           20.0+1.9
  X + S.D.

Range of Daily Analytical Concentration    7.3-35.0
  (ppm)

Nominal Concentrationt (ppm)              20.7+2.6
  X + S.D.

Range of Daily Nominal Concentration     11.00-37.8

Average Analytical Concentration/
  Average Nominal Concentration            •   0.97

Number of Exposure Days Within _+10%
  of Target Concentration                      433

Number of Exposure Days Within +_25%
  of Target Concentration                       64

Total Exposure Days                            507§

Total Number of Exposure Analyses             1484
                   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.
                                      13-109

-------
              TABLE 13-30.  CUMULATIVE MORTALITY DATA OF MALE 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
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
1
1
2
2
3
3
4
5
5
6
6
8
9
12
15
26
34 .
38
47
59t
68
72
81
86
0
0
1
1
1
2
6
12t
13t
14t
16t
18t
19t
22t
24t
28t
39 1
47t
56°
63t
76T
83t
85t
94t
96t
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.
                                      13-110

-------
             TABLE 13-31.  CUMULATIVE MORTALITY DATA OF FEMALE RATS
           EXPOSED BY INHALATION FOR 2 YEARS TO ACRYLONITRILE VAPORS*
                              (Quast £t 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
80 1
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.
                                      13-111

-------
              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
  fibroadenoma/adenofibroma

Mammary gland
  adenocarcinoma

Mammary gland tumors
  (benign and malignant)

Brain and/or spinal cord
  glial cell tumors
  (benign and malignant)
 2/100


 2/99



 0/100



 1/96




 0/100


79/100



 9/100


88/100



 0/100
     4/100               11/100
                (P = 0.009)*

     2/20                15/98
                (P = 7.03 x.lO~4)*
     4/99                22/99
                (P = 5.71 x 10~8)*
     0/14                7/89
                (P = 0.0251)*
     1/100               11/100
                (P = 3.65 x!0~4)*

     96/100              75/100
(P = 2.06 x 10~4)*
     8/100


     96/100
     8/100
(P = 0.003)
         20/100
(P = 0.022)*

         85/100
         21/100
(P = 3.32 x 10~7)*
    *P-values calculated by Fisher's Exact Probability Test.
                                     13-112

-------
    In summary, in the Dow study acrylonitrile induced a statistically




significant increased incidence of tumors in male and female rats following




exposure by inhalation.
                                      13-113

-------
  13.5.1.4  Cell Transformation Study




  13.5.1.4.1  Parent and Casto (1979)




      Parent and Casto (1979)  reported the effect of  acrylonitrile  on  Syrian




  golden hamster embryo cells  (EEC),  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.
                                      13-114

-------
TABLE 13-33.
ENHANCEMENT OF SA7 TRANSFORMATION BY TREATMENT OF HEC WITH ACN*
             (Parent and Casto, 1979)
Time of ACN
treatment
18 hr before SA7




18 hr before SA7




5 hr after SA7



ACN
ug/ml
200
100
50
25
0
200
100
50
25
0
200
100
50
25
0
Surviving
fraction!
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
SA7
foci§
1
32
41
45
50
6
19
31
37
20
26
75
37
41
41
Enhancement
ratioK
0.3
1.8
1.3
1.2
1.0
1.6
2.1
1.5
1.8
1.0
8.9
874
1.5
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.
                                        13-115

-------
    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.
                                      13-116.

-------
                 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 HEC 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.
                                       13-117

-------
 13.5.2  Epidemiologic Studies

 13.5.2.1  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 Caraden,  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.
                                      13-118

-------
 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  respiratory cancer,  5 versus  1.4




 (P  < 0.05).   Thus,  this study  provides  some  evidence that  acrylonitrile 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 cancer 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.
                                      13-119

-------
     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 (GAG) 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 non-lung 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
    *The 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 nasopharyngeal, 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.
                                      13-120

-------
cohorts, the relative incidence of lung cancer in the plant population is
    (1)
  = 1(0.30)I0 + 10(0.50)IQ + 20(0.20)I0 = 9.3 IQ
where Io 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 I
where Ig 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)
                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
                                       13-121

-------
 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 (!„/!„),
                                                                     IT   o


 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 smokers 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.






 13.5.2.2  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
                                      13-122

-------
in. two areas with, potential exposure to acrylonitrile.  The first area was a




nitrile 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).
                                     13-123

-------
     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.








 13.5.2.3 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 Rheinhessen-Pfalz, and the
                                      13-124

-------
Federal Republic of Germany (FRG).   A total of 89 deaths was 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 (ll




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.
                                       13-125

-------
    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.








13.5.2.4  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, though 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
                                       13-126

-------
 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  results  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  through  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 resulting in an accumulation of 3,241




person-years of which only  780 person-years  occurred to individuals age 55  and
                                      13-127

-------
 over.  Like  several  other  studies,  this  is a  relatively youthful cohort and




 the number of expected deaths  is  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.









 13.5.2.5  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).
                                      13-128

-------
13.5.2.6  Zack. (1980, unpublished)



    Judith A. Zack (1980, unpublished) 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, was 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.
                                       13-129

-------
 13.5.2.7  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 makes any



conclusions at best tentative.
                                      13-130

-------
     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.









 13.5.2.8  Kiesselbach ejt 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
                                      13-131

-------
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
                                                t


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 translator's 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.
                                      13-132

-------
    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, supervisory, 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.
                                       13-133

-------
     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 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  froiff 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.
                                      13-134

-------
13.5.2.9  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
                                      13-135

-------
 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.








 13.5.2.10  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.
                                      13-136

-------
    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."
                                       13-137

-------
13.5.3  Quantitative Estimation

    This quantitative section deals with the unit risk for acrylonitrile in air

and water and the potency of acrylonitrile relative to other carcinogens that

the GAG has evaluated.  The unit risk estimate for an air or water 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 they breathe

or to 1 ug/1 in the water they drink.  These calculations are 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 are taken from one or both of

the following:  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 pn
                                      13-138

-------
our current understanding of the mechanisms of carcinogenesis for guidance'as to




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 that the quantal type of biological response, which 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
                                      13-139

-------
 regarded as conservative, representing the most plausible upper-limit for the



 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.








 13.5.3.1.1 Animals




 13.5.3.1.1.1   Description of  the Low-Dose  Animal Extrapolation Model




     Let P(d)  represent  the  lifetime risk (probability)  of  cancer  at  dose  d.   The



 multistage model has  the form
P(d)
                    - exp  [-(q
                              0
where
                           q± _> 0, i = 0, 1, 2, ..., k
Equivalently,
           Pt(d) = 1 - exp [(qid + q£d2 + ... + q dk)
where
                                        - P(o)
                                     - P(o)
is the extra risk over background rate at dose d or the effect of treatment.
                                      13-140

-------
         point estimate of the coefficents q.,  1=0,  1, 2,  ..., k, and




 consequently the extra risk function Pfc(d) at  any given dose d, is calculated




 by maximizing the likelihood function of the data.




     The  point estimate and the 95% upper confidence limit of the extra risk,




 Pt(d), are calculated by using the computer program GLOBAL79 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,  q*,  on  parameter q1 .




 Whenever q^^ > 0,  at low doses the extra  risk Pt(d) has  approximately the




 form Pt(d)  = qx x d.   Therefore,  q* x d  is a 95% upper  confidence




 limit on the extra risk and R/q*  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,  q*, is  calculated by increasing




 q^  to a  value q*  such that  when the log-likelihood is remaximized




 subject  to  this fixed value,  q*,  for the linear  coefficient,  the resulting




 maximum  value of  the  log-likelihood LI satisfies  the equation
2 (L0 -
                                       = 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 Pt(d) is




an improvement on the Crump jit 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 the section calculating the risk estimates, Pt(d)



will be abbreviated as P.)
                                      13-141

-------
    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 dose not fit the data sufficiently well, data
at the highest dose is deleted, and the model is refit 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
                            £
is calculated where Nj[ is the number of animals in the itb- dose group, X-j^
is the number of animals in the itn dose group with a tumor response, P^ is
the probability of a response in the 1th 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 X2 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.


13.5.3.1.1.2  Selection of Data
    For some chemicals, several studies in different animal 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
listed as follows.
                                       13-142

-------
     L.  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

 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  that gives the highest estimate of the lifetime carcinogenic  risk, q*,

 is selected in most cases.   However,  efforts are made to  exclude  data sets that

 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 that has  the larger sample size is

 selected for calculating  the carcinogenic potency.


     2.  If there are two  or  more  data sets of comparable  size  that 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 Aj_, A£, ..., A^ is defined  as



            --—-— "~(Ai  x A2  x ... x
W


     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.




 13.5.3.1.1.3  Calculation of Human Equivalent Dosages from Animal Data

    Following the  suggestion of Mantel and Schneiderman (1975), 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 two-thirds power of the

weight as would be the case  for a perfect sphere,  the exposure in mg/day per

 two-thirds power of the weight is also considered to be equivalent exposure.  In
                                       13-143

-------
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

Then, the lifetime average exposure is
                               d =
                                   LeTw273


13.5.3.1.1.3.1  Oral

    Often exposures are not given in units of mg/day and it becomes necessary to

convert the given exposures into mg/day.  For example, in most feeding studies,

exposure is in terms of ppm in the diet.  Similarly, in drinking water studies,

exposure is in ppm in the water.  In these cases the exposure in mg/day is


                                m = ppm x F x r


where ppm is parts per million of the carcinogenic agent in the diet or water, F

is the weight of the food or water 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.  Water demands are also assumed

proportional to the surface area, so that
or
                              m  oc ppm x
                                     m
                                        W2/3
                                             x r
                                      13-144

-------
    A.S a result, ppm in the diet or in water is often assumed  to  be an



equivalent exposure between species.  However, this is not  justified because the



calories/kg of food are very different in the diet of man compared to laboratory



animals primarily due to moisture content differences.  Consequently, the amount



of drinking water required by each species differs also because of the amount of



moisture in the food.  Therefore, an empirically-derived factor,  f = F/W, is



used which is the fraction of a species body weight that is consumed per day as



food or water.  We use the following rates:



                                          Fraction of Body

                                          Weight Consumed as
Species
Man
Rats
Mice
W
70
0.35
0.03
rfood
0.028
0.05
0.13
rwater
0.029
0.078
0.17
Thus, when the exposure is given as a certain dietary or water concentration in



ppm, the exposure in mg/W^/^ ^s
m
                     ppm x F   ppm x f x W         -   T,i /•?
                   = rt^ 9/n  = "	Vjo	 = ppm x f x W1/J
                       W2/3        W2/3
When exposure is given in terms of mg/kg/day = m/Wr = s, the conversion is



simply




                                     s x W1/3
                m
                             rW2/3




13.5.3.1.1.3.2  Inhalation



    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
                                      13-145

-------
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.



13.5.3.1.1.3.2.1  Case 1

    Agents that are in the form of particulate matter or virtually completely

absorbed gases, such as sulfur dioxide, can reasonably be expected to be

absorbed proportional to the breathing rate.  In this case the exposure in

rag/day may be expressed as


                                 m = I x v x r


where I - inhalation rate per day in m3, v = mg/m3 of the agent in 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 mice, I = 0.0345 (W/0.025)2/3 m3/day

                  For rats, I = 0.105 (W/0.113)2/3 m3/day


For humans, the value of 20 m3/day* is adopted as a standard breathing rate

(ICRP, 1977).
    *From "Recommendation of the International Commission on Radiological
Protection," page 9.  The average breathing rate is 10' cnr5 per 8-hour
workday and 2 x 10^ cm3 in 24 hours.
                                      13-146

-------
    The equivalent exposure in mg/W2'^ 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 previously

stated relationships, are tabulated as follows:

                                       W       i = I/W
Species
  Man
  Rats
  Mice
                                     70
                                      0.35
                                      0.03
        0.29
        0.64
        1.3
Therefore, for particulates or completely absorbed gases, the equivalent

exposure in mg/W2'^ is
                          m
        Ivr
                                       iWvr
^273

                                        273
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.



13.5.3.1.1.3.2.2  Case 2

    The dose in mg/day of partially soluble vapors is proportional to the 02

consumption, which in turn is proportional to W2'^ 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-'^ t where k is a constant independent of species, it

follows that

                             m = k W2' ^ x v x r
or
                             d =
                                 W
           m
          2/1
kvr
As with Case 1, in the absence of experimental information or a sound
                                      13-147

-------
 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/m3 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.



 13.5.3.1.1.4  Calculation of the Unit Risk from Animal Studies

    The 95% upper-limit  risk associated with d mg/kg2/3/day is obtained from

 GLOBAL79 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 the

 number of mg/kg^'^/day corresponding to one unit of X is determined and

 substituted into the  above relationship.  Thus, for example, if X is in units of

 ug/ra3 in the air, then for case 1, d = 0.29 x 701/3 x 10~3 mg/kg2/3/day,

 and for case 2, d - 1, when ug/nr3 is the unit used to compute parameters in

animal experiments.

    Exposures given in terms of ppm in air can be converted to units of mg/m3

by the formula

                  1 ppm - 1.2 x molecular weight (gas) mg/m3
                                molecular weight (air)


Note that 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 = 1, 2	k
                                       13-148

-------
and use mg/kg equivalents for the unit risk values.



13.5.3.1.1.4.1  Adjustment for Less Than Lifespan Duration of Experiment

    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,  q*, 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)]


 13.5.3.1.1.5  Interpretation of Quantitative Estimates

     For several reasons, the unit risk estimate based on animal bioassays is

 only an approximate indication  of the 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
                                       13-149

-------
  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.,
 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.


13.5.3.1.1.6  Alternative Methodological Approaches
    The methods used by the GAG for quantitative  assessment are consistently
                                      13-150

-------
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 GAG in the




following analysis but are included for comparison in the appendix.  The models




presented there are the one-hit, Probit, and Weibull.  The GAG feels that with




the limited data available from these animal bioassays, especially at the




high-dose levels required for testing, almost nothing is known about the true




shape of the dose-response curve at low environmental levels.  The position is




taken by the GAG that the risk estimates obtained by use of the linear




non-threshold model are plausible upper-limits and the true risk could be lower.




    In terms of the choice of animal bioassay as the basis for extrapolation,




the general approach is to use the most sensitive responder on the assumption




that humans are as sensitive as the most sensitive animal species tested.  For




acrylonitrile, the average response of all of the adequately tested bioassay




animals was used; this is because three well-conducted valid drinking water




studies using different strains of rats showed similar target organs and about




the same level of response.




