EPA-560/1-76-002
CHEMICAL TECHNOLOGY AND
ECONOMICS IN
ENVIRONMENTAL PERSPECTIVES
TASK 1-TECHNICAL ALTERNATIVES TO SELECTED
CHLOROFLUOROCARBON USES
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF TOXIC SUBSTANCES
WASHINGTON, D.C. 20460
FEBRUARY 1976
-------
EPA-560/1-76-002
CHEMICAL TECHNOLOGY AND ECONOMICS IN
ENVIRONMENTAL PERSPECTIVES
Task I - Technical Alternatives to Selected
Chlorofluorocarbon Uses
Contract No. 68-01-3201
Project Officer
Irving J. Gruntfest
Office of Toxic Substances
Environmental Protection Agency
Washington, D.C. 20460
Prepared for
Environmental Protection Agency
Office of Toxic Substances
Washington, D.C. 20460
February 1976
-------
NOTICE
This report has been reviewed by the Office of Toxic Substances, Environ-
mental 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 tradenames or commercial
products is for purposes of clarity only and does not constitute endorse-
ment or recommendation for use.
-------
PREFACE
This report presents the results of Task I, "Technical Alterna-
tives to Selected Chlorofluorocarbon Uses," of a project entitled "Chemical
Technology and Economics in Environmental Perspectives." The project is per-
formed by Midwest Research Institute (MRI) under Contract No. 68-01-3201
for the Officer of Toxic Substances of the U.S. Environmental Protection
Agency. Dr. Irving J. Gruntfest is the project officer. This program has
MRI Project No. 4101-C.
Particular credits for authorship of portions of this report in-
clude the following: Mr. G. J. Hennon, Chapters III and VIII-A; Ms. Kathryn
Lawrence, Chapters IV, VI-B, and VIII-B; Mr. Howard Gadberry, Chapter V;
Dr. Roderick E. Athey, Chapter VII; Dr. Ivan C. Smith and Dr. Thomas W.
Lapp, Chapters II, VI-A, and VIII-C,D. Ms. Cassandra Collins provided tech-
nical assistance for portions of this study. Dr. Lapp is the project leader
for this contract. This program is under the supervision of Dr. Edward W.
Lawless, Head, Technology Assessment Section, Physical Sciences Division.
MRI expresses its sincere appreciation to the many companies and
organizations that provided technical information for this report.
Approved for:
MIDWEST RESEARCH INSTITUTE
L. J/ Shannon, Assistant Director
Physical Sciences Division
March 11, 1976
iii
-------
TABLE OF CONTENTS
Page
I. Introduction 1
References to Section I ••• 4
II. Summary and Conclusions 5
A. Aerosols - 5
B. Refrigeration 8
C. Foam Blowing Agents 9
D. Cleaning and Drying Solvents 12
III. Refrigeration and Air Conditioning ..... 14
A. Chlorofluorocarbon Systems for Refrigeration
and Air Conditioning 14
B. Alternatives to the Chlorofluorocarbons Under
Study 19
C. Applicability of Alternatives in Specific Use
Areas 34
V
References to Section III . . 43
IV. Aerosol Industry 46
A. Development of the Aerosol Industry, Aerosol
Systems and Propellants 46
B. Evaluation of Alternative Systems ...... 55
C. Alternatives for Specific Use Areas 70
D. Production Data on Chemical Alternatives . . 77
References to Section IV 82
V. Foam Blowing Agents 85
A. Major Types of Plastic Foams 85
B. Pdlyurethane Foams - Rigid and Flexible ... 87
C. Polystyrene Foams 115
D. Olefin Foams 118
E. Other Plastic Foams 119
F. Aggregate Consumption of Fluorocarbon Foam
Blowing Agents 119
-------
TABLE OF CONTENTS (concluded)
G. Fluorocarbon Emissions from Foams ...... 122
H. Alternatives to Use of Fluorocarbon Blowing
Agents 128
References to Section V 139
VI. Cleaning and Drying Solvents 143
A. Chlorofluorocarbons Used in Cleaning and
Drying 144
B. The Dry-Cleaning Industry 156
References to Section VI 161
VII. Potential Rankine Cycle Uses 162
A. End Uses 162
B. Criteria for Selection of Working Fluids . . 166
C. Identification of Possible Alternate Working
Fluids for Rankine Cycle Engines 169
D. Advantages and Disadvantages of Selected
Working Fluids for Rankine Cycle Turbines
by End Use 172
References to Section VII 181
VIII. Direct Economic Consequences of Limitations on
Chlorofluorocarbons 183
A. Refrigeration 183
B. Aerosols 187
C. Foam-Blowing Agents 189
D. Degreasing and Dry Cleaning 190
References to Section VIII 193
Appendix A - Chlorofluorocarbon Manufacturing Processes .... 195
Appendix B - The ROVAC Automotive Air Conditioning System . . . 206
Appendix C - Persons and Firms Contacted 217
vi
-------
List of Figures
Figure Title • Page
V-l Urethane Foam: Raw Materials Supply Structure 1972
Estimated Materials Flows, Million Pounds 88
V-2 Flexible Urethane Foam Industry Structure 1972
Estimated Materials Flows, Million Pounds 89
V-3 Rigid Urethane Foam Industry Structure 1972 Estimated
Materials Flows, Million Pounds ... 91
V-4 Effect of Water Concentration on Foam Density .... 97
V-5 Effect of Fluorocarbon-11 Concentration on Foam
Density . 98
V-6 Conductivities of 1.8 In. Thick Cooler Panel Cores:
Through the 40°F Face 125
A-l Fluorocarbons and Their Principal Precursors 197
vii
-------
List of Tables
Table Title Page
II-1 Production and Major Uses of Chlorofluorocarbons
(1972) 6
III-l Chlorofluorocarbon Use as Refrigerants by End-Use
Products , . 15
III-2 Estimated Chlorofluorocarbon Use as Refrigerants .... 16
III-3 ASHRAE Refrigerants 20
III-4 Materials for Consideration as Alternative Refrigerants. 25
III-5 Applicability of Alternatives to End-Use Products ... 36
IV-1 Usage of Fluorocarbons in Aerosols ...... 51
IV-2 Physical and Chemical Properties, Toxicity and
Reactivity of Candidate Aerosol Propellants 53
IV-3 Eliminated Candidate Propellants 61
IV-4 Propellant Usage in Nonpersonal Product Aerosols .... 75
IV-5 Summary and Relative Ranking of Alternative Delivery
Systems for Aerosol Products 76
IV-6 Annual Production of Alternative Chemical Propellants
and Blend Constituents 78
IV-7 Manufacturers of Propellant Blend Constituents and
Alternative Propellants 80
V-l Types and Applications of Plastic Foams 86
V-2 Rigid Urethane Foam Consumption 93
V-3 Typical Rigid Urethane Foam Systems 100
V-4 Thermal Conductivity of Blowing Agent Vapors 102
i
V-5 Change in K-Factor of Foams on Aging 104
V-6 Effect of Blowing Agents on Thermal Conductivity .... 106
viii
-------
List of Tables (continued)
Table
V-7
V-8
V-9
V-10
V-ll
V-12
V-13
V-14
Title
Cellular Gas Analysis from Rigid Urethane Bun Stock
Flexible Urethane Foam Consumption (Million Pounds) .
Typical Formulations for Flexible Urethane Foams . . .
Cooling Effect of Fluorocarbon- 11 in Resilient Foam
Slabs
Estimated U.S. Use of Halocarbons in Plastic Foam
Fluorocarbon Usage in Plastic Foams 1973 and 1974 . .
Fluorocarbon- 11 Concentrations in Breathing Zone Air
Page
107
108
111
113
114
121
123
126
V-15 Fluorocarbons in Foam- -Consumption and Atmospheric
Emissions ..................... 127
V-16 Potential Substitutes for Fluorocarbon Blown Plastic
Foams ....................... 129
V-17 Foam Formulations and. Properties ..... ....... 134
V-18 Volatile Compounds Boiling Point -40 to 100°C .... 135
VI- 1 Properties and Costs of Commonly Used Solvent Vapor
Degreasing Materials ................ 146
VI-2 Typical Applications for Vapor- Degreasing Solvents . . 148
VI-3 Common Solvents and Degreasing Systems as Possible
Alternatives ................... . 152
VI-4 Typical Alkaline Cleaner Formulations for Various
Metals ....................... 154
VI-5 Commonly Used Dry-Cleaning Solvents ......... 158
ix
-------
List of Tables (concluded)
Table Title
VII-1 List of Priority Pollutants Developed at the
Intergovernmental Meeting on Monitoring at
Nairobi (1974) 163
VII-2 Rankine Cycle Turbines in Operation Using Working
Fluids Other Than Steam 164
VII-3 Alternative Working Fluids for Rankine Cycle Engines . 170
VII-4 Underwriters' Laboratories Classification of Compara-
tive Hazard to Life of Gases and Vapors 173
V1I-5 Classification of Comparative Flammabilities of
Various Gases and Vapors ........ 174
VIII-1 Comparative Cost of Alternative Delivery Systems . . . 188
A-l The Chemistry of Fluorocarbons and Their Principal
Precursors 198
A-2 Fluorocarbon ;and Fluorocarbon Related Production and
Employment 201
A-3 Companies Producing Fluorocarbons and Their Principal
Precursors 203
A-4 Location of Production Facilities for Fluorocarbons
and Their Principal Precursors 204
-------
I. INTRODUCTION
The technical controversy over whether certain chlorofluorocarbons
are destroying the earth's protective ozone shield has evolved into a public
issue. Legislation is being considered at both the state and federal levels
to limit or ban the use of certain of these materials. One state, Oregon,
has banned the use of chlorofluorocarbons in aerosol sprays. While many tech-
nical questions remain to be answered, responsible public officials and reg-
ulatory agencies are trying to determine the technical and economic feasibil-
ity of using alternative chemicals, or of substituting mechanical devices
where suspect chlorofluorocarbons are now used. In addition, they seek to
identify the social consequences resulting from the regulation of the various
uses of these materials.
There have been numerous articles in the press referring to the
ozone problem and to the effects of "fluorocarbons" on the depletion of the
ozone layer. In actuality, the problem is not specifically fluorocarbons but
rather the highly halogenated chlorofluorocarbons or broraofluorocarbons.
The terra "Freons" has also been widely used in the press as being synonymous
with fluorocarbons, which of course it is not since the term Freon® is the
registered tradename of E. I. du Pont de Nemours and Company.
It should also be noted that not all chlorofluorocarbons or fluoro-
carbons are considered to be equally detrimental to the ozone layer nor are
these compounds the only halogenated species in the upper atmosphere. In
very simplified terms, the current controversy lies in the mechanism and rel-
ative rates of chlorine and fluorine atom (free radical) reaction with ozone.
In the upper atmosphere, the halogenated species undergoes photodissociation
to produce halogen atoms (free radicals), which in turn react with ozone.
Originally it was thought that the chlorine reacted with ozone at a rate 10
times faster than fluorine. The five halocarbons covered in this report are
considered to be the most potentially hazardous of the commercially produced
chlorofluorocarbons. Other commercially available chlorofluorocarbons are,
at this time, considered to be less hazardous to the ozone layer and, in
fact, some of these materials have been suggested in this report as possible
alternative materials.
Previous studies on chlorofluorocarbons have dealt with stratos-
pheric effects, biological effects of ozone reduction, human health effects,
and a review of the "fluorocarbon industry.'ll' A study conducted by Syracuse
University Research Corporation offered an appraisal of environmental hazards
of one and two carbon fluorocarbons^/ Another major study was a preliminary
impact-assessment of possible regulatory actions to control atmospheric emis-
sion of selected halocarbons..2' It covered 24 halocarbons and identified the
major emission sources of those compounds, analyzed alternatives for emission
abatement, named the affected industries, and assessed economic impacts of
-------
regulatory options. Reports on other studies by the Department of Commerce
on the economic and energy impacts of chlorofluorocarbon regulation were
not received in time for use on this program.
The objective of this study is to explore more deeply than have
any previous studies, the technical feasibility of and the basic economic
considerations related to use of alternative chemicals or mechanical de-
vices for supplying the public with the same goods and services now pro-
vided by five of the most potentially hazardous halocarbons: chlorofluoro-
carbons 11 (CC13F), 12 (CC12F2), 13 (CClFs), 113 (CC12FCC1F2), and 114
(CC1F2CC1F2).
This study was not directly concerned with the current ozone de-
pletion hypothesis or the environmental ramifications of that hypothesis.
No judgment on that question should be inferred from any of the results of
this study.
Four major-use categories (aerosols, refrigerants, solvents, and
foam-blowing agents) account for over 9970 of the usage of these five mate-
rials. Working closely with industry, MRI has appraised the various substi-
tute chemicals or alternative mechanical devices available in each of the
use areas, should the five chlorofluorocarbons be banned from further use.
Recovery of these chemicals, although a plausible solution for some emis-
sion problems was not to be considered a viable solution in this study.
Costs while of intense concern to the consumer, were not used to exclude
technologically feasible alternative systems from further consideration in
this study. Willingness of the public, once informed of the risks, to forego
certain conveniences afforded by these chemicals, should not be underesti-
mated, as evidenced by the 25% reduction in sales of chlorofluorocarbon
propelled aerosols over the last few months. The following sections of this
report contain a discussion of technical feasibility for the use of the
alternatives in each of the four use categories.
This document is intended to be useful to policy makers and the
public in understanding the technical difficulties associated with the de-
velopment of acceptable substitutes for these five materials if or when
their use should be limited. It is organized in a manner which we feel will
provide the greatest utility to the reader. The next section summarizes the
findings of this study and presents conclusions with respect to the utiliza-
tion of alternative systems for each of the four present use areas of the
selected chlorofluorocarbons. Sections III through VI present a detailed
discussion of the technical considerations to be applied to alternative sys-
tems in the areas of refrigeration and air conditioning, aerosols, foam-
blowing applications, and degreasing and dry cleaning solvents. For each
of these areas, a range of alternatives are shown but no attempt has been
made to select the single most attractive alternative or to provide any
rank-order of the alternatives. Section VII is concerned with a possible
-------
major future use of chlorofluorocarbons, (i.e., in the Rankine Cycle engine)
and with the numerous alternatives which may be possible in this use area.
Section VIII presents a brief summary of some economic considerations for
.the use of various alternative systems in each of the four areas discussed
in Sections III through VI.
Throughout the literature and in discussions with personnel di-
rectly involved with the industry, several different designation systems
are widely used for the halocarbons. In all industries, other than refrig-
eration, the halocarbons are given an F prefix (fluorocarbon) while in the
refrigeration industry the compounds have an R prefix (refrigerant). In
some instances, the designation "P" is also used to denote an aerosol pro-
pellant but finds very limited usage. Thus, F-12 and R-12 are the same com-
pound, only the prefix has been changed to reflect the views of two differ-
ent industries. In addition, certain company trademark names such as Freon®
(E. I. du Pont de Nemours and Company), and others are used.
Information was received from a major manufacturer immediately
prior to the printing of this final report regarding two of the compounds
(F-21 and F-115) suggested as possible alternative materials. Current leg-
islative attitudes indicate that any chlorofluorocarbons not containing
hydrogen or a carbon-carbon double bond may be restricted in the future.
If this materializes, F-115 would be included in such action and would not
be acceptable as an alternative. Recent toxicological data on F-21 indi-
cates that the material probably is not acceptable as an aerosol propellant.
Since this information became available at a time which made the deletion
of these two compounds from the report impractical, references to these
materials do remain in the report. However, any suggested use of F-21 and/
or F-115 in this report should be modified in view of the above comments.
-------
REFERENCES TOSECTION I
1. "Fluorocarbons and the Environment," Report of Federal Task Force on
Inadvertent Modification of the Stratosphere (IMOS), Council of En-
vironmental Quality, June 1975; GPO No. 038-000-00226-1.
2. Howard, P. H., P. R. Durkin, and A. Hanchettj "Environmental Hazard
Assessment of One and Two Carbon Fluorocarbons," EPA Contract No.
68-01-2202, Technical Report TR-74-572 (1974); EPA Report No. 560/2-
75-003; NTIS No. PB-247-419.
3. Arthur D. Little, Inc., "Preliminary Economic Impact Assessment of
Possible Regulatory Action to Control Atmospheric Emissions of Se-
lected Halocarbons," EPA Contract No. 68-02-1349, Task 8, Publica-
tion No. EPA-450/3-75-073, September 1975;,NTIS No. PB-247-115.
-------
II. SUMMARY AND CONCLUSIONS
The chlorofluorocarbons are a unique class of chemicals, many
of which are chemically very unreactive, nearly odorless, and generally
nontoxic and nonflammable. Because of these properties and the desirable
boiling points, vapor pressure characteristics and solvent properties ex-
hibited by certain chlorofluorocarbons, they are in demand for a broad
range of domestic and industrial uses and are manufactured in quantities
of millions of pounds. The higher costs of these compounds, compared to
those of available substitute chemicals has been accepted by the public
in preference because they pose lower health risks and afford greater con-
venience in certain applications.
The purpose of this study was to identify technically feasible
substitute chemicals and/or alternative methods of delivering the goods
and services now provided through the use of suspect chlorofluorocarbons.
This study did not involve an assessment of the risks associated with en-
vironmental discharge of these commercially important chemicals. While
cost was not a primary consideration in selecting substitute chemicals
and/or alternative methods of delivering a good or service, it was not
ignored in this study. This study was limited to the five commercial com-
pounds believed to pose the greatest environmental threat (F-ll, F-12,
F-13, F-113 and F-114), if the current ozone depletion hypothesis is found
to be correct.
It has recently been stated in a governmental report that four
categories currently account for 99% of the usage of these compounds
(Table II-1).
Thus, the focus of this study is on those four use categories.
The Kankine Cycle engine is an application which appears to have consider-
able potential for expanding future usage of chlorofluorocarbons. Alterna-
tive chemicals for this application are discussed.
The results of this study can be summarized as follows for each
use category.
A. Aerosols
Chlorofluorocarbons F-ll and F-12 account for over 92% of the
chlorofluorocarbons used in the aerosol industry. Fluorocarbon-propelled
products are widely used by the public as convenience items (convenience
for which they are willing to pay). However some products would not be
on the market if aerosols were not available. Similar goods or services can
-------
TABLE II-1
a/
PRODUCTION AND MAJOR USES OF CHLOROFLUOROCARBONS (1972)-
, . Aerosol . Foaming,
0^ c/ c/
Compound Sales— Production Propellants Refrigerants Solvents— Agenta-
F-ll
F-12
285
415
300
439
215
220
F-13 Very small NA
F-113 50
F-114 20
700
809
20
451
19
132
Very small
10
161
50
55
50
45
95
_a/ In millions of pounds.
b/ Sales 5% less than production because of inventory buildups.
cl No authoritative data are available regarding the precise quantities
used for various applications or the compound used. Estimates
shown throughout this report may vary according to the sources.
Source: "Fluorocarbons and the Environment," Report of Federal Task Force
on Inadvertent Modification of the Stratosphere (IMOS), Council
on Environmental Quality and Federal Council for Science and
Technology, 1975.
-------
often be provided without chlorofluorocarbons. Sales of chlorofluorocarbon-
propelled aerosols have significantly diminished during the past few months.
This reduction in sales can be attributed for the most part to the economic
recession and is also partly indicative of the public's concern about poten-
tial environmental hazards of these materials, and indicates a willingness
to accept alternate products in some categories. It also reflects the tech-
nical and marketing efforts of some manufacturers of aerosol products.
In this study, potential substitute propellants or alternative
methods of delivering current aerosol products are identified and eval-
uated. Aerosol application of chlorofluorocarbons are shown to fall into
three categories: (a) personal products, (b) household products, and
(c) other products.
Personal products account for about 84% of the chlorofluorocarbon
propellants used in aerosols; hair care, antiperspirants and deodorants
account for 90% of that usage. Low flammability, low toxicity, and absence
of objectionable odor are important criteria for alternative candidates
to meet in these applications. Alternatives identified include other chloro-
fluorocarbons, hydrocarbons, compressed gases and mechanical devices.
Other chlorofluorocarbons that could be used as alternatives to
the F-ll and F-12 in such products include F-115,* FC-318, F-21,** F-22,
F-142b and F-152a. Candidates still in the experimental stage include:
F-123, F-124, F-132b, and F-133a. F-218, F-227a and F-3110 would require
a longer time period for development and implementation. Most of these
materials could be as much as 5 to 10 times more expensive than F-ll or
F-12.
Hydrocarbons afford the necessary performance features and are
much less expensive; however, their high flammability will deter their
usage for this purpose, unless suitable flammability suppressants can
be found. Shaving lathers are largely propelled by hydrocarbons already,
and could be converted entirely to these propellants with little problems.
Compressed gases such as nitrogen, carbon dioxide, and nitrous
oxide will require larger cans for the same net weight of product and
lower proportions of active product. Industry advances in nozzle design
and incorporation of various mechanical features (bladder systems, pistons,
etc.) may allow substitution of compressed gases for some uses of chloro-
fluorocarbons if the market is willing to accept some tradedown in using
these systems. Mechanical substitutes appear to be feasible and acceptable
alternatives for delivery of some personal care products (hair sprays,
most medicinals, pharmaceuticals, colognes and perfumes).
* Industrial sources recently indicated that F-115 may not be an acceptable
substitute because it contains only chlorine and fluorine, as such, may
be restricted in future legislation.
** A major manufacturer states that recent toxicological data indicate that
F-21 probably is not acceptable as an aerosol propellant.
-------
Household products such as room deodorants, cleaners, laundry prod-
ucts and waxes and polishes account for only 6% of the chlorofluorocarbon-
propellant usage in aerosols. In household applications, about 80% of the
aerosol propellants are, hydrocarbons. Some chlorofluorocarbons, hydrocar-
bons, and compressed gases could be substituted for the small usage of F-ll
and F-12 in these aerosol applications. However, some increased consumer
costs and/or hazards associated with the chemical alternatives could preclude
extensive substitution in aerosol products where they are now used. Mechanical
delivery systems, already widely used for these products, appear to be accept-
able alternatives.
Other products delivered with chlorofluorocarbon propellants account
for about 10% of all chlorofluorocarbon-propelled products. These products
consist predominantly of pesticides, coatings (e.g., paints), and industrial
products. Hydrocarbons are now the most widely used propellants for these
products, and could almost entirely replace the chlorofluorocarbons for these
uses. The more expensive chlorofluorocarbons F-115 and FC-318 have been and
are currently used (in relatively small total amounts) in combination with
^0 in some food products such as whipped cream. Mechanical delivery systems
are also available for most products in this category.
B. Refrigeration
Mechanical vapor-compression systems (MVCS) are used for both re-
frigeration and air conditioning, and account for the vast majority of the
refrigerating capability in the U.S. While sulfur dioxide and ammonia were
once used as the refrigerants in these systems, chlorofluorocarbons are now
used almost exclusively, because of their much lower toxicity and nonflam-
mability as compared to ammonia. However, ammonia is still used in some com-
mercial applications. If the use of F-ll, F-12, and F-114 as refrigerants
were to be banned two alternative approaches are available: (a) employ other
working fluids in MVC refrigeration, or (b) substitute other refrigeration
systems for MVC.
1. Chemical substitutes: Flammability, toxicity, chemical stabil-
ity and thermodynamic properties are primary considerations in selecting re-
frigerant fluids. Chemicals having the necessary physical and thermodynamic
properties have been identified and evaluated. The 1972 ASHRAE Handbook of
Fundamentals lists 78 potential refrigerants, including the chlorofluorocarbons.
At present, F-ll, F-12, and F-114 account for over 65% of the working fluids
used in MVC systems. One additional chlorofluorocarbon, F-22, is also used ex-
tensively as a refrigerant, and could replace F-ll and F-12 in many applica-
tions but only with substantial redesign of refrigeration systems. Redesign
and retooling would.be expensive and industry is showing a considerable re-
luctance to redesign or retool to use F-22 as a refrigerant until they are as-
sured that it too will not be banned.
-------
While many other chemicals may be technically feasible alterna-
tive refrigerants, few of them are near-term candidates, either because
their physical and chemical properties are not suitable or are undeter-
mined, adequate production capabilities are not available or production
costs would be very high. Many of them would probably not be allowed for
use in the home because of their toxicity or flammability, and Occupa-
tional Safety and Health officials may restrict their use in the indus-
trial workplace for the same reasons.
2. Alternative refrigeration systems; Alternative refrigera-
tion systems could be used in place of F-ll and F-12 mechanical vapor
compression systems for some applications. The lithium bromide absorption
system, while less efficient than MVC, could be used in air conditioning
and chilling applications.
Ammonia-absorption systems could be used for low temperature
applications, but have the same drawbacks as ammonia-MVC (i.e., toxicity
and -explosiveness).
Thermoelectric cooling is far too expensive except for a few
specialized applications.
Steam-jet refrigeration has limited potential as a substitute
refrigeration system for a few industrial applications.
The air cycle, although showing some potential as a long-term
substitute, has little or no short-term capabilities as a replacement
refrigeration system.
C. Foam Blowing Agents
Approximately 78 million pounds of chlorofluorocarbons were
used in 1974 for the fabrication of plastic foams. This application ac-
counts for about 8% of U.S. chlorofluorocarbon usage.
By far the largest usage of chlorof luorocarbon blowirlg agents
is in the manufacture of rigid polyurethane foams—mainly for thermal
insulation. Nearly 46 million pounds of F-ll, F-12 and other chlorofluoro-
carbons are used annually for rigid urethanes. Because a substantial por-
tion of the chlorofluorocarbon vapors are trapped within the closed cells
of rigid urethane foams, emissions to the atmosphere are significantly
lower than annual consumption. Total cumulative emissions of chlorofluoro-
carbons to the atmosphere since 1965 from all types of plastic foams is
estimated to be 256 million pounds, or roughly one half of the 483 mil-
lion pounds estimated cumulative consumption in foam blowing.
-------
Flexible urethane foams are the second largest user of chloro-
fluorocarbon blowing agents. Current consumption for flexible urethanes
is about 23 million pounds. Chlorofluorocarbons are used to augment blow-
ing from the carbon dioxide released by reaction of free isocyanate groups
with water. The additioh of chlorofluorocarbdns as auxiliary blowing agents
improves the softness of the foams and reduces hysteresis. Virtually all
of the chlorofluorocarbons used in preparing flexible foams is lost to
the atmosphere soon after manufacture. A few 'other plastic foams consume
small quantities of chlorofluorocarbons, although most other foams employ
other means of foam generation. Polystyrene foamed sheet and film pre-
pared from crystal consume nearly 7 million pounds of chlorofluorocar-
bons. Polyethylene foams account for about 1 million pounds of F-114,
F-ll, and F-12. Because other blowing agents (such as pentanes or methyl-
chloride) can also be used to prepare styrene and olefin foams, only about
5% of styrene beads and extruded board stock use any chlorofluorocarbon,
but 40 to 45% of foamed styrene sheet is chlorofluorocarbon blown. Vir-
tually all chlorofluorocarbons used in styrene, olefin, and miscellaneous
foams diffuse promptly into the atmosphere.
The distribution of chlorofluorocarbons used in plastic foams
is not known with accuracy. One major producer of foams estimated the
distribution for 1973 to be:
F-ll 53 million pounds
F-12 10 million pounds
F-113 ~ 2 million pounds
F-114 -' 3 million pounds
Total »' 68 million pounds
Fluorocarbon producers indicate that while the distribution
given is reasonable, total consumption is. probably 20 million pounds
greater. The major difference is believed to lie in substantially higher
use of F-12, which has grown quite rapidly in recent years.
Blowing agents represent a relatively small part of the total
cost of plastic foams. At current prices the nearly 78 million pounds
used in 1973 cost about $30 to $35 million; or roughly 4 to 6% of the
cost of raw ingredients for foam products.
At present, chlorofluorocarbons are considered essential in
achieving the low thermal conductivity required for insulating foams.
Because low density foams (2 Ib/cu ft) consist of about 97% vapor within
the cells and 37» actual polymer, it is the thermal conductivity of the
cell gases that determines the overall K-value. While a few other vapors
10
-------
(mostly halogenated compounds) are known that compare with the low thermal
conductivity of the chlorofluorocarbons, these agents are not used in pre-
paring insulating foams. It is conceivable that compounds other than the
chlorofluorocarbons now used almost totally for insulating foams could be
developed for use as auxiliary blowing agents, but to date no satisfac-
tory substitute has been found. The cost of possible alternative blowing
agents could be lower than or higher than the chlorofluorocarbons that
are used at present. If all of the 78 million pounds of chlorofluorocar-
bons used for foams were eventually replaced by alternative blowing agents
costing 80<£/lb more (roughly three times present prices), the increased
cost would total $62 million. For the present annual output of 898 million
pounds for all types of fluorocarbon-blown foams, the added cost to the
user would amount to about 7.0$/lb of foam. If chlorofluorocarbon usage
in thermal insulating foams were curtailed, foam usage in building con-
struction, refrigeration appliances, and industrial applications would
be affected markedly. The froth-pouring and froth-spraying techniques
which are growing in usage are believed to depend on the use of F-12 to
preexpand the plastic froth. Because foamed insulation is currently quite
cost-effective (especially for the application of insulation to already
existing structures and equipment), elimination of foamed insulation could
inhibit efforts toward energy conservation.
Chlorofluorocarbon blowing agents are also considered essential
in preparing soft flexible urethane foams. Without the use of auxiliary
blowing agents, flexible urethanes do not provide the cushioning charac-
teristics on which most automotive seating and some furniture applications
are based. The use of chlorofluorocarbons may also be required at present
in making super-soft foams for textile lamination, and for self-bonded
rug backings.
t •
A wide variety of other materials compete with plastic foams
for most applications. In fact, most of the present uses of plastic foams
represents a displacement by the foams of some material formerly used
for the same purpose. Substitution of .alternate materials would be pos-
sible for many products and applications now consuming plastic foams.
In many cases, such substitution would mean less satisfactory or more
costly products. For certain major uses of plastic foams, there is no
presently available substitute.
While it might be practical to develop systems to collect, purify
and reuse chlorofluorocarbons lost during plastic foam production in plants,
no proven system now exists for chlorofluoroca'rbon recovery.
11
-------
D. Cleaning and Drying Solvents
These applications represent areas in which the chlorofluorocar-
bons play a relatively minor role both with respect to the total quantity
of solvents used and to the overall consumption of chlorofluorocarbons
among the four use areas. The consumption of F-113 and F-ll represents
only approximately 6% of the total solvents used for cleaning and only
about 9% of the total production of the five chlorofluorocarbons.
The use of F-113 and F-ll in the cleaning field is concentrated
in the electrical and electronics industries for the defluxing of printed
circuit boards (PCB's) and the cleaning of electronic parts, electric mo-
tors, and delicate electronic and scientific instruments. The solvents
are also used to clean metal, glass, plastics, and electroplated manufac-
tured articles. A dry-cleaning process compatible with plastic buttons
and trim, furs, and leathers has been commercialized. F-113 and F-ll based
cutting fluids are widely used in industries such as aircraft construction
where precise tolerances must be maintained.
F-113 based displacement drying fluids are used to remove water
from metal, glass, plastic or plated parts to leave a spotless, residue-
free surface.
Designation of suitable alternatives for the current uses of F-113
and F-ll is difficult since each cleaning or drying problem is unique. Factors
that would influence the selection of a specific solvent are:
* Type of soil and/or grease to be removed—.ionic, nonionic or
both, particulate matter.
* Compatibility of the object cleaned or dried with the solvent.
* Temperature limits of the objects to be treated.
* Solvent flammability.
* Employee safety vis-a-vis solvent exposure—toxicity, TLV OSHA
regulations, effect on skin, and eyes.
* MIL Specifications governing choice of solvent and cleanliness
tests for the finished article.
* Solubility in water.
* Energy requirements.
* Reject rates.
* Waste disposal.
* Cost/unit object cleaned or dried.
In theory, the existing chlorocarbons, aqueous detergent or cleaner
systems and flammable solvents could be potential substitutes for some of the
current applications of F-113 and F-ll. The chlorocarbons to be considered as
12
-------
possible substitutes would include perchloroethylene, methylene chloride,
methyl chloroform, and trichloroethylene. Flammable solvents that may have
some application include alcohols, ketones, and hydrocarbons; however, the
use of these solvents would not be recommended for more general usage un-
less they are used in very small quantities under very controlled conditions.
Regardless of the alternative solvent, certain tradeoffs will occur among
the various factors stated earlier and the ultimate alternative should rep-
resent the most viable material with respect to overall system efficiency,
worker safety, environmental effects, and otheridirect concerns.
As a dry-cleaning agent, F-113 is used primarily for specialty
cleaning with only very minor usage in other areas. Other currently used
dry-cleaning solvents, such as Stoddard solvent or perchloroethylene, can
be used as substitutes for the cleaning of these specialty items.
To reiterate an earlier statement, the designation of suitable
alternatives for the current uses of chlorofluorocarbon cleaning and dry-
ing agents is difficult since each problem is unique. The potential of a
possible substitute must be considered on a case-by-case basis considering
the relative advantages and disadvantages of each solvent.
13
-------
III. REFRIGERATION AND AIR CONDITIONING
This section of the report addresses the technological considera-
tions for using alternative chemical and/or mechanical substitutes for se-
lected chlorofluorocarbons in refrigeration and air conditioning.
General considerations in choosing a refrigeration system can be
summarized as follows: (a) is the system technically capable of operating
in the temperature and humidity range necessary for the contemplated end
uses ("engineered refrigeration systems may have as a target any dry-bulb
temperature from +70°F (+21°C) down to -240°F (-151°C), and relative humid-
ities from 95 to 0%, or combinations of dry-bulb and humidity control".!');
(b) is the system adaptable to the physical size limitations of the end-use
product; (c) is the system economically feasible in terms of initial invest-
ment and operating costs; and (d) is the system safe to use in the desired
application?
The report discusses chlorofluorocarbon systems for refrigeration
and air conditioning, describes alternatives to the chlorofluorocarbons under
study, and describes the applicability of alternatives in specific use areas.
A. Chlorofluorocarbon Systems for Refrigeration and Air Conditioning
Chlorofluorocarbons are used in mechanical vapor compression re-
frigeration and air conditioning applications. A breakdown by end-use prod-
ucts is shown in Table III-l. Inasmuch as this program is primarily concerned
with alternatives for R-ll, R-12, R-13, R-113 and R-114,* the use breakdown
translates to the fact that R-ll and R-12 account for more than 88% (Table
III-2) of refrigerant usage for the five chlorofluorocarbons under study.
Accordingly, the largest portion of effort in this part of the study has
dealt with possible alternatives to R-ll and R-12 refrigerant systems.
1. Mechanical vapor compression systems; The ASHRAE Guide and
Data Book provides detailed descriptions of reciprocating, rotary, and cen-
trifugal type compressors. Very simply stated, the following actions take
place in these systems: (a) a refrigerant vapor is compressed, thereby
raising its temperature and pressure; (b) the high-pressure (warm) gas is
then condensed to a high pressure (warm) liquid and heat is dissipated from
the condenser; (c) the high-pressure liquid is pumped through a metering
device and into an evaporator where it vaporizes to a low pressure (cool)
gas extracting heat from the surroundings and producing the refrigeration
effect; and (d) the low-pressure gas is returned to the compressor for'
recycle.
The "R" designation is standard industry terminology for "refrigerant"
and, for the chlorofluorocarbons, replaces the "F" designation used
in other industries.
14
-------
TABLE III-l
CHLOROFLUOROCARBON USE AS REFRIGERANTS BY END-USE PRODUCTS
U«
Type of
Equipment
Major Appliances
Room A/0^
Dehumidifiers^
Freezers-
Re frigerators—
Other
Ice Makers2
Water Coolers2
Mobile A/ct/
Unitary Residential
Unitary Commercial A/G£/
Centrifugal Chillers^/
Reciprocating Chillers£/.
Unit Coolers!/
Food Store Refrig.die/
Mobile Refrig,£/
Beverage Refrig.£/
Packaged Terminal
Other
Total
Unit Estimates
Refrigerant
Commonly
. Used
Units in
Service
(millions)
Units
Shipped
1973
(millions)
Units
Scrapped
1973
(millions)
Average
Unit
Charge
(Ib/unit)
Original
Charge
1973
(million Ib)
22
12
12
12
29.0
5.1
22.4
68.7
5.35
0.65
2.42
6.77
4.00
0.16
1.10
4.00
2.00
0.84
1.25
0.63
12
12
12
22
22
11,12,22
11,12,22
12
12,22
12
12
22
1.3
3.8
45.0
11.4
4.0
0.05
0.2
1.7
0.2
0.5
23.8
1.0
0.22
0.39
7.53
2.15
0.62
0.005
0.02
0.19
0.008
0.07
3.57
0.15
0.03
0.16
7.00
0.40
0.17
0.001
0.003
0.12
0.006
0.03
0.65
0.03
2.00
1.00
3.80
4.5
36.0
2,500.0
350.0
14.0
675.0
14.0
1.2
2.5
10.7
0.5
3.0
4.3
0.4
0.4
28.6
19.8
22.3
12.5
7.0
2.7
5.4
1.0
4.3
0.4
9.0
132.3
c/
d/
e/
Factory-charged sealed units.
Mobil (automobile) air conditioning unit.
Field-charged sealed compressor units.
Field-charged open motor units.
Unit data are on a per store basis.
Source: Arthur D. Little Report^/ and a major manufacturer.
-------
TABLE III-2
ESTIMATED CHLOROFLUOROCARBON USE AS REFRIGERANTS
1973 End Use
Chlorofluorocarbon
R-ll
R-12
R-22
Other
Total
million Ib
18
168
90
24
300
%
6
56
30
8
100
Note: If R-22 is eliminated from the above table and the 8% in the "other"
category is assumed to be all R-13, R-113 and R-114, then R-ll and
R-12 are calculated to account for a minimum of 88.6% of the five
chlorofluorocarbons under study. Ten million pounds of R-114 was
used as a refrigerant in 1972,—' which, on the basis of the above
table, would account for an additional 3.3%.
Source: Arthur D. Little Report.—'
16
-------
a. Reciprocating compressors; Reciprocating compressors
(those in which a piston travels back and forth in a cylinder) are pres-
ently used in applications up to approximately 100 tons refrigeration ca-
pacity. This upper limit is due to considerations of reliability, life ex-
pectancy, and energy consumption. Generally, these systems use R-22 for
comfort air conditioning applications (evaporator temperatures of 40 to
50°F (4 to 10°C))_1/ and R-12 or R-502 (azeotrope of R-22 and R-115) for
lower temperature applications. End use products utilizing reciprocating
compressors and essentially 100% R-22 include room air conditioners, unit-
ary residential air conditioners, unitary commerical air conditioners, and
packaged terminal air conditioners. Other reciprocating comp.ressors utiliz-
ing R-22 are used in chillers and food store refrigeration. Reciprocating
compressors using essentially 100% R-12 or R-500 include dehumidifiers,
home freezers and refrigerators, ice-makers, water coolers, unit coolers,
mobile refrigeration, and mobile air conditionersiiit' (mobile denotes auto-
mobile and light trucks up to 10,000 Ib GVW).J/ Other reciprocating compres-
sors utilizing R-12 are used in chillers and food storage refrigeration.
For reciprocating compressor applications^ this study is primarily concerned
with the latter group of end-use products which use the R-12 refrigerant.
b. Rotary compressors; Rotary compressors (those in which
an eccentric rotates within a cylinder) are used for fractional tonnage ap-
plications in some appliances. Most of these units operate on R-12 refrig-
erant. R-114 can be used in rotary compressors for small appliances. Heli-
cal rotary compressors can be obtained in 120 to 350 ton capacity.-!'
c. Centrifugal compressors; Centrifugal compressors (those
in which an impeller^ rotating within a housing, draws in vapor and dis-
charges it at high velocity by centrifugal force) are usually broken into
three capacity ranges: lower capacity, 100 to 1,500 tons, commonly use
R-ll; intermediate capacity, 1,000 to 6,000 tons, commonly use R-12 or some-
times R-114; higher capacity, 5,000 to 15,000 tons, commonly use R-22j±'
R-113 can be used in systems above 50 ton capacity^' The practical upper
limit of capacity of centrifugal compressors is determined only by physical
size_I' Centrifugal units are used to cool a secondary refrigerant, such
as water or brine, which is then circulated to cool large enclosures. In
practice, R-22 is rarely used in centrifugal chillers because of the high
operating pressures. The chiller market, excluding absorption systems, is
approximately two-thirds R-ll and one-third R-12 and R-500 (azeotrope of
R-12 and R-152a)J/
2. Physical and chemical characteristics of mechanical vapor com-
pression working fluids; "The choice of the refrigerant for a particular
application often depends on properties not directly related to its ability
to remove heat. . . .As a rule, the selection of a refrigerant is a compro-
mise between conflicting desirable properties."-!' There are but a very few
truly technical limitations in selecting materials for use as refrigerants.
17
-------
"Anything is technically feasible if energy and cost are not considered fac-
tors."^/ "It is technically feasible to utilize nearly any organic compound
as a refrigerant.'!2' However, there are technical limitations on which mate-
rials can be used in a given temperature rangeul' (a) the boiling point of
a refrigerant indicates directly the general temperature range within which
a refrigerant can be used; (b) the freezing point of a refrigerant must be
lower than any contemplated usage; and (c) in refrigeration cycles involv-
ing condensation, a refrigerant must be chosen so that the condensation will
occur at a temperature somewhat below the critical temperature (temperature
above which a vapor cannot be condensed)..1'
In reality a great many factors influence the choice of a chemi-
cal compound for use as a refrigerant. The following considerations have
been stated and are presented not necessarily in order of importance:
Property
Chemical and Physical
Boiling point
Critical temperature
Density and viscosity
Chemical and thermal stability
in contact with materials of
construction
Latent heat of vaporization and
enthalpy
Flammability and toxicity
Vapor pressure and specific
volume
Chemical reactivity with oil
Miscibility with oil
Compatibility with wire insula-
tion, elastomers and plastics
Velocity of sound in vapor
Solubility of water in
refrigerant
Reason for Consideration
Determines temperature application range.
Determines upper limit of condenser tem-
perature application range.
Related to pumping rates.
Determines rates of decomposition.
Related to refrigeration flow rate.
Safety considerations.
Affects compressor design.
Must not react to degrade the refriger-
ant or the oil in an hermetic system.
Must carry oil to moving parts in an her-
metic system, especially from low side
back to the compressor.
Should not swell, dissolve, or extract
from these materials in an hermetic
system.
Limits pumping rates.
Some solubility is desirable so that, ice
will not restrict metering devices.
18
-------
Other Factors
Electrical properties Must not short-out electrical motor wind-
Ings in an hermetic system.
Leak detection Leaks should be readily detectable.
Cost The costs of the refrigerant, the system
components, and system maintainance
are prime considerations.
Energy requirements '>The saving of energy is as important as
the saving of dollars.".12' The energy
needed to manufacture and operate the
system is becoming increasingly impor-
tant.
Environmental acceptability Efforts are being made to reduce the use
of the materials which may pose a threat
to the ozone layer.
These considerations have been compiled from the ASHRAE Handbook
Q Q/ ———————
and from direct communications with industry., a *'
B. Alternatives to the Chlorofluorocarbons Under Study
There are two broad category alternatives to the use of the sub-
ject chlorofluorocarbons in refrigeration systems: (a) utilization of other
chemical refrigerants in the mechanical vapor-compression type systems; and
(b) utilization of refrigeration systems other than mechanical vapor compres-
sion. (A third choice, which is not herein considered, is elimination of
refrigeration and air conditioning,)
1. Chemical alternatives for the selected chlorofluorocarbons as
mechanical vapor compression working fluids; The ASHRAE Guide and Data Book
lists 78 materials which have been used or considered as working fluids in
refrigeration systems. Each of these 78 materials has been given considera-
tion as an alternative to fluprocarbons R-ll, R-12, R-13, R-113, and R-114.
It is beyond the scope of this program to provide a detailed technical analy-
sis of all the refrigerants. MRI has gone through the list of refrigerants
and recorded the readily apparent negative aspects of the given materials
(Table III-3).
Materials from the ASHRAE which have not thus far been disquali-
fied as viable alternatives as refrigerants, and 12 other compounds having
boiling points in the general range of the materials under study, are pre-
sented in Table III-4.
19
-------
TABLE III-3
ASHRAE REFRIGERANTS
(ASHRAE Standard 34-67, ANSI B79.1)
to
o
Refrigerant No.
Halocarbons
10
11
12
13
13B1
14
20
21
22
23
30
31
32
40
41
Chemical Name
Carbontetrachloride
Trichlorofluoromethane
Dichlorodlfluoromethane
Chlorotrifluoromethane
Bromot rif luorotne thane
Carbontetrafluoride
Chloroform
Dichlorofluoromethane
Chlorodifluoromethane
Trlfluoromethane
Methylene Chloride
Chlorofluoromethane
Methylene Fluoride
Methyl Chloride
Methyl Fluoride
Chemical Formula
CC13F
CC1F
CC1F3
CBrF-
CF4
CHC1
CHC12F
CHC1F-
CHF,
CH2C1F
CH
CH
Cl
CH3F
Negative Considerations
Stable;—' possibly reactive with ozone; toxic
Stable;£.' possibly reactive with ozone
Stable;—' possibly reactive with ozone
Stable;—' possibly reactive with ozone
Stable ;£L/ possibly reactive with ozone
Low critical temperature; can use in only very
low temperatures
Some toxicity; dissolves some insulation
Swells or disintegrates some plastics; high
compressor temperature; some toxicity
High speed in centrifugals, permeates some
elastomers; oil miscibility problems at low
temperatures with most oils; high compressor
temperature
Low critical temperature
Swells or disintegrates some plastics; large
compressor displacement; high compressor
temperature; some toxicity
Flammable; low critical temperature (can be
used as azeotrope)
Flammable; low critical temperature
Flammable, cannot be used with aluminum or
magnesium; high compressor temperature; toxic
Flammable
-------
Refrigerant No.
50*
111
112
112a
113
113a
114
114a
114B2
115
116
120
123
124
125
133a
140a
142b
143a
150a
Chemical Name
Methane
Pentachlorofluoroethane
TetrachlorodIfluoroethane
Tetrachlorodifluoroethane.
Trichlorotrifluoroethane
Trichlorotrifluoroethane
Dichlorotetrafluoroethane
Dichlorotetrafluoroethane
Dibrotnotetraf luoroethane
Chloropentafluoroethane
Hexa fluoroe thane
Pentachloroethane
Dichlorotrifluoroethane
Chlorotetrafluoroethane
Chlorotetraf luoroethane.
Pen ta fluoroe thane
Chlorotrifluoroethane
Trichloroethane
Chlorodifluoroethane
Trifluoroethane
Dichloroethane
TABLE III-3 (Continued)
Chemical Formula
Negative Considerations
CC1F2CC1F2
CC1
CH,
CC13CC12F
CC12FCC1_F
CC13CC1F
CC12FCC1F
CC13CF3
1F2 -
V
CBrF2CBrF2
CC1F2CF3
CHCUCCl
CHC12CF3
CHC1FCF
CHF2CC1F2
CHF2CF3
CH2C1CF3
CH3CC13
CH_CC1F0
CH3CHC12
Flammable
Stable; possibly reactive with ozone
Stable; possibly reactive with ozone; may produce
cardiac effects.!?'
Stable; possible reactive wAh ozone
Stable; possibly reactive with ozone
Stable; possibly-reactive with ozone
Stable; possibly reactive with ozone
Stable; possibly reactive with ozone
Stable; possibly reactive with ozone
Stable; possibly reactive with ozone
Low critical temperature
Toxicity not determined; high boiling point
Toxicity not determined
Toxicity not determined
Toxicity not determined
Toxicity not determined
Toxicity not determined
Limited flammability
Limited flammability
Limited flammability
Flammable
-------
Refrigerant No.
152a
160
170*
218
290*
Chemical Name
Dlfluoroethane
Ethyl Chloride
Ethane
Oc ta fluoropropane
Propane
Cyclic Organic Compounds
to
C316
C317
C318
Azeotropes
500
501
502
503
504
Hydrocarbons
50
170
290
600
600a
1150
1270
Dichlorohexafluorocyclo-
butane
Chloroheptafluorocyclo-
butane
Octafluorocyclobutane
Refrigerants
12/152a (73.8/2612)
22/12 (75/25)
22/115 (48.9/51.2)
23/13 (40.1/59.9)
32/115 (48.2/51.8)
Methane
Ethane
Propane
Butane
Isobutane (2-methyl-
propane)
Ethylene
Propylene
TABLE III-3 (Continued)
Chemical Formula
CH3CH3
CF3CF2CF
CH3CH CH3
C4C12F6
C4C1F7
C4F8
CC12F2/CH3CHF2
CHC1F2/CC12F2
CHC1F2/CC1F2CF3
CHF3/CC1F3
CH2F2/CC1F CF
CH3CH2CH2CH
CH(CH3)3
CH3CH=CH2
Negative Considerations
Flammable
Flammable
Flammable
No process
Flammable
information available
Poor oil solubility
R-12 portion stable; possibly reactive with ozone
R-12 portion stable; possibly reactive with ozone
R-115 portion stable; possibly reactive with ozone
R-13 portion stable; possibly reactive with ozone
R-115 portion stable; possibly reactive with ozone
Flammable
Flammable
Flammable
Flammable
Flammable
Flammable
Flammable
-------
Refrigerant No.
Chemical Name
TABLE III-3 (Continued)
Chemical Formula
Negative Considerations
Oxygen Compounds—
c/
ISJ
u>
610
611
Ethyl Ether
Methyl Formate
Nitrogen Compounds—'
630
631
Inorganic Compounds
702
704
717
718
720
728
729
732
740
744
744a
764
Unsaturated Organic
1112a
1113
1114
Methylamine
Ethylamine
Hydrogen
Helium
Ammonia
Water
Neon
Nitrogen
Air
Oxygen
Argon
Carbon Dioxide
Nitrous Oxide
Sulfur Dioxide
Compounds—
Dichlorodifluoi
Chlorotrif luort
Tetrafluoroethj
HCOOCH3
CH3NH
CrtH_NH«
He
Ne
0.21 02; 0.78
0.01 A
°2
A
co2
N20
S00
CC12=CF
CC1F=CF
Flammable
Flammable; toxic
Flammable; high compressor displacement
Flammable; high compressor displacement
Flammable; low critical temperature
Low critical temperature
Flammable; toxic; cannot use copper In system
High boiling point
Low critical temperature
Low critical temperature
Low critical temperature
Supports combustion
High operating pressures
High operating pressures
High operating pressures
Toxic; high compressor temperature; dissolves
insulation
May be toxic
May be toxic, flammable
Polymerizes, flammable
-------
Refrigerant No. Chemical Name
1120 Trlchloroethylene
1130
1132a
1140
1141
1150
1270
Dtchloroethylene
Vinylidene Fluoride
Vinyl Chloride
Vinyl Fluoride
Ethylene
Propylene
TABLE III-3 (Concluded)
Chemical Formula
CHC1=CC1.
