United States
Environmental Protection
Agency
Office of
Toxic Substances
Washington DC 20460
EPA-560/2-81 006
November 1981
Toxic Substances
A Survey of Plasticizers:
Epoxies, Linear Polyesters, and
Trimellitates
-------
CHEMICAL TECHNOLOGY AND ECONOMICS IN
ENVIRONMENTAL PERSPECTIVE
Task VI - A Survey of Plasticizers: Epoxies,
Linear Polyesters, and Trimellitates
FINAL REPORT
November 1981
EPA Contract No. 68-01-3896
MRI Project No. 444l-T(6)
For
Environmental Protection Agency
Office of Toxic Substances
401 M Street, S.W.
Washington, B.C. 20460
Attn: Mr. Roman Kuchkuda
Project officer
-------
NOTICE
This report has been reviewed by the Office of Toxic Substances,
Environmental Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the views and
policies of the Environmental Protection Agency. Mention of trade names
or commercial products is for purposes of clarity only and does not con-
stitute endorsement or recommendation for use.
-------
PREFACE
This report presents the results of a study to compile and analyze
published information for three classes of plasticizers: epoxy compounds,
linear polyesters, and trimellitates.
This study was performed by Midwest Research Institute as Task VI
under Contract No. 68-01-3896 for the Office of Toxic Substances of the
U.S. Environmental Protection Agency. Project officer for this study was
Mr. Roman Kuchkuda. Midwest Research Institute contributors to this study
were: Dr. Thomas W. Lapp (Task Leader), Mr. Charles E. Mumma, and Mr.
Joseph Chaszar. This contract is being performed under the supervision of
Mr. Thomas L. Ferguson, Head, Process Analysis Section.
Midwest Research Institute would like to express sincere appreciation
to the many industry sources who provided technical input to this study,
especially to Mr. J. T. (Jack) Lutz and Mr. J. E. Voit of Rohm and Haas
Company, Mr. Robert Radue of Monsanto Company, and Mr. Jesse Edenbaum of
Technor-Apex for their valuable assistance.
Approved for:
MIDWEST RESEARCH INSTITUTE
(£-6.-^
Bruce W. Macy, Acting Director
Center for Technoeconomic Analysis
November 10, 1981
11
-------
CONTENTS
Figures . iv
Tables ....'. v
1. Introduction 1
2. Summary 2
3. Methodology and Data Sources 3
Literature sources. ... 3
Computer data systems 3
Other sources 7
4. Physical and Chemical Properties 8
Epoxy compounds 8
Linear polyesters 15
Trimellitates 23
5. Production and Use 28
Epoxy compounds 28
Linear polyesters ......... 37
Trimellitates 44
6. Health and Environmental Effects .... 50
Health effects 50
Environmental effects ........ 54
7. Plasticizer Interchangeability 57
References . 63
111
-------
FIGURES
Number Page
V-l Schematic of the two basic epoxidation processes 31
V-2 Generalized process flow diagram for linear polyesters . . 40
V-3 Schematic flow diagram for production of trimellitate
plasticizers . . 46
IV
-------
TABLES
Number . Page
IV-1 Physical Properties of Commercial Epoxy Compounds 11
IV-2 Physical Properties of Linear Polyester Plasticizers ... 17
IV-3 Chemical Abstract Numbers for Polyesters 22
IV-4 Physical Properties of Trimellitate Plasticizers 25
IV-5 Trade and Chemical Names of Trimellitates 26
V-l Manufacturers of Epoxy Plasticizers. ... 29
V-2 Annual Production of Epoxy Plasticizers. . .......... 29
V-3 Manufacturers of Polyester Plasticizers. 37
V-4 Annual Production of Polyester Plasticizers 38
V-5 Manufacturers of Trimellitate Plasticizers 44
V-6 Annual Production of Trimellitate Plasticizers 45
VII-1 General Compatibility of Plasticizers. 58
VII-2 Interchangeability for Epoxy Plasticizers 59
VII-3 Interchangeability for Polyester Plasticizers 60
VII-4 Interchangeability for Trimellitate Plasticizers 61
-------
SECTION I
INTRODUCTION
Plasticizers are an integral component of many current thermoplastics.
This component imparts workability, flexibility, extensibility, and resil-
ience to the final product. Over 80% of all plasticizers for thermoplastics
are used with polyvinyl chloride (PVC); over 90% of the plasticizers under
study in this report are used in PVC. The entire classification of plasti-
cizers entails a wide range of compounds, most of which are esters of long-
chain (Cs or higher) acids. The Environmental Protection Agency, Office of
Pesticides and Toxic Substances (OPTS), is currently reviewing the various
materials employed as plasticizers and compiling information on these com-
pounds so that their relative importance may be evaluated.
Two major classes of plasticizers, triaryl and alkylaryl phosphate
esters and alkyl phthalate esters, have been or are currently under exten-
sive study by OPTS. However, a very large number of other plasticizers re-
main that are not under study, and little information has been compiled on
these materials. Three classes of plasticizers were selected for investi-
gation during this task. These were:
* Epoxy compounds derived from soybean oil, linseed oil, or tall oil;
* Linear polyesters derived from adipic, sebacic, phthalic or glutaric
acids;
* Trimellitic acid esters (trimellitates).
The objectives of this study were to compile and analyze published in-
formation for each individual plasticizer in the areas of physical and chemi-
cal data, potential occupational and environmental exposure, manufacturing
sites and processes, use patterns, environmental degradation, biological
effects, and plasticizer interchangeability.
This report is divided into eight major sections. Section I presents
a brief introduction to the report and outlines the overall program objec-
tives. Section II contains the summary. Section III outlines the method-
ology and data acquisition techniques employed in the study. Section IV
presents the physical and chemical properties. Section V describes the
manufacturing and use information. Section VI provides data on environmen-
tal and health effects. Section VII contains information on plasticizer
interchangeability.
-------
SECTION II
SUMMARY
Epoxide derivatives of soybean oil, linseed oil, and tall oil esters;
linear polyesters; and esters of trimellitic acid are three classes of plas-
ticizers employed in a variety of plastics (primarily polyvinyl chloride)
and elastomers. Small quantities are used in various coating applications.
In general, the three classes of plasticizers impart good high or low temper-
ature flexibility while exhibiting low volatility, low migration rates, and
good resistance to oil or water extraction from the plastic medium.
In 1979, production volumes of these plasticizers were stated to be:
130 million pounds for epoxy plasticizers; 55 million pounds for linear poly-
esters; and 31 million pounds for trimellitates. For epoxy compounds, pro-
duction processes use either performic acid or peracetic acid. With linear
polyesters and trimellitates, multipurpose equipment is used in the produc-
tion of esters from the corresponding acids and alcohols.
Little published information is available concerning the health effects
of any of the three classes of plasticizers. No occupational standards exist
for any of the three classes. Feeding studies using rats and dogs showed
no significant toxic effects for epoxy compounds at levels up to 5% in the
diet. Skin and eye irritation tests showed epoxy compounds to be either
mild irritants or to be nonirritants. Two-year chronic feeding studies with
a linear polyester using rats and dogs showed no significant toxic effects
at levels up to 1% in the diet. Skin and eye tests generally showed little
or no irritation. For trimellitates, no studies were found in the literature;
data were obtained from manufacturers. Acute oral LDso studies with rats, mice,
and rabbits showed these compounds to have relatively low toxicity. In one
inhalation study, during which the trimellitate was heated to 180°C, rats ex-
posed to the vapor died. These deaths were delayed for up to 3 days after
conclusion of the tests.
Environmental tests were limited to static fish toxicity studies for
two trimellitates and studies of fungal and bacterial growth on epoxy com-
pounds and linear polyesters. In the fish toxicity studies, the 96-hr tests
showed the trimellitates to be nontoxic towards fingerling rainbow trout
and bluegill. Numerous studies of plasticizer degradation were reported
for fungal and bacterial action on epoxy compounds and linear polyesters.
The results of all studies showed both classes of plasticizers to be very
susceptible to attack.
-------
SECTION III
METHODOLOGY AND DATA SOURCES
This section describes the methodology used and identifies the data
sources employed for the acquisition of information presented in the report.
LITERATURE SOURCES
A number of books and periodicals were investigated as potential sources
of information concerning the manufacture and use of the various classes of
plasticizers. Among the sources employed were the following books and period-
icals:
* Encyclopedia of Chemical Technology;
* Encyclopedia of Polymer Science and Technology;
* Chemical and Process Technology Encyclopedia;
* Encyclopedia of PVC;
* Chemical Process Industries;
* Industrial Chemicals;
* Directory of Chemical Producers;
. * Chemical Economics Handbook;
* Various trade publications such as Modern Plastics, Plastics Tech-
nology, Chemical Marketing Reporter, Modern Plastics Encyclopedia; and
* Government publications.
COMPUTER DATA SYSTEMS
Searches were made of various computer-based data storage systems for
information on each class of plasticizer and on each of the specific plas-
ticizers known to be in commercial production. For most searches, the Chemi-
cal Abstracts Service Registry Number was employed as the initial code. For
those systems that did not employ the registry numbers, the full names of
the specific plasticizers, as denoted in the TSCA Candidate List of Chemi-
cal Substances, were employed. Brief synopses of the various data storage
systems used for the search and the source of the data file are presented
alphabetically in the following subsections.
-------
BIOSIS Previews
This file contains citations from both Biological Abstracts and Bio-
research Index, the major publications of Biosciences Information Service
of Biological Abstracts. Together, these publications constitute the major
English language service providing comprehensive worldwide coverage of re-
search in the life sciences. Approximately 8,000 primary journals as well
as reviews, preliminary reports, selected institutional and government re-
ports, and research communications are included for all aspects of the bio-
sciences and medical research. The file is divided into two separate sec-
tions, one covering 1969 through 1971 and the other from 1971 to present.
Approximately 2,300,000 records are included in this data source of Biosci-
ences Information Service, Philadelphia, Pennsylvania.
Chemical Abstracts (CA) Search
The CA Search is an expanded data source which resulted from the merger
of two files: the CA Condensates file and the CASIA file. The CA Conden-
sates file contains the basic bibliographic information appearing in the
printed Chemical Abstracts volumes. The CASIA file contains the general
subject headings from a controlled vocabulary and the CAS Registry Numbers.
Other uncontrolled vocabulary terms and cross-referenced general subject
headings are also included. This data source is divided into three files:
1967-1971; 1972-1976; and 1977-present. Approximately 4,000,000 records
are included in this data source from Chemical Abstracts Service, Columbus,
Ohio.
Enviroline
This data file covers environmental information including management,
technology, planning, law, political science, economics, geology, biology,
and chemistry as they relate to environmental issues. The interdisciplinary
approach provides indexing and abstracting coverage of more than 5,000 in-
ternational primary and secondary source publications including periodicals,
government documents, industry reports, monographs, proceedings of meetings,
and rulings from the Federal Register. This file covers 1971 to the present
and contains about 75,000 citations. It is a product of the Environment
Information Center, Inc., New York.
Environmental Periodicals Bibliography (EPB)
The EPB data file covers the fields of general human ecology, atmos-
pheric studies, energy, land and water resources, and nutrition and health.
Approximately 250 periodicals are indexed for this data source. This file
covers 1973 to the present and contains about 137,000 records. It is a
product of the Environmental Studies Institute, Santa Barbara, California.
-------
Excerpta Medica
This data source is one of the major sources for searching the biomedi-
cal literature. It consists of abstracts and citations of articles from
over 3,500 biomedical journals published throughout the world. The file
covers the entire field of human medicine and related disciplines. An on-
line file corresponds to the 43 separate specialty abstract journals and 2
literature indexes. This data source covers June 1974 to the present and
contains in excess of 1,003,000 records. It is available through Excerpta
Medica, Amsterdam, The Netherlands.
Medline
This data source is a service of the National Library of Medicine in
Bethesda, Maryland. It contains approximately 600,000 references to bio-
medical journal articles published in the current and two preceding years.
The articles are from 3,000 journals published in the U.S. and 70 foreign
countries as well as a limited number of chapters and articles from selected
monographs. In addition to the current file from 1978 to are present, back
files are available to cover 1966-1968, 1969-1971, 1972-1974, 1975, and 1976-
1977. Total references in all files exceed 2,250,000.
National Technical Information Service (NTIS)
The NTIS data base consists of government sponsored research, develop-
ment, and engineering plus analyses prepared by federal agencies and their
contractors or grantees. It provides a means through which unclassified,
publicly available, unlimited distribution reports are made available from
governmental agencies. The data base includes material from both the hard
and soft sciences, including many topics of broad interest such as environ-
mental pollution and control, energy conversion, technology transfer, be-
havioral/societal problems, and urban and regional planning. This data
source covers 1964 to the present and contains about 730,000 citations.
The file is a product of NTIS, U.S. Department of Commerce, Springfield,
Virginia.
Pollution Abstracts
This data source is one of the primary resources for references to en-
vironment-related literature on pollution, its sources, and its control.
The file covers subjects such as air pollution, environmental quality, noise
pollution, pesticides, radiation, solid waste, and water pollution. This
source covers 1970 to the present and currently contains about 67,000 cita-
tions. It is a product of Data Courier, Inc., Louisville, Kentucky.
RAPRA Abstracts
The Rubber and Plastics Association (RAPRA) data file is a comprehen-
sive source covering the commercial, technical, and research aspects of the
rubber and plastics industries. It provides information on materials includ-
ing synthesis and polymerization, raw materials and monomers, and compounding
-------
ingredients. RAPRA includes information on applications of polymers, toxicity
reports, and potential environmental and industrial health hazards. This data
source covers 1972 to the present and contains approximately 110,000 records.
It is maintained by the Rubber and Plastics Research Association of Great
Britain, Shawbury, Shrewsbury, Shropshire, England.
Science Citation Index (SCISEARCH®)
This data source is a multidisciplinary index to the literature of sci-
ence and technology and includes all records published in Science Citation
Index (SCI®) and additional records from the Current Contents series of pub-
lications that are not included in the printed version of SCI. The file
contains reference from about 2,600 major scientific and technical journals
and covers all areas of the pure and applied sciences. This source covers
1974 to the present in two files; one file covers 1974-1977 and the other,
1978 to the present. Total citations from both files number about 2,700,000.
The data file is produced by the Institute for Scientific Information,
Philadelphia, Pennsylvania.
SSIE Current Research
The Smithsonian Science Information Exchange (SSIE) Current Research
is a data file containing reports of both government and privately funded
scientific research projects, either currently in progress or initiated and
completed during the most recent two years. Data are collected from the
funding organization at the inception of a research project in all fields
of basic and applied research in the life, physical, social, and engineering
sciences. Project descriptions are received from over 1,300 organizations
that fund research, including federal, state, and local government agencies;
nonprofit associations and foundations; and colleges and universities. About
90% of the information in the data base is provided by agencies of the federal
government. The most recent data file (1978-present) contains in excess of
250,000 citations. This source is maintained by the Smithsonian Science
Information Exchange, Washington, B.C.
Toxicology Data Bank (TDB)
This data source contains chemical, pharmacological, and toxicological
information and data on approximately 1,000 compounds. Information is being
prepared on an additional 1,500 compounds. Data for this file are extracted
from handbooks and textbooks and subject to review by a peer group of subject
specialists. This source is a service of the National Library of Medicine,
Bethesda, Maryland.
Toxline
The Toxicology Information Online source is a service of the National
Library of Medicine in Bethesda, Maryland. This source is a collection of
over 520,000 references from the last five years of published work on human
and animal toxicity studies, effects of environmental chemicals and pollu-
tants, and adverse drug reactions. Essentially all references have abstracts
-------
or indexing terms, and most compounds are identified by the CAS Registry
Numbers. The references employed in this source are from five major published
secondary sources and five special literature collections maintained by other
organizations. The current file covers 1974 to the present. Older material
(~ 400,000 references) is contained in a separate file designated Toxback.
OTHER SOURCES
Telephone and letter contacts were made with industry trade organiza-
tions, manufacturers, distributors, and users to identify and collect avail-
able information relating to the various classes of plasticizers. Informa-
tion was obtained on the physical and chemical properties of many of the
specific plasticizers of interest. In addition, information was obtained
relating to the manufacturing process, product purity, product losses, use
patterns, environmental effects, and health effects for each class of plasti-
cizer. Except for environmental effects and health effects, little informa-
tion can be found in the published literature for the other topics. The
information provided by the groups identified above was the sole source of
the data.
-------
SECTION IV
PHYSICAL AND CHEMICAL PROPERTIES
This section presents discussion of the characterization of each class
of plasticizer, a compilation of the physical properties of the commercial
products, and a discussion of the chemical properties of each class.
EPOXY COMPOUNDS
This subsection will provide a discussion of the nature of the raw
materials as well as a discussion of the physical and chemical properties
of the epoxy compounds.
Characterization
Soybean Oil--
Soybean oil is a triglyceride comprised of glycerol and several unsatu-
rated fatty acids. It is extracted from the crushed bean with a mixture of
petroleum hydrocarbons. The oil is further refined with sodium hydroxide
or sodium carbonate to remove excess fatty acids, and then the oil is bleached.
The chemical structure of this oil is quite complex owing to the combinations
and permutations of fatty acids that can be esterified at the three nonequiva-
lent (enzymatically) hydroxyl groups of the glycerol. A generalized triglyc-
eride has the following structure, without regard for optical activity:
R = fatty acid component
Soybean oil has the following composition with respect to the fatty acid
components of the triglyceride (Applewhite, 1980):
Acid Structure % Composition
Hexadecanoic (palmitic) CH3(CH2)14C02H 2.3-10.6
Octadecanoic (stearic) CH3(CH2)i6C02H 2.4-6
cis-9-Octadecenoic (oleic) CH3(CH2)7CH=CH(CH2)7C02H 23.5-30.8
-------
cis,cis-9,12-0ctadecadienoic CHsCCH^h.CCIkCHiCH^CCIkhCO^H 49-51.5
(linoleic)
cis,cis,cis-9,12,15- CH3(CH2CH=CH)3(CH2)7C02H 2-10.5
Octadecatrienoic (linolenic)
Linseed Oil--
Like soybean oil, linseed oil is also a triglyceride derived from
glycerol and unsaturated fatty acids. It is produced from the seeds of the
common flax plant; the seeds contain approximately 33-43% oil (Conan, 1980).