    Extrapolations from animals to humans could also be done on the basis of




relative weights rather than relative surface areas.  The latter approach, used




here, has more basis in human pharmacological responses; it is not clear which




of the two approaches is 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.  In the case of the




acrylonitrile drinking water studies, the use of extrapolation based on surface




area rather than weights increases the unit risk estimates by a factor of 5.8.
                                      13-151

-------
13.5.3.1.2  Humans—Model for 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 BJJ.  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




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 linear 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 BJJ is the increased probability of cancer associated with each unit
                                      13-152

-------
 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
                      = L- = A + BH  (X1 + x?)
                        P0   A + BH  Xl
or
RP0 = A + BH
                                            x2)
where x^ = lifetime average daily exposure to the agent for the general

population, X2 = lifetime average daily exposure to the agent in the

occupational setting, and PQ = lifetime probability of dying of cancer with no

or negligible acrylonitrile exposure.

    Substituting PQ = A + BH x^ and rearranging gives


                           BH = P0 (R - l)/x2


To use this model, estimates of R and x2 must be obtained from the

epidemiologic studies.  The value PQ is derived from the age-cause-specific

death rates for combined males found in the 1976 U»S. Vital Statistics tables

(U.S. Department of Health and Education,  and Welfare, 1976) 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.
                                     13-153

-------
13.5.3.2  Acrylonitrile Risk Estimates



13.5.3.2.1  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, B^, is




calculated as follows:






                              „_ _ pn (R - 1)
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 (GAG, 1978).




    The Du Pont Chemical Company follow-up study by O'Berg (1980) of 1,345




workers exposed to acrylonitrile between 1950 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, 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.
                                      13-154

-------
    Estimates of exposure at the plant are those suggested by J. Brown, an

industrial hygienist at NIOSH, 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 acrylonitrile 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
                        ic        8   240    9
                   Xl = 15 PPm x 24 X 365 x 60

                      = 0.5 ppm continuous equivalent lifetime exposure
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 BJJ [in (ppb)""1] is derived using the above estimates as

follows:
                           BH =
                              _ 0.036 (3.1 - 1)
                                     500 ppb
                              = 1.5 x 10~4 (ppb)"1


Therefore, the lifetime risk of cancer for people continuously exposed to 1 ppb

is 1.5 x 10~4.
                                      13-155

-------
    To express risk in terms of ug/m3 concentration, the conversion factor for

acrylonitrile is
ppb
                   .  x
                           acrylonitrile = 1.2(53.06) .
                        MW air
                               28.8
or

                              1 ug/m3 = 0.45 ppb

                                                                            o
Therefore the upper-bound risk associated with a lifetime exposure of 1 ug/m

in air is


                     P = 1.5 x 10~4 x 0.45 = 6.8 x 10~5




13.5.3.2.2  Unit Risk Estimate Based on Animal Studies

    Unit risk estimates are calculated based on both inhalation and drinking

water studies.                                                                >




13.5.3.2.2.1  Drinking Water Studies

    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).

These three studies are evaluated separately and then  combined to  determine  the

mean estimate.

    The  individual results  of  the Dow Chemical Company rat  study  (Quast £t al.,

 1980a)  are shown in Table  13-21  and 13-22.   Since  several  tumor  sites are

affected,  the  overall risk  of  tumors is  determined  from the number of animals

having  tumors  that are statistically significant  at any of  the  sites. This  is

 shown  in Table 13-35A.   Presented there  is  the percentage  of animals  by  sex  and

 by dose with the tumors  discussed above*   In addition, the small intestine and
                                       13-156

-------
  TABLE 13-35.  NUMBER (PERCENTAGE) OF RATS DEVELOPING TUMORS IN AT LEAST ONE
        TARGET ORGAN:  THREE ACRYLONITRILE DRINKING WATER BIOASSAYS, BY
                                DOSE AND BY SEX

                 A.  Dow Chemical Company (Quast et a^., 1980a)
                   (Target Organs in Tables 13-21 and 13-22)
           Males
Animal Dose   Number/Total
(mg/kg/day)    (Percent)
                  Females

             Including Mammary  Excluding Mammary
               Gland Tumors       Gland Tumors
Animal Dose    Number/Total       Number/Total
(mg/kg/day)     (Percent)          (Percent)
0
3.42
8.53
21.18
4/80 (5.0%)
18/47 (38.3%)t
36/48 (75.0%)t
45/48 (93.8%)t
0
4.36
10.76
4.27
59/80 (73.8%)
47/48 (97.9%)§
46/48 (95.8%)§
48/48 (100%)§
3/80 (3.8%)
24/48 (50.0%)t
37/48 (77.1%)T
45/48 (93.8%)t
                         B.  Bio/Dynamics Inc.  (1980a)
                         (Target Organs in Table 13-23)

Animal Dose
(mg/kg/day)
0
0.09
7.98
Males
Number/Total
(Percent)
6/100 (6.0%)
6/98 (6.1%)
36/98 (36.7%)

Animal Dose
(mg/kg/day)
0
0.15
10.70
Females
Number/Total
(Percent)
1/100 (1.0%)
5/100 (5.0%)
41/99 (41.4%)
                                     13-157

-------
C.  Bio/Dynamics Inc. (1980b)
(Target Organs in Table 13-24)

Animal Dose
(mg/kg/day)
0
0.11
0.25
0.81
2.49
8.15
Males
Number/Total
(Percent)
5/200 (2.5%)
4/100 (4.0%)
5/100 (5.0%)
7/100 (7.0%)
20/100 (20.0%)
34/100 (34.0%)
Females
Animal Dose
(mg/kg/day)
0
0.12
0.36
0.82
3.65
10.89

Number/Total
(Percent)
2/199 (1.0%)
2/100 (2.0%)
6/101 (5.9%)
10/97 (10.3%)
13/100 (13.0%)
33/98 (33.7%)
              13-158

-------
 mammary  gland,  which, are target sites for the female Sprague-Dawley rats,  are




 included.   For  the  mammary gland,  however,  because the  control group incidence




 is so  high,  the percentages for the females are tabulated separately,  and  the




 risk extrapolation  is  calculated excluding  mammary gland  tumors.




    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 male rat (Table 13-35A) is






                 3.42 mg/kg/day  t (70/0.350)1/3 = 0.6 mg/kg/day







 When these human dose  equivalents  are used  with the  animal response data in




 Table  13-35A, the multistage model yields a value  of q* = 9.9  x 10"1




 (mg/kg/day)"1 for the  males,  and q*  = 9.2 x 10"1 (mg/kg/day)'1 for




 the females  excluding  mammary gland  tumors  (Table  13-36).  (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  (70/0.35)1/3  =  5.8.)




    The results of the two Bio/Dynamics Inc.  studies using Sprague-Dawley




 (1980a) and Fischer  344  (1980b)  rats  are  summarized  in Table 13-35B and 13-35C,




 based  on tumors of the individual  target  organs in Tables 13-23 and 13-24,




 respectively.  The dose  levels  in Tables  13-35B and  13-35C are presented in




mg/kg/day.   This allows for a direct  comparison between animal studies and an




 estimate of human dose equivalents.  Dividing this animal dose by the factor 5.8




 (presented above) yields the human dose equivalent used in the linearized




multistage models  When these doses are used with the animal response data, the




multistage model yields values for the upper-limit estimates  of slope all in the




 range of 0.3 to 1 (mg/kg/day)'1.  The individual and combined upper-limit




slopes  are  presented in Table 13-36.  The geometric average for males is
                                        13-159

-------
               TABLE 13-36.  ESTIMATES OF 95% UPPER-LIMIT SLOPES
                    FOR THREE DRINKING WATER STUDIES, BY SEX
     Study
                                                 Sex
  Males
 Females
Dow Chemical Co.

Bio/Dynamics Inc., 1980a

Bio/Dynamics Inc., 1980b
                                         q* in (mg/kg/day)
9.9 x 10"1

4.0 x 10"1

4.0 x 10'1
-1


9.2 x 10"1

3.7 x 10"1

2.9 x 10"1
Geometric Average
5.4 x 10'1
4.6 x 10
                                                                  r-1
                                        13-160

-------
5.4 x 1CT1 (mg/kg/day)"1 and 4.6 x 10-1 (mg/kg/day)"1 for females.  The




former slightly higher value is chosen as the slope for extrapolation purposes.




    In order to estimate the unit risk for 1 ug/liter 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/liter acrylonitrile concentration in water is then







                  P = 1 - e~5'4 x i0'1 x 2'86 x 10~5




                    = 1.5 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 following is done for comparison purposes only.  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 for 1 ug/m3, P, can be found




using this value of d and the value of q* estimated above as follows:
P = 1 - e~5'4




  = 1.5 x 10~4
lc
   l
                                                  10
                                                    ~4
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 in the following section.
                                        13-161

-------
13.5.3.2.2.2  Inhalation Studies



    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 et al. study is used




for determining a unit risk estimate via the inhalation route.




    The tumor incidence data on individual tumor types have been presented




previously (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, the responses




of both sexes 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 -j-r x y = 3.57 ppm
Similarly, the lifetime equivalent dosage for the 80 ppm group is 14.29 ppm.



Using the linearized multistage model with incidence data for females, the




carcinogenic potency is q* = 3.35 x 10"^ (ppm)'-1-, and the upper-bound




estimate of the lifetime risk of cancer associated with 1 ug/rn-^ = 4.53 x




10~4 ppm of acrylonitrile is






                  P = 1 - exp (-3.35 x 10~2 x 4.53 x 10~4)




                    = 1.5 x 10~5
                                        13-162

-------
13.5.2>. 1.3  Summary of Unit Risks




    In summary, the upper-limit unit risk estimates for humans breathing




1 ug/rn-^ of acrylonitrile in ambient air (equivalent to 0.45 ppb) are




6.8 x 10~5 based on the occupational study, 1.5 x 10"^ based on the rat




drinking water study, and 1.5 x 10~^ 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 1.5 x 10"^ x 1/5.8 = 2.6 x 10~5, a value




which is between 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/liter of
acrylonitrile in drinking water is estimated to be 1.5 x 10'
                                                           ,-5
13.5.3.3  Relative Potency




    One of the uses of unit risk is to compare the potency of carcinogens.  To




estimate the relative potency on a per mole basis, the unit risk slope factor is




multiplied by the molecular weight and the resulting number expressed in terms




of (mMol/kg/day)"1.  This is called the relative potency index.




    Figure 13-7 is a histogram representing the frequency distribution of




potency indices of 54 chemicals evaluated by the GAG as suspect carcinogens.




The actual data summarized by the histogram are presented in Table 13-37.  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.  Animal oral studies are




selected over animal inhalation studies because most of the chemicals have




animal oral studies; this allows potency comparisons by route.
                                        13-163

-------
                              4th          3rd          2nd         1st
                            quartile  I   quartile  I   quartile   I  quartile
                                    1x10+1
4x10+2
2xlO+3
                          0       24       6
                         Log Of Potency Index
           s
Figure 13-7.  Histogram representing the frequency distribution of the potency
             indices of 54  suspect carcinogens evaluated by the Carcinogen
             Assessment Group.
                                      13-164

-------
TABLE 13-37.  RELATIVE CARCINOGENIC POTENCIES AMONG 54 CHEMICALS EVALUATED
  BY THE CARCINOGEN ASSESSMENT GROUP AS SUSPECT HUMAN CARCINOGENS1>2,3
Compounds
Acrylonitrile
Aflatoxin B^
Aldrin
Allyl Chloride
Arsenic
B[a]P
Benzene
Benzidine
Beryllium
Cadmium
Carbon Tetrachloride
Chlordane
Chlorinated Ethanes
1 , 2-dichloroethane
hexachloroethane
1,1,2, 2-tetrachloroethane
1,1, 1-trichloroethane
1,1, 2-trichloroethane
Chloroform
Chromium (W)
DDT
Dichlorobenzidine
1 , 1— dichloroethylene
Dieldrin
Slope
(ing/kg/day)"1
0.24(W)
2924
11.4
1.19xlO"2
15(H)
11.5
5.2xlQ-2(W)
234(W)
4.86
6.65(W)
1.30X10"1
1.61
6.9x10-2
1.42xlO-2
0.20
1.6x10-3
5.73x10-2
7x10-2
41
8.42
1.69
1.47x10-1(1)
30.4
Molecular
Weight
53.1
312.3
369.4
76,5
149,8
252.3
78
184.2
9
112.4
153.8
409.8
98.9
236.7
167.9
133.4
133.4
119.4
104
354.5
253.1
97
380.9
Potency
Index
1X10+1
9x1 0+5
4xlO+3
9X10"1
2x10+3
3xlO+3
4x10°
4x1 0+4
4x1 0+1
7x10+2
2x1 0+1
7xlO+2
7x10°
3x10°
3x1 0+1
2X10'1
8x10°
8x10°
4xlO+3
3x10+3
4x10+2
lxlO+1
1x1 0+4
- Order of
Magnitude
(logiQ
Index)
+1
+6
+4
0
+3
+3
+1
+5
+2
+3
+1
+3
+1
0
+1
-1
+1
+1
+4
+3
+3
+1
+4
                                           (continued  on  the  following  page)
                                  13-165

-------
                            TABLE 13-37.   (continued)
Compounds
Dinitrotoluene
Diphenylhydrazine
Epichlorohydrin
Bis(2-chloroethyl)ether
Bis(chloromethyl)ether
Ethylene Dlbromide (EDB)
Ethylene Oxide
Formaldehyde
Heptachlor
Hexachlorobenzene
Hexachloro butadiene
Hexachlorocyclohexane
technical grade
alpha isomer
beta isomer
gamma isomer
Methylene Chloride
Nickel
Nitrosamines
Dime thy Initrosamine
Diethylnitrosamine
Dibutylnitrosamine
N-nitrosopyrrolidine
N-nitroso-N-ethylurea
N-nitroso-N-methylurea
N-nitroso-diphenylamine
Slope
(mg/kg/day)"1
0.31
0.77
9.9xlO-3
1.14
9300(1)
8.51
0.63(1)
2.14xlO"2(I)
3.37
1.67
7.75xlO~2
4.75
11.12
1.84
1.33
7.8xlO-3
1.15(W)
25. 9 (not by q*)
43.5(not by q*)
5.43 L
2.13
32.9
302.6
4.92xlO-3
Molecular
Weight
182
180
92.5
143
115
187.9
44.0
30
373.3
284.4
261
290.9
290.9
290.9
290.9
84.9
58.7
74.1
102.1
158.2
100.2
117.1
103.1
198
Potency
Index
6xlO+1
1x1 0+2
9x1 O'1
2x1 0+2
1x1 0+6
2x10+3
3x1 0+1
6x10-!
lxlO+3
5x1 0+2
2x1 0+1
lxlO+3
3x10+3
5x1 0+2
4x1 0+2
7X1Q-1
7x1 0+1
2x10+3
4x10+3
9x1 0+2
2x1 0+2
4xlO+3
3x1 0+4
1x10°
Order of
Magnitude
(logic
Index)
+2
+2
0
+2
+6
+3
+1
0
+3
+3
+1
+3
+3
+3
+3
0
+2
+3
+4
+3
+2
+4
+4
0
PCBs
4.34
324
1x10
                                      ,+3
+3
                                               (continued on the following page)
                                      13-166

-------
                               TABLE  13-37.   (continued)

Compounds
Phenols
2,4, 6-trichlorophenol
Tetrachlorodioxin
Tetrachloroethylene
Toxaphene
Trichloroethylene
Vinyl Chloride
Remarks :
1. Animal slopes are 95% u

Slope
(mg/kg/day)"1
1.9 9x1 0~2
4.25xl05
3.5xlO~2
1.13
1.26xlO-2
1. 75xlO-2(I)

pper-limit slopes t

Molecular
Weight
197.4
322
165.8
414
131.4
62.5

>ased on the lim

Potency
Index
4x10°
1x1 0+8
6x10°
5xlO+2
2x10°
1x10°

aarized multisl
Order of
Magnitude
dogio
Index)
+1
+8
+1
+3
0
0

:aep inhHpl .
    They 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).  Human slopes are point estimates based on the linear non-threshold
    model.                                        '

2.  The potency index is a rounded-off slope in (mMol/kg/day)-1 and is calculated by
    multiplying the slopes in (mg/kg/day)-1 by the molecular weight of the compound.