CHC1=CHC1
CH2=CF2
. CH,
CH,
=CHC1
=CHF
CH =CH2
CH3CH=CH2
Negative Considerations
Toxic; high compressor displacement; dissolves
some insulations
Flammable; high compressor displacement; toxic
May be toxic
Flammable; toxic.; polymerizes
Flammable; toxic; polymerizes
Flammable; toxic; polymerizes
Flammable; toxic; polymerizes
aj "Stable" indicates stable in the lower atmosphere, other materials such as R-14 are likewise "stable" but contain
no chlorine.
_b/ Cardiac effects may also apply to several other chlorofluorocarbons in high concentration under conditions of
stress. This effect is not restricted to F-112.
_c/ These compounds are generally far more reactive than the materials under study, and in general, would not be ex-
pected to possess the required chemical stability.
Note: * These materials appear in the halocarbon section but they are not halocarbons.
-------
TABLE III-4
MATERIALS FOR CONSIDERATION AS ALTERNATIVE REFRIGERANTS^
a/
Refrigerant
No.
Chemical Name
Chemical
Formula
Boiling Point
Melting Point
22 Chlorodifluoromethane
123 Dichlorotrifluoroethane
124 Chlorotetrafluoroethane
124a Chlorotetrafluoroethane
125 Pentafluoroethane
133a Chlorotrifluoroethane
142b Chlorodi£luoroethane
218 Octafluoropropane
C316 Dichlorohexafluorocyclo-
butane
C317 Chloroheptafluorocyclo-
butane
C318 Octafluorocyclobutane
1112a Dichlorodifluoroethylene
1113 Chlorotrifluoroethylene
1132a Vlnylidine Fluoride
Perfluoropropylene Oxide
Perfluorobutane
Perfluorotrimethylamine
bis(Trifluoromethyl)sulfone
Pentafluoroethyl Sulfur
Pentafluoride
bis(Trifluoromethy1)sulfur
Tetrafluoride
Perfluoroacetone
Trifluoromethy1 Sulfur
Pentafluoride
Perfluoropropane
Perfluorodimethylether
Perfluorohexane
bis(Pentafluorosulfur)-
di fluoromethane
Tetrafluoroethane
CHC1F2
CHC12CF3
CHC1FCF3
CHF2CC1F2
CHF2CF3
CH2C1CF3
CH3CC1F2
CF3CF2CF3
-41.4
79
-
-
-
-
14.5
-38
-40.8
26
-
-
-
-
-9.7
-39
-256
-160
C4C1F7
CC12=CF2
CC1F=CF2
CH2=CF2
CF3CFCF20
(CF3)3N
(CF3)2S02
(CF3)2SF4
(CF3)2CO
CF3SF5
C3F8
(CF3)20
n-C6F14
CF2(SF5)2
CHF2CHF2
21.5
69
-17.5
-4
-6.0
-42.5
102
28
12.4
60
66
39
-2.2
10.9
15.6
19
-175
-198
-
-
-
20.5
•27.5 -188
•20 -124
-41.4
-283
128
-122
-87
-34.6
-74.2
136
75.3
-37
-59
58
59.6
-297
-
< -102 .
-94
-183
-
< -75
-70
-15
-26
_a/ Those materials with an assigned refrigerant number were taken from the ASHRAE list of
refrigerants; the other materials are, in general, experimental materials.
25
-------
The 12 compounds are shown for illustrative purposes and do not represent
an exhaustive search of the literature. In all probability, some of these
materials will be eliminated from consideration for given applications be-
cause of inappropriate properties such as boiling point, critical tempera-
.ture, vapor pressure, specific volume, and toxicity.
Chlorofluorocarbon producers are, at this point, interested in
alternatives for chlorofluorocarbons which could be produced in existing
facilities now producing R-ll and R-12. R-21 and/or R-22 could be produced
in facilities now producing R-ll and R-12 at a conversion cost of approxi-
mately 15% of the current replacement cost of those facilities. *•!' In addi-
tion, a large capital investment for new chloroform capacity will be re-
quired. Although there exists some negative considerations to the use of
R-22 in some applications, this material is by far the leading candidate
as an alternative to R-ll and R-12. Many of the end-product manufacturers
are hesitant to design for the use of R-22, however, for fear that R-22
will also come under control. Most of the other materials listed in Table
III-3 could not be produced in existing facilities without very significant
plant changeover. The technology of high volume production is said to be
unavailable for 03 and 04 series materials .-2' The cost of chlorof luorocarbons
is directly related to the number of F atoms in the molecule, and the produc-
tion of perfluorinated materials requires very high energy consumption.
It is also possible that blends or mixtures of some of the refrig-
erants listed in Table III-3 may be formulated that would have better end
properties than the individual components. These would not be azeotropic
mixtures but physical mixtures. Most refrigeration systems have sufficient
mixing action within the system that any leakage on the vapor side would
probably not result in any significant fractionation of the refrigerant
mixture.—
2. Cycles other than mechanical vapor compression; Five systems
other than mechanical vapor compression have been considered in this study.
They are absorption, thermoelectric, steam jet, liquid nitrogen, and the air
cycle. Detailed technical discussions of these systems are provided in the
ASHRAE Guide and Data Book. A later generation air cycle system, Rovac, is
now in the prototype stage of development. A technical paper describing this
system and the claims of the inventor is provided in Appendix B.
a. Absorption cycle;
"As in the vapor compression cycle, the refrigerant is pres-
surized, liquified, cooled and then allowed to vaporize in the
evaporator where it absorbs heat to accomplish refrigeration. How-
ever, instead of using a compressor to pressurize the refrigerant,
it is cooled to permit liquefaction at a low pressure by absorption
in a miscible absorbdnt fluid and pumped as a liquid to the genera-
tor. Heat is applied at the generator to drive off the refrigerant
as a vapor and to pressure it. The refrigerant is then cooled in the
condenser to form a pressurized liquid for use in the evaporator
again."!/
26
-------
Two absorption systems using two different combinations of
working fluids have been used, and are in use today. One is a lithium
bromide-water cycle; the other is the ammonia-water cycle.-i=' The lithium
bromide-water system is used in special applications for comfort air con-
ditioning, and cannot be used below the freezing point of water. The
ammonia-water system can:be used in both air conditioning and refrigeration
applications. Some important factors to be considered for these systems are
as follows:
1. The technology for manufacturing absorption equipment
is well established, but is used by only a limited number of companies.-i-='
2. Absorption systems are heavier and more expensive than
mechanical vapor compression systems. It is commonly quoted that these sys-
tems are less efficient than compression cycle systems. This misconception
arises because the COP values of the compression cycle is based on the heat
equivalent of the electrical energy fed to the motor and ignores the thermal
losses at the power plant producing the electrical energy. When absorption
and compression cycles are compared on the same basis, i.e., the Btu of cool-
ing per Btu of fuel energy used, the COP values are very nearly the same. —'
Absorption chillers are used only where waste heat is availableJr (The heat
probably does not have to be "waste," but the economics of the energy source
is a prime consideration.)
3. While the absorption equipment is larger than the equiva-
lent sized compression equipment, its selling point has been its lower cost
of operation. As the costs of both gas and electrical energy rise, it may
be that the lower operating cost of absorption systems will overcome their
higher initial costs.—'
4. Ammonia is toxic to humans, with a threshold limit value
(TLV) of 25 parts per million (ppm), as compared with the high TLV of 1,000
ppm for fluorocarbons. It also forms.an explosive mixture in air at 16% con-
centration. Ammonia is not suitable for use in units within buildings where f
human exposure is possible^'
5. Ammonia is very corrosive to brass and bronze, and has
a higher tendency to leak than do common chlorofluorocarbons.
6. Absorption units are less efficient in energy consumption
than mechanical systems, and (for appliances) could be used only where nat-
ural gas can be used as a heat source»i2'
7. An estimated 300,000 absorption refrigerators are sold
per year in the U.S. Most of these go into the recreational vehicle or motor
home market. An estimated 25,000 absorption units are sold per year for cen-
tral air conditioning and chillers accounting for approximately 1.25% of
the market. Of the 25,000 units, approximately 80% are for residential and
small contnerical applications. \3'
27
-------
8. In the chiller market, approximately 2% of the total is
absorption units. All of these units are lithium bromide systems, and the
cost is approximately twice that of chlorofluorocarbon systems J-'
9. Operating efficiencies of absorption refrigerators and
freezers are considerably below those of vapor-compression units.-!' The use
of "waste" heat may be an opportunity to reduce the energy input to the sys-
tem and reduce overall operating costs plus energy consumption.
10. Ammonia or lithium bromide can be used where steam is
available, but absorption systems are inefficient when fired with an oil
burner.-!^' v '
11. System should be perfectly level.
b. Thermoelectric system; When electrical energy is forced
to flow through a thermocouple junction (two dissimilar metal wires con-
nected at one end), heat is absorbed or rejected at the junction. The impor-
tant factors to be considered for thermoelectric cooling are as follows:
1. The heating and cooling functions of a thermoelectric
system can be interchanged by reversing the polarity of the direct current
applied to it.—'
2. Since there are no moving parts, there is nothing to wear
out or to generate
3. It is practical to make a system of very low refriger-
ating capacity or a system of large capacity if many couples are used.—
4. The system will operate in any position.!'
5. The system is much less efficient than a compression sys-
tem. li 15/ This low efficiency may be a major factor in considering these
systems as alternatives.
6. Most units are in 30 to 200 Btu/hr range-L^/ (12,000 Btu/
hr °- 1 ton).
7. System is reliable, safe, quiet, and provides weight and
size advantages in given applications.!^' However, a major appliance manu-
facturer disagrees with the reliability factor and states that these systems
lack reliability.!0./
8. While the units are economical to produce in low cooling
capacity ranges (~ $10/40 Btu unit), the scale-up cost is linear
9. Units are presently used as follows:—'
28
-------
a. To cool sensitive electronic systems such as sen-
sors in military missiles.
b. In commerical airliners to cool drinking water.
c. In very small refrigerators for recreational vehic-
les and boats.
d. To cool epoxy for dental applications.
10. The U.S. government has sponsored development of units
up to 9-ton capacity for use on submarines.
11. A Swedish firm is contemplating production of a 1,000
Btu/hr room air-conditioner using thermoelectric units (the smallest U.S.
produced room air-conditioner is ~ 4,000 Btu/hr) JL5'
12. There are some practical problems in thermoelectric sys-
tems with regard to the dissipation of heat and the practical application
of the cooling effect.
13. Larger units could be practical where very large amounts
of DC power is readily available.
c. Steam jet: The steam-jet refrigeration cycle is quite
similar to more conventional refrigeration cycles with an evaporator, a com-
pression device, a condenser, and a refrigerant as the basic system compon-
ents. Instead of a mechanical compression device, the system characteristi-
cally employs a steam ejector or booster to compress the refrigerant to the
condenser pressure level.
Water is used as the refrigerant, and the cooling effect is
produced in the steam-jet refrigeration cycle by the continuous valorization
of a part of the water in the evaporator at a low absolute pressure level.—'
Considerations are as follows:
1. The system is now used for applications such as freeze-
drying of foods, chemical crystallizers, etc.-='
2. The largest users are paper pulp operations in which C102
water is chilled to ~ 40°F for bleaching. Some units are used for hospital
operations J£'
3. The units are initially less expensive, less troublesome,
and essentially maintenance-free. An operating engineer does not have to be
on
29
-------
167
4. The system is simple, rugged, vibration free, highly re-
liable and low in cost J/
5. While a given manufacturer sells 5 to 20 units/year in
the 100 to 3,000 ton range, the system is technically feasible down to ~ 1-
ton capacity.-!£/
6. Large quantities of water are needed for the larger units
and utility costs are likely higher.-^ (Obviously, high-pressure steam must
be available.) The requirements of a high pressure steam source and large
quantities of water should seriously limit its general consideration as an
alternative_IP./
7. The system tends to be noisier than 'conventional systems,
8. The lower the application temperature, the less the sys-
tem is competitive with chlorofluorocarbon systemsJJS'
d. Liquid nitrogen; Liquid nitrogen (LN2> is used as a re-
frigerant for refrigerated trucks and trailers where a hard freeze is required
(and for cryogenic applications)»!'
Other important factors are as follows:
1. The major disadvantages of LN2 are constant replenishment
and the special care required in handling and distribution because of its
very low temperature.-='
2. LN2 refrigeration is technically feasible to adapt to any
required temperature range, but it is .not economical. No one uses LN2 at use
temperatures greater than -40°FJLZ'
3. Systems are not produced in large numbers. At present a
closed system the size of a home refrigerator or .air conditioner and operat-
ing at '~ 20°K (-253°C or -423°F) would cost ~ $15,000 JJ/
4. LN2 production requires a great amount of energy—the
efficiency ratio (energy output to energy input may be 1:100).-Li'
e. Air-cycle; Refrigeration in a simple open air-cycle sys-
tem is obtained by three basic steps: (1) compression; (2) heat transfer;
and (3) expansion accompained by work extraction. Air-cycle refrigeration
systems are more commonly used in the air-conditioning of aircraft than in
surface and stationary applications, because the lightweight compact equip-
ment typical of air-cycle .systems usually offsets its inherently low effici-
ency. The development in recent years of lightweight vapor-cycle refrigera-
tion, with high speed compressors, has resulted in a decrease in the use of
30
-------
air-cycle equipment in aircraft systems. Air-cycle air conditioning has not
been found to be economical in residential and commercial buildings because
of the high power required. However, it is used in other specialized appli-
cations where efficiency and operating costs are not the primary factor.
For example, air-cycle cooling units are used with portable gas turbine power
plants to provide environmental control for remotely located, temporary, mil-
itary bases. Air-cycle refrigeration may be designed and operated either as
an open or a closed system. Open air-cycle systems have been widely used both
in military and commerical airplanes since the end of World War IIJL'
The major problem for the standard air-cycle systems is the
inherent low efficiences.-!'
f. Rovac air-cycle; The Rovac system is now in the proto-
type stage of development, and uses air as the refrigerant. The system is
described by its developer* as a "multifluid mixed phase cycle which is a
hybrid of the reverse Brayton cycle (air-cycle) and the reverse Rankine
cycle (chlorofIuorocarbon)."-i2'
A large portion of the development work on the system was
accomplished on contracts to the U.S. Air Force for production of prototype
air conditioning units for aircraft.JiH/ The system is said to be technically
feasible for a wide range of end-use applications: automobile air condition-
ers; transport refrigeration; refrigerators; freezers; supermarket applica-
tions; residential air conditioners and heat pumps;JL2' and possibly commer-
cial chillers.-=£' A prototype window air conditioner has been produced for
the U.S. Army. Rovac's in-house efforts have been concentrated primarily
on the development of an air conditioning system for automobiles. Prototype
units have been or are being evaluated by Chrysler Corporation?^' and by
01 /
General Motors Corporation,-=i' and information on the system was solicited
from these companies. Rovac also has working arrangements with a number of
foreign automotive manufacturers.^/
Other applications for which Rovac is now designing proto-
types include residential heat pumps and units for transport refrigeration.
While major manufacturers of air conditioning and refrigeration systems are
understandably reluctant to speculate on a system which is still under de-
velopment, most of these companies have likely given the system preliminary
consideration.
A technical paper authored by the inventor and patent holder is attached
to this report as Appendix B. The paper gives a technical description
of the system and records the claims of the developer. The data in the
paper are said to be approximately 2-years old, as of this writing,
and many technological improvements are said to have been made in the
past 2 years. The system is further described in the November 1975
issue of Air Conditioning and Refrigeration Business.
31
-------
Technical data now available are primarily from one source,
Rovac; and those data are for the automobile air conditioning prototypes.
Some back-up information has been obtained from the U.S. Air Force. Rovac
claims that the automobile system could be marketed in its present stage of
development, but admits that there are areas in which improvement is needed:
1. While the system is now competitive from an efficiency
standpoint, computer programs indicate that substantially better efficien-
cies can be obtained.
2. The vane material must have extremely good dimensional
stability (thermal coefficient of expansion < 6 Hin/in/°F) over the operat-
ing temperature range, and, while the carbon vanes are now being used meet
this criterion, there is a problem with carbon "overfilms" (carbon dust com-
bines with grease to form a sludge).
>
3. Long-term reliability must be performed (the^longest
continuous operation of 300 hr provided satisfactory performance).—
Rovac claims for the automotive system can be summarized as
follows:
1. The cool-down rate is extremely fast as compared to con-
ventional systems.
2. The weight of the total system is approximately 40 Ib
as compared to approximately 12t Ib for a conventional system.
3. The cost should be approximately two-thirds that of a
conventional system.
4. Energy savings, in terms of gasoline usage, should be
approximately one-third.
5. The temperatures achieved in operation are said to kill
bacteria and some spores, and the system removes pollen and smoke from the
air.
6. The unit is capable of operating on either recirculating
or fresh-air modes.
7. The unit operates satisfactorily at all engine speeds,
although the cooling capacity is a function of the engine speed.
8. The system is satisfactorily quiet, although improvements
are still being made.
9. The system can be designed for cooling to -50°F.
32
-------
197
The project officer for the Air Force contracts— substan-
tiates Rovac's claims, and adds the following comments:
1. Rovac is probably being conservative in claiming energy
savings of one-third.
2. The system is easily maintained and repaired, and
tially no special training is needed for a good mechanic.
essen-
3. No great retooling cost is anticipated. The system has
fewer parts than conventional systems, and standard metal-working machines
are used for production.
4. The old air-cycle turbine used on aircraft operated at
60,000 to 80,000 rpm, and the Rovac prototype operates at less than 4,000
rpm.
5. High-compression ratio machines should be able to achieve
a temperature of -100°F.
6. The Rovac system will be flight tested within the next
3 years. The Air Force also expects to use the Rovac system in ground-level
air conditioning applications.
7. The cost of the aircraft units is much higher than that
of an automobile system because of 'higher capacities and closer tolerances.
Present aircraft systems cost $9,000 to $10,000, and the Rovac aircraft sys-
tem is projected to cost $3,500.
MRI personnel have not seen any Rovac system in operation or
performed any testing on the system. Dr. Thomas Edwards of the Rovac Corpora-
tion visted MRI, and made a slide and film, presentation to a group of scien-
tists and engineers. The overall consensus of this group was that the concept
appears viable from mechanical and thermodynamic standpoints, based on avail-
able information. Fabrication of component parts appears to be within the
present state of the art for mass-produced items; however, final materials-
selections may require development of special fabrication techniques. Concern
was expressed in the following areas:
1. Machine size and compression ratios do not appear to have
been optimized.
2. Inasmuch as the efficiency of the system is dependent
partly on the humidity of the intake air, questions still need to be answ-
ered on operation in low-humidity conditions.
3. Filtering of high-solids-loading intake air may present
a problem in terms of restricted air flow and closing* or dnmaj',ii»j; intr.rnal.
components.
T5
-------
4. There is some concern with the choice of the vane mate-
rial and the inherent losses because of the vanes cycling through the rela-
tively extreme temperature differential in the circulator.
5. Longer-term reliability tests should be made using in-
service conditions.
6. A situation might arise wherein engine combustion prod-
ucts could be aspirated into the passenger compartment through a leak in
the inlet air duct.
The question was asked of Rovac: "Why are you not now in
production?" The following statements have been excerpted and pharaphrased
from Rovac's lengthy written response:
". . . Rovac is a very small company developing a
rather large technology. Due to the size of the techno-
logical development relative to the size of the company
resources, the level of development needed for the gen-
eration of adequate warranty data has not been reached.
Rovac is increasing the performance of the machine with
each passing month and is still in a learning process in
methods of manufacture. Another consideration is the mat-
ter of design and procurement of production equipment.
"In the event Rovac was fully satisfied with the
state of development of the system, it would take 12 to
18 months to have the first machines off a production
line. In the event that a large company joins Rovac in
the final development of the technology, a much higher
degree of productivity could take place on a shortened
time frame. However, there would still be the minimum
1-year situation. Further, the pragmatics of U.S. auto-
makers are such that the technology would not be inte-
grated on a wide scale into that industry for approxi-
mately 2-1/2 to 3 years.
"In summary, Rovac is not in production because it
wants to fully prove the product and gain all warranty
data for its financial planning and to ensure that it
launches into the marketplace with a fully acceptable
product. . ."
C. Applicability of Alternatives in Specific Use Areas
In addition to the previously discussed attempts to determine
technically feasible alternative systems, manufacturers of a wide range of
34
-------
eno|-use products, as well as the respective trade associations, were con-
tacted. They were asked what they considered technically feasible chemical
or mechanical alternatives to the use of the selected chlorofluorocarbons.
Early in the program, efforts were also made to determine the overall eco-
nomic impacts and resulting net changes in U.S. employment.
The efforts to determine the economic impacts of converting to
alternatives for the selected chlorofluorocarbons in air conditioning and
refrigeration applications were not successful. The reasons for this are
severalfold:
1. In contrast to the aerosol industry, the air conditioning and
refrigeration industry has had, until the ozone question arose, no economic
driving force to convert from the chlorofluorocarbons. The overall trend
has, in fact, been away from the use of the nonchlorofluorocarbon alternatives
because of flammability, toxicity, cost, and energy-consumption considerations,
2. The most often mentioned alternative to the materials in ques-
tion is R-22; however, a refrigerant is chosen for a given end use because
of combination of previously described considerations, and, while R-22 could
undoubtedly be utilized in many applications currently using the materials
in question, substantial redesign and retooling would be required. Most manu-
facturers are concerned that R-22 will be added to the list of chlorofluoro-
carbons which may come under regulatory action, and therefore, have not fin-
alized the design work which would necessarily precede estimates on costs
of conversion.
3. The industry is now greatly concerned with energy conservation, .
and any systems conversions will need to be analyzed in terms of energy con-
sumption. These analyses are not generally available.
i
4. Individual appliance manufacturers and the Association of Home
Appliance Manufacturers a^re hesitant to reveal present costs and market
shares and there is conflicting information on projected sales figures. Ef-
forts to project accurate -sales figures are hampered by the present state
of the overall economy ajid by the already demonstrated effect of possible
regulatory controls on chlorofluorocarbons. (Chemical Week's projected mar-
ket for chlorofluorocarbons has dropped substantially since mid-year because
of the ozone question.)
The detailed economic data and product growth rates are so incomplete as to
render its incorporation into this report to be premature. Some detailed
information is available in the open literature^Ifj^Q/ Gross economic consid-
erations o,f converting to alternative systems are provided where applicable.
-------
TABLE III-5
APPLICABILITY OF ALTERNATIVES TO END-USE PRODUCTS
Alternate Systems
Type of Equipment
Appliances
Water Coolers
Ice Makers
Dehumldifiers
Mobile Air Conditioning
Centrifugal Chillers
Mobile Refrigeration
Coonerical Refrigeration
Pluorocarbon
Refrigerants
Now Used
8-12
R-12
R-12
R-12
R-12
R-11,R-12,R-22,R-500,R-114
R-12.R-502
R-12,R-502,R-22
Candidate
Vapor Compression
Refrigerants^/
R-22.R-S02,
R-22.R-502,
R-22.R-502,
R-22.R-502,
R-22.R-502,
R-22.R-502,
R-22.R-502,
R-22.R-502,
and others
and others
and others
and others
and others
and others
and others
and others
Absorption
Lithium
Bromide
No
Yes
No
Yes
Yes
no design
Yes
No
No
Ammonia
Yes,
Yes,
Yes,
Yes,
Yes,
Yes,
Yes,
but toxic
but toxic
but toxic
but toxic
but toxic
but toxic
but toxic
Thermoelectric
Yes,
Yes,
Yes,
Yes,
Yes,
Yes,
Yes,
but
but
but
but
but
but
but
but
Inefficient
inefficient
inefficient
inefficient
Inefficient
inefficient
inefficient
inefficient
Steam Jet
No
No
No
No
No
Limited
No
Limited
Air Cjcle j;Rovac)
Not at present
Not at present
Not at present
Not at present
Possibly in future (under dev«lop»eat)
Not at present
Possibly in future (under development)
Not at present
a/ See Hat provided In Table IU-4. p. 111-15.
-------
1. Appliances; Although R-12 was once used in room air condition-
ers, all U.S. made room air conditioners, commerical unitary coolers, and
residential central air conditioning systems (excluding absorption units)
now use R-22_i2' Accordingly, these appliances will not be further consid-
ered. Appliances now using R-12 (or R-502 in some commerical applications)
include refrigerators, freezers, dehumidifiers, ice makers, etc. Efforts in
this area were concentrated on refrigerators and freezers which, for the
category of appliances, represent the largest usage of the specified mate-
rials.
a. Absorption system; The ammonia-water absorption system
was once used in household appliances, and is a technically feasible alter-
native to the R-12 system. This system is currently used in home appliances
in some European countriesJLO/ but "most" U.S. made absorption refrigeration
systems go to the recreational vehicle or motor-home market-i^' Absorption
systems are heavier and more expensive than mechanical systems^' and are said
to be competitive only where natural gas is available as a heat source^' or
where "waste" heat may provide an opportunity to be used in conjunction with
other energy sources. Other considerations such as toxicity and flammability
*
have been previously discussed. No other usable refrigerants have been iden-
tified for the absorption system.
b. Thermoelectric; Thermoelectric cooling units are techni-
cally feasible for appliances, and very small refrigerators are now produced
for recreational vehicles and boats. The thermoelectric system is inefficient
in power consumption, and has some practical problems in the dissipation of
heat and application of the cooling effects. In all probability, it is not com-
petitive in typical appliance applications because of cost considerations.—•—
c. Steam jet; The steam-jet system is technically feasible
at minimum capacities down to 1 ton.J£' Steam is not generally available for
appliance uses, and the system is not cost-competitive with mechanical vapor-
compression systems at low temperatures.
d. Liquid nitrogen; The use of LN£ f°r appliances is tech-
nically feasible.8?I?/ The system would not be cost-competitive, and energy
consumption would be prohibitiveJLZ'
e. Air cycle; The conventional air cycle would not be adapt-
able to appliance usage because of size, noise, poor efficiency, etc. The
greatest drawback is the J.ack pf efficiencyJQ'
f. Rpvac; No prototype unit has been produced for the sub-
ject applications, but the system may be technically feasible for such appli-
cations.^' The system thermodynamics were reviewed approximately 1 year ago
by a major appliance manufacturer and, at that time, did not appear to be
efficiency-competitive..12/ The industry will continue to follow the develop-
ment of this system.
37
-------
g. Alternative vapor compression refrigerants; Appliance
manufacturers are not seriously considering the previously used refriger-
ants such as sulfur dioxide, methyl chloride, and hydrocarbons because of
toxicity and/or flammability 'considerations or carbon dioxide because of
high pressures and poor efficiencies.
R-22 is probably the leading candidate alternative refriger-
ant for appliance applicationsJL2jJLL/ R-22 was the second choice of a large
compressor manufacturer which preferred R-.502.5' General technology is avail-
able for both of these materials inasmuch as they are now being used in some
applications. However, R-502 is an azeotropic mixture of R-22 (48.8%) and
R-115 (51.2%); the R-115 component of R-502 is perhalogenated and may come
under regulatory control. R-22 is also more economical to produce than is
R-502. Besides the advantage of available technology, R-22 could be produced
in existing facilities with relatively small capital investment (~ 15% of
current replacement costs)-ii' plus the additional costs for the new chloro-
form capacity. In conversion from R-12 to R-22, the carbon tetrachloride
producers would be hard-hit, since R-12 is made from carbon tetrachloride
and R-22 is made from chloroform.
Although R-22 is a technically feasible alternative to R-12
for use in most appliances, several problem areas have been delineated in
designing for that conversion. R-22 systems operate at higher pressures and
higher compressor temperatures than do R-12 systems. The reliability and
life of a product are said to be inversely proportional to the compressor's
operating temperature^/ It has been stated, for R-22 systems, oil stability
is a problem^' and that R-22 stability is a problem.^/ One large manufac-
turer and other industry sources believes neither of these problems to be
a serious concern JiUi2' It is quite often stated that there'is a miscibility
problem with R-22 and the refrigeratipn oil at low temperatures^^' A chloro-
fluorocarbon producer—' states that an oil is now available that is completely
miscible with R-22 in all proportions at temperatures as low as -115°F.
One refrigerator manufacturer;!!' has previously produced proto-
type refrigerators operating on R-22. Data on field testing and accelerated
load testing were not available. There may be a problem with compressor valve
mechanisms operating at the high pressures of an R-22 compressor (the compres-
sion ratio in a refrigerator may be approximately 10:1 as compared to approxi-
mately 5:1 for a room air conditioner).A!'
Large costs of retooling would result from the conversion from
R-12 to R-22 urn' t-c-8,10,31/ Qne compressor manufacturer^' estimated the re-
tooling costs for a 300 compressor per hour line to be $10 to $12 million (it
presently operates a total of five lines). Another appliance manufacturer es-
timated its retooling costs to be approximately $50 million_i9-'
Prototype R-22 units would have to undergo field testing, en-
ergy consumption testing, and standard Underwriters Laboratory tests before
they could be marketed.
38
-------
2. Industrial air conditioners and chillers; Very little infor-
mation has been obtained on alternatives for this end-use category. A gen-
eral discussion of the problems involved appear in Reference 12.
a. Absorption; Absorption systems are technically feasible
for these applications within the limitations addressed in Section C (1)
"Appliances." Absorption systems are used where steam is available for en-
ergy input—' or other "waste" energy is available*^' Ammonia can be used
in reciprocating compressors where the previously discussed limitations can
be tolerated.
b. Steam jet; The steam jet is a technically feasible al-
ternative to those applications using R-ll, R-12, and R-13 where steam and
water are available in quantity, and where size limitations are not impor-
tant.
c. Alternative vapor compression refrigerants; While R-13,
R-114, and R-22 are used in some applications, the chiller market is gen-
erally broken down as two-thirds R-ll and one-third R-12 and R-500_Z' No
technology is available for mechanical vapor compression refrigerants other
than the above_Z' Therefore, R-22 would appear to be the most viable alter-
native to the materials in question. The problem of converting to R-22 is
discussed in Reference 12.
Some large chiller manufacturers would not discuss the prob-
lem on the telephone (time and cost considerations precluded personal con-
ferences) or asked not to be quoted on the small amount of information
offered.
The only other alternative mentioned as being technically
feasible was to use multiple R-22 reciprocating compressor units, currently
limited to approximately 100 tons maximum cooling capacity, to attain the
cooling capacity now obtained from centrifugal units using R-ll (100 to 1,500
tons) and R-12 (1,000 to,6,000 tons). R-22 is .now used in centrifugal units
greater than 5,000-ton capacity. (The source offering this alternative asked
not to be quoted.) The usage ranges for the various systems were provided by
Reference 4.
3. Automotive air conditioning; R-12 from mobile air condition-
ing is estimated to account for 33.3% of the total chlorofluorocarbon refrig-
erants being emitted to the atmosphere^' Accordingly, a proportionately
large amount of this program's effort was directed at identifying alternative
refrigerants or systems for this end use. Presently, approximately 7 million
automobiles (including light trucks) are being air conditioned each year.
Of this number approximately 1/2 million units are add-ons.-='
39
-------
Alternatives to the nonhermetic R-12 systems now being used which
have been considered include: R-22 systems; alternative vapor compression
sytems; other vapor compression refrigerants; absorption systems; hermeti-
cally sealed R-12 systems; and the air-cycle systems./.?/ The American Society
of Heating and Refrigeration and Air Conditioning Engineers (ASHRAE) is cur-
rently preparing a report on alternative refrigeration systems, time frames
needed for change, and affected industriesJQ' The pertinent information
will be presented at a seminar in June 1976 at the ASHRAE Annual Meeting,
Seattle, Washington^?'
a. R-22 systems; R-22 systems have been, and are now being,
strongly considered as an alternative to the R-12 system, and would likely
be the most readily adaptable alternative system.34-36/ Chrysler Corporation
marketed automobiles equipped with an R-22 system in 1952 and 1953 JI2/ This
system contained all copper fittings, and did not incorporate a clutch. The
mechanism was mechanically engaged in the spring and disengaged in the fall;
therefore, the unit ran throughout the summer months with a constant load on
the automobile engine. This type of operation would very probably be unaccept-
able in the modern market. More recently, prototype R-22 units have been built
by the after-market (add-on) industry.^' The pragmatics of the after-market
industry differ from those of factory installers, £nd R-22 add-on units could
likely be marketed in a much shorter time from the date of a decision to con-
trol R-12.
Development of a nonhermetic R-22 system for automotive use
would require a number of considerations which are delineated, not necessar-
ily in order of importance, as follows:
1. R-22 systems operate at higher pressures than do R-12
systems; leakage through flexible rubber hosing is a problem.7|20»35,36/
Water-vapor permeability 6f the hosing is also a concern*^' These problems
can be minimized by proper design.
2. The compressor would have to be redesigned to accomodate
the increased pressure and vibration..?Q'
3. The condenser would have to be enlarged, or auxiliary
blowers would have to be used.22'
4. Controls would have to be redesigned .-=2'
5. Manufacturers are concerned that R-22 may be controlled.—
6. Very small cars which could use R-12 systems may not have
the power to operate a R-22 system^H'
40
-------
7. The time-frame for conversion by auto makers is estimated
at 3l20/ to 5^4- ' years from the time of decision. Most of this time would be
needed for retooling. —
8. The increased cost of an after-market (add-on) R-22 sys-
tem would be approximately 7 to 10%.^' No cost figures are available on
factory- installed units.
9. R-22 does not have the worldwide availability of R-12
b. Alternative vapor compression systems; The mechanical
vapor compression system using ammonia has been discounted because of the
toxicity of that refrigerant. This would be especially hazardous in the
aftermath of an automobile collision.-^'
c. Other vapor compression refrigerants; The automotive
industry has no information on new refrigerants which might be forthcoming
from refrigerant producers . 20} 34/ ^jo technology is available for alterna-
tive refrigerants other than R-22.J' Application of a new refrigerant may
take 7 to 9 years *t' Hydrocarbons and other systems have been considered,
but the information is being retained for incorporation in the forthcoming
seminar at the ASHRAE Annual Meeting in Seattle, Washington, in June 1976.
d. Absorption system; Absorption systems have been consid-
ered for automotive ngp-jQ»33/ ^he ammonia system has been discounted be-
cause of its t-niHri t-y.20/ -jhe lithium-bromide system could be applied, but
approximately one-half of the heat-energy required would have to come from
a source other than that available from the automobile engine, and the con-
denser would have to be approximately double the size of the present con-
denser ,20.' Ford Motor Company is said to have presented a paper dealing with
automotive lithium bromide absorption systems to the Society of Automotive
Engineers. A copy of that paper has not been received as of this writing.
Work is also in progress on absorption systems other than ammonia and lith-
ium bromide,^' but we do not know who is performing the work.
e. Hermetically sealed vapor compression system; Hermeti-
cally %ealed units using either R-12 or R-22 are said to be technically fea-
sible for automotive air conditioning at an increased wholesale cost of ap-
proximately $150. The unit would operate on a 12-V system, and no hoses would
be necessary..?' A 5 hp hermetically sealed DC motor would be required and
would be difficult to package.^2'
f. Air-cycle system; The Rovac air-cycle system has been
or is being considered by Chrysler Corporation, General Motors Corporation,
and several foreign automobile manufacturers*!^' Rovac is now working on
the installation of a Rovac air conditioning system in a General Motors
automobile^?!' and technical data are not yet available from that source.
41
-------
Chrysler Corporation^ has performed initial tes'ting on a Rovac air con-
ditioning system installed in an automobile by Rovac. While Chyrsler be-
lieves the system may find extended use in the future, there were certain
reservations about the system in its stage of development at the time of
the testing. The following comments were offered for consideration:
1. Manufacturability at a reasonable cost on a mass produc-
tion scale has not been demonstrated.
2. The efficiency is linear with speed in a rotary system,
but not in a piston compressor. The basic problem is to obtain sufficient
volume displacement in the compressor.
3. The efficiency of a rotary compressor is inherently more
sensitive to leakage than is a piston compressor.
4. There was concern because of the close tolerances demanded
as well as possible thermal distortion of components.
i
5. The alignment of the circulator end plates is very crit-
ical.
6. The coefficient of performance (COP) of the test unit was
not substantially different from the average conventional unit.
After all problems are solved at RoVac, it is estimated that
3 years would be needed to test the system completely, and an additional
3 years would be needed to design and tool for manufacture.
4. Other end-uses; The same considerations provided for appli-
ances, industrial air conditioners and chillers, and automotive air condi-
tioners can generally be applied to other end uses such as food store ap-
plications, automatic merchandising equipment, and transport refrigeration.
One basic difference is the claim by Rovac that a transport refrigeration
system, now being designed, will take advantage of moisture-laden output
air for maintaining very high humidity conditions in applications such as
produce transportation.
42
-------
REFERENCES TO SECTION III
1. ASHRAE Guide and Data Handbook, American Society of Heating, Refrigera-
tion, and Air-Conditioning Engineers, Inc., Joseph D. Pierce, Chair-
man, New York (1972).
2. Arthur D. Little, Inc., "Preliminary Economic Impact Assessment of Pos-
sible Regulatory Action to Control Atmospheric Emissions of Selected
Halocarbons," EPA Contract No. 68-02-1349, Task 8, Publication No.
EPA-450/3-75-073, September 1975; NTIS No. PB-247-115.
3. "Fluorocarbons and the Environment," Report of Federal Task Force on
Inadvertant Modification of the Stratosphere (IMOS), Council on En-
vironmental Quality, June 1975; GPO No. 038-000-00226-1.
4. Personal communication with H. T. Gilkey, Air Conditioning and Refrig-
eration Institute
-------
15. Personal communication with M. Levine, Melcor Materials, Electronics
Products, Trenton, New Jersey.
16. Personal communication with J. Knoble, Croll-Reynolds Company,
Westfield, New Jersey.
17. Personal communication with J. Fortier, Cryogenic Technology, Inc.,
Waltham, Massachusetts.
18. Personal communication with T. Edwards, Rovac Corporation, Maitland,
Florida.
19. Personal communication with L. Midolo, Thermal Control Section, Flight
Dynamics Laboratory, WPAFB, Dayton, Ohio.
20. Personal communication with J. Holtslag, Chrysler Corporation, Detroit,
Michigan.
21. Personal communication with J. A. Dobb, General Motors Technical Cen-
ter, Warren, Michigan.
22. Appliance Manufacturer, January 1975.
23. 1972 Census of Manufacturers, Industry Series, U.S. DOC.
24. Current Industrial Reports, U.S. DOC, Bureau of the Census.
25. Merchandizing Week. May 12, 1975.
26. American Machinist, January 15, 1975.
27. Appliance Manufacturer, February 1975.
28. Appliance, November 1974.
29. Plastic World. April 21, 1975.
30. Personal communication with engineer at Amana Refrigeration, Amana,
Iowa •
31. Personal communication with J. McClean, Westinghouse Corporation,
Columbus, Ohio.
32. Personal communication with R. E. Shamel, et al., Arthur D. Little,
Inc., Cambridge, Massachusetts.
33. Personal communication with Mr. Ciricillo, Ranko Corporation, Columbus,
Ohio.
44
-------
34. Personal communication with G. F. Stofflet Product Safety Department,
General Motors Corporation, Warren, Michigan.
35. Personal communication with D. Wilson, Transical Division, Carrier Cor-
poration, Columbus, Ohio.
36. Personal communication with R. L. Maier, et al., Southwest Factories,
Oklahoma City, Oklahoma.
45
-------
IV. AEROSOL INDUSTRY
Aerosol products were first introduced on the consumer market in
1947, following the successful development of reliable pressurizing and de-
livery systems during World War II. The war-time aerosol insecticide "bomb"
used the volatile chlorofluorocarbon liquid, Freon® 12, as the propellant-i'
This development, supported by the adoption of additional propellants and
improved valve and atomizer designs, spurred the growth of what is today a
$3 billion industry^' Hundreds of personal, household, and commercial prod-
ucts are now available in aerosol form, and although many propellant systems
are employed, the chloroflurocarbons are the most widely used and are cur-
rently the cause of much environmental alarm.
The results of our study of possible chemical and mechanical al-
ternatives to the use of those chlorofluorocarbon aerosol propellants of
concern are presented in the following order: a brief description of the
nature of the aerosol industry today; an introduction to the kinds of aero-
sol technologies and to the specific propellant chemicals that are in use;
surveys of the requirements that a successful alternative must meet, and of
the candidate propellants or mechanical devices; a discussion of potential
alternatives for specific'aerosol uses; and finally, an analysis of produc-
tion data for potential alternatives for the subject chlorofluorocarbons.
A. Development of the Aerosol Industry» Aerosol Systems jmd Propellants
This section presents a brief history of the industry's develop-
ment, discusses the basic aerosol systems, and identifies the commonly used
propellants.
1. Development of the aerosol industry; Early research on aero-
sol technology produced patents as early as 1862. Notable patents were is-
sued on a perfume atomizer pressurized with carbon dioxide in 1903, on a
dimethyl ether propellant system in 1931, on fluorinated hydrocarbons pro-
pellants in 1933, and on propellant blends of hydrocarbons, methyl chloride,
and dimethyl ether in 1938 .-i'
The first commercial fluorocarbon,* Freon® 12, was developed by
General Motors in 1928 and commercialized by E. I. du Pont de Nemours and
Company in 193d2' as a refrigerant fluid. Freori® 12 (CF2C12) was also a model
propellant, possessing both ideal vapor pressure and boiling point. During
World War II, the first portable aerosol dispenser was developed and F-12
was used as the propellant^'
The term fluorocarbon has been widely used to'refer to chlorofluorocar-
bons.
46
-------
Aerosol products were not available to the consumer until 1947.
Production in that year was 4.3 million units, almost exclusively insecti-
cides. The development of additional fluorocarbon propellants (e.g., F-ll
and F-114) made possible, through blends, the formulation of propellant
mixtures having a wide range of vapor pressures and boiling points. These
propellant blends are nontoxic, nonflammable, and compatible with a range
of products. Because of these qualities, the aerosol product market grew
tremendously between 1947 and 1974. By 1974, United States sales of aero-
sol products reached almost 3 billion units and now includes a variety of
products. The chlorof luorocarbon propellants are used in approximately 50%
of the aerosol units, hydrocarbon propellants are used in 45% of the units,
and nitrous oxide and carbon dioxide are used in the remaining 5% of the
aerosol units Ji'
Market surveys showed that in 1971, the average number of aerosol
units (including food products) per household reached 40.8. Approximately
one-half of these units were personal products and about one-fourth were
household cleaners, laundry products, etc«-= Thus, in less than 30 years,
the aerosol products have become a major consumer market.
2. Basic aerosol systems: Aerosol propellants are of two basic
types: liquefied gases and compressed gases. The liquefied gases usually
have boiling points between -45°F (-43°C) and 105°F (41°C) and vapor pres-
sures which can range up to 125 psig (139.7 psia) at 70°F (21°C)^' The
most commonly used liquefied gas propellants have a vapor pressure of 13.4
to 85 psia at 70°F (21°C) and boil below 70°F (21°C)^/ A compressed gas,
by Department of Transportation definition, has a vapor pressure which ex-
ceeds 25.3 psig at 70°F (21°C) or 89.3 psig at 130°F (54.4°C). In the aero-
sol industry, a propellant gas is considered a compressed gas if it has a
vapor pressure above 125 psig at 70°F
The liquefied gases are desirable propellants because they often
also function as a self-pressurizing solvent in the aerosol formulation.
They maintain a fairly constant pressure within the container throughout
product use. In contrast, the compressed gas propellants do not function
as solvents, and in addition, the internal container pressure drops as the
product is used and the gas space increases. Presented below are general
discussions of the liquefied and compressed gas propellants and the typical
aerosol systems used for each.
a. The liquefied gas aero.sol system;-=i£' Liquefied gas
propellants generally operate in either a two-phase or three-phase system
depending upon the propellant. Each of these systems will be discussed sep-
arately.
(1) The two-phase system; In the two-phase system,
the active ingredient is dissolved in the liquid propellant or a propellant/
solvent mixture. The two-phase system is so termed because the propellant is
47
-------
present in both the liquid and vapor state within the container. When the
aerosol valve is opened, the vapor pressure of the propellant forces the
active ingredient mixture through the valve, which in turn mechanically pro-
duces the spray.
Many of the "space sprays" (i.e., ultra-fine aerosol
sprays) operate in the "two-phase" system. These products generally have 2
to 20% concentrate, containing the active ingredient, and 80 to 98% propel-
lant. Pressure within the aerosol container is generally between 34.7 to
54.7 psia at 70°F (21°C) and spray particles range in size from less than
1 to 50 U-.
"Surface-coating sprays" also use the two-phase system,
but in contrast to the space sprays, contain less low-boiling-point propel-
lant (25 to 80% by weight) and more concentrate (20 to 75%). Particle size
varies between 50 and 200 41. Typical surface-coating sprays include hair
sprays, residual insecticides, perfumes and colognes, paints and coatings,
and topical sprays.
Current typical propellants of the two-phase spray sys-
tems include the chlorofluorocarbons or fluorocarbons F-ll (CC^F), F-12
(CC12F2), F-114 (CC1F2-CC1F2), F-142b (CH3CC1F2), and F-152a (CH3-CHF2).
(2) The three-phase system; Unlike the two-phase sys-
tem, the three-phase system can use propellants not miscible with the liquid
components of the aerosol formulation. The three-phase system consists of
a liquefied propellant, a water solution of active ingredient, and a vapor
phase. A fluorocarbon propellant is generally at the bottom of the water-
active ingredient solution, while a hydrocarbon propellant floats on top.
The three-phase system is present either in a two-layer
or a foam system. The two-layer three-phase system is further divided into
an aqueous or a hydroalcoholic active ingredient solution. None of the pro-
pellant is introduced into the aqueous solution. The aerosol spray is pro-
duced exclusively by "mechanical" breakup in the actuator via constriction,
valve orifice size, and tangential channels. Products of this type contain
from 5 to 10% propellant, and have an internal pressure of 15 to 20 psig
at 70°F (21°C); examples include moth-proofing sprays and many household
cleaners. Hydroalcoholic solutions use a mixture of fluorocarbon and hydro-
carbon propellants (e.g., butane, isobutane, and propane) which float on
top of the active ingredient solution and are dispersed by shaking. Many
insecticides and room deodorants contain hydroalcoholic active ingredient
solutions.
The foam three-phase system generally uses 15% or less
propellant (by weight) which is emulsified with the product. Most products
use between 6 and 10% propellant and operate at 30 to 40 psig at 70°F (21°C)<
48
-------
Both hydrocarbon propel lants and mixtures of F-114 and F-12 are commonly
used to propel shave creams, shampoos, and some topical Pharmaceuticals.
1 o ^ /
b. The compressed gas aerosol systems;*. * ' The com-
pressed gas propellants have a much higher vapor pressure range, 500 to 1,000
psig (514.7 to 1,014.7 psia) at 70°F (21°C), than do the liquefied gas pro-
pellants. The two most common compressed gas propellants are nitrous oxide
(0) and carbon dioxide
Compressed gases dispense products in a solid stream, wet
spray, or a foam. Because there is little or no reservoir of gas, when the
contents of the aerosol unit are expelled, the volume of the vapor phase
increases and the pressure drops. The solid-stream delivery system, which
represents a very small portion of the market, may use nitrogen as the pro-
pellant. The nitrogen is both insoluble and immiscible with the concentrates
which are generally in a semi-solid form. Products which at one time were
dispensed by this system include dental creams, medicinal ointments and
creams, cosmetic creams, vitamins, and some food products. Initial container
pressure at 70°F (21°C) is between 90 and 100 psia.
Dispensers of most foam food products, (e.g., whipped creams
and toppings) employ either nitrous oxide or carbon dioxide as the propel-
lantj both are soluble in the concentrate.
Compressed gas spray dispensers use a mechanical breakup ac-
tuator to produce a wet spray. Choice of a compressed gas propellant depends
upon the nature of the active ingredient solution and the compressed gas'
potential reactivity with the active ingredient solution (e.g., C02 may form
carbonic acid with water-based active ingredient solutions). Some medicinals
in aqueous solutions, some residual insecticides, foods, and beverage concen-
trates are delivered by this system.
3. The chlorof luorocarbon and fluorocarbon propellants; Most
chlorof luorocarbons and f luorocarbons are nonpolar, generally chemically in-
ert compounds that maintain a nearly constant delivery pressure and a large
expansion ratio. Chlorof luorocarbons are miscible with most nonpolar com-
pounds and many are solvents over a wide range of temperatures. Due to the
propellants1 inertness, aerosol formulators need consider hydrolysis reactions
which may form hydrochloric acidH' and reactions that occur only under very
specific circumstances, e.g., the reaction of F-ll with ethanol.
Chlorof luorocarbons most commonly used as aerosol propellants are
F-ll, F-12, and F-114. F-115 is used only in propelling food products. FC-318
(perf luorocyclobutane), previously used as a food propellant, is currently
produced only in 100- or 1,000-lb batches and has almost completely been re-
placed by F-115 for economic reasons. Addition of the final fluorine molecule
in producing FC-318 greatly increased the per-pound price. Du Pont estimates
49
-------
that, even at high volume production, FC-318 would still cost about $5.00/
lb»Z' For comparison of total fluorocarbon usage in aerosols and their price
per pound, a summary of production and usage data is presented in Table IV-1.
Of the fluorocarbons listed in Table IV-1, only FC-318 is least
suspect in the current ozone depletion controversy. Further discussion of
this fluorocarbon, in addition to other candidate alternative fluorocarbon
propellants, can be found in Section B of this chapter.
4. The chlorocarbon propellants; In the past, three chlorocarbons
(methylene chloride, ethyl chloride, and vinyl chloride) have found limited
use as aerosol propellants; most of the remaining chlorocarbons are not suf-
ficiently volatile to function as propellants. In fact, methylene chloride
is employed primarily as a solvent and/or vapor pressure depressant in aero-
sol formulations rather than as the propellant. In addition to methylene
chloride, 1,1,1-trichloroethane (CH3CC13) and perchloroethylene (CC12:CC12)
have functioned as solvency constituents in aerosol formulations. Obviously,
vinyl chloride is no longer used in propellant systems. The chlorocarbons
are frequently used to replace or decrease the concentration of the more
expensive F-ll in propellant blends for insecticides, hair sprays, and room
deodorants^'
The feasibility and suitability of the chlorocarbons as alterna-
tive aerosol propellants are discussed in Section B of this chapter.
5. The hydrocarbon propellants; Hydrocarbons did not become im-
portant as aerosol propellants before 1954 because they were considered to
be too flammable and to possess an unpleasant odor. However, in 1954 two
events occurred which resulted in a rapid growth in the usage of hydrocarbon
propellants. First, Risdon Manufacturing Company developed a valve which
mechanically divided the product stream into a spray, and second, Phillips
Petroleum offered bulk quantities of "Pure Grade" (with little or no odor)
hydrocarbons. In addition, patent litigations encouraged many shave cream
marketers to reinvestigate hydrocarbon propellants.^' Thus, many heretofore
chlorofluorocarbon-dominated markets were opened to hydrocarbon propellants.
By 1974, hydrocarbons were used in 45% of the approximately 3 bil-
lion aerosol units. In 1974, approximately 116 million pounds of hydrocarbons*
were used as aerosol propellants. Hydrocarbons are used in approximately 87%
of the nonpersonal product units. With the exception of shaving lathers (ap-
proximately 90% of which are hydrocarbon propelled), hydrocarbons are used
only as constituents of propellant blends in personal product formulations.