Climatic conditions at the time the oil is developed in the seed affect the
degree of unsaturation; the lower the temperature, the higher the extent of
unsaturation. The flaxseed is normally reduced to 20-30% oil in a screw
press, and then the residual oil is extracted from the crushed seed with
hexane. Further treatment of the crude oil is usually the same as described
for soybean oil.
Linseed oil contains the same fatty acids as soybean oil but exhibits
a somewhat higher degree of unsaturation. The typical composition of linseed
oil and a comparison to soybean oil are as follows (Cowan, 1980):
% Composition % Composition
Acid (linseed oil) (soybean oil)
Hexadecanoic 6 ~ 2-11
Octadecanoic 4 ~ 2-6
cis-9-Octadecenoic 13-37 . ~ 24-31
cis,cis-9,12-0ctadecadienoic 5-23 ~ 49-52
cis,cis,cis-9,12,15-0cta- 26-58 ~ 2-11
decatrienoic
Tall Oil-
Tall oil is the major by-product of the kraft or sulfate pulping pro-
cess; it is a mixture of resin, fatty acids, and unsaponifiables. The crude
oil is obtained by the acidification of the resin and fatty acid sodium soaps
recovered from the concentrated black liquor resulting from the pulping pro-
cess. Crude oil derived from pine trees in the various regions of the U.S.
contains approximately 43-56% fatty acids, 39-51% rosin acids, and 5-7% un-
saponifiables (Tate, 1969).
Many refining methods have been developed for the separation of the
rosin acids and the tall oil fatty acids. These methods include physical
and chemical techniques, as well as combinations of both techniques. Most
of the components in crude tall oil are decomposed or transformed when sub-
jected to high temperatures so that distillations are usually performed under
high vacuum using superheated steam. This simple distillation procedure is
normally followed by fractional distillation to obtain a mixture of refined
tall oil fatty acids (Johnson, 1978). A typical composition of tall oil
fatty acids in a refined, low-rosin mixture is as follows:
-------
Acid
% Composition
Hexadecanoic (palmitic) 1
Octadecanoic (stearic) 2
cis-9-Octadecenoic (oleic) 48-51
cis,cis-9,12-0ctadecadienoic (linoleic, nonconjugated) 37-40
cis,cis-10,12-0ctadecadienoic (linoleic, conjugated) 5
Unknown 4
The percentage composition data are basically from the literature (Tate,
1969) but modified slightly by MRI to normalize the percentage.
Prior to formation of the corresponding epoxy compounds, the tall oil
acids are treated with £4 or higher alcohols, such as 2-ethylhexanol, n-octyl
alcohol, or n-butyl alcohol, to form the corresponding esters.
Epoxidation--
Each of the three classes of oils (or esters of the acids) are epoxi-
dized by a variety of methods.depending upon the specific manufacturer or
the desired properties of the resultant product. The production processes
for the formation of the epoxy compounds from these raw materials are dis-
cussed in detail in Section V.
Physical Properties
Each manufacturer of epoxy compounds was contacted for information re-
lating to the physical properties of their respective compounds. The re-
sults are compiled in Table IV-1. Generally, the test procedures were not
identified by the manufacturers. If identified, the procedures were either
American Society for Testing and Materials (ASTM, 1980) or American Oil
Chemists Society (AOCS, 1974) methods. No data were available for properties
such as vapor pressure, solubility in water or organic solvents, or octanol-
water coefficients that would provide some insight into the potential for
environmental transport. Chemical Abstracts numbers and synopses of selected
physical properties are provided in the following subsections.
Chemical Abstracts Service Registry Numbers--
Chemical Abstracts numbers (CAS numbers, CASRN) for the epoxy compounds
of interest to this study are as follows:
Compound CAS number
Epoxidized soybean oil 8013-07-8
Epoxidized soya oil 61788-96-3
Epoxidized linseed oil 8016-11-3
Epoxidized linseed oil, butyl ester 68991-46-8
Octyl epoxytallate 61788-72-5
2-Ethylhexyl epoxytallates 61789-01-3
10
-------
TABLE IV-1. PHYSICAL PROPERTIES OF COMMERCIAL EPOXY COMPOUNDS
Compound
Epoxidized soybean oil
Polycizer ESOd
Flexol EPOf
E.S.O.8
Nuoplaz 849 h
Plastolein 92321
A
Drapex 6.8J
Epoxol 7-4k
Vikoflex 71701
Peroxidol 780™
Admex 710 & 71l"
Paraplex G-60°
Pa rap lex G-62 °
Plas-Chek 775 P
Epoxidized linseed oil
Flexol tOEf
Drapex 10. 4 J
Epoxol 9-5 k
Approximate Refractive Acid
Solubility
molecular Specific Viscosity index Oxirane Iodine number Saponif ication In water
weight gravity (cps at 25°C) (25°C) oxygen (%) number (rag KOII/g) number (% by wt.)
1000
1000
1000
1000
NA
1000
NA
1000
1000
HA
1000
1000
1000
1000
1000
NA
0.993
(25/20°C)
0.9977
(20/20°C)
0.995
(25/25°C)
0.99
(25/25°C)
0.99
(20/20°C)
0.992
(25/25°C)
0.994
(25/25°C)
0.992
(25/25°C)
0.991
(23/15. 5°C)
0.994
(25/25°C)
0.980
(25/15°C)
0.993
(25/15°C)
0.998
(25/25°C)
1.030
(25/20°C)
1.0385
(25/25°C)
1.030
(25/25°C)
372
518
(20"C)
340
320
159
320
314
317-416
355
368
350
550
875
700
(20°C)
1000
619
1.4730
(20°C)
NA
1.472
1.4565
1.470
1.4720
1.4705
1.472
1.4720
(23°C)
1.471
1.472
1.471
1.472
NA
1.4788
1.4715
7.3
7.0
6.9
NA
6.5
7.25
7.4
7.0.7.2
7.3
NA
NA
NA
7.3
9.0
9.6
9.2
0.8
NA
1.5
NA
2.5
1.3
NA
NA
NA
NA
NA
NA
1.5
NA
2.0
NA
0.4
NA
0.5
0.47
1.0
0.5
0.10
0.3
0.33
0.3
0.6
0.4
0.5
NA
0.5
0.12
conl tutted
NAe
NA
183-185
NA
NA
NA
178.1
NA
NA
NA
182
183
NA
NA
NA
172
NA
<0.01
(25°C)
NA
NA
NA
<0.01
(20°C)
NA
NA
NA
NA
NA
NA
NA
<0.015
(25°C)
<0.01
(20°C)
NA
Water in
I' (% by ut.)
NA
0.55
(25°C)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.017
(25°C)
NA
NA
Freezin
1 point (°
NA
NA
NA
NA
-18
NA
NA
0
NA
NA
5
5
NA
NA
NA
NA
g Pour
C) point (°C)
-2
-2
-10 to -5
0
NA
0
-4
NA
-2
-4 to 2
NA.
NA
-4
-3
-5
-1
Fl
point
316
157
316
288
307
143
310
NA
316
310
310
^,
310
316
238
290
:ii3
-------
TABLE IV-1. (continued)
Approximate
molecular
Compound weight
Plas-Chek 795P NA
Adoex ELO" NA
Octyl esters of tall oil fatty acids
Nuoplaz 850 h 1,20
Peroxidol 781m 424
Drapex 4.4 420
Flexol EP-8(o)f 420
Plastolein 9215* NA
Refractive Acid
Specific Viscosity index Oxirane Iodine number Saponifical
gravity (cps at 2S°C)a (2S°C) oxygen (1) number (mg kOII/g) number
1.03 399 1.477 9.4 HA 0.3 NA
(25/25°C)
1.032 815 1.477 NA NA 1.0 NA
(25/25°C)
0.92 20 1.456S NA NA 0.75 NA
(25/25°C)
0.924 46.6 1.4580 4.9 NA 0.5 NA
(23/15. 5°C) (23°C)
0.992 20 1.4580 5.1 2.2 0.5 NA
(25/25cC)
0.9232 35.2 NA 4.7 NA NA NA
(20/20°C) (20°C)
0.82 62 1.463 4.2 4.0 2.0 - NA
(20/20°C)
Solubility
Lion In water Water in Freezii
(% by wt.) (% by wt.) point ('
NA NA NA
NA NA NA
NA NA NA
NA NA NA
<0.01 NA -20
(20°C)
<0.01 0.3 NA
(20°C) (20°C)
NA NA -7
ig Pour Flash
>C) point (°C) point (°C)t
NA 310
-1 310
-22 221
-15 235
NA 220
-8.5 193
NA 282
a cps = centipoise: gm x 100; at 20"C, the viscosity of water = 1.0 cps.
sec x cm
b Test method was ASTH 0 97-66.
c Host test methods were Cleveland Open Cup. Others did not list the specific method used. Test method was ASTH D 92-66:
d Harwicke Chemical Corporation.
e NA - Not Available.
f Union Carbide Company.
g FHC Corporation.
b Tenneco Chemicals.
i Emery Industries.
j Argus Chemical Corporation, a subsidiary of Uitco Chemical Corporation.
k Swift Specialty Chemicals.
1 Viking Chemical Company.
m Reichold Chemicals.
n Sherex Chemicals.
o Rohm & Haas Company.
p Ferro Corporation.
-------
Other epoxy compounds are listed in compilations such as the Toxic Sub-
stances Control Act (TSCA) Candidate List of Chemical Substances but are
not among the products in Table IV-1 for which data were supplied by pro-
ducers or distributors. Many reasons could account for these omissions,
including discontinued products, uses other than as plasticizers, captive
uses, special orders, and low volume specialty items.
Specific Gravity—
For epoxidized soybean oils, the specific gravity of essentially all
of the listed commercial products is in the range of 0.991 to 0.998. With
epoxidized linseed oil, all values except one were in the range of 1.030 to
1.039. Three of the four octyl tallates had values between 0.920 and 0.924.
Overall, the epoxidized linseed oils had the highest specific gravity, fol-
lowed by the epoxidized soybean oils; the octyl epoxytallates had the lowest
values. Identified test method was ASTM D 1298-67.
Viscosity—
The viscosity of the various epoxidized soybean oils showed a wide vari-
ance, ranging from about 60 to 875 centipoise. Epoxidized linseed oils also
showed a wide range of viscosity levels, ranging from about 400 to 1,000.
For the four octyl epoxytallates, the levels were considerably lower and
more uniform with the range being 20 to 47 centipoise. Identified test
methods were ASTM D 445-65 or ASTM D 2393-68.
Oxirane Content--
Oxirane content, is the percentage of oxygen incorporated in the mate-
rial during epoxidation that is present as the oxygen in the oxirane struc-
ture, a three-membered ring containing one oxygen and two carbon atoms.
The percentage oxirane content for almost all of the epoxidized soybean oils
ranges from about 6.9 to 7.4%. Epoxidized linseed oils show a greater per-
centage of oxirane content because of the greater degree of unsaturation in
the raw material triglyceride. Percentage levels for all but one linseed
oil range from 9.0 to 9.6%. Octyl epoxytallates show the lowest oxirane
content with levels from 4.7 to 5.1%. Identified test method was AOCS Cd
9-57. .
Iodine Number--
The iodine number represents the grams of iodine absorbed per 100 grams
of epoxidized material. This number is a measure of the unsaturation remain-
ing after the epoxidation process. Since the oxirane content measures unsatu-
ration lost to oxirane formation, these values and the iodine number should
show an inverse relationship. For essentially all epoxidized soybean oils
listed in the table, the iodine numbers are about 0.8 to 1.5. Iodine numbers
were given for only two epoxidized linseed oils and one octyl epoxytallate.
Published test method was AOCS Tg-1-64.
Acid Number--
Acid number or value is the quantity of base (e.g. potassium hydroxide)
required to neutralize the free fatty acid in one gram of epoxidized product.
For the epoxidized soybean oils, the values ranged from 0.1 to 0.6 mg except
for two products which showed values of 1.0 and 2.0 mg. These two products
are also the same ones with low oxirane oxygen content and high iodine number.
13
-------
This indicates that quantities of free unsaturated acid are likely present
in these two products. Acid numbers for the epoxidized linseed oils ranged
from about 0.1 to 1.0 mg; those for octyl epoxytallates ranged from 0.5 to
0.75 mg. Identified test methods were ASTM D 1045-80 and AOCS Te-2a-64.
Saponification Number-
The saponification number is the quantity of base required to saponify
the esters and acids in one gram of plasticizer. Very few numbers were pro-
vided for any of the epoxy plasticizers, but the values that were reported
ranged from 150 to 185. Test methods were not identified.
Solubility--
Very few data were reported for the epoxy plasticizers. In general,
the solubilities in water are quite low, usually in the range of 0.01% by
weight or less. This very low solubility may account for the few data re-
ported. The test methods employed were not identified.
Chemical Properties
No studies were found in the literature related to the chemical pro-
perties of the epoxy plasticizers in the pure state; however, considerable
work has been performed on these plasticizers as components of an overall
polymer system. Most testing has been done on plastic films or sheets which
contain the plasticizer as part of the system.
As would be expected, most of the studies are concerned with the changes
in the polymer properties rather than information on the components. It is
known that the epoxides in a polymer system undergo photooxidation near the
polymer surface and that these photooxidation products will not migrate back
into the film. The products remain on the film surface and form a crusty
layer. No information is available on the nature of the specific oxidation
products. The only apparent concern is how these products will affect the
properties and utility of the film.
Information from manufacturers of epoxy compounds shows that these mate-
rials react in much the same manner as would be expected for compounds con-
taining an epoxide group. Lewis acids react readily with the epoxide group
resulting in ring opening and polymerization. If polymerization does occur,
information from manufacturers states that the incidence of a hazardous
reaction is rare. If cross-linking of the epoxy compound is desired, Lewis
acids are common agents for this purpose.
Since triglycerides (i.e. soybean oil and linseed oil) are esters, the
oils are susceptible to acid or base hydrolysis. Acid hydrolysis occurs
rapidly at room temperature. Base hydrolysis, as used for saponification
measurements, occurs readily at approximately 70°C at a pH of about 8. No
rate data on these hydrolysis reactions were available in the published
literature or from manufacturers.
14
-------
LINEAR POLYESTERS
This subsection will present a discussion of the nature of the raw mate-
rials and a discussion of the physical and chemical properties of these plas-
ticizers.
Characterization
Polyester plasticizers are high molecular weight polymers derived from
a dibasic acid and a glycol. The reaction is terminated using either a long-
chain alcohol or a fatty acid in the Ci2-Cig range. This relation can be
shown schematically as follows:
x [H02C-A-C02H] + x[HO-G-OH] --6C-A-C-0-G-0}- + 2H2 0 ,
A
-f C-A-C-0-G-O-}- + 2 ROH -R-0-f C-A-C-0-G-O^-C-A-C-R
x alcohol terminated
-{-C-A-C-0-G-O}- + 2 RCOH -RC ^0-G-O-C-A-C^O-G-O-CR
acid terminated
where A is the dibasic acid, G is the glycol, and T is the terminating alcohol
or acid. Most polyester plasticizers are tailored to function in specific
roles and to have specific physical properties or combination of properties.
Because they are designed for specific purposes, it is very difficult to
precisely characterize each of these plasticizers.
Current polyesters generally employ adipic acid, sebacic acid, or azelaic
acid as the dibasic acid. It is believed that some manufacturers use phthalic
anhydride to lower the cost of the polyesters and improve compatibility but at
a sacrifice in plasticizer efficiency. Propylene glycol, trimethylene glycol,
or butylene glycol are generally employed as the glycols. Ethylene glycol is
seldom used because it produces a solid product. The terminating acids or
alcohols can be isodecyl alcohol, 2-ethylhexanol, stearic acid, oleic acid, or
other alcohols or acids which will produce the desired properties in the poly-
ester. The structures of the more common acids and glycols are as follows:
Dibasic acid
Adipic H02C(CH2)4C02H
Azelaic H02C(CH2)7C02H
Sebacic H02C(CH2)gC02H
Glycol
1,2-Propanediol HOCH2CHOHCH3
1,3-Propanediol HQCH2CH2CH2OH
1,3-Butanediol HOCH2CH2CHOHCH3
1,4-Butanediol HOCH2CH2CH2CH2OH
15
-------
Approximate molecular weights of the polyester plasticizers generally range
from 2,000 to 3,000, but some are as low as about 800 and as high as 8,000.
Physical Properties
Each manufacturer of linear polyesters was contacted for information
relating to the physical properties of their respective products. These
data are compiled in Table IV-2. Generally the test procedures were not
identified by the manufacturer. If identified, the procedures were either
ASTM or AOCS methods. No data were available for properties such as vapor
pressure, solubility in water or organic solvents, octanol-water coefficients,
or others that would provide some insight into the potential for environmental
transport. Chemical Abstracts numbers of the more commercially significant
polyesters and synopses of selected physical properties are presented in the
following subsections.
Chemical Abstracts Service Registry Numbers—
Chemical Abstracts numbers (CAS numbers, CASRN) for those polyesters
which have commercial significance are presented in Table IV-3. Many other
polyesters are listed in compilations such as the TSCA Candidate List of
Chemical Substances but are not employed to any significant extent as plasti-
cizers in plastics. It is estimated by MRI that the materials listed in Table
IV-3 comprise approximately 90% of the total quantity of polyesters currently
used as plasticizers in plastics.
