Source Reduction Research Partnership
Metropolitan Water District of Southern California
Environmental Defense Fund
Source Reduction of
Chlorinated Solvents
rl
Prepared for
Alternative Technology Division
Toxic Substances Control Program
California Department of
Toxic Substances Control
and
Pollution Prevention Research Branch
Risk Reduction Engineering Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
June 1991
-------
Source Reduction Research Partnership
Metropolitan Water District of Southern California
Environmental Defense Fund
Source Reduction of
Chlorinated Solvents
AEROSALS MANUFACTURE
Prepared for
Alternative Technology Division
California Department of
Toxic Substances Control
P .O. Box 806
Sacramento, CA 95812-0806
Pollution Prevention Research Branch
Risk Reduction Engineering Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
June, 1991
-------
Source Reduction Research Partnership
Metropolitan Water District of Southern California
Environmental Defense Fund
Source Reduction of
Chlorinated Solvents
AEROSALS MANUFACTURE
Prepared for
Alternative Technology Division
California Department of
Toxic Substances Control
P .O. Box 806
Sacramento, CA 95812-0806
Pollution Prevention Research Branch
Risk Reduction Engineering Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
June, 1991
-------
TABLE OP CONTENTS
I. ACKNOWLEDGMENT AND DISCLAIMER j
II. INTRODUCTION
III. INDUSTRY BACKGROUND 3
SOLVENT USAGE 5
PRODUCT DESIGN 18
U.S. AEROSOL PACKAGING FACILITIES 27
THE FILLING PROCESS AND SOURCES OF RELEASES 28
COSTS 34
RESULTS OF SURVEYS 35
REGULATORY TRENDS 38
INDUSTRY TRENDS 43
IV. SOURCE REDUCTION OPTIONS 45
PROCESS MODIFICATION 45
SOLVENT RECOVERY AND REUSE 48
PRODUCT SUBSTITUTION 53
CHEMICAL SUBSTITUTION 57
V. ANALYSIS OF SOURCE REDUCTION OPTIONS 76
CLASSIFICATION OF OPTIONS 76
NO FURTHER ANALYSIS OPTIONS 78
LIMITED ANALYSIS OPTIONS 79
REFERENCES 98
-------
LIST OF FIGURES
2.1 BASIC COMPONENTS OF AN AEROSOL CONTAINER 19
2.2 AEROSOL VALVE ASSEMBLY 21
2.3 SCHEMATIC DIAGRAM OF AN AEROSOL FILLING LINE 30
-------
i
ACKNOWLEDGMENT
The principal sponsors of this project, The
Metropolitan Water District of Southern California and the
Environmental Defense Fund gratefully acknowledge major support
from the Switzer Foundation and from the U.S. Environmental
Protection Agency, The California Department of Health Services,
and the City of Los Angeles, (Los Angeles Department of Water
and Power). Additional support was received from Southern
California Edison Company. -
The Environmental Defense Fund also gratefully
acknowledges the support of the Andrew Norman Foundation an the
Michael J. Conne].1 Foundation for the exploratory phase that led
to the formation of the Source Reduction Research Partnership
and the development of the research plan.
The principal project sponsors recognize the effort and
contributions of many people from industry and government who
helped in preparation of these reports. These efforts and
contributions are being gratefully acknowledged.
DISCLAIMER
The statements and conclusions of this report do not
necessarily represent those of the State of California, the U.S.
Environmental Protection Agency or any other contributors. The
mention of any commercial products, their source or their use in
connection with material reported herein is not to be construed
as either an actual or implied endorsement of such products.
-------
1.1.
PRE FACE
This report is one of twelve reports that evaluate the
opportunities for source reduction of chlorinated solvents in
twelve specific industries. The twelve reports are part of a
large-scale study sponsored by the Source Reduction research
Partnership (SRRP), a joint venture by the Metropolitan Water
District of Southern California and the Environmental Defense
Fund. The reports cover the following industries and industrial
practices:
1. Aerosols Manufacture
2. Adhesives Manufacture
3. Chemical Intermediates Manufacture
4. Dry Cleaning of Fabrics
5. Electronic Products Manufacture
6. Flexible Foam Manufacture
7. Food Products Manufacture
8. Paint Removal
9. Pesticides Formulating
10. Pharmaceuticals Manufacture
11. Solvent Cleaning
12. Textiles Manufacture
The objectives of the SRRP study include a survey and
evaluation of existing and potential techniques for reducing the
generation of halogenated solvent wastes, and thus their
potential release into the environment, across a wide range of
the industrial users of these solvents.
Each of the industry-specific reports begins with a
description of the industry and processes where halogenated and
solvents are used. Sources and causes of releases are described
-------
— 111 —
and regulatory regime discussed for waste streams of concern.
Subsequent sections focus on source reduction opportunities
through chemical substitution, process modification, product
substitution and recovery/reuse. For major solvent using
industries, select source reduction options were analyzed for
their economic feasibility.
The information in the reports was compiled and analyzed
by the SRRP project staff, employed by the Partnership to carry
out the project research. Each report was reviewed by industry
representatives and/or other experts familiar with the specific
industry and the relevant technologies and issues, and then
reviewed and edited by an additional expert consultant.
The intent of the sponsors is to provide all interested
parties with useful information on available and potentially
available methods for source reduction of halogenated solvents,
in the context of specific industries and processes, and an
evaluation in context of the various source reduction options.
-------
I. INTRODUCTION
The five major chlorinated solvents most widely used in
commerce today include trichioroethylerie (TCE),
perchioroethylene (PERC, i,i,i-trichloroethane (TCA), methylene
chloride (METH) and 1,1, 2—triChlOrOl,2, 2—trifluoroethane
(CFC-313). Of these, METH and TCA are Co ionly used in aerosol
products. PERC arid CFC-113 are also used to some extent, while
TCE is not used in any significant amount.
This report is one of several companion reports that
focus on source reduction of chlorinated solvents in the
industries where the solvents are widely used. This document
specifically addresses the aerosol packaging industry and
provides an assessment of the source reduction potential within
that industry. The aerosol products industry currently uses
about 57 thousand metric tons ( Tnt) of chlorinated solvents
annually. In recent years, there has been a movement away from
these solvents, particularly PERC and METH, in certain market
Segments. This is the result of increased regulatory pressure
in the form of air quality regulations and labelling
requirements. Aerosol fillers have replaced some chlorinated
solvent based formulations with water based or flammable
hydrocarbon solvent based products.
Much of the regulatory attention, particularly at the
state level (California and New Jersey, for instance), focuses
on the end use of the aerosol product, rather than emissions at
the aerosol packing plant. Of the total amount of solvent used
lfl aerosol products, very little is lost to the atmosphere
during container filling. Virtually all of the emissions occur
during product use at the consumer or industrial level.
In the balance of this document, we discuss the aerosol
industry in detail. Section II focuses on the design and
filling of aerosol products and discusses market segments,
-------
—2—
solvent usage and industry trends. Section III describes the
source reduction options that may be implemented to reduce
chlorinated solvent usage in the industry. Options fall into
four categories——process modification, solvent recovery and
reuse, product substitution and chemical substitution. Section
IV provides a case study of two chemical substitution options in
aerosol products, and limited analysis of other options.
Section V summarizes our findings.
a
-------
—3—
II. INDUSTRY BACKGROUND
An aerosol product can be any of a large number of
consumer or industrial use products packaged in a pressurized
container. “Aerosol” refers to the packaged form, rather than
the product contained therein. An aerosol product can be
dispensed as a mist, fine or heavy spray, a liquid stream, foam
or powder. Characteristics of the dispensed form are determined
by the combination of ingredients in the formulation and by the
choice of container hardware. Basically, an aerosol product
Consists of an active ingredient, a liquid propellant and an
appropriate solvent in a container. The propellant Is also
present as a vapor, creating the pressure in the container.
When the product is released, the propellant evaporates
immediately.
Aerosol packaging provides the user with convenience,
ease of product transfer, and efficient product distribution.
The container seals itself after each use. The product retains
its composition because there is no way for material to
evaporate from the container. Since the aerosol package is
always sealed, product contamination is avoided. User contact
With hazardous ingredients is limited to intentional use, and
accidental release of the product through leaks or spills is
less likely to occur than with other product forms.
This section consists of eight subsections. The first
addresses solvent usage in general and within market segments.
The second subsection describes aerosol product design. The
third subsection discusses U.S. aerosol packaging facilities.
In the fourth subsection, the filling process is described and
Sources of releases are identified. The fifth subsection
discusses costs of aerosol products. In the sixth subsection,
we review the results of our site visits and surveys. The last
two subsections address industry and regulatory trends that
impact solvent usage.
-------
TABLE 1.1
Source Reduction Option Summary
Option Advantages Disadvantages References
Correction of Reduces Low potential for Flanner, 1988
over and under production and waste reduction
filling management costs
Improved Low capital Martinez, 1987
maintenance expenditure
practices
Vapor recovery— Can reduce Martinez, 1987
refrig. solvent
condensation emissions from
storage tanks &
mixing by 95%
Vapor recovery— Can reduce Hazardous wastes Martinez, 1987
carbon adsorption solvent emissions are generated
from mixing by from carbon
95% recovery
operations
Solvent recycling Able to reduce Recycling Flanner, 1988
from cans production contents of some
material cans presents
requirements and fire and
costs explosion danger
unless proper
equipment used
Proper can-
puncturing equip.
is not cost—
effective for
many shops to buy
Cleaning solvent Low capital
reuse expenditure
Non—aerosol Potential for Use not as Geigel & Miller
packaging large air convenient as 1985
emission aerosol packaging
reductions
Some production ICF, 1987 a,b
still require Westate, 1987
haz. solvents as
active
ingredients
-------
—5—
SOLVENT USAGE
A wide variety of product types are packaged as
aerosols. An estimated 2.9 billion units of pressurized
products were filled in the U.S. in 1988 (CSMA, 1989a). The
major categories included in this total are household products,
personal care products, industrial and automotive products,
foods, paints and related products, and insect sprays.
Various sources have estimated chlorinated solvent
usage in the aerosols industry. As illustrated in Tables 2 1
through 2.3, these estimates vary widely. SRRP staff estimates
of 1988 solvent usage are summarized in Table 2.4. Table 2.1
lists TCA usage estimates obtained from different sources for
the years 1981 through 1988. The numbers range widely from year
to year, which could indicate that TCA use has fluctuated in
recent years. SRRP believes that this is not so, and that the
1987 value of 18 thousand metric tons (mt) significantly
underestimates actual use. Based on these figures and
discussions with industry experts, it is estimated that 34
thousand mt of TCA were used in the aerosols industry in 1988.
Estimates of METH usage also vary widely from year to
year (Table 2.2). Even within each year, the numbers are
inconsistent. For instance, one source estimates that 20
thousand nit of METH were used in 1987. According to another
source, 41 thousand mt were used. As broad as the ranges are
for estimated use of METH, the values indicate that METH use in
aerosols has, in general, declined from 1984 to 1988. Various
sources confirm this trend (ICF, 1987a; Dow, 1989a). We
estimate that 20 thousand nit of METH were used in aerosol
products in 1988.
-------
—6—
Table 2.1
TCA Consumption in the Aerosols Industry
Volume
Year ( thousand metric tons Source
198]. 27 CMR, 1982
1982 27 CMR, 1983
1984 23 Geigel and Miller, 1985
23—31 ICF, 1987a
1985 19 CMR, 1986
29 HSIA, 1987
1987 19—24 ICF, 1989
1988 32 CMR, 1989a
32 Cuzic, 1989
34 SRRP Estimate
-------
—7—
Table 2.2
METH Consumption in the Aerosols Industry
Volume
____ ( thousand metric tons Source
1983 52 HSIA, 1985
1984 29—39 ICF, 1987a
34—43 Geigel and Miller, 1985
1985 73 Chemical Week, 1987
45 ICF, 1987a
1986 42 HSIA, 1986
1987 20 Pfetzring, 1987
21 ICF, 1989
41 HSIA, 1989
1988 4]. CMR, 1989b
20 SRRP Estimate
-------
—8—
Table 2.3
PERC Consumption in the Aerosols Industry
Volume
Year ( thousand metric tons) Source
1984 4.5 Geigel and Miller, 1985
1987 4.3 ICF, 1989
1988 3 SRRP Estimate
-------
—9—
Table 2.4
Chlorinated Solvent ConsUmptiOn in the Aerosols Industry
1988 SRRP Estimates
Thousand Metric Tons
TCA
METH 20
PERC
CFC—1 13 0.6
TCE -
Total 57.6
-------
— 10 —
Only a small amount of PERC is consumed in aerosols.