The hydrocarbon component cannot exceed 10% and maintain nonflammability of
the propellant blend. The addition of 10% hydrocarbon and resultant reduction
of the chlorofluorocarbon propellant lowers the cost of propellant per aerosol
unit without adversely affecting spray characteristics.
MRI estimate based on data in Reference 3.
50
-------
TABLE IV-1
USAGE OF FLUOROCARBONS IN AEROSOLS 7'8>9/
Fluorocarbon
F-ll
F-12
F-114
F-115
FC-318
Production
1974
(millions Ib)
347
509
27-
W&
< 0.01
Percentage
to
Aerosols
82
60
95
10
100
Millions of
Pounds
to Aerosol
285
305
26
-
< 0.01
Price/
Ib (c)
38
42
49
51.6
500
£/ NA = Not available. F-115 production is included in the class "others."
"Other" fluorocarbons include F-13, F-14, F-21, F-115, F-116, and others
and had a 1974 production of 14 million pounds.
-------
Three hydrocarbon propellants account almost exclusively for the
116 million pounds used in aerosols. Approximately 84% of the total is iso-
butane, 16% propane, and < 0.1% is ji-butane. N-pentane and isopentane have
only very limited uses, primarily as solvents or propellant extenders. They
are both very volatile solvents, but do not have the necessary vapor pres-
sures required of a propellant at ambient conditions^ (See Table IV-2 for
physical properties.)
The hydrocarbon propellants may be utilized alone, in blends with
other hydrocarbons, or in blends with chlorofluorocarbons. They have greater
solubility and are much less expensive than the chlorofluorocarbon propel-
lants. The hydrocarbons have a density of less than one, are not miscible
with water, are not subject to hydrolysis, and are therefore useful in two-
layer three-phase aerosol systems. They are generally very stable and react
with halogens only under very severe conditions.-2
In pure or aerosol grades, the hydrocarbon propellants are virtu-
ally odorless, colorless, and nontoxic. Liquefied hydrocarbons propellants
have nearly constant delivery pressure and large expansion ratio, as do the
fluorocarbons JJ' At the same weight percent of concentrate, the dispersive
properties of the hydrocarbons are better than those of f luorocarbons^'
However, they are not widely used in personal products because of their
flammability and potential explosiveness in air. If a flame front forms
within an enclosure, a rapid expansion of gases will sometimes cause an ex-
plosion^' An aqueous solution can effectively suppress the hydrocarbon's
flammability and explosive potential; this fact and the lower cost explains
why 90% of the shave lathers (an aqueous solution of active ingredient) are
propelled by hydrocarbons. Hydrocarbons are also used to propel active in-
gredient solutions which are already flammable, formulated with some water,
or not directed towards the body. Representative products include paints,
shave lathers, shampoos, room deodorants, and household cleaners»2'
The physical properties of the major and several alternative hy-
drocarbon propellants are shown in Table IV-2. A discussion of these propel-
lants with respect to their feasiblity as replacements for F-ll, F-12, F-114,
and F-115 is presented in Section B of this chapter.
6. The compressed gas propellants; A propellant is often con-
sidered to be a compressed gas by aerosol scientists if the vapor pressure
exceeds 125 psig at 70°F (21°C). For use in aerosols, a compressed gas is
defined as a gas which can be liquefied only at very low temperatures or
very high pressures^' The compressed gases have very little expansion power
and therefore produce coarse, wet sprays. The initial pressure within the
container is usually 90 psig (104.7 psia)£/ but drops as the product is used
due to an increase in volume of the gas' vapor phase in the container's head
space. Products often propelled by these inexpensive, nonflammable, nontoxic
gases include windshield de-icers, furniture polishes, low-foam cleaners,
52
-------
TABLE IV-2
Vapor Pressure
Fluorocarbons
Number
14
21
22
23
115
116
132a
133
142b
01 152a
U> 218
227a
C318
502
3110
13E1
Vinylidene fluoride
Vinyl fluoride
Chlorocarbons
Methyl chloride
Ethyl chloride
Vinyl chloride
Methyl chloroform
Hethylene chloride
Dichloroethylene
Formula
CF4
CHCljP
CHC1F2
CHF3
CC1F2-CF3
CF3-CP3
CHC12CHFZ
CHJC1CF3
CH3CC1F2
CH3CHP2
C3F8
C3HF7
C4F8
Azeotrope
48.87. F22
51. n TllS
C4F10
CBrFj
Cl^-CFj
CHj-CHF
CHjCl
CHjCHjCl
CHj-CHCl
CH3CC13
CB2C12
CHC1-CHC1
Boiling
Point
CC/'F)
-128/-198
9/48
-4 1/ -41
-B4/-U9
-39/-3B
-78/-109
58/136
5/42
-10/15
-25/-13
-39/-3S
-26/ -16
-6/22
-46/-50
-2/-2B
-S8/-72
-86/-122
-72/-9S
-24/-11
12/54
-14/7
74/165
40/104
60/140
(psia)
70'F
(21'C)
23
138
6 IB
120
445
3
26
44
76
-
80
40
150
-
214
-530
370
74
20
46
2
7
NA
130*F
(54'C)
65
312
-
267
-
13
71
107
191
_-
193
107
342
-
449
-
760
174
56
126
7
24
NA
Flammability^'
(7. Vol. in Air)
None
None
None
None
None
None
None
None
9.0-14.8
5.1-17.1
None
None
None
None
None
None
NA
2.6-21.7
7.6-19.0
3.6-14.8
4.0-22.1
None
None
5.6-13.0
Solubility
Water
(25'C/77'F)
0.0015 wt I
0.95 wt I
0.13 wt I
75 cm3/ 100 ml
0.006 irt I
Insoluble
751 ml/ 100 g
91 ml/ 100 g
....
0.5 ml/100 g
1.0 ml/100 g
0.5 ml/100 g
NA
0.5 ml/ 100 g
.
0.0095 wt Z
(70'F/21'C)
Slightly soluble
Insoluble
400 cm3/ 100 ml
0.574 g/100 ml
(20'C/68'F)
255 ml/ 100 g
Insoluble
2 g/100 ml
(20'C/68'F)
Slightly soluble
Alcohol
(25'C/77°F)
394.5 cm1/ 100 ml
Soluble
NA
NA
NA
NA
NA
MA
Soluble
Soluble
3,500 cm3 /100 ml
48.3 g/100 ml
Soluble
Infinity
Infinity
Soluble
Hydrolysis
in Water
(e HCl/vear)
Without Steel
w£'
< 0.01 (86'F/
30' C)
< 0.01 (86'F/
30-'C)
0.005 (68'F/
20'C)
NA
0.005 (68°F/
20'C)
NA
NA
NA
NA
0.001 (68'F/
20'C)
0.005 (68'F/
20'C)
0.003 (68'F/
20'C)
< 0.01 (68'F/
20'C)
0.001 (68'F/
20'C)
NA
NA
NA
NA
NA
0.1 (68'F/
20'C)
NA
NA
NA
With Steel
Strips T
NA
5.2 (86'F/
30' C)
0.1 (86'F/
-' 30'C)
NA
0.05 (68'F/
20'C)
NA
NA
NA
NA
0.04 (68'F/
20'C)
0.06 (68'F/
20'C)
0.04 (68'F/
20'C)
< 0.1 (68°F/
20'C)
0.005 (68'F/
20'C)
NA
NA
NA
NA
NA
' 0.155 (68'F/
20'C)
NA
NA
NA
'OKtcttyfe
6
4-5
5a
6
6
6
NA
NA
NA
6
NA
NA
6
5a
HA
6
• HA
HA
4
4a
-
4-5
4
TIV
f_ (ppm)
NA
1,000
HA
HA
HA
NA
1,000
100
1,000
200
350
500
200
-------
TABLE IV-2 (Concluded)
m
Vapor Pressure
Fluorocarbons
Number
Hydrocarbons
n- Sutaae
Isobutane
Propane
Pentane
Isopentane
Ethane
Propylene
Compressed Gases
Carbon dioxide
Nitrogen
Nitrous oxide
Sulfur dioxide
Proprietary mixture^'*
Ethers
Dimethyl ether
Vinyl methyl ether
Misc. Cocpounds
Methyl formate
Ethyrylsulfur-
pectaf luoride
Formula
n'C4H10
l'C4H10
C3H8
C5H12
i-C5H12
CH CH
'3*6 '
co2
"2
N,0
s82
• -
CHjOCH,
CB^-CHOCHj
C2B4°2
HCsCSF5
Boiling
Point
{•C/-F)
-1/31
-12/11
-42/-44
36/97
28/82
1
-48/-S4
.
-196/-320
-88/-127
-10/14
NA
-25/-13
5/41
24/75
6/43
fosla)
70'F
(21'C)
31
45
124
8
12
555
171
757
-
r
755
50
-
78
1,348
593
-
130'F Flamnablllty1'
<54°C) (X Vol. in Air)
81 1.6-6.5
110 1.8-8. A
274 2.2-9.5
27 1.4-8.3
34 1.4-8.3
2.6-10.5
387 2.0-11.0
None
None
None
135 None
None
187 3.4-18.0
2.6-39.0
4.5-20.0
-
Hydrolysis in Hater
Solubility (8 RCl/year)
Water
<25'C/77'P)
15 cm3/100 ml
(17'C/63'F)
13 cm3/ 100 ml
(17°C/63°F)
6.5 cm3/100 ml
(IB'C/64'F)
0.03 g/100 ml
(16°C/61'F)
Insoluble
10.2 ml/100 g
Very soluble
8.2 ml/100 g
60 ml/ 100 g
56.7 cm3/ 100 cc
22.8 g/100 cc
(0"C/32°F)
Soluble
31.5 wt X
Slightly soluble
Slightly soluble
-
Alcohol
(25*C/77*F) Without Steel
1,813 cm3/ 100 ml NA
(17-C/63'F)
1,320 cm3/100 ml
(17'C/63'F)
790 cm3/ 100 ml
(17'C/63'F)
Infinity NA
Infinity NA
Slightly soluble 0
Very soluble NA
31 cm3/ 100 cc NA
(15-C/59'F)
Slightly soluble (68°F/20'C)
very slight
Soluble
Soluble . -
Soluble 0
Soluble
Very soluble
Slightly soluble NA
-
With Steel TL7
Strips Toxicity—^ (ppa)
NA 5
5b
•96 1,000
NA 1.000
NA
0
NA
NA 5a 5.000
(68*F/20'C) - -
very slight
1 ' S
0 ' HA 3.000-4.000
NA 3 100
-
a/ None • nonflammable compound.
b/ Underwaters' Laboratory Toxicity rating; I - moat toxic, 6 - least toxic.
c/ NA " not available.
d_/ Tabline Company, Berkeley, California.
-------
engine cleaners, car waxes, residual insect sprays, dog-groomers, and food
products such as whipped creams and toppings. Compressed-gas units account
for approximately 5% of the aerosol salesJt' Several once exclusively
fluorocarbon-propelied products, e.g., hair spray, and now using carbon di-
oxide as the propellant, are currently being test-marketed*Zil2' Recent re-
finements in aerosol valves may make carbon dioxide systems viable for
deodorants and some other personal products.-^'
Carbon dioxide and nitrous oxide are often used as a combination.
The compressed gases are only rarely combined with the fluorocarbon propel-
lants.,3' Chlorocarbons (such as methylene chloride and 1,1,1-trichloroethane)
may be added to nitrous oxide to help compensate for pressure loss during
product misuse, i.e., loss of the propellant if the actuator is depressed
while the container is not upright. However, the chlorocarbon:N20 mixture
still produces a coarse sprayJL'
Compressed gas/liquefied propellant systems have also been con-
sidered. Either C02 or NoO were used to pressurize the fluorocarbon propelled
system to 90 psig (104.7 psia). The initial spray is finely divided but spray
characteristics soon return to those of the liquefied system without the com-
pressed gas. Thus, there is no advantage in superior spray characteristics or
decrease in cost since the amount of liquefied propellant is not reduced^
A discussion of the advantages, disadvantages, and technical feasi-
bility of the compressed gases as alternative propellants is shown in Section
B-3 of this chapter. A summary of physical properties occurs in Table IV-2.
B. Evaluation of Alternative Systems
To qualify as an aerosol propellant, a compound must possess cer-
tain physical and chemical characteristics. In addition, the environmental
hazards, toxicity, manufacturing technology, and cost of the compound must
be considered. A brief discussion of each of the above criteria is presented,
followed by an initial compilation of candidates selected on the basis of
physical characteristics only. From this initial compilation, candidates
were eliminated on the basis of toxicity, incompatibility with active in-
gredient solutions, and/or potential environmental hazard. Lack of knowledge
of the required manufacturing technology and high cost of the candidate pro-
pellant were not considered to be prime elimination criteria. Any compound
considered to be a feasible aerosol propellant in terms of its physical,
chemical, and toxicological properties was not eliminated from considera-
tion. However, in some instances data are incomplete because of the compound's
early developmental stages. Such data gaps will be noted in the brief discus-
sion presented for each of the selected candidate propellants.
55
-------
Aerosol units are classified into the following three types:
1. Personal products which include:
a. Hair care;
b. Antiperspirants and deodorants;
c. Medicinal and pharmaceutical;
d. Colognes and perfumes;
e. Shave lathers; and
f. Others.
2. Household products which include:
a. Room deodorants;
b. Cleaners;
c. Laundry products;
d. Waxes and polishes; and
e. Others
3. Other products which include:
a. Insecticides;
b. Coatings;
c. Industrial;
d. Food products and pan sprays;
e. Automotive;
f. Veterinary and pet; and
g. Others.
Personal products account for approximately 50% of all aerosol
units, household products for 24% of the market, and all other products for
26% of the market. Of the personal products, 84% are currently propelled by
fluorocarbons, while only 18 and 19% of household and all other products,
9Q19/
respectively, are fluorocarbon propelled. >' '
Replacement of F-ll, F-12, or F-114 in household and "all other"
products should not present severe technical problems since these products
are predominantly propelled by hydrocarbons, although some use chlorocar-
bons and compressed gases. In many instances, individual product types are
exclusively nonfluorocarbon propelled.
The personal products represent the reverse situation. The shave
lathers are the only product type not predominately fluorocarbon propelled.
Consequently, alternative candidate propellants were selected to replace
chlorofluorocarbons in the other personal product classes. Many of the in-
itial candidate propellants were eliminated because they were not suitable
for use in personal products, although they are currently used in other
products.
56
-------
1. Selection criteria for candidate propellents; Physical and
chemical properties, consumer safety standards (i.e., low toxicity), envi-
ronmental hazard, industry knowledge of required technology plus the cost
of the propellant are all factors that must be considered by the aerosol
industry in identifying candidate alternative propellants. This section
presents the requirements and/or suitability ranges for each of these cri-
teria.
a. Physical and chemical propertied > •* *
(1) Boiling point and vapor pressure; The boiling
point of a liquefied gas propellant (or propellant blend) must generally
fall between -45°F (-43°C) and 105°F (41°C) to achieve suitable aerosol per-
formances The vapor pressure of the liquefied propellant, or blends, should
be between 10 to 70 psig (24.7 to 84.7 psia) at 70°F (21°C) for proper aero-
sol performance. Most viable propellants have a vapor pressure in the 20 to
50 psig (35 to 65 psia) range. Vapor pressures outside this range often give
undesirable spray characteristics, or are not within the pressure limitations
of the containers as defined by Department of Transportation regulations.
The vapor-pressure relationship with respect to tempera-
ture fluctuations must be considered to ensure adequate product performance.
If, for example, the vapor pressure of a propellant drops to 20 to 30 psia
at near freezing temperatures, it would not be a viable propellant for prod-
ucts such as windshield de-icers since internal container pressure would
not be sufficient to expel1 the active ingredients. Conversely, should vapor
pressure rapidly increase with temperature increase, storage in warm places
may result in hazard as a result of overpressurizing of the container.
The vapor pressure of a liquefied gas propellant can
be increased or reduced only by blending the propellants of varying vapor
pressures or by the use of vapor pressure depressants. Reducing the amount
of propellant per container is not effective in reducing internal pressure.
The amount of propellant included in the aerosol formulation, for economic
reasons, is generally near the minimum amount required to expell all the
product unless an ultra-fine spray is required for optimum product perfor-
mance.
(2) Flammability and explosiveness; Propellants for
personal products are generally required to be nonflammable, nonexplosive,
and nonsupportive of combustion. A flammable propellant can often be used
in aqueous-based personal products because the water adequately suppresses
the flamraability. However* the combination of a hydrocarbon with an alcohol
solvent, as in a hair spray, would greatly increase flammability and pose
a potential hazard.
57
-------
Flammability of a propellant (or propellant mixture)
is determined on the basis of Department of Transportation and the New York
City Fire Department standards. A flash point below 20°F (-7°C) (open cup
test) is classified as extremely flammable; a flash point between 20°F (-7°C)
and 100°F (38°C) (open cup test) is considered flammable. In addition, the
Department of Transportation tests include a flame extension test and open
and closed drum tests.
In general, hydrocarbon propellants are used when there
is no increase in flammability hazard. The direction of the spray during
the consumer use of the product is a factor. The product's flammability is
of greater importance if the spray is directed towards the body than if the
spray is directed away from the body, and is used either out-of-doors or in
a well-ventilated area. There are no governmental regulations against the
use of hydrocarbon propellants in personal products.
(3) Compatibility; A suitable propellant must be com-
patible not only with the active ingredient, but also with the container and
valve. In many cases, incompatibility with the container can be resolved by
either coating the inside or by inserting an impermeable container lining.
Valve incompatibility can often be eliminated by redesign. However, no re-
actions should occur which may shorten the product's shelf-life or produce
products that react with the aerosol formulation itself. Product stability
tests are conducted at room temperature, 100°F (38°C) and 120°F (49°C) to
assure a propellant's compatibility.
(4) Purity; Major propellants are available in 99%+
purity. Impurities can affect a product's odor and corrosiveness (and there-
fore shelf-life). A propellant used in most personal products must have a
very high odor threshold, or the aesthetics of the products may no longer
be acceptable.
1 ^ 7 /
b» Toxicity; L.*Jt" A potential propellant must also have
low toxicity, particularly for use in personal products. Before a propellant
is utilized in consumer products, tests must be conducted to determine both
acute toxicity (LCso) and chronic toxicity. The acute toxicity of most com-
mon personal product propellants is generally several hundred or thousands
parts per million, or greater than 20% in air, making them virtually non-
toxic.
Of greater importance than the LC$o is the effect of long-
term chronic inhalation of the propellant. A threshold limit value (TLV) or
a time weighted average (TWA) must also be determined and shown to pose no
long-term hazard. The fluorocarbon propellants usually have a TLV of 1,000
ppm. The consumer is typically exposed to between 10 and 400 ppm of the pro-
pellant during product use, depending on product type, mode of application,
and location of use. This level rapidly declines to nondetectable amounts
within seconds, and is always below TLV concentrations.
58
-------
Also included in chronic toxicity is determination of the
propellant's cardiac sensitization potential. Consumer misuse by deliberate
propellant inhalation can result in exposures well above the 1,000 ppm level,
i.e., 50,000 to 100,000 ppm. Under the stress conditions at which most mis-
use events occur (e.g., increased adrenalin, etc.), the level at which a
propellant causes cardiac sensitization must be high enough that some margin
of safety exists even under abuse situations.
In addition to low hazard in acute toxicity, chronic inhala-
tion, and cardie sensitization potential tests, a propellant must also be
nonmutagenic, noncarcinogenic and nonteratogenic.
c. Environmental hazard: On the basis of the current ozone-
depletion theory, candidate propellants of very high atmospheric stability
and a propensity for reaction with the ozone layer were considered to pra-
sent a potential environmental hazard. Such compounds were thus not included
in the selected list of candidate propellants.
d. Manufacturing technology and product cost:~ For a pro-
pellant to be introduced to the aerosol market in a limited time (i.e., less
than 5 years), a company must be familiar with its manufacturing technology.
Development of a new propellant where there is no familiarity with the manu-
facturing technology requires at least 5 to 10 years, and possibly longer
if raw materials are not available in bulk quantities.
The cost of the propellant is very important from a marketing
standpoint, and is often closely linked to knowledge of the manufacturing
technology.
Although not an elimination criterion in this report, inor-
dinately high product cost and, to a lesser degree, ignorance of the needed
manufacturing technology may result in abandonment of a propellant by the
company.
2. Compilation of potential alternative propellants; Compila-
tion of the initial list of candidate propellants was on the basis of phy-
sical properties only. Candidates for the list were assessed by consulting
references on past and current aerosol technology!^' industry technical
bulletins,JL2' and compendiums of physical and chemical properties.1^"1'/
Toxicology datai2juULl2P.' were also collected for later use as elimination
parameters.
Table IV-2 presents the candidate propellants and summarizes their
toxicity, reactivity, and physical and chemical properties.
3. Selected list,of alternative propellants; The initial list
of potential alternative propellants for personal product aerosol (see Table
IV-2), was evaluated first by eliminating (on the basis of physical and
59
-------
chemical properties) the unsuitable candidate propellants, secondly by iden-
tifying those candidates suitable for use in blends of aerosol propellants,
thirdly by identifying candidates which may be technically and commercially
feasible propellants in 5 to 10 years, and finally, by identifying those
candidates that may be (both technically and commercially) suitable as per-
sonal product propellants within 5 years. As discussed previously (pp. 56
to 57), alternatives for nonpersonal product aerosol were not considered
because (a) these products are currently available in mechanical delivery
systems, and (b) approximately 82% of these products use propellants other
than the chlorofluorocarbons.
Presented below is a discussion of the propellants in the four
categories along with the criteria for including them in their respective
categories.
a. Eliminated potential alternatives; Fifteen of the 39
candidate propellants wer|e eliminated from consideration. It should again
be noted that F-ll, F-12, F-13, F-113, and F-114 were not considered for
further usage in aerosols within the framework of this document because of
their potential environmental threat.
To function as an aerosol propellant, a candidate must pos-
sess the suitable physical properties. The desirable vapor pressure range
is 25 to 95 psiailtZ/ at 70°F (21°C), however, a propellant can have an am-
bient vapor pressure up to 140 psia.-?-' The boiling point of the compound
should be between -45°F (-43°C) and 105°F (41°C). Seven of the 15 candidates
were eliminated strictly on the basis of unsuitable physical properties (see
Tables IV-2 and IV-3).
F-14, eliminated on the. basis of unsuitable physical proper-
ties, also has a very long atmospheric residence, i.e., > 10 years. It is
possible the F-14's high stability may lead to environmental problems in
the future.
When using an aerosol product, the consumer may be exposed
to from 10 to 400 ppmZ' of the propellant depending upon the product and the
mode of use. Candidates which have a TLV of 400 ppm or less, are toxic by
inhalation, or which pose a potential health hazard were also eliminated.
Candidate propellants were downgraded from consideration if,
on the basis of the current ozone-depletion theory, they pose a potential
'environmental hazard. Compounds of this group are F-115, F-502 (an azeotrope
of F-22 and F-115), and F-13B1.
The critical properties and boiling point of F-115 would make
this compound a plausible candidate as a liquefied gas propellant. However,
60
-------
TABLE IV-3
ELIMINATED CANDIDATE PROPELLANTS
Elimination Criteria
Propellant
F-14
F-23
F-115
F-116
F-502
F-13B1
Vinylidene Fluoride
Vinyl Fluoride
Vinyl Chloride
Dichloroethylene
Sulfur Dioxide
Vinyl Methylether
Methyl Formate
Ethynylsulfur
Pentafluoride
Boiling
Point
Too low
Too low
Too low
Too low
Too low
Vapor
Pressure
Too high
Too high
Too low Too high
Too high
Too high
Too high
Too high
Too high
/
-lack of data-
Potential
Reactivity
With Ozone
Yes
Yes
Yes
Toxicity
May be toxic by in-
halation
A carcinogen
TLV 200 ppm
TLV 5 ppm
TLV 100 ppm
61
-------
Q/
the atmospheric residence of F-115 exceeds 10 years— and since F-115 con-
tains chlorine and no hydrogen, it may be reactive with the ozone layer.
F-502 contains 51.2% F-115 and the azeotrope would consequently have basic-
ally the same potential ozone reactivity as would pure F-115. F-13B1, the
bromine analogue of F-13, was also downgraded as a candidate propellant.
Bromine-containing compounds are more susceptible to photodecomposition than
chlorine compounds. Since bromine radicals are as effective as chlorine rad-
icals in terms of ozone scavenging, F-13B1 may also deplete the ozone layer.
Ethynylsulfur pentafluoride has been suggested as a possible
aerosol propellant (Chemical Abstracts. 616, 1967, 55005y). The company that
employed the inventor was contacted, but had no information on the candidate
propellant^Z' Additional data sourcesMp* 15-21/ an(j later volumes of Chemi-
cal Abstracts were consulted, but no information was found. Ethynylsulfur
pentafluoride could not be assessed as a potential propellant because of
lack of data, and was therefore eliminated from consideration as a selected
alternative candidate propellant. In addition, ethynylsulfur pentafluoride
is probably too expensive for most uses.
The proprietary compressed mixture reportedly developed by
Tabline Company, Berkeley, California,—' was also eliminated from consid-
eration as an alternative candidate propellant. Very limited data were avail-
able on the mixtures composition. It is said to have solubility in both al-
cohol and water which may present problems during product formulation. Thus,
due to insufficient data (considered to be proprietary), this compressed
gas system was eliminated from final consideration as an alternative. As
additional data become available, this mixture may prove to be a viable al-
ternative but no published information is available at this time.
The candidate propellaats eliminated or downgraded from con-
sideration are summarized in Table IV-3.
b. Candidates suitable as constituents of propellant blends
for personal products; Fourteen of the candidate propellants were judged
to be suitable as propellant-blend constituents. Many of these compounds
are currently used as propellants for shave lathers, household, and other
aerosol products. Several of these compounds are used as vapor-pressure de-
pressants with some fluorocarbons, while others may serve a dual function
of vapor depressant/solvent. This section presents a brief discussion of
the chemicals included in this category. Numerical order does not indicate
an order of preference.
(1) F-142b; This fluorocarbon is probably nontoxic
(complete toxicological data are not currently available) and has ideal phy-
sical properties from the .standpoint of a suitable aerosol propellant (see
Table IV-2 for physical properties). F-l42b is flammable at 9.0 to 14.87.
volume in air. However, flash point tests showed F-142b not to flash even
62
-------
to dryness^i' Industry sources indicate that up to 60% F-142b, by weight,
can be mixed with a nonflammable propellant without producing a flammable
propellant blend.2' F-142b may be excellent for use with several of the se-
lected alternative fluorocarbons (see pp. 64 to 66) to produce blends with
a wide range of vapor pressures.
(2) F-152a; F-152a has a higher vapor pressure than
does F-142b, is nontoxic, and is slightly more flammable than F-142b (i.e.,
5.1 to 17.17o volume in air). Mixtures of F-152a and nonflammable fluorocar-
bons would require relatively less F-152a than F-142b to maintain nonflamma-
bility. However, F-152a could be used in blends to produce a wide range of
vapor pressures.-^' If the blend remains nonflammable, it is suitable as
an alternative propellant system for personal products.
(3) n-Butane; _n-Butane is very flammable (flash point -
101°F (-74°C),1/ 1.6 to 6.8% volume in air flammability limits) but is non-
toxic (Underwriters' Laboratory Class 5) and has suitable vapor pressure and
boiling point properties. Its current volume of use in aerosol is not high,
i.e., less than 0.1% of the total hydrocarbon aerosol propellants^x' ji-Butane
is often used in blends with other hydrocarbons and may be used with nonflam-
mable fluorocarbons. The blend can contain up to approximately 10% n-butane
and remain nonflammable.
(4) Isobutane; A common propellant in nonpersonal prod-
ucts (and in shaving lathers), isobutane accounts for approximately 847» of
all hydrocarbons used in aerosols^' A personal-product propellant blend with
a nonflammable compound may contain up to 107. isobutane and remain nonflamma-
ble. A common blend of 45% F-ll, 45% F-12, and 10% isobutane is currently
used in many hair sprays without changing the spray characteristics produced
by the more expensive 50:50, F-ll/F-12 propellant mixture. There seems to be
no technical reason why isobutane cannot be used in a similar manner with
the selected alternative, nonflammable, fluorocarbons to produce a propellant
blend suitable for use in personal products.
(5) Propane; Propane accounts for approximately 16%
of the hydrocarbon propellant sales.2' It has a high vapor pressure (124 psia,
70°F/21°C) under ambient conditions and may be excellent in blends with low
vapor pressure fluorocarbons.
(6) j-Pentane; jv-Pentane has a very low vapor pressure
(8 psia at 70°F/21°C), is a good solvent, but is more flammable than either
propane or isobutane. In amounts of less than 10% of the blend, ji-pentane
may serve as a vapor depressant/solvent in combination with nonflammable
fluorocarbons.
63
-------
(7) Isopentane; Isopentane has properties very similar
to those of ji-pentane although ambient vapor pressure is somewhat higher (12
psia at 70°F/21°C). Isopentane can feasibly function in the same mode in a
propellant blend as can j}-pentane.
(8) Propylene; This hydrocarbon is not commonly used
as an aerosol propellant probably because of its high vapor pressure. Small
amounts of propylene with a low vapor pressure propellant could possibly be
used to produce a blend with a suitable vapor pressure.
(9) Ethane; Ethane is nontoxic but flammable. Ambient
vapor pressure is very high, i.e., 555 psia. Small amounts of ethane may
function as would propylene in a propellant blend.
(10) Dimethyl ether; Nonflammable propellant blends of
75% F-12, 10% F-ll, and 15% dimethyl ether have been used in aerosol prod-
ucts^l' -It is technically feasible that dimethyl ether may be used with the
selected alternative fluorocarbons to produce similar propellant blends.
(11) Ethyl chloride; This chlorocarbon is an extremely
powerful elastomer solvent, is nontoxic (TLV = 1,000 ppm), has a low ambient
vapor pressure (20 psia), hydrolyzes in water, and reacts with ethanol. It
is currently used in propellant blends and mixtures as a solvent and can
feasibly perform in the same manner with the selected alternative propellants.
(12) Methyl chloride; Methyl chloride is used in minute
amounts in presently marketed aerosol units Jzj2.' It is relatively toxic (TLV
= 100 ppm) and may hydrolyze in the body to form methanol^2' Methyl chloride
can probably be used with the alternative propellants in a role analogous
to its present use.
(13) Methyl chloroform; Methyl chloroform is nonflam-
mable but is fairly toxic (TLV =' 350 ppm). Because of its low vapor pressure,
small amounts of methyl chloroform are presently used as a pressure depres-
sant. It is possible that methyl chloroform can be used with alternative
propellants in the role of a vapor-pressure depressant.
(14) Methylene chloride; Methylene chloride is also
nonflammable, is less toxic than methyl chloroform, and does not hydrolyze
to a significant extent in water. It is a very strong solvent and at high
levels is a skin irritant. Methylene chloride can be used in small amounts
as a combination vapor depressant/solvent with the alternative propellants.
c. Candidates that may be suitable aerosol propellants in
5 to 10 years; Five of the candidate propellants may serve as excellent
alternatives in 5 to 10 years. These compounds, if currently available, are
only produced in experimental quantities. Data in most cases are incomplete
64
-------
making difficult if not impossible analysis of technical feasibility as can-
didate aerosol propellants. In addition, complete toxicological data are
not available.
Included among these compounds are the five following fluoro-
carbons: F-132a, F-133, F-227a, F-218, and F-3110. Table IV-2 may be con-
sulted for a summary of available physical and chemical properties.
d. Candidate alternative propellants; There are six candi-
date propellants suitable for use as personal-product propellants. All" these
candidates are nonflammable, have a low order of toxicity, and can be avail-
able in sufficient quantities to meet market demands in less than 5 years.
It should again be noted that alternative propellants for
nonpersonal products have not been considered. Many of the chemicals in-
cluded among suitable blend constituents function as acceptable propellants
alone, and could be used in nonpersonal products currently utilizing flam-
mable propellants. However, as discussed previously in this report, 84% of
the personal products are propelled by the fluorocarbons while only approx-
imately 18% of the remaining aerosol products use fluorocarbons. The switch
from 72 to 100% nonfluorocarbon propellants should offer no severe obstacles,
unlike an 84% replacement required in personal products. Therefore, atten-
tion has been focused on identification of alternative personal-product
propellants.
(1) F-21; F-21 is the first of the selected alterna-
tive propellants (consult Table IV-2 for properties). Some toxicological
data are lacking, but these should be available by June 1976.* The manufac-
turers must then formulate products using F-21 and conduct 90-day inhalation
studies. F-21 is a good solvent and may attack valves; however, industry
feels this problem can be overcome with valve redesign.-!'
F-21 can also be made at a reasonable cost. Plants form-
erly producing F-ll and F-12 can be converted to products F-21 and F-22 with
some capital investment (estimated to be about 15% of the current plant re-
placement value of $50 to $100 million).!/
(2) F-22; F-22 can also serve as a suitable aerosol
propellant. It can be blended with both F-142b and F-142a; blends of F-21
and F-22 may function as a1 replacement for F-ll/F-12 blends-Z' F-22 cur-
rently has a very limited usage in aerosols, i.e., in flat-tire inflation
cannisters .-=-£'
(3) FC-318; FC-318 has been used in the past as a food
.propellant and is approved as a food additiveJ;?-' The vapor pressure at 70°F
* A major manufacturer recently stated (February 1976) that toxicological
data indicates F-21 probably is not acceptable as an aerosol propellant.
65
-------
(21°C) is 40 psia; FC-318 could be blended with other fluorocarbons and blend
candidates of higher vapor pressures to produce propellant blends with a wide
range of ambient vapor pressures.
t
Although a technically feasible aerosol propellant, FC-
318 may be hindered from a marketing standpoint because of a high cost per
pound. Addition of the final fluorine and purification of the product are ex-
pensive processes (in both cost and energy). In addition, industry believes
that existing plants cannot be converted, and that FC-318 would require new
production facilities_Z'
(4) Nitrous oxide (N2
-------
combinations of low vapor pressure fluorocarbons and nitrogen may be suit-
able for propelling personal products.
(6) Carbon dioxide; Carbon dioxide has been used as
window cleaners, starch sprays, window de-icers,' engine starters, and resid-
ual insect sprays where a coarse spray is desirable, or where spray droplet
size is not highly related to effective product performance. However, recent
refinements in valve design (i.e., mechanical breakup into a finer spray)
have allowed test marketing of personal products once propelled only with
fluorocarbons. A carbon dioxide propelled hair spray (Alberto-Culver's
"TRESemme TWO") is now being test marketed, ^i,^' CC>2 may also be a suitable
propellant for deodorants. 10>llj27/ Qne industry source thinks that CC>2 could
be a viable propellant for one-fourth to one-third of the 3 billion aerosol
units sold annually.!!'
Carbon dioxide is not a viable alternative propellant
for every aerosol product. C02 can form carbonic acid in aqueous solutions.
A fine mist of space-spray size (1 to 50 U- particles) has not yet been pro-
duced with C02 systems. Therefore, C02 is not yet a viable alternative for
room deodorants, room insect sprays, antihistamines, asthma sprays, or paints
and coatings .-Li'
Carbon dioxide is much less expensive per aerosol unit
than are the fluorocarbons. Only 5 to 20 g of C02 are required per unit to
fully dispense the product, the cost of which is very low (~ 0.2
-------
a. Systems requiring hand application
(1) Collapsible tubes; Collapsible tubes were in many
cases the original form of product packaging for medicinals, shaving creams,
and many food products. Many products sold as aerosols are still marketed in
tubes. Of the total collapsible tube production in 1973 (~ 8 million gross), •
27.870 were used for medicinals and pharmaceuticals and 1.2% for shaving
creams. There were no food products packaged in tubes .^Z'
Collapsible tubes are a viable mechanical alternative
for many of the viscous products such as shave creams, hand creams, some
cosmetic products, cheese spreads, and medicinal and veterinary ointments.
(2) Glass and plastic bottles; These products are a
viable marketing system for hair products such as shampoos, setting lotions
and conditioners, colognes and perfumes, household cleaners, waxes, and laun-
dry products. The bottle was often the original form of product packaging.
The Modern Packaging Encyclopedia^"' and monthly issues
of Modern Packaging, and Soap/Cosmetics/Chemical Specialties, were consulted
for information on the market size of products offered in bottles which are
also offered as aerosols. All data found were for total market size, e.g.,
37.8% of all plastic bottles iri 1973 were used for packaging household clean-
ers.^2' Estimation of the proportion of products also dispensed in aerosol
form was not possible with this data. Consequently, although glass and plas-
tic bottles are often viable alternatives, no estimation was made on their
market magnitude.
(3) Creams, roll-ons, and sticks; These products were
the original product form for deodorants and antiper spirants. Sales of these
products in 1975 are above those in 1974. An industry source felt the dra-
matic increase in sales was due to the current ozone-depletion controversy,
and that if the growth continued, the market could easily quadruple by 1978.
Industry consumer sales in 1975 will probably exceed $25 million in creams
and $45 million in roll-ons^22/
Only deodorants have generally been offered in stick
form, and one company is almost to the point where it cannot meet .demands
for this product. To broaden its scope, the same company is currently test
marketing an antiperspirant in stick
Many colognes are also offered in stick form. Estima-
tions of the number of such products marketed in 1975 could be not obtained.
b. Barrier packs/bag-in-can, and piston containers; The
barrier pack system consists of a metal container with an inner plastic bag
that holds the product. A propellant is required, but is not mixed with the
product. The product is loaded, using conventional filling equipment into
68
-------
the inner bag. A propellant is then loaded through the bottom of the can,
surrounds the bag, and provides the dispersing force. The impermeable bag
therefore functions as a barrier between the product and the propellant.1»31»32/
Barrier packs were originally developed to dispense foods and
other very viscous products such as creams, ointments, lotions, toothpastes,
polishes, and cosmetic and pharmaceutical foams. The major bag-in-can systems
include the "Sepro" can (Continental Can Company), the "Sterigard" dispenser
(Sterigard Corporation), "Kang" (by Valois of France), "Power-Flo" (Sterigard
Corporation), and "Cebal" (by Cebal of France).1|31|3Z/
Another barrier pack type system is the piston container. The
piston container is a metal can constructed around a polyethylene piston.
The product is located above the piston; the propellant is inserted through
the bottom of the can, and is located below the piston. Nitrogen as well as
the other compressed gases function well in this system. This system is also
only suitable for viscous products. Major piston containers include the "Mira-
Flo" (American Can Company) and PTS, Pressure Tube System (Eyelet Specialty
Company).I*31/
The barrier pack systems are still in a developmental stage
and, as far as one industry source knew, are marketed only on a test and
experimental basis.-3-3-'
A spring-loaded mechanical piston device has been announced
in the media, but to the best of our knowledge, is still in a developmental
stage with the areas of specific applicability not yet defined.
c. The bladder system; The bladder system is basically a
barrier pack system, i.e., an impermeable bag within an outer rigid container.
However, this system does not require a propellant, but instead uses the con-
tracting force of the inner bag to dispense the product.
Kain's Research and Development Company has developed a blad-
der system called the Eco-Pure®. This system can disperse viscous products
with a low-pressure inner liner, and insecticides, deodorants, hair sprays,
^^disinfectants with a high-pressure inner liner. The cost of the Eco-Pure®
system should be lower than conventional systems because there is no need
for a propellant, and conventional containers and valves will require no
redesign.^;'
Another bladder system (developed by Plant Industries, Inc.)
termed SELVAC contains a two-layer inner membrane which functions as permea-
tion barrier and pressure source. The first SELVAC system has a discharge
pressure of 3 to 5 psi and is suitable for dispensing viscous products. A
variety of containers can be used (glass, plastic, metal, cardboard) pro-
vided they are rigid.35»36/
69
-------
Within the last 6 months, a second SELVAC system has been
developed that has a discharge pressure of 10 tov 20 psi. Delivery pressures
in this range would make the system suitable for many household products,
deodorants, and perhaps antiperspirants. Sales of both SELVAC systems are
estimated to be 1/2 million for
Filling of bladder system units can be performed using con-
ventional filling equipment, although a nominal amount of adaptor equipment
may be required, the cost of which should not exceed $10,000^2^'
d. Mechanical spray pumps; Mechanical spray pumps have long
been used in a wide range of products including household cleaners, insect-
icides, perfumes and colognes, medicinals, and hair care products such as
setting lotions. In 1972, the first hair spray using a mechanical pump was
marketed by ClairolJLZ'
Today's market (1975) is estimated to be between 18o!§/ and
300 million »n-n-g. 33,37.39.407 of this market, 40% of the units are used in
household products and insecticides, 10% in medicinal and pharmaceutical prod-
ucts, and 50% in hair care products, cosmetics, colognes, and perfumes. ^°»-^
The average mechanical spray pump costs about 10£. However, sprayers used
for a household product may cost only 7£, while those used in hair care and
cosmetic products may be priced from 13£ to 150. Industry dollar volume by
product type follows the same basic pattern.
The five leading manufacturers of mechanical spray pumps are:
Calmar (Division of Diamond International Corporation); VGA/ ARC (Dispenser
Division of the Ethyl Corporation); Bakan Plastics (Division of Realex Cor-
poration); Thiokol, The AFA Corporation; and Emson Pump Corporation^zl' One
company plans to double its plant capacity by March 1976. Construction of
entirely new production facilities is estimated to require approximately
$10 million of capital investment for the space, tools, etcJi2'
C. Alternatives for Specific Use Areas
This section presents a discussion of the delivery systems
may function as viable alternatives for the personal products. Table II-5
presents a summary and relative viability ranking for all aerosol products
by alternative system.
1. Hair spray; Hair sprays are currently marketed in mechanical
delivery systems and as chlorofluorocarbon propelled aerosols. Presented
below is a brief discussion of the mechanical systems and alternative pro-
pellants which may function as viable alternatives to the chlorofluorocarbon
propellants in question (i.e., F-ll, F-12, and F-114).
70
-------
a. Mechanical alternatives; Hair sprays dispensed by me-
chanical spray pumps have been marketed since 1972.37/ Nearly every major
hair spray marketer currently offers their product in both the fluorocarbon
aerosol form and in mechanical pump form, indicating some consumer acceptance
of the mechanical pump system. Criticism of the mechanical pump systems for
hair spray with regard to leakage^' etc., may be over-stated. With regard
to leakage, substantial improvements have been made in the pump systems.
Packaging methods are essentially the same as for aerosol containers and
thus are not a consideration as stated in a previous study.
The bladder system may feasibly function as a mechanical de-
livery system for hair spray. The manufacturer of Eco-Pure® suggests the
high pressure inner membrane may provide sufficient internal pressure to
propel a hair spray.^' The newly developed SELVAC system may also be a fea-
sible hair spray dispensing system.
b. Chemical and compressed gas propellants; Carbon dioxide
propelled hair spray is currently being test-marketed by Alberto Culver*^'
Much industry attention has been directed toward technical refinement of
valves, pressure drops, etc 111Q»H»33/ an(j j^as increased the technical fea-
sibility of CC>2 propellant usage in personal products.
Nitrous oxide is nonflammable, however, if can support com-
bustion;^' therefore it may not be suitable as a propellant in a product
directed toward the face.
The selected fluorocarbons may serve as excellent alternative
hair spray propellants. Combination of F-22 and hydrocarbons or other pres-
sure depressants, such as ethanol, can feasibly replace F-ll/F-12 blends
commonly used in hair sprays. F-142b, at up to 60% of the blend, or F-152a
could also be used in conjunction with F-21 (a major manufacturer recently
stated (February 1976) that toxicological data indicates F-21 probably is
not acceptable as an aerosol propellant), F-22, or possibly FC-318 in some
applications. The low vapor pressure uncommon propellants such as F-132a
could be used with F-22 as vapor depressants. There are many possible pro-
pellant blends achievable through combinations of chemicals included in the
"candidates suitable as blend constituents" with the selected fluorocarbon
propellants.
2. Antiperspirants and deodorants: Antiperspirants and deodor-
ants are available in mechanical delivery systems and in aerosol form (using
100% chlorofluorocarbon propellants). The technically feasible mechanical
delivery systems and alternative propellants are discussed below.
71
-------
a. Mechanical alternatives; Cream, stick, and roll-on de-
odorants and antiperspirants are a major personal product market. All three
delivery systems are suitable for both antiperspirants and deodorants. Con-
sumer sales in 1975 are far above those of 1974, and may exceed $25 million
for creams and $45 million for roll-ons.-^'
Mechanical spray pumps are technically feasible for deliver-
ing deodorants and antiperspirants and may prove to be aesthetically accept-
able to the consumer. Tussey Cosmetics, affiliate of Lehnx and Fink Product
Company, and Bristol Meyers ("Ban-Basic") currently offer a mechanical spray
pump delivery system.
A final technically suitable alternative mechanical delivery
system is the bladder system with the high pressure inner membrane. Both man-
ufacturers feel this system is suitable for deodorants»^Li2£.' Many antiper-
spirants are powders which require the propellant as a carrier JLSjJLI' Products
of this type would probably not be as suitable for use in the bladder system
since no propellant is used.
b. Chemical and compressed gas propellants; Carbon dioxide
is suitable only for delivering deodorant. Antiperspirants require the pro-
pellant as a carrier, and C(>2 cannot perform this function. *Pi H' However,
deodorants constitute a major portion of the approximately 620 aerosol de-
odorants and antiperspirants marketed in 1975 and C02 may be a technically
feasible, nontoxic, nonflammable, inexpensive propellant.
Deodorants and antiperspirants are generally propelled by
F-ll/F-12 with very small amounts of F-114. Combinations of ethanol/F-22,
F-22/F-142b (or F-152a), etc., may be a technically feasible propellants
for deodorants and antiperspirants. Again, as with the hair spray systems,
possible propellant combinations are numerous, and all have the characteris-
tics required for use in personal products, i.e., they are nontoxic and non-
flammable.
3. Shaving lathers; Shaving lathers are predominantly dispersed
by systems other than aerosols using chlorofluorocarbon propellants. This
section presents the viable mechanical alternatives and chemical propellants
for delivering shave lathers.
a. Mechanical alternatives; Shaving lathers were originally
available in collapsible tubes or as shaving lather bars/brush. In 1975, ap-
proximately 14 million* collapsible metal tubes were used for shaving lathers.
Data on the number of shave-lather bars sold in 1975 are not available.
* MRI estimate based on data in Reference 29.
72
-------
Barrier packs and bladder systems designed for viscous prod-
ucts may be suitable for delivering nonfoamed shave lathers. Development of
a foaming nozzle would increase the feasibility of the system for producing
a foam having the similar characteristics as aerosol delivered shaving lather
foams.
b. Chemical propellancs; Over 90% of the currently marketed
aerosol shave lathers are hydrocarbon propelled. The flammability of the .
hydrocarbons is suppressed to nonhazardous levels because the product is
water based. About 10% of the propellants used in shave lathers are chloro-
fluorocarbons, but these are always used in small amounts with the hydrocar-
bons «JLt2' Discontinuing use of these fluorocarbonsJLtiP.' in shave lathers
will probably not require replacement by alternative fluorocarbons although
these may technically function in this role. Therefore, since shave lathers
currently use an "alternative" propellant, conversion to 100% alternative
propellants should present no major obstacles for industry.
4. Perfumes and colognes; Perfumes and colognes can be delivered
by both mechanical and aerosol systems. The viable alternatives are discussed
below.
a. Mechanical alternatives; Perfumes and colognes are mar-
keted in several alternative mechanical systems. Included are glass bottles,
perfumed creams, solid, small roll-ons, atomizers, and mechanical spray
pumps. All of these systems provide excellent product delivery, and since
they require no propellant do not affect the odor in any way.
b. Chemical and compressed gas propellants; Carbon, dioxide
is odorless and nontoxic; data on the technical feasibility of delivering
colognes and perfumes were not available although use may be possible.
The selected alternative fluorocarbons have a low order of
toxicity and will provide suitable propellants for colognes and perfumes.
A wide range of desirable vapor pressures can be produced by combining the
various fluorocarbons such as ethanol/F-22, FC-318/F-22, F-142b/F-22, F-152a/
F-22, etc., still maintaining propellant nonflammability necessary for prod-
ucts which contain alcohol solvents.
5. Medicinal and, pharmaceutical products; Medicinal and pharma-
ceutical products are also available in aerosol and mechanical delivery sys-
tems. This section presents a discussion of currently marketed alternative
systems and other viable alternative delivery systems. It should be noted
that systems intended for use with medicinals and pharmaceuticals must come
under FDA sanction.
a. Mechanical alternatives; Collapsible tubes provide an
excellent delivery system for medicinal ointments. In 1973, 27.8% of 8.3
million gross collapsible tubes produced were used in packaging medinical
and pharmaceutical products•22.'
73
-------
Many medicinal products may also be dispensed by the barrier
pack and bladder systems. The barrier pack and bladder systems with low-
pressure inner liner were developed to accommodate viscous-type products.
Since there is no contact between product active ingredient and the propel-
lant, problems of chemical reactions with or combination of the product
should not occur.
Delivery of medicinals requiring application as a spray may
be possible with bladder systems equipped with high pressure inner liners
and with mechanical spray pumps. Products such as local antiseptics, burn
treatments, etc., can be effectively delivered by the 10 to 20 psi-iS' pres-
sures produced by bladder systems with high-pressure inner liners. Mechanical
sprayers already account for a major market, i.e., 18 to 30 million units in
1975 indicating satisfactory delivery and consumer acceptance.-•?•*138,407
b. Chemical and compressed gas propellants; Carbon dioxide
generally produces a coarse spray; however, valve refinements and redesign
are making possible finer sprays. As yet, ultra-fine sprays necessary for
antihistamines and asthma sprays are not yet achievable with C(>2 units.—
Thus, only medicinal sprays (e.g., antiseptics) which provide satisfactory
performance when applied in coarse sprays can at present use carbon dioxide
as a propellant.
The selected fluorocarbon propellents will probably replace
current use of F-ll and F-12 if these compounds are banned from use in aer-
osols. The selected fluorocarbons are nonflammable, inert (and thus should
not affect the product) and nontoxic. Through combinations and blends, a
wide range of vapor pressures are producible, and ultra-fine space-sprays
required for certain products can be achieved.
6. Miscellaneous small-volume uses; A summary of the percentage
of nonpersonal products propelled by fluorocarbons is presented in Table
IV-4. Fluorocarbon use in these products can be replaced by hydrocarbons,
compressed gases, or the selected fluorocarbons. Products not requiring
space-spray fineness can be delivered by bladder systems with a high-pressure
inner liner. The viscous food products such as cheeses can use the barrier
pack systems, bladder systems with low-pressure inner liners, or collapsible
tubes. These systems must have FDA approval. As discussed in Section B-4,
"Mechanical alternatives," household products use 40% of the mechanical spray-
ers, i.e., 72 to 120 million '."•"'tgi33.38,407 since fluorocarbons are not the
predominant propellant and many of the products are currently available in
mechanical alternatives, no major obstacles should be encountered if F-ll
and F-12 are banned for use in these aerosols.
Table IV-5 presents a summary of the viable alternatives discussed
for the five personal product types and for the smaller use volumes.