Specific Gravity—
The specific gravity for most of the linear polyesters generally is
within the range of 1.0 to 1.1. This indicates that the majority of these
plasticizers have approximately the same bulk weight as water. Published test
method was ASTM D 2111-71.
Viscosity--
The viscosity of the various linear polyesters shows a wide variance,
ranging from about 130 to 220,000 centipoise. This wide range is expected
because of the variety of molecular weights and the different chemical enti-
ties comprising each of the materials. Identified test method was ASTM D
1638-74.
Acid Number--
As defined earlier in this section, the acid number is a measure of
the free acid functionalities present in the material. For the polyesters,
the acid numbers generally range from ~ 0.5 to 4.0. Some commercial products
show relatively high acid numbers (20-30) indicating large quantities of free
acid groups within the polymer. Published test method was AOCS Cd 3a-63.
Saponification Number--
The saponification number is a measure of the quantity of ester and
acids present in the product. Relatively few numbers were available for
the polyesters. Those values which were reported generally range from 400
to 600. Identified test method was AOCS Cd 3-25.
16
-------
TABLE IV-2. PHYSICAL PROPERTIES OF LINEAR POLYESTER PLASTICIZERS
Nuoplaz6 6186
Nuoplaz 6187
Nuoplaz 6188
Morflex8 P-50
Morflex P-50A
Morflex P-51A
Plastolein1 9717
Plastolein 9720
Plastolein 9730
Plastolein 9731
Plastolein 9734
Plastolein 9746
Plastolein 9750
Plastolein 9761
Plastolein 9765
Plastolein 9772
Plastolein 9775
Plastolein 9776
Plastolein 9780
Plastolein 9783
Approximate
molecular
weight
NAf
NA
NA
3,000
3,000
2,000
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Specific
gravity
(25/25°C)
1.05-1.07
1.085-1.095
1.0-1.03
1.125±0.1h
1. 124+0. lh
1. 11610. Olh
1.02
1.03
1.06
1.06
1.04
NA
1.06
1.06
1.08
1.04
1.08
1.08
1.04
1.08
Refractive
index
(25°C)
1.4640
1.4670
1.4735
1.4709
1.4695
1.4660
1.469
1.462
1.483
1.483
1 . 483
NA
1.477
1.469
1.479
1.486
1 . 465
1.466
1.466
1.465
Viscosity
(cps at
25°C)3
434
2,317
139
12,353
11,343
4,950
260
213
943
933
1,681
NA
908
2,062
3,224
390
1,415
2,778
1,071
1,211
Acid Freezing
number Saponification point
(mg KOH/g) number (°C)
2.0
2.0
2.0
NA
NA
NA
2.5
3.0
3.0
3.0
3.0
2.5
3.0
3.0
3.0
3.0
3.0
2.0
3.0
2.0
continued
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
18
18
45
45
55
NA
32
10
35
16
52
-4
-2
55
Pour
point
(°C)
-32
-18
-34
2
7
-9
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Flash
point ,
(°C; COC)D
266
277
231
288
285
277
232
260
266
266
260
237
279
268
277
271
299
302
299
282
Firr
point ,
(°C; COC)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Color
APHAC
70
70
70
500
500
600
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Lolor ,
i <>
darilner
NA
NA
NA
NA
NA
NA
5
5
6
6
6
5
5
2
6
3
20.0
2
7
5
-------
TABLE IV-2. (continued)
oo
Approximate Specific
molecular gravity
weight (25/25°C)
Plastolein 9789
Plastolein 9790
UltramollJ I
Ultramoll II
Ultraraoll III
Ultramoll PP
Ultramoll TGN
Hercoflex1 900
Santicizer111 412
Santicizer 429
Santicizer 334F
Santicizer 409
Santicizer 411
Staflex0 RS-550
Staflex RS-802
Staflex RS-804
Staflex RS-809
Resoflexq R-296
Resoflex R-446
Resoflex R-460
Resoflex R-804
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
2,000
NA
NA
2,500
1,900
1,900
2,100
NA
NA
NA
NA
1.08
1.08
1. 075-1. 090k
1. 100-1. 115k
1. 100-1. 110k
1. 035-1. 045k
1.090-1.100k
1.22
1.030-1.060
1.080-1.110
1.080-1.084
1.080-1.084
1.104-1.110
1.069P
1.087P
1.075P
1.079P
NA
NA
NA
NA
Refractive
index
(25°C)
1.460
1.46
1.472
1.472
1.469
1.502
1.503
NA
1.453-1.463
1.460-1.470
1.4654
1.4654
1.4772
1.4658(23°C)
1.4819(23°C)
1.4743(23°C)
1.4760(23°C)
1.471
NA
NA
NA
Viscosity
(cps at
25°C)a
17,280
17,280
2,000-3,000
2,000-3,000
1,000-1,300
1,200-1,500
2,000-2,500
NA
240-250
4,300-6,700
3,100-3,800
3,100-3,800
7,900-9,800
2,850
4,450
2,600
4,800
NA
NA
NA
NA
Acid Freezing
number Saponification point
(mg KOH/g) number ,(°C)
5.0
5.0
S 1.0
S 1.0
S 1.0
1 0.5
§ 1.0
20-30
NA'
NA
NA
NA
NA
2.8
1.96
2.2
2.2
32
NA
NA
25
continued
NA
NA
490-510
510-540
510-530
300-320
300-320
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
-20
-20
NA
S -10
S -20
S -35
S -20
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Pour
point
(°C)
NA
NA
NA
NA
NA
NA
NA
NA
-20
-18
3
4
21.1
-12
-15
-15
-12
NA
NA
NA
NA
Flash
point .
(°C; COC)D
304
304
280-300
280-300
270-290
230-265
220-240
238
257
288
277
277
282
280
250
260
260
NA
NA
NA
NA
Fire
point ,
(°C; COC)
NA
NA
NA
NA
NA
NA
NA
NA
282
310
299
299
NA
NA
NA
NA
NA
NA
NA
NA
NA
Color Color ,
APHAC Gardner
NA
NA
NA
NA
NA
NA
NA
NA
250
250
100n
100"
100"
75
80
75
80
NA
NA
NA
NA
4
7
NA
NA
NA
NA
NA
6
' NA
NA
NA
NA
NA
NA
NA
NA
NA
3
NA
NA
4
-------
TABLE IV-2. (continued)
Approximate
molecular
weight
Resoflex R-766
Plasthallr HA7A
Plasthall P-630
Plasthall P-640
Plasthall P-644
Plasthall P-7035 . •
Plasthall P-7092
Plasthall MX-1202
Plasthall P-530
Plasthall P-540
Plasthall P-550
Plasthall P-643
Plasthall P-1070
Plasthall P-7046
Plasthall MX-502
Admex1 433
Admex 515
Admex 522
Admex 523
Admex 525
Admex 529
NA
NA
NA
NA
NA
4,500
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA .
NA
Specific
gravity
(25/25°C)
NA
1.53
1.08
1.09
1.11
1.07
1.11
0.957
1.10
1.11
1.05
1.074
1.068-1.076
1.105
0.995
1.090
1.050
1.060
1.100
1.035
1.122
Refractive
index
(25°C)
NA
1.4662
1.464
1 . 466
1.469
NA
NA
1.467
1.464
1.466
1.463
1.4649
NA
NA
1.4641
1.5050
1.4630
1.5040
1.5140
1.4609
1.4695
Viscosity
(cps at
25°C)a
NA
NA
2,350
5,800
76,000
11,300
24,000
550
2,250
5,750
2,850
2,650
5,000
11,000
420
1,900
575
795
3,960
310
4,150
Acid Freezing
number Saponification point
(rag KOH/g) number (°C)
NA
3.0
0.4
0.4
1.0
NA
NA
1.3
0.3
1.5
0.7
0.5
2.0
0.9
NA
2.0
2.5
2.0
2.0
3.0
3.0
continued
NA
585-592
519
517
548
510
NA .
257
559
558
474
486
455
317
NA
NA
NA
NA
NA
NA
NA
NA
NA
-18
-22
-28
-12
-20s
-32
-15
-22
-41
-5
-22
< -25
-25
NA
NA
NA
NA
NA
NA
Pour
point
(°C)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
-8
2
-21
-7
-34
0
Flash
point ,
(°c; cocr
NA
270
268
274
288
260+
271
172
304
271
279
282
218
266
182
235 .
246
241
232
274
280
Fire
point .
(°C; COC)"
NA
310
304
318
327
260+
.. . 316
198
316
310
332
316
249
310
196
260
291
260
254
291
306
Color
API1AC
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
150
NA
NA
NA
NA
Color ,
Gardner
NA
NA
2-3
< 1
< 1
4
8-9
NA
3-4
2-3
2-3
< 1
3-4
4-5
NA
3
NA
3
3
1
2
-------
TABLE IV-2. (continued)
Admex 752
Admex 760
Admex 761
Admex 770
Admex 775
Admex 890
Paraplex" G-25
Paraplex G-30
Paraplex G-31
N>
° Paraplex G-40
Paraplex G-41
Paraplex G-50
Paraplex G-51
Paraplex G-54
Paraplex C-56
Paraplex G-57
Paraplex G-59
Uniflex" 300
Uniflex 312
Uniflex 314
Uniflex 315
Approximate
molecular
weight
NA
NA
NA
NA
NA
NA
8,000
783
995
6,000
5,000
2,200
2,175
3,300
A, 200
3,450
4,900
NA
NA
NA
NA
Specific
gravity
(25/25°C)
0.975
1.150
1.110
1.110
1.095
1.097
1.06
1.10
1.11
1.15
1.13
1.08
1.11
1.08
1.11
1.099
1.127
1.009
1.076
1.1924
1.105+0.005
Refractive Viscosity
index (cps at
(25°C) 25°C)a
1.4614
1.4700
1.4800
1.4660
1.4670
1.4665
1.470
1.501
1.503
1.471
1.470
1.466
1.464
1.466
1.466
1.4661
1.4699
1.4660
1.4650
1.4791
1.4650
130
117,600
5,330
5,570
6,130
4,890
220,000
1,300
4,800
200,000
110,000
2,300
2,100
5,300
107,000
6,200
25,400
3,330
1,054
5,970
7,293
Acid Freezing
number Saponif ication point
(mg KOH/g) number (°C)
0.5
2.5
3.0
2.0
1.3
2.0
1.4
0.5
0.6
1.4
0.8
1.4
0.8
1.1
0.8
0.8
0.7
2.0
2.0
32.5±2.5
2.0
continued
NA
NA
NA
NA
NA
NA
450
430
426
585
550
500
553
535
562
526
571
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
15
-29
-20
< -18
-22
10
< -23
4
-10
1
7
NA
NA
NA
NA
Pour
point
(°C)
-12
3
2
-1
-18
2
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Flash
point ,
(°C; COC)
285
293
238
293
288
304
316
257
274
288
288
280
> 93
300
310
277
232
293
291
316
279
Fire
point .
(°C; COC)1
307
329
310
316
315
323
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
327
313
338
316
Color Color ,
APHAC Gardner
NA
NA
NA
NA
NA
NA
NA
130
160
180V
100V
NA
150
130
NA
100
NA
NA
NA
150
200
1
3
3
1
1-
1
8-'
NA
NA
NA
NA
4
NA
NA
2-
NA
4
5
2
NA
NA
-------
TABLE IV-2. (continued)
Approximate Specific
molecular gravity
weight (25/25°C)
Uniflex 320
Uniflex 325
Uniflex 327
Uniflex 330
Uniflex 331
Uniflex 337
Uniflex 338
NA
NA
NA
NA
NA
NA
NA
1.085+0.003
1.100+0.003
1.055+0.003
1 . 088+0 . 003
0.961+0.005
NA
0.998+0.005
Refractive
index
(25°C)
1.4660
1.4650
1.469
1.4660
NA
NA
NA
Viscosity
(cps at
25°C)a
2,713
5,830
2,954
5,766
NA
NA
3,693
Acid
number
(mg KOH/g)
2.0
2.0
2.0
2.0
4.0
10
3.5
Freezing
Saponification point
number (°C)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Pour
point
(°C)
NA
NA
NA
NA
NA
NA
NA
Flash
point ,
(°C; COC)b
279
293
263
291
291
NA
288
Fire
point ,
(°C; COO"
307
332
285
318
313
NA
313
Color
APHAC
NA
NA
200
150
NA
NA
NA
Color ,
Gardner
7
5
NA
NA
6+ .
6+x
7
a cps - centipoise; all values at 25°C except as follows: all Plastolein and Nuoplaz at 38°C and all Morflex at 99°C. See ASTM D 1638-74.
b Most test methods were Cleveland Open Cup. Others did not state the specific method used. See ASTM D 92-66.
c Color APHA = Standards based on dilutions of cobaltous chloride, potassium chloroplatinate, and concentrated hydrochloric acid in water (B. Hills, Monsanto
N> Chemical Company). See also ASTM D 1544-68. , , . ,.„ . . ..
•-1 d Color Gardner = A set of 18 solutions; the lightest standard (No. 1) is equal to 0.550 g/liter of potassium chloroplatinate. Each solution is 50% darker than.
the preceding solution (Gardner and Sword, 1962). See also ASTM D 1554-68.
e Tenneco Chemicals.
f NA = Not available.
g Pfizer Chemicals.
h @(20/20°C).
i Emery Industries.
j Mobay Chemicals.
k Density (g/cm3).
1 Hercules, Incorporated.
m Monsanto Industrial Chemicals Company.
n 50:50 in 95% ethanol.
o Reichold Chemicals.
p @(23/15.5°C).
q Cambridge Industries
r The C. P. Hall Company
s This is the point when the liquid reaches a viscosity of a stiff paste.
t Sherex Chemical Company, Incorporated.
u Rohm and Haas Company.
v 50% solids.
w Union Camp Corporation
x 1% in xylene.
-------
TABLE IV-3. CHEMICAL ABSTRACT NUMBERS FOR POLYESTERS
Compound
Chemical
Abstract No.
Adipic acid, polyester with 1,2-propanediol
Adipic acid, 1,3-butylene glycol, 2-ethylhexanol polymer
Adipic acid, 1,3-butylene glycol, palmitic acid polymer
Adipic acid, 1,3-butylene glycol polymer, 2-ethylhexyl
ester
Adipic acidj 1,3-butylene glycol polymer, isodecyl
alcohol modified
Adipic acid, butylene glycol polymer, isodecyl ester
Adipic acid, palmitate, 1,3-butanediol polymer
Adipic acid, phthalic anhydride, 1,3-butylene glycol,
1,2-propylene glycol polymer stearate
Adipic acid, phthalic anhydride, dipropylene glycol resin
Adipic acid, phthalic anhydride, 1,2-propanediol,
1,4-butanediol polymer, caprylate, caprate
Adipic acid, phthalic anhydride, 1,2-propylene glycol
polymer, diisodecyl ester
Azelaic acid, polyester with 2,2-dimethyl--l,3-propanediol
Azelaic acid, adipic acid, propylene glycol, polymer,
2-ethylhexyl ester
Azelaic acid, propylene glycol polymer
Sebacic acid, polyester with 1,3-butanediol
Sebacic acid, polyester with diethylene glycol
Sebacic acid, polyester with 1,2-propanediol
25101-03-5
63149-79-1
30918-39^9
69029-19-2
68441-97-4
69029-22-7
69029-21-6
68238-78-7
9011-80-7
68890-79-9
68511-08-0
29408-58-0
68071-01-02
29408-67-1
28606-47-5
25610-21-3
26222-20-8
22
-------
Flash Point-
Flash point is the temperature at which a liquid evolves a vapor suf-
ficient to form an ignitable mixture with the air near the surface of the
liquid. Most of the tests for the polyesters used the Cleveland Open Cup
method. Flash points for most of the linear polyesters were within the
range of approximately 230 to 300°C. Published test method was ASTM D
92-66.
Chemical Properties
No studies were found in the literature which were directly related to
the chemical properties of linear polyesters in the pure state. As with
the epoxy compounds, considerable work has been performed on these materials
as components of an overall polymer system. Most testing is done on plastic
sheets or film containing polyesters as the plasticizer.
Little information was obtained from the manufacturers and all of it
was in general terms with no specific data. Linear polyesters are suscep-
tible to base hydrolysis, usually above pH 8 and at approximately 70-80°C.
.These compounds, in general, were described by the manufacturers as being
relatively stable towards acid hydrolysis and exhibiting good thermal stabil-
ity. No experimental data or detailed information on acid hydrolysis and
thermal stability for specific products was available from the manufacturers.
Polyesters are generally unreactive, except as noted above, with other
chemicals which would result in degradation of the plasticizer. These mate-
rials do not undergo hazardous polymerization. Material Safety Data Sheets
supplied by the manufacturers state that polyesters do not decompose at high
temperatures to produce any hazardous products except those products nor-
mally associated with the burning of organic compounds (CO, C02, etc.).
Ultraviolet light will degrade the polyesters in vinyl film, but no data
are available on the pure polyester or products of the degradation.
TRIMELLITATES
This subsection will present a discussion of the nature of the raw ma-
terials and a discussion of the physical and chemical properties of these
plasticizers. '
Characterization
Trimellitates are esters derived from trimellitic acid (1,2,4-benzenetri-
carboxylic acid) and alcohols in the range of Cy to CIQ. During actual pro-
duction of these esters, trimellitic anhydride is used instead of the acid.