The figures in Table 2.3 indicate that use ranges from 3 to 5
thousand mt. PERC use is expected to decline in the future
because of concerns over the health effects of PERC exposure.
CFC-113 usage in 1988 was less that one thousand mt. No
significant amount of TCE is used in aerosols. Because of its
relatively small volume of use, source reduction of TCE will not
be discussed in the balance of this report.
It is important to note that the pattern of solvent
usage has changed rapidly in recent years and continues to
change in response to regulatory pressures, economic
considerations and consumer preferences. Consumption data, even
from as recent as 1988, may not accurately reflect current
solvent usage.
The aerosols market can be described in terms of
general product categories, with each category further divided
into product segments. One industry trade group identifies the
following eight categories:
— insect sprays,
— paints and finishes,
— household products,
— personal products,
— animal products,
— automotive, industrial and miscellaneous household
— food products, and
— miscellaneous products.
Table 2.5 summarizes the number of containers filled in each of
these categories in 1988 and their ontribution to the total.
Several studies have analyzed chlorinated solvent usage in
aerosol products. Among them are Geigel and Miller, 1985;
-------
— 11 —
Table 2.5
Containers Filled in Eight Categories in 1988
Personal Products
Household Products
Automotive and Industrial
a
Paints and Finishes
Insect Sprays
Food Products
Miscellaneous
Animal Products
Total
Number of
Units Filled
(in millions)
1,100
650
440
331
190
157
31
8
2,907
Percent of
Total
37.8
22.4
15.1
11.4
6.5
5.4
1.1
0.3
100.0
Source: csMJ4, 1989a.
-------
— 12 —
Westat, 1987; and ICF, 1989. Some of these report detailed
analyses of chlorinated solvent concentration in product
formulations, and estimate solvent usage by industry segment.
We will refer to these studies where appropriate but do not
reproduce all of the data here. The pattern of solvent usage
has changed in recent years, and continues to change at
present. Solvent consumption estimates from as recent as two
years ago cannot be considered reliable estimates of current
usage trends. We attempt to address solvent usage in the
industry in general, and use specific examples to illustrate
successes or difficulties in implementing various source
reduction options.
Based on U.S. market estimates, personal products
represent over a third of all aerosol products filled in 1988
(CSMA, 1989a). Included in this category are hair sprays,
mousse and other hair products, deodorants and antiperspirants,
shaving lather, colognes and perfumes, and other personal care
products. Medicinal and pharmaceutical products are also
included in this category. Other than a small amount of METH,
no chlorinated solvents are used in personal care products. One
source places 1987 METH consumption in personal care products at
0.4 thousand mt (IcF, 1987a). In the past, METH was widely used
as a co-solvent in hair spray products. When the industry moved
away from CFC propellants in the late 1970s, the flanunable
hydrocarbon propellants isobutane and propane replaced the
CFCs. METH was found to be very good at reducing the
flammability of the newer hair sprays, and also served to
solubilize the resins and propellants into an alcohol base.
Because of its volatility, It causes the hair spray resin, when
applied, to dry and set quickly (Aerosol Age, 1989a). One
estimate places METH use in hair sprays at about 13.6 thousand
mt in 1983 (Geigel and Miller, 1985). Because of concerns over
the health hazards associated with METH exposure, and pending
-------
— 13 —
federal regulations requiring labelling of consumer products
containing METH or an outright ban on the use of METH in
cosmetic products, its use in hair sprays declined
dramatically. A solvent trade group states, in comments to the
U.S. Food and Drug Administration (FDA), that “all or virtually
all manufacturers or formulators” had stopped using NETH in
cosmetics. According to FDA, however, at least one manufacturer
is currently using METH in aerosol cosmetics (Aerosol Age,
1989a). By and large, METH and the other chlorinated solvents
are not widely used in personal care products, but we are aware
of at least one pharmaceutical product that contains TCA.
Chlorinated solvent usage in personal care products is not
expected to increase in the future.
The second largest use of aerosols in the U.S. is in
the household products category. Its share of the 1988 U.S.
aerosol products market has been estimated at 22 percent of the
total market (CSMA, 1989a). Included in this category are
household cleaners (i.e. for rugs, appliances, walls and other
surfaces and fabrics), laundry products (pre-wash sprays, spot
removers, etc.), room deodorants, waxes and polishes, water
repellants and water proofing products for shoes and other
items, and miscellaneous other products. Some of these products
are likely to contain TCA, PERC or METH. For instance, any of
these three can be used as the active ingredient in laundry spot
removers. In cleaning products, a high degree of solvency is
desired, which is why PERC, NETH and TCA are good choices.
Water repellants are usually silicone based products in which
TCA or PERC is used as a solvent. Many fabric protectors and
sprayable shoe polishes are formulated with TCA. Estimates of
solvent consumption in household aerosol products are 0.5
thousand mt of METH, 7.5 thousand nit of TCA, and 1.0 thousand nit
of PERC (ICF, 1987a).
-------
— 14 —
Automotive and industrial products make up the third
largest use of aerosol products in the U.S. Based on 1988 data,
this market accounted for 15 percent of the aerosol products
filled (CSMA, l989a). Automotive products include carburetor,
choke, fuel injector, and brake cleaners, engine starting fluid,
tire inflator and sealant, automotive spray undercoatings,
engine degreasers, cleaners for vinyl, leather, upholstery and
tires, and windshield and lock spray de-icers. Refrigerants,
lubricants, mold releases, adhesives and other products are also
inc1u 1ed in the automotive and industrial category. In man’y
automotive and industrial products that contain chlorinated
solvents, such as cleaners, the solvents are present as active
ingredients. Chlorinated solvent usage is higher in this
category than in any other. TCA is used most often, followed by
METH, PERC and CFC-1l3. Solvent consumption in automotive and
industrial aerosol products has been estimated at 8 thousand mt
of TCA, 7 thousand mt of NETH, and 1.7 thousand mt of PERC (ICF,
1987a)
TCA is formulated into engine cleaners because its high
dielectric strength allows engines and appliances to be cleaned
while they are operating. Lubricants include metal cutting
fluids, mold release agents and silicone compounds. There are
many different types of these products, and they are used in a
wide variety of applications and environments. General purpose
lubricants are likely to contain TCA, METH, or petroleum
distillates. For technical reasons, TCA or the petroleum
distillates are favored. METH is less often used. Silicone
lubricants designed to be used in broad temperature range
applications are generally formulated with TCA and METH as
solvents. For these products also, TCA is reportedly preferred
over METH. TCA, being a slower evaporating solvent, tends to
spread the silicone lubricant better before all of it
evaporates. Conversely, METH is preferred over TCA as the
-------
— 15 —
solvent in aerosol mold release agents because it evaporates
more quickly. Often, a blend of NETH and TCA is used in
industrial and automotive aerosols to achieve desired product
properties.
There are many types of adhesives available in aerosol
form. Of those that contain chlorinated solvents, METH is
almost always the solvent used. Some formulators have switched
to TCA as the solvent in sprayable adhesives although its slower
drying properties make it less desirable. Some adhesives
containing METH cannot be reformulated. Where a fast drying
adhesive is required, only METH provides good solvency
characteristics, fast drying, and an acceptable spray pattern.
CFC-ll3 is used in only one segment of the aerosols
market. In industrial applications, CFC—113 and TCA are used in
cleaners for electronic equipment and electrical contacts.
In the automotive market there are a number of aerosol
products that contain NETH, TCA and PERC. In many cases,
chlorinated solvents are used in cleaning and degreasing
products because they are such good cleaners. They serve partly
as the active ingredient, and partly as the solvent in the
blend. For some of these products, TCA is preferred since its
slower drying allows the cleaner to remain on the part’s surface
longer and seep into cracks or small joints. Engine degreasers
are commonly made with a blend of all three solvents, although
PERC has historically been used more than TCA or METH. It has
been reported that PERC use is dropping in this product segment,
with TCA being used in its place.
Battery cleaners may contain METH, TCA and PERC in
small amounts. Spray undercoatings contain METH, alone or in
combination with TCA, or may use petroleum distillates as
-------
— 16 —
solvents. Most brake cleaners contain a blend of TCA and PERC.
METH and TCA are used in aerosol tire cleaners. The chlorinated
solvents swell the rubber in tires and make cleaning easier.
METH is frequently used in carburetor cleaners.
Paints and coatings accounted for approximately 11
percent of the aerosol products filled in the U.S. in 1988
(CSMA, 1989a). This category includes traditional spray paints,
primers, varnishes, rust inhibitors, wood stains, paint
strippers, and artificial snow and other decorative product .
About 80 percent of these products are spray paints. Spray
paints consist of pigments which provide color and capacity, a
resin system that forms the continuous paint film upon drying,
solvents and propellants. Four different kinds of materials
must be contained in solution or in suspension within the
aerosol container. METH is the solvent of choice in many
aerosol paint products. Water is used as well, as is discussed
in Section III. One source estimates that 10.5 thousand Tnt of
METH were used for aerosol paint products in 1987 (ICF, 1987a).
The amount of METH in a spray paint formulation depends
on several factors. Important among them are the type of paint
(e.g., flat, gloss) and the type and concentration of solids
(resin and pigment) in the paint. For instance flat paints
generally have a higher concentration of pigment and,
consequently, a lower percentage of solvent (usually METH).
Gloss paints contain less pigment and a higher concentration of
solvent (ICF, 1987a).
Insect sprays accounted for almost 7 percent of the
1988 aerosol products market (CSMA, 1989a). Chlorinated
solvents are used in many insect sprays. Products within this
category include room insecticides (such as total release
foggers), personal use sprays, pet flea and tick products and
-------
— 17 —
insect repellants. Chlorinated solvents are most widely used in
total release foggers, house and garden insect sprays, residual
insecticides, and flying insect killers. The function and
source reduction potential of pesticides is addressed in more
detail in the companion document “Pesticides Industry”.
Aerosol insecticides are formulated with toxic active
ingredients, solvents and propellants. A mixture of active
ingredients is used. A knock—down agent stuns the insect upon
contaçt, another toxic ingredient kills the animal, and a
synergist is added to improve the activity of the first two
components. TCA is reported to be the most effective knockdown
agent available for wasp and hornet sprays (CSMA, 1989b). It
evaporates and rapidly cools the insect’s body. The propellant
system is a mixture of propane and isobutane, or CO 2 . Other
hydrocarbons (base oils) are added to the blend as carriers, but
because of their poor solubility, do not function as solvents.
Careful solvent selection is important to solubilize the various
ingredients and propellants and to achieve the right droplet
size. METH and TCA are used as solvents in insect sprays,
although water and aromatic solvents will work in many cases.
1987 solvent use in aerosol insecticides has been estimated at 2
thousand mt of METH, and 1.7 thousand mt of TCA (ICF, 1987a).
Wasp killers represent one specialty product that uses
only CFC-113. When utility employees are working on high
voltage electrical lines, they need an effective, sprayable wasp
killer that will not be conductive. CFC-113 is used for this
purpose. EPA recognizes this product as an essential use of
CFC-l13, one in which alternative solvents are not feasible. It
is estimated that 600,000 cans per year of wasp killer are
produced, with CFC-113 concentratio in the product ranging from
70 to 85 percent (ICF, 198Th). TCA, because of its electrical
properties and nonflammability, is used in insecticides applied
near transformers and other electrical equipment.