74
-------
TABLE IV-4
PROPELLANT USAGE IN NONPERSONAL PRODUCT AEROSOLS3?2?7'9/
Percentage Propelled By
Aerosol Product Fluorocarbons Hydrocarbons Compressed Gases
Household Products
Room Deodorants and 20 80
Disinfectants
Cleaners 10 90
Waxes and Polishes 30 70 < 1
Laundry Products 30 70 -
Others 10 90 - .
All Other Products
Coatings 5 95
Automotive 15 75 10
Insecticides 30 40 30
Industrial 50 50
Foods and Pan Sprays 25 5 70
Veterinary and Pet 50 50
Others '40 60
S Some percentages are MRI estimates based on References 2, 9, and
industry contacts.
75
-------
TABLE IV-5
SUMMARY AMD RELATIVE RANKING OF ALTERNATIVE DELIVERY SYSTEMS FOR AEROSOL PRODUCTS^
Mechanical
Alternatives
Ch
emical Propellants
Bladder System
Personal Products
Hair Sprays
Antipersplrants and
Deodorants
Medicinal and Pharma-
ceutical
Cologne and Perfume
Shax-e Lathers
Household Products
BDOO Deodorants
Cleaners
Laundry Products
Waxes and Polishes
All Other Products
Insecticides and Insect
Repel Ian ts
Coatings
Industrial
Foods and Pan Sprays
Automotive
Veterinary and Pet
Hand-
Applied*/
NA
a
a
a
a
NA
a
b
a
NA
a
-
-
a
a
Collapsible Roll-Ons
Tubes
NA
b
a
-
a
NA
a
-
c
NA
NA
-
a
.
a
and Sticks
NA
a
*
c
a
NA
a
MA
NA
-
•
b
NA
-
b
-
c
Barrier Low Pressure High Pressure
Packs I
NA
NA
b
-
b
MA
b
-
b
-
NA
-
b
-
b
nner Liner
NA
NA
b
-
b
NA
b
-
b
-
NA
-
b
- . '
b
Inner Liner
NA
b
b
b
.
NA
b
b
b
* • ' .
HA
b
b
b
b
Mechanical Carbon
Pumps Dioxide
a a
a"-' b£'
a . b
a b
NA HA
b b
a b
b b
a b
a a
MA c
b c
b a£'
b a
a b
Nitrogen
NA
NA
c
c
HA
MA
c
c
c
c
HA
c
c
c
c
Nitrous
Oxide Hydrocarbons
c
- • c
'a
c
NA a
MA a
a
a
c a
a ' " '
a ._.
a
a • a
a
- - a
Alternative
Fluorocarbons
b
b
b
b
c
b
e
, c
c
c
c
c
b
c
b
_*/ Code: * - currently employed; b - not presently used* but definitely feasible; c - technically feasible, however, formulation problems may exist; dashes indicate utility
and viability of the alternative is unknown; NA indicates the alternative is not applicable for the specified end-use product.
^b/ Hand-applied systems include creams, liquids in jars, and plastic bottles, etc.
£/ Applicable for foods but not for pan sprays-
-------
D. Production Data on Chemical Alternatives
This section presents production data (where available) and cur-
rent manufacturers of the candidates suitable as blend constituents and as
propellants as derived in Section B of this chapter.
1. Production estimates; The 1973 (or 1974 if available) produc-
tions of the alternative candidate propellants and the chemicals suitable
as constituents of propellant blends are summarized in Table IV-6.
In 1974, only very small amounts of F-22 of the alternative fluoro-
carbons were used as aerosol propellants. Only a very small amount of the
total F-22 production was used to propel aerosol products, i.e., the only
product being flat-tire inflation cannistersj:^/ FC-318 was not commercially
produced in 1974 in anything but experimental quantities!/ but stock, from
past production, was used to a minor extent (generally in combination with
nitrous oxide) in food products. The remaining fluorocarbons in Table IV-6
are not currently used to propel aerosol products.
1974 use of hydrocarbon propellants is estimated to be 116 mil-
lion pounds* divided between the three following chemicals: (a) propane -
19 million pounds; (b) isobutane - 97 million pounds; and (c) ji-butane -
<0.1 million pounds. The remaining hydrocarbons are not presently used, or
are used only to a very limited extent as aerosol propellants.
Fifteen percent (99 million pounds) of the total ethyl chloride
was used as solvents or exported; 32 million pounds of methyl chloride went
to solvents and exports; 47 million pounds of methyl chloroform were used
in aerosols, solvents for adhesives, etc.; and 47 million pounds methylene
chloride were used as solvents and vapor pressure depressants. The portion
of each figure attributable to usage in aerosols only is not available.-?-'
Production data for dimethyl ether and nitrous oxide were not
available in literature sources consulted. The percentages of total nitro-
gen and carbon dioxide production used as aerosol propellants were not avail-
able.
2. Manufacturers of chemical propellants: Numerous manufacturers
supply the chemical propellants (including the compressed gases). Many of
the selected fluorocarbons are not yet available in commercial quantities.
FC-318 is produced in 100 to 1,000 Ib lots or is drawn from previous produc-
tion supplies. F-132b, F-133a, F-218, F-227a, and F-3110 if available, are
only in experimental supplies. F-22, and to a lesser degree F-142b and F-152a,
are available in commercial quantities^!'
* MRI estimate based on Reference 3.
77
-------
TABLE IV-6
ANNUAL PRODUCTION OF ALTERNATIVE./CHEMICAL PROPELLANTS AND
BLEND CONSTITUENTS^^'43/
Chemical
Production ('millions of pounds)
1973 1974
F22
F21
Fl42b
Fl52a
FC-318
F132a
F133a
F218
F227a
F3110
Propane
n-Butane
Isobutane
Pentane
Isopentane
Propylene
Ethane
Dimethyl Ether
Ethyl chloride
Methyl chloride
Methyl chloroform
Methylene chloride
Nitrous oxide
Nitrogen
Carbon dioxide
136
141
Experimental
Quantities On4y
9,068
2,802
1,092
b/
b/
9,884
5,485
< 0.01
Experimental
Quantities Only
660
544
548
520
660
458
590
591
16,380
2,770
at Individual production estimates are not available. This production
figure is for other fluorocarbons which include F13, 14, 21, 115,
116, and others.
b/ 1973 Production of pentane, isopentane and all C_ hydrocarbon mix-
tures is 493 million pounds.
c/ Production data are not available at this time.
78
-------
A summary of the major manufacturers of chemical candidates suit-
able as blend constituents and candidates selected as alternative propellants
is presented, by chemical compound, in Table IV-7.
79
-------
TABLE IV -7
MANUFACTURERS OF PROPELLANT BLEND CONSTITUENTS AND ALTERNATIVE
Chemical
1. Fluorocarbons 21,22,
132b, 133a, 142b,
152a, 218, 227a,
3110
2. Ethane
3. n-Butane
oo
o
4. Isobutane
5. ti-Pentane
6. Isopentane
7. Propane
Producers
E. I. du Pont de Nemours and Company, Allied Chemical Corporation, Kaiser Aluminum
and Chemical Corporation, Racon, Inc., and Union Carbide Corporation
Allied Chemical Corporation, Atlantic Richfield Company, Dow Chemical Company,
Exxon Chemical Company, Monsanto Company, Olin Corporation, Amoco Production
Company, Phillips Petroleum Company,- Puerto Rico Olefins, Mobil Chemical Company,
Texaco, Inc., and National Distillers and Chemical Corporation, U.S. Industrial
Chemicals Company Division
Atlantic Richfield Company, B. F. Goodrich Chemical Company, Commonwealth Oil and
Refining Company, Inc., Coastal States Petrochemical Company, Olin Corporation,
Amoco Production Company, Phillips Petroleum Company, Shell Oil Company, Mobile
Chemical Company, Suntide Refining Company, Sun Oil Company, and U.S. Industrial
Chemicals Company • .
Atlantic Richfield Company, B. F. Goodrich Chemical Company, Coastal Sttftues Petrochemi-
cal Company, El Paso Products Company, Olin Corporation, Amoco Production Company,
Phillips Petroleum Company, Shell Oil Company, Sun Oil Company, and U.S. Industrial
Chemicals Company
Atlas Processing Company and Phillips Petroleum Company
Amoco Production Company and Phillips Petroleum Company
American Oil Company, Ashland Oil, Inc., Atlantic Richfield Company, Crown Central
Petroleum Corporation, Commonwealth Oil and Refining Company, Inc., Commonwealth
Petrochemicals, Inc., Cosden Oil and Chemical Corporation, Cities Service Oil
Company, Coastal States Petrochemical Company, Exxon Chemical Company, Champlin
Petroleum Company, Marathon Oil Company, Texas Refining Division, Olin Corporation,
Amoco Production Company, Phillips Petroleum Company, Puerto Rico Olefins, Mobil
Chemical Corporation, Suntide Refining Company, Charter International Oil Company,
Sun Oil Company, Texaco, Inc., Union Oil Company of California, and U.S. Industrial
Chemicals Company
-------
TABLE IV-7 (concluded)
Chemical
8. Propylene
9. Dimethyl ether
10. Ethyl chloride
11. Methyl chloride
12. Methyl chloroform
13. Methylene chloride
14. Nitrous oxide
15. Nitrogen
16. Carbon dioxide
Producers
Amoco Chemicals Corporation, Allied Chemical Corporation, American Oil Company,
Ashland Oil, Inc., B. F. Goodrich Chemical Company, Cities Service Company,
Petrochemical Division, Chemplex Company, Cities Service Oil Company, Dow
Chemical Company, E. I. du Font de Nemours and Company, Inc., Eastman Kodak
Company, Texas Eastman Company Division, El Paso Products Company, Gulf Oil
Chemicals Company, Jefferson Chemical Company, Inc., Marathon Oil Company,
Monsanto Company, Northern Petrochemical Company, Phillips Petroleum Company,
Shell Oil Company, Standard Oil Company-of Ohio,-Mobil Chemical Company, Charter
International Oil Company, Sun Oil Company, Texaco, Inc., and Union Carbide
Corporation
Chemical Solvents Corporation and Union Carbide Corporation
American Chemical, Dow Chemical Company, E. I. du Pont de Nemours and Company,
Inc., Ethyl Corporation, PPG, and Shell Oil Company
Allied Chemical Company, Conoco, Diamond Shamrock, Dow Chemical, Dow Corning,
E. I. du Pont de Nemours and Company, Inc., Ethyl Corporation, General Electric, •
Stauffer, Union Carbide Corporation, and Vulcan Materials
Dow Chemical Company, Ethyl Corporation, PPG, and Vulcan Materials
Allied Chemical Company, Diamond Shamrock, Dow Chemical Company, Dow Corning,
E. I. du Pont de Nemours and Company, Inc., Stauffer, and Vulcan Materials
Airco, Inc., Air Products and Chemicals, Inc., and Standard Air Company of New Jersey,
Inc.
50 Producers including Chemetron Corporation, Union Carbide, United States Steel
Corporation
63 Producers including Allied Chemical, Chemetron Corporation, E. I. du Pont de
Nemours and Company, Inc., Kaiser Aluminum and Chemical Corporation
-------
REFERENCES TO SECTION IV
1. Sanders, P. A., Principles of Aerosol Technology, Van Nostrand Reinhold
Company, New York (1970).
2. "Further Information on the Fluorocarbon Industry and on Potential
Economic Impacts of Restrictions of Fluorocarbon Production," Inter-
agency Task Force on Inadvertent Modification of the Stratosphere
(1975).
3. Johnson, M. A., W. B. Dorland, and E. K. Dorland, Aerosol Handbook,
McNair-Dorland Company, Inc., New York (1972).
4. "Aerosol Makers Brace for Ozone Ordeal." Chemical Week, pp. 14-15, June
11, 1975i
5. 1972 Du Pont Aerosol Market Report.
6* Remington's Pharmaceutical Sciences. 13th ed., Mack Publishing Company,
Easton, Pennsylvania (1965).
7. Personal communication with J. C. Bray, Jr. et al., Freon® Products
Division, E. I. du Pont de Nemours and Company, Inc., Wilmington,
Delaware.
8. Howard, D. K., P. R. Durkin, and A. Hanchett, "Environmental Hazard
Assessment of One and Two Carbon Fluorocarbons," Environmental Protec-
tion Agency, Contract No. 68-01-2202, September 1974; EPA Report No.
560/2-75-003; NTIS No. PB-247-419.
9. Arthur D. Little, Inc., "Preliminary Economic Impact of Possible Regu-
latory Action to Control Atmospheric Emissions of Selected Halocar-
bons," EPA Contract No. 68-02-1349, Task 8, Publication No. EPA-450/3-
75-073, September 1975; NTIS No. PB-247-115.
10. Confidential communication with a major industry source involved in
aerosol product development.
11. Personal communication with Mr. Charles S. Hayes, Applications Engineer,
Chemetron Corporation, Chicago, Illinois,
(,
12. Chemical Marketing Reporter, 208(9):9, September 1, 1975.
13. Du Pont Freon® Technical Bulletins B-2 and S-16.
82
-------
14. ASHRAE Guide and Data Handbook, American Society of Heating, Refrig-
eration, and Air-Conditioning Engineers, Inc., Joseph D. Pierce,
Chairman, New York (1972).
15. International Critical Tables of Numerical Data, Physics, Chemistry.
and Technology, 1st ed., Vol. Ill, McGraw-Hill Book Company, Inc.,
New York (1928).
16. The Chemical Rubber Company, Handbook of Chemistry and Physics, 45th
ed., Cleveland, Ohio (1964).
17. Kirk-Othmer Encyclopedia of Chemical Technology, 2nd ed., Vol. 5, Inter-
science Publishers, Division of John Wiley and Sons, Inc., New York
(1964).
18. Christensen, H. E., and T. L. Luginbyhl, eds., Suspected Carcinogens;
A Subfile of the NIOSH Toxic Substances List, U.S. Department of
Health, Education, and Welfare, June 1975.
19. Christensen, H. E., and T. L. Luginbyhl, ed.<=., The Toxic Substances
List. 1974 Edition. U.S. Department of Health, Education, and Welfare,
June 1974.
20. Hawley, G. G., The Condensed Chemical Dictionary, 8th ed., Van Nostrand
Reinhold Company, New York (1971).
21. Hoy, K. L., "New Values of the Solubility Parameters From Vapor Pres-
sure Data," J. Paint Techno1.. ££(541) (1970).
22. Documentation of the Threshold Limit Values for Substances in Workroom
Air, 3rd ed., American Conference of Governmental Industrial Hygienists
(1971).
»
23. Personal communication with Tabline Company, Berkeley, California.
24. Personal communication with Arthur D. Little, Inc.
25. Title 21, Code of Federal Regulations, 121.1065, 121.1181, and 121.101,
supplied by John E. Thomas, Assistant to the Director, Division of
Regulatory Guidance, Bureau of Foods, Food and Drug Administration.
26. "Pressure Packaging," Soao/Cosmetics/Chemical Specialties, p. 85,
September 1975.
27. "Carbon Dioxide as an Aerosol Propellant," Soap/Cosmctics/Chemicai
Specialties, p. 105, October 1975.
83
-------
28. "An Aerosol Filler's Experience with CO as a Propellent," Soap/Cos-
metics/Chemical Specialties, pp. 94-96, October 1975.
29. Modern Packaging Encyclopedia and Planning Guide. 47(12), December 1974.
30. Personal confidential communication with a major manufacturer of per-
sonal products.
31. "Room at the Top," Aerosol A^e. 'IjKD; 14-16. January 1974.
32. Diamond, G. B., "The Development of New Aerosols. The Impact of Regula-
tions." Aerosol Age. 19(2):22-23, February 1974.
33. Mr. Dan Massey, Marketing Manager, Valve Corporation of America, Baton
Rouge, Louisiana.
34. "Propellant-Less System Uses Hydraulic Pressure." Aerosol Age, 20J2):
25, February 1975.
35. "SELVAC is Back." Aerosol Age. 17(6):27-28, June 1972.
36. PersonaJ. communication with a major industry source that wishes to
remain anonymous.
37. "Tiny Pumps Thrive on Aerosol's Woes," Business Week, p. 28, August 25,
1975. /
S8. Mr. Frederick Muller, Bakan Plastics Division, Realex Corporation,
Kansas City, Missouri.
\
39. "Comeback for Mechanical Spray Cans?" Kansas City Star, p. 5D, August
24, 1975.
40. Mr. Donald Knox, Calmar Division, Diamond International Corporation,
New York, New York.
41. "Plastic Bottle Sprayers Take Hold." Soap/Cosmetics/Chemical Special-
ties, pp. 105-107, October 1975.
42. "Activites of Federal Agencies Concerning High Volume Chemicals," U.S.
Environmental Protection Agency, Office of Toxic Substances, EPA
560/4-75-001, Washington, D.C., February 1975.
43. "Synthetic Organic Chemicals, United States Production and Sales, 1973,"
United States International Trade Commission Publication 728 (1975).
44. "1974 Directory of Chemical Producers, United States of America," Chemi-
cal Information Services, Stanford Research Institute, California.
84
-------
V. FOAM BLOWING AGENTS
Fluorocarbons* are used in the manufacture of foams made from
polyurethane, polystryene, and polyolefin. These compounds are primarily
used as the'blowing agents which foxm the cellular structure of these
foams. Other type of plastic foams (primarily vinyl, urea, phenolic, and
epoxy foams) are produced in much smaller quantities, and do not typi-
cally involve much usage of fluorocarbon blowing agents.
In 1974, approximately 78 million pounds of fluorocarbons were
used in the production of plastic foams. This represents slightly more
than 8% of total U.S. fluorocarbons production for that year. Rigid and
flexible urethane foams together account for more than 90% of the fluoro-
carbon blowing agents in plastic foams. Rigid urethane foams consume
twice as much fluorocarbons as do the flexible foams. Rigid urethane
foams of low density, produced primarily for thermal insulation, use
more than half of all fluorocarbons used as plastic foam blowing agents.
A. Major Types of Plastic Foams
Each of the commercially important plastic foams (flexible ure-
thanes, rigid urethanes, polystyrene foams, and other plastic foams) have
differing technical requirements and economic constraints on the blowing
agents used. The resin systems, the production processes employed, and
the end-use applications of these plastic foams currently dictate the
choice of the type and quantity of blowing agents employed. To provide
overall perspective before discussing the requirements for individual
types and classes of foamed plastics, Table V-l presents a summary of
major types of plastic foams by process and application areas.
Each of the major foams systems currently employing fluoro-
carbon blowing agents will be discussed separately. A detailed treat-
ment of the technology and chemistry involved in plastic foam produc-
tion is beyond the scope of the present study; and has been addressed
in recent surveys and reviews.
Hence, the following discussion will emphasize those factors
directly affecting the use of blowing agents, and the range of alterna-
tives to be considered should it prove desirable to reduce fluorocarbon
emissions from plastic foam manufacture.
The term fluorocarbon is used in this section tb mean chlorofluoro-
carbon.
85
-------
CD
TABLE V-l
TYPES AND APPLICATIONS OF PLASTIC FOAMS
Tvoe of Plastic Foam
Polyurethanes
Foam Production Process
Location of Foaming Blowing Agents
Operation Used
Low density rigid *
polyurethane (1.6 to
3.0 pcf)
Medium density rigid
polyurethane
Polyurethane structural
foams (8.0 to 30.0 pcf)
- Pour (bun or continuous slab) Plant
F-ll, CO , F-113
Cavity filling (pour or froth) Plant or job site F-ll, CO , F-12, F-114
Spray on (spray or froth spray) Job site F-ll F-.12, CO , F-114
- Pour, mold, cast
Flexible polyurethane foams -
Polystyrene
Extruded film and sheet
Expanded board stock
Extrusion from crystal
Steam expansion to logs
Expanded shapes
- Bead board molding
Polvolefin
Medium density polyethylene - Extrusion
Miscellaneous
Phenolic foams
Epoxy foams
Urea foams
PVC plastesol foams,
polyisocyanurate foams,
polyester foams
- Pour or slab
- Pour, mold or syntactic
- Foam gun
- Various processes
Plant
Mold, cast, reaction injection Plant
molding
Pour (bun or continuous slab) Plant
Mold
Plant
Plant
Plant
Plant
F-ll, CO , F-114
CO., F-ll, others
C02, F-ll,
Pentane, F-12, others
Isopentane, pentane,
CH^Cl, hexane, F-12,
. others
Pentane, F-12, others
F-114, F-12, F-ll,
others
Plant or job site CO., F-ll, pentane
Plant
F-ll, others
Plant or job site Surfactant, CO
Plant F-ll, others, organics
Typical
Product .
Applications
Building panels, board stock,
packaging, foam logs, pipe
Insulation
Refrigerators, boats, building
walls
Industrial tanks, pipelines,
roofing
Pattern stock, decorative
shapes
Furniture parts, industrial
automotive parts
Auto seats, bedding, rug pad-
ding
Meat trays,' egg cartons, pack-
aging disposables
Building insulation, flotation,
packaging
Cups, packaging, dunnage, molded
board, foam parts
Flotation, crash padding,
packaging
Building insulation, pipe in-
sulation
Specialty foams
Mobile homes, building walls.
Specialty foams
-------
B. Polyurethane Foams - Rigid and Flexible
1. Structure of the urethane foam industry; The structure
of the industries that collectively are called the "plastic foam indus-
try" varies substantially according to the resin involved. The complex-
ity of industry structure increases markedly as the number of end uses
and raw materials increases. The urethane foam industry is unusually
complex and difficult to characterize because the urethane polymer does
not come into existence until the foam is created, usually at the fabri-
cation stage. The versatility of these foams has created numerous prod-
uct forms, application methods, raw material types, competitive situa-
tions— and, therefore, corporate involvements. Three distinct segments
make up the urethane foam industry: (a) the suppliers of basic raw ma-
terials and foam systems; (b) the fabricators and suppliers of flexible
foam; and (c) the producers and fabricators of rigid foam.
a. Materials and systems! The major chemical producers
and the chemical subsidiaries of other industries dominate the raw ma-
terials supply picture. Many of these raw materials suppliers, such as
Olin and Goodyear, are forward-integrated into liquid urethane systems,
broadly defined as packages of reactive components for ultimate conver-
sion into foam. A number of intermediate-size chemical firms—Stepan,
Reichhold, and Witco—supply urethane systems. The chemical divisions
of other firms are also major suppliers of urethane systems. It should
be emphasized that systems are most important to the rigid urethane foam
industry, while the flexible foam fabricators generally purchase raw ma-
terials and formulate their own foams.
The suppliers of raw materials and systems are normally
not engaged in fabrication, nor vice versa. Major exceptions to this
rule include Upjohn and PPG Industries, both of whom supply systems
and also engage in fabrication. Upjohn typifies those firms that are
extensively vertically integrated. Figure V-l portrays schematically
the structure of the raw materials portions of the urethane industry.
b. Flexible urethane foam; Major production of flexible
foams passes through a bun and slab stock fabrication stage. Almost all
flexible foam producers rely on the basic raw materials. About half of
the dashboard padding and cushioning used in the automotive industry are
captively produced, employing urethane systems. The remainder is pur-
chased principally from the tire and rubber companies who are large
enough to produce their own flexible foam systems. The structure of
the flexible foam producing industry with estimated material transfers
is shown in Figure V-2.
87
-------
Isocyanates
Mobay
Allied Chem.
Union Carbide
Wyandotte
DuPont
Upjohn
Kaiser Chem.
384
Polyols
Dow
Jefferson
Union Carbide
Wyandotte
Witco
Olin
Atlas Chem.
' ' <
348
i '
Liquid Urethane Systems
Chemical Companies Other
Wyandotte Upjohn
Union Carbide PPG Ind.
Reichhold Cook Paint &
Stepan Chem. Inmont
Witco M-R Plastics
Olin Pelron Corp. ,
348
i ' •
FLEXIBLE & RIGID
620
Blowing Agents, *
Catalysts & Additives
Allied Chem.
Union Carbide
Air Products
DuPont
Wyandotte
Abbott
Jefferson
Stauffer, etc.
i
Yarn.
etc.
103*
f
* (Some of the blowing agents
(such as Freons, etc.) are
volatilized and lost from
final poundage. Figure of
103 million pounds used
for materials balance.)
759 •
URETHANE FOAM
(see Figure V-2) 1107
Figure V-l - Urethane Foam: Raw Materials Supply Structure
1972 Estimated Materials Flows, Million Pounds
Source: MR1 1974 Survey.
88
-------
Raw Materials 673 Liquid Urethane Systems 131
i
\
330
i
330
Furniture & Bedding
Kroehler,
Mohasco ,
Sealy, American
Seating, etc.
673 35
i i '
Bun & Slab Stock 708
Major Tire & Rubber Firms,
Firestone, Upjohn (CPR Div.),
E.R. Carpenter, Tenneco
(General Foam Div.), etc.
"
228
\ -
228
Apparel, Carpeting,
Packaging, etc.
Reeves Bros., Mohasco,
Scott Paper, E.R. Carpenter,
etc.
131
i
,
1
150
i i
96
i
260
Transportation
General Motors,
Ford
, Chrysler,
White, Boeing, etc.
Figure V-2 - Flexible Urethane Foam Industry Structure
1972 Estimated Materials Flows, Million Pounds
Source: MRI 1974 Survey.
89
-------
c. Rigid urethane foam; Fabrication originates from
both raw materials producers and suppliers of liquid systems. This is
because rigid foam is more diverse in application than flexible foam
and there are literally hundreds of applicators and molders, each in-
dividually small, who find it more economical to purchase their mate-
rials in pre-engineered, prepackaged systems. At the same time rigid
foam is growing rapidly in building insulation and structural foam
parts for furniture. Consequently, many major producers of rigid foam
board stock are the building materials suppliers. Because rigid ure-
thane foam competes directly in insulation markets formerly dominated
by rock wool and fiberglass, firms such as Owens-Corning Fiberglas are
now among rigid foam producers. Most of the large refrigerator and ap-
pliance producers buy raw materials, while smaller appliance firms pur-
chase systems. The transport carrier industry, makers of railroad
freight cars and insulated truck trailers, rely totally on the purchase
of foam systems. Furniture manufacturers purchase most foam structural
parts from custom plastic processors, even though larger manufacturers
have established their own foam production lines. Details of the rigid
urethane foam industry structure are shown in Figure V-3.
2. Rigid urethane foams
a. Consumption of rigid polyurethane foams; Consumption
of rigid polyurethane foams for all applications has been growing at a
substantial ra'te over the past 10 years and is anticipated to show fur-
ther growth through 1980 dt nearly 10%/year.
By far the largest application of the rigid urethane
foams is for thermal insulation. The most important application areas
are in building construction, refrigerators, freezers, transportation
carriers (truck trailers and freight cars) and industrial applications
such as the insulation of tanks and pipelines. For insulation uses,
urethane foams are typically produced at densities of 1.6 to 2.2 Ib/ft
(pcf); and virtually all of the insulating foams employ fluorocarbon
blowing agents.
Other uses of rigid urethane foams including packaging
and dunnage, boats and flotation, and molded structural parts, may or
may not employ fluorocarbon blowing agents. In particular, the medium
to high density (8 to 26 Ib/ft ) rigid urethane moldings have grown
rapidly in application between 1967 and the present. These molded parts
replace wood in furniture, television cabinets, decorative interior pro-
files, and a variety of industrial parts. In the higher density ranges
little or no fluorocarbon blowing agents are employed.
90
-------
Raw Materials 105
75
l '
75
Board Stock
Upjohn
Celotex
Owens Corning
National Gypsum
1
75 39
\ ' '
114
Building
Construction
Prefabricators
Bldg Contractors
30
' \
35
65
Refrigeration
GE, Whirlpool,
Frigidaire,
Westinghouse,
Philco
Liquid Urethane Systems 198
Appli
i
50
cators
50
Upjohn (CPR Div.)
Armstrong Cork
PPG,
71
21
21
Flotation
Boats
Logs
i
etc.
15
< \
11
26
Packaging,
Industrial,
Other
Various
\
24
24
Molders
Numerous
Plastic
Fabricators
16
I \
24
40
Furniture
Case Goods
Cabinets
Television
37
37
Transportation
ACF,
Freuhauf,
Brown,
GATX
Figure V-3 - Rigid Urethane Foam Industry Structure
1972 Estimated Materials Flows, Million Pounds
Source: MRI 1974 Survey.
91
-------
Table V-2 summarizes the statistics for rigid urethane
consumption since 1965, with current estimates for 1980 consumption lev-
els. Annual growth rates for the period 1973 through 1980 are also in-
dicated. Total quantities of rigid urethane foams consumed are in rea-
sonable agreement with vthe estimates of production reported by various
trade sources. The breakdown of consumption by individual end applica-
tions is however subject to considerable variation depending upon the
format used in reporting, and the variability of different trade reports.
This table also includes an estimate of the quantity of rigid urethane
foam blown with fluorocarbon agents. The basis of this estimate is the
sum of thermal insulation plus a quantity of foam equivalent to that
consumed in packaging and flotation uses. The industrial applications
of rigid urethane foams are currently expanding at a rapid pace as a
direct consequence of energy conservation programs. Trade sources pre-
dict that consumption for tank insulation, pipe lagging and insulated
petroleum pipelines will exceed 30 million pounds by 1976, roughly
double the consumption for 1974.
b. Rigid urethane foam production processes; There has
been substantial technical change and progress in rigid foam making since
the urethanes were first introduced from Germany at the close of World War
II. The initial resin systems based on castor oil have been replaced by
polyether and polyester types of urethane systems. Polyether foams such
as those using sucrose-based polyols have been found to give properties
very similar to the polyester foams but at a lower cost.—- Systems based
upon halogenated polyesters or phosphorous-containing polyols are widely
used where maximum fire resistance is important. However, the polyether
foams can also be rendered flame resistant by the incorporation of a suit-
able flame retardant. Blowing agent use with all types of resin systems is
essentially similar. However, all foam formulations are sensitive to the
choice and quantity of auxiliary blowing agent used. The use of different
fluorocarbons or other volatile solvents will require adjustments in the
silicone surfactant used to control cell size, arid the water content of the
resin system.
Rigid urethane foams are prepared by mixing two or more reac-
tive liquid ingredients together. The chemical reaction which results forms
a polymeric material which entraps gas bubbles to provide the desired cellular
structure. The ga's required for foaming is derived from carbon dioxide re-
leased during the polymerization reaction, or from the volatilization of an
added blowing agent vaporized by the heat of reaction.
92
-------
TABLE V-2
RIGID URETHANE FOAM CONSUMPTION
vO
Thermal Insulation
Building Construction
Refrigeration Appliances
Transport Carriers
Industrial (tank and pipe)
Total Insulation
Other Applications
Furniture
Packaging
Marine and Flotation
Miscellaneous
Total Other
Total Rigid Urethane Foam
Quantity Reported by Trade
Fluorocarbon Blown Urethane
Foam-
Percent Fluorocarbon Blown
(Million Pounds)
1965
17.0
19.0
14.8
3.0
53.8
1.0
4.6
7.0
12.6
66.4
71.0
1966
24.5
24.1
27.5
5.0
81.1
4.0
7.6
13.0
24.6
105.7
106.0
1967
31.7
28.2
26.1
7.0
93.0
4.0
5.0
9.9
16.0
34.9
127.9
134.0
1968
47.5
36.6
25.1
8.0
117.2
11.1
6.0
11.9
9.0
38.0
155.2
155.0
1969
58.0
44.9
34.7
12.0
149.6
20.5
8.0
14.4
15.0
57.9
207.5
209.0
1970
71.2
51.2
29.8
12.1
164.3
25.3
7.2
15. 9S/
26. 2k/
74.6
238.9
241-0
1971
83.3
57.8
29.1
10.5
180.7
31.3
6.0
11. Zi/
7.2k/
55.8
236.5
236.6
1972
113.7
63.4
37.2
10.0
224.3
37.0
7.0
14. 4* /
7.0
65.4
289.7
285.9
1973
150.1
72.3
51.3
11.0
284.7
46.3
9.1
15. 42/
6.9
77.7
362.4
349.4
1974
154.2
74.8
49.6
15.8
294.4
42.9
12.1
13. 8»/
8.1
76.9
3-71.3
369.6
175.0
72.6
48.4
15.4
311.4
35.2
13.2 .
13.2*'
8.8
70.4
382.8
399.0
Growth Rate
1980(est) 1980/1973(Z)
275.0 9.0
115.0 6.9
70.0 4.3
33.0 17.0
493.0
115.0
30.0
2B.C2'
17.0
190.0
683.0
827.0
8.2
13.9
18.5
8.9
13.7
13.6
9.5
13.1
59.4 92.7 107.9 135.1 172.0 187.4 197.9 245.7 309.2 320.3 337.8
89.5 87,7 84.4 87.0 82.9 78.4 83.7 84.8 85.3 86.2 88.2
551.0
80.5
8.6
Source: MRI 1974 Survey updated 6-15-75.
Notes: a/ Due to changes in reporting format, the scope of Industrial, miscellaneous and nonboat flotation varies; in some years foam logs were
Included with board stock; mine sealing foam not reported after 1972.
t>/ Uses included in miscellaneous applications was reallocated for years subsequent to 1970.
£/ Fluorocarbon blown foam estimated to Include all Insulation applications plus the equivalent of flotation and packaging.
-------
The liquid ingredients are mixed in a foam machine consist-
ing of a pumping unit capable of accurately metering the components through
a continuous mixer which blends the components uniformly. Preparation of
the urethane system components, metering and complete mixing are all criti-
cal to the proper manufacture of polyurethane foams. Machinery for the pro-
portioning and dispensing of urethane foams is highly developed, with equip-
ment being offered by some 48 manufacturers. The basic forms and variations
among commerically available urethane foaming machinery has been reviewed
recently by Johnson,— and will not be considered in detail. After the mix-
ture is dispensed from the metering and mixing head, the method of applica-
tion, and the means used to contain the foaming mixture, differ widely ac-
cording to the application. These factors in turn influence the choice of
blowing agent and the formulation of the urethane system.
The four basic dispensing methods for rigid urethane foams
are pour, froth, spray^ and injection. Molding and casting are both gen-
erally considered as a form of foam pouring. The frothing process can be
applied either to foam pouring, or froth spraying. Thus the basic dispens-
ing methods can be divided into two broad classes—spraying and pouring.
Rigid urethane foams can be manufactured by batch processes, by a continu-
ous slab or "bun" process, by foaming or frothing in place, and by spraying.
In the U.S., the continuous slab production and foaming in
place are the most common. Continuous production of rigid foam board stock
(commonly used for building insulation) involves pouring the liquid mixture
as a thin layer onto a continuously moving conveyor where it expands to
form a continuous block of foam. After oven curing, the foam may be sliced
to specific thicknesses by a variety of sawing methods. An alternative of
increasing importance involves the continuous application of surface skins
(paper, plastic films, or metal) with the panel expanding to standard
thickness.
Foaming in place involves pouring the liquid mixture of
foam components from the dispensing head directly into a cavity (such as
the walls of a refrigerator or boat hull). The chemical reaction causes
the blowing agent to expand, and the foam completely fills the cavity.
Considerably accuracy and uniformity is required in dispensing the ure-
thane into cavities and molds. Each volume of the liquid urethane expands
30 times in volume and therefore must be accurately measured and precisely
placed within the mold or-cavity in order to insure accurate filling, and
avoid excessive pressure on the sides of the mold or product.
94
-------
The froth process introduced in the early 1960's is being
used to an increasing degree to minimize these problems. The frothing pro-
cess involves the addition of another blowing agent, usually f luorocarbon-
12 (alternatively, F-114 or F-115 may be used) injected directly into the
mixing head. The mixing head operates at a pressure of about 100 psi to
keep the frothing agent liquid until it emerges from the dispensing noz-
zle into the cavity or mold. The reaction and further expansion of the
foam roughly doubles the froth volume, thus filling the cavity or mold
without excessive pressures, and permitting more accurate filling of
cavities.
Sprayed polyurethane foams are often used for on-site ap-
plication of rigid thermal insulating foams. The liquid urethane ingredi-
ents emerge from the mixing head and are sprayed onto the desired surface
to expand. As in pouring operations, froth spraying has become a conve-
nient method of achieving the desired urethane thickness and maintaining
control over the desired density and skin thickness.
The frothing pour process has become increasingly important
for cavity and mold- filling applications. Refrigerators and freezers rep-
resent one of the most important applications of rigid froth pour- in-p lace
foams. Thin-wall refrigerators can use approximately one-half the usual in-
sulation thickness, thus resulting in up to 35% greater interior capacity.
The average foam usage is 8 to 10 Ib/unit. Froth foam is placed in the cav-
ity between the refrigerator liner and the outer steel shell to bond the
two surfaces together and completely fill the cavity.
When nonfrothed resins are used to generate foams of 2
Ib/ft density, complex jigs and supports must be used to contain the
molds or sheets and prevent distortion during foaming. Pressures devel-
oped by foaming in cavities 1 to 2 in. thick may reach 2.5 to 3.8 psi
on the walls of the mold, because of the high expansion ratio, and be-
cause about 57» excess urethane must be added to insure a complete foam
'
When froth poured foam is used, the foam expansion exerts
only about 0.4 psi on the surrounding walls. Simple, flat walls on build-
ing panels or mobile homes and trailers require only minimum supports and
jigs — usually plywood support is adequate. Refrigerator doors containing
shelves, separators, and egg trays may require more complex jigs and fix-
tures. The foams have sufficient flowability to fill narrow and complex
mold cavities such as thin shelves and egg frame protrusions of a verti-
cally filled refrigerator door, but small holes must be positioned to al-
low the escape of air during the foaming process. These holes seal them-
selves as the foam gels.
95
-------
c« flowing agents in rigid urethane fornrulations; There
are two general types of blowing agents used in the manufacture of ure-
thane foams.
* Carbon dioxide generated in situ by the reaction of
isocyanate with water.
* Auxiliary blowing agents—volatile liquids such as
fluorocarbons, particularly trichlorofluoromethane,
vaporized by the exothermic reaction between the iso-
cyanate and hydroxyl components.
High-density foams are most frequently made using carbon
dioxide (derived from organic additives) as the blowing agent, while low-
density foams providing the best insulation properties are generally pre-
pared using fluorocarbons for this purpose. For frothed foams, fluorocarbon-
12 is used in addition to fluorocarbon-11.
Selection of the blowing agents in the foam formulation,
and control over its vaporization directly affect foam density and thermal
insulation effectiveness. Density influences most, foam properties (strength,
thermal, acoustical, electrical, and other properties). Therefore choice of
the amount and type of blowing agent is one of the most important controls
for foam properties. The effect of water and fluprqcarbon-11 concentration
' 107
on foam density is shown' in Figures V-4 and V-5.—
The most significant difference resulting from the selection
of either carbon dioxide or fluorocarbon as blowing agent lies in the ther-
mal insulation effectiveness. If the cell size and shape, and the percentage
of closed cells are the same, the K factor (initial) of fluorocarbon-11-
blown foams is 0.10 to 0.15, while that of CO^-blown foams is 0.22 and 0.24.
This difference in K factor is quite evident when comparing the thermal
conductivities of cut, rigid polyester urethane foam specimens.— The lower
K factor for fluorocarbon-blown foams means that the amount of insulating
material needed for a given application may be reduced by as much as 50%.
The nature of the blowing agent also influences electrical
properties. When water is used to generate carbon dioxide, water vapor may
become entrapped within the cells and subsequently absorbed by the polymer
increasing its dielectric constant and dissipation factor. Water in the foam,
from any source increases the absorption of RF radiation at radar frequencies.
In reproducing rigid foams for radomes and microwave applications, special
precautions are taken to minimize the presence of water in the foam.
96
-------
(90% Confidence
Limits Shown)
0.02
0.01 La-u
0.6 0.8 1.0
2 4 6 8 10 20 40 60
DENSITY, lb$./cu. II.
Figure V-4 - Effect of Water Concentration on Foam Density
97
-------
(90% Confidence '
Limits Shown)
0.6 0.8 1
2 4 6 8 10
DENSITY, lbs./cu. It.
Figure V-5 - Effect of Fluorocarboh-11 Concentration on Foam Density
98
-------
Thus, combinations of water and fluorocarbon blowing agents
are used in different types of rigid urethane foam depending upon the ap-
plication. Table V-3 gives typical starting formulations for rigid urethane
foam systems intended for different applications. The total fluorocarbon
content of these urethane systems ranges from less than 1% up to 17% by
weight. Low-density foams for thermal insulation represent 80% of total
output. The fluorocarbon content of systems for thermal insulation foams
more typically is between 12 and 16% by weight.
d. Effect of blowing agents pn thermal properties of foam;
More than 80% of rigid urethane foams are produced for thermal insulating
applications. Achieving and maintaining the lowest practical thermal conduc-
tivity (K value) is of critical importance since the quantity of insulating
foam required increases.directly with thermal conductivity.
In low density plastic foams, the thermal conductivity of
the gas within the cells is, of course, the controlling factor determining
the conductivity of the foam. The cell gases make up about 97% of the total
volume of the foam, while the resin network makes up only 3% of the foam.
The thermal conductivity of the gas depends upon the gas composition, and
the gas composition is often hard to define. Carbon dioxide blown foams
initially contain nearly pure C02 within the cells. But C02 diffuses readily
through urethane, and much or all of the initial GC>2 in the foam cells will
eventually be replaced by air.
The low thermal conductivity of halocarbon vapors, coupled
with relatively slow diffusion through urethane resin is largely responsible
for the nearly universal use of fluorocarbon blowing agents in preparing
urethane insulating foams. The relationships between foam density, cell size,
composition of the vapor in the cells, structural variables, and aging ef-
fects on thermal properties has been intensively studied for more than 20
years.
In order to understand the effect of blowing agents on ther-
mal conductivity of foams, it is useful to separate the components of heat
transfer through a foam structure. While more elaborate and complex expres-
sions have been developed,-- thermal conductivity may be considered as:
K = k; + k- + k;. + k- (i)
99
-------
TABLE V-3
TYPICAL RIGID URETHANE FOAM SYSTEMS
Type of Rigid Foam
Ingredient
(Parts'by wt)
Polyol
Polyester
Flane retardant
Isocyanate
^ Surfactant
O Catalyst
° Catalyst diluent
Water
Fluorocarbon F-ll
Other blowing agent
Total
% Fluorocarbon by
Continuous
Slab Stock
Polyether
100.0
-
-
46.0
1.0
0.2
-
3.6
16.0
.
166.8
weight 9.6*
2.0 pcf Board
Flame Resistant
Polyether
70.0
-
30.0
108.0
1.0
1.5 .
1.5
-
30.0
3.0(F-113)
245.0
13.5%
1.4 pcf Insulation
Flame Resistant
Polyether
70.0
-
30.0
108.0
1.0
1.5
1.5
-
40.0
'-
252.0
15.97.
Halogenated
Polyester
Flame Resistant
_
100.0 (Br)
-
92.0
2.0
0.5
-
-
35.0
1
229.5
15.3%
1.6 pcf Frothed
Polyether
100.0
-
-
78
1.2
0.65
-
-
25.0
12.0 (F-12).
217.9
17.0%
12.0 pcf Structural
Foam Molding
115.5
-
-
132.0
0.02
15.0
- -
-
2.0
-
264^5
0.76%
-------
where K = the thermal conductivity of the foam in Btu/(hr)(ft2)(°F/in)
k' = conductivity through the polymer network of the foam
k' = thermal conductivity of the gas within the cells
O
k' = the component due to thermal radiation through foam
k' = the component due to connection within the foam
As a good first approximation^— for low density foams,
heat conduction through the solid and gas phases can be expressed as the
product of the thermal conductivity of each phase times its volume frac-
tion. Solid urethane polymers exhibit K values ranging from 0.7 to 2.0;
with a typical value of 0.9 to 1.0. Thus, the conductivity through a foam
containing 3% urethane would be about:
k^. = 0.03 x 0.9 = 0.027 K units
Similarly, the conductivity through the cell gases would
be: k' = 0.97 x K value of the gas. Table V-4 gives thermal conductivity
values for some of the gases found in urethane foams. It can be seen that
the values for halbcarbon vapors are considerably lower than those of air,
nitrogen or carbon dioxide.
Foam producers wish to use blowing agents that measurably
reduce the effective K value of the insulating foam. However, the blow-
ing agent gases within foam cells are not normally pure halocarbon vapors,
and the composition of these gases changes over time altering the thermal
conductivity of the foams. The thermal conductivity of gaseous mixtures
can be approximated as:
where k = the conductivity of the mixture
m
N. and N = the mblfractions of gases 1 and 2
k. and k = the respective conductivities of the pure gases
101
-------
TABLE V-4
THERMAL CONDUCTIVITY OF BLOWING AGENT
f\
Thermal Conductivity, kg (Btu/(hr)(£t )(°F/in))
Gas or Vapor 32°F 77°F
Carbon Tetrachloride 0.0384 0.0450
Chloroform 0.0456 0.0516
Methylene Chloride 0.0468 0.0540
F-ll 0.0468 0.0575
F-12 0.0576 0.0660
F-21 - 0.068 (at 86°F)
Benzene 0.0624
Methyl Chloride 0.0636 0.0738
Acetone 0.0684 -
F-114 - 0.0770 (at 86°F)
Methyl Acetate 0.0708
F-22 - 0.0806 (at 86°F)
Iso-Pentane 0.0864 0.0996
Carbon Dioxide 0.1020 . 0.1156
Nitrogen 0.168
Oxygen 0.170
Air 0.168 0.181
102
-------
When cured foams are aged exposed to air, the K factor
is degraded by migration of gases such as 02, N£, 1^0 (vapor) and possibly
some C02 into the foam.* Unless the foams are protected by surface skins
or other barriers, this dilution of the fluorcarbon vapors within the
cells will eventually result in poorer insulating value. Producers of
foams are attempting to develop foams which retain their initial low K
factor. A comparison of conventional foams and experimental foams, and
self-skinning sprayed foams on aging in air is shown in Table V-5.—
If the foam is contained within a sealed cavity (refrigerator cabinet,
or building panel) or if the thick polymeric skin is left on the foam
surface (as usually happens with sprayed foams), there will be rela-
tively little change in composition of the gases within the foam. Cut
and oven-aged board stock, on the other hand, undergoes a significant
change in thermal properties as a result of diffusion.
The effect of using different blowing agents on the ther-
mal conductivity or urethane foams can be estimated with reasonable ac-
curacy. Experimental values generally agree within 5 to 10% of K values
calculated by various mathematical approximations.— It is more diffi-
cult to predict the extent and rate of gaseous diffusion and change in
K value that will occur during aging if nonfluorocarbon blowing agents
are used. The present state of knowledge about the diffusion of organic
vapors through foam structures is not adequate^' to anticipate how such
materials would behave over time.
* Note: There are two reasons that atmospheric gases initially diffuse
into the foam, diluting the fluorocarbon vapors rather than
diffusion of the FC out of the foam cells.
1. The diffusion rate constants for fluorocarbon vapors through
urethanes are much lower than those of oxygen or nitrogen,
and nearly 100-fold lower than that for C02« The rate of
loss of the large, heavy fluorocarbon molecules by diffu-
sion from foams is quite low.
2. Gaseous pressure within the cells of rigid foams after cool-
ing is considerably below atmospheric pressure. Unless
the compressive strength of the foam is adequate, chilling
the foam may result in it being crushed by air pressure.
2
Diffusion Constant (cm /gee)
Oxygen 1.12 x 10~
Nitrogen 6.27 x 10"
F-ll 2.25 x 10
103
-------
TABLE V-
CHANGE IN K-FACTOR OF FOAMS ON AGING
(Cut surfaces exposed to air at 68 F;
K in Btu/(hr)(ft2)(°F/in)
K Aged K Aged
Foam Type ' K Initial 10 Days 120 Days
Conventional Foams
Standard Commercial 0.120 0.150 0.165
F-ll Blown
Commercial Foam 0.120 0.140 0.158
Supplier A
Commercial Foam 0.120 0.154 0.159
Supplier C
Experimental Foams
Experimental J 0.115 0.138 0.153
Experimental N 0.100 0.129 0.139
Experimental 0 0.101 ' 0.120 0.138
Experimental P 0.105 0.123 0.133
Foams Sprayed On
Wood at 2.01 lb/ft3 0.105 0.140
(Core 1.86 lb/ft3)
Steel at 2.07 lb/ft
(Core 1.84 lb/ft3)
Aluminum at 2.06 11
(Core 1.92 lb/ft3)
Steel at 2.07 lb/ft3 0.103 0.131
Aluminum at 2.06 lb/ft3 0.104 0.137
104
-------
Despite these limitations, the effect of various blowing
agents on thermal conductivity of urethane foams has been estimated for
a few simple systems in order to show the magnitude of differences in
thermal properties. Table V-6 compares the K values computed from Eqs.
(I) and (2), using data from Table V-4 for vapor conductivities of blow-
ing agents. Calculated values are generally in agreement with reported
thermal conductivities of freshly prepared or 1-day aged foams. The re-
sults of cellular gas analysis studies on rigid urethane bun stock are
shown in Table V-7.
It can be seen that air, carbon dioxide and most hydro-
carbons are not capable of providing the desired low K values in ure-
thane foams. For insulating use within the temperature range of 0 to
150°F, some halogenated hydrocarbons are theoretically able to contri-
bute low thermal conductivity to foams. However, any such alternative
blowing agent would also have to satisfy other important criteria—such
as compatability with the formulation, effect on foam formation, effect
on foam properties and aging, acceptable flammability, toxicity, etc.
At lower temperatures (below about -10°F) it has been re-
ported that F-12 is markedly superior to the other fluorocarbons, and
may offer lower thermal conductivity than any other blowing agents.—
3. Flexible urethane foams
a. Consumption of flexible urethane foams; Flexible
urethane foams are produced primarily for cushioning, textiles and car-
pet underlay applications. Table V-8 shows recent trends in consumption
of flexible urethane foams for each of the major end uses. Because the
largest uses depend upon automotive sales and home furnishings, growth
in flexible foams is more sensitive to the state of the U.S. economy
than is the case with rigid foams.
b. Production technology for flexible urethane foams;
The technology and chemistry involved in producing flexible urethanes
is quite similar to that for rigid foams. Despite these similarities,
the manufacture of flexible.foams has become a highly specialized seg-
ment of the industry, usually separate from the production of rigid
urethane foams. The principal differences in production and marketing
which characterize flexible urethanes versus rigid foams include:
105
-------
TABLE V-6
EFFECT OF BLOWING AGENTS ON THERMAL CONDUCTIVITY
Computed and Experimental K-Values for 2.0 Ib/ft Urethane Foams
* At 2.0 Ib/ft foam is 3% polymer; 97% gas by volume
* Radiation component k1 = 0.02; convection fc1 = 0
* Solid urethane polymer k = 0.924; k1 = 0.03 x 0.924 or 0.0277
Blowing Agent Vapor
or Gas in Cells
A. Pure Gases in Cells
Air
co2
Iso-Pentane
B. Pure Halocarbons in Cells
F-ll
F-12
Methylene Chloride
Chloroform
Methyl Chloride
C. Mixed Gases in Cells^
(Vapor 60.4% halocarbon)
F-ll
F-112
Chloroform
Thermal Conductivity of Foam at 77°F (Btu/(hr)(ft )( F/in))
Experimentally Determined Calculated From Table V-
(where data are available) Using Eqs. (1) and (2)
0.227
0.170-0.185
0.110-0.126
0.127
0.2234
0.1600
0.1443
0.1022
0.1117
0.1001
0.0978
0.1193
0.1255
0.1313
0.1228
aj Average analysis of gases in 24-hr foams: 7.3% air; 30.3% CO.; 60.4% F-ll; 2% unknown (see Table V-7
for details).