A generalized trimellitate, or trimellitic acid ester, is as shown:
to
23
-------
In terms of characterization, this class of plasticizer is different from
either the epoxy compounds or the linear polyesters in that trimellitates
represent compounds with definable molecular weights. While many of these
materials are esters resulting from 2-ethylhexanol, isooctanol, 1-methyl
heptanol, or decanol, mixtures of Cy to Cg and Cg to CIQ alcohols are com-
monly used. In this case, the resultant products will be a mixture of
esters.
Physical Properties
Each manufacturer of trimellitates was contacted for information relating
to the physical properties of their respective products. These data are com-
piled in Table IV-4. In general, the test methods were not identified. Some
manufacturers identified the procedures as ASTM methods. In contrast to the
two previous classes of plasticizers, only slightly more than 50% of the
manufacturers responded with data on physical properties. For the other two
classes, the response was generally greater than 90%.
The few data reported show that, as a class, trimellitates exhibit low
vapor pressures even at elevated temperatures. At temperatures of 250-260°C,
vapor pressures generally range from 0.2 to 5 mm Hg; at room temperature,
the pressures were usually stated to be negligible. Solubility of these
esters in water at approximately room temperature ranges from negligible to
about 0.1%.
Chemical Abstracts numbers of the commercially significant trimellitates
and synopses of selected physical properties are presented in the following
subsections.
Chemical Abstracts Service Registry Numbers--
Chemical Abstracts numbers (CAS numbers, CASRN) for those trimellitates
which have commercial significance are presented in Table IV-5. Other tri-
mellitates may be listed in compilations such as the TSCA Candidate List of
Chemical Substances but were not among the products for which data were sup-
plied by producers or distributors.
Specific Gravity—
The specific gravity for the trimellitates generally fell within the
range of 0.97 to 0.99. This is slightly lower than the values observed for
the linear polyesters, epoxidized soybean oil, and epoxidized linseed oil.
The trimellitates have a bulk weight slightly less than water. Referenced
method was ASTM D 1045-80.
Viscosity--
The viscosity of the trimellitates ranged from about 50 to 300 centipoise.
This is considerably less than for the linear polyesters and epoxidized linseed
oil but about the same as that for epoxidized soybean oil. The octyl epoxy-
tallates have the lowest viscosity values.
Acid Number--
Acid number is a measure of the free acid functionalities present in the
product. For trimellitates, the acid numbers range from 0.05 to 0.1. This
range is lower than the ranges for any of the other classes of plasticizers
and indicates a low level of free acid groups in the finished product.
24
-------
TABLE IV-4. PHYSICAL PROPERTIES OF TRIMELLITATE PLASTICIZERS
to
Ui
Approximate Specific
molecular gravity
weight
PX-337d 546
PX-338 546
PX-336 NA
HATCOL TOTM8 546
NUOPLAZ TOTM1 546
KODAFLEX TOTM-" 547
STAFLEX TOTMm 547
STAFLEX TIOTM 547
STAFLEX NONDTM 578
Rucoflex 26TM° 546
Rucoflex NTM 560
Rucoflex C7-C9 NA
Uniflex TOTMq NA
Uniflex TCTH NA
Santicizer 79TMr 548
a cps = centipoise = '
(25/25°C)
0.987
0.987
0.972
0.987
0.986-0.992
0.989k
0.986"
0.986"
0.974"
0.992k
0.983k
0.990k
0.991k
0.977k
0.982-0.987
Refractive
index Viscosity
(25°C) (cps
1.485
1.485
1.482
1.4848
1.4850
1.4832
1.4855
1.4852
1.4830
1.4846
1.4829
1.4845
1.4830
1.4780
1.481-1.484
at 25°Cr
220
216
103
300h
NA
194
210
200
90
244
100
NA
213
242
113
^ x 100; at 20°, the viscosity of water
X Cflfl
b Test methods were Cleveland Open Cup
c Color APHA = Standards
chloroplatinate, and
Monsanto Chemical Co.
d USS Chemicals.
e NA = Not available.
f @ 1 mm Hg
. See ASTM D 92-66
based on dilutions of cobaltous
concentrated
. ). See also
chloride,
hydrochloric acid in water (B
ASTM D 1209-79
Acid Acidity as . Acidity Ester
number trimel Litic as acetic content
(mg KOH/g)
NAe
NA
NA
NA
0.1
NA
NA
NA
NA
0.05
0.05
0.05
0.1
0.1
NA
= 1.0 cps.
potassium
. Mills,
acid (%)
NA
NA
NA
NA
NA
0.02
0.010
0.010
0.015
0.02
0.02
NA
NA
NA
NA
g Hatco Chemical Corporation.
Boiling
point
Freezing
point
Pour
point
acid (%) (% min) (3mm Hg; °C) (°C) . (°C)
0.01 99.0
0.01 99.0
0.007 99.0
0.01 99.0
0.1 99.0
NA 99.0
NA NA
NA NA
NA NA
NA NA
NA NA
0.017 NA
NA NA
NA NA
NA NA
h @ 20°C
i Tenneco Chemicals.
j Eastman
k @ 20/20°C.
1 @ 760 mm Hg.
m
n @ 23/15. 5°C.
o Ruco, Div. Hooker
p @ 1 . 5 mm Hg
260f
260f
275f
283
283
600 l
311
327
335
283
289P
NA
NA
NA
263s
Chemicals
Reichold
Chemicals .
NA
NA
-17
NA
-46
-38
NA
NA
NA
-35
-45
NA
-35
-35
NA
Chemicals ,
-40
-46
NA
-45
-32
NA.
-45
-48
-7
NA
NA
NA
NA
NA
-50
Flish
point , Color
(°C; COC) APHA
254
254
278
260
260
263
258
2r,8
266
257
260
NA
254
246
263
100
100
. 75
75
75
NA
75
75
80
150
200
150
100
100
100
Incorporated
q Union Camp Corporation.
r Monsanto Company.
s @ 10mm Hg.
-------
TABLE IV-5. TRADE AND CHEMICAL NAMES OF TRIMELLITATES
Trade name
Chemical name
CAS number
PX-336
PX-337
PX-338
HATCOL TOTM
NUOPLAZ TOTM
KODAFLEX TOTM
STAFLEX TOTM
STAFLEX TIOTM
STAFLEX NONDTM
RUCOFLEX 26-TM
RUCOFLEX NTM
RUCOFLEX C7-C9
UNIFLEX TOTM
UNIFLEX TCTM
SANTICIZER 79-TM
n-octyl, n-decyl trimellitate 34870-88-7
triisooctyl trimellitate 27251-75-8
trioctyl trimellitate 89-04-3
tri(2-ethylhexyl)trimellitate 3319-31-1
tri(2-ethylhexyl)trimellitate 3319-31-1
tri(2-ethylhexyl)trimellitate 3319-31-1
1,2,4-benzenetricarboxylic acid; 2-ethylhexyl 68186-31-2
ester
1,2,4-benzenetricarboxylic acid; isoctyl ester 68186-32-3
1,2,4-benzenetricarboxylic acid; octyl, decyl 34870-88-7
ester
tri(2^ethylhexyl)trimellitate 3319-31-1
tri(mixed-n-alkyl)trimellitate not listed
tri(C-7, C-9 alkyl)trimellitate 68515-60-6
tri(2-ethylhexyl)trimellitate 3319-31-1
tri(l-methylheptyl)trimellitate not listed
tri(C-7, C-9 alkyl)trimellitate 68515-60-6
26
-------
Flash Point—
For the trimellitic acid esters, the flash points are in the range of
245 to 280°C. This range is lower than that for epoxidized soybean oil and
linseed oil and approximately the same as for the linear polyesters. Octyl
epoxytallates had the lowest flash point range. Measurements for the tri-
mellitics were by the Cleveland Open Cup method. Procedure stated to be ASTM
D 92-66.
Chemical Properties
No studies were found in the literature which were directly related to
the chemical properties of trimellitates in the pure state. In contrast to
the other two classes of plasticizers, little information was found on chemi-
cal properties of these materials as components of an overall polymer system.
In general, the data supplied by the manufacturers and distributors
were approximately the same as for the epoxy compounds and the linear poly-
esters. Trimellitates are incompatible with oxidizing agents and nitric acid.
They are hydrolyzed by alkali, but usually temperatures of about 100°C are
required. The trimellitates are more resistant to acid hydrolysis by non-
oxidizing acids than to hydrolysis by alkali. Manufacturers would provide
no experimental details on the alkali or acid hydrolysis data. No hydrolysis
of these esters was found after 96 hr in boiling water.
Trimellitates exhibit little additional incompatibility with other chemi-
cals which would result in degradation of the plasticizer. These materials
do not undergo hazardous polymerization. In general, the compounds show
good thermal stability with only 0.04% decomposition to the corresponding
acid after heating the pure compound for 2 hr at 200°C. However, at higher
temperatures, more rapid dissociation to the corresponding acid may occur,
which can present potential health problems for local personnel. This aspect
will be discussed more thoroughly in the section concerned with health effects
(Section VI).
27
-------
SECTION V
PRODUCTION AND USE
This section contains information on the production locations, quanti-
ties manufactured, production methods, plasticizer uses, estimated worker
exposure, occupational standards, and environmental transport and degrada-
tion. Information is provided for each of the three classes of plasticizers
included in this study.
EPOXY COMPOUNDS
Producers and Quantities
A listing of the manufacturers of epoxidized plasticizers, their produc-
tion locations, and production method is shown in Table V-l. The largest
volume producers of this class of plasticizer are Union Carbide Corporation,
Witco Chemical Corporation, and Rohm and Haas Company. These three producers
account for approximately 75% of the total production capacity for the indus-
try, which is an estimated 245 million pounds per year.
The estimated annual U.S. production quantities for epoxidized plasti-
cizer are presented in Table V-2 for 1975-1979. During that time period,
the overall production trend was towards increased production of this class
of plasticizer. Industry sources estimated that the data in Table V-2 are
accurate to + 10%. However, the production of 130 million pounds in 1979
represents only approximately 50% of the total operating capacity of the
industry (Sigan, 1980). Preliminary data in Modern Plastics indicated that
for 1980 the estimated consumption of epoxy plasticizers will be about 121
million pounds, which is a decrease of approximately 9 million pounds from
1979. Manufacturers indicated that annual production is approximately equal
to consumption and that little stockpiling occurs.
Production Process
The principal epoxidized plasticizer products currently produced in
the United States are derived from vegetable oil substrates (e.g., soybean
oil) or from alkyl fatty acid ester substrates, (such as octyl esters of
tall oil fatty acids (Lutz, 1980).
Epoxidation is defined as a chemical reaction in which double bonds in
unsaturated (e.g., olefinic) compounds are converted to cyclic three-membered
ethers by an active oxygen agent. This is shown schematically the following
equation (Thompson, 1977).
28
-------
TABLE V-l. MANUFACTURERS OF EPOXY PLASTICIZERS
Company Production site Production method
Rohm and Haas Co. Philadelphia, PA P'F'h
Union Carbide Corp. Taft, LA P.A.
Witco Chemical Corp. Taft, LA P.A.
Viking Chemical Co. Blooming Prairie, MN P.A.
FMC Corporation Baypprt, TX P.A.
Ferro Chemical Co. Bedford, OH P.F.
Sherex Chemical Co. Mapleton, IL P.A.
Swift Specialty Griffith, IN N.A.
Chemical Co.
Q
, P.F. = Performic acid
P.A. = Peracetic acid
N.A. = not available
TABLE V-2. ANNUAL PRODUCTION OF EPOXY PLASTICIZERS
(million pounds)
1975 1976 1977 1978 1979
Plasticizer ITC" MP ITC MP ITC MP ITC MP ITC MP
Total Epoxy
Soybean Oil
Linseed Oil
All others8
97. 6C
77.6
N.A.
20.0
117
-
-
-
117
91
6
19
.4
.4
.4
.6
117
-
-
-
120.5
92.5
5.2
22.8
119
-
-
-
120
89
6
23
.2d
•9f
.4 .
.9
126
-
-
-
N.A.e 130
-
-
-
, ITC - International Trade Commission data.
MP = Modern Plastics data.
r*
, Does not include epoxidized linseed oil.
Sum includes linseed oil sales data.
,. N.A. = not available.
Data are for sales of epoxidized linseed oil.
° Primarily epoxytallate esters.
29
-------
R-C=t-R >• R-6JS-I
The principal oxygen sources for this conversion are peracetic acid or per-
forraic acid. Both peracids result from treatment of the normal acids with
hydrogen peroxide (Thompson, 1977).
In the United States, epoxidized plasticizer materials are produced
primarily by reacting the unsaturated substrate with either peracetic acid
(peroxyacetic acid) or with performic acid (peroxyformic acid) (Lutz, 1980).
Peracetic acid can be prepared by the oxidation of acetaldehyde with hydro-
gen peroxide. Performic acid can be prepared by a similar oxidation of form-
aldehyde. Although the use of performic acid is discussed in the literature,
peracetic acid is the principal reagent in the use at the present time.
Epoxidation processes used in the United States can be divided into
two basic types: either the peracid is preformed or it is formed in situ,
(i.e., in a primary reaction vessel) (Lutz, 1980). At present epoxy plasti-
cizer production in the United States is roughly evenly divided between these
two processes. Batch production methods are used extensively for both pro-
cesses. Each process has its own advantages and disadvantages. The product
composition and performance can be affected by proprietary processes involving
peracid formed in situ or preformed peracid using co-solvents, especially
selected olefinic substrates and catalysts, methods of addition of components,
and post-treatment of the epoxide (Lutz, 1980).
Figure V-l shows a representative schematic of the general process flow
arrangement used for each of the two basic production processes. A discus-
sion of these process methods follows.
Epoxidation with Preformed Peracid—
Peracid is formed as the first step (peroxidation) in this process, as
shown in Figure V-la. For example, peracetic acid can be prepared from acetic
acid and hydrogen peroxide according the following reaction:
+ H20
Water
In this process, using preformed peracetic acid, a catalyst is not nec-
essary for the epoxidation step and the reaction can be conducted at 20° to
80°C according to the reaction:
CH3COOH + RCH=CHR HRM-^HR + CH3COH
Peracetic Substrate Epoxy Acetic acid
acid compound
H202 +
Hydrogen
Peroxide
OH j
Acetic
acid
H
•rH«rn<>H
Peracetic
acid
30
-------
RCOOH H2O2
i r
*
H2SO4 RCOOC
Reaction Aqueous
Products Peracid
1
Epoxidat'
RCOOH
i
Epoxy Com
i
1 Strippe
Water, Residues
/ •_
/
ion |«« — H2SO4
)H
*(C=C)XR
Unsaturated
on Substrate
f" "• Weak Base
DOund
H2SO4 H2O2
•» 1 '
*
*
Organic
Acid
RCOOH
• 1
Solvent
(Optional)
1 r R(C=C)XR
f ? Unsaturated
Residual Epoxidation Substrate
RCOOH
Residual
HoOo '
1 Epoxy Compound 1
1
r 1 Stripper
Water,
Salts,
ter Residues Filter
T
Waste
Product
a. Preformed Acid Process
Source: Midwest Research Institute
Waste
Product
b. In Situ Process
Figure V-l. Schematic of the two basic epoxidation processes.
-------
By-product acetic acid is recycled to the peroxidation step. The epoxy com
pound is combined with a weak base and treated in a stripper to separate
water and residues as waste materials. The crude product is then filtered
to yield the final purified product. A major disadvantage of this process
is the fact that the preformed acid presents an explosion hazard and cannot
be safely stored (Lutz, 1980).
Epbxidation with Peracid Formed in situ —
Many production techniques for i.n situ epoxidation have been developed.
In general, a peroxide solution (35 to 70% of hydrogen peroxide in water)
containing a small amount of a mineral acid catalyst (e.g., sulfuric acid
or phosphoric acid) is added to a mixture of an epoxidizable substrate and
acetic acid or formic acid. As the reactants are mixed, hydrogen peroxide
and the organic acid react in the presence of the acid catalyst to form the
peracid. An example, with formic acid and hydrogen peroxide as reactants,
is shown in the following reaction:
H202 + HCOH - - - -CH203 + H20
Hydrogen Formic Performic Water
Peroxide acid acid
The peroxide solution is added incrementally with agitation to prevent an
uncontrolled exothermic reaction. The reaction temperature is maintained
at 50° to 65°C for 10 to 40 minutes per addition of peroxide. Using this
operating procedure, only small amounts of peracid are formed in the pres-
ence of the unsaturated substrate. Since the peracid reacts with the un-
saturated portion of the molecule, the peracid is quickly depleted and a
buildup of detonatable quantities of peroxide compounds is avoided.
After the iodine number of the substrate is reduced to a predetermined
level, the reaction is stopped and the epoxidized substrate is separated
from an aqueous layer. This aqueous layer, which contains a mixture of or-
ganic peracid and some hydrogen peroxide, can be recycled to the next batch
as part of the charge. In the epoxy material, the acid catalyst is neutral-
ized with a mild base, and residual peroxide is decomposed. The crude epoxy
compound is then washed and transferred to a stripper for removal of water
and nonproduct residues. Following the stripping step, the epoxy compound
is purified in a filtration step to obtain the finished product.
As a process option, solvents compatible with the substrate can be used
to facilitate epoxidation. For example, heptane or octane serve well with
soybean oil substrate and aromatic solvents (e.g., toluene) can be used with
linseed oil (Lutz, 1980).
Process Modifications--
There are a number of modified production processes based on the two
general processes. A brief discussion of two of the most common is presented
in the following paragraphs.
32
-------
Processes using sulfuric acid as catalyst—In. situ production methods
for epoxidation of soybean oil based on use of acetic acid with sulfuric
acid as catalyst have been developed by the Archer-Daniels-Midland (ADM)
Company and the FMC Corporation (Lutz, 1980).