3.  Not all of the carcinogenic potencies presented  in this table represent the same
    degree of certainty.  All are subject to change  as new evidence becomes available.
                                       13-167

-------
    The potency index for acrylonitrile based on the O'Berg study of Dupont



workers is 1.3 x 10+1 (mMol/kg/day)"1.  This is derived as follows.  The



slope estimate from the O'Berg study, 6.8 x ICT^^g/m3)"1, is first



converted to units of (mg/kg/day)"-*-, assuming a breathing rate of 20 m3 of




air per day and a 70 kg person.
6.8 x ID'5 (ug/m3)'1 x -^S2- x
                                       1 ug
                       20 m
                                     10
                                        ~

                                       ~3
                                             x 70 kg
                                                     = 0.24 (mg/kg/day)
                                                                       -1
Multiplying by the molecular weight of 53.1 gives a potency index of 1.3 x



10+1.  Rounding off to the nearest order of magnitude gives a value of 10+1,



which is the scale presented on the horizontal axis of Figure 13-7.  The index



of 1.3 x 10+1 lies at the bottom of the third quartile of the 54 suspect



carcinogens.



    Ranking of the relative potency indices is subject to the uncertainty of



comparing estimates of potency of different chemicals based on different routes



of exposure to different species using studies of different quality.



Furthermore, all the indices are based on estimates of low-dose risk using
           '                                                                (


linear extrapolation from the observational range.  Thus, these indices are not



valid to compare potencies in the experimental or observational range if



linearity does not exist there.
                                        13-168

-------
 13.5.4  SUMMARY




 13.5.4.1  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 matabolism 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  13-38.  In  addition,  ten




 epidemiologic studies  of  cancer  incidence  have been reported.  These are




 summarized  in Table  13-39.  A  short description  of these  studies follows.




     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
                                        13-169

-------

















CO

^*

0 a)
CU rH
e ?
4-J *^>
cd a)
CU 43
M a
H W
^
rH

cu


(1)
CO
o
o
c
0
•rl
H [ ^ i
0 cd
a) 4-1
4-1 CO
3 tH
O 0
£•
rrH
 i — 1 Cd rH £•*>
CO 60rH Cd rH
cd 3 B 0
B rH 13 CU 0
o cd 0 MH
4J 43 cd s_x 13 CO
>, g rH 0 CU
0 >, 60 13 cd rH
0 Nl 0 0 rH cd
M v_x O cd tO S
4-J 0 /- * t-H v_x-
CO rH ^ 0 60 p~,
cd cd 0 M co
v^ cd 43 iH >^ cd cd
Cd O 4J M CU 4-> CU
0cjcdJHCdS'HM
•H B O B 60 3 CJ
CdMOP.B04-10
M cd 4-J cd o -H cd
pq pq co S H CM PM



•
o
B


^"1
rH
0
o

CO
0)
rH
cd
S
CU
m
s«x
CO
3
M
0)
4-1
n


























































13

s~ "•* rH
CO 60
cd
B rH
O cd
4-1 43
>» B
CJ >k
O tsl
4-1
CO rH
cd cd
0 0
•rl
cd M
!-! Cd
•(—
t^
£

LO









B
p.
p.

0
o
m




j_j
cu
0 4-J
•rl Cd

pj
O 60
•H 0
4J -H
CO ,ij
CU 0
60 T-l
0 $-1
•rl 13




J-l
CU
•rl
Pd

CO
CU
rH

cd
43
CJ







X— \ *
0
00

, — 1


e
rH
cd

4-1
CU

CO
CU
rH
•H
rH
CU
PQ


Q
cd

CO 60
cd
B rH
. o cd
4-> 43
>> 0
CJ (=-,
O IS]
!-i s-'
4J
CO rH
cd cd
s-' 0 43
cd cj
0 cj cd
Cd rl O
1-1 Cd 4-1
pa w co
0
e
o

1

1m^








B
p.
p.

0
0
i — i




p
cu
0 4J

^
rj
O 60
•rl 0
4-1 -rl
CO A!
CU 0
60 -rl

•H "^3









0
cd
4J

cd
p.




^- s
cd
o
00
cr>
i — i
v^

9
CJ
0
M

CO
CJ

cj
cd
0
Q

O
•H
M


13
0
cd
^"N r~H
CO 60
cd
B rH
O cd
4-1 43
>> B
CJ >->
O Csl
}-i s_x
4-1
CO H
cd cd
^ 0 43
cd cj
0 cj cd
Cd rl O
M Cd 4->
M pa co
0
a
vD
CM
1
CO
CM





B

P.

0
o
I—I
1
o
r-l




p
• cu
0 4->

J5j
0
O 60
•H 0
4J -H
CO ^
CU 0
60 -rl
0 M
•rl *^3






;*
CO

j_i
CU
f*
o
CO
•rl
Pn



/^-v
43
O
CO
CTi
i — 1
V— '

e *
O
0 '
M

CO
0
•rl
B
cd

o"
*^
0
•H
pq
3
r-l
0
O
CO
CU
rH
cd
cu
MH
0
cd
rH
60
&
Mamma


















































































v»l
r~1
0
x-x O
13
0 CO
cd cu
x-M-H rH
co 60 cd
cd E
B rH CU
O Cd 4-1
] 1 O Srf f
>. B
CJ >-> 13
O N 0
MS-^ cd
4J rH
CO rH 60
cd cd
s-' 0 43 >,
cd cj M
0 o cd cd
•H SB
cd M o B
M cd 4J cd
pq [il co 5;
i
o
CM
1

I— 1



K**t
cd

*«*^.
60
^

60
B

o
r- 1












CU
60
cd
>
cd
60









0
cd
4J
M
cd
p.
CO



X— N
o
o
OO
ON
r—l
V^ '

9
CJ
0
I-H

CO
o
•rl
c]
cd
0
p"
*^^.
0
•H
PQ
cu
60
p.
60
5
"J*
0
rH
._(
0^
SMH
*r! 0)
4-J 2
25
0 e
§ o
60 •*
^ CU
0 2
8-S
s g
cu v
43
4-1
•
CU 0
CO -rl
o
13 13
cu
< [ ^
0) M
& CU
O CO
rH 43
o
CU
43 CU
4-1 M
cu
4_) J5
cd
CO
13 M
cu o
> S
M 3
CU 4-)
'CO
43 CU
0 43
4-1
CU
fa • «t
Q) ^">
[5 ^O
r3
£>-, 4-J
*& CO
S
4-> OJ
CO >
•H
^ 4J
0 0
Ctf r3
Q) *^
l-l JH
O P^
m 
-------



































/"X
tJ
CU
3
d
•H
JJ
d
o
o
**^


•
CO
CO
1
ro
i»H

w

i

























'





















CO
S-i
o

H
MH
0

CU
4-1
•H
CO



4-1
d cu
CU rH
B 3
4-J *d
03 0)
cu ,d
S-i G
H CO



•K
rH
CU

CU


CU
CO
o
n




d
o

MH 4-J
O 03
S-I
CU JJ
4J CO

o 'd
* j

 60

T\ ^
a vi
td cd
o ^
CO £J

•
I_»^
^ fe
Cd ""••*.
*t3 03
""^ ^**
S-i cd
,£5 T3

, 3
JJ iH Cd S-i 60
^•x CO 60 P 03 d
>^ >-. JJ S O
rH CO iH B JJ
d 03 rH Cd
O CO ,Q 03 B "•
3 S d ^x
CO 0 >> -H •« >,
CU > Cxi 4J ^ rH /-x
rH S-I x^ CO >i d >*
03 CU CU iH O rH
S d rH JJ d d
CU Cd d O CO O
MH rH d -H CU
x-' Cd Cd O CO r-H CO
s-i o s-j cu cd cu
d JJ JJ rH g rH
•H d s-i co cd a) 03
4i> S
cd ~-.
CN *d CO -sf"
»-H **^ ^-, CM
SJ 03
SJ ,d 13 M
O 0
MH vO in MH





B
P.
ex

o
C30
1
o
CM








d
0
•H
4-J
03
rH
5
d
•H


a)
rH
cd
Q
1
a)

60
cd
J_l
ex
CO


^^^
,Q
0
00

1 — (


•



JJI
OJ

4J
CO
0)

01











•
0
•H
4J
t.
fll
ti
CU
60
13
d
0
0
CU
CO

cu
,d
4J
•
cu d
CO -H
0
cu

CU S-I
5 CU
O CO
rH ,£>
O
cu
,d CU
JJ S^
0)
JJ £
cd
CO
T3 S4
CU O
> e
S-i 3
cu JJ
CO
,0 CU
O rf3
4J
CU
J-I ••.
cu >•>
S •«
>, 4J
13 CO
3
4-1 CU
CO >
0 O
cd 3
CU "T3
0
M M
O P4
M-l OJ
S-<
•T3
CU d
JJ 0
cd -H
0 JJ
•H cd
d a)
•H d
cu
CO 60
S-i 1
o cu
§CU
Vi
4J J3
4-1
cu
J3 0)
4-J
CO
rH 03
rH 5
03
CO
4J -H
O ^
* -1-
13-171

-------
















a
CO
CO
[d
CO
CO
w
co
M
oi

Cd
•1
r-l
04
M
•z
O
S
CJ
23
M

O
Cd
S

M
I
CO
Cd
M
a
CO
o
M

O

O
M
a
Cd
a
M
PM
rd


^
cr>
co
1
CO
[d
S
•3
F.






























CO
CO
iH
O
S
iH
CJ
•H
 CO d
d CU *rl 4-1
cd ,d 4-> d
O 60 CO
•H -rl 4-1 6
<4H rd CO 4J
•H CO CO
d rd 60 3
60 4-1 d •>-)
•rl -rl O 13
CO & rH cd



/-s
in
o
•
o

V
5
,-).
•
rH
in





J_l
cu
o
cd
cu o
•H 60
4J d
•rl 3
CO rH
o
PM









^^
o
OO
CT\
i-H


60
H
cu
pq

o


t
to

cu
60
o
d
•rl
CJ
to
cd
o

to
cu
o
o
4-1
co
to
3
CO
0
ft
X
cu

rH
cd
•H
4J
d
cu
4-1
0
PM



s**\
m
o
*
o

V
PM
^f
•
1 — I
^fr





to
cu
o
cd
cu o
•H 60
4-1 d
•H 3
CO rH
O
PM




r*
O
CO
d
o


*d
rj
cd

rH 
N rH

cu
o






















































CO
•H
4-1
•H
CO
O
PM









»
r-l
cd

4J
CO

CO
CO
co
•H

H


9
to

CU CO
^ d
to -rl
O S rH
5 cd >i
r-l d
d Cd «rH
60rd >
•H 4->
CO rd 13
rl ft d
o cd cd
^ f .
d cd co
O 4-1 d
CO O
§"° "£
I o cd
S 4J CJ
o o '
rH CO H
rH CU 13
O 5-1 >->
IH 3 rd
CO
cu o cj
4-1 ft -H •
CO X rH CU
rH CU CJ -d
6 13 0 to
o co >-. o
O X rH rH
d -H 0 rd
M S ft 0



X""N X"N
in m
o o
• *
o o

V V
PM PM
-* -d-
• •
-3: i— i
crT^t




H
to co
CO CJ
o d
ss
CJ
60 ft
3 r*^
rH rH
















x—\
O
OO
CT>

s_^







13
co
d
•rl
,£3
e
0
CJ


•
ft
1
o
rH
t-H
0
4-1
0
rd
CO

£>^
rH
CO
>
•H
4J
3
cu



X"N
m
o
•
o

V
PM
1^,
•
o
rO





^
cu
o
cd
co o
•H 60
4-> Cl
•rl 3
CO rH
0
PM





M
CO
4-1
to
cd
CJ

*d
£
rij s*^
•-H
J-( OO
0) C^
gl— t
X.--'
cu









d
o
•H
4->
cd
N
•H
r-l
1
O
ft
CO
e
o
to
IH

d •
O CO
•H 0)
4-1 'H
Cd r-l
rH 0
ft CJ
o cd
PM <4H



/""N
m
o
V
o

V
p-l
CTi
o
i — 1
in


to
CU
CJ
d
cd
o
rd
a
o
4J'
CO


























•
CO
d
cu
60
o
d
•H
CJ
to
cd
a

to
CO
4-1
o
o
4-)
01
r-l
CO
0
ft
X
cu

r-l
cd
•rl
4-1
d
CU
4-1
O
PM





















CO
^
•H
CO
3
rH
O
d
o
a
d
M









^**
OO
r-

i— i
^s

d
0
CO
d
o
S




d
o

ft
3
1
fe
0
rH
rH
O
<4H
4-1
to
o •
^ 4J
co to
O
. 0
co o
N
CO O

4-1 CO
to to
O CU
rd r^
0 B
o cu
B
rH
rH 4->
Cd CO
e Q
co 6





















CO
£>
•rl
CO
3
rH
0
d
o
o
d
M








s~>*
13
CU
*» fj
O co
00 -H
CTt *— 1
i— 1 pQ
N*— "^ 1-3
ft
rW d
0 3
ct)
M
CO

CU
JD
e
3
d •
CO
co co
60 CU
to £>•>
cd o
rH rH
ft
to S
cu cu
^ *TJ
rH CU
O CO
d o
t-H ft
cu
• d
co 3
N
•rl to
CO O