-------
— 18 —
Total release foggers are designed to quickly release
their entire contents in one application. They are used to
exterminate insects in large areas. Total release foggers have
traditionally consisted of 15 percent active ingredient and 85
percent propellant. CFCs were the propellants of choice until
the late 1970s. Since the phaseout of CFCs in aerosol products,
the formulations have changed. Typically, total release foggers
now contain one to three percent active ingredients (toxicant)
and about sixty percent of a METH/TCA blend. The percentage of
relative METH and TCA varies by product.
In response to consumer concerns about the safety of
total release foggers that contain flammable propellants, large
quantities of TCA are added to the formulations. TCA is used
because it is nonflammable (Johnsen, 1988). As in other
products, the use of METH is declining.
The remainder of the major aerosol product categories
are animal, food and miscellaneous products. Animal and food
products do not contain chlorinated solvents to any significant
degree.
PRODUCT DESIGN
An aerosol product is a combination of (1) the desired
product to be dispensed, (2) a propellant and solvent system,
and (3) the container hardware. Careful selection of each of
these components is critical to the successful development of a
product in aerosol form.
A typical aerosol container is shown in Figure 2.1.
Major Components of the hardware in 1ude the container itself,
valve assembly and actuator. Most containers are made of
-------
— 19 —
Figure 2.1
Basic Components of an Aerosol Container
Source: Geigel arid Miller, 1985.
-------
— 20 —
tinpiate, although aluminum containers are available
(approximately 13 percent of the aerosols market). Tinpiate and
aluminum containers are sometimes lined to prevent contact with
a product that may react with the metal, or which may cause
pitting or corrosion of the can. Glass and plastic containers
are also used, primarily for personal care products such as
perfume and cologne. U.S. Department of Transportation
regulations don’t permit the use of nonmetallic pressurized
containers larger than 118.3 cc’s (3.5 to 4 fluid ounces).
The purpose of the valve system is to dispense the
product in the desired form, pattern, and dosage. The valve
system also serves to keep the top opening of the container
closed and to retain adequate pressure within the container. A
valve assembly is illustrated in Figure 2.2. The choice of
valve assembly determines the form and dose of the dispensed
product. Valves are manufactured in standard sizes and are
selected to produce the desired pattern, i.e. spray, foam or
specialty valves. Hardware components must be chosen to be
nonreactive with all ingredients in the blend, to function
properly and to not contaminate the product in any way. The
type of hardware used can influence the flammability of the
aerosol product (Cuzic, 1989).
While proper selection of the valve system is important
to proper dispensing action, selection of the formulation
components is equally important. Components that must be
considered include the active ingredients, propellant and
solvent blends, and additives. Propellants used in aerosol
products are either liquifi.ed or compressed gases with vapor
pressures greater than atmospheric pressure. The propellant is
present in the container in both the liquid and vapor phases.
When the actuator button is pushed, the valve opens and the
propellant vapor expands, providing the driving force to expel
the contents from the container.
-------
- 21 —
Figure 2.2
Aerosol Valve Assembly
Gwcr
Source: Adapted from Kawai and Flynn, 1979.
-------
— 22 —
Selection of a propellant involves consideration of its
vapor pressure, solubility, chemical compatibility with other
product ingredients, and wetness or dryness of the desired
product. Often, a blend of propellants is chosen to achieve the
right properties of product delivery, solubility, cost, vapor
pressure and degree of flammability. The propellant vapor
pressure inside the container determines the product delivery
rate, which is also influenced by the chemical composition of
the propellant and its concentration. Droplet size and the
“wetness” of the spray are also influenced by the choice of
propellant. The propellant affects the solubility of other
components in the formulation.
For effective product delivery, it is desirable to have
the active ingredient, solvent and propellant in solution in the
liquid phase. Liquified propellants don’t usually go into
solution well, making careful solvent selection important.
Traditional liquified propellants include the
chiorofluorocarbons CFC-ll, CFC-12, and CFC-114; and flammable
hydrocarbons such as propane, n-butane and isobutane. An
advantage to using liquified propellants is that they maintain a
constant pressure in the container until the contents are
completely exhausted. CFCs are no longer used in aerosol
products because they are known to deplete the stratospheric
ozone layer. This is discussed in more detail in the section on
regulatory trends.
The liquified petroleum gases are widely used as
propellants, even more so since the use of CFCs has diminished.
Because liquified petroleum propellants are flammable,
flammability suppressants must be added.
Compressed gas propellants perform differently in an
aerosol container. They are present only in the gas phase, not
in the liquid phase. Carbon dioxide (C0 2 ), nitrous oxide
-------
— 23 —
(N 2 0), and nitrogen (N 2 ) are commonly used as compressed gas
propeliants. They are nontoxic, nonflammable, inert gases and
are lower in cost than ].iquified propellants. As an aerosol
package containing a compressed gas propellant is dispensed, the
vapor pressure within the container diminishes. Since the
propellant pressure released to the atmosphere provides the
force to push product out of the container, this gradual
decrease in vapor pressure effects the spray characteristics and
rate of flow. This causes performance problems with certain
types of formulations.
Most propellants, especially halocarbons and
hydrocarbons, tend to have poor solubility characteristics. One
or more solvents must be added to the blend to mutually
solubilize the active ingredients and the propellant. Selection
of the appropriate solvent or solvent blend is important in
order to ensure that the active ingredients are present as a
homogeneous solution with the propellant, or to solubilize a
compressed gas propellant into an ingredient concentrate.
Choice of the solvent also influences the rate of spray and
droplet size. In some products, the solvent itself performs as
an active ingredient.
Substances used as solvents in aerosol products include
water, acetone, methyl ethyl ketone (MEK), alcohols, chlorinated
hydrocarbons, and other organic compounds. An indication of the
relative solubility of different solvents can be obtained by
comparing their Kauri-Butanol values. A Kauri—Butanol value is
a measure of how well a solvent solubilizes a standard solution
of Kauri resin in butanol. A high Kauri—Butanol value indicates
strong solvency properties. In general, chlorinated solvents
have high Kauri-Butanol values when compared to other organic
solvents. Table 2.6 lists Kauri—Butanol values for the
chlorinated solvents under study, as well as other common
solvent characteristics.
-------
Table 2.6
SOLVENT PROPERTIES
Chemical MEH TCA CFC-113 PERC ACETONE MEK Water Xylae
Formula Ch Cl CH CC]. CC]. FCC1F CC]. Ccl CH COCH CH COC H H 0
Molecular Weight 85 133.4 187.4 165.9 58.1 72.]. 18
Boiling Point (°F) 104 165 118 252 133 175.4 212 140
Denaity (g/cc 70°F) 1.33 1.33 1.57 1.62 0.80 0.81 1.0
Solubilityin Water 2 0.07 0.02 0.02 100 27 N/A
(wt % @ 70°F)
Kauri—Butanol Value 136 124 31 90 95
Flash Point None None None None -15 23 N/A 79
(°F, Closed Cup)
Vapor pressure 300 125 18 185.5 70.9 17.5
(MMhG @ 20°C)
Solubility Parameter 9.7 4.64 7.3 9.3 10 9.3 23.4 8
-------
— 25
In some products, the solvent itself performs as an
active ingredient. TCA and METH are widely used in aerosol
products because they possess unique characteristics that
improve product quality, efficacy and safety. When used with
flammable hydrocarbon propellants, TCA and METH serve to reduce
the flammability of the product so that it can be safely used.
They solubilize compressed gas propellants, and depress the
vapor pressure of flammable propellants, allowing more
propellant to be put in the container. This results in more
uniform spraying, less waste and longer product shelf life.
Both TCA and METH are especially good solvents for use with
CO 2 propellants.
TCA and NETH have excellent solvency properties, and
can solubilize most active ingredients. Most resins, such as
those in paints and adhesives, are not highly soluble in water
or in the common hydrocarbon propellants. They are often
soluble in chlorinated solvents, especially METH, which accounts
for its use in aerosol adhesives and spray paints. Chlorinated
solvents keep solid materials in suspension longer than other
solvents. This helps to improve spray characteristics and
reduce valve clogging (Cuzic, 1989). Chlorinated solvents
provide more weight to each package because they are denser than
most other commonly used solvents. Higher density is
beneficial because it retards product settling, and also helps
prevent valve clogging.
Because of its high evaporation rate, METH is a very
good solvent for products that require quick drying properties.
Rapid drying improves atomization, yielding smaller particles
and a drier spray. TCA evaporates a little more slowly, and is
useful in products that require less rapid drying. PERC
evaporates even more slowly, and is used in products where it is
desirable to have the product stay on a surface longer.
Automotive engine degreasers, for example use PERC.
-------
— 26 —
Of the chlorinated solvents used in aerosols, only METH
is used as a component of the propellant System where it serves
as a propellant co-solvent. There are several grades of METH
available. One has an inhibitor added to Scavenge hydrochloric
acid (HC1) that may form when METH reacts With water or
degrades. This is designed to be used in aerosol containers
susceptible to corrosion and leak formation, or where
contamination of the product could result.
The use of METH grew steadily from 1973 through 1982.
Part of this growth was a result of the decrease in CFC
propellant use. As hydrocarbon propellants replaced the CFCs,
more chlorinated solvent was used to solubilize the
propellants. Since 1982, METH use in aerosols and particularly
in personal care products has declined. Reasons for this
decline are discussed in the subsection on regulatory trends.
There are two general types of aerosol mixtures. A
homogeneous aerosol mixture is one in which all of the
components are mutually soluble under pressure. To achieve
this, more than one solvent is usually required. A homogeneous
aerosol is a single, liquid phase mixture in equilibrium with
the vapor phase in the container. The package does not need to
be shaken before use. Examples of homogeneous aerosols include
hair sprays and some insecticides (Geigel and Miller, 1985).
In a heterogeneous aerosol mixture, the ingredients are
not mutually soluble. This type of aerosol product usually
consists of three or more phases in a single container.
Examples of heterogeneous aerosols include antiperspirants, foot
sprays and other powdered products. Powder aerosols are more
difficult to produce than other aerosol types. Sedimentation
and agglomeration of the powder ingredient within the liquid
phase can occur, both in the container and as it is dispensed.
Powders can clog the valve system and cause leaks. Formulation,
-------
— 27 —
selection of hardware and the filling process must be carefully
examined to successfully formulate powder aerosols. Powder
sprays consist of an insoluble powder (active ingredient)
suspended in a liquid propellant. Oily dispersants and
suspending agents aid in keeping the powder dispersed.
An emulsion is a mixture of two or more liquid phases,
usually an oil phase and water, in equilibrium with a vapor
phase. Oil-in—water emulsions are dispensed as foams. The
prope .lant is present in the internal, dispersed phase. Wh n
the product is released from the container, the propellant
vaporizes and expands into the continuous, aqueous liquid
phase. Shaving lather is an example of an emulsion foam.
U.S. AEROSOL PACKAGING FACILITIES
Aerosol packing facilities are identified in SIC Codes
relating to the products they fill (i.e., paints, chemicals,
cleaning preparations, etc.), in SIC Code 3411, Metal Cans, or
or in SIC Code 7399, “Business Services Not Elsewhere
Classified”. Large fillers usually produce a wide variety of
products from several product categories. Smaller fillers, and
some dedicated facilities owned by large companies, may fill as
few as one or two product. types. Custom fillers are
manufacturers that fill aerosol packages for other companies.
A 1984 industry survey indicated that there were 217
companies in the U.S. with aerosol filling lines (CSMA, 1985).
According to that survey, 130 of these facilities are reported
to account for 88 percent of the total number of containers
filled. In a study done for the U.S. EPA, 212 aerosol fillers
were identified (Martinez, et al, 1987). Another EPA report
states that there are 117 U.S. fillers. let another study for
EPA reports 215 aerosol packing plants (ICF, 1988). A more
recent industry survey conducted by CSMA reports that 151
-------
— 28 —
fillers represented 94 percent of all the Containers filled in
the U.S. in 1988 (CSMA, l989a).