In the ADM process, the catalyst is added last and is admixed with gla-
cial acetic acid. Epoxy ring opening by the sulfuric acid is minimized since
the system is heterogeneous and involves the interaction of an oil phase
and an aqueous phase containing acetic acid and hydrogen peroxide.
In the FMC modification of the basic in situ process, an inert solvent
(e.g., hexane) is used to reduce the effect of sulfuric acid in catalyzing
epoxy ring opening.
Repeated-resin process—In this process, a relatively large quantity
of poly(styrenesulfonic acid) resin is used as catalyst; however its reuse
is required in succeeding epoxidation batches to achieve good economics.
Advantages claimed for this process include high epoxy yields, little by-
product formation, almost complete elimination of unsaturation, low reaction
temperatures (60°C), and short reaction periods (Lutz, 1980).
The process involves mixing the fatty oil or ester, glacial acetic acid
acid, and dry resin. Hydrogen peroxide is added slowly so that a reaction
temperature of 60°C is not exceeded. The reaction medium is maintained at
the maximum temperature for about 4 hr and then separated from the resin
catalyst by decantation or filtration. The resin catalyst remains in the
reactor for succeeding runs. The catalyst can be reused for approximately
6 to 8 runs. With each succeeding run, degradation of the catalyst produces
fine particles which may introduce problems in the filtration procedures.
If products with maximum epoxy oxygen values are not required, the gen-
eral practice is to use much less resin. A smaller amount of resin can be
economically discarded following each run. This procedure is termed the
minimal-resin technique.
By-Products and Contaminants--
The only major contaminants in the product plasticizer are by-products
formed during the process reaction. Since the epoxidation process is revers-
ible, there is a potential for the occurrence of undesirable side reactions.
Although the epoxidation reaction is generally conducted at the lowest temper-
ature and shortest time consistent with the desired product, the following
side reactions can occur (Lutz, 1980).
?CH_-tl
H+
CH3COOH
H20
CH3COOOH
H202
HO OOCCH3
Rrwrvn? x
H9 OH J
Rpuptro *c
JJCH- R
HO OOOCCHs
-^RfHCHR
HO OOH
*Rf!Hr.HR
H?0
33
-------
The specific composition of the by-product mixture will vary considerably
depending upon the specific reaction process, reaction temperature, pressure,
and other variables. According to a major epoxy manufacturer, the two by-
products at the top, the hydroxy ester and the glycol, would probably consti-
tute the major components of most by-product mixtures. If the by-products
are not volatile (almost none are volatile), they are processed with the
epoxy compound and will remain in the final product. For most reaction pro-
cesses, the approximate quantity of by-product found is less than 5% of the
total product material.
Production Losses—
Industry sources estimate that the overall loss of plasticizer due to
processing and transfer is approximately 2% of the total production quantity.
The losses occur during four processes: stripping, filtering, neutralization,
and washing. The largest contribution to product loss probably occurs in
the filtering process and the next largest is in stripper losses. No informa-
tion is available concerning the actual quantities lost by each process.
Losses resulting from neutralization and washing are estimated to be very
small and probably less than losses from the stripper. The filter material
containing the epoxy plasticizer is likely disposed in a landfill.
Uses
Epoxy plasticizers are usually employed as secondary plasticizers, which
means they are always used in conjunction with other plasticizers to provide
specific properties to the finished plastic or to perform certain functions
within the plastic matrix.
Information from manufacturers varied with respect to the percentage
utilization of epoxy plasticizers. No published data were available regarding
the quantities of epoxy compunds used in specific products. The following
estimate represents a combination of data from several manufacturers, not
the views of any single company.
Quantity in 1979
Use Category % (million pounds)
Polyvinyl chloride 85-90 111-117
Other polymers 8-13 10-17
Miscellaneous 2 3
Other polymeric systems included nitrocellulose, chlorinated rubber, chlori-
nated polyethylene, and acrylics. Miscellaneous uses include adhesives,
sealants, pesticide formulations, and a myriad of uses employing small vol-
umes. Usage of epoxy plasticizers in adhesives and sealants is for purposes
of stabilization; very small quantities are employeed in this area. No pub-
lished information was found relating to uses of epoxy compounds in specific
products of polyvinyl chloride (PVC). Producers stated that epoxy compounds
are used to impart good flexibility properties, particularly at lower tempera-
tures, and good hydrogen chloride stability. One specific area of use for
epoxy compounds is with flat vinyl film or sheet, including film for food
wrapping.
34
-------
The use of epoxy compounds as secondary plasticizers can be exemplified
by considering the change in formulation which would occur. A standard PVC
formulation for a flexible film and a formulation employing a secondary plas-
ticizer are as shown:
Standard Formulation
Formulation with Epoxy
PVC resin
: 100 parts
Plasticizer (e.g. DOPa): 50 parts
Stabilizer : 3 parts
PVC resin : 100 parts
Plasticizer (eg. DOP) : 45 parts
Epoxy Compound : 5 parts
Stabilizer : 3 parts
DOP=dioctyl phthalate
Epoxy plasticizers are very good acid stabilizers and react readily
with hydrogen chloride which is generated in the PVC as the plastic degrades.
As stated in the discussion of chemical properties (Section IV), epoxide
rings are quite susceptible to ring opening due to acid attack. The action
of hydrochloric acid, or hydrogen chloride, on the epoxide ring can result
in formation of the corresponding hydroxy chloro compound or formation of a
cross-linked chloro ether, as shown in the following equation:
I
— C-C- or -i-C-
Cl Cl
HC1
If the ether cross-linkage becomes extensive within the plastic, loss in
flexibility and production of a brittle product could result.
Quantities Released During Processing—
Essentially all processes used with the high volume production of for-
mulated PVC resins are highly automated. Wilkinson, et al. (1979) performed
an extensive study of the production and use of selected aryl and alkyl aryl
phosphate esters. In this study, no evidence was found of any consistent
loss of plasticizer during resin formulation. Losses due to equipment mal-
function or breakage occur but are very difficult to quantify. Since both
phosphate esters and epoxy compounds are used primarily with PVC resins and
both are liquids, the processing methodologies should be very similar.
Ultimate Disposal—
Although the useful life of plastics varies considerably from one product
to another, most plastic products will be discarded within a relatively short
period of time (a few years). These plastics will become solid waste and
subjected to either incineration or landfill. The very low volatility of
the epoxy plasticizers and their very low migration from plastics would pre-
clude any appreciable loss of the plasticizer from the product during its
use. However, plasticizers can migrate from one plastic to another if their
solubility in the second plastic is greater.
35
-------
It has been estimated that of the phthalate plasticizer containing plas-
tics that are disposed of as solid waste, 10 to 20% are destroyed by high
temperature incineration and 2% are subjected to low temperature incineration
or open burning (Peakall, 1975). The remainder of the plastic would be dis-
posed in a landfill. Since epoxy compounds are employed as secondary plasti-
cizers, often with phthalates, these disposal figures would appear to be
valid for this class of plasticizer. Within the landfill, the epoxy plasti-
cizers are subject to rapid attack by fungi and bacteria (see Section VI,
Environmental Effects). The exact lifetime of the epoxy compounds in the
landfill is unknown, but probably is less than 2-3 years based on the micro-
organism degradation studies reported in Section VI.
Worker Exposure and Occupational Standards
Occupational Standards--
No information was found concerning any occupational standards for epoxy
plasticizers. Sources searched for these data were the National Institute
for Occupational Safety and Health (1978a, 1978b), American Conference of
Governmental Industrial Hygieiiists (1979), and Sittig (1979).
Worker Exposure--
The National Institute for Occupational Safety and Health (1980) has
estimated the total number of workers exposed to certain chemicals in plants
of selected industries. This National Occupational Hazard Survey (NOHS)
estimated that a total of 1,257,364 workers covering 202 occupations were
exposed to one or more plasticizers. NOHS estimated that a total of 264,112
workers covering 177 occupations were exposed to epoxy plasticizers. The
individual data are as follows (number of workers rounded as appropriate by
MRI):
Number of Estimated Workers
Epoxy Compound Occupations Exposed
Epoxidized oils 100 232,700
Epoxyesters 57 28,000
Epoxidized butyl oleate 18 3,400
Epoxytallates 2 70
Total 177 264,170
It is entirely possible that overlap has occurred in the number of occupa-
tions and therefore in the total number of exposed workers. However, it
would be extremely difficult to resolve this overlap because insufficient
information is presented in the survey.
Environmental Transport and Degradation
Very few quantitative data were found for properties which would be
directly related to environmental transport and bioaccumulation. Epoxy plas-
ticizers have very low vapor pressures at ambient conditions so that volatil-
ity would not be a major factor in their transport through the environment.
In addition, these plasticizers exhibit a very low solubility in water. No
36
-------
information was available on mobility in soils or sediments. No data were
available for octanol-water partition coefficients so that no inferences
can be made with respect to bioaccumulation or biomagnification.
Environmental degradation can occur by photolysis, hydrolysis, and bio-
logical methods. As discussed in Section IV, no photolytic data are available
for pure epoxy compounds, but plastics containing these plasticizers were
subjected to photolytic degradation. It was determined that the epoxy plasti-
cizer was being degraded but no products were identified. In the same section,
the conditions for hydrolysis were also discussed. The effects of fungi and
bacteria on this class of plasticizer are discussed in Section VI, Environ-
mental Effects.
LINEAR POLYESTERS
Producers and Quantities
A listing of the manufacturers of linear polyester plasticizers and
their production sites, as supplied by the companies, are shown in Table
V-3.
TABLE V-3. MANUFACTURERS OF POLYESTER PLASTICIZERS
Company Production site
Emery Industries, Inc. Cincinnati, OH
Rohm and Haas Company Philadelphia, PA
Knoxville, TN
Monsanto Company Everett, MA
Reichold Chemicals, Inc. Carteret, NJ
C.P. Hall Company Chicago, IL
Pfizer, Inc. Greensboro, NC
Union Camp Corporation Dover, OH
Cambridge Industries N.A.
Sherex Chemical Co. Mapleton, IL
N.A. = not available.
o
Believed to have discontinued production at this site in
1980.
The largest volume producers are Emery Industries and Rohm and Haas. Industry
sources estimate that the combined market shares of these two companies prob-
ably account for approximately 60% of the total market with the remaining
seven companies sharing 40% of the market. Production capacity data are
very difficult to ascertain because the plants are multipurpose facilities
capable of producing many types of ester plasticizers.
37
-------
The estimated annual U.S. production levels for polyester plasticizers
are presented in Table V-4 for 1975-1979. One industry source stated that
the International Trade Commission production figures for adipic acid type
should actually be 37.5 and 42.0 million pounds, respectively, in 1977 and
1978. This source also estimated that only approximately 40-45 million pounds
of the polyester during 1977 to 1979 is being use for plasticizer purposes
and the remainder is starting material in urethane foams.
TABLE V-4. ANNUAL PRODUCTION OF POLYESTER PLASTICIZERS
(million pounds)
Plasticizer
Total Production
a
b
Adipic Acid Type
All Other
ITC = International
MD — M/i/lov^ Dloff--;^
ITC
38.
-
1975
4
Trade
V,
MPb
46
-
1976
ITC
52.9
33.3
19.6
Commission
1977
MP
51
-
data
ITC
48.
10.
37.
MP
0 53
5 -
5 -
1978
ITC
54.
12.
42.
. MP
2 53
2 -
0 -
1979
ITC MP
N.A.C 55
-
N.A. = not available
Preliminary data in Modern Plastics indicate that for 1980, the esti-
mated consumption of polyester plasticizers will be about 48.5 million pounds,
which represents a decrease of approximately 6 million pounds from 1979.
Manufacturers indicate that annual production is approximately equal to con-
sumption and that little stockpiling occurs.
Production Process
Linear polyesters, which are also known as polymeric plasticizers, are
produced by the polymerization/esterification of an aliphatic dicarboxylic
acid, or acid anhydride, with either pure or mixed aliphatic alcohols (glycols),
The dicarboxylic acids commonly used include adipic, sebacic, and azelaic.
Phthalic anhydride is believed to be used in certain instances. Glycols are
usually propylene glycols or butylene glycols. Ethylene glycol is seldom used
because it gives a solid product. Terminating alcohols and acids include
stearic acid, oleic acid, 2-ethylhexanol, isodecanol, and many others.
The method of production stems from the basic esterification reaction,
but is more complex. In general, the reaction is between a dibasic acid
and a dihydric alcohol, usually at an elevated temperature (e.g., ^ 200°C).
Variations in the reactants used, rather than the mode of preparation, deter-
mine the structure and properties of the polyester. An excess of dihydric
alcohol gives a polyester with terminal hydroxyl groups, whereas an excess
of dibasic acid gives terminal carboxyl groups. Thus, the chain length of
the product can be varied by changing the proportions of the reactants.
The use of a monohydric alcohol or a monobasic acid as a third reactant lim-
its the chain growth and provides esters which are termed "alcohol end-stopped"
38
-------
and "acid end-stopped", respectively. If equiraolar amounts of acid and alco-
hol are used, the reaction is non-end-stopped and usually leads to the produc-
tion of higher molecular weight polymers. This condition, along with an
increase in viscosity, can be achieved by prolonged heating. The average
molecular weight of common ester products is about 2,000. Products with
average molecular weights of 800 to well over 6,000 are produced for spe-
cialty applications.
The polymeric esters have greater permanence as plasticizers than do
monomeric plasticizers (e.g. dioetyl adipate) because they have very low
volatility, low extractability by solvents, and low tendency to migrate into
other organic materials. Thus, polymeric plasticizers have found wide com-
mercial acceptance for those applications in which durability and stability
are primary considerations.
Most polymeric plasticizers are synthesized by simple esterification
reactions, which can be conducted in the liquid phase using heated reaction
vessels with stirring and water take-off provisions (Thompson, 1977). The
production operations are conducted using batch processess.
A generalized process flow diagram for production of linear polyesters
is shown in Figure V-2. The feed materials, consisting of a dicarboxylic
acid and an aliphatic dihydric alcohol (glycol) are fed into a esterification
reactor. The reaction is usually conducted at a temperature of about 200°C.
In some processes a catalyst (e.g., sodium acetate) may be used. An excess
(30 to 40%) of glycol is used in carrying out this reaction. A fractionat-
ing column is operated in conjunction with the reactor to separate by-product
water and unreacted glycol from the reaction mixture. Separated water is
sent to waste treatment and disposal; the recovered glycol is recycled to
the esterification step. The crude esterification product, which has a low
acid number, is sent to a polymerization step.
The polymerization step is carried out by a suitable combination of
applied heat and high vacuum to accomplish an ester interchange. The process
temperature is normally held at about 200°C. Some of the initial reaction
product consists of an undesirable material with the generalized designation
of T-G-T, in which G is the glycol and T is a monobasic terminator acid.
The purpose of this step is to convert the T-G-T material and other low molec-
ular weight esters to a desired crude product with the formula of T(G-A-G) T,
in which A is a dibasic acid by ester interchange. The goal of this step
is to drive this ester interchange to obtain a minimum of 90% completion of
reaction.
The liquid reaction material discharged from the polymerization step
is treated in a sparger unit to separate residual trace amounts of alcohol
and odoriferous components. The sparging medium may consist of a high boil-
ing liquid, a noncondensable gas, or steam. The overhead material from the
sparger passes through a condenser to a vent; recovered glycol can be recycled
to the esterification step. When steam is used as the sparging material,
steam condensate is withdrawn from the sparger and sent to wastewater treatment
and disposal.
39
-------
Recycle Alcohol
Glycol-^-
Dicarboxlylic
Acid
o
Catalyst
1
Esterification
^ Reactor
h
fc.\A/~f~, *„ u/^*~ To H'9n
4 k
ctionation
umn
2 o
i U
T i. i j Vacuurr
Treatment and ,.
P.. , Source
Disposal A
k Polymerization
? Step
t
Heat
i
— *
c
I
Atm.
Vent
t
Condenser
t
Sparger
Unit
-»„,,-,!
t 1
iparging Filter Cake
\Aedium to Waste
Disposal
Product
'roduct - ^ ° ' .
ma&> Packaging
storage — ^ Qnd
Shipping
Source: Midwest Research Institute
Figure V-2. Generalized process flow diagram for linear polyesters.
-------
From the sparger unit the liquid reaction product is sent to a filter
for removal of suspended matter and turbidity. A filter aid (e.g., diatoma-
ceous earth) is commonly used in this filtering operation. Activated carbon
can also be used in this step to absorb undesirable components (e.g., the
color of the product can be controlled). The used filter cake is either
sent to a landfill or used as an ingredient in the compounding of PVC resin.
The product is stored in dedicated storage tanks equipped with heaters and
special transfer pumps. The product is withdrawn from storage and sent to
packaging and shipping operations.
Multipurpose production equipment is commonly used in U.S. facilities
so that plasticizer products other than linear polyesters can be manufactured
in the same equipment.
By-Products and Contaminants--
The data compiled from manufacturers for product specification indicate
a general product purity for polyesters of 99.5% or higher. The principal
contaminants in the finished product generally consist of partially reacted
starting material and, in some cases, traces of catalyst residue. Even though
the polymerization catalyst is generally removed by chemical reaction, some
traces may remain with the polyester.
The sole by-product of this reaction sequence consists of water of es-
terification formed in the first process step. This water is removed by
distillation, with only trace amounts of polyester carried over with the
water.
Production Losses—
Industry sources estimate that the overall loss of plasticizer due to
processing and transfer is approximately 2% of the total production quantity.
The losses occur during three processes: fractionation, filtration, and sparg-
ing. The largest contribution to product loss probably occurs in the filter
process, followed by the sparging losses. No information is available con-
cerning the actual quantities lost by each process. Losses resulting from
the fractionation are estimated to be very small and probably less than from
the sparging process. The filter material containing the polyester plasticizer
is either disposed in a landfill or used in PVC compounding.