106
-------
TABLE V-7
CELLULAR GAS ANALYSIS FROM RIGID URETHANE BUN
Air
C02
Cell Gas Composition (mol %)
CC13F
Unknown
(H20?)
K Value of
Foamk/
1
2
3
4
. 5
6
Average
8
9
6
7
6
8
7.3
34
27
31
32
30
28
30.3
57
61
61
59
62
62_
60.3
2
2
2
2
2
£
2.0
0.127
a/ All foams identical formulation, prepared on bun machine. Density of
~" all samples is 2.0716 Ib/ft3, except No. 4, which was 1.7716 lb/ft3,
b/ Determined by garded hot plate method.
107
-------
o
00
Total reported by trade
Bedding
Furniture
Packaging
Textile laminates
(except carpet)
Transportation
Other
Rug underlay:
Prime
(recycled*')
TABLE V-8
FLEXIBLE URETHANE FOAM CONSUMPTION (million pounds)
1965 1966 1967 1968 1969 1970 1971 1972 J.973 1974 1975 (P) 1980 Eat
255 307 356 480 528 635.5 673 724.9 954.8 998.8 1,014
58.8 78.0 83.4 93.5 114.4 132.6
195.0 207.5 220.0 239.8 369.6 365.0
14.0 13.0 11.0 15.1 16.5 17.6
16.0 20.5 26.5 39.6 31.9 33.4
179.0 225.0 257.4 288.2 363.0 374.0
29.7* 59.8* 22.1 24.6 25.4 26.6
18.0 23 21.8 28.6 31.9 32.3
(25) (27) (36) (50 est.)(60) (NA)
1975 (P)
1,014
136.4
352.0
19.8
26.4
341.0
17.6
30.8
(NA)
1980 Est.
1,690
175
670
50
60
440
50
110
(250)
Growth Rate
1980/1973
8.57.
6.27.
8.9%
17.07. "
9.5*
3.07.
10. OX
19.01
Total flexible foam (prime)
Total all flexible urethane foam
511.1 626.8 643.0 729.0 952.7 981.5 924.0 1.555 7.3Z
536.1 653.8 679.0 779.0 1.012.7 - - 1,805 8.6%
a/ Prompt scrap and reclaimed foam shreaded and rebonded to form controlled density, resilient padding.
* Note: Other applications reallocated starting in 1971; cushioned rotovlnyl flooring, recreational cushioning, special pads
and cushions, etc.
Source: Compiled by Midwest Research Institute from 17 recent trade reports and forecasts.
-------
* Almost all flexible urethanes are produced in plants;
on-site foaming is rare.
* Large production runs are made of specifically engi-
neered products and shapes.
* Mechanical properties and dynamic behavior of flexible
foams are much more important.
* A wide range of foam densities and stiffness (resil-
ience) is produced.
* Product specifications are tighter, requiring much
greater use of instrumentation and automation.
* Virtually no equivalent of froth foaming process is
used.
* Interconnecting, open-cell structures are predominately
produced.
Of special importance for flexible foams used in automotive
seating, furniture cushioning and textile applications are specifications
and testing for: .
* Foam density;
* Cell structure;
* Indentation load deflection;
* Compression set;
* Tear resistance;
* Elongation;
* Tensile strength;
* Flexural fatigue;
* Chemical and solvent resistance;
* Aging and oxidation resistance;
* Flammability;
* Resilience (falling ball, or pendulum); and
* Humidity' aging.
The emphasis placed on achieving uniformity and predictable
properties in flexible slab stock or molded foams makes the formulation
and processing of flexible foams more critical than is the usual case for
rigid foam. Any changes in foam blowing technology would have to be thor-
oughly tested and proven before being generally adopted.
c. Blowing agents for flexible urethane foams; All com-
mercial flexible urethane foam systems employ as the principal blowing
agent, carbon dioxide which is generated by die reaction of water with
free isocyanate groups. The amount of water in the formulation will de-
termine the density, and to a lesser extent the hardness of the foam.
109
-------
t.
While the density of flexible foams can be lowered simply by increasing
the isocyanate and water levels, the range of control is limited because
excess water changes the polymer structure of the foam. Therefore, aux-
iliary blowing agents are widely used to obtain lower foam density. For-
mulations typical of both CO^ blown and fluorocarbon assisted flexible
foams of different types are given in Table V-9. The level of FC blowing
agent ranges from 3.6 to 12.5% for most flexible systems, substantially
lower than blowing agents in rigid foams.
The use of volatile solvents to augment CO- blowing per-
mits foams to be produced at lower cost, and with both improved proper-
ties and superior aesthetic qualities. When foams are blown with CC^
alone, higher proportions of the most expensive main ingredient—the
isocyanate—must be used simply to provide the C02 for foaming. Fluoro-
carbons and other volatile solvents permit conserving use of isocyanates.
There has been a continuing shortage of aniline capacity in the U.S.
which has kept certain isocyanates in short supply. Given these economic
considerations, foam producers have strong incentives to utilize fluoro-
carbon assisted foaming systems.
Possibly of even greater importance is the control over
foam softness and compression that is provided by the use of fluorocarbon
blowing agents. Concentrations of 2 to 67, of F-ll in the formulation pro-
duce a considerable amount of softening. The density and load-bearing ca-
pacity of a given formulation can be varied, within limits, by merely in-
creasing or decreasing the amount of fluorocarbon that is introduced at
the mixing head.
Increasing the amount of fluorocarbon in a conventional
one-shot formulation results in a decrease in both density and load-
bearing properties. If high concentrations of fluorocarbon are used,
both the catalyst concentration and the silicone stabilizer level must
be increased in order to efficiently trap the large volume of gas which
is generated.
Fluorocarbon-auxiliary-blown foams are always softer than
foams of the same density,which are blown only with the carbon dioxide
generated by the isocyanate-water reaction. This difference in modulus
occurs because the C02-blown foam has a greater crosslink density and
thereby, a stiffer polymer chain in the foam structure.
Fluorocarbons also provide a desirable cooling action
during foaming, which has an important effect on the state of cure. Ad-
ding F-ll to the foam formulation reduces the extent of heat buildup,
110
-------
TABLE V-9
TYPICAL FORMULATIONS FOR FLEXIBLE URETHANE FOAMS
Po lyo 1
Isocyanate
Tin Catalyst
Amine Catalyst
Surfactant
Cell Opening Agent
Water
Fluorocarbon
Percent FC
Density, pcf
Tensile, psi
Elongation, %
Tear, Ib/in.
Compression Set
50% Deflection
90% Deflection
Indentation Load Deflection
25% Deflection
65% Deflection
Ratio ILD 65%/25%
Rebound Falling Ball, %
Hepteresis Return, %
FC Blown
Polyether
Slab*/
105.0
43.7
0.4
0.3
2.0
-
3.5
0.0
0.0
1.6
21.0
430
4.0
-
10
36
66
1.8
40
-
FC Blown
Polyether
Slabt/
100.0
46.0
0.4
0.2
1.0
-
3.6
6.4
4.0
1.4
14.0
220
2.2
6
5
30
57
1.9
38
-
C02 + FC
Blown
Polyether
Slab£/
100.0
54.3
0.3
0.1
1.0
-
4.3
6.0
3.6
1.21
16.6
192
2-7
-
4.9
34
62
1.8
-
66
CC>2 Blown
High Load
Bearing^/
100.0
64.9
0.4
0.1
1.2
5.0
5.3
_
0.0
1.2
15.3
150
5.4
-
-
40
55
1.4
-
37
FC Blown
High Load
Bearing3/
100. 0
51.0
1.0
-
1.2
5.0
1.5
10.0
5.8
5.3
21.2
101
1.8
1.6
'
79
170
2.2
10
70
Extra -
Supersoft
Foam3-/
100.0
34.1
0.2
0.2
3.0
-
2.4
20.0
12.5
1.25
6.2
149
1.2
10.3
_
9
20.4
-'
-
-
£/ Frisch and Saunders (1972).
b/ Benning (1969).
c/ Dow (1974).
-------
as illustrated in Table V-10. As the concentration of this fluorocarbon
increases, the maximum temperature in the slab decreases, and the time
to reach maximum temperature increases. The heat capacity and vaporiza-
tion of the fluorocarbon undoubtedly are contributing factors to the re-
duction of the exotherm. Formulations containing high levels of fluoro-
trichloromethane may yield foams with higher than normal compression set.
Such foams are usually given additional postcures at 250 to 275°F for 1
hr or more extended postcures at room temperature.
The use of a fluorocarbon as an auxiliary blowing agent
in one-shot foams also serves to increase the 65:25 percentage indent
load deflection (ILD) ratios. Specifications based on 65:25 ILD ratios
greater than 2.0 have been instituted for most automotive cushioning.
The importance of the "feel" of flexible foams must be
recognized. The dynamic behavior of cushioning used for bed pillows,
chair seats, mattress toppers and, especially in clothing as fabric/
foam laminates, is exceedingly critical. The.earliest attempts to use
flexible foams in garments were not successful largely because users
judged the foam layers to be too harsh, lacking in drape, with low re-
silience and high hysteresis. Within the past 10.years, considerable
improvement has been achieved in the production of softer foams having
better tactile and mechanical properties.
A wide range of foam softness classes is produced today.
The normal means of classifying flexible foams is according to the load
required under standard test conditions (called the Indent Load Deflec-
tion; ILD-257,).
ILD Range
at 25%
(Ib)
Extra Firm 44 + 6
Firm 34 + 4
Medium 27 + 3
Soft 21 + 3
Super Soft 15+3
Extra Super Soft 9+3
Since about 1970, some manufacturers have introduced even
softer grades sometimes called "Hyper Soft." The ILD-25% does not ade-
quately characterize the differences between these special low hystere-
sis foams and the normal range of soft foams. Either the ILD at 10%, or
the ratio of ILD at 65%:25% gives a better measure of the properties of
Hyper Soft foams.
112
-------
TABLE V-10.2V
COOLING EFFECT OF FLUOROCARBON-11 IN RESILIENT FOAM SLABS
CFCLj in Foam Formulation
(parts/ 100 parts resin)
-
0
2.5
5.0
7.5
Maximum Exotherm Temperature
in Slab£/ (°F)
224
223
215.5
210.5
Time to Reach Maximum Exotherm
Temperature (sec)
200
260
300
300
af Starting temperature of foam intermediates was 75°F.
-------
The use of fluorocarbon blowing agents has been essential
in achieving extreme softness and low hysteresis. Table V-ll compares
the properties of CC>2 (water) blown soft foams and those utilizing high
proportions of F-ll as the blowing agent. At present, no other blowing
agent is known that can provide the desired properties for soft, low
hysteresis foams in the super soft range and softer (i.e., ILD-257» be-
low 14 Ib).
TABLE V-ll
PROPERTIES OF SOFT FLEXIBLE FOAMS
A. Blown Without Fluorocarbons
Density (lb/ft3) 1.6 1.5 1.5
ILD-25% (Ib) 14 16 20
ILD-65% (Ib) 26 30 36
ILD Ratio 65:25% 1.85 1.88 1.80
B. Blown With 20% F-ll
Density (lb/ft3) 1.0 1.0 1.0
ILD-25% (Ib) 5 6 7
ILD-65% (Ib) 11 12 14
ILD Raio 65:25% 2.2 2.0 2.0
d. Other auxiliary blowing agents; Methylene chloride
is the only auxiliary blowing agent other than the fluorocarbons which
has been widely used commercially.— Because of its lower molecular
weight, methylene chloride furnishes a slightly greater volume of vapor
per unit weight. Its higher boiling point and higher polarity cause
methylene chloride to remain dissolved in the foam at the end of the
rise. A special inhibited grade of methylene chloride is offered as a
blowing agent for flexible urethane foams.— In both slab stock and
molded foams, it is claimed that the inhibitor minimizes deterioration
of the foam by oxidation and subsequent scorching, producing a nonyel-
lowing foam^=- Methylene chloride is also the solvent generally used
to wash out urethane production machinery for cleanup.
114
-------
C. Polystyrene Foams
Four commercially important classes of styrene foams are widely
produced in the United States. These include:
1. Extruded boardstock (Styrofoam®);
2. Expandable bead foam products;
3. Extruded foam sheet and film; and
4. Molded (high density) structural foams.
The first three types of foams utilize some form of volatile organic vapor
as the foam blowing agent. The relatively high density, structural foams
(used for bottle carriers, industrial pallets, furniture and machine parts,
etc.) do not use volatile solvents, but typically rely on thermally un-
stable organic blowing agents. These structural foams will not be consid-
ered here. The technologies used in blowing the first three types of sty-
rene foams will be discussed briefly, with emphasis on the use of volatile
blowing agents.
1. Extruded boardstock; The first commerical polystyrene foam
was extruded expanded board, developed by Dow during World War II. Molten
polystyrene containxng methyl chloride as the blowing agent is extruded
to form large foam logs, which are then cut into the desired standard
board shapes. Major uses include building construction, cold storage in-
sulation, and the floral trade.
Equipment for extruding expanded styrene board using volatile
or gaseous solvents to generate the foam comprises (a) a holding chamber,
(b) a mixing conveyor discharging through an orifice into (c) a tunnel
within which the foam expands and is shaped. Because flammability is so
important in the construction field, foams containing flame retardants
(such as acetylene tetrabromide) are widely produced. These foams gen-
erally incorporate blue pigments which serve as nuclei for vapor forma-
tion and also identify the self-extinguishing grades. —
Several systems have been demonstrated that can generate sty-
rene foams on-site for building construction, tank insulation and other
applications. To date this type of foam production in the field has not
been widely used.
Typical styrene boardstock has a density of 1.7 to 2.0
Nearly 40 million pounds of expanded styrene board were produced in 1974.
The foam is relatively brittle, which is an advantage for florists be-
cause flower stems can be forced into the green foam supports. Closed
cell structure and low water permeability make styrene logs suitable
for dock flotation and other marine uses.
115
-------
For critical'insulating applications, a special "low K-factor"
grade of foam was developed. This foam contains fluorocarbon vapor from
the blowing agent trapped within the cells, effectively reducing the ther-
mal conductivity of the insulation foam.
The incorporation of some F-12 along with methyl chloride re-
duces the tendency of freshly prepared foams to warp.—' To achieve sta-
ble, low warpage foams, sufficient fluorocarbon is added so that diffu-
sion of blowing agent vapor out of the foam is .several times higher than
the rate of diffusion of air into the foam. In some cases, the use of
normally liquid blowing agents such as methylene chloride, benzene or
acetone has been practiced instead of the more common use of methyl chlo-
ride and low boiling point fluorocarbons.—
Extruded styrene logs and board represent only a minor usage
of fluorocarbon blowing agents compared with other types of styrene foams,
or with urethanes. Less than 3 million pounds of extruded boardstock is
blown in part with F-12, and these products typically use roughly 5 Ib
of fluorocarbon per 100 Ib of foam.
2. Expandable bead foam products; The styrene foams produced
in the greatest volume are generated by expanding and molding small sty-
rene beads. In 1974, over 215 million pounds of bead foam was produced.
The styrene bead cup for hot drinks typifies these molded items.
Styrene beads are generally expanded using hydrocarbon blowing
agents such as the pentanes. Only a minor quantity of bead foams are blown
using fluorocarbon agents. Total consumption of fluorocarbons is bead mold-
ing is estimated to be about 0.5 million pounds per year, which represents
10% of the fluorocarbons used in styrene foams.
Expandable styrene beads were first intrpduced about 1954, and
volume production began after 1959. Billets of "bead board" for insula-
tion was one of the first major products, although molded products, drink
cups, and packaging applications now claim over 707» of production. The
trade generally recognizes three size ranges of expandable beads. The
smallest size designated as "cup beads," slightly larger beads called
"molder's beads," and the largest sieve size range known as "board beads."
Bead foam products are molded from prepared beads of polystyrene
which incorporate about 5 to 7% of volatile solvent. Blowing agent may be
incorporated into the beads by a variety of methods, and the volatile agent
is retained by the styrene polymer until expanded by heat. In fabrication,
beads are usually pre-expanded (using either one-step pre-expansion, or
multistage pre-expansion) with steam. As a separate step, the "pre-puff"
is then charged into a mold where steam is injected causing further expan-
sion as the beads fuse together in the desired shape. After a brief cooling
step the molded part is ejected.
116
-------
Pentanes are the agents most widely used for bead blowing, at a
typical level of 6.3 to 7.07, by weight, although butanes, hexanes and other
hydrocarbons are sometimes used to a minor extent. Nucleating agents such
as sodium bicarbonate plus citric acid are mixed into the beads to initiate
bubble formation by the volatile solvent. Foam formation is quite sensitive
to the quantity of blowing agent used, and an excess of about 1% is suffi-
cient to cause collapse of most foams. Neo-pentane, boiling at 9.5°C pro-
duces exceedingly fine cells, and is normally used only for specialty prod-
ucts, or in mixtures with other blowing agents. Iso-pentane, boiling at
28°C, gives small, uniform cells, and is the agent generally preferred for
producing .the lowest density foams. Iso-pentane, despite its lower boiling
point and higher vapor pressure, is retained within the beads better than
normal pentane. For most bead foams, normal pentane, boiling at 38°C, is
used as the primary blowing agent.
Occasionally fluorocarbons are employed for special purposes.
The use of F-ll and F-12 has become fairly we,ll established since 1967
for blowing foams of lower density; for insulating foams; and in flame
resistant products. In certain types of suspension polymerized beads,
F-114, boiling at 3.8 C, is incorporated just as the polymer reaches
the bead forming stage. >^' Sometimes a mixture of 10% methylene chlo-
ride plus 90% pentane is used to blow the foam.
Providing that fluorocarbons are properly incorporated into
the beads, they are retained as long as pentane, because they are less
soluble in the styrene polymer, and permeation out of the bead or foam
is thus retarded. However, the higher cost of fluorocarbons, together
with the smaller volume of vapor produced per unit weight of blowing
agent restricts their use to a small quantity of the bead foam produced,
primarily in applications where fire hazards are a major consideration.
3. Extruded foam sheet and film; By far the largest use of
fluorocarbon blowing agents in styrene foams is for production of foam
sheet or film from crystal polystyrene. First introduced in the mid 1960's,
consumption of foamed extruded sheet grew to about 70 million pounds by
1969. By 1974, use more than doubled, exceeding 200 million pounds. Egg
cartons and foamed meat trays are typical of this class of styrene foams.
A high molecular weight, crystal polystyrene is the starting material, so
that these styrene foams^are often referred to as "foamed from crystal."
Foamed sheet and film are produced by a single step extrusion
process. Crystal resin is melted in the extrusion machine; blowing agent
under pressure is introduced into the mixing zone of the barrel; and the
resin mixture is extruded through a circular die as it expands forming
a cylinder. The foam is pulled over a sizing mandrel, slit and wound as
a flat sheet. Thicknesses typically range from 0.050 in. up to 0.125 in.
or more. The sheet or film is then postheated and can be vacuum formed
or pressed into the final shape.
117
-------
In smooth or texturedi sheets, as well as molded forms, this
extruded foam competes with a variety of paper and molded paper pulp
products. Food packaging, disposable plates and other servicewares,
decorative ribbons, and waffle-textured sheets used to protect furni-
ture in shipment are growing applications. The foam sleeve jacketing
recently introduced for soft drink bottles has been a notable develop-
ment for the glass container industry.
Either pentanes or fluorocarbons can be used in extruding
styrene foams from crystal. Blowing agent levels are typically 6.0%,
which is similar to the proportion used for bead foams. While pentanes
are used for more than 607, of all extruded sheet foams, the use of
fluorocarbons, especially F-ll and F-12, has become more common since
about 1967. Use of fluorocarbons permits the production of products of
lighter weight and desirable tactile properties of drape and "feel."
Mixtures of pentane and the fluorocarbons are sometimes used. One com-
mercial process for foamed polystyrene sheet utilizes a 4:1 mixture of
propane and tetrachloroethylene, plus the customary mixture of sodium
bicarbonate and citric acid as the nucleating agents.—'
Extruded polystyrene sheet foamed from crystal is estimated
to consume from 6 to 10 million pounds of fluorocarbons per year. After
a year or two, practically all fluorocarbon has been lost from polystyrene
foam, regardless of type.
D. 01efin Foams
In the early 1950's, plastic foams based on polyethylene and
other polyolefins were first introduced. The original use, and still the
largest volume application was as foamed medium density polyethylene
foams for electric cable insulation (i.e., television lead-in and multi
conductor telephone cables). The dielectric constant of the insulation
layer is markedly improved by foaming the olefin polymer because of the
contribution of the gas within the closed cells. For cable insulation,
the polymer is expanded 6nly about 100%'using organic blowing agents
(Celogen OT, Kempore, Unicel NDX, etc.). In 1973, over 25 million pounds
of polyolefin was foamed for cable insulation, more than was consumed
in all other olefin foams'.
There are many processes for foaming olefins using organic pneu-
matogens that release nitrogen, COj or other gases. Neither the organic
blown foams, nor high density structural foams will be considered.
118
-------
Polyethylene and other olefins are also fabricated by extru-
sion into foam planks, rounds, tubes and sheets. About 13 million pounds
of these olefin foams were produced in 1973, over half of this quantity
was fluorocarbon blown. The most successful process for expanding low
density polyethylene foams is based on the use of fluorocarbon blowing
agents. *— The Dow Chemical Company's Ethafoam - a low density (2 lb/
ft-*), closed cell foam having cells ranging from 0.5 to 1.0 mm in size,
is typical of this class of foams.
Olefin foams are produced by forming a gel consisting of heat
plasticized polyethylene having a high activation energy of flow and a
volatile agent under pressure, and extruding the composition so as to
obtain a controlled foam expansion. The most widely used blowing agents
are F-114, F-12, and F-22. »36' Formerly, special purpose polyethylene
and polypropylene foams were produced using perfluorocyclobutane (FC-318)
as the blowing agent.— The foaming agent is fed into the mixing section
of the screw, and on extrusion to the atmospheric pressure or reduced
pressure evaporation spontaneously cools the viscous mixture to form a
stable polyolefin foam. Most of the blowing agent diffuses out of low
density olefin foams soon after formation.
E. Other Plastic Foams
Among the miscellaneous plastic foams that may employ fluoro-
carbons as blowing.agent, are epoxy foams,— pyranyl foam,— vinyl
plastisol foams,— and some phenolic and urea foams. It is likely that
the fluorocarbons have been tried as blowing agents for nearly every
kind of plastic and elastomer. The total quantity of such specialty
foams is not reported, but annual production probably does not exceed
2 million pounds. Fluorocarbon usage is undoubtedly less than 0.2 mil-
lion pounds.
F. Aggregate Consumption of Fluorocarbon Foam Blowing Agents
There are no official data on the quantities of fluorocarbons
used in plastic foams, but various estimates have been published. Several
of the recent estimates will be compared, and then revised estimates for
each major type of foam will be presented.
Chemical Marketing Reporter has published estimates for fluoro-
carbon consumption by end-use distribution three times over the past 12
years, expressed as percentage of total FC production. When combined with
production data, consumption would be:
119
-------
Total
FC Used as Fluorocarbons FC Used as
Blowing Agents U.S. Production^' Foam Blowing
Year ("/„ of Total) (million Ib) (million Ib)
1963 5 423 21.2
1968 4 661 26.4
1973 7 1,056 73.9
_a/ Shamel, R. E., et al., "Preliminary Economic Impact of Possible Regu-
latory Action to Control Atmospheric Emissions of Selected Halocar-
bons," Arthur D. Little, Inc., Contract No. 68-02-1349, Task 8,
for Office of Air Quality, U.S. Environmental Protection Agency,
Table II-1, p. II-3, September 1975; similar data from Tariff Com-
mission source omits some fluorocarbons and results in somewhat
lower production totals.
41/
The IMOS Report— has compiled several estimates of fluoro-
carbon consumption by end uses. Based on data from U.S. Tariff Commis-
sion Reports, plus information from Stanford Research Institute and
Du Pont, plastic foaming agent consumption for 1972 was placed at 50
million pounds of F-ll and 45 million pounds of F-12. This total of
95 million pounds represents 10.6% of the 889.6 million pounds of pro-
duction estimated for 1972. One producer has.indicated that actual use
of F-12 in foams is substantially less than the 45 million pounds esti-
mated for the IMOS Report.
The recent Arthur D. Little Report includes a table for usage
of halocarbons in plastic foams for 1973 (reproduced here as Table V-12).
The total of 90.2 million pounds was rounded down to indicate aggregate
use of 88 million pounds of halocarbons which includes unspecified quan-
tities of methylene chloride and methyl chloride.
43 /
A knowledgeable industry source— provided estimates for 1973
and 1974 blowing agent consumption as follows:
1. F-ll constitutes 80% of all fluorocarbons used in plastic
foams during 1974. Foam blowing consumed nearly 25% of the F-ll produced
in the United States.*
* Note: Subsequent industry discussion leads MRI to believe that F-ll
constitutes 70 to 75% of blowing agents, and that foams use
only 15 to 17% of the F-ll production.
120
-------
TABLE V-12
ESTIMATED U.S. USE OF HALOCARBOSS IN PLASTIC FOAM MANUFACTURE - 1973^ (million Ib)
Type of
Plastic Foam
Rigid
po lyurethane
Flexible
po lyurethane
Extruded
polystyrene
sheet
Expanded
polystyrene
(EPS)
Extruded
polystyrene
board
Polyolefin
Total
£/ Although
b/ Methylene
Source: Hod.
Halocarbons
Primarily
Used
F-11,12
frothing
F-11,113
mixture
methylene
chloride^
F-12, F-ll,
pentane
F-12
pentane
Methyl
chloride
fluoro carbon
F-114,12
(1)
Total Quantity
of Foam
Produced
350
960
140
190
35
11
1,686
(2) (3)
Estimated I Weight of Foam
Using Using Halocarbon
Halocarbons Blowing Agent
85 300
55 530
50 70
5 10
100 35
65 7
952
almost all of these emissions occur immediately, figure given is for
chloride used both as blowing agent and as solvent.
Plast., January
(4) . (5) (6) (7)
Percent of Halo- Estimated % Estimated Quantity
Carbon Commonly Quantity of of Halocarbon of Halocarbon
Used in Formu- Halocarbon Lost to the Lost to Atmos-
lation Used Atmosphere phere*/
15 44.8 20 9.0
. 7 35.4 95 32.9
7 4.9 95 4.7
6 0.5 90 0.4
10 3.5 100 3.5
-16 1.1 95 1.1
88 - 51
(adds to 90.2)
emissions within 10 years.
1975, and Arthur D. Little, Inc., estimates.
-------
2. For 1973, plastic foams consumed F-ll plus F-12 at levels
of:
19 million Ib flexible polyurethane foams
37 million Ib rigid polyurethanes
7 million Ib all other plastic foams
63 million Ib F-ll and F-12 for blowing agents
3. In 1973, blowing agent use can be allocated as:
Agent Million Lb
F-ll 53
F-12 10
F-113 Small (< 2)
F-114 Small (< 3)
Total < 73
To compare with previous estimates, MRI developed the figures
given in Table V-13 for 1973 and 1974 usage of chlorofluorocarbons for
plastic foams. Changes between 1973 and 1974 were not particularly sig-
nificant. The level of blowing agent used in those flexible and rigid
urethanes that employ fluorocarbons increased slightly due to greater
production of frothed foam and soft flexibles, but this was partially
offset by a corresponding rise in water blown urethanes. Somewhat higher
consumption of FC in polystyrene sheet foamed from crystal was experi-
enced in 1974. Total fluorocarbon use in all plastic foams for both years
is believed to be close to 78 million pounds. Of this total, about 55 to
59 million pounds (roughly 70%) was F-ll, while F-12 usage was between
12. and 14 million pounds. Estimates for other fluorocarbons used in smal-
ler quantities are subject to large uncertainties, but probably does not
exceed 6 to 7 million pounds.
G. Fluorocarbon Emissions from Foams
Blowing agents are trapped and retained within rigid urethane
and epoxy foams for relatively long times. In contrast, blowing agents
diffuse out of the foam structure very rapidly from flexible foams and
more slowly from the nonpolar polymers such as polystyrene and polyethyl-
ene. In addition, the foam production processes entail varying losses of
blowing agent during fabrication. Thus the quantity of blowing agent
lost to the atmosphere during manufacture, subsequent use, and ultimate
122
-------
Type of Foam
Ure thanes
Flexible
Rigid
TABLE V-13
FLUOROCARBON USAGE IN PIASTIC FOAMS 1973 AND 197A
1973
Foam Used Lb FC (avg.)/
(Million Pounds) Percent Using FC 100 Ib Foam
952.7
362.4
50
86
5.0
14.0
FC Used
(Million Pounds)
23.8
43.6
67.4
Expanded beads 202.4
Extruded board 37.4
Extruded foamed sheet 145.2
Olefins and Others-/ 12.0
Total Fluorocarbons for Plastic Foams
Urethanes
4
5
45
0.55
Olefins and Others^' 13.5
Total Fluorocarbons for Plastic Foams
1974
60
6.0
5.0
6.5
15.0
15.0
0.5
0.1
4.2
1.0
1.2
4.8
73.2
Flexible
Rigid
Styrenes
Expanded beads
Extruded board
Rvt-ntripri foamed sheet
998.8
374.0
217.8
39.6
200.2
45
88
•
4
5
48
5.0
14.5
6.0
5.0
6.5
22.5
47.7
0.8
0.1
6.2
70.2
7.0
78.4
Other foams include:Phenolic, Urea, Expoy, Polyester, Polyisocyanurate, Pyranyl, and PVC
Plastisol foams.
123
-------
disposal of plastic foams varies greatly. Among the factors that influence
losses and emissions of fluorocarbons to the atmosphere are:
• Type of plastic foam.
• Density and structure of the foam.
• Percentage of open and closed cells.
• Solubility and diffusivity of blowing agent in the polymer.
• Surface area to volume ratio of finished foams.
• Enclosure or surfacing of foam with impermeable skins.
• Lifetime of foam in use, and ultimate disposal.
It has been confirmed that fluorocarbns are retained in rigid
urethane, both by chemical analyses, ^i-*' and by thermal measurements
over long periods, as illustrated in Figure V-6.— Fluorocarbon losses
during foam manufacture depend on the equipment used, and on whether a
poured on frothed foam is employed.
Urethane foam plants must work with the relatively toxic iso-
cyanates. Because the maximum permissible threshold limit values (TLV's)
for isocyanates, are so low,* all plants are well ventilated and the foam
production equipment is usually hooded to protect the workers. Fluorocarbon-
11 concentrations, monitored in the breathing air near various stages of
foam manufacture are shown in Table V-14.— These levels of fluorocarbon
vapor are well below the allowable industrial exposures for fluorocarbons
(TLV = 1,000 ppm).
In considering the quantities of fluorocarbons emitted to the
atmosphere frora'plastic foams, the retention properties of rigid and en-
closed foams must be taken into account. Despite the lack of data on such
emissions, it can be shown that the retention and slow release over a pe-
riod of years is not trivial, but reduces losses by perhaps 40%.
A recent estimate placed FC losses from all rigid urethane
foams over a 10-year period at 207<£2/ (see Table V-12). MRI estimates
that 15 to 20% of the blowing agent is .lost within days of production,
and that an additional 10% diffuses out over the life of the foam. To
simplify estimation, this 10% loss of blowing' agent was allocated to
the 5th year after foam production. If the fluorocarbons from all other
foams are assumed to be lost to the atmosphere immediately or shortly
after foaming, the expected emission pattern would be that shown in
Table V-15.
* TLV ceiling for TDI is 0.02 ppm, ANSI Z 37.36 (1973).
124
-------
0.17
1 0.16
§ 0.15
g 0.14
E
-------
TABLE V-14
FLUOROCARBON-11 CONCENTRATIONS IN BREATHING ZONE AIR
FOAM MANUFACTURING FACILITIES
Location
Mix tanks
Foam head
Expansion - Begin
Middle
End
Cutting
Curing racks
Product storage
General building
Packaging/shipment
Laboratory/office
FC-11, ppm v/vi/
Flexible Foam
Rigid
Foam
Panel
Produe-
tion
NA
113
94
147
192
64
NA
12
70
30
60
High
Production
Rate
Regular
NA
178
146
182
181
181
355
NA
61
95
Normal Production
Rate
Regular
Foam
NA
62
31
25
24
15
280
10
26
10
Supersoft
Foam
NA
88
139
198
156
130
586
10
47
10
as general building
al All measurements from corrected POVA readings. Agreement with grab
samples typically within + 5%.
NA--Not available, e.g., no curing racks, tanks outside, etc.
126
-------
- .TABLE V-15
FLUOROCARBONS IN FOAM--CONSUMPTION AND ATMOSPHERIC EMISSIONS
Fluorocarbons (millions of pounds)
Used .in
Plastic Foams
Year
1.9.65
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
Total
All
Foams
11
16
19
29
35
43
46
5.6
73
78
(P) 77
Total 483
a/
Assuming 100%
Rigid
Urethane
Only
7.7
12.1
14.0
17.6
22.4
25.3
26.7
33.2
43.6
47.7
48.8
.299
Annual Emissions
to Atmosphere—
5
6
8
15
18
24
26
31
40
42
41
•256
prompt emissions from foams
Cumulative Emissions
from 1965^
5
11
19
34
52
76
102
133
173
215
256
other than rigid uretha
plus 207» prompt emissions for rigid urethanes, with an additional
10% allocated to the 5th year after production. .
b/ Use of FC-blowing agents negligible prior to 1960. Cumulative emissions
from 1960 to 1964 are estimated to be less than 5 million pounds.
Source: MRI estimates based on Tables V-.2, V-r8, and V-13.
(P) = Preliminary
127
-------
Although the numbers shown in Table V-15 are estimates, not based
on well established data, they do illustrate a robust conclusion: atmospheric
emissions are substantially less than fluorocarbon blowing agent usage. On
either an annual basis, or cumulative since 1965 (fluorocarbon use prior to
1960 or 1961 was negligible), loss to the atmosphere is roughly one-half of
total consumption of fluorocarbons in plastic foams.
The difference depends on two basic facts—(a) rigid foams use
twice as much fluorocarbons as do flexible foam and other types of foamsj
and (b) a major portion of blowing agents used in rigid urethanes- are re-
tained for years. Eventually, on disposal of these foams, the retained
fluorocarbons would presumably enter the atmosphere. Meanwhile, fluoro-
carbon losses to the environment are significantly less than total use
in foams.
H. Alternatives to Use of Fluorocarbon Blowing Agents
If present uses of fluorocarbons in plastic foam fabrication
were restricted, what technically feasible alternatives exist, or might
be developed?
Several general possibilities should be considered:
1. Substitution of other materials for fluorocarbon blown
foams.
2. Use of alternative blowing agents.
3. Modification of foam production technology.
4. Minimizing release of fluorocarbons to the atmosphere.
1. Substitute materials; Plastic foams have been introduced,
largely since 1945, as preferred replacements for other materials. Either
a return to former materials, or more likely, the introduction of new re-
placements for fluorocarbon blown foams is a distinct possibility.
Table V-16 presents a partial listing of current foam applications
with some indication of possible substitutes for fluorocarbon blown foams.
In many cases acceptable replacements are currently available. Certainly,
there are important considerations such as practical fabrication, cost dif-
ferences, or esthetic preferences that currently dictate the consumption
patterns for plastic foams. Switching to alternative materials in the auto-
motive seating, furniture and bedding, flotation, and packaging industries
would cause problems. Consumers would certainly have to pay for changeover
costs, and for more costly materials and fabrication.
128
-------
TABLE V-16
POTENTIAL SUBSTITUTES FOR FLUOROCARBON BLOWN PLASTIC FOAMS
NJ
vO
Current Foam Application
Building insulation
Refrigeration appliances
Industrial tanks
Pipe and pipelines
Transportation carriers
Flotation and marine
Packaging, molded shapes
Packaging, dunnage and cushioning
Bedding
Furniture and seating
Rug underlay
Vinyl flooring, bonded foam
Textile laminates
Flexible packaging foam
Electric cable insulation
Substitute Materlals-
FiberglaS®, urea foams, phenolic foams, Styrofoari®, perllte, vermlculite. (No
satisfactory replacement for some uses.)
Pyranyl foam, Fiberglas®, microcellular silica, styrene foams. (No completely
satisfactory equivalent.)
Foamglass, phenolic foams, .syntactic foams, mineral wool blankets or spray.
(No equivalent for low temperature applications.)
Fiberglas®, cellular rubber, phenolic foams, mineral fiber lagging.
Styrene foams, Fiberglas@l (No equivalent for subsequent insulation.)
Styrene foams, phenolic foams, olefin foams.
Styrene foams, thermoforraed sheet, molded paper pulp, blister pack and shrink
film packaging.
Olefin foams, styrene fill, sealed air bubble films, shaped corrugated board
blocks, vermlculite, Excelsior, Shredded newsprint, creped paper.
Innerspring mattresses, latex foams,' Dacron Fiberfil, feathers and down.
Latex foams, rubberized hair, pocketed springs, zig-wire springs, elastomer
cords, pneumatic cushioning.
Hair-jute padding, rubber latex ripple underlay, sponge rubber.
No general substitute.
Quilted Fiberfil, Pellon, special knits. (No equivalent material.)
Latex foams, sealed air cap films, vinyl foams, rubberized hair.
Radiation cross-linked olefin, neoprene inorganic insulations, pressurized
gas insulation polybutencs.
j/ Representative materials which could be substituted for the presently used foams. A few of the substitute
foams such as pyranyl and olefin foams currently are FC blown; most of the other foams do not require
the use of fluorocarbon-blowing agents.
-------
There are in addition, several applications for which no accept-
able or equivalent substitute is now known. Most critical of these are
thermal insulation foams for construction--especially on-site application
of rigid urethane foams. Other areas in which thermal properties depend
critically on fluorocarbons include: refrigeration appliances and cold
storage, industrial tanks for cryogenics or LNG, and application of thermal
insulation to existing railroad cars and barge carriers.
While other plastic foams, notably the ureas and phenolic foams
are available for on-site application as thermal insulation, they do not
approach the K-factor provided by fluorocarbon vapors within the urethane
foam cells. Foam thickness is restricted in thin building panels and cur-
tain wall construction. Lower K-value foams would require up to double
the thickness to provide equivalent insulation. Energy conservation ef-
forts in residential, commercial and industrial buildings could be seri-
ously hampered by the extra costs involved.
The fluorocarbon-blowing agents are not the only vapors capable
of contributing low thermal conductivity to plastic foams. As discussed
in Section B-2-d (pp. 99 to 105), other halogenated vapors also show
low thermal conductivity (see Table V-4). Providing that these or similar
compounds could be used in low density foams for thermal insulation, the
K-factors theoretically obtainable should be acceptable. However, there.-
are major practical problems that would have to be solved before satis-
factory rigid plastic foams could be produced for thermal insulation
without the use of the present fluorocarbons. (See discussion of require-
ments for any alternative blowing agent in Section H-2, pp. 131 to 137.)
Although it may be possible in the future to develop a low K-value foam
that does not require the use of fluorocarbon-blowing agents, there is
no proven substitute presently available for use in insulation applications.
Refrigeration appliance manufacturers have stated that if fluoro-
carbon foamed urethane was not available, they would not choose to make wall
cavities twice as thick and use CC^-blown foams, but would probably return
to fibrous insulation blankets. This would penalize the purchaser in terms
of usable interior space, and in high energy cost of operation.
Large storage tanks such as those for liquid natural gas, not
only depend on urethane insulation foams, but also on the froth spraying
method of job-site application which uses additional fluorocarbon. Pipe-
line insulation applied in the field similarly depends on fluorocarbon
expanded foam. Many railroads, now modernizing their rolling stock by
insulation, have become dependent on spray applied foams for tank cars
and hopper carriers.
130
-------
Most flexible foams can be produced by CC^ blowing. This in-
creases foam cost by about 15% because of increased isocyanate consump-
tion. Mechanical properties of water blown foams would suffer in several
respects, but for most noncritical cushioning uses, nonfluorocarbon blown
foams would be acceptable.
There is no known method of producing the super-soft flexible
foams without using auxiliary fluorocarbon-blowing, agents. Whether non-
fluorocarbon agents can be developed for this purpose remains to be seen.
Soft foams with low hysteresis and excellent recovery have become almost
essential in automotive seating. So far as can be determined, it is not.
possible to produce highly resilient flexible foams of low density without
the use of fluorocarbons.
Acceptable textile laminates having a soft hand and good gar-
ment draping characteristics are currently dependent on fluorocarbon
blowing. To our knowledge, no suitable substitute blowing agent is avail-
able for this use.
f '
Urethane rug padding, which has largely replaced latex foams
and hair padding could be produced without fluorocarbons when made as
a separate underlayment. However, many carpet manufacturers have devel-
oped integral or self-backed carpeting that uses resilient foam lami-
nated directly to the backing. Producers were unable to speculate on
the feasibility of making similar cushioned carpeting without the use
of fluorocarbon expanded foams.
2. Use of alternative blowing agents; Foam-blowing agents
must satisfy a variety of requirements. Lassman-£2 has listed the prop-
erties desired in a physical blowing agent such as the volatile solvents:
1. Agent should be odorless and nontoxic.
2. Agent should be noncorrosive.
3. Agent should be nonflammable.
4. Agent should not affect the physical and chemical prop-
erties of the polymer, although it may be soluble in, or swell, the poly-
mer at low temperatures. Must not affect rate of polymerization or cure.
5. In gaseous form, the agent must be thermally stable and
chemically inert.
131
-------
6. Agent should possess a low vapor pressure at room tempera-
ture and thus lend itself to easy handling under ambient conditions.
i
7. Agent should have a fast evaporation rate (low specific
heat and low latent heat of vaporation). Must vaporize over a definite
temperature range.
8. Agent should have a low molecular weight.
9. Agent should have a high specific gravity.
10. In gaseous form, agent should possess a slower diffusion
rate through the polymer membrane than air.
11. Agent should be readily available commercially.
12. For thermal insulating foams, vapor should have the lowest
possible thermal conductivity.
It is unlikely that any volatile compound can satisfy all of
the requirements for the ideal blowing agent. Moreover, the practical
trade-offs among factors such as flammability, toxicity, latent heat
of vaporization, vapor volume per unit of blowing agent, water solu-
bility, etc., are not easily established. It is relatively difficult
to predict in advance how the addition of any particular volatile agent
will affect the foaming process, or exactly what the effect on foam prop-
erties will be. The established fluorocarbon agents have been proven
over the years to be exceptionally satisfactory for both rigid and flex-
ible urethane foams. It is logical that any alternative agent should ap-
proach the characteristics of these fluorocarbons as closely as practi-
cal in several key properties.
Remarkably little information on the suitability of nonfluoro-
carbon blowing auxiliaries has been published. In 1958 Klesper was among
the first to report using volatile solvents for urethane foams.— He
tried acetone, ether and pentane, plus F-ll and other fluorocarbons.
While all of the volatile agents did produce foams, acetone tended to
soften the rigid urethane; pentane caused "a boiling action;" and ether
had its characteristic odor and high flammability. The results obtained
with F-ll were so superior that further investigations have been neglected.
Ten years later, Boucher and Murray made a study of 27 differ-
ent solvents as blowing agents for a low-viscosity, highly exothermic
rigid urethane foam system,^' They found that the times and temperature
of such phenomena as cream, rise, cup, and gel are strongly influenced
by the boiling point of the blowing agent as well as its chemical struc-
ture. Similar effects may be expected in flexible systems. Satisfactory
132
-------
foams were obtained with several of the higher boiling solvents, while
many solvents produced faulty foams, or caused the foam to collapse.
It is worth noting in this case that the particular foam system under
study did not produce acceptable foams with F-ll or with methylene
chloride or several other agents that are known to be satisfactory for
the usual urethane foam systems. It is believed that the unusually high
exotherm of the urethane reaction, and the effect of thermocouple probes
was responsible for the poor results obtained with some solvents.
A few other solvents have been reported as blowing agents: .
* 2-Chloropropane;
* Chloroform;
* 1,1,1 Trichloroethane; and
* Perfluorocyclobutane (FC-318).
The use of methylene chloride as a blowing agent in flexible
foams is well established. Methylene chloride can be used as the sole
auxiliary blowing agent, or it can be used in combination with F-ll.
Because of its lower molecular weight, methylene chloride produces more
vapor per unit weight than does F-ll, so that only two-thirds as much
agent is required (see Table V-17). In many cases methylene chloride
can be substituted directly in the reactive system without adjustment
of proportions. Table V-17 compares the formulation and foam properties
obtained using methylene chloride versus fluorocarbon-11. Foam density,
ILD values, and other important properties of both foams were equivalent.
Blowing agent vapors exhibiting low thermal conductivity, and
little tendency to diffuse out of the foam cells are essential for insul-
ating foams. Some of the halogenated compounds listed in Table V-4 show
relatively low K-values, and would be potential candidates for trial
in rigid urethane foams. It may be possible to utilize blends or mix-
tures of several solvents to more nearly match the properties desired
in various foam systems. Combination of ethyl chloride (b.p. 12.5°C),
methylene chloride (b.p. 40.1°C) and 1-chloropropane have been suggested.
In general, the volatility requirements limit blowing agents for poly-
urethane foams to materials boiling between 10 and 50°C.
Solvents other than the conventionally used fluorocarbon blow-
ing agents are available to match almost any boiling point range desired
(Table V-18). Some of the fluorocarbons listed are seldom used in foam
production because of their higher cost. If these agents produce accept-
able foams they might be suitable replacements for any fluorocarbon that
is restricted. The cost of blowing agents used in foams is a small part
of the cost of the finished foam but is still an important factor.
133
-------
a/ Parts per hundred parts polyol.
b/ F-ll blowing agent.
£/ Methylene chloride blowing agent,
TABLE V-17
FOAM FORMULATIONS AND PROPERTIES^
Sample Formulations, PHP— A"
VORANOL Polyol 3010 100 100
Water 4.3 4.3
L-540 Silicone 1.0 1.0
NIAX A-l Amine Catalyst 0.10 0.10
T-9 Stannous Octoate .20-.30 .20-.30
Isocyanate Type: HYLENE TM/80/20 54.3 54.3
(TDI Index 108)
F-ll 6.0 -
MeCl U.G. , -
165.9
Physical Properties Values Averaged
Density, lb/ft3 1.21 1.27
Tensile, psi 16.6 16.5
Elongation, % 192 193
Tear, Ib/in 2.7 2.9
Resiliency, % DB 47 47
Compression set 90% 4.9 4.3
4 in. ILD 25% 34 33
4 in. ILD 65% 62 61
% Hysteresis Return 66 66
Modulus 1.82 1.85
Air Flow, Middle . 5.0 5.8
Processing Conditions
Thruput, Ib/min 150
Polyol/TDI Temperature, °F 75/75
S.itrrer RPM 4,800
A.B.S. Humidity, g/lb 70
134
-------
TABLE V-18
VOLATILE COMPOUNDS .BOILING POINT -40 TO 100°C
Fluorocarbon
No. Compound B,P. ° C
Propane -42.1
22 CHCLF2 -40.8
115 C2C1F5 -38.7
GH3'CH2F -37.1
Cyclopropane -32.9
12 CC12F2 -29.8
CH2F - CF3 -26.5
.Methyl Chloride -24.1
Dimethyl Ether -24.1
CHBrF2 -15.0
Isobutane -11.7
•Methylaraine - 6.5
FC-318 Perfluorocyclobutane - 5.9
CBrClF2 - 4.0
Butane - 0.5
114 C2C12F4 3.8
.Methyl Bromide 4.6
Vinyl Methyl Ether 5.0
133 CH2C1CF3 6.1
21 CHC12F 8.9
.Neopentane 9.5
Ethylene Oxide 10.7
Ethyl Methyl Ether 10.8
Ethyl Chloride 12.5
.Cyclobutane 13.1
:Ethylamine 16.4
1,1-Dimethylcyclopentane 20.6
;Perfluorocyclopentane 22.0
11 CC13F 23.8
CBr2F2 24.5
Isopentane 27.9
123 CHC12CF3 28.2
Methyl Formate 31.4
/Fur an 31.6
Diethyl Ether 34.6
135
-------
TABLE V-18 (Concluded)
Fluorocarbon
No.
114B2
113
112
Compound
n-Pentane
Ethyl Bromide
Methylene Chloride
1-Chloropropane
Carbon Bisulfide .
C2Br F4
C2C13F3
2,2-Dimethylbutane
Perf luorocyclohexane
Ethyl Formate
Acetone
Ethylene Bichloride
Methyl Acetate
2 , 3 -Dimethylbutane
Chloroform
CH2ClBr
2,2-Dichloropropane
1,1,1 -Trichloroethane
Benzene
Cyclohexane
Perf luorocycloheptane
2 -Methy Ihexane
Methylene Bromide
ri-Heptane
B.P. °C
36.0
38.0
40
46
46
47
47.6
49.7
50.0
54.2
56,
57,
57.4
58.0
61.0
68.5
70.0
,1
,3
74,
80,
80,
82,
90,
92.8
98.2
98.4
136
-------
The cost of all materials for foams similar to those of Table
o
V-17 ranges from $0.50 to $0.60/ft of foam. At current prices, the F-ll
blowing agent represents about 4% of ingredient cost; or less than
ft . If comparable ammounts of some other blowing agent costing three
times as much as F-ll were used, ingredient cost would be increased by
about 5#/ ft3, or nearly 10%. The cost of fabricated foam to the user
would be increased by about half this percentage. If all 78 million
pounds of f luorocarbons used in 1975 to produce foams were replaced by
agents costing about $1.20/lb (an additional $0.80/lb), the increased
cost would total $62 million. For an annual output of 898 million pounds
of f luorocarbon-blown foam, the added cost would amount to 6.5^/lb.
3. Modification of foam production technology; For certain
purposes it may be possible to produce plastic foams by means other than
internal generation of vapors and gases. Some attempts have been made to
mechanically whip polymer systems into a stable foam. The presence of dis-
solved air or gas in urethane foam reactants provides nucleation indis-
pensable to foaming, — ' but efforts to use gases such as nitrogen, carbon
dioxide, argon, or air as auxiliary blowing agents in slab production
have not been commercially successful.
The Carborundum Company has developed a process for expanding
urethane foams in a vacuum. — The properties of vacuum-blown foams are,
unusual and offer promise of wide applicability. These foams are blown
in vacuum without any blowing agent; they are tougher, have a higher re-
sistence, have better resilience, and lower compression set. However,
these rigid foams have been used thus far only for special purposes.
At the present time, techniques for fabricating plastic foams
without the use of either CC^ or auxiliary blowing agents have not been
developed to a practical technology. Since foams for thermal insulation
depend upon the low conductivity of vapors within the cells, it is not
likely that any new foam fabrication process would be suitable for pre-
paring insulating foams.
4. Minimizing release of f luorocarbons; A certain amount
of blowing agent is lost during any type of on-site foam application.
Even where wall cavities are filled on-site, some release of volatile
agent is unavoidable. Self-skinning foams as applied on tanks and roof
decks help to prevent subsequent release to the atmosphere, as does the
application of weatherproof coatings.