Table 2.7 lists 104 U.S. custom fillers, their
locations, and the types of products they package. Other
manufacturing plants that fill their own aerosol products are
not included in the list.
Within Southern California, U.S. EPA reports that there
are 17 aerosol packers. Other sources count 22 facilities in
Los Angeles County, 5 in Orange County, and 2 in San Bernardino
County.
THE FILLING PROCESS AND SOURCES OF RELEASES
Aerosol container filling is an automated process. A
schematic diagram of an aerosol filling line is shown in Figure
2.3. Empty cans, open at the top, are automatically fed from
cartons or skids onto a conveyorized belt. Product components,
such as active ingredients and other additives, are mixed in a
batch mixing tank according to a recipe or standard procedure.
The individual ingredients are poured into the mixing tank from
drums, or are transferred from bulk storage tanks. After
thorough mixing, the material is pumped to the filling station.
Alternatively, a portable mixing tank can be moved directly next
to the filling line.
At the first stop in the filling line, a controlled
amount of the product mixture is dropped into the empty aerosol
can. The feed nozzle is positioned directly over the container,
or is inserted inside the container. The conveyorized belt
moves the partially filled containers to the next station,
located in close proximity to the first. At this point, solvent
is pumped from bulk storage and a small amount is added to the
can. Immediately afterwards, the valve assembly is inserted
-------
— 29 —
into the container. Because the three steps of active
ingredient addition, solvent addition and valve insertion are
done in quick succession, very little product or solvent vapor
escapes to the atmosphere.
Some fillers add the solvent directly to the mixing
tank with other product components. This practice can result in
increased solvent vapor emissions, since there is more
opportunity for volatilization to occur during material
transfers. Sometimes custom fillers receive premixed
concentrates in bulk storage containers shipped to them by their
customers. For these products, on-site blend make—up does not
occur.
After the valve assembly has been placed in the
container, the can passes to a crimping station where the valve
is mechanically tightened to the container. Next, the can is
conveyed into a separate gassing room where the propellant is
added. Aerosol propellants are stored in large bulk storage
tanks outside the facility. In the case of flammable
hydrocarbon propellants, the tanks are located a distance from
the building. Gassing rooms are explosion proof rooms designed
to be used only for adding propellant to the containers. No
other operations are conducted in the gassing room. The
propellant is charged under pressure into the small valve
opening on the container. Then, outside the gassing room, the
valve is covered and product filling is completed. Cans are
automatically weighed; underfilled or overfilled cans are
removed from the conveyor. The filled containers are conveyed
through a water bath heated to 180’ to 200’F, to test for
leaks. Cans are visually inspected for leaks as they pass
through the water bath, then cleaned and dried. Labels and caps
are attached and the cans are packed for shipment.
-------
Empty
Ca ii S
On to
Conveyor
Filled,
Capped
Cans to
Labelling
and
Shipping
Crimping
Station
FIgure 2.3
Concentrate
Propellant
(from
Concentrate
Filling
Station
Valve
Insertion
bulk
storage)
Over &
Under
Weight
Cans
Gassing
Room
--
Leak Test
Button and Water Bath Weight
Cap Inserters Check
C
I
Simplified Schematic of Aerosol Filling Line
-------
— 3]_ —
There are very few releases or waste streams from the
filling process. As mentioned above, the cans are covered
almost immediately after active ingredients and solvent have
been added. Vapor emissions from this step are negligible.
Most filling lines are equipped with local exhaust systems that
pull air away from process equipment and vent directly to the
atmosphere. There may be fugitive emissions from storage tanks,
transfer equipment and lines, but these are considered
insignificant.
The only place where substantial solvent vapor
emissions would be likely to occur is during mixing--only in
those cases where solvent is added into the mixing tank.
Available data indicate that most mixing operations do not have
dedicated process vents or vapor control equipment (Martinez et
al, 1987). This may be especially true for smaller operations
(Pfetzing, 1987)
There is a potential for waste generation in the hot
water bath leak testing procedure. Any material that does leak
from a can will remain in the bath or may volatilize into the
atmosphere. When asked during the site visits where the waste
water is discharged to, most operators didn’t know or stated
that the water goes to the sewer. They did not feel that
contamination of the water was a problem. Some aerosol fillers
treat the water from the leak test bath. One filler uses a
filter system originally designed for the electroplating
industry. Water from three baths is routed through one filter
to remove particulates before the water is sewered. The
contaminants are collected on diatoinaceous earth, which is
periodically replaced. The system works best for particulate
collection, but does not collect organic solvents very well.
-------
— 32 —
Cans that are underweight are filled or discarded, and
overweight cans are discarded. One industry expert estimates
that 0.75 to 1.5 percent of all filled containers are discarded
either by the filler or the distributor (Johnsen, 1989). Other
reasons for discarding a can include leakage, appearance
defects, past expiration date, or because it was only a sample.
For the largest fillers, this source estimates that 4,000 to
10,000 cans per day are discarded prior to sale, totalling
28,000,000 units a year industry—wide.
When aerosol cans are to be discarded, they are
typically punctured, emptied, and the liquid contents are
collected. [ ed. note: While this is commonly done, puncturing
is technically prohibited under the Federal Insecticide,
Fungucide, and Podenticide Act (City of Los Angeles, 1990).
With few exceptions, the empty, drained cans are not considered
hazardous waste. If, however, the cans contained listed
hazardous wastes such as ?4ETH or PERC, the liquid contents must
be collected and sent for incineration. There are several
devices on the market designed to puncture aerosol cans and
collect the liquid contents. In one kind of unit, discarded
cans are conveyorized to a crusher/roller device. As cans are
crushed, the propellant gases are exhausted upward through a
short stack. A blower installed in the stack dilutes the
propellants with air and releases them to the atmosphere.
Liquids from the containers are collected below the device,
sometimes diluted with water, and drummed to storage. The
shredded cans are hauled away for disposal. The cans may or may
not be rinsed with water. If the cans previously contained
hazardous waste constituents, the rinse water may be hazardous.
Some fillers use a simpler’ method to discard unwanted
containers. The cans are simply punctured in the bottom with a
nail or other sharp object. Liquid is collected and drummed.
Propellant gas and other vapors are emitted to the atmosphere.
-------
— 33 —
This method could be dangerous to use with flammable propellants
or products. There is also some danger of explosion even with
nonflammable constituents. Some local air districts may require
air pollution control permits for releasing the pressurized
contents of aerosol containers.
Discarded cans that are not punctured may be classified
as hazardous waste if their contents are flammable according to
U.S. Department of Transportation (D.O.T.) standards. About six
percent of all waste generated at an aerosol packing facility
(other than sewage) is estimated to be hazardous waste (Johnson,
1989)
By far the most significant releases of solvents occur
during aerosol product use. All of the components of an aerosol
product are intended to be dispersed when the product is used.
Whatever volume of material remains in a container is discarded
as waste along with the container. Consequently, the entire
volume of chlorinated solvent consumed by the industry is
eventually released to the environment. According to one study
that looked at METH and TCA use in aerosols, 90 percent of both
the solvents consumed in aerosols are released to the air,
including releases from consumer product use (IEI, 1988).
COSTS
Aerosol packaging contributes to the final market cost
of a product, perhaps more so than other container types.
Consumers are willing to pay more for aerosols because of the
convenience that aerosol packaging provides. Two industry
studies have estimated the cost contribution of aerosol
products. A report by Geigel and Miller (1985) estimates that
the can itself accounts for 27 percent of the manufactured cost
of a product; the valve system, gasket and other hardware
components cost an additional 20 percent. The remainder of the
-------
— 34 —
cost is split between the concentrate (33 percent) and labor and
overhead (20 percent). Of the concentrate, solvents represent a
third of the cost. Chlorinated solvents are reported to
contribute about half the total solvent cost. Raw material,
labor and overhead costs vary according to product, operating
practices, and facility.
Another report (ICF, l987a) provides similar estimates,
although it does not break down concentrate costs between active
ingredients and solvents. The results of both studies are
summarized in Table 2.7.
RESULTS OF SURVEYS
SRRP staff visited four aerosol packers in the Southern
California area. In addition, questionnaires were sent to
approximately 100 fillers in the U.S. Nine responses were
received, eight of which were from fillers that use chlorinated
solvents. In total, then, detailed information was collected
from twelve facilities that use chlorinated solvents. The
plants range in size from three employees to more than three
hundred. Yearly production ranges from 25,000 to 40 million
units filled, generating from less than one million to more than
40 million dollars in annual sales.
Products filled by these facilities included personal
care products, automotive and industrial products, paints and
coatings, foods, household products and insecticides. Use of
all five chlorinated solvents was reported by the questionnaire
respondents. Although some respondents did not provide
information on volume of solvent used, the data we have
indicates that solvent usage varies widely by plant, depending
on the products filled and their formulations and on operating
-------
— 35 —
Table 2.7
MANUFACTURING COST COMPONENTS
Percent of Total Cost
I II
Can 27 < 35—40
Other Hardware 20
Concentrate 33 35_45
Labor and Overhead 20 20-25
Sources: I — Geigel and Miller, 1985; II — ICF, 1987a.
-------
— 36 —
practices. Some facilities, for instance, use TCA to flush
transfer lines and clean tanks. Other facilities have dedicated
equipment and don’t need to clean after each batch. In general,
though, solvent usage ranges from 500 gal/yr to more than five
million gal/yr.
For the facilities that provided data, TCA usage was
the highest overall, followed by PERC, METH and TCE. CFC-113
usage was reported to be the lowest. This information reflects
only those plants that provided chemical usage data. Of those
that did not, at least two are high volume manufacturers that
potentially use large quantities of chlorinated solvents.
SRRP’s estimates (Table 2.4), based on communications with our
industry contacts and literature review, indicate a different
distribution of solvent consumption than the questionnaire
responses revealed. In both cases, TCA consumption is highest.
It is estimated that much more MET!-! is used than PERC, that
CFC-]i.3 use exceeds TCE, and that TCE consumption is very
small. These differences can be explained by the fact that the
questionnaire responses were voluntary and random. SRRP’s
solvent consumption estimates are “top-down” estimates based on
total solvent production, and allocation among various
industries. No facilities that were visited use TCE; this is
attributed to the stringent air quality regulations in Southern
California.
Waste generation appears to be primarily dependent on
operating practices, and secondarily on production volume.
Facilities that manufacture only one product do not need to
clean their process equipment unless it has somehow become
contaminated. Some multiple product facilities have dedicated
process equipment so that frequent cleaning is avoided. Others
clean equipment with TCA, a process that generates significant
volumes of liquid waste that is sent to an incinerator or cement
-------
— 37 —
kiln for destruction. Four facilities reported sending their
liquid waste to an off—site solvent recycler. Two reported
sending their waste to a landfill. One of the facilities we
visited uses a detergent/water blend to clean process equipment,
and alcohol to dry the equipment. The plant does not generate
liquid hazardous waste.
At the facilities that responded, generation of
off—spec or damaged cans amounts to 50 to 210,000 pounds
annua .ly. Two of the smaller plants reported that they dispose
of waste cans in the trash. Most of the others ship the cans or
their contents to a solvent recycler or to an incinerator or
cement kiln. One plant sends the waste cans to a landfill, but
sends the contents to a solvent recycler.
None of the plants is equipped with vapor control or
recovery devices. In some cases, the gassing rooms, filling
stations and mixing tanks are equipped with local exhaust that
vents to the atmosphere. Some plants have no local exhaust, and
only one had vents for all three areas. When asked if they had
contaminated water from a leak test bath to dispose of, five
plants reported yes and five said no. Of the facilities that
answered yes, one said that the water goes “down the drain”.