Uses
Linear polyesters are employed principally as primary plasticizers which
require very low volatility, low oil or water extraction, and low migration
characteristics from the plastic. Information from manufacturers varied
with respect to the percentage utilization of the polyesters. No published
data were available regarding the quantities of polyesters used in specific
products. The following estimate of polyester use as a plasticizer repre-
sents a combination of information from several sources.
41
-------
Quantity in 1979
Use Category % (million pounds)
Polyvinyl chloride ~85 38
Rubber 8-10 4-5
Adhesives 4-5 ~2
Coatings 1-2 ~1
The quantities used in 1979 are based on a volume of 45 million pounds, as
indicated by a manufacturer as the level employed for plasticizers.
Uses of polyvinyl chloride which employ polyesters as plasticizers in-
clude gaskets for appliances, refrigerators, and automotive vehicles, high
temperature wire coatings, high quality vinyl upholstry, electrical tape,
coaxial cable coatings, applications requiring oil and gasoline resistance,
coated fabrics, apparel and footware, and many others. Food uses of polyester-
plasticized PVC include food wrapping film, beverage hoses, milk tubing, milk-
ing machine components, bottle cap liners, and food conveyor belting.
Applications of polyesters in the rubber industry as a plasticizer in-
clude primarily those areas in which the vulcanizate requires good resistance
to swelling and plasticizer migration. Typical rubbers using polyesters
include styrene-butadiene and nitrile.
Adhesive applications are in areas such as pressure sensitive adhesive
backings and hot melt adhesives. Coating applications include usage of the
polyesters as pigment grinding vehicles and dispersion agents.
Quantities Released During Processing—
Essentially all processes used with the high volume production of for-
mulated PVC resins and synthetic rubber are highly automated and employ auto-
matic equipment and pumping systems for liquid plasticizers. No information
was available from the literature or industry concerning quantities released
during the formulation of resins or rubbers. Since the use of automated
equipment usually precludes consistent loss, it is estimated that processing
in the plastics and rubber industry does not lead to significant release of
polyester plasticizer. Losses due to equipment malfunction or breakage oc-
cur but are very difficult to quantify.
No information was found concerning losses during processing in the
adhesives or coatings industries. Losses in adhesive applications and in
coatings applications would be anticipated to be small due to the physical
nature of the polyesters and the volume of material employed in these two
industries. In general, polyesters are viscous liquids with a very low vapor
pressure and are not easily volatilized. These properties are not conducive
to significant losses in situations where mixing would be the principal pro-
cess. In addition, the estimated overall use of polyesters in these two
areas is quite small compared to the uses in PVC resins and synthetic rubbers.
42
-------
Ultimate Disposal--
Although the useful life of plastic and rubber products varies consider-
ably from one product to another, most products will be discarded within a
relatively short period of time (a few years). These products will be solid
waste and subjected to either incineration or landfill. The very low volatil-
ity of the polyester plasticizers and their very low migration from plastics
would preclude any appreciable loss of the plasticizer from the product dur-
ing its use.
It has been estimated that of the phthalate plasticizer containing plas-
tics that are disposed of as solid waste, 10 to 20% are destroyed by high
temperature incineration and 2% are subjected to low temperature incineration
or open burning (Peakall, 1975). The remainder of the plastic would be dis-
posed in a landfill. Since polyester plasticizers are employed in many of
the same applications as phthalates, these disposal figures would appear to
be valid for this class of plasticizer. Within the landfill, the polyester
plasticizers are subject to rapid attack by fungi and bacteria. The exact
lifetime of the polyesters in the landfill is unknown but probably is less
than 2-3 years based on the microorganism degradation studies reported in
Section VI.
Worker Exposure and Occupational Standards
Occupational Standards--
No information was found concerning any occupational standards for poly-
ester plasticizers. Sources searched for these data were the National Institute
for Occupational Safety and Health (1978a, 1978b), American Conference of
Governmental Industrial Hygienists (1979), and Sittig (1979).
Worker Exposure—
The National Institute for Occupational Safety and Health (1980) has
estimated the total number of workers exposed to certain chemicals in plants
of selected industries. This National Occupational Hazard Survey (NOHS)
estimated that a total of 1,257,364 workers covering 202 occupations were
exposed to one or more plasticizers. NOHS estimated that a total of 161,500
workers covering 74 occupations were exposed to polyester plasticizers.
Environmental Transport and Degradation
Very few quantitative data were found for properties which would be
directly related to environmental transport and bioaccumulation. Polyester
plasticizers have very low vapor pressures at ambient conditions, so that
volatilility would not be a major factor in their transport through the en-
vironment. In addition, these plasticizers exhibit a very low solubility
in water. No information was available on mobility in soils or sediments.
No data were available for octanol-water partition coefficients; there-
fore, no inferences can be made with respect to bioaccumulation or biomagni-
fication.
Environmental degradation can occur by photolysis, hydrolysis, and bio-
logical methods. As discussed in Section IV, no photolytic data are available
43
-------
for pure polyester compounds but plastic containing these plasticizers have
been subjected to photolytic degradation. It was determined that the plasti-
cizer was being degraded but no products were identified. In the same section,
the conditions for hydrolysis were also discussed. The effect of fungi and
bacteria on this class of plasticizer is discussed in Section VI, Environmental
Effects.
TRIMELLITATES
Producers and Quantities
A listing of the U.S. manufacturers of trimellitates (trimellitic acid
esters) and their production sites, as supplied by the companies, are shown
in Table V-5. The large volume producers are reported to be Hooker Chemical,
Reichhold Chemicals, USS Chemicals, Monsanto, and Technor-Apex.
TABLE V-5. MANUFACTURERS OF TRIMELLITATE PLASTICIZERS
Company Production site
Hooker Chemical Corporation Hicksville, NY
Reichhold Chemicals, Inc. Carteret, NJ
USS Chemicals, Div. of U.S. Steel Neville Island, PA
Monsanto Company Everett, MA
Technor-Apex Hebronville, MA
Brownsville, TN
C.P. Hall Company Bedford Park, IL
Inolex Corporation Philadelphia, PA
Tenneco Chemicals, Inc. Fords, NJ
Eastman Kodak Company Kingsport, TN
Pfizer, Inc. Greensboro, NC
Exxon Baton Rouge, LA
BASF Wyandotte Corp. Kearny, NJ
a Technor-Apex produces trimellitates under contract to other
companies and does not offer these plasticizers for direct
sale to consumers.
These five companies are estimated to control approximately 90% of the cur-
rent market. The remaining seven companies control a very small share of
the market. Since trimellitates are generally manufactured using multipur-
pose process equipment capable of producing many other esters, it is very
difficult to estimate plant capacities.
The estimated annual U.S. production levels for trimellitate plasticizers
are presented in Table V-6 for 1975-1979. Preliminary data in Modern Plastics
indicated that for 1980, the estimated consumption of trimellitate plasticizers
will be approximately 28.7 million pounds. This represents a decrease of
slightly more than 2 million pounds from 1979. Manufacturers indicate that
annual production is approximately equal to consumption and that little stock-
piling occurs for this class of plasticizer.
44
-------
Production Process
Trimellitate plasticizers are monomeric compounds and consist of esters
of trimellitic acid. The production process is normally conducted on a batch
basis. The basic raw material used in the manufacture of these plasticizers
is trimellitic anhydride, derived from trimellitic acid.
TABLE V-6. ANNUAL PRODUCTION OF TRIMELLITATE PLASTICIZERS
(quantities in million pounds)
1975
1976
1977
1978
1979
Plasticizer
Total Trimellitates
Trioctyl Ester
Tri-n-octyl, n-decyl
Ester
All other
ITC"
16.2
6.1
N.A.
10.1
MP ITC
20 23.1
9.3
- N.A.
13.8
MP ITC
24 27.3
12.5
1.2
- 13.6
MP ITC
26 32.8
- 15.8
1.1
- 15.9
MP ITC MP
31 N.A.C 31
-
-
™ ™"
a International Trade Commission data.
b Modern Plastics data.
c N.A. = not available.
The typical production process in the United States is essentially the same
as that used for production of phthalate plasticizers (i.e., esterification
of an anhydride compound).
The process chemistry is based on an esterification reaction between
trimellitic anhydride (TMA) and an alcohol (C7 to Ci2) to produce the corre-
sponding ester and by-product water. The overall chemical reaction is shown
below:
+ 3ROH
_cataly$t
2H20
Trimellitic
anhydride
A generalized process flow sheet
plasticizers is shown in Figure V-3.
of TMA and an alcohol, are fed in the
lyst into an esterification reactor.
commonly used. A single catalyst is
type such as toluenesulfonic acid, or
is amphoteric. In Figure V-3, it is
Reactor effluent is sent directly to
Trimellitate
By-product water
for domestic production of trimellitic
The liquid raw materials, consisting
desired proportions along with a cata-
A stoichiometric excess of alcohol is
generally used, either a proton acid
a tetra alkyl titanate compound, which
assumed that an acid catalyst is used.
a wash tank system, and the acid catalyst,
45
-------
Acid
1
nl
Mitic— ».
Jride
yst
Recycle Alcohol
I
Esterification
Reactor
Recovered
Alcohol
Storage
^ —
Alkaline — i
Solution 1
Decanter
Separator
System
r
Type
i
i
Water
ater
Catalyst Removal
(Washing Tanks)
1
enser
i
Steam
(Batch)
*— Steam
Spargii
CaO— |
Filter -C
Polish Fi
n9
r- Filte
)ry
zation/
Itration
Waste water to
Treatment and
Disposal
Filter Cake to
Waste Disposal
Final Product
to Packaging
and Shipping
Source: Midwest Research Institute
Figure V-3. Schematic flow diagram for production of trimellitate plasticizers.
-------
along with a small amount of unconverted trimellitic anhydride, is separated
from the crude product. The washed reaction product is then transferred to
a batch-operated steam stripper, where unreacted alcohol (along with some
water) is removed overhead and condensed. The stripper condensate is then
treated in a decanter-type separator system to remove water from the alcohol.
Recovered alcohol is reclaimed from a storage tank and recycled to the ester-
ification reactor. Wastewater from the washing operation is combined with
separated water from the decanter-separator system and sent to secondary
treatment and disposal. An alternative stripping operation consists of a
continuous distillation unit in which the washed reaction product is con-
tacted countercurrently with steam.
Crude product discharged from the stripper is then treated in a multi-
purpose filtering operation to neutralize any residual traces of acid, remove
suspended solids and clarify the final product material. Calcium oxide (CaO)
is commonly used as a neutralizing material (e.g., as a precoat on the filter
media). Filter aids are also commonly used (e.g., clay or diatomaceous earth)
The filter cake is normally disposed as a solid waste material or used in
the compounding of PVC resins. The finished product, which generally has a
purity in excess of 99.5%, is then transferred to packaging and shipping
operations.
By-Products and Contaminants--
The data compiled from manufacturers for product specifications indicate
a general product purity for trimellitic acid esters of 99.5% or higher.
The principal contaminants in the finished product generally consist of un-
reacted or partially reacted starting material and, in some cases, traces
of catalyst residue. Although the catalyst is generally removed by chemical
reaction, some traces may remain with the trimellitate final product.
The by-products of this reaction sequence consist of water of esteri-
fication and partial esters formed in the first process step. The water is
separated during the reaction step in the esterification reactor and during
the subsequent steam stripping step. Only trace amounts of the trimellitate
are carried over with the water during the stripping process. The partial
esters are removed during the final neutralization/clarification step.
Production Losses--
The overall loss of plasticizer due to processing and transfer of tri-
mellitic acid esters results primarily from the catalyst removal procedure,
the steam sparging process, and the final filtration. No data were found
in the literature concerning specific percentage losses from each of these
sources. Because the basic processing and transfer procedures employed for
the trimellitates are very similar to those for epoxy compounds and linear
polyesters, the percentage losses may be very similar. Overall processing
and transfer losses for epoxy compounds and linear polyesters were estimated
by the manufacturers to be approximately 2% of the production quantity.
These losses occur in processes such as fractionation, steam stripping, fil-
tration, neutralization, and washing. Since the same processes are employed
in the manufacture of the trimellitates, it is estimated that approximately
the same overall percentage loss should occur.
47
-------
During processing of the previous two classes of plasticizers, the fil-
tration step was estimated by manufacturers to account for the greatest loss,
followed by the stripping (or sparging) procedure. Losses due to fractionation,
neutralization, and washing were estimated by producers to be very small
compared to the other methods of loss. For the trimellitate plasticizers,
these same relative rankings would likely be valid.
Uses
Trimellitates are employed as primary plasticizers which require low
volatility, low water extraction, low migration characteristics, good high
temperature performance, and good electrical properties. No published data
were available regarding the quantities of trimellitate used in specific
areas. The following estimates of trimellitate consumption by use area were
derived by MRI based on information supplied by various producers of the
trimellitates: PVC 94-95%; ABS 1-2%; and miscellaneous 3-5%. Miscellaneous
uses include other polymeric systems (e.g. chlorinated polyethylene, cellulose
nitrate, cellulose acetate) and coating applications.
Within the very large consumption category of polyvinyl chloride (PVC),
the applications are estimated to be as follows: 90% in communication cable
coating and electrical wire and cable coating including 90°C and 105°C rated
coatings; 2-3% bonded PVC and ABS rubber used in automotive and truck crash
pads; 1% in speciality tapes and electrical tape; and 1% in plastisol uses
for electrical purposes. The specialty tapes are often PVC-polyvinyl ace-
tate copolytners.
Trimellitates are used as plasticizers in the ABS for automotive and
truck panels and chlorinated polyethylene in high temperature, wire coating
applications. Trimellitates are used in lacquers (coatings) to waterproof
copper wiring in electric motors and generators.
Quantities Released During Processing—
Essentially all processes used with the high volume production of formu-
lated PVC resins and synthetic rubber are highly automated and employ auto-
matic equipment and pumping systems for liquid plasticizers. No information
was available in the literature or from industry concerning quantities re-
leased during the formulation of resins or rubbers. Since the use of auto-
mated equipment usually precludes consistent loss, it is estimated that pro-
cessing in the plastics and rubber industry does not lead to significant
release of trimellitate plasticizer. Losses due to equipment malfunction or
breakage occur but are very difficult to quantify.
Ultimate Disposal--
Although the useful life of plastics and other polymer products varies
considerably from one product to another, most products will be discarded
within a relatively short period of time (a few years). These products will
be solid waste and subjected to either incineration or landfill. The very
low volatility of the trimellitate plasticizers and their very low migration
from plastics would preclude any appreciable loss of the plasticizer from
the product during its use.
48
-------
It has been estimated that of the phthalate plasticized plastics that
are disposed as solid waste, an estimated 10-20% are destroyed by high tem-
perature incineration and 2% are subjected to low temperature incineration
or open burning (Peckall, 1975). The remainder of the plastic would be dis-
posed in a landfill. Within the landfill, the trimellitate plasticizers are
probably subject to attack by fungi and bacteria. The exact lifetime of
the trimellitates in the landfill in unknown but based on the studies dis-
cussed in Section VI, the lifetime probably exceeds that for either the epoxy
or polyester plasticizers.
Worker Exposure and Occupational Standards
Occupational Standards--
No information was found concerning any occupational standards for tri-
mellitate plasticizers. Sources searched for these data were the National
Institute for Occupational Safety and Health (1978a, 1978b), American Confer-
ence of Government Industrial Hygienists (1979), and Sittig (1979). A bulletin
has been published concerning the health effects of trimellitic anhydride
(TMA), the raw material for trimellitate production (NIOSH, 1978c).
Worker Exposure--
The National Institute for Occupational Safety and Health (1980) has
estimated the total number of workers exposed to certain chemicals in plants
of selected industries. No data were found for trimellitate plasticizers.
Environmental Transport and Degradation
Very few quantitative data were found for properties which would be
directly related to environmental transport and bioaccumulation. Trimellitate
plasticizers have very low vapor pressures at ambient conditions, about the
same as the linear polyesters, so that volatility would not be a major fac-
tor in the transport through the environment. In addition, these plasticizers
exhibit a very low solubility in water. No information was available on
mobility in soils or sediments. No data were available for octanol-water
partition coefficients so no inferences can be made with respect to bioaccumu-
lation or biomagnification.
Environmental degradation can occur by photolysis, hydrolysis, and bio-
logical methods. No photolytic data were available nor have any studies
been performed on degradation by microorganisms. In Section IV, conditions
were discussed for the basic and acidic hydrolysis of trimellitate plasti-
cizers.
49
-------
SECTION VI
HEALTH AND ENVIRONMENTAL EFFECTS
HEALTH EFFECTS
This section summarizes the information available from the literature
and manufacturers concerning the health effects of each of the three classes
of plasticizers. In general, very few data were available in the litera-
ture for any of the three classes; no data were found for the trimellitates.
Data from manufacturers normally stated only species tested, type of test,
dosage level, and effects. Information was not available on testing proto-
col.
Epoxy Compounds
Many of the epoxy compounds have been approved by the Food and Drug
Administration (FDA) for use as an indirect food additive. These additives
are usually chemicals that constitute a relatively minor ingredient in the
packaging materials, and contamination of foodstuffs would result from migra-
tion of the chemical from the packaging material into the food product.
The FDA was contacted but they provided no information that had not been
obtained from the literature.
Larson, et al. (1960) performed chronic toxicity studies on two epoxi-
dized soybean oils, Paraplex G-60 and G-62 (products of Rohm and Haas).
They performed two-year feeding studies at levels of up to 5% epoxidized
oil in the diet of rats and one-year feeding studies in dogs at the same
dietary levels.