Foams produced under controlled conditions in plants offer
the possibility of collecting and recovering most of the blowing agent
released during fabrication. While all slab stack and bun foam machines
are hooded, and vapors are removed through ventilating ducts, virtually
no attempt has yet been made to recover volatile blowing agents because
of the very high anticipated cost of recovery systems.
137
-------
Because the vapors from foam production lines contain highly
reactive and toxic isocyanates, and other organic compounds plus water
vapor, many types of solvent recovery systems cannot be used. It would
appear technically feasible to prescrub the vapors evolved during foam
production, thus removing the contaminants that interfere with fluoro-
carbon recovery and leaving the chemically stable fluorocarbons suffi-
ciently free of contaminants that collection and purification for reuse
would be possible. Recovery of 90% of the fluorocarbons from flexible
foams would collect about 20 million pounds per year, valued at $8 mil-
lion. Much less fluorocarbon recovery would be possible from rigid foams;
perhaps 5 to 6 million pounds per year could be collected from boardstock
lines and curing rooms.
However^ practical processes for vapor pretreatment, recovery,
and purification of blowing agents remain to be developed. The equipment
required would need to be corrosion resistant. Because process equipment
for fluorocarbon recovery has not yet been developed, consideration of
economics is premature. However, in certain cases where it has been nec-
essary to install other vapor recovery systems to comply with regulations,
the savings derived from recovery have largely justified the costs of the
extra equipment and processing required.
138
-------
REFERENCES TO SECTION V
1. Frisch, K. C., and J. H. Saunders, Plastic Foams, Dekker, New York
(1972).
2. Doyle, E. N., The Development and Use of Polyurethane Products, McGraw-
Hill, New York (1971).
3. Bruins, P. E., Ed., Polyurethane Technology, Interscience, New York
(1969).
4. Benning, C. J., Plastic Foams, Wiley-Interscience, New York (1969).
5. Wilson, J. E., Mod. Plast., 36jl51, September 1959.
6. Wilson, J. E., and R. H. Fowler, Science, 128jl343 (1958).
7. Wilson, J. E., H. M.' Truax, and M. A. Dunn, J. Appl. Polym. Sci.,
3j343 (1960).
8. Johnson, V., "Specifying Urethane Foam Machinery, Plast. Technol.,
pp. 104-190, mid-April 1975.
9. Kuryla, W. C., et al., "The Mold-Pressure Characteristics of Rigid
Urethane Foams," Journal of Cellular Plastics, pp. 532-537, December
1967.
10. Stengard, R. H., "Rigid Urethane Foams," Section IX, Handbook of
Foamed Plastics, R. J. Bender, Ed. (1965).
11. LeBras, L. R., Society of Plastic Engineers Journal, j.6_:420 (1960).
12. Norton, F. J., "Thermal Conductivity and Life of Polymer Foams,"
Journal of Cellular Plastics, pp. 23-27, January 1967.
13. Kevy, M. M., '!Moisture Vapor Transmission and Its Effect on Thermal
Efficiency of Foam Plastics," 21st ANTEC, Volume II, Society of
Plastic Engineers Journal (1965). •
14. Nadeau, H. G., et al., "A Method for Determination of the Cellular
Gas Content of Rigid Urethane Foams and Its Relationship to Thermo-
conductivity," Journal of Cellular Plastics, Proceedings of Con-
ference, Natick, Massachusetts, April 13-15, 1966, NAS/NRC,
Washington, D.C. (1967).
139
-------
15. Patten, G. A., and R. E. Skochdopale, "Environmental Factors in
Thermal Conductivity of Plastic Foams," Mod. Plast., pp. 149-191,
July 1962.
16. .Gorring, R. L., and S. W. Churchill, Chem. Eng. Progr. , 57^:53 (1961).
17. ASHRAE Guide and Data Book, Table 3-278, pp. 3-206, American Society
of Heating, Refrigeration, and Air-Conditioning Engineers, Inc.,
Joseph D. Pierce, Chairman, New York (1972).
18. Freon® Technical Bulletin B-9A, E. I. DuPont de Nemours and Company,
Inc. (1962).
19. Skochdopale, R. E., Chem. Eng. Progr., 5J7:55-59, October 1961.
20. Recktenweld, G. W., and W. R. Andrews, Polyurethane Technology,
pp. 70-61, P. F. Bruins, Ed., Interscience, New York (1969).
21. Traeger, R. K., "Physical Properties of Rigid Polyurethane Foams,"
Journal of Cellular Plastics, pp. 405-418, September 1967.
Analytic expressions have been developed by Harding (1962, 1964),
Patten (1962), Skochdopale (1961), and Levy (1965).
22. Steinle, H., "On the Behavior of Polyurethane Foams in Refrigera-
tor Cabinets," IIR Proceedings. Washington, D.C. (1971).
23. "Methylene Chloride, Urethane Grade: Blowing Agent Solvent, Pre-
polymer Thinner," Dow Chemical Company (1974).
24. Gmitter, G. T., and E. M. Moxey, "Polyurethane Production," in Poly-
urethane Technology, P. F. Bruins, Ed., Interscience, New York
(1969).
25. "Methylene Chloride and Refrigerant 11 as Urethane Foam Blowing
Agents," Mobay Technical Bulletin 75-F-29 (1964).
26. U.S. Patent 3,281,379 to Dow Chemical Company (1966).
27. Ballastj D. E., and J. D. Griffen, U.S. Patent 3,188,295 to Dow
Chemical Company (1965).
28. Nakamura, M., Belgian Patent 642,533 to Dow Chemical Company (1964).
29. Dynarait-Nobel, A. G., Belgian Patent 654,348 (1963).
30. British Patent 1,063,333 to Products Chimiques Peching - Saint Gobain
(1967).
140
-------
31. U.S. Patent 3,265,643 to Kanegafuchi Chemical Industry Company (1966).
32. Nakamoru, K., U.S. Patent 3,293,196 (1966).
§33. Skochdopale, R. E., and L. C. :Rubens, U.S. Patent 3,067,147 to Dow
Chemical Company, August 28, 1967.
34. Sneary, L. D.., U.S. Patent 3,102,865 to Phillips Petroleum Company,
September 3, 1963.
35. Sundquist, D. J., "Polyolefin Foams," Plastic Foams, Frisch and
Saunders, eds., pp. 280-281 (1972).
36. Rubens, L. C., J. D. Griffen, arid D. Urchick, U.S. Patent 3,067,142
(1962).
37. Skochdopale, R. E., and L. C. 'Rubens, Belgian Patent 669,041 (1960).
38. Tooky, R. P., Chetn. Eng. Progr., 57< 10), October 1961.
39. Benning, C. J., Plastic Foams, Chapter 10, Wiley-Interscience, New
York (1969).
40. Lineberry, D. D.., U.S. Patent 3,052,643 to Union Carbide Chemical
Company, September 4, 1962.
41. "Fluorocarbons and the Environment," Report of Federal Task Force
on Inadvertent Modification of the Stratosphere (IMOS), Council
of Environmental Quality, GPO No. 038-000-00226-1, June 1975.
42. Arthur D. Little, Inc., "Preliminary Economic Impact of Possible
Regulatory Action to Control Atmospheric Emissions of Selected
Halocarbons," Table IV-9, EPA Contract No. 68-02-1349, Task 8,
Publication No. EPA-450/3-75-073, NTIS No. PB-247-115, September
1975.
43. Personal communication with Charles F. Kloss, Uniroyal Chemical Com-
pany.
44. Steinle, H., "On the Behavior of Polyurethane Foams in Refigerator
Cabinets," Figures 2 and 6, IIR Proceedings, Washington, D.C. (1971),
45. Lohmeyer, S., and G. Muller, "Bestmmung der Porengasmenage und
Zusammensetung in Polyurethan-schaumen," Kaltechnik-Klimatisierung,
2^:291-294 (1970).
46. Harding, R. H., "Predicting the Performance of Foam Insulated Con-
tainers," Journal of Cellular Plastics, pp. 206-213, July 1966.
141
-------
47. Ward, R. B., "The Safety of Fluorocarbon Blowing Agents for Poly-
urethane Foams."
48. Lassman, H. R., Modern Plastics ,Encyclopedia, pp. 394-402 (1966).
49. Klesper, E., Rubber Age. 84j84-89 (1958).
50. Boucher, R. E., and J. Murray, Journal .of Cellular Plastics. 4_:216
(1968).
51. Buist, J. M., R. Kurd, and A. Lowe, "Vacuum Degassing of Urethane
Pre-Polymers Prevents Foaming with Standard Recipes," Chem. Ind.
(London), 5^1:1544 (1960).
52. Orchon, S., "New Method of Polystyrene Foam Formation," Carborundum
Company, Niagara Falls, New York (1966).
142;
-------
VI«, CLEANING AND DRYING SOLVENTS
.This section addresses the technical .considerations for the use
of alternative chemicals and/or mixtures in the .areas of chlorofluorocar-
bon solvent cleaning and drying. This market can ,be .broken down into five
applications:
Defluxing and electronics cleaning;
Degreasing;
Displacement .drying;
Dry cleaning; .and
.Miscellaneous specialty .applications.
The current overall utilization in each application can be deemed
minor with respect .to both ;.the .total overall chlorofluorocarbon consumption
and the consumption .of other solvents in .each ,of .the applications under con-
sideration. However, in many cases, the .chlorofluorocarbons are either the
solvent of choice based on value-in-use or the only solvent capable of pro-
viding the cleanliness levels .required by .military .specifications.
From data presented in the A. D. Little .report, various issues of
the Chemical Marketing Reporter» and .adjustments of these figures by person-
nel directly .related to the cleaning and drying industry but not related to
the actual manufacture of these products, the estimated 1975 consumption of
F-ll (and F-112), and F-113 .in the applications is very small (approximately
1/20 or 5%) when compared to the competitive .chlorocarbons, as shown in the
following tabulation.
Fluorocarbons, Chlorocarbons,
Chemical Million Pounds Million Pounds
F-ll (and F-112) 5.5-6..0
F-113 54-62
1,1,1-Trichloroethane 360-390
Trichloroethylene 370-390
Perchloroethylene 515-575
Methylene Chloride 110-120
Total 59.5-68 1,355-1,475
From the discussions presented in the three previous sections on
refrigeration and air conditioning, aerosol industry, and foam-blowing
agents, it can be readily observed that the utilization of chlorofluorocarbons
143
-------
in these areas (~ 63 million pounds) constitutes a small percentage of the
overall chlorofluorocarbon consumption (~ 97»)»
In addition to their thermodynamic advantages of high vapor pres-
sure and low boiling points, the chlorofluorocarbons F-ll and F-113 have
very useful solvent properties, which make them suitable for the cleaning
and drying industries. Alternative chemicals are available and widely used
in both of these industries. However, the two chlorofluorocarbons afford
the advantages of: nonflammability, high vapor pressure which results in
rapid drying, low heat of vaporization, low order of toxicity, high density,
low surface tension, chemical inertness, chemical stability to heat and ma-
terials, purity, and selected solvency as compared to many of the alterna-
tives. In many instances, it is the specific combination of properties of-
fered by the pure chlorofluorocarbons, blends, or azeotropes which make them
an economically justifiable cleaning system. Although alternate chemicals
(chlorocarbons) are available and widely used in these markets, compromises
in employee health and safety, technical performance, energy utilization,
and waste disposal often*reduce their overall value-in-use.
In the subsequent subsections, the areas of solvent cleaning and
drying will be discussed individually with respect to the technical consid-
erations for utility, current status of the various chlorofluorocarbon sol-
vents, and alternative systems for the two chlorofluorocarbons, F-ll and
F-113.
A. Chlorofluorocarbons Used in Cleaning and Prying
The selection and evaluation of alternatives in the cleaning and
drying industry is quite difficult to assess primarily because each problem,
or cleaning and drying application, is unique in that a particular type of
soil and/or grease must be removed from a specific item or piece of hardware
which in turn is fabricated from a range of materials (metals, metal oxides,
plastics, glass, elastomers, etc.). Consequently, degreasing solvents for
a particular application often have been tailored to a specific problem.
Thus identification of specific alternatives to F-ll and F-113 is very dif-
ficult in the absence of detailed knowledge concerning the specific degreas-
ing problem.
! .
1. General criteria for degreasers; There are basically three
systems or methods for material cleaning and drying: (a) cold cleaning;
(b) vapor degreasing; and (c) flushing. The first two of these methods may
be supplemented by the use of ultrasonic equipment if deemed necessary for
a specific application.
In very simplified terms, cold-cleaning systems consist of a con-
tainer for the solvent and some mechanism to immerse and remove the mate-
rial to be cleaned. The solvent remains basically at room temperature and
144
-------
the cleaning action, without ultrasonic equipment, is based entirely on
the solubility properties of the solvent. A more detailed discussion of
the mechanisms of cold-cleaning and vapor-degreasing systems has been given
in the Arthur D. Little report!' so that any detailed discussion at this
point would be redundant. The general criteria to be considered for cold-
cleaning applications are as follows:
* Compatibility of the solvent with the material to be cleaned.
* Volatility.
* Stability.
* Cost of solvent.
* Toxicity.
* Flammability, although not as critical as for vapor degreasing.
* Solubility parameters.
* Recoverability.
* Solvent composition; azeotropes or nonblends are desirable to
eliminate composition change.
Some properties of commonly used solvent materials are shown in Table VI-1.
For vapor degreasing, the solvent is heated to form a hot vapor
zone above the liquid. The function of the vapor zone is two-fold. It serves
as both a cleaning medium and a means of vapor emission control. In properly
designed and properly applied vapor degreasing systems, solvent loss can be
controlled quite effectively. In these systems, vapor density becomes an ex-
tremely important factor. The higher the vapor density the easier it would
be to retain solvent vapors within the system.
Typically, vapor degreasers are of two basic types: those designed
with only a boiling sump or vapor generating means and those containing a
secondary rinse chamber as well. In a simple vapor degreasing operation, the
part to be cleaned is immersed into the vapor zone where hot solvent vapor
condenses on the cooler part. Soil and grease are removed from the part and
are collected in the boiling chamber. Condensate cbllected in the condensing
zone of the degreaser is first passed through a water separator where any
moisture generated in the condensing process is separated from the condensed
solvent. This water separator also, functions as a source of distilled solvent
in the event a distilled solvent spray is desired. Following separation of
water, the solvent is returned to the boiling sump.
145
-------
Methylene Chloride
Perchloroethylene
TTichloroefhnne (1,1,1)
Trichloroethylene
Trichlorotrifluoroethane
(F-113)
TABLE VI-1
PROPERTIES AND COSTS OF COMMONLY USED SOLVENT VAPOR DECREASING MATERIALS"/
Flash Point OSHA TWA
CP) (ppm)^/
Boiling
Point
CF)
Latent
Heat
(Btu/lb)
Specific
Heat
(Btu/lb/°P)
Vapor
Density
(Air - 1)
Solvency
(KB Value)
Stability
in Presence
of H2(£/
Restric-
tions
Cost
Per Gallon^'
($/Cal.)
None
None
None
None
None
500
100
350
100
100
102-106
250-254
162-190
188-190
117.6
141
90
102
103
63
0.28
0.21
0.25
0.23
0.22
2.96
5.71
4.55
4.53
6.47
136
90
124
130
31
S.I
76.7
9.0
Nil.
el
11
1.95-1.98
2.22-2.26
2.12-2.14
2.155-2.195
6.53
if Data supplied by a major producer of chlorofluorocarbons.
W OSEA - 8-hr time weighted average expressed in parts per million of solvent vapor.
£/ Tests run using 200 ml of solvent containing 0.1 g of iron powder and refluxing through a water separator in a modified Soxhlet extractor for 100 hr.
At the end of the test, the increase in the test system (HC1 equivalent) was determined. Hethylene chloride and perchloroethylene were not run, but
. methylene chloride is considered to be* "Fairly stable in the presence of water," and perchloroethylene to be the "least susceptible" of the chlori-
nated solvents. * Selecting Alternative Chlorinated Solvents. Archer - Metal Progress, p. 133, October 1974.
i/ Chenieal Marketing Reporter.
£/ There is some concern developing regarding the elevation of carboxyhemoglobin following exposure to methylene chloride. Refs. Carboxyhemoglobin
Elevation After Exposure to Dichlorone thane - Stewart, Terrance, Fisher et al., Science. 176, 295 (1972). Experimental Human Exposure to Methylene
Chloride - Stewart, Terrance, Pisher et al., Archives of Environmental Health. 25, 342 (1972).
_£/ Dioxaue, an inhibiting agent commonly used in trichloroethane, is listed as a suspect carcinogen by the National Institute of Occupational Safety and
Health (NIOSH). It has been placed on the carcinogenic surveillance list of the American Conference of Governmental and Industrial Hygienists (ACGIH).
The ACGIH TLV for dioxane is 50 ppm; the OSHA TWA is 100 ppm.
g/ Under Los Angeles County Rule 66, emissions of trichloroethylene were limited to 40 Ib/day/piece of equipment. The EPA under the Clean Air Act of 1970
has recommended that Rule 66-type controls be implemented in those other geographic areas where trichloroethylene emissions are to be limited.
-------
In a two-stage system, the part may be immersed in both the boil-
ing liquid and the rinse liquid prior to rinsing in the vapor zone. In the
event ultrasonic cleaning of the part is desirable, this modification may
be made to the rinse chamber. Advantages of this system, in addition to the
increased cleaning efficiency of multiple bath exposure, stem from increased
efficiency in use of ultrasonics through solvent vapor loss control of the
ultrasonically agitated liquid, once degassed.
The criteria for selecting candidates for vapor-degreasing sys-
tems are generally better defined than for cold-cleaning solvents and would
include the following:Jtl§'
* Nonflammable and nonexplosive.
* Low toxicity.
* Boiling point consistent with the temperature sensitivity of
the material to be cleaned and the desired vapor zone clean-
liness level.
* Low latent heat of vaporization.
* Low specific heat.
* High vapor density.
* Effective removal of ionic, nonionic, and particulate soils.
* Compatibility with materials to be cleaned.
* Stability of the system in the presence of water.
Some typical examples where chlorocarbons and chlorofluorocarbons are used
in vapor-degreasing applications and the factors affecting the selection
of the solvent are shown in Table VI-2.
2. Current use areas of F-113 and F-llt The very small current
use of F-ll and F-112 is limited either to highly specialized, low volume
applications or as a replacement for F-113 if it is unavailable to the user.
Consequently, all discussions in this subsection will be directed towards
F-113 with the understanding that, in certain isolated instances, F-ll or
F-112 are used as a replacement. Azeotropes of F-113 and F-112 with other
common solvents are also .used; the specific azeotrope is.dictated by the
type of grease or other foreign material to be removed. Some of the common
F-113 and F-112 azeotropes and blends, currently used for either vapor de-
greasing or cold-cleaning applications, are shown in the following list.
147
-------
TABLE VI-2
TYPICAL APPLICATIONS FOR VAPOR-DECREASING SOLVENTS^/
•P-
00
Application
Removal of oils, greases,
and waxes from parts
Removal of high-melting oils,
greases, and waxes
Removal of water films from
metals
Cleaning coils and components
for electric motors
Cleaning temperature-
-sensitive materials
Cleaning components -for
rockets or missiles
Cleaning with ultrasonics
Solvent
Trichloroethylene
Perchloroethylene
Ferchloroethylene
Freon IP 35
Freon T-DA 35/TF
Methyl chloroform^
Trichlorotrifluoroethane
Perchloroethylene
Trichlorotrifluoroethane
Methyl chloroform^/
Trichloroethylene
Perchloroethylene
Methyl chloroform^/
Trichlorotrifluoroethane
Trichloroethylene
Perchloroethylene
Methyl chloroform^/
Trichlorotrifluoroethane
Approximate
Vapor Temp.,
"F
188
250
250
118
HA.
165
118
250
118
165
188
250
165
118
188
250
165
118
Factors Affecting Selection
Most commonly used degreasing solvent.
Used where higher operating temperature is desirable.
Rapid and complete drying in one operation.
Used only at room temperature for drying and only when
small quantities of water are to be removed.
Used where large quantities of water are involved and
spot-free drying is desired.
Solvent must not damage wire coating or sealing agents.
Selection should be based on preliminary trials.
Used where parts must not be exposed to higher vapor
temperature during cleaning.
Cleaned parts must be free of soils or residues that
might react with oxidizers. White room cleaning.
For cleaning efficiency beyond that obtained from
standard vapor degreasing. Solvent must be kept
clean by continuous distillation and filtration
during use. Selection should be based on prelimi-
nary trials.
a/ Metal Progress. T. J. Kearney and C. E. Kircher, p. 93 (modified), May 1960.
b/ Use of methyl chloroform requires special equipment design and corrosion-resistant materials of construction.
Source: See Reference 4.
-------
Vapor Decreasing Agents
Composition
32% F-112 + 68% isopropanol
85.5% F-112 + 14.5% n-propanol
F-113
F-113 (elec. gr.)
96% F-113 + 4% isopropanol
96.25% F-113 + 3.75% ethanol
95.5% F-113 + 4.5% ethanol
55% F-113 + 42% methylene chloride
+ 3% methanol
50.5% F-113 + 49.5% methylene chloride
89% F-113 + 11% acetone
94.05% F-113 + 5.7% methanol +
0.25% nitromethane
40% F-113 + 51% methylene chloride
+ 9% cyclopentane
95% F-113 + 5% acetonitrile
95.2% F-113 + 3.8% SDA-30 ethanol
+ 1.0% nitromethane
Trade Name
Alpha 1001
Alpha 1003
Freon TF, Genesolv D
Freon PGA, Genesolv D-elec. gr.
Genesolv DI
Freon TE
Genesolv DE
Genesolv DTA
Freon TMC
Freon TA
Freon TMS
Genesolv DMC
Genesolv DA
Freon TES
Cold-Cleaning Agents
Composition
90% F-113 + 10% isopropanol
65% F-113 + 35% isopropanol
64.7% F-113 + 35% isopropanol +
0.3% nitromethane
90% F-113 + 10% SDA-30 ethanol
85% F-113 + 15% SDA-30 ethanol
65% F-113 + 35% SDA-30 ethanol
91.5% F-113 + 6% water + 2.5% surfactant
Trade Name
Genesolv DS-10
Genesolv DI-35
Freon T-P 35
Freon T-E 10
Genesolv DE-15
Freon T-E 35
Freon T-WD602
The Alpha series are products of Alpha Metals, Inc.,—' while the Freon
series are produced by E. I. du Pont de Nemours and Company, Inc., and the
Genesolv products are by Allied Chemical. The information shown above is
for illustrative purposes only and is not intended to be a complete compila-
tion. Its utility is to exemplify some of the numerous types of mixtures
of F-112 and F-113 that are available commercially and the other components
that are often added to the chlorofluorocarbons.
149
-------
At the present time, the primary use of F-113 and its azeotropes
is for the vapor degreasing and cleaning of components in the electrical
and electronic industry.1 It is commonly used in systems where ultrasonics
are employed in conjunction with the vapor degreasers or cleaners. Because
it can be prepared with very high purity, has low toxicity, and possesses
excellent material compatability, F-113 based materials are utilized in
white-room metal cleaning applications. In addition, the low boiling point
of F-113 (48°C) allows its usage with temperature-sensitive materials or
component systems. A former major use of F-113 was in the aerospace in-
dustry but, since the termination of the Apollo program, limited quantities
have been used in this application. However, spin-off industries generated
during the space program, which depend on state-of-the-art technology, use
F-113.
The major specific use area for F-113 is the flux removal and
general cleaning of printed circuit boards (PCB) for the electronics in-
dustry. These solvents are well-suited for this application as they effec-
tively remove the fluxes used in high volume production of machine soldered
boards. These solvents are also well suited for flux removal following hand-
soldering operations generally associated with PCB repair and rework. Other
specific areas of application for F-113 and its azeotropes included'
* Electronic relays * Gyroscopes
* Switches * Accelerometers
* Potentiometers * Precision valves
* Electrical connectors * Bearings
* Electrical motors * Magnetic tape heads
* Semiconductors * Vacuum tubes
* Amplifiers * Nuclear power equipment
* Memory discs * Hydraulic systems
* Integrated circuits * Missile fuel systems
In addition to applications directed primarily at the manufacturing aspects,
F-113 and its azeotropes are also used for in-place maintenance by spray
cleaning while the equipment is completely assembled or in operation. Ex-
amples of such applications include:
* Electrical control panels
* Telephone switch gear
* Electrical motors
* Fork lift trucks
* Precision bearings
3. Alternative systems; The selection of substitute solvents or
alternative cleaning procedures to replace the current chlorofluorocarbons
(F-ll and F-113) is a difficult task due to the variety of unique cleaning
and drying problems. Metal degreasing is an art that has achieved its present
150
-------
state of development by ferial and error methods of,evaluation of the avail-
able solvents and aqueous surface active agents. In almost direct contrast
to the other areas of application, the area of solvent cleaning and degreas-
ing is much more of an art than a science.
The advantages and disadvantages, relative to F-ll, F-112, and
F-113, of some of the more common viable substitutes are discussed in Table
VI-3."***' Most of these solvent and degreasing systems are widely used at
the present time and were the choice before being replaced in specific ap-
plication by chlorofluorocarbons.
Formulations of alkaline cleaners and detergents for use in aque-
ous systems are complex mixtures of primarily inorganic compounds that are
tailored to the specific application. Two typical formulations for unheated
alkaline cleaning solutions are shown belowi^'
Component % by Weight
Solution A
Borax 28-30
Sodium tripolyphpsphate 13-17
Sodium orthophosphate mono-
hydrogen 8-12
Sodium carbonate, anhydrous 25-30
Sodium nitrite 6-12
Wetting agents 8-10
Solution B
Sodium sesquicarbonate 27-33
Sodium pyrophosphate 27-33
Sodium tetraborate 18-22
Sodium nitrite 14-16
Polyoxyethylene glycol 2.9-3.1
Petroleum solvent 1.9-2.1
The variation in alkaline cleaner formulations for various metals with re-
spect to the specific metal and the method of application is exemplified
by the data shown in Table VI-4. From the data in this table and the above
information, it is readily apparent that, while alkaline cleaners may be
acceptable substitutes in certain applications, no single formulation will
suffice and each application will require a specific formulation.
A review of the patent literature (via Chemical Abstracts) from
1967 to 1975 was conducted for new azeotropes and blends of chlorocarbons
and/or chlorofluorocarbons with applications in the solvent degreasing
field. The review revealed very few new systems and only one that did not
incorporate either F-112 'or F-113. A system of 35 to 58% perch lor oe thy lene
and the balance methylene chloride (CA, 79^ 80658g, 1973) was patented for
metal cleaning applications.
151
-------
TABLE VI-3
COMM3N SOLVENTS AND DECREASING SYSTEMS AS POSSIBLE ALTERNATIVES
System
F-113 CC12FCC1F2
to
Trlchloroe thylene
ClBC:CCl2
Perchloroethylene
C12C:CC12
Application
a. Cleaning
• Electronic parts
• PCS
• Belays and thermal switches
• Computer discs
• Electric motors
• White room components
• Optical lenses
• Waveguides
• Precision bearings
• Hydraulic systems
• Integrated circuits
• Semiconductor devices
• Business machines
b. Drying
• Coin blanks
• Plated parts
• Jeuelery
• Glass
• Ultrasonic cleaning
• Wool degreasing
• Electronic parts cleaning
• Solder flux removal
• Electric motor cleaning and
maintenance
• Business machine cleaning
• Flush solvent for liquid
oxygen systems
• Missile and rocket component
cleaning (white room)
• Electronic parts cleaning
• Solder flux removal (Incomplete)
• Optical lens cleaning
• Wool scouring
• Cleaning of electric motors
and component parts
• Ultrasonic cleaning
• Missile and rocket (component
cleaning (white room)
• Component of proprietary solvents
Relative Advantages
Nonflammabllity
Low toxicity
Chemically stable without
inhibitors
Low surface tension
High liquid and vapor
density
High purity
High dielectric strength
Recoverable by distillation
and carbon absorption
Does not normally require
venting
Less expensive
Readily applicable to current
equipment
High cleaning efficiency
High boiling point permits
metal drying, reduces stain-
ing, removes waxes, pitches
and other gross high melting
contamination
Stable to hydrolysis
Less expensive
Good solvent power (KB value
of 90)
Best carbon adsorption re-
covery of chlorocarbons
May be operated in air-cooled
equipment but: not recommended
Relative Disadvantages
High initial cost
Limited use as cold cleaning agent
Limited solvency
Low boiling point
Photochemlcally reactive
Probable incompatibility with some
plastics, elastomers, paints, etc.
High boiling point
TLV - 100 ppm
Reacts with sodium hydroxide to pro-
duce dichloroacetylene which is
toxic and spontaneously combustible
Requires engineer to operate
boilers for vapor degreasers
(40-60 pslg steam)
High heat Input (steam)
High boiling point presents burn
hazard, high solvent consump-
tion, may warp or structurally
change heat-sensitive parts
Requires special controls for
aluminum alloys
Steam distillation causes water
contaminat ion
TLV • 100 ppm
-------
Methylene chloride
CH2C12
Methyl chlorofoxn
CH,CC1,
Aqueous systems
Enxilslon cleaners
Alkaline cleaners
Detergents
Neutral synthetic
cleaners
Application
Thermal switches
Removal of photo-resistant ink
(PCB)
Cleaning of temperature-sensitive
equipment
Removal of solder flux
Cleaning electrical coils and
motors
Cleaning office equipment
Cleaning temperature-sensitive
equipment
Ultrasonic cleaning
Ultrasonic cleaning
FCB
Waveguides
Switch components
Instrument connector pins
Ring bearings
Vacuum tube cleaner
Coated and uncoated lenses
TABLE VI-3 (Concluded)
Relative Advantages
• Highest solvent power for paint
and varnish
• Virtually unreactive under atmo-
spheric conditions
High solvent-efficiency
Less expensive
Suitable for cold cleaning
Hay be used on PCB and hermetic
statore, depending upon con-
struction material
Less expensive
Nontoxic
Removes both organic and
inorganic contaminants
Almost complete material
compatibility
Nonflammable
Relative Disadvantages
Incompatible with plastics,
elastomers and paints
Low vapor density results in
excessive solvent loss without
a high freeboard
High evaporation rate
TLV » 100 ppm
Moderate boiling point
Requires good condensation system
and temperature control
Carbon adsorption recovery system
depletes stabilizers
Must be stabilized against hydrolysis
Cannot be used with gross water con-
tamination
Photochemically reactive
Unstable in the presence of freshly
machined aluminum
Requires large operating areas
Requires multiple rinsing and dry
oven
Increases process time
Presents potential for moisture
and corrosion' problems
Can present disposal problem; must
be separated from water
Requires high energy input to heat
or boil
Difficult to remove from blind holes,
crevices, and densely packed printed
circuits
Not applicable for military specifica-
tion applications
-------
TABLE VI-4
TYPICAL ALKALINE CLEANER FORMULATIONS FOR VARIOUS METALS"
Aluminum
Soak Spray
Builders
Sodium hydroxide, ground
Sodium carbonate, dense
Sodium bicarbonate 21 24
Sodium tripolyphosphate 30 30
Tetrasodiun pyrophosphate
Sodium metasilicate, 45 45
anhydrous
Surface-Active (Wetting) Agents
Sodium resinate
Alkyl aryl sodium sulfonate 3
Alkyl aryl polyether alcohol - -
Non ionics high in ethylene 1 1
oxide
Soak
20
18
-
-
20
30
5
5
2
-
Copper Cu Plate
Electro- Electro-
Spray lytic lytic
Composition
15 15 55
8
34 34
10
10 - 10
40 40 25
-
-
1
11 1
Iron and Steel
Soak
Spray
of Cleaner, 7.
20
18
-
20
-
30
5
5
2
-
20
29
-
20
-
30
-
1
Electro-
lytic
by WefRht
55
8.5
-
10
-
25
-
1
•"
0.5
Magnesium
Soak Spray
20 20
18 29
-
20 20
-
30 30
5
5 - .
Z -
1
Zinc
Electro-
Soak Spray lytlc
15 15
-
35 34
90 10 10
-
40 40
5 -
5 -
.
- - 1
Other Conditions
Operating temperature of 160 160
solution, *F
Concentration of cleaner, 4 1
180
8
170 160 180
18 8
200
8
170
1
180
8
200 170
8 1
180 170 180
41 6
oz/gal H20
-------
Additional chlorofluorocarbons and fluorocarbons are presently
available in research quantities which may prove to be acceptable substi-
tutes. It must be stressed that the suggested applicability of these mate-
rials is purely speculative and based on very limited available information.
The materials discussed below are all products of E. I. du Pont de Nemours
and Company but similar products may be available from other fluorocarbon
manufacturers.
Perfluorodimethylcyclobutane (C51-12) has a boiling point of 113°F
(45°C) and is a stable fluid for heat transfer. It is compatible with elec-
trical insulation so that, if it has good solvency properties, it may be
useful for electrical motor and components cleaning applications. Its use
as a heat-transfer medium would indicate good thermal stability and compat-
ibility with common metals.
Two other compounds, F-132a and Freon® E-l, may find application,
based strictly on boiling points, as suitable solvents but no information
is available with regard to their solvency powers or compatibilities with
various materials. F-132a is l,l-dichloro-2,2-difluoroethane and Freon® E-l
is the highly fluorinated ether, CF3CF2CF2OCHFCF3.
As stated earlier in this section, the identification of speci-
fic alternatives is very difficult because each cleaning or drying problem
is unique. The chlorocarbons are still widely used in the degreasing and
solvent field and may be acceptable substitutes in some applications
presently employing the chlorofluorocarbons; however, there are cleaning
problems where no acceptable substitutes for chlorofluorocarbons exist. Use
of the chlorocarbons in defluxing and electronic cleaning, major industries,
has been limited by solvent/electronic component compatibility problems aris-
ing from their higher boiling points and greater solvent power.
The initial conversion from chlorocarbons to chlorofluorocarbons
was due to a combination of technical and aesthetic reasons. A major factor
influencing the conversion of many systems from chlorocarbon to fluorocarbon
was the concern for employee safety. Technical factors include situations
where chlorocarbons resulted in residue contamination problems, compatibility
problems, water spotting, and incomplete cleaning. Other factors included
multistep cleaning procedures, cooling periods prior to handling of cleaned
parts, higher rejection rates, and overall reduction in production volumes
and increased labor costs. Increased emphasis on worker environment will
necessitate the use of better vapor and liquid control measures should a
return to the use of chlorocarbons occur. These increased control measures
will result in additional capital outlay for new equipment. In addition to
additional capital outlays, the increased control measures dictated by a
conversion to chlorocarbon solvent systems will also result in increased
solvent consumption and costs for chemical monitoring and control. Imple-
mentation of freeboard chilling systems or carbon adsorbers will deplete
155
-------
inhibiting systems commonly used in methyl chloroform and methylene chloride.
Control of solvent vapors emitted from trichloroethylene and perchloroethyl-
ene degreasers will dictate the use of exhaust systems, with attendant loss
and environmental problems. The lower cost of chlorocarbons as compared to
chlorofluorocarbons should offset a portion of the monetary cost of control
equipment. The amount of'compensation will be determined by the adaptability
of present equipment to the new solvent and the specific solvent application.
Other common solvents, such as ketones, alcohols, or hydrocarbons,
are presently used in very small volume operations where a close control can
be maintained on their utilization. Since essentially all of these solvents
have low flash points and explosive limits, extreme caution would be neces-
sary if their use were expanded or more generalized and would necessitate
the use of very efficient vapor control and recovery equipment. Due to the
inherent danger, flammable solvents are not considered to be acceptable al-
ternatives. The observance of toxicity limitations would also be a factor
in the utilization of these materials as in the case of the chlorocarbons.
As with the chlorocarbons, no specific alternative can be suggested for wide-
spread applicability since each degreasing operation presents its own set
of circumstances.
The ultimate selection of an appropriate solvent obviously is not
a well-defined situation. In some instances the widely used chlorocarbons
may be acceptable alternatives, and in other cases the use of aqueous solu-
tions of cleaners and detergents could be satisfactory. A major application
of fluorocarbon solvents is the removal of rosin flux from printed circuits
intended for military use. With the many different materials involved in the
fabrication of printed circuits and the problem of qualifying a new solvent
to meet military specifications, it is unlikely that a ready substitute is
available. Our general assessment of this use category for chlorofluorocarbons
is that substitute solvents and/or cleaning technology may be available for
some applications presently employing chlorofluorocarbons, but due to the
unique problems associated with each cleaning or drying problem, the consid-
erations should be made on the basis of each individual Use.
B. The Dry-Cleaning Industry
This area of application consumes very small quantities of chloro-
fluorocarbons in the United States. It was stated in the introdduction to
this section that approximately 2 to 4 million pounds of F-113 were used in
1975 in the dry-cleaning .industry, as compared to an estimated 400 million
pounds of perchloroethylene. In Europe, F-ll is used as a dry-cleaning agent
but not in the United States.
156
-------
There are three basic types of dry-cleaning equipment systems:
transfer type, dry-to-dry type, and coin-operated. A detailed discussion
of dry-cleaning equipment and solvent mileage can be found in the Arthur D.
Little Report.-'
Perchloroethylene can be used in all of the three equipment
types. Fluorocarbon* 113 is used in dry-to-dry (often called "Valclene"
machines) systems. Care is taken to minimize loss of F-113 because of
its high cost. Petroleum distillates (Stoddard solvent and 140-F) are
used in the transfer type (washer-extractor) systems. According to a
recent survey of the International Fabricare Institute,— the most com-
monly used systems are of the washer-extractor type.
10.12/
1. General criteria required of a dry-cleaning solvent: —
To qualify as a dry-cleaning solvent, a candidate must possess certain
physical and chemical characteristics. The National Institute of Dry-
Cleaning has set seven general criteria for a dry-cleaning agent. These
criteria are as follows:
a. The dry-cleaning agent must be safe to use on all common
textile fabrics and dyes.
b. It must be a good solvent for fats and oils.
c. The candidate must be free of objectionable odors.
d. It must be chemically stable under all use conditions.
e. The dry-cleaning solvent should be noncorrosive towards
ordinary metals used in machine, pipe and pump construction; it must also
not swell or dissolve a wide range of plastics.
f. The candidate should be sufficiently volatile to permit
economical reclamation by distillation and to permit rapid, economical,
and safe drying conditions.
g. The solvent must also be compatable with a wide range of
detergents so as to enhande its cleaning ability.
In conjunction with the above criteria, the solvent must be low
in toxicity to minimize an,y hazard to the worker during processing and to
the customer should residual solvent be left in the clothing. The solvent
should have a flash point above 100°F (38°C) or should be nonflammable.
The dry-cleaning agent should not bleed dyes or weaken, dissolve, or shrink
textile fibers,!3-/
* The term fluorocarbon designates both the fluorocarbons and the
fluorochlorocarbons.
157
-------
.1.13-15/
1 tkxneitlc Flaih
Dry-cleaning Point
Solvent Markets' CC/'F)
Perchloroethylene 72 None
Stoddard tolvent 1 41/105
I 26
140-F 1 59/138
j
Fluorochlorocarbonf '
F-113&/
Cn
00 F'u
Trichloroethylene
Methylene chloride
Methyl chloroform >
None
None
2
32/90
None
Kone
.
Boiling Dry End
Point Point
CC/'F) CC/'F)
121/250 123/254
152/305 177/350
181/358 202/396
48/118 NA£/
24/75 NA
87/189 NA
40/104 NA
75/165 NA
Spec If ic
Gravity
CC/60'F)
1.623
0.766
0.789
1.574
1.494
1.462
1.320
1.385
Aromatic
Density Content
Ib/gal Volume
(25'C) q)
13.55 0
6.38 11.6
6.57 12.1
13.16 0
12.34 0
12.16 0
(20V20-C)
0
11.50 0
Beat of
Vaporization
Corroaiveneas (Btu/lb)
Slight on metal 90
None - 500
None ~ 500
None 63
None 78.31
None HA
NA 142
NA NA
Toxicity
TLV
(pro)
100
200
200
1.000
1.000
100
250
500
Odor
Ether- like
Sweet
Mild
Like CCU
Slight
ethereal
Chlorofom-
llke
Ether- like
Chloroform-
like
Color
Water white
Water white
Water white
Water white
Water white
Water white
Water white
Water white
Vapor
Demlty
(Air -
l.QO)
5.3
1.0
1.0
6.3
5.0
HA
NA
NA
MRI estimates based on d*ta ID References 2 and 3.
b/ Valcleae® (E. 1. du Pont de Nemours and Company trade name) • P-113 + 0.1Z catlonic detergent.
c/ NA • not available or not known.
-------
2. Solvents currently used for dry-cleaning; Solvents commonly
used in domestic dry-cleaning are within one of three chemical classes:
petroleum distillates, chlorinated hydrocarbons, or chlorofluorocarbons.
a. Petroleum distillates; The two major petroleum distil-
lates still used extensively in domestic dry-cleaning are Stoddard solvent
and 140-F. Stoddard solvent is a water-white, basically odor free petroleum
distillate with a flash point of 100°F while 140-F is a petroleum distillate
solvent so named because its flash point is 140°F. In specialized equipment,
140-F can be used in locations where Stoddard solvent is not allowed. It is
somewhat more expensive than Stoddard solvent and is slower drying. A sum-
mary of the physical properties of these petroleum distillates can be found
in Table YI-5.^0-'
b. The chlorinated hydrocarbons; The major domestic dry-
cleaning agent used is perchloroethylene. Perchloroethylene is nonflammable
and safe for use with a wide range of textile fibers. However, many spe-
cialty items such as furs and leathers cannot be cleaned with perchloro-
ethylene. In addition to perchloroethylene, there are three other chlori-
nated hydrocarbons used as dry-cleaning agents: trichloroethylene, methyl
chloride, and methylene chloride. None of these are used in significant
quantities in the United States, because of incompatability with many syn-
thetic fibers (e.g., trichlorethylene tends to bleed dyes from acetate
fabrics).— Physical properties and additional information for the chlori-
nated hydrocarbon dry-cleaning solvents can be found in Table VI-5.
c. The chlorofluorocarbons; To date, only F-113 has been
used domestically as a dry-cleaning solvent. F-ll is used only in Europe.
E. I. du Pont de Nemours and Company offers a F-113/detergent combination,
Valclene®. This compound is expensive in comparison to the petroleum dis-
tillage and chlorinated hydrocarbon solvent and is used primarily for spe-
cialty cleaning, i.e., furs and leather. The petroleum distillates can
also be used as specialty cleaners but, unlike F-113, the petroleum solvents
are flammable. Perchloroethylene is nonflammable but its solvent properties
are too strong to function as a speciality cleaner.
3. Alternatives to the chlorofluorocarbon solvents; Only a
very small segment of the dry-cleaning industry would be affected should
further usage of F-113 not be allowed. Such discontinued use of F-113
would affect approximately 2% of the total market. In the event F-113 is
regulated, other chlorofluorocarbons or the petroleum distillates could
be used as specialty dry-cleaning agents. In the final processing of
leather goods, it is possible to treat the leather finish in a manner
that will allow the use of perchloroethylene as a cleaning agent^iS/ A
solvent, having a Kauri-Butanol (KB) value in the range of 30 to 40 and
satisfying the general criteria discussed previously in this chapter,
would be suitable as a specialty item cleaner and function as an alter-
native for F-113.
159
-------
As previously mentioned, many of the petroleum distillates
can serve as dry-cleaning solvents for furs and leathers. The KB values
of Stoddard solvent, 140-F, hexene, pentane, etc., vary between 27 and
45, thus making them possible alternatives to F-113 with respect to sol-
vency power. The petroleum distillates are flammable and cannot be used
in shopping center dry-cleaning establishments. This problem could be
generally adverted by locating specialty cleaning operations in nonre-
stricted areas*
160
-------
REFERENCES TO SECTION VI
1. Arthur D. Little, Inc., "Preliminary Economic Impact of Possible Regu-
latory Action to Control Atmospheric Emissions of Selected Halocarbons,"
EPA Contract No. 68-02-1349, Task 8, Publication No. EPA-450/3-75-073,
NTIS No. PB-247-115, September 1975.
2. Chemical Marketing Reporter, Chemical Profiles, various issues.
3. Chemical Eng. Progress. (39, 85 (1973).
4. Surface Preparation and Finishes for Metals, James A. Murphy, Ed.,
McGraw-Hill, New York (1971).
5. Metal Progress. April 1975.
6. Personal Communication, Dow Chemical Company.
7. Electronic Packaging and Production, p. 78, April 1975.
8. Technical Data Sheets, E. I. du Pont de Nemours and Company and Allied
Chemical Company.
9. Metals Handbook. 8th Ed., Vol. 2, Taylor Lyman, Ed., American Society
for Metals (1964).
10. Kirk-Othmer Encyclopedia of Chemical Technology. 2nd Ed., Vol. 7,
John Wiley and Sons, New York, p. 307 (1965).
11. The IFI Special Reporter. Vol. 3, No. 1, International Fabricare In-
stitute (1975).
12. "Valclene®," Technical Bulletin T-458, National Institute of Dry-
Cleaning, February 1970.
13. E. I. du Pont de Nemours and Company, Technical Bulletins 5-16, FS-
1, FST-1, and FST-5A.
14. Properties of Common Organic Solvents. The Chemistributors, The Sol-
vents and Chemicals Company (1972).
15. The Condensed Chemical Dictionary. 8th Ed., G. G. Hawley, Ed., Van
Nostrand Reinhold Company, New York (1971).
j
16. Personal Communication, Mr. Lloyd, International Fabricare Institute.
161
-------
VII. POTENTIAL RANKINE CYCLE-USES
In June 1972, at the United Nations Conference on the Human En-
vironment in Stockholm, Sweden, representatives of some 113 nations met to
translate their environmental concerns into a plan for action to preserve
and enhance the human environment. The Stockholm Conference set the stage
for concerted world action to solve environmental problems of concern to
both developed and developing nations. One of the results has been to es-
tablish a list of pollutants, in order of priority, that must be controlled.
Table VII-1 shows that organochlorine compounds in the air ranks second in
order of priorityJ^'
One possible source of organochlorine compounds in the air is
through losses from Rankine cycle turbines using chlorinated fluorocarbons
as the working fluid. More specifically, F-ll, F-12, F-13, F-112, F-113
and F-114 are potential causes for concern. These particular chlorofluoro-
carbons are sufficiently stable to pass through the lower atmosphere and
interact with the ozone layer (under ultraviolet light).
A. End Uses
The use of vapor turbines operating on the Rankine cycle with
fluorocarbons as the working fluid is somewhat limited at the present time
but is anticipated to show future growth. An examination of the state of
the art indicates that most of the fluorocarbon turbines are experimental
models with a rating of a few hundred kilowatts. No large turbines are cur-
rently being marketed to handle fluorocarbons as the working fluid.
Attempts to use fluorocarbon turbines on a commerical basis began
about 1955 in the United States and other developed countries like Japan and
the Soviet Union. A list of some of the more important fluorocarbon turbines
is presented in Table VII-2.
In addition to the end uses indicated in Table VII-2, several
other uses may materialize in the near future. These end uses can be placed
in the following categories:
1. Power Generation
a. Central station power plant
1. Use in conjunction with solar electric generating
plant.
2. Use in conjunction with geothermal electric gene-
rating plant.
3. Use in conjunction with steam turbines in a dual
cycle power plant.
162
-------
TABLE VII-1
Order of
Priority
LIST OF PRIORITY POLLUTANTS DEVELOPED AT THE INTER-
GOVERNMENTAL MEETING'ON MONITORING AT NAIROBI (1974)
Pollutant
Sulfur dioxide plus suspended partic- Air
ulates
Radionuclides Food
Medium
Ozone
DDT, Organochlorine Compounds
Air
Biota, Man
Cadmium and compounds
Nitrates and nitrites
NO and N02
Mercury and compounds
Lead
Carbon dioxide
Food, Man, Water
Drinking Water, Food
Air .
Food, Water
Air, Food
Air
6
7
Carbon monoxide
Petroleum Hydrocarbons
Fluorides
Asbestos
Arsenic
Mycotoxins
Microbial Contaminants
Reactive Hydrocarbons
Air
Sea
Fresh Water
Air
Drinking Water
Food
Food
Air
163
-------
TABLE VII-2
RANKINE CYCLE TURBINES IN OPERATION USING
WORKING FLUIDS OTHER THAN STEAM
Date of
Completion
1928-1950
1961
1966
Country
U.S.
U.S.
Israel
and
Africa
Location
Various
Trunkline
Gas Company
Houston,
Texas
Remote
Locations
Purpose
Topping Cycle
Pipeline Gas
Compressor
Water Pumping
(solar collec-
tors used for
Working
Fluid
Mercury
R-ll
Monochloro-
benzene
Scale of
Turbine
7.5-20 Mw
472 KW
600 w
1966
1967
1968
1968
1969
1971
Japan
USSR
U.S.
Japan
Japan
France
IHI
Kamchatka
Japan Gas
Chemical
Mizushima
Plant
Iwa taya
Department
Store
(Kyushu)
heat)
200RT Turbo-
Refrigerator
Drive
Geothermal
Generating
Unit
Demons tration
Purposes
Refrigeration
Power Genera-
tion for Chemi-
cal Plants
800RT Turbo-
Refrigerator
Drive
Radar Relay
Power (radio-
isotopic heat
source)
R-ll
R-12
190 KW
750 KW
Dowtherm A 6 KW
R-ll 3,800 KW
R-ll
475 KW
Monochloro- 4 KW
benzene
164
-------
TABLE VII-2 (Concluded)
Date of Working Scale of
Completion Country Location Purpose Fluid Turbine
1975 U.S. Currently Electricity Toluene 100 KW
Being Mar- Generation
keted Around (fossil fired)
the Country
1975 U.S. Sandia, Labs Electricity- Toluene 40 KW
1 Albuquerque, Generation
New Mexico
165
-------
b. On-site power plant
1. Use in conjunction with solar electric generating
equipment.
2. Use with fossil-fired boilers for apartments, mo-
tels, hospitals, and shopping centers.
2. Transportation
a. Automobile power units
b. Mobile home and recreational units
c. Propulsion power for use with ships
3. Industry
a. Pipeline gas compressor
b. Use in a dual cycle unit for plant compressors
c. Use with solar collectors for pumping applications
B. Criteria for Selection of Working Fluids
The choice of a working fluid depends on many factors, some of
which have no thermodynamic significance. In order to make an intelligent
decision as to the best working fluids for a specific end use, it is neces-
sary to develop a set of selection criteria. The following criteria are
applicable to all of the end uses previously discussed.
1. Stability; Thermal stability is the first criterion in choos-
ing a working fluid. A practical working fluid must be stable against decom-
position under prolonged heating in contact with conventional construction
materials. Since thermal stability of a chemical substance diminishes with
increasing temperature, this factor is usually the limiting parameter in the
selection of a peak cycle temperature, and therefore limits the potential
energy conversion efficiency of the Rankine cycle.
The products of thermal decomposition form noncondensible gases
which accumulate in the condenser. This inhibits its performance and deter-
iorates the work output of the system. The acceptable rate of thermal decom-
position will depend on whether the decomposition products are corrosive,
and how easily they can be removed.
2. Flammability; The flammability of the working fluid is im-
portant because the turbine will operate near large boilers and high tem-
perature steam pipes. However, as the fluorocarbon system will normally be
166
-------
sealed, it is anticipated that this criterion can be relaxed slightly. Flam-
mability need not necessarily be a prohibiting factor provided that adequate
precautions are taken.