4-1 r>*i
I-I rH
O rH
rd cd
o B
O TH
d
r~n »rH
CO
S 4-1
CO O





















01
t>
•H
CD
3
I — 1
CJ
d
o
o
d
M



•d
cu
to ,d
to to
3 -H
Cd rH
to ,0
4-1 3
CO ft
d
-a 3
Cl
cd •>
i — i
>-,00
0) CTi
<4H r-l
MH >-^
cd
O
^- ^
co
60
cd
ft

60
d
H
fe
o
rH
rH
O
IH

cu
4-1
d
o
13
0)
d
•H
4-1
d
o
o
s^x
























































13-172

-------






































CU
3
C
•H
4-1
fi
O
O
v-x

•
CO
1
CO
i — i

w
H










































CD
CU
•H
O
fi
CU
•H
O
•H
<4_j
CU
n

Td
c
cd

CO
4-1
pfi
00
•H
iH
^
00
•H
W

Td
cu
4J
0
CU
&,
{^
w
Td
cu
fc
cu
CD
,£>
O





CO
4-1
r— 1
3
CO
d)
pzj















^Sj
Td
jd
4-1
Crt



X"N
i— 1

O
4-J

CU
3
Td

^1
•U
•H
, — |
cd
4-1
5-1
O
0
Td
cu
0

Td
cu


t**!
r-l
Td
cu

5-1

^


















CO
•rH
CO
3
rH
O
fi
0
O
c
M

•
r-l
cd

4->
CU

fi
0
cd

rH
CU
CD
CO
CU
•H
W
&
CU 3
r-l 4-1
rQ CO
•H
CO 14H
CD O
O
^vt cj
0
x"~* *H
CM 4-1
cd
Td fi
fi T-l
cd fi

~ CU
4-1 4-1
o cu
cu Td
"4-1
TU.S

5-i co
cu cd
^ -H

0
5 fi
o
r^*> *H
rfi 4J
4-1 CJ
rH CU
Cd rH
CU CU
rfi CO





























CU
ft
CD
•ri
rH
^
3
PH
fi


A
O
oo
O"\
T— t
X. X



cd
•H
^_i
O
4-1
•iH
Td
cu

rH
cd
O
•H
4J
•H
^_J
CJ

m
o

•3
td



•
fi
o
•H
4-J •
cd &
rH CU
3 iH
p. >
o cu
P. 5-1
















































4-1
fi

g
K^l
0
rH
p.
&
0)

M-l
0

fi
o
•H

cd

rj
Td
4-1

o

CO

£•*•»
rH
CU
>
•H
4-1
cd
rH
CU



















CU
>
CO
3
rH
O
fi
O
o
fi
M






•V
1— 1
oo
CTi
< — 1
V-X

C
nj
£
J-t
 cu
r-l <~{
x_/ CO
•H
Td rH
5-1 XI
cd 3
rH ft
•H fi
cd 3
4-1
CO
13-173

-------
reproduction study performed at Litton-Bionetics by Bellies et_ al_. (1980);




three studies by Bio/Dynamics Inc. (1980a, b, c) in which acrylonitrile was




administered 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




atmosphere 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




(Table 13-39):  five published [Monson (1978), 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 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




subgroups of the populations studied.  All four have problems with the




methodology, definition and/or size of the population, whether exposure to




other carcinogens occurred, and short follow-up intervals.  In three of the




four studies [Delzell and Monson (1982), Thiess et_ al. (1980), Werner and




Carter (1981)], the problems were sufficient to cast doubt on the finding of
                                        13-174

-------
 significantly elevated risks of lung cancer found in each study.   In the




 fourth study  by O'Berg,  the  problems were insufficient  to obscure the




 significant finding  of lung  cancer.   After adjusting for latent  factors  and




 evaluating the contribution  due to  smoking,  the  finding of a  statistically




 significantly 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




 significantly 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




 carcinogenicity 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."









 13.5.4.2  Quantitative Assessment




    Three  unit risk estimates for air are calculated; one based on a human




occupational study (O'Berg, 1980) and two based on rat cancer bioassays (Quast




et_ ai_., 1980a, b).   The upper-bound lifetime risk of cancer associated with a




lifetime inhalation exposure of 1 ug/m3 is 6.8 x 10~5 from the human study
                                        13-175

-------
and 1.5 x 10"-' from the rat inhalation study.  The value based on the rat




drinking water study is 1.5 x 10"^ (or 2.6 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




relatively 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/liter of acrylonitrile in drinking




water is 1.5 x 10" , based on the mean value of three drinking water studies




in male rats.  The estimate based on the mean of three female rat studies is




nearly the same.




    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.

-------
    The carcinogenic potency of acrylonitrile is in the third quartile among




54 compounds evaluated by the Carcinogen Assessment Group as suspect




carcinogens.




    Using the International Agency for Research on Cancer (IARC)




classification 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.
                                        13-177

-------
13.5.6  Appendix—Comparison of Results by Various Extrapolation Models




    The estimates of unit risk from animals presented in the body of this




document are all calculated by the use of the linearized multistage model.




The reasons for its use have been detailed therein.  Essentially, it is part




of a methodology that estimates a conservative linear slope at low extrapolation




doses and is consistent with the data at all dose levels of the experiment.




It is a nonthreshold model holding that the upper-limit of risk predicted by




a linear extrapolation to low levels of the dose-response relationship is the




most plausible upper-limit for the risk.



    Other models have also been used for risk extrapolation.  Three




nonthreshold models are presented here:  the one-hit, the log-Probit, and the




Weibull.  The one-hit model is characterized by a continuous downward curvature




but is linear at low doses.  It can be considered the linear form or first




stage of the multistage model because of its functional form.  Because of this




and its downward curvature, it will always yield estimates of low level risk




which are at least as large as those of the multistage model.  Further,




whenever the data can be fit adequately by the one-hit model, estimates from




the two procedures will be comparable.




    The other two models, the log-Probit and the Weibull, are often used  to




fit toxicological data in the observable range, because of the general "S"




curvature.  The low-dose upward curvatures of these two models usually yield




lower low-dose risk estimates than those of the one-hit or multistage models.




    The log-Probit model was originally proposed for use in problems of




biological assay such as the assessment of potency of toxicants  and drugs and




has usually been used to estimate such values as percentile lethal dose or




percentile effective dose.  Its development was strictly empirical, i.e., it
                                      13-178

-------
was observed that several log dose-response relationships followed the




cumulative normal probability distribution function.  In fitting the cancer




bioassay data, assuming an independent background, this becomes:






             P(D;a,b,c) = c + (1-c) $ (a+blog10 D)   a,b > 0 _< c < 1






where P is the proportion responding at dose D, c is an estimate of the




background rate, a is an estimate of the standardized mean of individual




tolerances, and b is an estimate of the log dose-Probit response slope.




     The one-hit model arises from the theory that a single molecule of a




carcinogen has a probability of transforming a single noncarcinogenic cell




into a carcinogenic one.  It has the probability distribution function:
                       P(D;a,b) = l-exp-(a+bd)   a,b > 0
where a and b are the parameter estimates.  The estimate a represents the




background or zero dose rate, and the parameter estimated by b represents




the linear component or slope of the dose-response model.  In discussing the



added risk over background, incorporation of Abbott's correction leads to






                          P(D;b) = l-exp-(bd)   b > 0






Finally, a model from the theory of carcinogenesis arises from the multihit




model applied to multiple target cells.  This model has been termed here the




Weibull model.  It is of the form






                        P(D;b,k) = l-exp-(bdk)   b,k > 0
                                      13-179

-------
For the power of dose only, the restriction k > 0 has been placed on this model.




When k > 1, this model yields low-dose estimates of risks usually significantly




lower than either the multistage or one-hit models, which are linear at low




doses.  All three of these models usually project risk estimates significantly




higher at the low exposure levels than those from the log-Probit.




    The estimates of added risk for low doses for the above models are given




below for the various acrylonitrile data sets.  Both maximum likelihood




estimates and 95% upper confidence limits are presented.  All estimates




incorporate Abbott's correction for independent background rate.




    The results (Tables 13-40, 13-41, 13-42, and 13-43) show that the maximum




likelihood estimates of risk for the log-Probit model are all less than those




for the other models.  The one-hit model yields estimates identical with those




of the multistage in six of the eight data sets and higher in the other two.




The Weibull model yields risk estimates higher than those of the one-hit for




five of the eight data sets.  Based on past experience, this last result is




unlikely.  The Carcinogen Assessment Group feels that estimates based on




the linearized multistage model represent the plausible upper-limits of risk.
                                      13-180

-------
CO
1-3
M
Q
O
g

co H
H 3

Pi pi
w
>-l ptl

t£ H

S I-H
pq (^j
fV] JVJ
Q
Q

^ O
p- >
W EH
lJ CO

g^ i^i
o

EH H
0 — i cu

fV] 4_j
O CD
cd
W 3
W O-




















MH
O

CO
CU
4-1
cd
e
•H
4-1
CO
W

Tr3
o
o
r0
•1-1
T-l

-------



CO
rJ
H§
 (1)
J2J J_( TQ
^H &i 0
Id -H
Id Q

S Q m
tt H P
H CO 0
O -H

p 3J £
fy. Q
O O
5*5 55
CO M CO

P5 53 •*•*

Cd pi 0
CO S rQ
Q ,0

? 'id
ts o CD
O CO 4J
rJ OS Cd

to *~s o
O 0.
• J-l
CO O O
Cd 55 t)
r j i i r*
<« *H
S co

H M S

cd 55 cd
r?- a
£^ *iH
• Q T '
rH ^ M
-* O tt)
1 M
CO M rH

Id <^

i E_$
pa
H 0
cd
to

















M-J
0
.,_,
•H
•r) CO
CO
tt) -i-j
y (2
P
0) rH
13 cd
•H C
MH O
P -H
O 4-1
U -H
J-l 'Tj
a) — 1
^
as
















tf-i
0

CO
0)
4-)
cd
s co
•H _sj
4J CO
CO *H
Id prf

TJ rH
O Cd
O P
J2 O

rH 4J
tt) iH
,vj »i3
•H Tj
i-4  O
t »«r«
T— 1 i* •
3
2




0) i-l
CO — .
O 60
Q 3

CO
•—i
b
!•!
CN

in



m
b
rH
CO
CO











VO
b
I— 1
t— 1
1— 1





in
•— i
1
b

X
r^*
•
rH




OO
I
O
1 — 1
X

9
cs



r-^
I
b
i— i
X
-l
i-l •
cd o
S


CT*
0
t— i
in

m



m
b
i — i
•— i
in











in
b
i—*
X
rH
i — 1





rH
i-H
|
O
i-H
X

•
CvJ




J-s^,
1
O
•— I

F-^
e

-------
CO
1-4
W
O
O
g
EH

W
co pi
H w
 Pr . Cj
1 3
W S 0
£3 O r-l
Q Pi 60
i^C ,NTt ,M
Pi o
PM Q cd
CO W r°
^>
W M 4-1

-l 0)
Q a.
Q C3 CU

<3 co d
•1-1
w pi
r-3 w r-i
S EH O
^r* ^ti 'I i
£3
|Tj pd
EH Q5 O
O !3 -H
CQ M 4-1
M O
S S cu
O M S-(
pi Pi M
Cl4 Q O
o
co cd co
M O -
Pi oo 4-1

pc] rH O
CO .0
O ~rO
Q • *^
1 iHl
ts cd cu
O 4-1
,j 4-i cd
cu| J-i
pg-f Q
o 4J a.
CO VI
co cd O
W 3 o
EH cy 0
^yj
i— l >-i co
H g CU

p£J P_( Cd
S £
O -H
• O 4-J
CM CO

CO O i-l
r-l M i-l .
w s 5

i— 1 iTl
PQ U
3
H [3
0


W
od
H
Q
Pi
fe


















14-1
o
4-1
•H
£3
•H CO

CO
CU -H
o pi
0
OJ t-H
nzJ cfl
*n1 Cj
HH 0
O -H
O 4-1
CJ -i-l
13

CU i-l
W Pi

"3 iH
o cd
0 S
J"j Q
•H -H
rH 4J
CU -H
^ *T3
^^

ri
3
S
•H
X
cd
j£Tj





















4-1
•H

0 i-l
jj cu
' T 0
60 S
o





.-1
r-l
3 H
JO CU
•H T3
CU O





4_)
•H

1 CU
CU TJ
a o
0 g

cu
60
cd
4-1
CO i-l
•H CU
4-1 13
r-l 0
3 S






4-1
•H
,0
O i— 1

fS.[ r^
1 0
60S
O
1 ~\




i-H
i-l
3 i-H
f CU
•rl 13
cu o





4-1
•H
Cd 1-1
1 CU
CU T3
CJ O
O iS<


cu
60
cd
4-1
CO i— 1
•H CU
4-1 13
i-H 0

S





0) i-H
CO »-.
0 60
0 3










0






in
l
0
x
1 — t









vO
1
0
i— i

OO
4
CM















O








vO
|
0
1 — 1
CO
•
CM



vO
I
0
1— 1
CO
•
CM




vO
1
0
I— 1

CO

CM

i— 1
60
3
CO
CU —1
1— 1 •
cd O
S







vO
CM
O
i — l
O
•
t— 1




|
0
1 — 1
CO
•
CM




•o-
1
o
1 — 1
X
CO

CM


i-H CO
-*. CU
60 i-H
3 cd
£j
Cs cu
i-H PL*







t-^ OO
CM ^H
0 0
.— i i-H
in CM
• •
T— 1 •— 1



*^ <1"
1 t
O 0
1 — 1 1 — I
x x
t-H in
• •
i-H VD









vo m
1 1
0 0
.— 1 .-H
X X
^•O vO
• •
CM CM















0 O








in 
-------

CO
.J
co la
H 0
*^C f"^ ^*\
<=2 2 •
CO
^•f £— f 4-J
w is cd
5 §
^^ ^O *^J
O Cu C
1 to 3
[a M O
O 60
< OS ^i
os a o
PU o «
CO tr, XI

w 2 -w
.-3 o e
^c oS cu
2 Cl4 T3
Cd C
fa O 0)
d3 O*
Q ^ O
j2j l_ 1 *TJ
<3 p£j pj
W -H
W Q
(J M
2 Q >*-<

S3 H C
H CO O
SOSS
Ci3 O
2 R CU
O — 3 ^
CO
CU -rl

C
CU rH


M-l O
C -rl
O 1 '
O -H

f-i T3
CU <3
0.
cx
£3

5^2
in
CT.

















t[ ]
O
CO
cu
4-1
cd
e co
•rl J«!
4-1 CO
CO -rl
Pd pS

13 rH
o cd
0 C
XI O
•rl -H
rH 4-1
CO -rl
31
£•
3

•H















4-1
•rl
o
0 rH
M CU
p j nj
1 0
602
o





rH
rH
3 rH
XI CU
•rl 13
CU O
Cs 2




4J
•rH
S3 rH
1 cu
CU TJ
C 0

^^.
cu
60
cd
CO rH
•rl CO
4J *XJ
3 2
2





4J
•rl
O rH
!J CU

1 0
602
O
.J




rH
rH
3 rH
XI CO
•rl *rj
CU O
ts 2


4-1
•rl
S3 rH
1 CU
CO 13
d o
o 2


co
60
cd
4J
CO rH
•rl CO
4-1 "TJ
rH 0
3 2
2




CO rH
CO ~~-
0 60
Q 3

O
i — i
t
O
i— I

cr,
a
CO



in
1
0
i — [

o

|«s.