Another explained that the waste water passes through a
clarifier to the sewer. One plant that produces exclusively
spray paints said that the paint is not water soluble. It
floats on top of the leak test bath and is periodically skimmed
off.
REGULP TORY TRENDS
In this section, the discussion focuses on regulatory
trends that impact the aerosols industry and, more specifically,
solvent usage within the industry. These regulatory issues fall
into the general areas of air quality and other environmental
-------
— 38 —
concerns, consumer health and safety, and worker health and
safety.
In the 1970s, scientific studies first determined that
emissions of CFCs contribute to depletion of the earth’s
stratospheric ozone layer. In 1978, the U.S. EPA banned the use
of CFCs in aerosols except for certain applications determined
to be essential. In the 1960s and early 1970s, CFCs were the
propellants of choice for certain personal care products,
particularly hairsprays, deodorants, and antiperspirants.
Hydrocarbon propellants were used for most other aerosol
products. Because the hydrocarbon propellants are flammable,
the CFCs were preferred in personal care products where consumer
safety was of special concern. Prior to the regulatory ban,
however, many aerosol producers had already switched away from
CFC propellants. In response to concern about the role of CFCs
in the environment, consumers bought less aerosol products and
favored nonaerosol type packaging. By 1978, 98 percent of all
aerosol products in the U.S. did not contain CFCs.
Among the uses determined to be essential, and for
which CFCs could continue to be used were products for specific
industrial and military electronics applications and other
miscellaneous industrial applications (such as mold release
agents).
In accordance with the Montreal Protocol, EPA has
promulgated regulations that, in July 1989, capped production of
fully halogenated CFCs at 1986 levels. Future production will
be further reduced to 50 percent of the 1986 levels by 1998.
CFC producers have agreed to phase out production of fully
halogenated CFCs, including CFC-113’, by the end of the century.
Approximately two percent of U.S. aerosol products contain
-------
— 39 —
CFC-113. Therefore, the CFC regulations and phase out will not
have a significant impact on the U.S. aerosols industry.
More recently, state and local regulators have
developed or proposed stringent VOC emissions regulations for
many industries and sources. Some air quality agencies have
specifically listed aerosol products as a target for VOC
emissions reduction regulations. In California, both the
California Air Resources Board (ARB) and the South Coast Air
Quality Management District (SCAQMD) have expressed the intent
to regulate aerosols. To date, SCAQMD has postponed the
development of aerosol products regulations, pending regulatory
action at the state level. ARB has passed regulations that,
over the next few years, severely limits the VOC content of
aerosol antiperspirants and deodorants. This rule will have
limited impact on chlorinated solvent consumption since these
products do not contain chlorinated solvents. The new rule also
specifies that the VOC content of these products cannot be
replaced with ozone depleting substances (such as CFC—113 and
TCA). There is speculation by many in the industry that this
kind of regulation will be expanded to other aerosol products in
the near future.
In 1988, the State of New Jersey proposed to limit the
VOC content of air fresheners, disinfectants and consumer
insecticides. When the final regulations were promulgated, only
air fresheners were affected, 75 percent of which are estimated
to be in aerosol form. Similar regulations were proposed in New
York and Dallas/Fort Worth, Texas.
Attempts to control VOC emissions from aerosol products
affect chlorinated solvent usage indirectly. VOC reductions
will require that aerosol producers either reformulate their
products or discontinue them. Reformulation impacts solvent
-------
— 40 —
usage directly, since any change in propellant or active
ingredients requires a change in solvent composition or
concentration. In some products, use of hydrocarbon propellants
or solvents can be reduced by replacing the solvents with TCA.
While the outcome of these imposed changes remains uncertain,
there is a possibility that one result may be an increase in the
consumption of chlorinated solvents, particularly TCA, in
aerosol products.
In California, METH was recently identified as a toxic
air contaminant. As a result of this determination, METH use
and emissions will be strictly controlled throughout the state.
ARB is in the process of developing regulations that are likely
to severely curtail METH use in California. PERC is currently
under consideration as a toxic air contaminant. A determination
as to its status is expected shortly. Other states across the
country are considering similar action.
For some aerosol products, TCA is a potential
substitute solvent for METH. The future of TCA, though, is
questionable at present. There is some uncertainty whether the
regulation of the fully halogenated CFC5 and halons will be
sufficient to prevent significant damage to the ozone layer. In
response to this concern, the U.S. EPA recently published an
Advance Notice of Proposed Rulemaking (ANPR) stating their
intent to evaluate TCA as an ozone depleter. This could mean.a
freeze on TCA production or total phase out at some future
date. EPA has taken no further action on the ANPR to date.
On another front, EPA has requested submission of
available health and safety data from pesticide registrants
whose products contain METH. Some f these pesticides are
supplied in aerosol form.
-------
— 41 —
Two federal agencies have taken action on consumer
products containing METH. The Consumer Product Safety
Commission (CPSC) considered a ban on consumer products
containing METH. In December 1987, the commission voted not to
restrict METH use in these products, but rather to work with
industry to develop warning label standards and a consumer
education program. The warning label requirements are in place,
and a consumer education program has been implemented. Aerosol
products most affected by this requirement are spray paints and
furniture strippers, both of which commonly contain METH. In
1988, the CPSC’s actions were challenged in court by a consumer
group favoring a complete ban on METH in consumer products. The
District of Columbia Circuit Court of Appeals rejected the
group’s challenge.
In December 1985, the U.S. Food and Drug Administration
(FDA) proposed a ban on METH in cosmetic products, particularly
hair sprays. Their proposal was based on results of scientific
studies that indicate inhalation of METH causes cancer in
laboratory animals. The ban became effective on August 28,
1989.
METH and PERC are listed on California’s Proposition 65
list of chemicals known to the state to cause cancer, birth
defects or other reproductive harm. Their listing requires that
products containing METH or PERC be labeled. In addition,
businesses cannot knowingly discharge these chemicals into any
potential sources of drinking water and must provide warning to
individuals exposed to significant amounts of the chemicals.
similar legislation is pending in other states.
In the area of employee health and safety, recent
action by the federal Occupational Safety and Health
Administration (OSHA) affects employers using chlorinated
solvents. The permissible exposure limit (PEL) for PERC was
-------
— 42 —
lowered from 100 ppm to 25 ppm. OSHA has announced its intent
to lower the current PEL of 500 ppm for METH.
There are national, regional and local building and
fire codes that stipulate how and where aerosol products are
filled, stored and displayed. Manufacturing facilities, storage
warehouses and retail stores are subject to these codes which
may take the form of industry standards or enforceable
regulations. These codes must be considered when evaluating the
costs of reformulation strategies that will change the
flammability of products.
INDUSTRY TRENDS
A significant decline in the number of pressurized
products filled occurred in the years 1976 through 1978, and
again in the period 1980 through 1982. Since 1982, product
sales have steadily increased. The number of aerosol packages
filled increased 6.8 percent in 1988 over 1987 figures (CSMA,
1989a). Much of the gain has come from market segments that
don’t traditionally have chlorinated solvents in the products.
Significant growth has occurred in personal care products, most
notably in new hairsprays, antiperspirants and saline sprays for
contact lens care.
One reason for the continued growth in product sales is
an increase in consumer preference for “do-it-yourself”
products. Consumers are buying more automotive and industrial
products for home use. CSMA also attributes growth to
“continuing consumer confidence in the aerosol form of
packaging” because it offers convenience and performance, and a
healthy econOmY in general (CSMA, 1989a).
-------
— 43 —
Sales of aerosol food and pharmaceutical products are
expected to increase in the future, although aerosol food
products have been more widely accepted in Europe than in the
u.s. These products don’t traditionally contain chlorinated
solvents.
New packaging and delivery systems have been introduced
and others are under development. The new technology permits a
wider variety of products to be packaged and dispensed in
aerospi form, including pharmaceuticals, cosmetics, foods a id
toiletry products (CMR, 1989c). Some of the new developments
allow a wider range of propellants to be used, or in at least
one case, no propellant at all. New developments that influence
chlorinated solvent usage are discussed in Section III.
-------
— 44 —
III. SOURCE REDUCTION OPTIONS
This section examines source reduction options that, if
implemented, could reduce solvent usage and emissions in the
aerosols industry. Four categories of potential options have
been identified and summarized in Table 3.1. They are process
modification, solvent recovery and reuse, product substitution
and chemical substitution. Each of these alternatives is
addressed in a separate subsection.
PROCESS MODIFICATiON
One way to reduce solvent use and release is to modify
production processes. In the case of aerosol packing, process
modification provides limited source reduction opportunities.
Solvent emissions during filling operations are minimal. Very
little emissions reductions would be achieved by changing the
filling process.
Two options are considered here for their potential to
reduce solvent releases. The first is a method to minimize the
disposal of over- or underweight cans. The process of filling
aerosol cans with product virtually never results in over- or
under—filling. Adding propellant to the cans is less precise
(Flanner, 1988). Cans can be over or under filled with
propellant or damaged and crushed by the propellant filler.
Some of these problems can be corrected. Excess
propellant can be released from overweight cans, and propellant
can be added to under filled cans. This reduces the volume of
cans discarded because they are out of specification. According
to Flanner (1988), this practice can be more cost-effective than
disposal for some plants.
-------
45 —
Table 3.1
SOURCE REDUCTION OPTIONS
Process ModifiCatiOfl
Correction of Over and Under Filling
Improved Maintenance Practices
Solvent Recovery and Reuse
Vapor Recovery: Refrigerated Condensation
Carbon Adsorption
Liquid Solvent Recovery
Product Substitution
Non-aerosol Packaging
Chemical Substitution
Hydrocarbon Solvents
Water
Reformulate Active Ingredients
Recycled Solvents
Alternative Propellant Technologies
-------
— 46 —
Improved Maintenance Practices
Another way to reduce solvent use is to identify and
control fugitive emissions sources. Fugitive vapor emissions can
be controlled and minimized through improved maintenance practices.
Frequent leak detection inspection and regular equipment
maintenance should be included in normal plant operating
procedures. The study by Martinez, et al (1987) specifically
identifies periodic leak detection and equipment repair as an
effective means to reduce fugitive emissions. Achievable emission
reductions were estimated to be 59 percent to 73 percent for
various types of process equipment. Leak detection requires
additional manpower, but little or no capital equipment purchases.
Automatic sensors distributed around the plant that trip an alarm
when a leak is detected can also be helpful in minimizing solvent
releases from a leak.
Total solvent consumption can also be reduced by changing
filling equipment cleaning procedures. The use of dedicated
transfer mixing and equipment eliminates the need for thorough
cleaning between process batches. Alternatively, cleaning solvent
can be collected and reused many times before it is disposed of.
Solvent life can be prolonged by filtering out solids after each
cleaning cycle. A third way to reduce chlorinated solvent
consumption for cleaning is to use another solvent or water
instead. One plant we visited cleans process equipment with water,
followed by an alcohol rinse to dry.
-------
— 47 —
SOLVENT RECOVERY A!’ D REUSE
Chlorinated solvents can be recovered from
manufacturing processes in three ways. Liquid solvents can be
collected from the contents of scrapped cans that are crushed
arid emptied before being shipped off—site. Solvent vapors that
are emitted during the filling process can be collected and
condensed. Solvents used to clean process equipment can be
collected and reused. All three techniques reduce the amount of
chlorinated solvent sent off—site as waste. The recovered
solvents can be used on-site in the formulation of new products
or to clean process equipment. Alternatively, the collected
solvents may be sent to an off—site recycler. In this
subsection, we discuss solvent recovery techniques and solvent
reuse for cleaning. In the section to follow on chemical
substitutes, we address the issue of using recycled solvents in
aerosol products.
A study of chlorinated solvent emissions from aerosol
filling operations determined that significant solvent emissions
occur only in the mixing step (Martinez et al, 1987). The study
included emissions data from eleven aerosol facilities that use
PERC, NETH and TCE (one facility). The data were provided in
response to a U.S. EPA request for information under Section 114
of the Clean Air Act. Emissions can occur as solvents or other
ingredients are added to the mixing tank and while mixing takes
place. Releases can also occur when the mixed blend is
dispensed into the empty aerosol cans. There are fugitive
emissiOns from transfer equipment and storage tanks as well.