In the study with Paraplex G-60, the highest dose (5%) produced rela-
tively minor toxic effects in rats. These effects were manifest as early
depression in weight gain and increased relative liver weight but no hepatic
histopathology. The early depression in weight gain was recovered as the
tests proceeded. A dose level of 2.5% did not produce any effects on the
rats. The dogs appeared to be more sensitive to weight loss than the rats.
Those dogs fed at the 5% level lost weight (or gained less than the controls)
because they consumed less food than the controls due to an apparent aversion
to the dosed feed. Those dogs fed at a dietary level of 1% oil were not
adversly affected with respect to weight loss.
The results from Paraplex G-62 were similar but this product appeared
to be somewhat more potent in its effects. Rats fed at a 1, 2.5, or 5% die-
tary level of the epoxidized oil had an initial depression of weight gain
50
-------
but recovered later in the tests. Liver and kidney weights were increased,
relative to the controls, in several of the higher dose groups; however since
there was an absence of any significant histopathology, this result is of
little toxicological importance. Dogs fed at the 5% dietary level lost weight
in comparison to the controls, but those animals fed at the 1% level appeared
to be normal with respect to weight levels.
In summary, the feeding study for Paraplex G-60 and G-62 on rats (two
years) and dogs (one year) produced minimal nonspecific effects on weight
gain, even at dietary levels of 5%. No effects were observed on survival
and histologic examination of tissues of the heart, lung, liver, kidney,
spleen, thyroid, adrenal, pancreas, gonads, muscle, and bone marrow showed
no lesions attributable to treatment.
Weil, et al. (1963) conducted a massive study including preliminary
acute toxicity tests on 60 compounds and skin-painting carcinogenesis studies
on 28 of the compounds. Compounds of interest to this report included epoxi-
dized soybean oil and seven epoxidized tall oil derivatives.
In the acute tests, all compounds of interest showed very little toxicity.
Acute oral LDso values for rats were in excess of 20 ml/kg body weight.
Exposure of rats to the concentrated vapors for eight hours produced no deaths.
Minimal skin irritation was produced on uncovered rabbit stomach. In an
eye irritation test only one rabbit showed evidence of corneal injury from
any of the eight compounds tested. None of the eight compounds showed any
sensitization of guinea pigs. In the skin painting carcinogenesis study,
neither the epoxidized soybean oil nor the two epoxidized tall oil esters
produced any tumors.
In summary, the tested compounds were not totally inert, but they pro-
duced few toxic effects. These effects were limited to skin irritation effects
even at relatively large doses.
Kotin and Falk (1963) reviewed the effects of various epoxides and per-
oxides and related materials on neoplasia, including incomplete mouse studies.
Very few data were produced, and the data could not be evaluated due to a lack
of control data. In addition, the loss of 60% of the test mice (30/50) during
the course of the tests made evaluation of the results difficult.
Kieckebusch, et al. (1963) performed a structure-activity study of vari-
ous epoxidized soybean oils. In this study, the degree of epoxidation (de-
fined in terms of mg/epoxy oxygen/dose) was the independent variable. A
dose-response relationship was found for weight gain and for death. In this
study, no data were available to allow a comparison of the composition of
these test materials with the commercial products currently on the market
in the U.S. so that the results have no significant utility.
Arffman (1964) conducted a study of the effects of modified fats on
newts as a potential screening method for carcinogenicity. The experimental
methods for the animal testing were not described. The dependent variable
51
-------
in the tests was epidermal reaction. Heat polymerized soybean oil produced
negative results on the test species, and highly peroxidized oils showed a
toxic effect. The compositions that are normal for commercial usage were
negative.
Data from a producer for skin sensitivity for epoxidized soybean oil
showed that patch tests on humans with undiluted epoxide for a 5-day period
and 2-day repeat tests produced no irritation. A plastic film containing
19% epoxide patch tested on humans for 2 days with a repeat test of 4 days
showed no cutaneous reactions (Rohm and Haas, 1980).
Linear Polyesters
Some of the linear polyesters available on the current market have been
approved by the FDA as plasticizers for selected homopolymers used in contact
with food. Information was solicited from FDA for seventeen specific linear
polyesters, but information was available for only one compound, an azelaic
acid-propylene glycol polymer. However, all data were designated as
"privileged" and unavailable.
Only two reports were found in the literature on linear polyesters,
Mallette and von Haam (1952) and Fancher, et al. (1973).
Mallette and von Haam (1952) conducted a study of the toxicity and skin
effects of 25 plasticizers used in the rubber and plastics industries. Two
linear polyesters, Paraplex G-25 and Paraplex G-40, were included in the
study. For each of the two materials, intraperitoneal toxicity was negligible.
Only one foreign body granuloma was found after a 6-g/kg injection. However,
this study reported that when the two compounds were diluted in mineral oil
or propylene glycol severe dermal irritation and moderate dermal sensitization
effects were produced in humans. In the dermal irritation tests, they were
the most toxic of all compounds tested. However, the latter results have
been refuted by the manufacturer of the materials.
Information from the manufacturer, (Rohm and Haas, 1980) states that
Paraplex G-25, either in undiluted form or as a 25% ointment, produced no
skin irritation to human subjects after 48-hr contact. Tests employing 70%
polyester in 30% toluol or 30% Solvesso 100 produced both skin irritation
and sensitization due to the solvent. Tests using Paraplex G-40 as a 25%
ointment produced no skin irritation to human subjects after a 48-hr contact
period.
Fancher, et al. (1973) performed a two year chronic feeding study on
rats and dogs and a three-generation reproduction study on rats using a 1,3-
butylene glycol adipic acid polyester terminated by a 16% by weight mixture
of myristic, palmitic, and stearic acids (Santicizer 334F). For all studies,
the doses were 0, 0.1, 0.5, and 1.0% in the feed.
The toxicity tests showed no consistent toxic effects. A few parameters
in the three-generation reproduction study were statistically different from
the control animals, but there were no consistent effects over the course
of the study. The overall results showed that there were no toxicological
effects for this material in any of the chronic tests, even at a dose level
of 1% in the feed.
52
-------
Monsanto Company supplied unpublished toxicity information for five
linear polyester plasticizers. The acute oral LDso in rats was greater than
50 g/kg for Santicizer 409, greater than 10 g/kg for Santicizers 334F and
411, 9.42 g/kg for Santicizer 412, and 20.8 g/kg for Santicizer 429. Acute
dermal LD5o for rabbits was estimated to be greater than 7.94 g/kg for San-
ticizers 334F and 411 and greater than 10 g/kg for the other three materials.
Neither Santicizer 334F nor 411 showed eye irritation in rabbits; Santicizers
412 and 429 showed slight irritation, and Santicizer 409 showed mild irritation.
Only Santicizer 411 showed slight skin irritation to rabbits for a 24-hr
contact period; all others showed no irritation.
Rohm and Haas Company (1980) supplied toxicity information on three
linear polyesters: Paraplex G-54, G-56, and G-57. The acute oral LD50 in
rats was greater than 30 ml/kg for G-54 and greater than 5 ml/kg for G-56
and G-57. All three materials showed acute dermal LDso levels in rabbits
of greater than 3 g/kg. All three materials showed mild skin irritation to
rabbits after a 24-hr contact period. Only G-57 showed mild eye irritation
to rabbits; the other two produced no eye irritation. Patch tests of 24-hr
duration conducted on 50 human subjects showed no signs of primary irritation
or sensitivity reactions for Paraplex G-54.
Trimellitates
No information was available in the literature for this class of plas-
ticizer. Data were supplied by two manufacturers of the trimellitate.
Monsanto Company provided information on Santicizer 79TM plasticizer.
Acute oral LD5o in rats was found to be greater than 15.8 g/kg and acute
dermal LDso in rabbits was greater than 7.94 g/kg. Slight eye irritation
was produced when the undiluted material was placed in the conjunctival sac
of the rabbit. No skin irritation was detected on rabbits after a 24-hr
contact period.
Eastman Kodak Company stated that their trioctyl trimellitate had oral
and intraperitoneal LDso greater than 3.2 g/kg in both rats and mice. Liquid
placed in contact with"guinea pig skin for 24 hr resulted in only slight
irritation with no evidence of absorption (skin LDso greater than 20 ml/kg).
The skin of the guinea pig was also not sensitized"during testing. During
tests with rabbits slight eye irritation was produced by the undiluted ma-
terial.
Rats survived a 6-hr exposure to 10 ppm of the trimellitate and exhib-
ited only mild irritation. The animals gained weight in a normal manner
during a two week observation period following exposure. However, rats ex-
posed for 6-hr to calculated atmospheric concentrations of 118 ppm generated
at 180°C resulted in death. These deaths were delayed for as long as three
days.
53
-------
The National Institute for Occupational Safety and Health (1978) has
recommended that trimellitic anhydride (TMA) be handled as an extremely toxic
agent. Exposure to this compound may result in noncardiac pulmonary edema
(apparently without a pulmonary irritation warning), immunological sensitiza-
tion, and irritation of the pulmonary tract, eyes, nose, and skin. It could
be suggested that heating the trioctyl trimellitate to 180°C may have resulted
in the thermal dissociation of a small amount of the ester to trimellitic
anhydride. Therefore, the death of the rats may have resulted from inha-
lation of the anhydride and not from the trioctyl trimellitate.
ENVIRONMENTAL EFFECTS
No information was found in the literature concerning the effects of
any of the three classes of plasticizers on fish, plants, birds, or mammals.
Information was available relating to the effects of the epoxy compounds
and linear polyesters on protista. The effects of two linear polyesters on
fish have been studied by Monsanto Company. No environmental information
was found for any of the trimellitates.
Effects on Fish
Unpublished data on the effects of two linear polyesters, Santicizer
409 and 429, were reported by Monsanto Company. In a 4-day static fish
toxicity study using Santicizer 409, the 96-hr LDso was calculated to be
100 ppm for fingerling rainbow trout and 125 ppm for bluegill.
A 4-day static study was also conducted for Santicizer 429. The 96-hr
LDso was calculated to be greater than 100 ppm for the fingerling rainbow
trout and the bluegill.
Microbiological Degradation
Berk, et al. (1957) conducted a massive study in which 99 acids and
their esters were among 127 compounds tested with 24 fungi. The object of
the study was an attempt to correlate the extent of fungus growth with
chemical structure. Among the compounds tested were two unidentified sebacic
acid polyesters, polypropylene sebacate, and two materials identified only
as polyesters. One of the unidentified sebacic acid polyesters was tied
with another compound for the highest average fungal growth rate (6.8 cm)
of all compounds tested. The other four polyesters also showed high fungal
growth with average rates ranging from 5.1 to 6.0 cm. Only 21 of the 127
test compounds had average fungal growth rates in excess of 5.0 cm.
Klausmeier (1966) studied the isolation of microorganisms capable of
degrading the ester plasticizers but incapable of using those esters as a
sole source of nutrient and energy in a mineral salts medium. Butylene glycol
polyadipate (EGA) was one of seven plasticizers used to isolate fungi. This
compound (EGA) showed very poor fungal resistance with only 1 of 51 fungal
isolates being adventitious (i.e., the isolate would degrade the plasticizer
only in the presence of an extraneous organic nutrient). Bacterial and yeast
cultures expected to be active against plastic materials were inoculated on
yeast extract-EGA and mineral salt-EGA media. With the bacteria, EGA was
54
-------
only adventitiously degraded; the yeast studies showed EGA to be adventi-
tiously degraded in four of the tests and utilized as the sole organic
nutrient in five of the tests.
Sterile strips of PVC film containing 50 parts of EGA per 100 parts of
polymer were incubated in the presence of an undefined organism adventitiously
capable of degrading EGA for various periods of time. The preliminary findings
indicated that there was a considerable stiffening (loss of plasticizer) in
the inoculated specimens after 23 days. No significant change occurred in
the sterile controls.
Booth and Robb (1968) reported the bacterial attack on plasticized PVC
in a soil environment for 8 weeks by species of Pseudomonas and Brevibac-
terium and the changes in some physical properties of the plastic that accom-
pany the biodegradation process. Weight loss, cyclic deformation, and stress
relaxation were the physical parameters tested.
Epoxidized soybean oil, with di-isooctyl phthalate (DIOP), and epoxy
esters (plus DIOP) were the two epoxy compounds tested. Of 13 plasticizers
in one test group, the two epoxy compounds ranked in approximately the middle
in terms of weight loss and for the stiffness tests. They showed essentially
no loss in the relaxation test. The relaxation test showed very little dif-
ference for all samples between the controls and the test material, so the
results of this test provided very little guide to deterioration of all plas-
ticizers tested. Overall, the epoxy compounds ranked in approximately the
middle of 13 plasticizers for bacterial degradation during the course of
this study.
Materials comprised of adipic acid plus propylene glycol and sebacic
acid plus propylene glycol were the two linear polyesters tested. In the
weight loss test and the stiffness test, the two polyesters ranked slightly
above average out of 13 plasticizers. In the relaxation test, neither plas-
ticizer showed any significant loss of relaxation. Overall, the polyesters
were among the top five plasticizers in terms of bacterial degradation.
Darby and Kaplan (1968) tested three linear polyesters, both as monomers
and as polymers with selected diisocynates, with six organisms for fungal
susceptibility. The three linear polyesters were polyethylene glycol adipate,
poly-l,3-propanediol adipate, and poly-l,4-butanediol adipate. The six fungi
were: Aspergillus niger, A^ flavus, A^ versicolor, Penicillium funiculosum,
Pullularja pullulans, and Trichoderma spp~!(mixed species) with added
Chaetomium globosum.
All polymers derived from each of the three polyesters were excellent
substrates for fungal growth. All of the polymers except one showed heavy
growth (60 to 100% covered). The exception showed moderate growth (30 to
60% covered). All three monomer polyesters showed heavy fungal growth.
Osmon, et al. (1969) studied the effects of 17 yeasts (8 different general)
on 13 different plasticizers, including one polyester. Butylene glycol poly-
adipate (EGA) was the only material from the three classes of plasticizers
of this survey. Of the 17 yeasts, 10 degraded the EGA regardless of the
55
-------
presence of yeast extract as a nutrient; 3 yeasts were reverse adventitious
(i.e., degraded plasticizer only in the absence of nutrient); and 4 yeasts
showed no degradation.
Three of the cultures (Torulopsis BY4, Pullularia BY10, and Candida
BY17) were selected for evaluation of their ability to degrade vinyl film,
containing EGA, in liquid culture. After 14 days, none of the cultures
significantly degraded the vinyl film even though all three cultures hydro-
lyzed EGA in the plate studies (described in the previous paragraph). Mix-
tures of cultures showed no improvement in the degradability of the vinyl
film over the individual cultures.
Rodriquez (1971) published a review article on biodegradability of a
number of components of plastics and rubber. Several of the previous arti-
cles on linear polyesters were briefly reviewed.
Lazar and loachimesca (1973) conducted a study in which a linear poly-
ester, polypropylene glycol adipate, was among nine plasticizers subjected
to testing for fungus attack. The polyester was among the most sensitive
materials tested. The sample was completely covered by fungus mycelia and
fructifications. Fungi employed in this study were not identified, Romanian
and French standard methods were stated to have been employed.
Potts, et al. (1973a) studied the biodegradability of synthetic polymers.
Polyvinyl chloride, containing epoxidized soybean oil plasticizer, was exposed
to a mixture of fungi (A. niger, A_^ flavus, C^ globosum, and P_._ funiculosutn)
for a period of 3 weeks. At the end of this period, the test sample showed
medium growth with 30-60% of the sample covered with fungal growth.
Potts, et. al. (1973b) studied the biodegradability of commercially
available plastics and additives commonly used in these plastics. The fungi
and test methods were described in Pott, et al. (1973a). One epoxy plasti-
cizer, epoxidized soybean oil (Flexol EPO), and one linear polyester, Plas-
tolein 9765, were among the materials tested. Both plasticizers showed heavy
fungal growth; each sample was 60 to 100% covered.
Sewage Treatment
Saeger, et al. (1976) studied the biodegradability of three aliphatic
adipic acid diesters and one linear polyester, 1,3-butylene glycol adipic
acid (Santicizer 334F), in acclimated, activated sludge systems. Carbon
dioxide evolution procedures were employed to determine the biodegradability
of the polyester. Primary biodegradation rates were not determined for this
compound because of the lack of an applicable analytical method. At concen-
trations ranging from 20 to 56 mg/L, the extent of carbon dioxide evolution
from the polyester was comparable to that for dextrose. For the two different
carbon dioxide evolution procedures, gas evolution from the polyester after
35 days was 78.1 and 88.1% of theoretical, compared to 82.1 and 90.2% for
Santicizer 97A, di(heptyl,nonyl) adipate, and 93.8 amd 96.4% for dextrose.
56
-------
SECTION VII
PLASTICIZER INTERCHANGEABILITY
The entire concept of interchangeability among plasticizers within vari-
ous resin systems is extremely complex and dependent upon the specific appli-
cation for the particular product. A very large number of the plasticizers
currently available on the market are basically directed towards specific
uses or the incorporation of specific properties into a particular resin sys-
tem. While general purpose plasticizers are commonly found on the market,
the vast majority of the plasticizers are developed for specific purposes.
Data on plasticizer compatibility were compiled from the published lit-
erature (Modern Plastics Encyclopedia, 1979-80). Information presented in
Table VII-1 relates to the overall compatibility of various classes of plas-
ticizers with specific resins. The table is not comprehensive with respect
to all classes of plasticizers but does incorporate all of the major classes.
For those plasticizer classes which are stated to be not used with a resin,
the data from the Modern Plastics Encyclopedia showed that none of the
individual plasticizers within the specific class are used with the resin.