\*D
|
O
«— i
X
1 — 1

1— (









T— J
|
O
r— i

p^.
•




m
1
o
i — i

vO
0
i — r


">
O
t — i
O
cr.



1
0
1— 1
X
o

ON
rH
60
3
CO
CU rH
rH

2


r--
1
0

X
CN
9
CN



^-»-
|
O
1 — I

vO
9
CO










m
1
0
1 — 1
1— 1

l 	 ,









OO
1
o
r~{

I — 1
9
CN



m
1
o
i — i
X
CO

<*


VD
O
i — |
O
CT»



|
O
i— 1
X
o
*
<^

i-H
•^^
AO
*y

,-lj



LO
|
O
T— i

VO
p
CO



CO
1
0
t — 1
JxJ
CO

»— t










-3"
|
O
1 — 1
M
r-H

i— |









vO
•
o-
I—I
X
r— 1
^
in



^
I
0
'x
m

in


m
1
o
! 	 1
X
o
O"\



m
1
o
i — i
X
o

°"

rH
^^
00


o
1 — 1
                       CO
                       CU
                      rH
                       cd

                       §
                              X
                              CN
       in

        O

        "x
        co
                             m     CN
                             o
                                    X
                                    vo
 o-

  o
  I—t
   X
  CN

  m
                                         co

                                          o
                                          CO
                            r--.

                             o
                             CN


                             OO
                                    X
                                   CN
             in

              o
              !—I
              X
              CM
                                   OO     OO
                            00

                             o
                             1—I
                             X
                             CO
                            in

                             o
                             I—t
                             X
                             CN
       'o    'o
       I—I    .—I
        bd     bd
       (Ti    CO


       CO    rH
            CO


             O
                             X
                             CO
                            o

                            "x
                            CO
        X
       CM
       O

       "x
       CO
      vO

       o
       1—I
       X
       CO
  X
 CO
m

 o

 "x
 co
in

 o

 X
 CO
 60   rH

 3    ^~-
       60
rH     3


O    rH
 60
 3
13-184

-------
                                  REFERENCES
Aarato, S.E., and Bittera, E.  (1972).  Determination of acrylonitrile in the air
     in the presence of dimethylformamide and  ammonia.   Munkavedelen  18;50-51.
     (In Hung.) Taken from:  Chem. Abst. 79_:57280b, 1973.             ~

Abreu, M.E.,  and Ahmed, A.E. (1980). Metabolism of acrylonitrile to cyanide:  Iri
     vitro studies.  Drug Metab. Disp.  8^:376-379.

A.D. Little, Inc.  (1974).  A Model Economic and Safety Analysis of the  Trans-
     portation of Hazardous Substances in Bulk, Final Report (NTIS COM-74-11271)
     264 pp.  (Cited in Miller and Villaume, 1978.)

Ahmed, A.E.,  and Patel, K.  (1979).  Pharmacokinetics, distribution and binding of
     l-l4C-acrylonitrile in rats.  Pharmacologist ^1(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.  97.:l8598p.

Ahmed, A.E.;  Farooqui,  M.Y.H.; Upreti, R.K.;  and El-Shabrawy, 0. (1982).  Distri-
     bution  and covalent  interactions  of  [1-14C]  acrylonitrile  in  the  rat.
     Toxicology 23:159-175.  Taken from:  Chem. Abst.  97;67436t.

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.

Axelson, 0.   (1978).  Aspects on confounding in  occupational  health and  epide-
     miology.  Scand. J.  Work Environ. Health.   4:  85-89.
            <6>
Babanov,  G.P.  (I960).   Problems  of  industrial  hygiene  in  the production  of
     synthetic butadiene-acrylonitrile rubbers.  Gigiena Truda i  Professional.
     Zabolevaniya M12):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. Zabol. .3:55-7.   (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 .17 (1-2): 37-46.   (In Slo.)   Taken from:   Chem.  Abst. 88:44531v,
     1978.                                                          —
                                      R-1

-------
Bandt, H.J. (1953).
     Ger.)
[Acrylonitrile as  an effluent.]  Z. Fisch. ^:457-46l.  (In
Barrett, W.J.  (1974).   Sampling and Analysis  of Four Organic Compounds Using
     Solid Sorbents.  Southern Research Institute Report, SORI-EAS-74-301, 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-
     nitrile.  Ann.  Occup. Hyg. _22:429-435.

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 24,  1980,  including minutes  of the meeting  held  to Review
     Recently Completed Studies on  the Epidemiology of Acrylonitrile, held in
     Brussels,  April 23 and 24, 1980.

Beaumont, G.P., and Garrido, C.H.  (1979).  Respirator cartridge  test system and
     test-results for  benzene and  acrylonitrile.   Amer.  Ind.  Hyg.  Assoc. J.
     40(10):883-87.

Bellies, R.P.;  Paulin, H.J.; Makris, N.G.; and Weir, R.J. (1980).  Three-Genera-
     tion Reproductive Study of Rats Receiving Acrylonitrile  in  Drinking Water.
     Prepared by Litton Bionetics, 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. Epidemiol. Microbiol. Immunol. 3>:106-ll6.   (In Ger.)

Benes,  V.,  and  Sram,  R.  (1969).   Mutagenic  activity of  some  pesticides in
     Drosophila melanogaster.  Ind. Med. Surg. 38(12):442-444.

Berck,  B.  (I960).   Fumigant residues:   Retention  of acrylonitrile and  carbon
     tetrachloride by shelled walnuts fumigated with Acrylon.  J. Agr.  Food Chem.
     8^(2): 128-31.

Berck,  B.  (1962).    Polarographic  determination  of  methyl  bromide,  ethylene
     dibromide,  acrylonitrile,  chloropicrin,  and carbon tetrachloride in  air.
     J. Agr. Food Chem. 3.0:158-62.

Berck,  B.  (1965).    Determination  of  fumigant   gases  by gas  chromatography.
     J. Agr. Food Chem. 113(4):373-77.

Berck,  B.  (1975).   Analysis  of fumigants and fumigant residues.   J.  Chromatogr.
     Sci. 13(6):256-67.

Bio/dynamics Inc.  (1980a).   A Twenty-Four  Month Oral Tqxicity/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-
     1745 for Monsanto Company, St. Louis, MO.
                                      R-2

-------
Bio/dynamics Inc.  (1980b).   A Twenty-Four Month Oral  Toxicity/Carcinogenicity
     Study of Acrylonitrile Administered  in  the Drinking Water to Fischer  344
     Rats, Vols. 1-4 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.

Bio/dynamics Inc.  (1980c).   A  Twenty-Four Month Oral  Toxicity/Carcinogenicity
     Study of Acrylonitrile 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. 77-1746  for
     Monsanto Company, St. Louis, MO.

Bocek,  K.  (1976).    Relationships  among  activity  coefficients,   partition
     coefficients and solubilities.   Experientia, Suppl.  23:231-40.

Bogaczek, J., and Jaworski, M.  (1970).  Continuous polarographic  determination
     of acrylonitrile. Fresenius1 Z. Anal. Chem. 251(3):174-76.  (In Ger.)  Taken
     from:  Chem. Abst. 73_:94406u,  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 thiocyanate. Arch. Biochem. 39:7-26.

Boyland,  E.,  and  Chasseaud,  L.F.  (1967).    Enzyme-catalyzed conjugations  of
     glutathione with unsaturated compounds.   Biochem.  J.  104:95-102.

Braun,  D., and  Vorendohre,  G. (1963).   Thin-layer  chromatography  of some
     unsaturated polymerizable  compounds.   Z.  Anal. Chem.  199(1);37-41.   (In
     Ger.) Taken from:  Chem. Abst.  60:4253d, 1964.

Brieger, H.; Rieders, F.; and Hodes, 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-
     containing  polymers  and  food  simulating solvents.   J. Assoc.  Off.  Anal.
     Chem. 61(6):1383-388.

Burg, S.P., and  Burg, E.A. (1967).   Molecular  requirements for  the  biological
     activity of ethylene.  Plant Physiol. 42:144-152.
                                      R-3

-------
Burkhart, P.; Bartels, U.;  and Hoyme,  H.  (1961).   Analytical  chemistry of  poly-
     acrylonitrile fibers.   I.   Analytical  studies on the separation  of  vinyl
     monomer mixtures.   Faserforsch.  Textiltech.  1.2:581-87.   (In Ger.)  Taken
     from:  Chem. Abst. j57_:1107i, 1962.

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. _4_0(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 p.

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. 31:343-346.

Chaigneau, M., and LeMoan, G.  (1974).   Pyrolysis of plastic materials.   VIII.
     Polyacrylonitrile and copolymers.   Ann.  Pharm. Fr.  32.(9-10):485-90.   (In
     Fr.) Taken from:  Chem.  Abst.  8_2:736l4f,  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.  £>5_:10669g,  1966.

Chekhovskaya, E.V.;  Gorban, N.S.; Yakovleva, L.I.; Nazarenko,  G.P.jand  Nechaeva,
     A.V. (1966).  Data for experimental studies  of toxicity of wash waters from
     acrylonitrile  production.   Vodosnabzh.  Kanaliz.    Gidrotekh.    Sooruzh.
     Mezhued. Resp. Nauch.  Sb. No.  JL:83-88.  (In  Russ.) Taken from: Chem.  Abst.
     66:88487k, 1967.
Chemical  Marketing  Reporter   (1980).
     February 4, 1980, p. 9.
Chemical  profile:
Acrylonitrile.
Cherry, A.B.; Gabaccia, A.J.; and Senn, H.W. (1956).  The assimilation behavior
     of certain toxic  organic  compounds  in natural water.  Sewage  Ind.  Wastes
     ^.(9): 1137-46.

Chiou, C.T.; Freed, V.H.;  Schmedding, D.W.;  and Kohnert, R.L. (1977).  Partition
     coefficient and bioaccumulation of  selected organic chemicals.   Environ.
     Sci. Technol. 11:475-478.

Chopra, S.K.; Vishnoi,  S.C.;  Kapoor, V.B.; and Gupta, P.L. (1978).  A gas chroma-
     tographic method for  the estimation of  residual monomers in styrene-acrylo-
     nitrile  copolymer.    Res. Ind. _23(1):27-30.   Taken  from:   Chem.  Abst.
     89_: 16409 7a, 1978.

Chudy, J.C., and Crosby, N.T. (1977).   Some  observations on the determination of
     monomer residues in foods.  Food Cosmet. Toxicol. 15(6);547-51.

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  Salmonella
     typhimurium.  Environ. Mutag. 1:269-276.
                                      R-4

-------
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. ^3_:28605g,  1970.

Cox, C.R.   1972.   Regression model and  life tables.  J.  Roy.  Stat. Soc. B.
     34:157-167.

Grume,  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.   (1979).   Dose-response  problems  in carcinogenesis .
     35:157-167.
                                                                    Biometrics.
Crump, K.S. and Watson, W.W.   (1979).   GLOBAL 79.   A fortran program to extra-
     polate dichotomous animal  carcinogenicity data to low dose.  Natl.  Inst.  of
     Environ.  Health Science Contract No. 1-ES-2123.

Crump, K.S.; Guess, H.A.;  and Deal, L.L.   (1977).  Confidence intervals and test
     of  hypotheses  concerning dose-response  relations inferred  from  animal
     carcinogenicity data.  Biometrics.  33:437-451.

Dahle, L.; Hill, E.; 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).   Identification of  metabolites  of acrylonitrile.   Testimony
     before FDA.   (Cited in Miller and Villaume, 1978.)

Daues, G.W., and Hamner, W.F. (1957).  Determination of  small amounts of acrylo-
     nitrile in aqueous industrial streams.  Anal. Chem. 29.: 1035-37.

Daugherty, P.M., Jr., and Garrett, J.T. (1951).  Toxicity levels of hydrocyanic
     acid and some  industrial  by-products.  Texas J. Sci. ^:391-396.

Davis, J.H.; Davies, J.E.; Raffonelli,  A.; and Reich, G. (1973).  The investiga-
     tion of fatal acrylonitrile intoxications.  In:  Pesticides and the Environ-
     ment:  A Continuing Controversy.  Intercontinental  Medical Book Corp., Vol.
     2, pp. 547-556.

Delzell, E. and Monson, R.R.   (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 mutagenicity
     of acrylonitrile.  Toxicology 3.3(1) : 7-15.