While the study did not include information on the
sizes and production capacities of the facilities that
responded, it did provide emissions estimates. As noted above,
mixing operations generated the highest emissions, on the order
of 200 to 18,100 kg/yr per facility (1986 and 1987 data).
-------
— 48 —
Emissions from solvent bulk storage tanks ranged from 200 to
6,600 kg/yr per plant. One facility that stored solvent in
pressurized tanks estimated that it had no releases from
storage. Equipment leaks, from valves and pumps for instance,
were 100 to 4,200 kg/yr per facility. Since the report was
generated for a specific, limited purpose, it is not clear that
these data would be representative of the industry as a whole.
In the study, refrigerated condensation was identified
as a control measure for reducing solvent emissions by 95
percent from storage tanks and mixing operations. Carbon
adsorption was identified as an alternative control measure for
mixing operations that could also achieve a 95 percent reduction
of emissions over systems without such control measures. Both
options, which are discussed further below, have the potential
for solvent recovery and reuse on-site or by an off—site solvent
recyci er.
Refrigerated Condensation
Refrigerated condensation involves installing ductwork
from the mixing tank or storage tanks to route emissions to the
condenser. Collected vapors are refrigerated to below their dew
point, generally using CFCs as refrigerants. As the vapor
stream cools, chlorinated solvents, water and other vapors
present from the mixing condense. The liquid mixture can be
shipped off-site as hazardous waste for treatment or
incineration. Alternatively, the chlorinated solvent can be
separated out of the liquid mixture either on site or by a third
party recycler. Both METH and TCA contain stabilizers that are
water soluble and separate from the solvent in the condensation
process. For reuse on-site, METH or TCA needs to be reblended
with stabilizers. TCA hydrolyzes in the presence of water,
creating hydrochloric acid that can damage the condenser and
cause additional waste handling problems. There may also be
-------
— 49 —
problems with any other vapors (i.e. active ingredients) from
the mixing or filling operations. Depending on what the other
chemical constituents are, fractional distillation might be
necessary in order to separate the various components.
Carbon A sorttion
In a carbon adsorption unit, vapors collected from the
process area are passed through a carbon bed. Organic materials
are adsorbed on the carbon. When the bed has adsorbed its
capacity of organics, the bed is heated, usually with steam.
The orgaflics are desorbed from the carbon into the steam. The
steam is condensed; and the organic material can be separated
from the water. As with refrigerated condensation, there are
difficulties in restabilizing the NETH or TCA for use on—site,
and there are problems with TCA in water streams. Carbon
adsorption is thus not a feasible option for recovery of TCA.
It can be used to capture PERC, and, with some difficulty, to
recover NETH.
Another problem associated with the use of carbon
adsorption is that the efficiency of the Unit depends on the
concentration of solvent vapor in the gas stream. Given the
relatively low levels of solvent emissions, whether NETH or PERC
is present, the unit efficiency will be low and operating costs
will be high.
A carbon adsorption unit transfers organic materials
from an exhaust (air) stream to the carbon bed and then to a
water stream. Some residual solvent or product active
ingredient will remain on the carbon after it has been
desorbed. The carbon can only be regenerated a finite number of
times and will then need to be disposed of as waste. Depending
on the chemicals present and their concentrations, the carbon
may be hazardous waste. There are also water pollution concerns
-------
— 50 —
to be addressed. After solvent is removed from the condensed
water stream, residual contaminants remain in the water.
Depending on local sewer regulations, the waste water may
require treatment before it can be released to the sewer.
Capital and operating costs are major obstacles
impeding the implementation of refrigerated condensation and
carbon adsorption in many aerosol packaging plants. Given the
relatively low volume emissions that are present, both methods
can be prohibitively expensive. Technical problems also exIst.
Neither method is designed for capturing TCA. PERC is easier to
recover, and doesn’t require the addition of stabilizers or
inhibitors, but much less of it is used in the aerosols
industry. Reblending stabilizers into recovered METH and TCA
for on—site use may be too complicated for many facilities to
adopt. For instance, some of the filling plants we surveyed
have ten or fewer employees and may be unlikely to devote
manpower resources to reblending operations. These facilities,
though, could conceivably send the recovered liquid to an
off—site recycler for further processing.
Solvent Recovery From Can Disposal
When off—spec or damaged cans are crushed, the liquid
contents are collected. The liquids are either shipped off-site
for solvent recovery, incineration or for use as supplement
fuel. If kept on-site, the liquids can be reused in two
different ways. First, if the concentrate came from only one
type of product or from very similar products, it may be
possible to use the recovered material directly in the
formulation of new product. In some cases, reuse of the
concentrate may require a primary treatment step to remove
unwanted components. If the separation is complex, for
instance, if fractional distillation would be required, it is
very unlikely that fillers would adopt this practice. Another
-------
— 51 —
way to reuse the recovered concentrate is to use it for cleaning
process equipment. In this way, less virgin solvent would be
needed for cleaning.
As noted in Section II, some fillers use a homemade
device with a nail to puncture waste cans, and a drum to collect
the liquid contents that drain from the cans. There are also
commercial units available in which cans are punctured within an
enclosed chamber and the liquid contents drain to a drum. The
propellant gases are released to the atmosphere. An improvement
over this method is a newer device that crushes cans within a
chamber and collects the liquid contents and 99 percent of the
propellant gases. The liquids and propellants are stored under
pressure and transferred to a pressurized storage vessel. One
manufacturer provides these devices on two sizes. The smaller
unit processes 600 to 1,000 cans per hour and the larger unit
processes 1,200 to 2,000 can per hour (Beacon Engineering,
undated).
These units are relatively expensive, and are only
cost-effective for large volume filling operations. Smaller
facilities are not likely to purchase one of these can crushing
devices. One advantage of the automated units is that solvent
(and other components of a blend) are fully contained and
collected. The “homemade” can puncturing units are open to the
atmosphere, allowing some solvent to evaporate and not be
collected, and are also dangerous to use for cans containing
flammable hydrocarbon propellants.
A supplier of can evacuation units reports that waste
disposal firms have shown more interest in the devices than have
most aerosol fillers. Waste disposal companies collect waste
cans from fillers and crush them at the disposal facility. They
charge the filler whose waste they take as much as one dollar
per container (Shields, 1989). Because of the larger volume of
-------
— 52 —
containers they process and the fees received, purchasing an
automated crushing device is much more cost-effective for them
than for individual fillers.
Solvent Reuse
Several aerosol fillers who were surveyed by the SRRP
staff reuse their cleaning solvents. In some cases, solvent
that is used to clean equipment after certain product batches is
segregated, stored and blended into the product when the next
batch of it is processed. In other cases, the solvent is simply
used over and over again for cleaning until it has become too
dirty. A representative at one facility indicated that his
plant has a dedicated, closed loop solvent cleaning system.
Solvent is used to clean equipment until it is contaminated. It
is sent to an off-site recycler and recycled solvent is
purchased for cleaning use.
PRODUCT SUBSTITUTION
Virtually all solvent emissions from aerosols occur as
the products are dispensed in use. This may be in an industrial
setting or in the home. Control of solvent vapors from aerosol
product use is, for all practical purposes, impossible to
implement. In fact, chlorinated solvents are present in aerosol
products because they enhance and influence how the active
ingredients are dispensed. At the use level, then, solvent
emission reductions can only practically be achieved by not
using aerosol products. Instead, products can be purchased and
used in other packaging forms.
Previous studies have evaluated the availability and,
to a lesser extent, the cost of non-aerosol alternatives
(Westat, 1987; ICF, 1987b). Since this information is
available, it is not recreated here for individual market
-------
— 53 —
segments and products, but rather discussed the benefits and
limitations of product substitution. Some examples are
prec ented for illustrative purposes.
Many consumer products are available in spray pump
containers. When the nozzle of a spray pump is manually
depressed, pressure is exerted on the liquid in the container,
forcing it out of the spray nozzle. Less versatility in spray
patterns is available with pump sprays. No propellant is used
in spray pumps. This reduces performance demands on the
solvent, which means that a wider variety of solvents may be
appropriate for the formulation. The choice of solvent still
depends very much on the active ingredients in the product. A
product purchased in a spray pump package can be less expensive
than the same product in an aerosol container.
Among the many products that are available in pump
sprays are hair sprays, cleaners, furniture polishes, fabric
protectors and laundry prewash sprays.
Atomizers are another package form in which the product
is dispensed as a spray. Colognes are commonly packaged in
atomizer containers. Like pump sprays, no propellant is
required. Atomizers work by mixing air with the liquid phase as
the nozzle is manually depressed. With this design, solvent
requirements may be less stringent, depending on the solubility
of the active ingredients. For both spray pumps and atomizers,
it may be necessary to retain the chlorinated solvent in the
formulation in order to suppress flammability or to completely
solubiliZe and dilute the active ingredients. For products in
which a chlorinated solvent is the active ingredient, the
solvent must be retained in the formulation, unless alternative,
more benign chemicals are available.
-------
— 54 —
Other product substitution options are also available.
For instance, drain openers can be purchased in solid rather
than aerosol form. Furniture polishes and other cleaners are
available as creams or waxes. Solids, creams and waxes may not
contain any chlorinated solvents.
There are a number of non—aerosol type insecticides on
the market. These include liquids, powders and cakes.
Nonchemical options include traps, baits and sticky tapes. The
non-aerosol alternatives work differently than aerosol
products. In some cases, they are intended for different
purposes and do not truly replace the aerosol types. In
households where small children are present, it may be unsafe to
leave traps or solid insecticide cakes where children can reach
them.
There are three general disadvantages associated with
the replacement of aerosol products by other packaging types.
The first is the loss of convenience offered by aerosol
containers. Consumers have shown that they are willing to pay
for the added convenience of products packaged in aerosol form.
The dosage and direction of product released is well
controlled. When a product such as a cleaner is supplied in a
cream, wax, liquid or solid form, it must be poured or otherwise
manually applied to a surface. Then it must be physically
removed. This can be messy, and can generate additional waste.
Non—aerosol fabric protector is an example. In its aerosol
form, a fabric protector is sprayed directly on the item or part
of an item to be treated. The product is dispensed as a mist or
spray and the solvent and propellant components evaporate. In
liquid form, the product must be poured or brushed onto the
item, or the item can be dipped into the liquid. In either
case, the product is likely to be “wetter” than the aerosol
version. It can run, drip, be messy to apply and take longer to
dry. To treat an upholstered chair, for instance, the aerosol
-------
— 55 —
product offers easier handling, more direct control of
application, and, in general, added convenience.
The second limitation to non—aerosols relates to
safety. Aerosols completely contain hazardous products like
pesticides, caustic cleaners and flammable materials. Spills
and leaks are less likely to occur than with glass or plastic
bottles, jars or other containers. With few exceptions (in the
case of leaking aerosol cans), the product is only released
inten ionally (i.e. when the valve is manually activated),
minimizing accidental exposure of consumers.
The third problem with non-aerosol products relates to
their efficacy. Some non—aerosols do not perform as well as
aerosol versions of the same products.
According to testimony submitted to Congress, pump type
products cannot replace aerosol products in many applications.
In some cases where pump sprays can substitute for aerosols,
consumers have indicated less acceptance of the non—aerosol
form. The testimony also pointed out that elderly and arthritic
individuals may have difficulty using a pump type product
(Aerosol Age, 1988a).
Nonaerosol products that can be used in place of
aerosol products often contain chlorinated solvents. Examples
are liquid and paste spot and stain removers that contain TCA
(Westat, 1987). A review of the Westat data indicates that
liquids without chlorinated solvents are available, but no data
were obtained as to the applicability of the alternate
products. Aerosol tire cleaners are another product type for
which non-aerosol alternatives exist. METH and TCA are used in
the aerosol version because they swell the rubber and make
cleaning easier. Chlorinated solvents have been removed from
the non-aerosol pump spray type container and replaced with
alcohols and 2—butoxyethanol.