Within each class of plasticizer, an estimate was made of the general compati-
bility of that class with the specific resin. This does not imply that all
plasticizers within a certain class will conform to the compatibility rating
given for the class. In addition, not all plasticizers within a given class
may be used with a particular resin. For example, adipic acid esters as a
class are partially compatible with polymethyl methacrylate. Some specific
esters may be incompatible and some may be compatible; but overall, the major-
ity of the specific esters are partially compatible. However, for azelaic
acid esters, none of the specific esters are used with polymethyl methacrylate.
Tables VII-2 to VII-4 present general interchangeability information
for each of the three classes of plasticizers in this study, i.e., epoxy,
polyester, or trimellitate. In these tables, the resins which show compat-
ibility with one of the three classes of plasticizers are compared with all
other classes of compatible plasticizers. In Table VII-2, epoxy plasticizers
are stated to be generally compatible with cellulose acetate butyrate, cellu-
lose nitrate, ethyl cellulose, polyvinyl chloride, and vinyl chloride acetate.
Numerous other classes of plasticizers are shown which are also compatible
with these resins. As in Table VII-1, not all of the individual plasticizers
within a given class are completely compatible with a specific resin. These
data present generalizations for entire classes.
57
-------
TABLE VII-1. GENERAL COMPATIBILITY OF PLASTICIZERS
Compatibility
Plasticizer
Adipic acid derivatives
Azelaic acid derivatives
Benzoic acid derivatives
Polyphenyl derivatives
Citric acid derivatives
Epoxy derivatives
Fumaric acid derivatives
Glutaric acid derivatives
Glycerol derivatives
Glycol derivatives
Isophthalic acid derivatives
Laurie acid derivatives
Maleic acid derivatives
Trimellitates
Myristic acid derivatives
Oleic acid derivatives
Paraffin derivatives
Petroleum derivatives
Phosphoric acid derivatives
Phthalic acid derivatives
Polyesters
Ricinoleic acid derivatives
Sebacic acid derivatives
Stearic acid derivatives
Sucrose derivatives
Sulfonic acid derivatives
Tall oil derivatives
Terephthalic acid derivatives
CA
pb
C
P
I
C
I
N
P
C
I
I
N
N
I
I
I
C
I
C
I
I
I
I
I
C
C
P
I
CAB
C
C
C
C
C
I
N
C
C
C
C
C
N
C
C
C
C
I
C
C
C
C
P
C
C
C
N
C
CN
C
C
P
C
C
C
N
P
C
C
C
C
N
C
C
C
P
I
C
C
C
C
C
C
C
C
P
C
EC
C
C
C
C
C
C
N
C
C
C
C
C
N
C
C
C
C
C
C
C
I
C
C
C
C
C
C
P
PM
P
N
C
N
N
I
N
P
C
C
P
N
N
C
N
P
C
C
C
C
I
P
C
I
C
C
N
N
with plastics3
PS
C
C
P
C
C
I
N
C
I
C
C
C
N
P
C
C
C
C
C
C
I
P
C
C
C
P
N
C
PVA
P
P
P
C
C
I
C
P
C
C
P
N
C
I
N
P
C
C
C
I
P
C
P
C
C
C
P
C
PVB
P
I
P
C
C
I
N
P
C
C
C
C
N .
P
N
C
P
P
C
C
I
C
P
C
I
C
C
P
PVC
C
C
C
C
C
C
C
C
I
C
C
C
N
C
N
C
C
P
C
C
C
C
C
C
C
I
C
C
VGA
C
C
P
C
C
C
C
C
I
C
C
C
N
C
I
P
C
C
C
C
C
C
C
C
C
C
C
C
a Code for,Plastics: CA=Cellulose acetate; CAB=Cellulose acetate butyrate;
CN=Cellulose nitrate; EC=Ethyl cellulose; PM=Polymethyl methacrylate;
PS=Polystyrene; PVA=Polyvinyl acetate; PVB=Polyvinyl butyral;
PVC=Polyvinyl chloride; VCA=Vinyl chloride acetate.
Code for Compatibility: C=Compatible; P=Partially compatible;
I=Incompatible; N=Not used in this plastic.
58
-------
TABLE VII-2. INTERCHANGEABILITY FOR EPOXY PIASTICIZERS
Compatible plastics3
Plasticizers
Epoxy derivatives
Adipic acid derivatives
Azelaic acid derivatives
Benzoic acid derivatives
Polyphenyl derivatives
Citric acid derivatives
Fumaric acid derivates
Glutaric acid derivatives
Glycerol derivatives
Glycol derivatives
Isophthlaic acid derivatives
Laurie acid derivatives
Trimellitates
Myristic acid derivatives
Oleic acid derivatives
Paraffin derivatives
Petroleum derivatives
Phosphoric acid derivatives
Phthalic acid derivatives
Polyesters
Ricinoleic acid derivatives
Sebacic acid derivatives
Stearic acid derivatives
Sucrose derivatives
Sulfonic acid derivatives
Tall oil derivatives
Terephthalic acid derivatives
CAB
c
Cb
C
C
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
CN
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
EC
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
PVC
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
VGA
c
c
c
(
c
c1
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
Code for Plastics: CAB=Gellulose acetate butyrate; CN=Cellulose nitrate;
EC=Ethyl cellulose; PVC=Polyvinyl chloride; VCA=Vinyl chloride acetate.
C=Compatible.
59
-------
TABLE VII-3. INTERCHANGEABILITY FOR POLYESTER PLASTICIZERS
Compatible plastics3
Plasticizer
Polyesters
Adipic acid derivatives
Azelaic acid derivatives
Benzoic acid derivatives
Polyphenyl derivatives
Citric acid derivatives
Epoxy derivatives
Fumaric acid derivatives
Glutaric acid derivatives
Glycerol derivatives
Glycol derivatives
Isophthalic acid derivatives
Laurie acid derivatives
Trimellitates
Myristic acid derivatives
Oleic acid derivatives
Paraffin derivatives
Petroleum derivatives
Phosphoric acid derivatives
Phthalic acid derivatives
Ricinoleic acid derivatives
Sebacic acid derivatives
Stearic acid derivatives
Sucrose derivatives
Sulfonic acid derivatives
Tall oil derivatives
Terephthalic acid derivatives
CAB
Cb
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
CN
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
PVC
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
VGA
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
Code for Plastics: CAB=Cellulose acetate butyrate; CN=Cellulose nitrate;
PVC=Polyvinyl chloride; VCA=Vinyl chloride acetate.
C=Compatible.
60
-------
TABLE VII-4. INTERCHANGEABILITY FOR TRIMELLITATE PLASTICIZERS
Compatible
Plasticizer
Trimellitates
Adipic acid derivatives
Azelaic acid derivatives
Benzoic acid derivatives
Polyphenyl derivatives
Citric acid derivatives
Epoxy derivatives
Fumaric acid derivatives
Glutaric acid derivatives
Glycerol derivatives
Glycol derivatives
Isophthalic acid derivatives
Laurie acid derivatives
Myristic acid derivatives
Oleic acid derivatives
Paraffin derivatives
Petroleum derivatives
Phosphoric acid derivatives
Phthalic acid derivatives
Polyesters
Ricinoleic acid derivatives
Sebacic acid derivatives
Stearic acid derivatives
Surcose derivatives
Sulfonic acid derivatives
Tall oil derivatives
Terephthalic acid derivatives
CAB
Cb
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
CN
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
EC
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
plastics3
PM
C
C
C
C
C
C
C
C
C
C
C
PVC
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
VGA
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
Code for Plastics: CAB^Cellulose acetate butyrate; CN=Cellulose nitrate;
EC=Ethyl cellulose; PM=Polymethyl methacrylate; PVC=Polyvinyl chloride;
VCA=Vinyl chloride acetate.
C=Corapatible.
61
-------
An evaluation of specific plasticizer usage and interchangeability on
an individual plasticizer basis would be extremely complex because of the
many factors which are involved in the selection of a plasticizer for a par-
ticular resin. Physical properties of the plasticizer, resin component com-
patibility, physical properties of the resultant plastic, effect on product
specifications, and overall process economics are a few of the factors which
must be considered if a change of a specific plasticizer for a specific resin
is contemplated. An evaluation of all possible alternatives for each individ-
ual plasticizer for each specific application is beyond the scope and intent
of this study.
62
-------
REFERENCES
American Conference of Governmental Industrial Hygienists (ACGIH). 1979.
TLVs Threshold Limit Values for Chemical Substances and Physical Agents in
the Workroom Environment:with Intended Changes for 1979. Cincinnati, Ohio.
American Oil Chemists Society (AOCS), 1974. Official and Tentative Methods
of the American Oil Chemists Society: Sampling and Analysis of Vegetable Oil
Source Materials. Champaign, Illinois.
American Society for Testing and Materials (ASTM), 1980. Annual Book of ASTM
Standards. Philadephia, Pennsylvania.
Applewhite, T.H. 1980. "Fats and Fatty Oils" in Kirk-Othmer Encyclopedia
of Chemical Technology, 3rd Edition, Volume 9, John Wiley and Sons. p. 795-
804.
Arffmann, E. 1964. Heated fats and allied compounds as carcinogens, Acta
Path, et Microbiol. Scandinav., 61:161-180.
Berk, S., H. Ebert, and L. Teitell. 1957. Utilization of plasticizers and
related organic compounds by fungi. Ind. Eng. Chem., 49(7):1115-1124.
Booth, G.H. and J.A. Robb. 1968. Bacteria degradation of plasticized PVC-
effect on some physical properties. J. Appl. Chem., 18:194-197.
Cowan, J.C. 1980. "Drying Oils" in Kirk-Othmer Encyclopedia of Chemical
Technology, 3rd Edition, Volume 8, John Wiley and Sons, p. 130-150.
Darby, R.T. and A.M. Kaplan. 1968. Fungal susceptibility of polyurethane.
Appl. Microbiol., 16(6):900-905.
Gardner, H.A. and G.G. Sword. 1962. Paints, Varnishes, Lacquers, and Colors,
12th Edition. Gardner Laboratory, Inc., Bethesda, Maryland, pp. 23-24.
Johnson, R.W. Jr. 1978. "Fatty Acids from Tall Oil" in Kirk-Othmer Encyclo-
pedia of Chemical Technology, 3rd Edition, Volume 4, John Wiley and Sons,
P. 859-861.
Kieckebusch, W., K. Jahr, G. Czok, E. Degkwitz, and K. Lang. 1963. Physio-
logical properties of epoxidized oils. Fette-Seifen-Anstrichmittel, 65:919-
924.
63
-------
Klausmeier, R.E. 1966. The effect of extraneous nutrients on the biodeteri-
oration of plastics, in "Microbiological Deterioration in the Tropics",
SCI (Soc. Chem. Ind., London) Monogr., 23:232-243.
Kotin, P. and H.L. Falk. 1963. Organic peroxides, hydrogen peroxide, epox-
ides, and neoplasia. Radiation Research Supplement, 3:193-211.
Larson, P.S., J.K. Finnegan, H.B. Haag, R.B. Smith, Jr., and G.R. Hennigar.
1960. Chronic toxicity studies on two epoxidized soybean oils in the rat
and dog. Toxicol. and Appl. Pharmacol., 2:649-658.
Lazar, V. and M. loachimescu. 1973. Contributions to the study of rubber
biodeterioration. II. Researches concerning the fungus resistance of in-
gredients used in the rubber industry. Rev. Roum. Biol.-Botanique, 18(4):
227-233.
Lutz, J.T. 1980. "Epoxidation." in Kirk-Othmer Encyclopedia of Chemical
Technology, 3rd Edition, Volume 9. Interscience, New York. p. 251-266.
Mallette, F.S. and E. Von Haam. 1952. Studies on the toxicity and skin
effects of compounds used in the rubber and plastics industries: II. Plas-
ticizers. A.M.A. Arch. Indust. Hyg. Occup. Medicine, 6:231-236.
Modern Plastics Encyclopedia. 1979-1980. Chemicals and additives charts:
plasticizers. Volume 56, Number 10A, McGraw-Hill Incorporated, New York.
p. 685-697.
Mills, R.H. 1980. Monsanto Industrial Chemicals Co., St. Louis, Mo. Telephone
communication with Mr. Joseph M. Chaszar, October 24.
National Institute for Occupational Safety and Health (NIOSH). 1980. National
Occupational Hazard Survey - 1977; Quarterly Hazard Summary Report, March
1980. DHEW Publication No. 78-114. U.S. Department of Health, Education,
and Welfare, Washington, D.C.
National Institute for Occupational Safety and Health (NIOSH). 1978a. Regis-
try of Toxic Effects of Chemical Substances (RTECS). U.S. Department of
Health, Education, and Welfare, Washington, D.C.
National Institute for Occupational Safety and Health (NIOSH). 1978b. Pocket
Guide to Chemical Hazards. DHEW Publication No. 78-210. U.S. Department
of Health, Education, and Welfare, Washington, D.C.
National Institute for Occupational Safety and Health (NIOSH). 1978c. Cur-
rent Intelligence Bulletin 21: Trimellitic Anhydride (TMA). DHEW (NIOSH)
Publication No. 78-121. U.S. Department of Health, Education, and Welfare,
Washington, D.C.
Osmon, J.L., R.E. Klausmeier, and E.T. Jamison. 1969. The ability of selected
yeast cultures to degrade plasticized polyvinyl systems. in "Developments
in Industrial Microbiology", American Institute of Biological Sciences,
Washington, D.C. p. 447-452.
64
-------
Peakall, D.B. 1975. Phthalate esters: occurrence and biological effects.
Residue Rev., 54:1-41.
Potts, J.E., R.A. Clendinning, W.B. Ackart, and W.D. Niegisch. 1973a. The
biodegradability of synthetic polymers, in "Symposium on Polymers and Ecologi-
cal Problems," James Guillet, editor, Plenum Press, New York, p. 61-79.
Potts, J.E., W.B. Ackart, R.A. Clendinning, and W.D. Niegisch. 1973b. The
effect of chemical structure on the biodegradability of plastics. Lecture
presented at the Conference on Degradability of Polymers and Plastics, London,
England, November 27-28.
Rohm and Haas Co., 1980. Data included in letter to Thomas W. Lapp, July 1980.
Rodriquez, F. 1971. The prospects for biodegradable plastics. Chem. Tech.,
July: 409-415.
Saeger, V.W., R.G. Kaley II, 0. Hicks, E.S. Tucker, and J.P. Mieure. 1976.
Activated sludge degradation of adipic acid esters. Appl. Environ. Microbiol.,
31(5):746-749.
Sigan, J. 1980. Sherex Chemical Co., Columbus, Ohio. Telephone communica-
tion with Mr. Charles E. Mumma, June 4.
Sittig, M. 1979. Hazardous and Toxic Effects of Industrial Chemicals. Noyes
Data Corporation, Park Ridge, N.J.
Thompson, C.M. 1977. Industrial process profiles for environmental use:
Chapter 13. Plasticizer industry. EPA-600/2-77-023m. Environmental Protec-
tion Agency, Industrial Environmental Research Laboratory, Cincinnati, Ohio.
Tate, D.C. 1969. "Tall Oil." in Kirk-Othmer Encyclopedia of Chemical Tech-
nology, 2nd Edition, Volume 19, John Wiley and Sons, p. 614-629.
Weil, C.S., N. Condra, C. Haun, and J.A. Striegel. 1963. Experimental car-
cinogenicity and acute toxicity of representative epoxides. Amer. Indust.
Hyg. Assoc. Journal, 24(4):305-325.
Wilkinson, R.R., H.V. Ellis, and F.G. Hoffmeister. 1979. Assessment of
the need for limitations on triaryl and trialkyl/aryl phosphates. Draft Final
Report. Environmental Protection Agency, Office of Pesticides and Toxic
Substances, Washington, D.C.
65
-------
30J72-TO'
REPORT DOCUMENTATION
PAGE
j 1. REPORT NO.
I "EPA 560/2-81-006
3. Recipient*« Accession No.
4. THIe and Subtitle
A Survey of Plasticizers: Epoxies, Linear Polyesters and
Trimellitates
5. Report Date
November 1981
7. Authorts)
Thomas W. Lapp, Charles E. Mumma, Joseph Chaszar
a. Performing Organization Rept. No.
». Performing Organization Neme and Address
Midwest Research Institute
425 Volker Blvd.
Kansas City, Missouri 64110
10. Projeet/Task/Work Unit No.
Task VI
11. Contract(C)
(C) 68-01-3896
(C)
No.
12. Sponsoring Organization Name and Address
Environmental Protection Agency
Office of Pesticides and Toxic Substances
Washington, D.C. 20460
13. Type of Report & Period Covered
Final Report
14.
IS. Supplementary Note*
Roman Kuchkuda, Project Officer
16. Abstract (Limit: 200 words)
Study investigated the published literature for selected areas in three classes
of plasticizers: epoxies, linear polyesters, and trimellitates. Areas of interest in-
cluded physical and chemical properties, production and use, health and environmental
effects, and plasticizer interchangeability. Current production methods, sites, and
annual volumes are presented for each class. Little information is available in the pub-
lished literature on health effects,, No occupational standards exist for any of the
three classes. Unpublished health data are available for selected tests from manufacturers
for specific materials„ Environmental effects were limited to static fish toxicity
studies for two trimellitates and studies of fungal and bacterial growth with epoxies
and linear polyesters. Both classes of plasticizers are very susceptible to fungal and
bacterial attack.
17. Document Analysis a. Descriptors
Epoxies Production
Polyesters Exposure
Trimellitates Toxicity
Plasticizers Plastics Additives
b. Identlfiers/Open-Ended Terms
Degradation
c. COSATI Reid/Group
18. Availability Statement
Unlimited Distribution
19. Security Class (This Report)
Unclassified
20. Security Class (This Page)
Unclassified
21. No. of Pages
72
22. Price
OBTIONM. FORM Z72 (4-7T)
------- |