DeMeester,  C.;  Poncelet,  F.;  Roberfroid, M.;  and  Mercier,  M.  (1978).   Muta-
     genicity of  acrylonitrile.  Toxicology 11:19-27.
                                      R-5

-------
Deur-Siftar, D., and Svob, V. (1976).  Procedure for gas chromatographic deter-
     mination of acrylonitrile as a  residual monomer  in polymerization  process
     waste  waters  and in polymers.   Kern.  Ind. _25(12):703-08.   (In Croatian.)
     Taken from:  Chem. Abst. 8_7:58072n, 1977.

DiGeronimo, M.J.,  and Antoine, A.D. (1976).   Metabolism of acetonitrile  and
     propionitrile by Nocardia rhodochrous LL100-21.  Appl. Environ. Microbiol.
     Jl(6):900-906.

Dilling, W.L. (1977).  Interphase  transfer  processes.  II.  Evaporation rates of
     chloromethanes, ethanes,  ethylenes,  propanes,  and propylenes  from dilute
     aqueous solutions.  Comparison with theoretical predictions.  Environ. Sci.
     Technol. 11(4):405-09.

DiLorenzo, A., and Russo, G.  (1969).  Gas chromatographic analysis of mixtures of
     acrylonitrile with  acetonitrile and water.  Chim.  Ind. j51( 2): 170-71.   (In
     Ital.) Taken from:  Chem. Abst.  £0:120896p, 1969.

Dinu, V.  (1975a).   Activity of  glutathione  peroxidase  and  catalase  and  the
     concentration of lipid peroxides in acute intoxication with acrylonitrile.
     Rev. Roum. Biochim. 12(1);11-14.

Dinu, V. (1975b).  Intracellular thiol concentration in acrylonitrile intoxica-
     tion.  Rev. Roum. Biochim. 3.2(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. 16J3(1):133-40.

Doll, R.   (1971).  Weibull distribution of cancer.   Implications for models of
     carcinogenesis.  J. Roy. Stat. Soc. A.  13:133-166.

Doll, R. and Hill, A.B.   (1952).  A  study of the aetiology of carcinoma of  the
     lung.  Br. Med. J.  2:1271-1286.
Dripps,  R.D.;  Eckenhoff,  J.E.;  and  Vandam, L.D.
     anesthesia,  the  principles of  safe practice.
     Company, Phil. Pa.  p. 121-123.
(1977).    Introduction  to
 5th Ed.   W.B.  Saunders
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  Quality
     Criteria Document for Ammonia.  Prepared  for the Environmental  Criteria and
     Assessment Office, U.S. Environmental Protection Agency,  Cincinnati,  OH.

Ehrenberg, L.  and  Ostermar-Golkar.   (1980).  Alkylation of macromolecules  for
     detecting mutagenic agents.   Terato. Carcinog. and Mutag.   1:105-127.
                                      R-6

-------
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.
     Altman and Dorothy S. Dittmen.  Federation of American Societies  for Experi-
     mental Biology.  Bethesda, MD. Library of Congress  No.  72-87738.

Fasitta, V.A.,  and Licciardello,  G. (1977).   Determination of  acetonitrile and
     acrylonitrile in the air by gas chromatography.  Ann. 1st. Super. Sanita.
     13(1-2):245-48.  (In Ital.)   Taken from:  Chem.  Abst.  89_:11151p,  1978.

Fassett, D.W. (1963).  Cyanides and nitriles.  In:  Toxicology. Fassett, D.W.,
     and Irish,  D.D., editors.  Vol. 2 of Industrial Hygiene and Toxicology, 2nd
     ed.  New York:  John Wiley and Sons,  Inc., pp.  2009-2012.

Federal Register  (1978a).  Emergency temporary standard  for  occupational expo-
     sure   to    acrylonitrile    (vinyl   cyanide):       Notice  of    hearing.
     _43(11):2586-2607.

Federal  Register  (1978b).    Occupational  exposure  to  acrylonitrile  (vinyl
     cyanide).   43(192):45762-45819.
Federal   Register   (1979).
     44(233):69479-84.
Method   603,   acrolein   and   acrylonitrile.
Finkel, J.M.;  James,  R.H.; and Miller, H.C. (1979).  Residual Monomers in Acrylic
     and Modacrylic  Fibers  and  Fabrics.    Draft  Final  Report.    Prepared by
     Southern Research Institute, Birmingham, AL, under Contract No.  68-01-4746.
     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, N.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.

Gagnon, Y.T.,  and Posner,  J.C.  (1979). Recovery of acrylonitrile from charcoal
     tubes at low levels.  Amer.  Ind.  Hyg. Assoc. J.  40(10);923-5.

Gaffey, W.R. and Strauss, M.E.  (1981).  A mortality study  of  workers  potentially
     exposed   to   acrylonitrile   during   startup,   Monsanto   Decatur  plant.
     Unpublished report.
                                      R-7

-------
Gale, D.M.  (1970).   Mechanistic aspects  of  the photochemistry of  unsaturated
     nitriles.  J. Org. Chem. 35(4):970-72.

Gandini, A.,  and Hackett, P.A.  (1978).   The  photochemistry of  acrylonitrile
     vapour at 213.9 run.  Can.  J. Chem.  56 (15): 2096-98 .
Garny, A.L.; Poncelelet, F.;  and Mercier,  M. (1981).
     nitrile in mouse.  Toxicol. Lett. 7:329-334.
Mutagenicity of acrylo-
Garrison, P.P. (1978).  Preliminary studies on the effects of acrylonitrile  and
     potassium thiocyanate  on the  seagrass  Ruppia maritima  potamogetunaceae.
     ASB (Assoc. Southeast Biol.)  Bull. 25(2);68.

Gawell,  G.B.M.   (1979).   Determination of  acrylonitrile monomer  in  plastic
     packaging  and beverages  by headspace  gas  chromatography.   Analyst  104
     (1235):106-10.

Giacin, K.; Gilbert, S.; and Antoine, A.D. (1973).  Utilization of acrylonitrile-
     based polymers and related monomers by soil  fungi.   Abstracts  of  the 73rd
     Annual  Meeting of  the American  Society  of Microbiologists,  Miami,  FL,
     p. 170.  (Abstract No. P175.)

Gisclard, J.B.; Robinson, D.B.; and Kuczo,  P.J. (1958).  Rapid empirical proce-
     dure  for determination of acrylonitrile  and acrylic esters in the  atmo-
     sphere.  Amer. Ind. Hyg. Assoc. J. 19;43-45.

Glass, E.H., and Crosier,  W.F.  (1949).   The viability of seeds fumigated with an
     an  acrylonitrile-carbon   tetrachloride   mixture.    J.   Boon.   Entomol.
     jJ2(4):646-649.

Going,  J.E.; Kuykendahl,  P.; Long,  S.;  Onstot, J.; and  Thomas,  K.  (1979).
     Environmental Monitoring Near Industrial Sites:  Acrylonitrile,  Publication
     No. EPA-560/6-79-003.  Prepared by Midwest Research Institute, Kansas City,
     MO,  under  Contract  No.  68-01-4115,  Final  Report.   U.S.  Environmental
     Protection Agency, Office of Toxic Substances, Washington, DC.

Gorokhovskaya, A.S.,  and  Geller,  B.E.   (1962).  Polarographic  determination of
     acrylonitrile.   Zavodsk.  Lab.  _28:809-11.  (In Russ.)   Taken from:  Chem.
     Abst. 58:11960b, 1963.

Grasselli, J.G., and Ritchey, W.M., editors (1975).  CRC Atlas of Spectral Data
     and  Physical  Constants  for Organic  Compounds,  2nd ed.    Cleveland,  OH:
     Chemical Rubber Co., Vol. 2, p. 102.

Green, M.H.L.; Bridges, B.A.; Rogers, A.M.; Horspool,  G.; Muriel, W.J.;  Bridges,
     J.W.; and Fry, J.R.  (1977).  Mutagen screening  by  a simplified  bacterial
     fluctuation test:  Use of microsomal preparations and whole  liver cells  for
     metabolic activation.  Mutat. Res. _4:28 7-29 4.

Green, M.H.L., and  Muriel, W.J. (1976).  Mutagen testing  using TRP+ reversion in
     Escherichia coli.  Mutat. Res. 38:3-32.
                                      R-8

-------
Groet, L.T.  (1978).   Acrylonitrile.   In:   Kirk-Othmer Encyclopedia of Chemical
     Technologyj 3rd ed., Grayson, M., and Eckroth, D.,  editors.  New York:  John
     j. ^%-.i.j.iiwa.v^
-------
Hoffman, P.; Kleine, D.; and Franzen, E.  (1976).
     exposure:  Detoxification of  acrylonitrile
     acid.]
                                    [On the proof of acrylonitrile
                                    by coupling with D-glucuronic
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
     Villaume, 1978.)
Hosaka,  S.,  and  Wakamatsu,  S.  (1968).
     Tetrahedron Lett. 2:219-220.
                              Photodimerization of  acrylonitrile.
Howie,  S.J.  (PEDCO).    (1982).    Ambient  acrylonitrile  levels  near major
     acrylonitrile production and use of facilities.  Final Report.  U.S. EPA,
     U.S. EPA/RTP, No.

Hovious, J.C.; Waggy, G.T.;  and Conway, R.A.  (1973).  Identification and control
     of petrochemical pollutants inhibitory  to anaerobic processes.  Available
     from:   National Technical Information Service, Springfield, VA (NTIS  PB-
     222-287).  (Cited in Miller and Villaume,  1978.)

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.  68-02-1874.  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).

IARC  (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,
     IARC, Vol. 19, pp. 73-113.

ICRP  (International  Commission on  Radiological Protection)   (1977).  Recommen-
     dation of the International Commission on Radiological Protection,  Pub.  No.
     26, adopted Jan. 17, 1977.  Pergamon Press,  Oxford, England.

Ivanenko,  G.S.,  and  Lukashevskaya,  A.V.  (1976).    Chromatographic  method  of
     determining bound acrylonitrile in butadiene-nitrile rubbers.  Metody Anal.
     Kontrolya Proizvod. Khim. Prom-sti. ^:19-20.  (InRuss.)  Taken from:  Chem.
     Abst. 84:l8l403m, 1976.

Jackson, S.,  and  Brown,  V.M.  (1970).   Effect of  toxic wastes on treatment pro-
     cesses  and  watercourses.  Water Pollut. Contr. 69_(3): 29 2-313.   (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. 39(2);l6l-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
                                      R-10

-------
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  aegypti.    J.  Econ.
     Entomol. _55(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. 3.1(5): 161-64.   (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.

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.; Rostov, G.; Milina,  R.; and Mikhailov, M.Kh. (1974).  Determina-
     tion of the composition of ternary copolymers by IR spectroscopy  and pyro-
     lytic  gas chromatography«    Vysokomol.  Soedin., Ser.  A.,  16(9):2162-165.
      (In Russ.) Taken from:  Chem. Abst.  8_2:32109b, 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. 85:67783?, 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-C-77-0009.   U.S.  Consumer Product Safety  Commission,
     Washington, DC.

Keresztesy,  J.; Taylor, T.J.; Katz, A.;  Haddock,  J.T.; Byer, W.L.;  and Malseed,
     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.   (1982a).        	
     Monsanto  Company,  St.  Louis.
     800282-ML-80-441.
Ames/Salmonella mutagenicity  assay of  acrylonitrile.
             Report  No:   MSL-2063,  Job/Project  No:
 Kier,  L.D.    (1982b).    Ames/Salmonella mutagenicity  assay of  acrylonitrile
      epoxide.  Monsanto Company, St. Louis.  Report No: MSL-2064, Job/Project No:
      800283-ML-80-442.

 Kiesselbach, N.; Korallis, U.; Lange, H.J.;  Neiss, A.; and Zwingers, T.  (1980).
      Bayer-ACN-Studie  1977.   Zentralblatt  fur  arbeitsmedizin,  arbeitsschutz,
      prophylaxe und  ergonome, Band 7-
                                      R-11

-------
Kihlman,  B.A.  (1961).   Cytological effects  of phenylnitrosamines.   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 priority
     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. J75:1304lla, 1971.

Klyaev, V.I.;  Slisarenko,  F.A.; and Finkel'shtein,  A.V.  (1966).  Polarographic
     determination of acrylonitrile.  Fiz.-Khim. Issled. Prir.  Sorbentdv Ryada
     Anal.  Sist.  1.: 27-33.  (In Russ.) Taken from:  Chem. Abst. 6£:64465n, 1968.

Knoblock, K.; Szendzikowski, S.; Czajkowska, T.; and Krysiak,  B. (1971).   Acute
     and subacute toxicity of acrylonitrile.  Med.  Pracy 22(3):257-69.

Kobayashi, Y. (1956). Determination of acrylonitrile vapor by means of detector
     tubes.  Yuki Gosei  Kagaku Kyokai Shi 14:673-75, 1956.   (In Japan.) Taken
     from:  Chem. Abst. 51:72401,  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.
     75:129033r, 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 synthe-
     sis 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. 51:7246e, 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. Tekhnol.
     .20(11) :16 38-640.  (In Russ.)  Taken from:  Chem. Abst. 88:63091t, 1978.
                                      R-12

-------
 Lambotte-Vandepaer,  M.;  Duverger-Bogaert, M.;  de Meester, C.; Poncelet, F.; and
      Mercierj  M.  (1980).  Mutagenicity  of urine from  rats and mice treated with
      acrylonitrile.   Toxicology J6.-67-71.

 Langvardt,  P.W.;  Putzing, C.L.; Young,  J.D.; and Braun, W.H. (1979).  Isolation
      and identification  of  urinary metabolites of vinyl-type compounds:  Appli-
      cation 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 cyanide in industrial
      effluents.   Barwniki-Srodki Promocnicze  _2_H4):87-91.    (In Pol.)   Taken
      from:   Chem. Abst. 89_:2203l8n,  1978.

 LeBlanc,  G.A.  (1980).   Acute  toxicity of priority  pollutants to  water  flea
      (Daphnia  magna).  Bull. Environ. Contain. Toxicol. 24:684-91.

 Lee,  W.R.,  Abrahamson,  S.,  Valencia,  R., Von Halle, E.S., Wurgler, F.E.  and
      Zimmering, S.   In press.   The sex-linked recessive lethal  test  for  muta-
      genesis in Drosophila melanogaster.  Mutat. Res.

 Leo,  A.;  Hansch,  C.;  and Elkins,  D.  (1971).   Partition coefficients  and  their
      uses.  Chem. Rev. 7J(6);525-616.

 Leonard,  A.; Garny,  F.;  Poncelet, F;  and Mercier, M.   (1981).   Mutagenicity of
      Acrylonitrile in Mouse.  Toxicol. Lett.  7^:329-334.