3. Corrosivity; Corrosion is one of the problems associated with
the design and maintenance of any power plant. The corrosive properties of
any new working fluid must be carefully evaluated before it is accepted for
power plant application.
A particular problem of corrosion may result if the new working
fluid should leak into the steam circuit where high purity is essential.
Another problem is the possible contamination of the working fluid by steam,
cooling water, or decomposition products. Because the turbine systems are
hermetically sealed, it is expected that the corrosion criterion is some-
what eased in that all moisture and air can be eliminated initially from
the system.
4. Freezing; For a system to operate satisfactorily over a wide
range of ambient temperatures, it is essential that the working fluid freeze
below the minimum ambient temperature encountered by a power system. In fact,
it should have a freezing^point well below the minimum ambient temperature
in order to minimize the startup power requirement of a system containing a
highly viscous fluid near its freezing point.
5. Critical temperature; It is essential to the operation of a
Rankine cycle turbine that the working fluid not be heated above the criti-
cal temperature. This requirement may be the limiting parameter in the se-
lection of a peak cycle temperature, and therefore limit the potential energy
conversion efficiency of the cycle. A theoretical treatment by Horn and
Norris-t' indicates that the minimum acceptable critical temperature is ap-
proximately 340°F.
6. Toxicity; The chosen working fluid in the Rankine cycle tur-
bine should, if possible, be nontoxic. However, mercury has successfully
been used as a working fluid in a number of power stations since 1928. In
a plant using a toxic working fluid, special precautions must be taken to
avoid leakage. In order to decide on a toxicity limit it is necessary to
compare the increase in the plant cost and difficulty of maintenance and
operation against the advantages of using the toxic fluid.
7. Turbine size; A reduction in turbine size is desirable in
order to reduce the cost and complexity of the turbine. This can be accomp-
lished if the working fluid has a low specific volume and a high latent heat
of vaporization. While there does not appear to be a specific size reduction
that is considered to be "right," a reduction in the exhaust area of at
least 4-5;l is desirable.
167
-------
An approximate idea of the relative machine size of different
fluids can be obtained from basic fluid propertiesi-t'
[M PC Lc3/2]steam
Exhaust area ratio = - -
[M PC Lc3/2]fluid
where M = Molecular weight
PC = Pressure of the vapor in the condenser
Lc = Latent heat of vaporization
8. Molecular weight; When a gas expands adiabatically, heat
energy is convered into kinetic energy. This can be presented by 1/2 Y£ =
AH (for a unit mass), where VL is the velocity of the fluid leaving the
turbine and AH is the heat drop. It can be shown thermodynamically that the
AH obtainable from a fluid is inversely proportional to the molecular weight
of the fluid. Thus, by choosing a working fluid of sufficiently high molecu-
lar weight, the exit velocity of the fluid can be reduced to the point where
velocity or pressure compounding is unnecessary to reduce the turbine blade
speed to practical limits. It is best if the molecular weight of the fluid
exceeds 100.
9. Boiling point; The boiling point of the working fluid has
a considerable influence on its performance in a turbine. The most important
influence is the disc and blade function. It can be shown that the friction
of the turbine is proportional to the vapor pressure in the condenser. There-
fore, the friction can be reduced by having a fluid with a low vapor pres-
sure, i.e., a high boiling point vapor. A cautionary note should be added:
the fluid should not have too low a vapor pressure to avoid high vacuums in
the condenser. A vapor pressure range of 1 to 15 psi is desirable.
10. Temperature - entropy diagram; The T-S diagram for water
shows that the slope of the vapor-liquid line is negative. This means that
when saturated steam expands isentropically, the steam becomes wet. If this
wetness exceeds a certain amount, it can cause erosion of the turbine blades.
Wetness loss is a purely practical issue not predicted by theory
and only discoverable in its amount by careful testing. It is not known
whether fluids other than steam exhibit wetness loss, but there is no rea-
son to suppose that they would not. The wetness loss in steam is about 1%
loss of stage efficiency for each 1% of water present. For a steam turbine
this loss amounts to approximately 6%.
It is desirable to select a working fluid which has a slightly
positive vapor-liquid line on the T-S diagram. This assures that the fluid
168
-------
expands into a dry region as it passes through the turbine. Both erosion of
the turbine blades and wetness loss will be decreased.
11. Number of atoms/molecule; The cycle efficiency of the Rankine
cycle can be improved if, by a judicious selection of working fluid and cy-
cle conditions, the turbine exhaust lies on the saturation curve (vapor-liquid
line). This offers the joint advantages of requiring very little superheat
to avoid wetness losses, and very little regenerative heating is necessary.
Consequently, fluids are desired which have a nearly vertical vapor-liquid
saturation line.
Theoretical studies have shown that the slope of the vapor-liquid
boundary is a function of the number of atoms in the molecule. Molecules with
a small number of atoms would have T-S diagrams similar to water, while mol-
ecules with a large number of atoms would have a strongly positive slope. A
vertical T-S diagram calls for a rather small number of atoms, 5 to 10. Com-
bining a small number of atoms with a high molecular weight requires .heavy
atoms such as chlorine or bromine.
C. Identification of Possible Alternate Working Fluids for Rankine Cycle
Engines
Having identified the possible end uses of Rankine cycle turbines
using working fluids other than steam, and the criteria for selection of
acceptable fluids, candidates for use in these engines were identified. A
large number of possible fluids were considered. However, where the pub-
lished literature indicated that a fluid was undesirable because of insta-
bility, corrosiveness, etc., it was rejected from further consideration.
Table VII-3 presents a partial list of viable candidates for use
in a Rankine cycle turbine. All of these fluids have been evaluated by other
researchers during various studies*iliZ'
The first group of six fluids represents the chlorinated fluoro-
carbons which are to be considered banned. These fluids were included to
serve as a basis against which to measure the properties of the alternate
fluids (second group). The third grouping consists of alternate halogenated
compounds. It is not entirely certain that these alternate halogenated com-
pounds offer any advantage over the first group. Careful testing of the sta-
bility of each compound will be required to determine the danger to the
ozone layer.
The properties of each alternate fluid were obtained from vari-
ous sources and may not reflect the latest available data. However, the num-
erical values are sufficiently accurate to permit use of the selection cri-
teria.
169
-------
TABLE VII-3
ALTERNATIVE WORKING FLUIDS FOR BANKINE CTCLE ENGINES
Fluid
Fll
.F12
F13
F112
F113
F114
Formula
CCljF
CC12F2
CC1F3
CC12F-CC12F
CC12F-CC1F2
CC1F2-CC1F2
Molecular
Weight
137.4
120.9
104.5
203.8
187.4
170.9
Boiling
Point
CF)
(1 atm.)
74.9
-21.6
-114.6
199.0
117.6
38.8
Freezing
Point
(*F)
-168.0
-252.0
-294.0
79.0
-31.0
-137.0
Critical
Temperature
(*F)
388.4
233.6
83.9
532.0
417.4
294.3
Liquid
Dem ley
( Ib/f t3)
92.1
81.8
81.1
102.1
97.7
90.9
Vapor
Density
(lb/ft3)
0.367
0.395
0.438
0.438
0.461
0.489
Latent
Heat of
Vaporization
(Btu/lb)
78
71
64
67
63
57
Slope of
Vapor Pressure
Curve
+
+
+
+
•+
+
TonieltT
5a
6
6
4-5
4-5
6
Plaams-
btlitr
A
A
A
A
A
A
Water
Ammonia
Ethane
Propylene
Propane
Butane
laobutane
Benzene
Blphenyl
Toluene
Ethyl Ether
Trlaethyliaobutene
KITS
Orthoxylene
Ethylene Oxide
Methyl Formate
Acataldehyde
Methyl Mercaptan
IthjUnlne
Dowthenn A
Carbon Dioxide
Sulfur Dioxide
Mercury
Fotaaalua
NH3
C3Hg
(CB3)3CB
C6H6
CjHjCBj
C(C2H5)2
(CR3)2CHCfl:C82
C15H16
BCOOCH3
CHjCHO
CH3SH
CjReNR
26.5*(C6H5)2/-
73.51
s°2
Hg
K
18.0
17.0
30.0
.42.1
44.0
58.1
58.1
78.0
154.2
92.0
74.1
70.0
196.0
106.0
44.0
60.0
44.0
48.0
45.0
44.0
64.0
200.6
39.1
212.0
-28.0
-127.5
-53.7
-44.2
31.3
10.0
176.0
492.6
230.0
94.3
172.2
560.0
291.2
51.3
89.2
68.4
42.7
61.9
495.8
-109.0
(76.4.
14.0
672.0
1,400.0
32.0
-107.9
-297.0
-301.0
-305.8
-211.0
-229.0
41.0
159.8
-139.0
-177.2
-165.7
-68.0
-16.6
-169.1
-146.2
-190.3
-189.4
-217.7
53.6
-69.6
pal)
-104.0
-38.0
147.0
705.0
271.4
90.0
197.2
206.3
306.0
' 273.0
552.0
961.0
605.4
381.0
341.0
675.0
381.0
418.0
370.0
361.0
369.0
930.0
87.8
315.0
2,800.0
3,951.0
62.4
42.6
7.6
2.2
6.8
5.8
37.6
54.8
54.1
49.9
44.5
44.0
54.9
56.0
61.2
48.9
53.7
57.6
66.0
97.6
91.2
845.2
44.6
0.037
0.056
0.129
0.147
0.145
0.167
0.175
0.145
0.176
0.192
0.048
0.031
1,050
589
212
191
182
166
158
17S
153
153
150
151
240
200
245
230
159
256.3
172.3
125
892
6
2
5b
4-5
Sb
Sb
5b
4-5
1
4-5
4-5
4
3
1
4-5
1
Sb
5a
1
1
1
B
B-C
B-C
D
C
B
-------
TABLE VII-3 (Concluded)
Fluid
Formula
F14
F21
F22
F23
F115
F116
PlUBj -
F-C318
FC-75
Dibrotaodlfluoro-
methane
Ethyl Chloride
Metbylene Chloride
Methyl Chloride
Methyl Bromide
Boron Trichloride
Hexaf luorobenzene
Monoch lor obenz ene
Tribronof luoro-
sllane
RC-1
RC-2
CF4
CHC12F
CHC1F2
CHF3
CC1F2-CP3
CF3-CF3
CBrF2-CBrF2
C4F8
CgFiftO
CBl^Fji
O2H5C1
CH2C12
ffljCl
CB3Br
BClj
C6F6
CftHsCl
Br-jFSl
Cff6. CjHFj
Molecular
88.0
102.9
86.5
70.0
154.5
138.0
259.8
200.0
416.0
210
64.5
84.9
50.5
95.0
117.2
186.0
112.5
289.0
175.3
44.3
Boiling
Point
CF)
(1 atm.)
-198.0
48.1
-41.4
-115.0
-38.4
-108.0
117.1
21.0
212.0
76.1
54.0
105.2
-10.8
38.2
54.5.
194.0
269.6
185.0
172
200
Freezing
Point
-299.0
-211.0
-256.0
-247.0
-159.0
-149.0
-166.8
-42.0
-76.0
-217.7
-142.0
-144.0
-136.6
-160.6
-67.0
-49
-112.0
-44
-40
Critical
Temperature
CF)
-50.2
353.3
204.8
78.0
175.9
67.0
418.1
239.0
-440.0
369.0
421.0
289.4
376.0
354.0
471.2
678.2
460.0
675.0
Liquid
Density
(Ib/ft3)
82.2
85.3
•74.5
42.0
80.6
99.0
135.0
94.0
110.5
57.6
82.8
62.6
108.1
89.5
100.3
96.7
61.5
Vapor
•Density
(lb/ft3)
0.476
0.285
0.295
0.291
0.522
0.562
0.601
0.209
0.162
0.376
0.904
Latent
Hest of
Vaporization
(Btu/lb)
58
104
100
103
54
50
45
50
35
52
159
142
180
110
67
140
79
378
Slope of
Vapor Pressure
Curve
CD
+
+
+
+
-
-
CD
- . "
CD
+
- -
Toxteity
6
4-5
5a
6
6
6
5a
5a
4-5
4-5
4
2
4-5
4-5
6
4-5
5b
Flam
bill!
A
A
A
A
A
A
. C
A
C
B
A
D
A
B
-------
Both the toxicity and flammability of each alternate working fluid
are presented in somewhat qualitative terms in Table VII-3. Table V1I-4 is
a semiquantification of the toxicity of each fluid based on the classifica-
tion used by Underwriters Laboratory. Table VII-5 is a definition of the
classification of the comparative flammability of various vapors.
D. Advantages and Disadvantages of Selected Working Fluids for Rankine
Cycle Turbines by End Use
The following discussion presents the relative advantages and
disadvantages of selected working fluids for Rankine cycle turbines by spe-
cific end use. It is extremely unlikely that one particular fluid will be
found which will satisfy all of the requirements for each application.
Tables VII-3 through VII-5 illustrate this point by showing that, although
some fluids may be ideally suited for a compact turbine or for heat exchanger
design, they are completely unsuitable on chemical grounds. This situation
suggests that the final selection of one particular fluid for each applica-
tion will be difficult. The decisions may well rest on features such as tox-
icity, flammability, or stability.
The conclusion reached as a result of this analysis is that the
"best" working fluid will be chosen based on the specific purpose of the
turbine, and on the turbine design. Therefore, it is not a useful exercise
to attempt to indicate the reasons for the rejection of all other working
fluids. Rather, the following analyses are generic in nature, and present
only the advantages and disadvantages of the most promising candidates for
selection within each end use category.
1. Power generation
a. Benzene; The advantages of benzene as a working fluid
in a Rankine cycle turbine are:
* Excellent stability.
* Low to moderate toxicity.
* Noncorrosive.
* Satisfactory low temperature startup capability.
* T-S diagram has a positive slope to the vapor-liquid line
(expands into dry region).
* Reduction in turbine exhaust area compared to steam tur-
bine is approximately 15.
172
-------
TABLE VII-4
UNDERWRITERS' LABORATORIES CLASSIFICATION OF COMPARATIVE
HAZARD TO LIFE OF GASES AND VAPORS
Group Definition
1 Gases or vapors which in concentrations of about 1/2 to 1% for
duration of exposure of about 5 niin are lethal or produce serious
injury.
2 Gases or vapors which in concentrations of about 1/2 to 1% for
durations of exposure of about 1/2 hr are lethal or produce
serious injury.
3 Gases or vapors which in concentrations of about 2 to 2-1/2%
for durations of exposure of about 1 hr are lethal or produce
serious injury.
4 Gases or vapors which in concentrations of about 2 to 2-1/2%
for durations of exposure of about 2 hr are lethal or produce
serious injury.
Between Appear to classify as somewhat less toxic than Group 4.
4 and 5
Much less toxic than Group 4 but somewhat more toxic than Group 5.
5a Gases or vapors much less toxic than Group 4 but more toxic than
Group 6.
5b Gases or vapors which available data indicate would classify as
either Group 5a or Group 6.
6 Gases or vapors which in concentrations
-------
TABLE VII-5
CLASSIFICATION OF COMPARATIVE FLAMMABILITIES
OF VARIOUS GASES AND VAPORS
Group Definition
A Nonflammable
B Small Flaimnability Range
C Large Flannnability Range
D Dangerous
174
-------
* Molecular weight is < 100, but greater than water.
* Satisfactory condenser pressure.
The disadvantages of benzene as a working fluid in a Rankine cycle turbine
are:
* Dangerously flammable - must use an inert gas system for
sealing turbine shaft, etc.
* High freezing point - this can be overcome if the turbine
is located inside a heated structure.
* Moderate value of critical temperature may limit use in
some design applications.
* Heat transfer properties are not as good as water - this
means that boiler and heat exchanger sizes are large.
* Maintenance on turbines, pumps, condenser, valves, etc.,
will have to be handled carefully due to the flammability
factor.
b. Toluene; The advantages and disadvantages of toluene as
a working fluid in a Rankine cycle turbine are similar to those of benzene
with the following changes:
* The freezing point of toluene is such that system freez-
ing is not a problem.
* The molecular weight of toluene is higher than that for
benzene.
* A slightly higher critical temperature permits higher
temperatures in the operating cycle.
c. Monochlorobenzene: The advantages and disadvantages of
monochlorobenzene as a working fluid in a Rankine cycle turbine are similar
to those of benzene with the following changes:
* Stability is not quite as good as benzene.
* The freezing point of monochlorobenzene is such that sys-
tem freezing is not a problem except in extreme environ-
ments.
* The molecular weight of monochlorobenzene is satisfactory.
175
-------
* A higher critical temperature permits higher temperatures
in the operating cycle.
* T-S diagram has a vertical slope to the vapor-liquid line
(expands along saturation line).
* Unacceptable costs at the present time.
d. FC-75: The advantages of FC-75 as a working fluid in a
Rankine cycle turbine are:
* Chemically inert.
* Excellent stability.
* Noncorrosive.
* Nonflammable.
* Very high molecular weight.
* Satisfactory low temperature startup capability.
* T-S diagram has a positive slope to the vapor-liquid line
(expands into dry region).
* Satisfactory condenser pressure.
* Satisfactory boiling and freezing temperatures.
The disadvantages of FC-75 as a working fluid in a Rankine cycle turbine
are:
* Low critical temperature may severely limit use in power
plant applications.
* Poor heat transfer characteristics may result in large
heat exchangers.
* Low heat of vaporization.
* Unacceptable costs at the present time.
e. Ammonia; The primary advantages of ammonia as a working
fluid in a Rankine cycle turbine is:
* Acceptable for use with low temperature cycles such as
are encountered in solar electric power plants.
176
-------
The disadvantages of ammonia as a working fluid in a Rankine cycle turbine
are:
* High toxicity.
* Low molecular weight.
* Reacts with moisture.
* Flammability.
f. Mercury; The advantages of mercury as a working fluid
in a Rankine cycle turbine are:
* Excellent stability.
* Nonflammable.
* Extremely high critical temperature puts no limit on the
cycle operating temperatures.
* High molecular weight.
* Boiling point and freezing point are acceptable.
* Extremely high cycle efficiency because of the high tem-
perature applications (topping cycles).
t
* Extensive operating experience with mercury as the working
fluid.
*
The disadvantages of mercury as a working fluid in a Rankine cycle turbine
are:
* Highly toxic - must prevent leakage from turbine shaft,
etc.
* Maintenance on turbines, pumps, condensers, valves, etc.,
will have to be handled carefully due to the toxicity
factor.
* Requires that austenitic steel be used to avoid corrosion
problems.
* Condensing temperature cannot be below about 400°F due
to vapor pressure considerations - this limits the use
in some power plant applications.
177
-------
* T-S diagram has a negative slope to the vapor-liquid line
(expands into a wet region). The terminal wetness at the
turbine exhaust is approximately 20%, thus causing signifi-
cant erosion of the turbine blades.
2. Transportation
a. Monochlorobenzene; The advantages and disadvantages of
monochlorobenzene as a working fluid in a Rankine cycle engine for use in
transportation are the same as in the discussion on power generation, with
the following exceptions:
* Monochlorobenzene is mildly toxic such that containment
following an accident must be considered during design.
* The extreme flammability of monochlorobenzene makes it
similar to gasoline, with the result that this factor
must be considered a potential problem.
* Mild steel is attached by monochlorobenzene at elevated
temperatures > 600°F.
b. Toluene; The advantages and disadvantages of toluene as
a working fluid in a Rankine cycle engine for use in transportation are the
same as in the discussion on power generation, with the' following exceptions:
* Toluene is mildly toxic such that containment following
an accident must be considered during design.
* The extreme flammability of toluene makes it similar to
gasoline, with the result that this factor must be consid-
ered a potential problem area.
«
c. RC-1 (60 mole % pentafluorobenzene - 40 mole % hexafluor-
obenzene); The advantages of RC-1 as a working fluid in a Rankine cycle
engine are:
* Excellent stability.
* Low to moderate toxicity (uncertain).
* Noncorrosive.
* Satisfactory condenser pressure.
* Nonflammable.
178
-------
* High molecular weight.
* The freezing point of RC-1 is such that system freezing
is not a problem except in extreme environments.
* T-S diagram has a positive slope to the vapor-liquid line
(expands into dry region).
The disadvantages of RC-1 as a working fluid in a Rankine cycle engine are:
* Low heat of vaporization leads to high flow rates.
* Low cycle efficiency leads to low miles per gallon.
* Unacceptable costs at the present time.
d. RC-2 (65 mole % water - 35 mole % 2-methylpvridine);
The advantages of RC-2 as a working fluid in a Rankine cycle engine are:
* Low to moderate toxicity.
* Satisfactory condenser pressure.
* High latent heat of vaporization.
* Only slightly toxic (similar to LP gas).
* The freezing point of RC-2 is such that system freezing
is not a problem except in extreme environments.
* T-S diagram has a positive slope to the vapor-liquid line
(expands into dry region).
* Higher cycle efficiency than RC-1 (higher miles per gallon).
* Lower cost than RC-1.
The disadvantages of RC-2 as aa working fluid in a Rankine cycle engine are:
* Moderate flammability (less than conventional fuels).
* Low molecular weight.
* Corrosive to mild steel.
179
-------
3. Industry
a. Monoch1orobenzene; The advantages and disadvantages of
monochlorobenzene as a working fluid in a Rankine cycle turbine for use in
pumping are the same as in the discussion on power generation.
b. MethyLene chloride; The advantages of methylene chlor-
ide as a working fluid in a Rankine cycle turbine for use in operating a
compressor are:
* Excellent stability at temperatures below 550°F.
* Low to moderate toxicity.
* Noncorrosive (particularly to aluminum).
* Satisfactory condenser pressure.
* Nonflammable.
* Molecular weight moderatley high.
* The freezing point of methylene chloride is such that
system freezing is not a problem.
The disadvantages of methylene chloride as a working fluid in a Rankine
cycle turbine are:
* T-S diagram has a negative slope to the vapor-liquid line
(expands into wet region).
* Moderate value of critical temperature may limit use in
some applications.
c. Toluene; The advantages and disadvantages of toluene as
a working fluid in a Rankine cycle turbine for either pumping or operating
a .compressor are the same as in the discussion on power generation.
d. Benzene; The advantages and disadvantages of benzene as
a working fluid in a Rankine cycle turbine for either pumping or operating
a compressor are the same as in the discussion on power generation.
180
-------
REFERENCES TO SECTION VII
1. McChesney, I. G., and R. H. Shannon, "The Use of Benzene as a Thermo-
dynamic Working Fluid for a Nuclear Power Plant," presented at ASME
Meeting, Paper No. 55-5-29, April 18, 1955.
2.: Tabor, H., and L. Bronicki, "Establishing Criteria for Fluids for
Small Vapor Turbines," presented at SAE National Transportation,
Power Plant, and Fuels and Lubricants Meeting, Paper No. 931C,
October 1964.
3. Ray, S. K., and G. Moss, "Fluorochemicals as Working Fluids for Small
Rankine Cycle Power Units," Advanced Energy Conversion, 6:89-102
(1966).
4. Horn, G., and T. D. Norris, "The Selection of Working Fluids Other
Than Steam for Future Power Generation Cycles," Chem. Eng., pp.
298-305, November 1966.
5. Luchter, S., "A Quantitative Method of Screening Working Fluids for
Rankine-Cycle Power Plants," presented at ASME Meeting, Paper No.
67-ST-12, March 5, 1967.
6. Wood, B., "Alternative Fluids for Power Generation." Proceedings of
the Institution of Mechanical Engineers 1969-1970, 184, Part 1,
No. 40. ~
7. Ichikawa, S., "Use of Flourocarbon Turbine in Chemical Plants." Chem.
Econ. and Eng. Rev., pp. 19-26, October 1970.
8. Angelino, G., and V. Moroni, "Perspectives for Waste Heat Recovery by
Means of Organic Fluid Cycles," J. Eng. for Power, pp. 75-83, April
1973.
9. Miller, D. R., H. R. Null, and Q. E. Thompson, "Optimum Working Fluids
for Automotive Rankine Engines," U.S. Environmental Protection Agency,
APTD 1563, June 1973.
10. Slusarek, Z. M., "The Economic Feasibility of the Steam-Ammonia Power
Cycle," Department of the Interior, Office of Coal Research and
Development, Report No. 7.
11. Wigmore, D. B., and R. E. Niggemann, "The Specification of an Optimum
Working Fluid for a'Small Rankine Cycle Turboelectric Power System,"
7th Intersociety Energy Conversion Engineering Conference 1972, pp.
303-314, September 25, 1972.
181
-------
12. Bronicki, L. Y., "The Ormat Rankine Power Unit," 7th Intersociety Energy
Conversion Engineering Conference 1972, pp. 327-334, September 25,
1972.
13. Niggemann, R. E., "3,000 Hour Endurance Test of a 6 KWE Organic Rankine
Cycle Power System," 7th Intersociety Energy Conversion Engineering
Conference 1972, pp. 288-293, September 25, 1972.
14. Adam, W. A., and J. Monahan, "100 Kw Organic Rankine Cycle Total Energy
System," 8th Intersociety Energy Conversion Engineering Conference
Proceedings, pp. 126-130, August 13, 1973.
15. Werner, D. K., and R. E. Barber, "Working Fluid Selection for a Small
Rankine Cycle Total Energy System for Recreation Vehicles," 8th
Intersociety Energy Conxersion Engineering Conference Proceedings,
pp. 146-151, August 13, 1973.
16. Paul, F. W., and N. A. Macken, "New Boiler Concepts for Advanced Auto-
motive Rankine Cycle Power Plants," 8th Intersociety Energy Conversion
Engineering Conference Proceedings, pp. 214-220, August 13, 1973.
17. Barber, R, E., "Small Rankine Cycle Total Energy System for Recreational
; Vehicles: A Comparison of Three Possible Approaches," 8th Intersociety
Energy Conversion Engineering Conference Proceedings, pp. 138-145,
August 13, 1973.
18. Jensen, C. E., D. W. Brown, and J. A. Miralbito, "Earthwatch," Sci.,
190^:432-438, October 31, 1975.
182
-------
VIII. DIRECT ECONOMIC CONSEQUENCES OF LIMITATIONS
ON CHLOROFLUOROCARBONS
This chapter focuses on the primary economic effects that would
occur if limitations on the use of selected chlorofluorocarbons necessi-
tated the adoption of alternative chemicals or devices. These consequences
include the direct costs to industries that currently use, or make products
that use, these chlorofluorocarbons, and the costs to the consumer that
purchase products or services that are currently based on chlorofluorocar-
bon technologies. Economic impacts will be analyzed for refrigeration,
aerosol, foam blowing and solvent cleaning applications of the selected
chlorofluorocarbons.
A. Refrigeration
Costs associated with the use of substitute chemicals and al-
ternative refrigeration systems are considered in the following sections.
1. Substitute refrigerants; Few plausible chemical substi-
tutes for R-ll and R-12 have been identified for use in mechanical vapor
compression (MVC) refrigeration and air conditioning systems. Ammonia,
sulfur dioxide, hydrocarbons and chlorinated hydrocarbons are technically
feasible substitutes for the above chlorofluorocarbons; toxicity, flam-
mability, materials compatibility problems, and the necessity for systems
redesign are likely to preclude any substantial conversion to the use
of such materials as refrigerants.
The most plausible substitutes for R-ll and R-12 as refrigerants
appear to be R-22 and R-502, even though designs and hardware for using
these refrigerants in home appliances and automotive refrigeration units
are not available at present. Redesign and retooling costs of converting
to R-22 for these end uses can be only crudely estimated.
A few major manufacturers of home appliances have given esti-
mates for the costs of redesigning and retooling their own production
lines varying from $10 to $50 million. When the number of production
lines operated by each of the manufacturers is taken into consideration,
this equates to about $10 million per production line. These numbers are
admittedly crude because of a lack of available design data, and because
of the reluctance of manufacturers to develop such data (they fear a pos-
sible future banning of the use of R-22 or R-502 in MVC refrigeration sys-
tems). While the costs of redesign and retooling appear to be quite high,
183
-------
they would ultimately be passed on to consumers in the form of higher
unit prices. Such costs might have an adverse short-term economic im-
pact, but little long-term economic impact on the refrigeration in-
dustry. Some of the smaller appliance manufacturers might be placed
in serious economic jeopardy by the need for redesign and retooling
capital. The appliance industry is very competitive, and operates on
a relatively small margin, a fact which would be to the further dis- •
advantage of the smaller manufacturers.
The conversion of existing R-12 appliances to R-22 would be
prohibitively expensive if not technically unfeasible. No estimates are
available for the cost of retrofitting existing units, but they could
be more per unit than its current value. If further changeover were made
through attrition, manufacturers would have to develop and maintain a
parts inventory for both types of units, over an estimated 20-year pe-
riod from the time R-12 was banned. Again, no estimates of the cost of
maintaining such an inventory are available.
The average refrigerator requires 0.65 Ib of refrigerant,
while a freezer requires about 1.25 Ib. Assuming that the average unit
charge of refrigerant would be the same as for R-12 and R-22, the added
cost of refrigerant per unit would be about $0.50 for a freezer and $0.25
for a refrigerator. However, since R-22 has a somewhat greater cooling
capacity than R-12, less refrigerant may be required—thus reducing this
cost differential. Such costs are nearly insignificant, assuming an av-
erage retail price of $300 to $400 for an appliance unit. It is apparent,
therefore, that a suitable substitute refrigerant costing 10 or 20 times
as much as R-12 would have only a small relative impact on the price of
appliances. The cost of recharging a defective unit in place with a higher
priced substitute would be proportionately, somewhat higher to the con-
sumer, but the substantial cost of the service call greatly reduces the
direct impact.
To convert future units from R-12 to R-22, one major home ap-
pliance manufacturer, who supplies over 20% of the appliance market, es-
timated that the redesign and retooling cost of $50 million for retool-
ing their plants would translate into an added cost of $5/unit to the
consumer.
Obviously, costs to some of the larger manufacturers would be
less than those to smaller manufacturers. Taking all factors into consid-
eration, we believe that an average appliance will increase in consumer
cost by $2 to $10—depending on the make and model. This range translates
to a 1 to 37« increase in price.
184
-------
Comparative energy efficiencies have been estimated for various
end-use products. Such efficiencies vary, however, with the intended use,
size, and other design parameters. For example, comparative refrigerant
performance per ton based on 5 F (-15 C) evaporation and 86°F (30°C) con-
densation indicates that R-22 requires a compressor horsepower of 1.011
as compared to 1.002 for R-12. Such a small change in energy efficiency
would have a far greater economic impact when accumulated over the life-
time of the product than would the initial unit cost. Higher operating
pressures and temperatures of R-22 units would also significantly reduce
reliability, and result in higher maintenance and repair costs.
Substitution of R-22 or R-502 for R-12 in mobile air condition-
ers would also require design and'retooling of production lines. One in-
dustry source estimated that an add-on auto air conditioner using R-22
would cost about 7% more than the current R-12 unit. It has been sug-
gested that low-horsepower automobiles may not have sufficient power to
handle R-22 refrigerant systems--indicating that energy requirements for
R-22 systems are greater than for R-12 systems, 'and again posing the pos-
sibility that long-term operating costs would be significantly greater
with the substitute.
Percentage increases in the manufacturing and consumer costs
for dehumidifiers, water coolers, beverage refrigeration, and similar
type units are estimated to be lower than those of the major home ap-
pliances, because existing applicable R-22 technology and hardware are
available. The costs for ice-makers, however, would be comparable to
those for refrigerators and freezers.
R-22 is currently used in some large cooling units. Consequently,
design and retooling costs should not be as great as for the home appli-
ances and other systems in which R-12 is used almost exclusively. However,
operating efficiency and unit reliability would be of even greater impor-
tance. No information was developed on relative operating efficiency of
such units. The higher operating pressures and temperatures would probably
again result in reduced reliability and higher repair costs.
2. Alternative refrigeration systems; Absorption refrigera-
tion systems could replace MVC units for many home appliance uses and
in many industrial applications, i.e., ammonia-water systems for low-
temperature operations and the lithium bromide-water system for higher-
temperature applications.
185
-------
Absorption refrigeration units currently account for only a
small fraction of these markets (3.3% of the appliance market—mostly
mobile home and recreation vehicles--and 1.2570 of the air conditioning
and chiller market). While the initial investment in such units is 50%
higher than for a comparable MVC unit, operating costs have been lower
primarily because of prices for natural gas (the energy source most com-
monly used for such units) have been unrealistically low. Recent rises
in natural-gas prices have reduced the operating cost differential be-
tween the two types of units and further rises are possible. Absorption
units deliver approximately 1.5 Btu of cooling per watt hour of energy
input compared to approximately 3.0 Btu of cooling per watt hour of en-
ergy input for MVC units. Thus, absorption refrigeration is only half
as energy-efficient as MVC refrigeration; however, under specific cir-
cumstances, the use of "waste" heat may provide an opportunity to re-
duce the energy input.
The poor energy efficiency alone will be a strong deterrent
to substitution of absorption for MVC refrigeration, particularly when
we consider that consumers typically pay $20 to $80/year for electricity
to operate a vapor-compres son refrigerator-freezer.
Absorption units have little or no anticipated utility for auto-
motive usage because of their size, unit weight, and higher energy require-
ments. They might have utility for large mobile units, but again energy
costs would be a strong deterrent.
Absorption units currently account for about 27» of sales of
chillers, based on tonnage (100 to 1,200 ton units), and cost about twice
as much as MVC units. Rising energy costs and the lower energy efficiency
of absorption units will be an even greater deterrent to their substitu-
tion for MVC systems in these larger units than in home-appliance applica-
tions.
Steam-jet refrigeration is used primarily where waste steam
is available. The unit is less expensive and more maintenance-free than
MVC systems. It is most useful for higher-temperature applications such
as chillers or air conditioning. This system is probably competitive
with MVC only in industrial applications.
Thermoelectric refrigeration, because of its high initial and
operating costs and its need for direct current, offers little prospects
for replacing MVC refrigeration systems in the foreseeable future.
The conventional air-cycle refrigeration systems used for air-
craft and remote military operations likewise offer little or no pros-
pect for replacing MVC refrigeration systems because of their inherent
low efficiency.
186
-------
A relatively new refrigeration system (the Rovac system), de-
scribed by its developer as a "multifluid mixed phase cycle which is a
hybrid of the reverse Brayton cycle and the reverse Rankine cycle," is
currently in the prototype stage. The system is claimed to be technically
feasible for a wide range of applications, including automobile air con-
ditioners, transport refrigeration, home appliances, residential air con-
ditioners and possibly commerical chillers.
Data on a prototype automobile air conditioner is said to show
that the unit compares favorably to the conventional automobile air con-
ditioner; it is only 327,, as heavy, is 1.7 times as efficient, and will
cost only two-thirds as much according to the manufacturers. Although
these claims are impressive, additional development and testing is needed
to demonstrate reliability and end-use applicability of this system.
B. Aerosols
Typical chlorofluorocarbon propelled aerosols use 0.3 to 0.4
Ib of propellant per unit. Of the many types of products dispensed as
aerosols, a few hair sprays, antiperspirants and deodorants, medicinals
and pharmaceutical, colognes and perfumes, pan sprays, and a few other
specialty items rely almost exclusively or chlorofluorocarbons as propel-
lants.— In all but a few cases alternative methods of delivering these
products are available and on the market. These include the use of car-
bon dioxide and hydrocarbons as the propellant and mechanical applicators.
While mechanical delivery systems do not produce identical re-
sults to the chlorofluorocarbon-propelled aerosol products, they are gen-
erally less expensive and have obtained a degree of consumer acceptabil-
ity. Table VIII-1 shows the types of products predominantly propelled
with chlorofluorocarbons, the number of units sold annually, the costs
of the delivery systems (excludes cost of active ingredient concentrate
and solvents), the costs of using F-22 or FC-318 as propellants and es-
timated costs of viable alternative delivery systems. Candidate substi-
tutes F-22 and F-C318 currently cost about 1.5-times and 10-times as
much, respectively, as F-ll or F-12.
The number of applications of the product per unit package
provided by the various delivery systems are uncertain. Therefore, the
cost data presented in the table assumes equal numbers of applications
fbr the aerosol and mechanical applicators, although it appears that
most mechanical delivery systems would provide a substantially larger
number of applications per unit, thus making them even more economically
favorable per unit application.
187
-------
TABLE VIII-1
COMPARATIVE COST OF ALTERNATIVE DELIVER? SYSTE
,3-5/
(Excludes Cost of Active Ingredient Concentrate)
Millions of Percent Propelled Average Weight Estimated Cost Average Cost
Product Types
Personal Products
Hair Care
*~* Antiperspirants and
QQ Deodorants
Medicinal and
Pharmaceutical
Colognes and Perfumes
~
Pan Sprays
Units Filled by Fluorochloro-
(1974) carbons
460 95
595 95
70 > 90
145 95 .
d/
302 95-98
Propel lant/Unit of F-ll or F-12
(Ib)
0.40
0.40
0.32
0.05
0.52
Delivery System
39*
39*
36*
15*
e/
49*^'
of Carbon Dioxide
Delivery System
28*
b/
28*2'
26*
_ I
U**'
Applicability
Unknown
Estimated Cost Estimated Cost'
of F-22 of FC-318
Delivery System Delivery System
41* 173* .
43* 192*
46* 140*
15* . . 36*
48* 278*
Cost of
Mechanical
Delivery
Systems'
(1) 26*
(4) 35* .
(1) 27*
(2) 11-12*
(4) 35*
(1) 26*
(3) 8*
(4) 35*
(1) 27*
(4) 35*
(1) 26*
(4) 35*
,b/
£/
_d/
_e/
The specific mechanical system(s) viable as alternatives are coded as follows:
(1) Mechanical spray pumps;
(2) Sticks, creams, and roll-ons;
(3) Collapsible tubes; and
(4) Bladder system, high pressure inner liner. Cost includes container system, valve, and cap where applicable.
C02 is a viable propellant only for deodorants.
Feasibility unknown with glass containers; price estimate assumes use of present containers.
MRI estimate based on data from Reference 1 and communication with an Industry source (2/76).
F-114 serves as the propellant in pan sprays.
-------
Chlorofluorocarbons may afford distinct advantages over other
delivery systems for a few products, primarily medicinals and pharma-
ceuticals. Alternative propellants such as those approved for food use
by FDA (F-115 and FC-318),— although more expensive, and F-22 could
be substituted for F-ll and F-12 for such product types. In view of
the small volumes of chlorofluorocarbons consumed, their continued use
in medicinals and Pharmaceuticals should be considered.
Also of economic importance is the cost of converting aerosol
filling equipment now using chlorofluorocarbons to alternative propel-
lants or mechanical delivery systems. The cost of converting to alter-
native fluorocarbons or chlorofluorocarbons that have about the same
thermodynamic properties as F-ll and F-12 should be negligible.
Conventional filling lines can also be converted to carbon
dioxide pressurizing units with add-on equipment at a cost estimated
at $7,000^' to $50,000 per line.-' Most industry estimates of the ad-
ditional equipment cost per line were approximately $10,000. One com-
pany which has converted a chlorofluorocarbon filling line to carbon
dioxide reports a daily savings of $3,000 because of the lower pro-
pellant cost*2' Thus, the cost of add-on conversion equipment can be
rapidly recovered. Conventional fill lines can also be adapted to fill
a bladder delivery system at a cost of not more than $10,000.—
C. Foam-Blowing Agents
The use of viable substitutes for the chlorofluorocarbon blowing
agents in foams, and hence any economic evaluations, is very difficult
to assess as delineated earlier in Chapter V. Chlorofluorocarbons impart
specific properties or meet specifications that no other known substitute
can duplicate, so that any suggestion of.viable substitutes and the resul-
tant economic considerations would be based on the assumption that such
alternative systems will provide a foam with identical or acceptable prop-
erties. In many instances this assumption is, at best, rather tenuous.
In very general terms, if all 78 million pounds of the current
chlorofluorocarbons used in foams were eventually replaced by alternative
blowing agents costing three times the current blowing agent price, the
overall increased cost would total $62 million. At the current annual out-
put of 898 million pounds of chlorofluorocarbon-blown foams, the added
cost to the user would be about $0.065/lb. For a typical flexible foam,
this would amount to an approximate 16% price increase. This example is
for illustrative purposes only as no single material is known that could
serve as an across-the-board replacement for chlorofluorocarbons. The prin-
cipal utility of this example is to place emphasis on the fact that the
cost of the blowing agent represents a minor factor with respect to total
raw material cost.
189
-------
In 1974, approximately 457» of all flexible urethane foam and
about 88% of all rigid urethane foam utilized chlorofluorocarbons as
blowing agents. These two applications accounted for 90% of the total '
utilization of chlorofluorocarbons as blowing agents.
In flexible urethane foams, carbon dioxide and methylene chloride
can be substituted as blowing agents in many noncritical applications. For
carbon dioxide-blown foams, the cost of the foam would show an approxi-
mate 15% increase in cost due to the necessity for increased isocyanate
consumption in the foam developing process. As discussed earlier in Chap-
ter V, Dow Chemical Company has stated that flexible urethane slab foams
can be prepared, using methylene chloride as a blowing agent, with approxi-
mately the same foam properties as produced with F-ll. The cost of the
methylene chloride blowing agent is approximately one-half that of F-ll
and the resultant foam cost, based on raw materials, is $0.365/lb as op-
posed to $0.372/lb for the chlorofluorocarbon-blown foam. Urethane grade
methylene chloride from other suppliers would likely produce similar re-
sults. The use of other blowing agents for flexible foams has not been
developed to a sufficient state of technology to predict their successful
application and, hence, the quantity of blowing agent necessary to pro-
duce an acceptable foam. If it is assumed that the concentration of these
materials would be approximately the same as the current blowing agents,
then the cost differential would be expected to be very small since, at
current prices, the F-ll blowing agent represents only about 4% of the
total raw material cost. If the use of substitute blowing agents led to
an increase in the consumption of isocyanate, then the final cost of the
foam would show a marked increase in cost per pound.
In rigid urethane foams, the chlorofluorocarbons are considered
essential in achieving the low thermal conductivity required for insulat-
ing foams. While a few other materials (mostly chlorocarbons) are known
that compare with the low thermal conductivity of the present blowing
agents, these materials are not currently used in the preparation of in-
sulating foams and it is not known if they could produce a foam with ac-
ceptable physical characteristics. It is possible that compounds, other
than the chlorofluorocarbons now used almost exclusively for insulating
foams, could be developed for use as auxiliary blowing agents; however,
to date no satisfactory substitute has been found.
D. Degreaslng and Dry Cleaning
In 1975, an estimated 60 to 68 million pounds of F-ll, F-113
and F-113 blends and azeotropes were used in cleaning and drying appli-
cations, including dry cleaning. Using an average weight of 11.6 ib/gal,
calculated from F-113 and six of the common blends and azeotropes, this
represents approximately 5.2 to 5.9 million gallons of solvent. Current
190
-------
suggested retail price data, based on tankcar quantities (30,000 Ib), are
presented below for suitable chlorocarbons, F-113, and the six blends or
azeotropes. In this listing, the F-113 mixtures are products of E. I.
du Pont de Nemours and Company, Inc. Similar products produced by other
manufacturers, such as Allied Chemical, would be competitively priced
with those shown in the list. The composition of each of these blends or
azeotropes was stated previously in Section VI.
j'roduct Price per Gallon
1,1,1-Trichloroethane $2.12
Trichloroethylene 2.155
Methylene chloride 1.95
Perchloroethylene 2.22
Alkaline cleaners 2.00
F-113 (100%) 6.53
TMC 4.80
TA 5.69
TE 6.28
T-E 35 4.73
T-WD 602 6.30
T-P 35 4.66
Since a compilation of the total quantity of each of the F-113
products consumed was not available, an average price of $5.57/gal was
used to calculate a current market of $29 to $33 million in 1975. The
possible chlorocarbon and alkaline cleaner substitutes show an average
price of $2.09/gal, and have a market value of $10.9 to $12.3 million.
The market differential would be approximately $18 to $21 million.
For those applications in which chlorocarbons may be considered
as substitutes, however, additional costs would likely be incurred. Due
to their higher toxicity, vapor-control systems may be required. It has
been stated in the Arthur D. Little report- that vapor-control systems
may initially cost from $50 to $9,000 per degreasing unit, depending upon
the specific system, with yearly operating costs in the range of $150 to
$400/unit. However, the vapor-control systems would decrease the amount
of solvent loss and result in some cost savings. Aside from vapor-control
systems, additional equipment such as solvent baths, spray systems, water
baths, or oven dryers may be necessary dependent upon the specific solvent
degreasing application. In addition to equipment costs, added costs may be
incurred including higher labor costs resulting from multistep cleaning
processes instead of the current single step process, higher incidence
of incompletely cleaned products which must be returned to the cleaning
191
-------
process, additional solvent cost since most multistep cleaning processes
use two or more different solvents, and other costs unique to the specific
cleaning application. The total additional costs cannot be realistically
estimated since such costs would vary considerably depending upon each
specific application.
For dry-cleaning applications, perchloroethylene and petroleum
solvents (e.g., 140-F) would be the possible substitute for specialty clean-
ing applications. While both of these solvents are considerably cheaper in
price than the F-113-based "Valclene" solvent, the greater solvent mileage
obtained with "Valclene" results in approximately equal costs.— Perchloro-
ethylene can be used on leather products only if they have been specially
treated in the finishing process. Petroleum solvents, such as 140-F, may
be used to clean furs but, as stated in Section VI, these solvents are
somewhat restricted with regard to cleaning establishments.
The area of dry-cleaning applications is very small and the use
of substitutes would probably have little overall economic effect unless
the dry-cleaning establishment was entirely "Valclene"-based. In this
case, the economic impact would be dependent upon the financial stability
of the individual establishment.
192
-------
REFERENCES TO SECTION VIII
1. Arthur D. Little, Inc., "Preliminary Economic Impact of Possible
Regulatory Action to Control Atmospheric Emissions of Selected
Halocarbons," EPA Contract No. 68-02-1349j Task 8, Publication
No. EPA-450/3-75-073, NTIS No. PB-247-115, September 1975.
2. Title 21, Code of Federal Regulations, 121.1065, 121.1181, and 121.101,
Supplied by John E. Thomas, Assistant to the Director, Division of
Regulatory Guidance, Bureau of Foods, Food and Drug Administration.
3. Howard, D. K., P. R. Durkin, and A. Hanchett, "Environmental Hazard
Assessment of One and Two Carbon Fluorocarbons," EPA Contract No.
68-01-2202, Technical Report No. TR-74-572, EPA Report No.
560/2-75-003, NTIS No. PB-247-419 (1974).
4. Personal communication with the American Can Company, Shawnee Mission,
Kansas, November 1975.
5. Modern Packaging Encyclopedia and Planning Guide. Vol. 47, No. 12,
December 1974.
6. Personal communication with Mr. Charles S. Hayes, Applications Engi-
neer, Chemetron Corporation, Chicago, Illinois, September 1975.
7. Confidential communication with industry source, September 1975.
8. "An Aerosol Filler's Experience with C0~ as a Propellant," Soap/
Cosmetics/Chemical Specialties, pp. 94 to 96, October 1975.
9. Confidential communication with industry source, September 1975.
193
-------
APPENDIX A
CHLOROFLUOROCARBON MANUFACTURING PROCESSES
195
-------
Any significant reduction in the production of fluorocarbons* will
have an impact on the employment and sales of producers of the fluorocarbons
and their principal precursors. The precursors are shown diagrammatically in
Figure A-l. The chemical reactions behind this diagram are listed in Table
A-l. The precursors considered in this analysis are hydrofluoric acid used
in the production of all fluorocarbons; carbon tetrachloride used in the
production of F-ll, F-12 and F-13; and carbon disulfide used to produce some
of the carbon tetrachloride. The rest of the carbon tetrachloride is produced
as a co-product with perchloroethylene by the chlorination of propylene.
Chloroform is produced by the chlorination of methanol. Chloroform is also
considered in this analysis, as it is used to make F-22, which is assumed
to be a permissible substitute for F-ll and F-12 in refrigeration. Perchlor-
oethylene, from which F-113 and F-114 are produced, is not included in this
analysis because f luorocarbons consume only about 87,i' of the perchloroethy-
lene production.
Fluorspar, from which hydrofluoric acid is produced, is also ex-
cluded from this analysis because most of the fluorspar consumed in the
United States is imported (97%), primarily from Mexico. Fluorspar currently
carries a high import duty, while hydrofluoric acid carries none. As a re-
sult, any new facilities built to produce hydrofluoric acid for U.S. con-
sumption will probably be built outside of the U.S. to avoid the tariff*—'
A tariff adjustment might be considered as part of a package to ameliorate
the impacts of reduction in fluorocarbon production.
Capacity, production, and employment estimates for fluorocarbons
and their principal precursors are presented in Table A-2. These figures do
not include the new Du Pont "Kel-chlor" facility currently in operation in
Corpus Christi, Texas. Du Font's plans show the completed facility rated
at 500 million pounds of fluorocarbons annually, which represents almost
a 50% increase over present U.S. capacity.
In 1973, about $275 million worth of fluorocarbons were produced
in the U.S. This figure includes F-ll, F-12, F-13, F-21, F-22, and F-113
and F-114. Direct and indirect employment supported by fluorocarbon produc-
tion n'as been estimated at about 6,200 (see Table A-2). Carbon disulfide
production comes under SIC-2869, Industrial Organic Chemicals (2869 is the
SIC for fluorocarbons and chlorocarbons as well). The 1972 Census of Manu-
facturers shows about $90,000 of product produced in that year for each
employee in SIC-2869. On the basis of figures, fluorocarbon-related employ-
ment in carbon-disulfide production can be estimated at about 100.
In this section, the term fluorocarbon is intended to denote chlorofluoro-
carbons.
196
-------
CCL4
F-ll, F-12,
F-13, F-14
C3H6 CI2
T I
CH3OH HCI CI2
Chlorination
of Propylene
CCI,
Cl,
H2O
HCI
C2CI4
F-113, F-114,
F-115, F-116
T
Chlorination
of Methanol
HF
•CH3CI
•CH2CI2
CHCN
F-21, F-22,
F-23
CHLOROFLUOROCARBONS
Figure A-l - Fluorocarbons and Their Principal Precursors
197
-------
TABLE A-l
THE CHEMISTRY OF FLUOROCARBONS AND THEIR PRINCIPAL PRECURSORS^'
Fluorocarbons from
CC14 +
nn'l T? j
Uv»J-«r \
CC12F2
ppl -I? J
HF .
h HF
+ HF .
1- HF
Carbon Tetrachloride
^ CCloF + HCl
— > CC12F2 + HCl
•v CC1F + HCl
^. CF/ + HCl
F-ll
F-12
F-13
P-U
Fluorochlorination of Methane (this method is not used commercially in
the U.S.).