-------
— 56 —
Flying insect killers are sprayed directly at live
insects to kill them. To be effective, the product must be
dispensed as very small particles suspended in a fine mist.
Pump sprays are reported to be ineffective in producing the fine
particle size required.
There are other packaging technologies under
development. One company has introduced a mechanical spray
system that is reported to perform like an aerosol, but without
the u e of a chemical propellant. As noted above, the remo al
of the propellant may increase the number and type of solvents
that can be used. The device has a twist cap that is connected
to an expandable rubber bladder. The product is pulled into a
chamber under the cap when the cap is twisted to the left. When
the cap is twisted to the right, the product is forced under
pressure down into the rubber bladder. Then, when the valve
stein is depressed, the product is dispensed as a continUOUS
spray. The system is said to be compatible with any size or
shape container. The developer plans to license the technology
for use with consumer products (Geigel and Miller, 1985).
CHEMICAL SUBSTITUTION
There are five ways by which reformulation can be
implemented to reduce chlorinated solvent usage in aerosol
products. The first two approaches are to simply substitute
another solvent or water for the chlorinated solvent portion of
a blend. The third option is to replace the active ingredient
in a blend so that another solvent can be used. The fourth
option is to use recycled solvent in place of virgin solvent.
The fifth option is to switch to a propellant that has
different, or less stringent, solvency requirement and permits
the replacement of the chlorinated solvent with another
solvent. Each of these approaches has been adopted, at least to
some extent, by the aerosols industry. All five alternatives
-------
— 57 —
present difficulties that must be overcome if they are to be
successfully implemented. There are some aerosol products for
which these chemical substitution options are not viable.
There are previous studies that have examined the
potential for chemical substitution in aerosol products; among
them are Geigel and Miller (1985) and ICF (1987a and 1989).
Some earlier studies listed formulations in terms of weight
percent of different components and, perhaps, compared costs of
the different blends. In this section general considerations of
chemical substitution are discussed, with focus on solvent
replacement and active ingredient reformulation in specific
product segments where chlorinated solvent usage is highest.
Also discussed is the possibility of replacing virgin solvent
with recycled solvent and the use of new propellants and
propellant blends that can be adopted by the industry.
General Issues
When a chlorinated solvent is present in a blend solely
because of its solvent properties, it may be relatively easy to
replace it with another solvent. Even when another solvent can
be used, it is rarely a direct 1:1 replacement. The
concentrations of other components in the blend are likely to
change as well. It is uncommon that a chlorinated solvent is
selected only because of its solvent characteristics. As
discussed in Section II, METH and TCA in particular are often
used in household, automotive and industrial products because of
their other unique characteristics as well. When a solvent
contributes other properties to a product, it becomes more
difficult to replace. Furthermore, if the chlorinated solvent
is the active ingredient in a blend, it becomes very difficult
to find a substitute for the solvent and achieve a comparable
level of product performance.
-------
— 58 —
When reformulation of a product is considered, it is
necessary to evaluate the compatibility of the new blend with
existing valve hardware and the container itself. For example,
NETH chemically attacks certain types of plastics that are often
used for making valve hardware. Aerosol products containing
METH are fitted with hardware made from plastics known to be
resistant to METH. If METH is removed from a blend and replaced
by another solvent, it may be necessary to use different
hardware. For instance, the existing hardware might react with
the new solvent. It might be unnecessarily expensive for the
new blend, or it may not provide the desired spray pattern.
When a formulation is changed, the way a product is dispensed is
likely to change as well. The evaporation rate of the product
may be different, spray characteristics are very dependent on
the composition and concentrations of ingredients in the blend.
The spray pattern may be affected, as well as the “wetness” or
“dryness” of the spray.
Product performance, stability and aging effects are
among the other factors that must be tested before a new product
can be introduced to the market. Products that are regulated by
a government agency may need to be reviewed by that agency prior
to sale. The Food and Drug Administration (FDA) regulates
pharmaceuticals, foods and cosmetics, though these are not
significant uses of chlorinated solvents. Pesticides are
regulated under the Federal Insecticide, Fungicide, and
Rodenticide Act (FIFRA), which is administered by EPA. Aerosol
disinfectants are included in this category. If these types of
products are reformulated, they must be reregistered. State
agencies have their own registration programs as well.
In the subsections below, chemical substitution of TCA,
METH, PERC and CFC—113 is discussed. For many products, use of
hydrocarbon solvents may be a chemically viable substitution
option. These solvents are flammable and their use in aerosol
-------
— 59 —
products poses increased risk of fire or explosion. There are
three areas where fire hazards need to be considered. The first
is during use, especially for personal care and household
products, and for products that may be used near a source of
ignition (near electrical equipment, for example). Second, and
of importance to manufacturers, distributors and retail
marketers is the risk of fire where flammable aerosol products
are stored or displayed in bulk. An increased volume of
flammable products in storage may require installation of
additional fire suppression and firefighting equipment. Third,
handling of flammable materials presents problems for aerosol
fillers.
It should be noted that chemical substitution can
result in increased manufacturing costs. There may be
associated equipment changes for process equipment or for
storage tanks. Additional fire suppression systems or explosion
proof equipment may be required.
personal Care Product
Chlorinated solvents are no longer widely used in
personal care products. However, a review of the history of
chemical substitution in hair spray products is useful as an
illustration of the critical interrelationships between product
ingredients, hardware selection and product performance.
When CFC propellants were removed from hair sprays in
the l970s, they were replaced by flammable hydrocarbon
propellants, often in combination with alcohol. METH was
included in the newer formulations to reduce flammability. It
also served as a drying aid and cosolvent to solubilize the
polymer resins used in hair sprays. METH has been removed from
essentially all hair sprays, resulting in increased flammability
and lower polymer solubility. New hair spray resins have been
-------
— 60 —
developed that are more compatible with the hydrocarbon/alcohol
blends. Also, water has been added to hair spray formulations
as a cosolvent to further improve resin solubility (Rouse and
Novack, 1988). It has been reported that the ethanol/water
blends dried more slowly than the METH-based hair sprays (Geigel
and Miller, 1985). This has been corrected by incorporating a
different valve system that changes the evaporation rate of the
product as it is dispensed.
The fact that hair spray sales exhibited strong growth
in 1987 and 1988 indicates consumer confidence in the
reformulated products. That growth is also considered by
industry sources to reflect changes in hair style preference.
Household Products
There are a variety of household products that contain
chlorinated solvents. In many of these, MET!-!, TCA or PERC serve
as active ingredients. Chemical substitution in these products
is more complex. With household products, toxicity is an
important consideration because there is a greater chance of
exposure to products used in the home. For household products,
then, chemical substitutes must be carefully evaluated in terms
of toxicity, chemical reactivity and product safety.
- An ICF report (1989) identif led three household aerosol
products that contain TCA for which no chemical substitutes are
available. The products are spot removers, water and oil
repellants, and suede protectors. Aerosol spot removers are
powder sprays used to remove spots and stains from clothing and
fabrics instead of laundering. PERC, METH and TCA have been
used in spot removers as active ingredients. Reports vary as to
the extent to which PERC and METH are used. The ICF report
states that TCA is the only chlorinated solvent in spot
removers. Later in the same report, though, it is stated that
-------
— 61 —
PERC, in combination with TCA, has largely replaced METH as the
primary active ingredient. It appears that the use of METH in
spot removers is declining. It remains uncertain whether PERC
or TCA are replacing METH, or if the two are used in
combination.
The degreasing action of METH, PERC and TCA make them
good cleaners in spot removers. As ICF reported, there appears
to be no chemical substitute for chlorinated solvents in aerosol
spot removers. Flammable hydrocarbon solvents are undesirable
for two reasons. First, they don’t clean as well as the
chlorinated solvents. Second, their flammability poses dangers.
Water and oil repellant applications are also
identified as essential uses of TCA (ICF, 1989). Water and oil
repellarits are used primarily for protecting textiles and
upholstery. Their use and composition differs from “water
repellaritS” used for waterproofing shoes, textiles and other
products. Water and oil repellants are composed of 90 percent
or more TCA. To replace the TCA with flammable hydrocarbon
solvents would increase risk of fire. This is especially
dangerous near combustible textiles and furniture.
Suede protectors are the third product type that
contain TCA for which chemical substitution is especially
difficUlt. High concentrations of TCA are reported in aerosol
suede protectors (Westat, 1987; ICF, 1989). Some spray shoe
polishes contain TCA or METH. There are aerosol shoe polishes
without chlorinated solvents. No information as to the
alternative solvents used is available. Many other aerosol
household products have brands that contain chlorinated solvents
and brands that do not. The question to be resolved is whether
the products can be used interchangeably for the same purposes,
with comparable performance.
-------
— 62 —
Automotive and Industrial Products
Automotive and industrial cleaning products in aerosol
form often contain chlorinated solvents as the primary active
ingredient. In these products, where TCA, PERC, or METH is used
not just to solubilize other components, it is more difficult to
find nonchiorinated solvent replacements.
The ICF (1989) report identified three automotive and
industrial aerosol product categories that contain TCA and or
which no chemical substitute is available. The three product
segments are electric motor cleaners, mold releases and brake
cleaners. As in household products, chlorinated solvents are
often present as active ingredients.
PERC and TCA are good degreasing and cleaning agents.
They are used in aerosol products to remove grease and dirt from
electric motors and other electrical equipment. Most brands on
the market contain both PERC and TCA to take advantage of PERC’s
strong degreasing action and slow evaporation rate. Products
that contain only TCA are used for smaller cleaning jobs where
faster evaporation is less of a problem. No alternative active
ingredients/solvents that can be used in aerosol products of
this type have been identified. CSMA reports that one of the
reasons TCA is so useful in engine cleaners is because its high
dieleOtric strength and nonflammability allow motors to be
cleaned while running (CSMA, 1989b).
Brake cleaners are another aerosol product category
traditionally utilizing TCA and PERC as active cleaners.
Usually the two solvents are used in combination. Chemical
substitutes must have good cleaning and degreasing properties,
should be nonflammable and work quickly. No solvent cleaners
that can replace TCA and PERC in aerosol brake cleaners have
been identified. Xylene and alcohol solvents have been tested,
-------
— 63 —
but they do not provide the same degree of cleaning capability
(ICF, 1987a)
CFC-l13 is used in aerosol cleaners for the electronics
industry. It is also used in mold release products. ICF
reports that METH and TCA are the two chlorinated solvents used
in ]flC)ld releases. Aerosol mold releases are sprayed onto molds
to release products formed in those molds. High concentrations
(40 to 50 wt. %) of TCA and METH have been reported in aerosol
mold releases. The chlorinated solvents are favored for thIs
application in industrial facilities because of their
nonflammability. There are currently no substitutes for CFC—113
in aerosol electronic cleaners, mold releases and other
industrial applications (Strobach, 1988).
Other aerosol products used in automotive and
industrial applications contain chlorinated solvents. With the
exception of the three products noted above, there are existing
aerosol product alternatives formulated without chlorinated
solvents. Once again, the flammability and efficacy of these
products must be evaluated for specific desired end uses.
Sometimes the alternative formulations exhibit better
performance. Examples of successful chlorinated solvent
replacement include aerosol lubricants, carburetor and choke
cleaners and some adhesives. METH is used in some spray
adhesives because of its very good solvency and fast evaporation
rate. It also defines the spray pattern of the dispensed
product. The alternatives to METH, usually acetone, xylene and
MEK, don’t have comparably good solvency characteristics.