 Levin, D.E.,  Hollenstein,  M.,  Christian, M.F.,  Schwiers,  E.A.  and  Ames,  B.N.
      (1982).  A new Salmonella tester strain (TA102) with A:T  base pairs at the
      stie of mutation detect oxidation  mutagens.  Proc. Natl. Acad.  Sci.   USA.
      79:  7445-7449.

 Lezovic, A., and Singliar, M.  (1977).  Polarographic determination of  acrylamide
      and  acrylonitrile.   Petrochemia 17(5-6):128-32.    (In  Slo.)  Taken from:
      Chem. Abst.  89_:l6338q,  1978.

Lindgren, D.L.; Vincent,  L.E.;  and Krohne, H.E. (1954).  Relative effectiveness
      of  ten  fumigants to adults  of eight  species  of stored-product insects.
     J. Econ. Entomol. 47:923-26.

Linetskii, V.A.,  and Serebryakov,  B.R. (1965).   The  alkaline  hydrolysis of
     nitriles.   Khim. Prom.  4_1_(7):498-99.

Loveless, A. (1951).   Qualitative aspects of the chemistry and  biology of radio-
     mimetic (mutagenic)  substances.   Nature  (London)  167:338-42.
                                     R-13

-------
Loveless, L.E.;  Spoerl,  E.;  and Weisman, T.H.  (1954).   A  survey of effects of
     chemicals   on   division   and  growth  of  yeast  and   Escheriohia  coli.
     J. Bacteriol. j6£:637-44.

Ludzack, F.J.; Schaffer, R.B.; Bloomhuff, R.N.;  and Ettinger, M.B. (1958).  Bio-
     chemical  oxidation  of some  commercially important  organic cyanides.   I.
     River  water  oxidations.    In:     Proceedings  of  the  Industrial  Waste
     Conference, pp. 297-312.

Ludzack, F.J.; Schaffer, R.B.; and Bloomhuff, R.N. (1961).   Experimental treat-
     ment  of organic  cyanides by  conventional processes.   J. Water  Pollut.
     Contr. Fed. 33:492-505.
Lysyj,  I.  (I960).
     .32:771.
Gas-chromatographic  analysis  of  nitriles.    Anal.  Chem.
McCann,  J.;  Choi,  E.;  Yamasaki,  E.;  and Ames,  B.N.  (1975a).
     carcinogens  as mutagens in the Salmonella/microsome  test:
     chemicals.  Proc. Nat. Acad. Sci. USA 72:5135-5139.
                                             Detection of
                                             Assay of 300
McCann, J.; Spingarn, 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. 72(3):979-83.

McMahon, R.E.; Cline, 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, W.C.; Breder, C.; and Sphon, J.A. (1979).  Gas-solid chroma-
     tographic-mass spectrometric  confirmation of low levels  of  acrylonitrile
     after  distillation  from food-simulating solvents.   J.  Assoc. Off.  Anal.
     Chem. i62(l): 41-46.

McNeil, T.,  and Broader,  C.V.  (1981).  The determination of sub-ppb quantities of
     acrylonitrile in 3% acetic acid using manual headspace gas solid chromato-
     graphy.  U.S. Food and Drug Administration, draft procedure,  June, 1981.

McOmie, W.A.  (1943).   Comparative toxicity of  methacrylonitrile and  acrylo-
     nitrile.  J. Ind. Hyg. Toxicol. j}l: 113-116.

Magos, L. (1962).  A  study of  acrylonitrile  poisoning  in relation  to methaemo-
     globin-CN complex formation.  Brit. J. Ind. Med. 19;283-286.

Mallette, F.S. (1943). Industrial hygiene in synthetic rubber manufacture.  Ind.
     Med. 12(7):495-499.

Maltoni, C.; Ciliberti, A.; and DiMaio, V. (1977). Carcinogenicity bioassays on
     rats of  acrylonitrile administered by inhalation and by  ingestion.   Med.
     Lavoro 68_(6 ): 401-411.

Mantel, N. and Schneiderman, M.A.   (1977).  Estimating "safe" levels, a hazardous
     undertaking.  Cancer Res.  35: 1379-1386.
                                     R-14

-------
Marano,  R.S.;  Levine, S.P.; and  Harvey,  T.M. (1978).   Trace determination of
     subnanogram amounts of  acrylonitrile  in  complex matrices by gas ehromato-
     graphy with a nitrogen-selective detector.  Anal. Chem. J50(13):1948-50.

Marconi, S., and lonescu, M.  (1974).  Toxication of acrylonitrile on fish.  Stud.
     Prot. Calitatii Apelor  16:7-22.  (Cited in Miller and Villaume, 1978.)

Markelov, M.A.,  and  Semenenko,  E.I. (1976).   Determination  of microconcentra-
     tions  of  substances migrating  from polymer  materials  to water.   Plast.
     Massy 3.:57-59.   (In Russ.)  Taken from:  Chem. Abst. 8J»:155747g, 1976.

Markelov, M.E.; McMaster, A.; and Meyer,  N.  (1981).  Analysis and identification
     of  organic  impurities  in water by  a headspace  gas  chromatographic tech-
     nique, presented at the 8th Annual Meeting of the Federation of Analytical
     Chemistry and Applied Spectroscopy Societies,  Philadelphia,  PA,  September
     20-25, 1981.

Marrs, D.R.; Campbell, R.L.;  Boyko,  J.M.; Parrish,  A.L., Jr.; and Standish, N.W.
     (1978).  Measurement of acrylonitrile  in  ambient  air in the area of plastic
     processing equipment.  Paper presented at the 1978 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. (1978).  Acrylonitrile.  Job Safety Health 6/3):4-13.

Mekhtiev, S.I.; Mamedova, V.M.; Guseinov, A.G.; Abashin, V.G.; and Agaeva, N.E.
     (1968).   Polarographic determination  of acrylonitrile  and  methacryloni-
     trile.  Azerb.  Neft.  Khoz.  47(6):33-34.   (In Russ.)  Taken from:   Chem.
     Abst. 68_:113370p, 1968.

Miller, L.M.,  and  Villaume,  J.E.  (1978).  Investigation  of  Selected Potential
     Environmental Contaminants:   Acrylonitrile.   Publication No.  EPA-560/2-78-
     003.     Prepared  by   the   Franklin  Institute  Research   Laboratories,
     Philadelphia, PA, under Contract No. 68-01-3893.   U.S.  Environmental Pro-
     tection Agency,  Office of  Toxic  Substances,  Washington,  DC.   Available
     from:  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.  449-64.
Milvy, P., and Wolff, M. (1977).
     Res. 48(3-4):271-278.
Mutagenic studies with acrylonitrile.  Mutat.
Minami, M.; Sato, M.; Suzuki, A.;  Yamamura,  K.;  and Ishizu,  S.  (1973).   Studies
     on acrylonitrile poisoning.   1.  On the result of exposure  to acrylonitrile
     vapor.   Tokyo Joshi Ika  Diagaku Zasshi _4_3(10) :849-855.    (In Jap.; Eng.
     summary.)

Monson, R.R.  (1978).  Mortality and cancer morbidity among chemical workers with
     potential exposure to acrylonitrile.
                                     R-15

-------
 Morris, E.D., Jr.,  and  Niki,  H.  (1971).  Reactivity of hydroxyl radicals with
      olefins.  J. Phys.  Chem.  .75 (23): 36 40-41.

 Murray, F.J.; Nitschke,  J.A.;  John, F.A.; Smith, F.A.;  Quast,  J.F.; Blogg, C.D.;
      and Schwetz, B.A.  (1976). Teratologic  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:    CMA,
      Washington,  DC.

 Murray, F.J.;  Schwetz,  B.A.;   Nitschke,  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);547-552.

 Narasimhan, K.S.;  Shurpalekar, S.R.;  and Venkatesh, K.V.L.  (1972).  Use of fumi-
      gants  in  prevention of  insect  infestation  and mold  growth  in papads.
      J. Food Sci.  Technol. 9_(3) :134-37.

 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.)

 Nazarova,  V.I., and Nakrap, L.K. (1978).  Determination of acrylonitrile in air.
      Metody Anal.  Kontralya  Kach.  Prod.  Khim.  Prom-sti.  5:28-29.   (In Russ.)
      Taken from:   Chem. Abst.  90_:2826lt, 1979.

 Neely,  W.B.j Branson,  D.R.;  and Blau, G.E.  (1974).  Partition  coefficient  to
      measure bioconcentration  potential of organic chemicals in fish.  Environ.
      Sci.  Technol. 8/13):1113-15.

 Nelson, G.O., and Harder, C.A.  (1974).  Respirator  cartridge efficiency studies.
      V.  Effect of solvent vapor.  Amer. Ind. Hyg.  Assoc. J.  35(7);391-410.

 Nestler,  H., and  Berger, W.  (1965).    Gas  chromatographic  analysis,  of  water
      containing solutions of  acrylonitrile and various esters in dimethyl-forma-
     mide.   Chem. Tech.  17(3):169-70.   (In Ger.)  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.
                                     R-16

-------
NIOSH  (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, NIOSH (April
     17).  Submitted to Branch Chief of Industrial Workers Safety Board, NIOSH.

Norris, M.B.  (1962).  Acrylonitrile.  Encycl. Ind. Chem. Anal. 4_:368-8l.

O'Berg,  M.  (1980).  Epidemiologic  study  of workers exposed  to  acrylonitrile.
     J. Occup. Med. 22:245-252.

Ofengard, J. (1967). The  function of pseudouridylic acid in transfer ribonucleic
     acid.  I.   The specific  cyanoethylation of pseudouridine,  inosine,  and 4-
     thiouridine by acrylonitrile.  J. Biol. Chem. _24_2 (21): 5034-45.

Ofengard, 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. J.:21-23.   (In Russ.)  Taken from:   Chem. Abst. 74_:43238d, 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. §2(4):1025-29.

Patterson, R.M.;  Bornstein, M.I.; and Garshick, E. (1976).  Assessment of Acrylo-
     nitrile as  a  Potential Air  Pollution  Problem, Publication No.   GCA-TR-75-
     32-G(6).   Prepared  by  GCA Corporation,  GCA/Technology Division,  Bedford,
     MA, under Contract No. 68-02-1337.  U.S. Environmental  Protection Agency,
     Research Triangle Park, NC.

Paulet, G., and  Vidal, Mrs. (1975).   [Toxicity of some acrylic  and methacrylic
     esters of acrylamide and polyacrylamides.]  Arch.  Mai.  Prof. Med.  Trav.
     Secur. Soc.  36(1-2):58-60.   (In Fr.)   Taken from:   Chem. Abst.  83:9l832a,
     1975.                                                          —

Pell, S.;  E.I.  duPont de Nemours  and  Co.    (1980).   Letter  to  the  Carcinogen
     Assessment Group,  U.S.  Environmental  Protection  Agency  (July 23).

Perry,  R.H., and Chilton, C.H.,  editors (1973).  Chemical Engineers' Handbook,
     5thbed.  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.  52:211-213.

Petrova, L.I.; Maksimova, N.S.; and  Boikova, Z.K.   (1978).   Determination  of
     traces of styrene, acrylonitrile, and  methacrylate monomers  in binary mix-
     tures.   Gig.  Sanit.  4_:84-86.   (In  Russ.)   Taken  from:    Chem.  Abst.
     9_0:6935p, 1979.

Pilotti, A.;  Ancker,  K.; Arrhenius,  E.;  and  Enzell,  C. (1975).   Effects  of
     tobacco and tobacco  smoke constituents on  cell  multiplication in  vitro.
     Toxicology jp(l):49-62.
                                     R-17

-------
Plieninger, H.,  and  Sharma,  S. (1978).  Note on a  simple  detection  of  aorylo-
     nitrile in solvents.  Justus Liebigs Ann. Chem. ^:535-36.   (In Ger.)  Taken
     from:  Chem. Abst. 89_:52913g, 1978.

Pokrovskaya, L.A., and Frolova, G.S. (1969).  Chromatographic method  in  produc-
     tion  control  of  butadiene-acrylonitrile  rubbers.     Fiz.   Khim.  Khim.
     Tekhnol., pp. 201-06.  (In Russ.)  Taken from:  Chem. Abst. 75:99134e, 1971.

Ponomarev, Yu P.; Anufrieva,  T.L.; and Shkorbatova,  T.L.  (1974).   Polarographic
     determination   of  acrylonitrile  in   waste   waters   using  extraction.
     Vodosnabzh., Kanaliz.,  Gidrotekh.  Sooruzh.  17:67-70.   (In  Russ.)    Taken
     from:  Chem. Abst. !32:34803x, 1975.

Portmann, J.E., and Wilson, K.E. (1971).  The toxicity of  140 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.j 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.

Quast,  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 Oncogenicity 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.

Quast, J.F.; Schuetz, D.J.; Balmer,  M.F.; Gushow, T.S.; Park, C.N.;  and McKenna,
     M.J.  (1980b).   A Two-Year Toxicity and  Oncogenicity Study with  Acrylo-
     nitrile Following Inhalation Exposure of Rats.   Prepared by the  Toxicology
     Research Laboratory, Health and Environmental  Sciences,  Dow Chemical, USA,
     Midland, MI,  for  the Chemical Manufacturing Association,  Washington,  DC.
     Available from:  CMA, Washington,  DC.
Rabello-Gay, M.N., and Ahmed,  A.E.  (1980).   Acrylonitrile:
     studies in mice and rats.   Mutat.  Res.  79:249-255.
In vivo cytogenic
Radimer, G.F.; Davis, J.H.; and Ackerman, A.B.  (1974).   Fumigant-induced toxic
     epidermal necrolysis.  Arch.  Dermatol.  110(1):103-04.
                                     R-18

-------
Rajendran, S., and Muthu, M. (1976).  Toxioity of acrylonitrile to the adults of
     five  species   of  stored  products  insects.     Bull.  Grain   Technol.
     14(3):179-181.

Randall, T.L., and Knopp, P.V. (1980).   Detoxification of specific organic sub-
     stances by wet oxidation.  J. Water Pollut. Contr. Fed. 52:2117-2130.

Reichle, A., and Tengler, H. (1968).   Gas-chromatographic procedures for deter-
     mining residual monomers in  the polymer.  Determination of acrylonitrile in
     polyacrylonitrile.  Plastverarbeiter j.9_(12):921-24.   (InGer.)  Taken from:
     Chem. Abst. 70:97206c, 1969.