CH4 + HF + C12 > CHC12F F-21
+ CHC1F2 F-22
+ CHF3 F-23
+ CC13F F-ll
+ CC12F2 F-12
+ CC1F3 F-13
+ CF4 + HCl F-14
Fluorocarbons from Chloroform
CHC13 + HF 5» CHC12F + HCl F-21
CHC12F + HF > CHC1F2 + HCl F-22
CHC12F + HF > CHF3 + HCl F-23
198
-------
TABLE A-l (Continued)
Fluorocarbons from Perchloroethylene
C2C14'+ 3HF + C12 > C2C13F3 + 3HC1 F-113
C2C13F3 + HF > C2C12F4 + HC1 F-114
• C2C12F4 + HF > C2C1F5 + HC1 F-115
C2C1F5 + HF > C2F6 + HC1 F-116
Carbon Tetrachloride and Perchloroethylene from Propylene
7C12 > CC14 + C2Cl4 + 6HC1 ,
Chloroform from Methanol
CH3OH + HC1 ^ CH3C1 + H20
CH3C1 + C12 > CH2C12 + HC1
CH2C12 + C12 > CHC13 + HC1
Carbon Tetrachloride from Carbon Disulfide
CS9 + 3C17 7 > SoCl, + CC!A Carbon Tetrachloride
i L catalyst l 2. '*
CS2 + 2S2C12 > 6S + CC14
Carbon Disulfide
C + 2S - > CS2
(in conjunction with carbon tetrachloride production)
Or CH4 + 4S - > ,CS2;+ 2H2S
carbon disulfide
199
-------
TABLE A-l (Concluded)
Hydrofluoric Acid From Fluorspar
CaF2 + H2S04 > CaSO^ + 2HF Hydrofluoric Acid
a/ The principal precursors described in this table are those with more
than 10% of their present U.S. production devoted to fluorocarbon
production.
b_/ Sources: Noble, H. L., "The Chemical Marketing Research Association
Review and Forecast," CMRA Paper No. 766, May 1972.
Faith, W. L., D. B. Keyes, and D. L. Clark, Industrial
Chemicals, 3rd ed., John Wiley and Sons, Inc., New York
(1965).
Personal communication with chlorofluorocarbon manufacturer.
200
-------
TABUE A-2
FLUOROCARBON AMD FLUOROCARBON RELATED PRODUCTION AMD EMPLOYMENT
Product
F-ll and F-12
F-21 and F-22
F-113 and F-114
Subtotal Fluorocarbons
Carbon Tetrachlorlde
Fron CS2
From Chlorlnatlon of
Methane
Total
Chloroform from Chlorlna-
tion of Methane
ro Carbon Dlsulflde
O
Hydrofluoric Acid
Subtotal for Principal
Precursors
. Total
Indirect Employment^'
' Million
Capacity—'
97oV
160fe/
686'
1.198*'.
650*'
928
1,578*'
293e/
850S' •
814*'
of Pounds Per Year
Quantity Percent of 1973
Consumed In Production Price
Fluoro- Denoted to Per
1973 carbon Fluoro- Pound
Production Production carbons (?)
834V 0.19 (F-11)V
0.24 (F-12)
13&V 0.45
584' 0.55S'
1,028
3538'
694
1,047V i>047 99. 9l/ . 0.06
253^' 205 81. OS' 0.07
77*1' 194 25. Ol/ O.OSS'
73lV ' 305*' 41.6
1973 Dollar
Volume of
Production Estimated
(millions) Employments'
180.7
61.2
31.9
273.8 2,100^'
<-
62.8 611
17.7 172
38.8 43 lB'
135. 4i' 800*'
254.7 2.014
Direct and Indirect Employment Supported by Fluorocarbon Productions'
Employment
Supporting
Fluorocarbon
Production
2.100E'
6113'
13*3'
108
330*'
1,188
3,288
2,897
6,185
a/ Capacity does not include Du Font's new Kel-chlor facility under construction In Corpus Christ!, Texas. Preliminary plans show the eventual capacity
of this plant to be 500 million pounds of F-ll, F-12, F-13, F-21, anS F-22 annually.
b/ MRI estimate based on Q/t d_/, and e/
c.1 ADL Table V-8.
d/ ADL Table V-6. r,
e/ SRI 1975 Directory of Chenical Producers. ft
ij Figures from e_/; includes some produced by chlorlnation of methane.
a/ 1972 production estimated from the CS2 production at 1,100 Ib of CS2 producers 1 ton of CC14- W. L. Faith, Donald B. Keyes, and Ronald L. Clark (1965),
Industrial Chemicals. Third Edition, p. 229.
h/ U.S. International Trade Consnlssion (1975), U.S. Production and Sales of Miscellaneous Chemlca*ls for 1973.
i/ Ninety-five percent of carbon tetrachlorlde is used to produce F-ll and F-12. Essentially all of the remaining 57. la converted to perchloroethylene by
pyrolysis and used to make F-113 and F-114.
J/ 1972 production from the Chemical Marketing Reporter. July 24, 1972.
k/ ADt Table V-13.
II ADL Table V-12.
m/ Employrcnt estimate based on $90,000 of product per employee per year. This figure was derived from the 1972 Census of Manufacturers for SIC 2369,
Industrial Organic Chemicals.
n/ KR1 estimate based on current prices.
o/ Obtained fron "direct employment supporting fluorocarbon production," using a multiplier of 1.881 derived from the U.S. Department of Cotmerce (1968)
~~ Input-Output Structure of the I'.S. Economy: 1967 Cor Industry classification 27.01 Industrial Organic and Inorganic Chemicals.
£/ ADL Table 1-5.
-------
But this employment generates additional activity in the rest of
the economy. The Input-Output Structure of the U.S. Economy! 1967 (pub-
lished in 1974 by the U.S. Department of Commerce) measures these indirect
effects in 367 different industry categories. The numbers in these tables
for industry classification 27.01, Industrial Organic and Inorganic Chemi-
cals, indicate that each job in organic and inorganic chemicals supports
about 0.881 job in other segments of the economy, or that roughly 2,900
jobs altogether are supported by fluorocarbon production. Many of these
jobs will be located near the facilities which produce fluorocarbons and
their principal precursors.
These calculations indicate that approximately 6,200 people in
the U.S. today are employed directly or indirectly by the production of
fluorocarbons. This analysis is based on 1973 information, when about 1 bil-
lion pounds of fluorocarbons were produced.
A small change in production would probably have no identifiable
impacts on employment in the industry. On the other hand, if no fluorocar-
bons were made next year, employment would be seriously affected. Any sig-
nificant reduction in fluorocarbon production will create an economic impact
somewhere in between these two extremes. Assuming that the reduction in
fluorocarbon production is significant, alternative employment will be needed
for three chemical workers for each million pounds of fluorocarbons no longer
produced. If alternative employment is not found for these chemical workers,
an additional three jobs in other segments of the economy will be affected
for each million-pound reduction.
The companies producing both fluorocarbons and their principal
precursors are listed in Table A-3. It was assumed for the purposes of this
study that F-22 would be an allowable alternative to F-ll and F-12 in refrig-
eration. F-22 is made from chloroform, while F-ll and F-12 are made from
carbon tetrachloride. Chloroform and carbon tetrachloride are co-products
in the chlorination of methane. However, only a small percentage of the pro-
duction of chloroform is manufactured in this manner because the freedom
to adjust the ratio of the final products is limited. Most chloroform is
co-produced with methylene chloride by the chlorination of methanol. This
process allows more freedom in adjusting the ratio of the final products
according to market demand. Those facilities producing carbon tetrachloride
from carbon disulfide cannot make this adjustment. However, the facilities
producing carbon disulfide will not be as seriously affected, since the pro-
duction of carbon tetrachloride accounts for only 25% of the utilization of
carbon disulfide.
The locations of the facilities producing fluorocarbons and their
principal precursors and details on population and employment in each of
these areas arc shown in Table A-4. Because of economic considerations, some
of these production facilities will close as a result of any sizable reduc-
tion in fluorocarbon production.
202
-------
o
L--
TABLE A-3
COMPANIES PRODUCING FLUOROCARBONS AND THEIR^INCIPAL PRECURSORS
Company
Allied Chemicals
Alcoa
Diamond Shamrock
Dow
Du Pont
Essex Chemicals
FMC
Kaiser Aluminum
and Chemicals
Kewanee Oil
Lehigh Valley
Chemicals
Olin
Pennwalt
PPG
Racon
Stauffer
Union Carbide
Vulcan
Total
Carbon Tetrachloride
from from
Fluoro- Chloro- Chlorination Carbon
carbons form of Methane Disulfide Total
310 30 8 8
18
130 275 275
500S/ 500^
300£/ 300
50
115
20
75 70 350 420
200
40 75 75
1,195 293 928 650 1,078
Hydro-
Carbon fluoric
Disulfide Acid
224
110
200
22
180
100
36
10
26
10 50
60
600 36
- — —
850 814
a/ Does not include the new Kel-chlor facility under construction in Corpus Christi, Texas, which is
designed for another 500 million pounds per year.
b/ Includes carbon tetrachloride and chloroform.
c/ FMC produces some carbon tetrachloride from carbon disulfide and some from chlorination of methane.
Source: SRI 1975 Directory of Chemical Producers.
-------
TABLE A-4
LOCATION OF PRODUCTION FACILITIES FOR FLUOROCARBONS AND THEIR PRINCIPAL PRECURSORS
Population
% Change
1970 1960-1970
1.
.. -
2.
3.
NJ ,
0 4.
*-
5.
7.
8.
9.
10.
11.
12.
13.
14.
Alabama
Le Moyne Mobile SMSA
California
El Segundo Los Angeles/Long Beach SMSA
Antioch I san Franciso/Oakland SMSA
Pittsburgh)
Delaware
Delaware City I Wilmington, Delaware/ New
North ClaymontJ Jersey /Maryland , SMSA
Illinois
Joliet - Chicago SMSA
Danville - Vermilion County
Indiana
East Chicago Chicago SMSA
Kansas
Wichita - Wichita SMSA
Kentucky
Louisville - Louisville SMSA
Calvert City - Marshall County
Louisiana
Baton Rouge - Baton Rouge SMSA
Geismar - Ascension County
Cramer cy - St. James County
Plaquemine - Iberville County
376,690
7,040,697
3,108,026
499,493
6,977,611
42,570
6,977,611
389,352
826,553
20,3813.'
285,167
37,086^
19., 733-'
30,746-
3.7
16.6
17.3
20.5
12.2
1.7
12.2
2.0
14.0 ,
21. B^
24.0
32. 8^
X 7'4a/
2.7-'
Employment
7, Change
1970 1960-1970
93,178
2,826,565
1,267,643
192,061
2,852,017
16,856
2,852,017
149,169
321,180
7 , 143&/
102,577
10, 805S-/
4.976S/
8,018-
-12.3
8.1
17.8
39.9
13.6
5.4
13.6
18.0
25.7
40.6
30.6
45.4
15.3
3.1
Related Firm
Stauf fer
Allied
DuPont
Allied/Dow
Stauf fer
Allied
Olin
Allied
DuPont
Racon/Vulcan
DuPont/Stauf fer
Pennwalt
Allied
Allied/Vulcan
Kaiser
Dow
-------
TABLE A-4 (Concluded)
Population
Employment
N>
o
Michigan
15. Montague Muskegon/Muskegon Heights SMSA
New Jersey
16. Elizabeth I Newark SMSA
Newark .)
17. PaulsboroV Philadelphia, Pennsylvania,
Thorofajrej New Jersey SMSA
18. Deep Water - Salem County
New York
19. Niagara Falls - Buffalo SMSA
Ohio
20. Ceveland - Ceveland SMSA
Pennsylvania
21. Glendon Philadelphia, Pennsylvania
New Jersey SMSA
Texas
7, Change
1970 1960-1970
157,426
1,859,096
60, 346^
1,349,211
1,349,211
4,822,245
284,832
1,985,031
11,997
17, 831-'
5.0
10.0
2.8*>
3.2
3.2
11.0
6.8
40.0
3.3
7.5*/
% Change
1970 1960-1970
56,081
762,303
23.2033/
509,789
828,585
1,878,497
96,252
797,421
4,826
5..835S/
63.2
12.0
7.9
7.0
19.7
14.2
38. 7
69.5
9.0
13.3
Related 1
DuPont
Allied
Vulcan
Essex
Pennwalt
DuPont
Stauffer
Kewanee (
Lehigh 1
DuPont
Pennwalt/
DuPont
Dow
Alcoa
22.
23
24.
25.
26.
27.
28.
-- Corpus Christi Corpus Christi SMSA
Greens Bayou~V Houston SMSA
La Porte _J
Freeport - Brazoria County
Point Comfort - Calhoun County
West Virginia
Belle ~~1
Nitro \Charleston SMSA
South Charleston 1
So. Charleston - InstituteJ
Moundsville - Marshall County
Natrium - Marshall County
284,832
1,985,031
11,997
17, 831-'
229,515
13,560
37,598-
6.8
40.0
3.3
7.5*/
-9.3
-10.6 ,
-1.2^
96,252
797,421
4,826
5..835S/
82,127
4,896
1 3,224^'
38. T
69.5
9.0
13.3
-1.0
7.8
15.0
Diamond Shamrock
Allied
FMC
Union Carbide
Allied
Pittsburgh Plate Glass
a_/ County data.
Source: United States Census of Population, 1960 and 1970, General Population Characteristics, General Social and Econ. Characteristics.
-------
APPENDIX B '
THE ROVAC AUTOMOTIVE AIR CONDITIONING SYSTEM
Reprinted with permission, "Copyright® Society of Automotive Engineers, Inc.,
1975, All rights reserved."
206
-------
The Rovac Automotive
Air Conditioning System
Thomas C. Edwards
The Rovac Corp.
Automotive Engineering Congress and Exposition
Detroit, Michigan
February 24-28,1975
750403
207 i
-------
750403
The Rovac Automotive
Air Conditioning System
Thomas C. Edwards
The Rovac Corp.
THE VAST MAJORITY of conventional refrigeration and air
conditioning systems in use today embody the well known and
highly developed reversed Rankine (vapor compression) cycle.
Of much lesser use is the reversed Brayton or "dense air"
cycle. Until recently, the air cycle has been shown to be of
value only in restricted uses such as the thermal management
of turbine-powered aircraft.
The vapor compression cycle operates on the principle of
working fluid phase change to approximate isothermal heat
transfer at the source and sink conditions in order to approach
the reversed Carnot efficiency. The ideal reverse Brayton
cycle, on the other hand, does not have the thermodynamic
benefit of working fluid phase change. However, investiga-
tions of the rotary vane air cycle system performed by the
author, as well as the activities of other investigators, have
verified that the air cycle system, when employed with proper
hardware and accounting for properly natural moisture phase-
change effects, can produce exceptional performance.
The earliest development work (1)* verified that the inter-
nal thermodynamic performance of the rotary vane air cycle
system was quite promising, but that the system at that time
was harnessed by extremely low mechanical efficiency rotary
vane hardware. Thus, thermodynamic achievability of the
rotary vane air cycle system was proven, but the serious me-
chanical constraint in the form of mechanical friction was
also fully exposed. Continued development efforts, however,
have yielded effective means of increasing the mechanical
efficiency of rotary vane air cycle hardware, and therefore
has permitted greatly increased total performance.
As is well known in the design of rotary sliding vane hard-
ware, the most serious source of mechanical friction lies at
the interface of the vane tip and the stator housing. (2) Not
only does this tip friction vector induce a direct rubbing me-
*Numbers in parentheses designate References at end of
paper.
ABSTRACT-
The ROVAC air conditioning system, a new system that
employs air as the refrigerant, is a combination rotary com-
pressor/expander unit. A prototype has been modeled, de-
signed, fabricated, laboratory tested, and field tested in a full
size four door 1973 Dodge Coronet. The description of the
new system, the analysis, design and actual test results are
reported here.
The objective of the engineering program was to demon-
strate and prove the capability of the ROVAC system to ef-
fectively and efficiently air condition automobiles. The
prototype system installed in the Dodge Coronet produces
delivered cooling capacity on the order of one to one and a
half tons per thousand rpm and has produced delivered coef-
ficients of performance at relatively high humidity levels (150-
180 grains water per pound of dry air) rivaling the best de-
veloped conventional vapor compression air conditioning
systems.
While the present system reported herein has not reached
the levels of performance predicted by detailed computer
models, continued hardware improvement is facilitating ac-
tual performance very near the levels predicted to be prac-
tically achievable.
During actual in-car jury tests, the prototype ROVAC air
conditioning system brings the average passenger compart-
ment temperature from a thermally soaked condition of 107°F
down to 72° F in less than two min with five passengers at an
average road speed of 30 mph.
208
-------
Fig. 1 - First generation cam-guided circulator
chanical loss, but it also increases the "cocking" of the vane
inside the rotor slot. This serves to further intensify frictional
losses between the vane sides and the rotor slots.
Various approaches to eliminating the vane tip friction vec-
tor were considered; the embodiment that presented the best
promise for sliding vane machines involves cam-guided vanes
wherein the kinematics of the vanes are dictated entirely by
end-plate cam paths that engage in a rolling element fashion,
small solid protrusions from the vanes. Within this concept,
since the vane motion is dictated entirely by the cam profile
(independent of the stator profile), the rotating assembly and
cam profile can be designed such that the vane tip just clears
the stator wall during operation. This, of course, eliminates
all tip friction and wear, and thereby greatly increases the
mechanical efficiency and longevity of the unit. This concept
was transformed into hardware and is pictured as first em-
bodied in Fig. 1.
As shown in Fig. 1, the cam path is of the double-acting
type with small needle-bearing mounted polyamide rollers
fitted to the vanes. While this particular unit did not meet
the specified design tolerances, it did demonstrate the basic
mechanical fesibility of the concept and did produce a COP
in the vicinity of 1.0, almost twice the original Purdue ma-
chine. Fig. 2 shows this unit in operation. Since the fabrica-
tion and test of this first cam-guided machine, a considerable
amount of activity has been dedicated to analysis, design, and
fabrication of more sophisticated machines.
Preliminary analytical results involving predictions of actual
mechanical efficiency of rotary cam vane machines indicated
that mechanical efficiencies for both the compressor and ex-
pander sides could approach or exceed 90%. (This would in-
dicate a total mechanical efficiency for the air cycle machine
would be greater than 80%.) Subsequent in-depth detailed
computer-aided analyses recently performed by a United
States Air Force contractor (3) has predicted that even greater
mechanical efficiencies can be achieved with real machines.
Namely, compressor and expander mechanical efficiencies on
the order of 95% arc predicted with Coulomb coefficients of
friction of 0.1-0.05. Reasons for the somewhat higher pre-
dicted mechanical efficiency than might have been expected
are apparently couched in the following facts:
Fig. 2 - First cam-guided machine in operation-note formation of ice
crystals in output manifold
1. Very little mechanical motion takes place for a large
change in volume. (For example, the compression or expan-
sion process takes place in approximately one-tenth of a rev-
olution.)
2. At operating speeds, the vanes appear to float or oscillate
azimuthally wjthin the rotor slot (apparently due to the ab-
sence of the vane tip friction vector and the pressure-opposing
Coriolis acceleration component).
This last consideration is lending explanation to the very
small vane, wear that has been observed in the various tests
performed on the first cam-guided machine (Fig. 2) which
has logged in excess of 300 h and displayed maximum vane
side wear of 0.0003 in during this time.
The obvious item that must be considered when appraising
this type of rotary vane machine is that of leakage. The im-
portance of this consideration has been borne out during the
development of ROVAC air cycle machines. It can be easily
shown, for example, that vane tip to stator clearances exceed-
ing approximately 0.001 in will seriously inhibit the total
performance of the air cycle system at low rotor speed due to
the relatively long residence times of a given charge of air
within a vane segment. This problem received considerable
attention in design and fabrication, and was ultimately over-
come through refined rotating component design and en-
hanced non-rotating component fabrication techniques.
Upon verification of both the thermodynamic fesibility and
the mechanical operability of the cam vane machines, it be-
came apparent that ROVAC air cycle systems definitely had
the potential of being developed into numerous highly com-
petitive air conditioning and refrigeration products. However,
since that realization, another significant discovery was made
regarding the basic actual thermodynamic processes upon
which ROVAC hardware operate. This discovery and its
ramifications are discussed briefly in a subsequent section.
First, however, a brief review of the classical reversed Brayton
cycle is presented.
THE REVERSED BRAYTON CYCLE
The well-kn6wn conventional reversed Brayton cycle em-
ploys a single phase working fluid that undergoes four basic
209
-------
w
\
/Con
Heat Exchanger
Wfl
Compressor
\
Fig. 3 - Open reversed Brayton cycle system
processes: 1. compression, 2. heat rejection, 3. expansion,
and 4. heat acceptance. In the open-cycle form, the heat ac-
ceptance process occurs via cold mixing of the working fluid
which the ambient gas (usually air) and in the closed-cycle
form, heat is accepted through a heat exchanger. Fig. 3 sche-
matically presents the open-cycle form of the reversed Brayton
cycle.
As depicted in this figure, ambient air is drawn into the
compressor, compressed, pumped into a heat exchanger (in-
tercooler), pumped into the expander, expanded, and finally
emerges in a greatly cooled state. For the reversed Brayton
cycle, the coefficient of performance (COP), defined as the
cooling obtained divided by the energy provided, appears as:
COP
QL
F - F
comp exp
where:
QL = cooling capacity
ECO = power provided to compressor
exp = Power provided to expander
Under ideal conditions, with polytropic perfect gas behavior,
the cooling capacity of the system shown in Fig. 3 can be rep-
resented as:
QL = MCf
and the COP can be represented as:
COP = (r""1 - I)'1
where:
n = the polytropic process index
r = volume ratio
M = mass flow rate
C_ = specific heat at constant pressure
Study of these equations does provide considerable insight
into the ideal reversed Brayton cycle. However, when the
numerous real effects such as leakage, heat transfer, various
pressure losses, and other real effects are considered, the anal-
ysis becomes considerably more involved. For a discussion of
these matters, the reader is referred to References 1, 2, and 3.
The previous detailed analysis performed by the author con-
tained one significant assumption, that, when considered
more carefully, has proven to hold a key to a new thermody-
Expander namic cycle. Namely, it was assumed in previous analyses,
that atmospheric moisture naturally existing in the inlet air,
would not, due to the rapid process rates involved, condense
during expansion. Actual experimental data has proven this
^ to be an invalid assumption and further analysis has disclosed
that, properly employed, the actual process can greatly im-
prove the operation of ROVAC systems. Before proceeding
to this and other new thermodynamic considerations, how-
ever, consider the following qualitative description of the
system to see how it achieves the processes required to achieve
cooling.
THE ROVAC SYSTEM
Fig. 4 illustrates in a generalized fashion the substance of
the ROVAC system. In the open cycle form, the system con-
sists of three basic components:
1. Input-Output Duct.
2. Rotary Vane Combination Compression-Expander-
Circulator.
3. Heat Exchanger.
The input-output duct serves to provide a path for both the
incoming flow of warm air and outgoing'cold air. This ducting
system is equipped with various processing media such as fil-
ters, sound control, or moisture separation means, depending
upon the actual application of the particular system.
The circulator unit, which can be considered to be the heart
of the system, consists mainly of two items: the stationary
portion, termed the stator, and the rotating rotor-vane assem-
bly. The stator housing is also equipped with two endplates
which hold the rotor in its axial position and aid in contain-
ing the air that circulates through the machine. The rotor-
vane assembly consists of a cylindrical rotor equipped with a
series of sliding vanes. These vanes are arranged so that they
can slide inside the rotor slots as the rotor-vane assembly
rotates within the stator, and thereby maintain continuous
(2) near-contact with the stator wall.
The heat exchanger, also termed the intercooler, serves to
dissipate the heat generated during the operation of the sys-
tem. This heat exchanger is principally a simple externally
finned tube. For some applications, internal finning can also
(3) be used to increase the rate of heat transfer.
0)
210
-------
WARM AIR IN
INLET-OUTLET
DUCT
VANE SLOTS
INLET PORTS
OUTLET PORTS
HEAT EXCHANGER
INLET PORTS
STATOR
HEAT EXCHANGER
OUTLET PORTS
FLOW
COOL AIR OUT
HEAT EXCHANGER
FLOW
Fig. 4 - Basic ROVAC system
In order to understand how the system functions, consider
Fig. 5 which illustrates the system in cross-sectional schematic
form. Imagine air (any gas will work, however) entering the
system through the input duct at atmospheric pressure and
temperature. As the rotor turns counter-clockwise (in this
example), air is drawn through the inlet leg of the duct and
into the expanding inlet volume Vj. As rotor motion contin-
ues, the maximum inlet volume Vj is filled. The air which is
now trapped is then compressed by further rotor rotation to
V2- At V^ tne a'r h35 reached an elevated pressure and tem-
perature. The air is then pumped through the heat exchanger,
and it cools to a temperature near that of the surroundings,
thus transferring heat to the surroundings. (When the system
is being employed as a heat pump, the heat exchanger warmth
is transferred to the area being heated.) The cooler air then
emerges from the heat exchanger (still at an elevated pres-
sure) and enters the expanding volume segment, V^. Finally,
this air is expanded back to atmospheric pressure at Vj by
continued rotor motion. During this expansion process, a
large fraction of the work of compression is recovered, thereby
greatly reducing the air temperature. Continued rotor rota-
tion then forces the cold air through the output ports and
down the output leg of the duct. When the ROVAC system
is being employed as an air conditioner or refrigerator, this
cold air is circulated to the area being cooled.
THE EFFECTS OF MOISTURE
Early experience with experimental ROVAC systems dis-
closed that the actual physics and thermodynamics of the air-
water mixture was not well understood by the author. Even
at the time of this paper, some details of the process are not
fully comprehended. To begin the discussion of the effect of
moisture, consider on a qualitative basis what happens to the
natural ambient moisture as it flows through the cycle. As-
sume, for example, that the inlet humidity ratio is 100 grains
of water per pound of dry air. (Corresponding, for example,
to an inlet temperature of, say, 95°F, 40% relative humidity.)
During the compression process, the mixture of air and water
vapor can generally be considered to become even more super-
heated because of the higher temperatures occuring during
compression. (The higher pressures, of course, tend to lower
the superheat. However, the effect of temperature is domi-
nant. Whether superheat or saturation occurs, of course, de-
pends upon the actual compression process path.)
During the intercooling process, the pressure loss is quite
211
-------
WARM AIR IN
TEMPERING
VANE (OPTIONAL)
EXPANDER
SECTION
COMPRESSOR
SECTION
V,
\
V«
TO FROM
HEAT HEAT
EXCHANGER EXCHANGER
Fig. 5 - Generalized schematic end-view of the ROVAC system
small, (actual systems exhibit only about 0.3-0.75 psi drop)
but the temperature drops quite significantly; very near to
ambient temperature. Now depending upon actual condi-
tions, part of the condensation occurs inside the intercooler
because the relative humidity towards the low temperature
side of the heat exchanger will generally attempt to exceed
100%. (Under usual operating conditions, for example, ap-
proximately 30 grains of water would be condensed per
pound of dry air flowing through the intercooler.)
Next consider what occurs as the saturated moist air enters
the expander and expands back down to atmospheric pres-
sure. As expansion begins, both the pressure and temperature
of the mixture begins to drop. Since again the temperature
drop is thermodynamically more significant as far as the mois-
ture is concerned, the moisture begins to condense further.
When this occurs, the energy associated with the condensing
water is transferred to the surrounding air. This energy ex-
tracted from the vapor, (which caused condensation in the
first place) serves to increase the temperature of the air-water
mixture. This also, of course, increases its pressure over what
it would have been at the same volume with dry air. This
means that the expander will have to expand to a larger vol-
ume in order to reach the ambient pressure than it would have
if no energy was added to the air (remember, of course, that
energy was extracted from the moisture and that is why it
condensed). This results in a most important event: The re-
covered expansion work (/ y5 PdV) has increased. This effect
is portrayed in Fig. 6 where the work process is shown as a
poly tropic process. (Subsequently, the justification for con-
sidering this process to behave in a polytropic fashion will be
presented.)
An important additional consequence of the higher energy
recovery during expansion is that the final energy state after
expansion is lower. Of course, attainment of a low final en-
ergy state in the outlet flow is the fundamental goal of the
system. One additional important consideration is that
through the condensation of moisture at the intercooler,
more ac'tual energy is rejected by the system. This results in
a yet further reduction in total outlet enthalpy because the
expander inlet enthalpy begins at a lower state than the dry
air, if one considers the fact that moisture removed is one of
the requirements of most air conditioning applications.
The foregoing discussion yields the conclusion that natural
moisture is actually an important benefactor for ROVAC sys-
tems, in direct contrast to conventional vapor compression
units.
THE MIXED MULTI-PHASE POLYTROPIC PROCESS -
After some consideration, it was decided that the most expe-
dient approach to modeling the mixed multi-phase expansion
process would be to view it as a polytropic. The justification
of this assumption is couched in the fact that, insofar as the
air is concerned, the process is simply influenced by heat
transfer-an effect known to be modeled quite well by the
polytropic.
After deciding to model the process as a polytropic, the
problem is reduced to finding the correct polytropic index.
In the author's approach, the polytropic index of expansion
212
-------
I
ENERGY DIFFERENCE - EXPANSION
V
Fig. 6 - Expansion work recovery: wet vs dry air
of wet air is found by considering the amount of work pro-
duced during expansion from a consideration of the conserva-
tion of energy (first law of thermodynamics: work = Au) by
considering initial and final state point conditions. The same
work is then calculated from the quasi-static work integral
through the definition of the polytropic index. Since the
polytropic index appears in both expressions, it becomes an
iterative parameter. That is, using the polytropic index as a
parameter, the first law work is matched with the polytropic
work. This matching is done in an iterative fashion in a simple
computer program until the correct index causes both the
work and final state points to match.
Analytical results show that the polytropic index varies only
a small amount over an extremely large range in inlet humid-
ity ratio. The reason for this is that the largest reduction in
polytropic expansion index is brought about by the amount
of water vapor left to condense during expansion. Due to the
fact that the normal intercooler operating pressure is in the
vicinity of 43-50 psia with an outlet temperature in the range
of 80°-100°F, the air can support water vapor content be-
tween 52 grains and 70 grains. This relatively small variance
in expander inlet vapor content over an extremely large range
of inlet humidity ratio confines the polytropic index to the
vicinity of 1.24-1.28. This is obviously important because the
total power demand of the system will remain constant even
though the relative humidity changes markedly. This is also
important because the total expander temperature difference
will remain approximately constant throughout a large range
of ambient humidity. Of course, the delivered cooling capac-
ity will rise significantly with an increase in inlet humidity
ratio due to the high pressure moisture condensation. It is
thus apparent that the ROVAC air conditioning system ex-
hibits a very favorable characteristic in that the sensible cool-
ing remains approximately constant at all humidity levels.
This is, of course, in solid contrast with conventional Freon
systems where moisture decreases the sensible cooling delivery.
The approximate properties of the multi-mixed -phase ther-
modynamics were factored into a pre-existing detailed model
of the system. Results of the advanced computer program
indicated very significant gains in the predicted achievable
performance. While the current system test performance falls
short of the computer predictions, the actual performance of
ROVAC test systems now rivals the performance of high
quality factory-installed automotive air conditioning systems.
A comparison between predicted and actual performance ap-
pears in the following section.
213
-------
Fig. 7 - Model 308/45/9.5-4 circulator
ROVAC SYSTEM PERFORMANCE
Based upon design information provided by computer-aided
analysis, third and fourth generation anti-friction circulators
were designed, fabricated, tested, and partially developed.
Fig. 7 shows a photograph of the model 308/45/9.5-4 Circu-
lator (4th generation) and Fig. 8 shows in operation the model
209/45/84 machine.
Table 1 presents both predicted and actual data for a typical
test of the model 308 machine.
TEST FACILITY AND PROCEDURE - Fig. 9 shows a
photograph of the laboratory test facility employed to eval-
uate the performance of various prototype ROVAC air condi-
tioning systems. The facility consists of a 5 HP DC motor
driven by a SCR control and monitoring unit. This prime
mover system permits infinitely variable speeds and continu-
ous armature current readout. Due to the properties of the
DC driven motor, the delivered torque is linearly proportional
to the armature current. Therefore,-a knowledge of the arma-
ture current permits calculation of the output torque. Since
the test facility includes circulator RPM readout, and the drive
ratio is known, the input drive horsepower is easily deter-
mined.
In addition to power input determination instrumentation,
Table 1 - ROVAC Model 308/45/9.54 Performance
Geometry of CIRCULATOR:
Major Axis = 9.50 inches (24.13 cm)
Minor Axis = 6.717 inches (17.06 cm)
Active Rotor Length = 4.0 in (10.16 cm)
Ambient Operating Conditions:
Temperature = 110°F
Pressure = 14.69 psia
Relative Humidity = 57%
Parameter
Compressor Outlet'
Temperature
Pressure
Expander Inlet
Temperature
Pressure
Expander Outlet
Temperature
Pressure
Rotor Speed
HP Drive
Air Mass Flow Rate
Cooling Capacity
COP
Computer Predicted
Performance
English Units S.I.
316°F
47.75 psia
1*5.5°F
47.10 psia
9°F
14.65 psia
1510
3.875
451.0 Ibm/h
29,979 BTUH
3.04
Actual Measured
Performance
English Units S.I.
315°F
48.2 psia
lf8°F
47.5 psia
34° F
14.65 psia
1510
5.75
453 Ibm/h
23,284 BTUH
1.59
instantaneous heat exchanger inlet and outlet pressures are
indicated directly by Heise pressure gages. Ambient relative
humidity is provided by an Albion hygrometer. Temperature
readout is provided by an Ircon digital copper-constantan
thermocouple thermometer. In addition, two channels of
transient temperature and pressure measurements are provided
through a Hewlett-Packard Model 1704A oscillograph and
copper-constantan thermocouple probes. Sound level mea-
suring equipment in the form of a Triplett Model 370 sound
Fig. 8 - Model 209/45/8-4 circulator in operation
Fig. 9 • ROVAC system test facility
214
-------
Fig. 10 - ROVAC-Dodge installation
level meter permits measurement of the noise level produced
by the system. Mass flow rate is measured by conventional
pilot-static methods as well as a Roots type positive displace-
ment flow meter. (Flow tests, however, have proven that the
circulator mass flow volumetric efficiency is between 98-99%.)
The chamber which houses the test facility is equipped with
an electric heater, humidifier, and air conditioner? Prior to
the test, the chamber conditions are set through the use of
these items. When the desired chamber conditions are reached
and recorded, the test is begun.
Testing the unit is accomplished by simply bringing the rotor
speed to the desired level and recording the data indicated by
the instrumentation. Immediately after the test, various in-
ternal and external surface temperatures are measured and
recorded with a special miniature thermocouple probe. The
test unit is often then disassembled for visual and metrologjcal
inspections.
DISCUSSION OF PERFORMANCE - From Table 1, several
conclusions are immediately apparent. For example, the ac-
tual delivered COP of each unit rivals the COP of well de-
veloped Freon automotive air conditioning system which are
reported to produce delivered-to-car COP's on the order of
1.0. Also, it is obvious that while the actual ROVAC COP's
are quite good by current standards, they fall considerably
short of the performance predicted by computer simulator.
This data mismatch is discussed subsequently in a following
section.
THE DODGE CORONET ROVAC AIR CONDITIONING
SYSTEM - In April of 1973, Chrysler Corp. contracted with
The ROVAC Corp. to produce an appraisal ROVAC automo-
tive air conditioning system and install it on a Chrysler-sup-
plied Dodge Coronet. Fig. 10 is a photograph of the ROVAC-
Dodge installation.
In the Dodge system, air flows from either the car's interior
or the outside into the circulator, a model 308/45/9.5-4 ma-
chine. After passing through the compressor section, the
circulator air passes through the intercooler placed in front
of the radiator and then re-enters the circulator on the ex-
pander side. After expansion, the cold air emerges into the
car's existing ducting system through an insulated duct. This
•cold air then passes through the humidity control element
where it subsequently is mixed with interior or exterior air
and propelled through the existing air conditioning plenum
and registers to the passengers via the existing heater fan.
Condensed moisture leaves the automobile from the existing
condensate drain.
Tests performed in Florida on the ROVAC-equipped Dodge
Coronet yielded significantly positive results. The delivery
of cold air to the passenger compartment was nearly instan-
taneous and more than adequate cooling capacity was quite
apparent even at low road speeds under very hot and humid
conditions. A typical jury test was performed after thermally
soaking the automobile in 95°F ambient Florida sun for ap-,/
proximately three h wherein the interior car temperature rose
to 105°F. Five adult passengers (including one Chrysler En-
gineer) entered the car and the windows remained closed. The
Coronet was driven for approximately 10 min at 30 mph with
windows closed and no fan air. The interior temperature of
the car rose to 107°F and the system was engaged. Within
one minute and 53 s, the interior temperature of the automo-
bile dropped to 72°F. Engine speed averaged about 1300
rpm (approximately 30 mph) throughout the slow speed cool-
down test. The idle condition cooling level was also judged to
be fully adequate. Bearing in mind that the Dodge Coronet
was the first automobile in history to be effectively air con-
ditioned through the use of atmospheric air as the refrigerant,
the prognosis for the future is promising.
GENERAL DISCUSSION: CONCLUSIONS
AND RECOMMENDATIONS
The major problem area in the design and fabrication of
ROVAC circulators is devising simple embodiments and tech-
niques that minimize mechanical friction while simultaneously
maximizing internal air sealing of the machine over a wide.
range of rotor speeds. This translates directly to three areas:
1. maximizing structural rigidity (especially the rotating com-
ponents), 2.. determining proper contour relations, and 3.
dimensional tolerance.
As might be expected, therefore, during the construction of
this machine, numerous fabrication problems were encoun-
tered, both in the non-rotating and rotating subcomponent
areas. At the outset, the most obvious serious fabrication and
difficulties were with the precision contouring of the stator
and the alignment and concentricity of the cam profiles. After
these problems were solved, attention was focused upon the
fabrication of the rotating elements. The problems associated
with this area consisted of maintaining good surface finish,
true radial element rotor slots and axle-centered vanes. Tech-
niques for fabricating these subcomponents were also devel-
oped which pave the way for mass production of the parts.
Once processes and techniques were developed which pro-
vided design-print hardware, the actual measured performance
of the prototype systems was seen to be, by current compari-
sons with conventional automotive air conditioning systems,
favorable. Nonetheless, the prototype systems' performance,
considering the computer performance predictions, has a
great margin for further improvement. Specifically, current
215
-------
9
automotive ROVAC performance is seen to be in the range of
about 1.3 at high specific humidity ratios.\ It is predicted that
this number can increase to the vicinity of 3.2; over three
times that of conventional laboratory system performance.
The present approach to optimizing ROVAC performance is
three-fold:
1. Instrumenting all internal Pressure-Volume and Tempera-
ture-Volume processes in order to determine precisely the ac-
tual port location requirements.
2. Further improvement of the mechanical design of all
Circulator subcomponents.
3. Isolate and minimize adverse heat transfer.
The first area of improvement is very important; present
port locations, while determined with a sophisticated com-
puter program, cannot be expected to be completely accurate.
This is due primarily to the fact that the physics and thermo-
dynamics of the multi-mixed phase processes are modeled only
on a simplified basis and that the actual cut-off point for port
flow is not precisely determinable because of the vane tip con-
tour.
The second area of improvement is also important and will
entail the design of more rigid vanes and rotors and trimming
in contours to account for thermally induced stator dimen-
sional changes.
In addition to the direct mechanical-alterations, attention
is also being given to optimizing the thermal flow within the
Circulator. Also, some activity has been directed towards
port noise abatement. Preliminary development approaches
have yielded large decreases in sound output. For example,
the 306 and 308 circulators produce approximately 105 db
(A-scale) at a distance of one foot without manifolds. Minor
port modifications on the 209 circulator reduced the level to
85 db without manifolds. Due to this rewarding result, the
port design will receive further attention. The goal for auto-
motive circulator sound generation is 75 db without mani-
folds and 65 db with manifolds. Actual tests within the
Dodge Coronet shows that no measurable sound is transferred
to the interior of the car through the ducts. However, noise
emanating from the manifolds in the engine compartment,
reaches the passenger area from road reflection and the hood
area. Measurements indicate that passenger area noise level
with the 308 circulator is approximately 6-8 db above the
ambient cruise noise level; obviously an unacceptable level.
While a model 209 circulator has not been installed in an auto-
mobile by writing time, the low db output from this machine
with manifolds indicates that noise levels, both by-pass and
interior, would be fully acceptable by present standards.
The humidity control module, which is basically a simple
filter element, has not seen long enough use to determine if a
long-term clogging problem will eventually develop. From an
optimistic viewpoint, however, there is a basis for belief that,
due to condensation, the filter will be constantly washed. If
this is not the case, periodic element replacement would be
required.
The only material area that the ROVAC automotive system
has not demonstrated absolute credibility is circulator longev-
ity. Most test data on the later generation machines are based
on individual operating runs of no more than 15 min. Total
individual logged operating time on the third and fourth gen-
eration machines does not exceed 10 h. However, the second
generation machine (Fig. 1) logged over 300 total hours with-
out failure or notable wear. Naturally, due to the short op-
erating times, no measurable wear has been noted on the latest
machines.
In spite of the absense of long term reliability data at print-
ing, there exists data and operating information on materials
and machines operating under similar, and in some respects
more severe, conditions. For example, "smog" pumps, dry
rotary vane motors and dry rotary vane vacuum pumps, all
operate at respectable speeds and under varying conditions.
Dry operating rotary vane machines (with vane tips rubbing
against the stator housing) are available off-the-shelf with
guaranteed lifetimes of 4000 hours. These facts, coupled with
design knowledge of the circulator materials, and existing test
data builds a credible basis for expecting ROVAC machines to
demonstrate quite favorable longevity.
In conclusion, the past development of the ROVAC system
has demonstrated without question the capability of employ-
ing ordinary atmospheric air to serve as an exceptionally ef-
fective refrigerant for the air conditioning of automobiles. It
now appears within sound judgement to anticipate the reality
of the major contribution brought forth by the advent of
ROVAC.
ACKNOWLEDGMENTS
The author gratefully acknowledges the opportunity of per-
forming this pilot program with the Chrysler Corp. Specific
thanks and acknowledgement is directed to Messrs. Gerald
Davis, James Holtslag, and Timothy Scott of the Chrysler
Automotive Air Conditioning Section for their foresight and
encouragement during this program.
REFERENCES
1. T. C. Edwards, "A Rotary Vane Open Reversed Brayton
Cycle Air Conditioning and Refrigeration System." Ph.D.
Thesis, School of Mechanical Engineering, Purdue University,
June 1970.
2. C. R. Peterson and W. A. McGahan, "Thermodynamic
and Aerodynamic Analysis Methods for Oil Flooded Slidding
Vane Compressors." Purdue Compressor Technology Con-
ference, 1972.
3. United States Air Force report (not available at this
time).
216
-------
APPENDIX C
PERSONS AND FIRMS CONTACTED
217
-------
AIR CONDITIONING AND REFRIGERATION CONTACTS
1. Mr. Frank Vergasi
Air Conditioning, Heating, and
Refrigeration News
Birmingham, Massachusetts
2. Mr. H. T. Gilkey
Air Conditioning and
Refrigeration Institute
Arlington, Virginia
3. Mr. H. L. Noble
Allied Chemicals Corporation
Morristown, New Jersey
4. Mr. Louis Marz
Amana Refrigeration
Amana, Iowa
5. Mr. James Stevens
Appliance
Elmherst, Illinois
6. Mr. Arnold Consdorf
Appliance Manufacturer
Chicago, Illinois
7. Mr. R. J. Destiche
Mr. John Hester
Arkla Industries
Little Rock, Arkansas
8. Mr. Roger Shamel
Mr. R. Williams
Mr. Robert Green
Arthur D. Little, Inc.
Cambridge, Massachusetts
9. Mr. Herb Phillips
Association of Home Appliance
Manufacturers
Chicago, Illinois
10. Mr. Ben Wilson
Bahnson Company
Winston-Salem, North Carolina
11. Mr. Huddleson
Mr. David Wilson
Carrier Corporation
Columbus, Ohio
12. Mr. James Holtslag
Chrysler Corporation
Detroit, Michigan
13. Mr. John Knoble
Cro11-Reynolds Company
Westfield, New Hersey
14. Mr. Jerry Fortier
Cryogenic Technology, Inc.
Waltham, Massachusetts
15. Dr. Hans Borchardt
Mr. Clark Hoffman
Du Pont Freon® Products Laboratory
Wilmington, Delaware
16. Mr. Billy Owens .
Federal Energy Administration
Washington, D.C.
17. Mr. Gerald Stofflet
Mr. J. A. Dobb
General Motors Corporation
Warren, Michigan
18. Mr. Todd Merrill
International Mobile Air
Conditioning Association
Dallas, Texas
218
-------
19. Mr. Milton Levlne
Melcor Materials Electronic
Products
Trenton, New Jersey
20. Mr. Ciricillo
Ranko Corporation
Columbus, Ohio
21. Dr. Thomas Edwards
Rovac Corporation
Maitland, Florida
22. Mr. H. L. Munde
Mr. R. L. Maier
Southwest Factories'; Inc.
Oklahoma City, Oklahoma
23. Mr. J. V. Weigle
Mr. R. D. Randall
Dr. K. S. Sanvordenker
Tecumseh Products Company
Tecumseh, Michigan
24. Trane Company
La Crosse, Wisconsin
25. Dr. Lawrence Midolo
United States Air Force
Flight Dynamics Laboratory
WPAFB, Dayton, Ohio
26. Mr. J. McLean
Westinghouse Corporation
Columbus, Ohio
27. Dr. Gale Cutler, et al.
Whirlpool Corporation
Benton Harbor, Michigan
28. Mr. W. T. Ferguson
York Division oi Borg-Warner
York, Pennsylvania
219
-------
AEROSOL INDUSTRY
10.
11.
1. Mr. Bob Tucker
President
Container Division
Plant Industries, Inc.*
Montreal, Canada
2. Mr. Donald Knox
Calmar Division
Diamond International Corporation
New York, New York
3. Mr. Dan Massey
Marketing Manager
Valve Corporation of America
Baton Rouge, Louisiana
4. Mr. Frederick Muller
Executive Vice President
Bakan Plastics Division
Realex Corporation
Kansas City, Missouri
5. Mr. Hal Goldman
Marketing Research Manager
The Mennon Company
Morris town, New Jersey
6. Mr. Pat Adamo
Division of Regulatory Guidelines
Bureau of Foods
Food and Drug Administration
Washington, D.C.
7. Mr. Charles S. Hayes
Applications Engineer
Chemetron Corporation
Chicago, Illinois
Mr. Edward Kalat
Research and Development
Aerosol Packaging Component Division
Risdon Manufacturing Company
Thomas ton. Connecticut
220
Amway Products
Distribution Center
Kansas City, Missouri
Mr. Charles S. Booz, Jr.
Public Affairs Department
E. I. du Pont de Nemours and
Company
Wilmington, Delaware
Mr. John C. Bray (and others)
Market and Product Development
Manager
Aerosol Propellants
Freon® Products Division
E. I. du Pont de Nemours and
Company
Wilmington, Delaware
-------
DECREASING AND DRY-CLEANING
1. Mr. Francis Figiel \
Technical Specialist
Technical Service
Specialty Chemicals Division
Allied Chemical Corporation
Morristown, New Jersey
2. Mr. Robert Orseo
Allied Chemical Corporation
Buffalo, New York
3. Mr. Ralph Dermott
American Society of Metals
Metals Information Department
Metals Park, Ohio
4. Mr. Anderson
Anderson Myers Company, Inc.
Kansas City, Missouri
5. Mr. Arthur
Baron Blakeslee
Vapor and Ultrasonic Degreasers
Cicero, Illinois
6. Mr. Geroald Gershon
Chemical Commodities; Inc.
Olathe, Kansas
7. Mr. T. Kearney
Senior Engineer
Mr. George Leith
Manager, Special Products Sales
Detrex Chemical Industries, Inc.
Detroit Michigan
8. Mr. Ken S. Suprendnt
Chlorinated Solvents
Dow Chemical, USA
Midland, Michigan
9. Mr. Curt O'Hogan
Dow Chemical, USA
Johnson County, Kansas
10. tar. Maltrnan
Solvent Section
E. I. du Pont de Nemours and
Company, Inc.
Wilmington, Delaware
11. Mr. Bill Fisher
Joliet, Illinois
Mr. Lloyd
Silver Spring, Maryland
International Fabricare
12. Livingston Industries, Inc.
Lenexa, Kansas
13. Mr. Ishmial Akbay
Mr. Mack Sharp, NASA
George C. Marshall
Space Flight Center, Alabama
14. Missouri Solvents and Chemicals
Kansas City, Missouri
15. Solvent Supply Company
Kansas City, Missouri
16. Mr. Tom McDonald
Vic Manufacturing Company
Minneapolis, Minnesota
221
-------
FOAM BLOWING INDUSTRY
1. Mr. H. Lee Noble
Allied Chemical Company
Morristown, New Jersey
2. Mr. Joseph Hope
Allied Chemical Company
Director of Fluorocarbon Marketing
3. Mr. Ken Me inert
Manager, Corofoam Division
Cook Paint and Varnish Company
North Kansas City, Missouri
4. Mr. Tom Morriss
Corofoam Division
Cook Paint and Varnish Company
North Kansas City, Missouri
5. Dr. Charles F. Kloss
Uniroyal, Inc.
Naugatuck, Connecticut
6. Mr. Richard B. Ward
Freon Products Laboratory
E. I. du pont de Nemours
Wilmington, Delaware
7. Mr. Arthur Chivas
E ecutive Director, Urethane Safety Group
SP1
New York
8. Mr. McLaughlin
SPI
New York
222
-------
TECHNICAL REPORT DATA
(Please read luxtritclioiis on Ilie reverse before completing)
1. RtPORT NO.
EPA 560/1-76-002
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE ANDSUOTITLE
Chemical Technology and Economics in Environmental
Perspectives; Task I - Technical Alternatives to
Selected Chlorofluorocarbon Uses
6. REPORT DATE
March 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Thomas W. Lapp
Ivan C. Smith
G. Joe Hennon
Kathryn Lawrence
Howard M. Gadberry
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
Contract No. 68-01-3201
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
Office of Toxic Substances
Washington, D.C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
Final June 1975-January 197(
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The purpose of this study was to identify technically feasible substitute chemi-
cals and/or alternative methods of delivering the goods and services presently pro-
vided through the use.of the five commercial chlorofluorocarbons FC-11, -12, -13,
-113, and -114. This study did not involve an assessment of the risks associated with
environmental discharge of these chemicals. For the purposes of this study, recovery
and/or recycling were not to be considered as eligible alternatives. Four categories
currently account for 99% of the usage of these compounds. These areas are: refrig-
eration and air conditioning, aerosol propellants, plastic foam blowing agents, and
cleaning and drying applications. In addition to these four areas, the Rankine Cycle
engine was considered as it appears to have potential for expanding future consump-
tion of chlorofluorocarbons. Limited evaluations were made on the direct economic
consequences which could result from a reduction or elimination of chlorofluorocarbon
consumption in each of the current use areas.
KEY WORDS AND DOCUMENT ANALYSIS
DF.SCRIPTORS
Freons®
Blowing agents
Refrigeration
Air conditioning
Solvents
Alternatives
IK. orl)
Unclassified
70. Sl.OJHITY C\.A.W>
Unclassified
. COSATI Held/Group
Chemistry
Organic
chemistry
21. NO. of r.v.i -j
232
22. IMMCh' ~~
lil'A I oi-m 22?O-I
------- |