Renn, C.E.  (1955).   Biological properties and  behaviors  of cyanogenic wastes.
     Sewage Ind. Wastes j27(3):297-310.   (Cited in Miller and Villaume,  1978.)

Roberfroid, M.; Poncelet,  Fa,'  Lambotte-Vandepaer,  M.;  Duverger-VanBogaert,  M.;
     DeMeester, C.; and Mercier,  M. (1978).  Acute biotoxic effect of styrene on
     rat liver.   Correlation with enzyme-mediated mutagenicity  of  benzypyrene
     and acrylonitrile.  Scand. J. Work Environ. Health jKsuppl.  2):l63-68.

Rogaczewska, T.  (1964).   Polarographic determination of  acrylonitrile  in air.
     Chem. Anal. 2(3):417-23.   (In Pol.)  Taken from:  Chem. Abst.  62:995d, 1965,

Rogaczewska, T.  (1975).   [Percutaneous  absorption of acrylonitrile vapors  in
     animals.]  Med. Pracy 26:459-65.   (In Pol.)

Rogaczewska,  T.  (1976).    Acrylonitrile determination  in  air.   Med.  Pracy
     27(2):115-26.  (In Pol.)  Taken  from:   Chem.  Abst. 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. Pharmacol. ^7(4): 559-65.

Roy, S.S.  (1977).   Titrimetric determination of residual  monomers  in  styrene-
     acrylonitrile copolymers. Analyst 3.02(1213):302-05.

Russell, J.W.  (1975).  Analysis  of air pollutants  using sampling tubes and  gas
    • chromatography.  Environ. Sci. Technol.  2d3) = 1175-78 .

Russkikh, A.A.  (1971).   Determination  of acetone cyanohydrin and acrylonitrile
     in air.  Fiz.-Khim. Metody Anal. Kontr.  Proizvod.,  Mater. Konf. Rab.  Vuzov
     Zavod. Lab. Yugo-Vostoka  SSSR, 4th _3:100  (Pub.  1972).   (InRuss.)   Taken
     from:  Chem. Abst. 8JL:l6301g, 1974.

Russkikh, A.A.  (1973).  Photometric determination of acrylonitrile and  meth-
     acrylonitrile in  air.   Zavod. Lab. 39_(l):5-6.   (In  Russ.)   Taken  from:
     Chem. Abst. 72=1015666, 1973.

Sakurai, H.;  Onodera, M.; Utsunomiya,  T.; Minakuchi, H.; Iwai, H.;  and Matsumura,
     H.  (1978).   Health effects  of acrylonitrile  in acrylic fiber  factories.
     Brit. J. Ind. Med. 35:219-22.

Sandberg, E.  Ch.,  and Slanina,  P. (1980).  Distribution of [1- C]-acrylonitrile
     in rat and monkey.  Toxicology Lett. 6:187-191.
                                     R-19

-------
Sansone, E.B.; Tewari, Y.B.; and Jonas,  L.A.  (1979).   Prediction of removal of
     vapors from air by adsorption on activated carbon.  Environ. Sci. Technol.
     13(12):1511-13.

Santodonato, J.; Basu, D.;  and Howard,  P.H.  [1979].   Multimedia human exposure
     and carcinogenic risk assessment for environmental PAH.  In:  Chemistry and
     Biology  of Polynuclear Aromatic Hydrocarbons.    (Proceedings  of the  4th
     International  Battelle Symposium  on Polynuclear  Aromatic  Hydrocarbons,
     Columbus,  OH,  October  2-4,  1979.)   Ann Arbor,  MI:   Ann Arbor Science
     Publishers, Inc. (Book in press.)

Sarofim, A.F.; 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
     ^7:184-87.  (In Ital.)

Scheddel, R.T. (1958). Infrared quantitative analysis  data:  Analysis of acrylic
     acid,  ethylbenzene,  methyl  methacrylate,   °-methylstyrene,  styrene,  and
     acrylonitrile mixtures.  Anal. Chem. 30:1441.

Sevast'yanova,  I.G.,  and Tomilov,  A.P.  (1963).    Reduction of  nitriles on  a
     dropping mercury  cathode.   Zh.  Obshch.  Khim. ^3(9):2815-19•    (In  Russ.)
     Taken from:  Chem. Abst. 6Cl:1583e, 1964.

Sevast'yanova, I.G.; Tomilov, A.P.;  and  Yamaleev, I.Ya.  (1966).   Polarographic
     determination  of acrylonitrile  in  industrial   solutions.    Zavod.  Lab.
     _32(10):1210.  (In Russ.)  Taken from:  Chem. Abst. 66:34651r,  1967.

Silverstein, L.G.  (1977).   Validation of abcor GASBADGE  for acrylonitrile  and
     improved desorption efficiency.  Amer. Ind.  Hyg.  Assoc. J. 38(8);412-13.

Sloof, W. (1978).  Biological monitoring based on fish respiration for continuous
     water quality control.  In:   Aquatic Pollutants:  Transformation and Bio-
     logical Effects.  Hutzinger,  0.; VanLelyveld, L.H.;  and Zoeteman, B.C.J.,
     editors.  Elmsford,  NY:  Pergamon Press, pp. 501-06.

Smith, C.J. (1975).  Solvent vapour phototoxicity. Lab.  Pract. 24:9-11.

Smyth, H.F., Jr.,  and Carpenter, C.P.  (1948). Further experience with the range
     finding  test in  the  industrial toxicology  laboratory.    J.  Ind.  Hyg.
     Toxiool. ^0:63-68.

Smyth, H.F.,  Jr.; Weil,  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.
     14(2):340-07.
Stallard, C.  (1982). Acrylonitrile epidemiology.
     Unpublished report.
Standard Oil Company of Ohio.
                                     R-20

-------
Stanford Research Institute  (1978).   1978  Chemical Economics Handbook.
     Park, California,  1978.   p.  607.5032J.
            Menlo
Steere, N.V.  (1968).   Safety in the  chemical laboratory.
     chemicals data.  J. Chem. Ed. 45(3):A199-A205.
XLVI.   Hazardous
Stefanescu, T., and Ursu,  Gh.  (1973).  Method for acrylonitrile and acetonitrile
     determination in residual waters.  Mater. Plast.  10(6):330-34.   (In Rom.)
     Taken from:  Chem. Abst. 8_0:40770j, 1974.

Steichen, R.J.  (1976).  Modified  solution  approach  for the gas chromatographic
     determination  of  residual monomers by  head-space analysis.  Anal.  Chem.
     48(9): 1398-1402.

Suta, B.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 Quality
     Planning and Standards, Research Triangle Park, NC.

Suta,  B.E.  (1982).   Letter  to  J.E.  Hadley,  Jr.  (LaRoe,  Winn,  and  Moerman,
     Attorneys  at Law, 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. _3_3(1) :46-51.

Szabo,  S., and  Selye,  H.  (1971).   Adrenal  apoplexy  and  necrosis produced by
     acrylonitrile.  Endokrinologie 57(3):405-08 .

Szabo,  S.,  and Selye, H.  (1972).   Effect  of phenobarbital and steroids on the
     adrenal  apoplexy produced  by acrylonitrile  in  rats.    Endocrinol.  Exp.
     6^(3): 141-46.

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. J37:133.   (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.  79(1):32-37.

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,  R.G.,  editor.
     Toronto,  Ontario,  Canada:  International  Congress on  Toxicology.

 Tanaka, M.;  Nishimura, F.; and Shono, T. (1975).  Pyrolysis-gas chromatography of
      vinyl chloride-methyl methacrylate and vinyl chloride-acrylonitrile copoly-
     mers.  Anal. Chim.  Acta 74(1):119-24.
                                      R-21

-------
Taubinger, R.P.  (1969).  Direct  determination of  free  acrylonitrile in  aqueous
     copolyraer latexes.  Analyst 94(1121);628-33.

Terent'ev, A.P.,  and  Obtemperanskaya, S.I.  (1956).   Quantitative  determination
     of acrylonitrile with  sodium sulfite.  Zhur. Anal.  Khim _ll:638-39.   (In
     Russ.) Taken from:  Chem. Abst. _51:11935h,  1957.

Thiess, A.M., and Fleig, I.  (1978).  Analysis of chromosomes  of workers  exposed
     to acrylonitrile.  Arch. Toxicol. _4l(2):149-52.

Thiess, A.M.; Frentzel-Beyme, R.; Link,  R; and Wild, H.   (1980).  Mortalitats-
     studie Bei Chemiefacharbeitern Verschiedener Produktionsbetriebe Mit Expo-
     sition Auch Gegenuber Acrylonitrile Zentralbl. Arbeitsmed. 30:359-267.
Thompson,  K.R.   (1974).
     5(3):18-20.
Process  mass  spectrometry.
DuPont  Innovation
Troxell, T.C.  (1980).   Letter from T.C.  Troxell,  Division  of Food 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(4):975-80.

Uhde, W.J., and Koehler, U.  (1967).  Testing of plastic articles.  Determination
     of monomeric compounds  in the  volatile  fraction of styrene polymers.  Z.
     Lebensm.-Unters. Forsch 135(3):135-40.  (In Ger.)  Taken from:   Chem.  Abst.
     67:117599b, 1967.

U.S. DHEW/PHS/NCHS  (U.S. Department  of Health,  Education and Welfare/Public
     Health  Service/National  Center   of  Health  Statistic).     1976.     Vital
     Statistics of the United States.   Vol. II  - Mortality. Part B.

U.S. DHEW (U.S. Department of Health, Education and Welfare).  (1973). Adult Use
     of Tobacco, 1970.   National  Clearinghouse  for Smoking  and Health.

U.S. EPA (Environmental Protection Agency)  (1976b).  EPA Method 624,  Purgeables,
     44 Fed. Reg. 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. 68-01-4646  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, 44  Fed.  Reg.  69479-69484, Dec.  3,  1979.

U.S. EPA (Environmental Protection Agency) (1980a).   EPA Method 1624.  Volatile
     Organic Compounds by Purge and  Trap Isotope  Dilution   GC-MS,  U.S. EPA
     Effluent Guidelines Division, June,  1980.
                                     R-22

-------
U.S.  EPA (U.S.  Environmental Protection Agency) (1980b).  Ambient Water Quality
     Criteria Document:  Acry^Lonitrile.  U.S. EPA.  EPA 440/5-80-017,  Oct.,  1980.
     Available from:  National  Technical Information Service,  Springfield,  VA.

U.S. EPA (Environmental Protection Agency).   (1980c).   Carbaryl  Decision Docu-
     ment.  Fed. Reg.  _4_5(241): 81869-81876.

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. EPA (U.S. Environmental Protection Agency) (1982).   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.;   Kizilova,  L.I,;  Gavrilina,   L.Ya.; Malakhov,  V.V.;   and
     Yashin, Ya.I.  (1977).  Gas-chromatographic  analysis of products  from  the
     oxidative ammonolysis of propylene.  Izv. Sib.  Otd. Akad. Nauk SSSR,  Ser.
     Khim. Nauk _2:115-18.   (In Russ.)  Taken from:   Chem. Abst. 8_7_:77980y, 1977.

Ved Brat, S. and Williams, G.M.  (1982). Hepatocyte-mediated production of sister
     chromatid  exchange  in cocultured  cells  by  acrylonitrile:   evidence  for
     extra-cellular transport  of a s.table 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. Fish. Rejs. Board  Can.
     36:1040-48.
Venitt,  S.  (1978).
     57(D:107-09.
Mutagenic  studies  with  acrylonitrile.    Mutat.  Res.
Venitt, S.; Bushell,  C.T.; and Osborne, M. (1977).  Mutagenicity of acrylonitrile
     (cyanoethylene)  in Escherichia coli.  Mutat.  Res.  45(2):283-88.

Werner, J.B., and Carter, J.T. (1981).  Mortality of United Kingdom acrylonitrile
     polymersation workers.   British Journal of Industrial Medicine  38;247-253.

Wilcox,  W.S.,  and Goldstein,  J.H.  (1954).   The microwave  spectrum of  vinyl
     cyanide.  J. Chem. Phys. _22(3): 516-18.

Wilson, R.H. (1944).   Health hazards encountered in the manufacture of synthetic
     rubber. J. Amer. Med. Assoc. 124:701-03.

Wilson, R.H.; Hough,  G.V.; and McCormick,  W.E.  (1948).  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.

Wright, P.L.  (1977).   Studies on  the  metabolism  of acrylonitrile.   Testimony
     before the FDA.   (Cited in Miller and Villaume,  1978.)
                                     R-23

-------
Wronski,  M., and  Zbigniew,  S.   (1974).    Thiomercurimetric  determination  of
     acrylonitrile in water.  Chem. Anal. 19_(3):633-38.   (In Pol.)  Taken  from:
     Chem. Abst. 81175988c, 1974.

Young, J.D.; Slauter, R.W.; and Karbowski, R.J. (1977).  The Pharmacokinetic and
     Metabolic Profile of l4C-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.  (1980).   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 intraperi-
     toneally injected  acrylonitrile.on liver, kidney and^xbrain.  Acta Pharma-
     col. Toxicol. 49:412-415.  Taken  from  : j'Chem.  Abst.- |H6:117119u.
                                      R-24
» U«OO«TmilCHTF)WiWa OFFICE,'19M - 759-102/10705

-------

-------
                                                                       •JJ O

                                                                       3  5
                                                                       Q)  O
                                                                       3-g
                                                                          en
                                                                       75'
                                                                       P' CD
                                                                       m 0>
                                                                       C
                                                                       en
                                                                       CD

                                                                       •en
                                                                       w
        a. =:
        2-e
o
o
oo
oo
to
o
o
T|
-. " •-
S"

II i
sll
|S 3
              •  O) 0}
              IT O W
              m 3- o
          3-2
          s§.
          S-S
        a o
        Q> ^
        a. m
        a. o
                                                                                            > m i
                                                                                           (O 3
                                                                                            0> <
                                                                                            U ^'
                                                                                            o o
                                                                                           •< 3
                                                                                              3
                                                                                              0)
                                                                                              D
                                                                                              r-*
                                                                                              o
                                                                                              CD
                                                                                              §
                                                                                              o'
                                                                                            OS"
                                                                                            1°
                                                                                            0) r»
                                                                                            -• o
                                                                                            O13
                                                                                            X
                                                                                            .p.
                                                                                            en
                                                                                            to
                                                                                            o>
                                                                                            oo

-------