Paints and Finishes
There are adequate alternatives to chlorinated solvents
used in aerosol paints and finishes. Products formulated with
exhibit acceptable quality spray patterns and product solubility
-------
— 64 —
characteristics. There has been rapid substitution of METH in
spray paints since 1985. In some simpler paint formulations,
!“IETH has been replaced with acetone. The products are reported
to be more flammable, less dense and slightly runnier on
application (ICF, 1987a). For some uses, these drawbacks may be
acceptable. Solvent blends made with acetone and TCA are also
utilized. For applications that require better quality paints,
aerosol formulators have developed spray paints based on
acetone/ester solvent combinations.
Water is an acceptable solvent in some spray paints.
There are problems that arise whenever water is used in aerosol
containers. Water can cause corrosion, so containers are
generally lined with a protective coating. Water is not
necessarily a very good solvent for all of the types of
ingredients found in aerosol products. Hydrocarbon propellants,
for instance, are generally insoluble in water. Water based
spray paints containing water insoluble propellants are reported
to have some performance problems, such as low hiding power,
chemical or mechanical instability, poor finish, and foaming
(Bartlett, 1988). A later subsection on new propellants
discusses how this problem has been overcome. We are not aware
of any aerosol paints or finishes for which chemical
substitution cannot be accomplished.
• Aerosol paint strippers contain significant amounts of
NETH. One industry report indicates that no solvent
alternatives are available for aerosol strippers (ICF, 1988).
Insecticides
The four major types of insect sprays that contain
chlorinated solvents are foggers, house and garden insect
sprays, flying insect killers and residual insecticides. We
have already discussed in Section II the specialty use of
-------
— 65 —
CFC-113 on flying insect sprays used near high voltage electric
lines. This is considered an essential use of CFC—113. A
chemical substitute for CFC113 for this application has not
been found. CSMA reports a similar use of TCA in insect sprays
applied near transformers and other electrical equipment (csMA,
1989b)
Reformulation of aerosol pesticides is estimated to
take 3 years or more before the products can be marketed.
Research and development takes a lot of time, followed by
product development, product testing and EPA registration
procedures. Significant costs can be incurred in each step of
the reformulation process.
In some insect sprays, chlorinated solvents sometimes
are added as active ingredients. TCA is one of them. It is an
active ingredient not because of its toxicity, but because it
evaporates quickly and rapidly cools an insect’s body. TCA is
considered a knockdown agent in this application. CSMA states
that it is the most effective knockdown agent available for
hornet and wasp sprays (CSMA, 1989b). Water can be used as a
solvent for flying insect killers, but not in products that may
potentiallY be dispensed near electrical equipment. Water has
been used in some aerosol pesticides since the 1960s. It costs
less to use than other solvents, is less irritating if inhaled
and is nonflammable. It does not, however, serve as a knockdown
agent, but only as a solvent.
Because of regulatory pressures and concerns over the
health effects of METH, pesticide manufacturers are removing
METI-I from formulations. Total release foggers traditionally
contained high concentrations of NETH or TCA. Petroleum
disti]lates have been tested as replacement solvents. They were
determined not to be good alternatives for two reasons. First,
the products become flammable and dangerous for use. Second,
-------
— 66 —
the petroleum distillates can damage fabrics such as carpets,
upholstery and drapes. Total release foggers based on water are
being researched. When water is used, a propane and isobutane
mixture is used as the propellant. This tends to make the
product flammable. Water based total release foggers have
excessive fallout. That is, too much of the product drops out
and settles right around the container. It doesn’t get
propelled throughout the room to be treated, and therefore is
inefficient to use. Water based total release foggers are still
being researched. Research efforts are directed at product
storage stability, physical characteristics of the spray and
hardware systems, and biological efficacy of the product.
Residual insecticides are applied to areas where
insects are located. They are generally not sprayed directly at
insects to kill them. Residual insecticides contain a high
percentage of base oils and solvents with 1.0 to 2.0 percent
toxicant. Chlorinated solvents are not used as knockdown agents
or active ingredients in these products. Rather, they are
present only as solvents (ICF, 1987a). Desired characteristics
of residual insecticides are low odor, little or no tendency to
stain, and good effectiveness. As was discussed, solvent
substitution is easier to achieve in products where the
chlorinated solvent is not present as an active ingredient.
Residual insecticides are examples of this kind of
substitution. There are two solvent alternatives. The first
alternative is petroleum distillates, which are used in some
formulations. Water is the other alternative. It can be used
in two different ways. The first way is in combination with a
new propellant, DME. This use is discussed ma later
subsection. The other alternative is to use water as the
solvent in aerosol foams. The foams are reported to be safer,
more efficient, and less expensive to use. They can be
difficult to apply in hard to reach areas because a foam will
not travel as far as a sprayed mist or stream (ICF, 1987a). On
-------
— 67 —
the plus side, water based products have less odor and are
unlikely to cause staining.
Use of Recycled Solvent
The use of recycled chlorinated solvent represents a
viable source reduction option. On a technical level, there is
no reason why recycled solvents cannot be used in most aerosol
products. Recycled solvents are not widely used in the industry
because of institutional, rather than technical problems.
Ten survey respondents answered a question about their
purchases of recycled solvents. Seven said that they do not
purchase recycled solvents and three said that they do.
Information from the surveys and site visits indicates that
recycled chlorinated solvents are sometimes used in industrial
and automotive products and in spray paints. One survey
respondent said that recycled solvents are used in their
household aerosol products as well.
Respondents cited three reasons why recycled solvents
aren’t used more often in aerosol products. First, they said
that products that must be registered by the U.S. EPA or FDA
cannot contain recycled solvents. This would include products
such as pharmaceuticals and pesticides. It appears to SRRP,
however, that the regulations may not specifically prohibit the
use of recycled solvents.
The second reason why fillers are reluctant to use
recycled solvent is because solvent purity a high priority.
This is especially true when the solvent is an active ingredient
in pharmaceuticals and personal care products. There is a
general concern among aerosol fillers that the purity and
quality of recycled solvent is questionable. One way to
implement recycled solvent use in plants with this concern is to
-------
— 68 —
employ them, at least at first, in the less critical
operations. If reliable recycled solvent meeting the
specifications of critical operations could be regularly
obtained, its use could then be expanded to more operations in
the plant.
The third reason that those surveyed resist recycled
solvent use is based on cost. Several respondents indicated
that recycled solvents cost almost the same as virgin, and so
there is little economic reason to change. In order to achieve
the purity required for aerosol applications, recyclers must
process the used solvent more so than they might if it was going
to be used in a gross metal cleaning application. The further
processing increases the cost of the recycled material.
Alternative Propellant Technologies
Another way to approach chemical substitution is to
change the propellant in an aerosol formulation. The use of a
different propellant system changes the solvent characteristics
of the product, and may permit the replacement of chlorinated
solvent with another solvent or water. The development of
alternative propellant technologies has been driven primarily by
the phase out of CFCs. There are few aerosol products sold in
the U.S. that still contain CFC propellants. A consumer
preference for less flammable products has also been a driving
force to new propellant development.
Much of the new technology consists of replacements for
CFC-ll and CFC—12. There is considered to be no alternative for
CPC-113 used as a propellant (Strobach, 1988). In this
subsection we discuss four propellants that are being used more
often. The four are dimethyl ether (DME), HCFC-22, HFC-152 and
HCFC-l42b. An emerging new aerosol technology that physically
separates active ingredients and propellants is also described
in the subsection.
-------
Table 3.2
PROPELLANT PROPERTIES
Chemical DME HCFC-142b HFC-152a CFC—22 Propane Iso—butane N-butane
Formula C l ! OCH Cl! CC1F CH CHF CHC1F C H i-C H N-C H
Molecular Weight 46.1 100.5 66.1 86.5 44.1 58.1 58.1
Boiling Point (°F) —12.7 14.4 —13.0 —41.4 —43.7 10.9 31.1
Density (g/cc e 70°F) 0.66 112 0.91 1.21 0.50 0.56 0.58
Solubility in Water 34 0.5 1.7 3.0 0.06 17 20
wt%@70°F)
Kauri—Butanol Value 60 20 11 25.0
Flash Point —42 0 0 None —156 —120 —100
(°F, Closed Cup)
-------
— 70 —
Chlorinated solvents are often used in aerosol products
because they reduce flammability which arises when flammable
hydrocarbons are used as propellants. If nonflammable
propellants can be used, the need for chlorinated solvents can
be reduced or avoided. The solvency characteristics of the
reformulated product must be considered. It is sometimes
necessary to include chlorinated solvents in blends because of
their good solvent properties. The ideal propellant system,
then, is one that reduces product flammability and solubilizes
the components, or permits a different solvent blend to be
used. The four propellaritS listed above are used because they
are nonflammable or are less flammable than propane, n-butane
and isobutane. Properties of these four are listed in Table
3.2. Properties of the traditional hydrocarbon solvents and the
chlorinated solvents are provided for comparison.
Dimethyl Ether The use of DME as a propellant was
commercialized in Europe and Japan in the 1970s. It had been
tried in the U.S. during the 1940s, but there were problems with
its use and it was discontinued. DME is a strong solvent. When
initially tried, it caused most valves not to function properly
(Strobach, 1988). Since its reintroduction in the 1970s,
protective internal liners and new hardware materials that can
withstand the strong solvency of DME are available. DME is less
flammable than traditional hydrocarbon propellants. A primary
advantage of DME is its water solubility. It is 35 percent
soluble in water at 70 degrees F. Most other propellants have
very limited water solubility. Its flammability, therefore, is
offset by the ability to formulate DME with water as the
solvent, resulting in products that are less flammable.
DME/CFC-ll blends are used in hairsprays, deodorants,
and antiperspirants. (Under the Montreal Protocol, the use of
CFC-ll will be phased out. As noted above, however,
replacements for CFC-ll are available). Because of its high
-------
— 71 —
solvency, DME solubilizes many resins used in paints and
adhesives. It can function as the propellant and solvent in
some types of products. A system consisting of DME propellant
and hydrocarbon solvent replaces a hydrocarbon propellant and
METII solvent based system in an aerosol adhesive application.
The new adhesive is reported to be safer (uses less toxic
ingredients), faster drying, and economically competitive (Dunn,
1988).
DME/water blends have been developed for total release
insect sprays. These products replace the more flammable
hydrocarbon solvent based formulations. Water dropout has been
reported to be a problem with the new products. Water can be
damaging to carpets, furniture or other surfaces on which the
container is placed. Substitution with propane or ethanol in
place of a small amount of the DME reduces dropout (Bartlett,
1989). Other insect sprays formulated with hydrocarbon
propellants and chlorinated solvents can be replaced using DME.
Blends of DME with hydrocarbon propellants or with cFC-22
propellant eliminate the need for another solvent in some
products.
Water based paints have been developed that contain DME
as a propellant and hydrocarbon solvents to improve solubility
of water insoluble components. The presence of hydrocarbon
solvents and DME enhances the evaporation of water from sprayed
paint. Slow drying of water based paints is commonly cited as a
reason why they are not more widely used. The addition of DME
is also reported to reduce foaming, another common problem of
aqueous spray paints.
DNE is used in other aerosol products that
traditionally contain chlorinated solvents. Because of its high
solvency, it is used in laundry prespotters, leather and fabric
protectors, flying and crawling insect killers, flea sprays,
-------
— 72 —
paint foams, primers, and sprayable polyurethane (Bohnenn,
1988)
The use of DME is not without its drawbacks. DNE is a
flammable liquified gas. Process equipment used with DHE must
be classified as Class I, Division I, Group C. Pump gaskets and
seals must be made from materials that can tolerate DME’s
solvency. The cost to convert an existing fitting line to
accommodate DME (that is, from Group D to Group C) can range
from 1,500 to $15,000, excluding the cost to retrofit pump ,
heat exchangers and other ancillary equipment (Johnsen, 1988).
DME is photochemically reactive. It is often used in
combination with hydrocarbon solvents that may also be
photochemically reactive. The use of VOCs is discouraged and
regulated in many parts of the country.
Other Propel].ants HCFC-22 (chlorodifluoroxnethane) is a
nonflammable propellant. It can be used alone or in blends with
other propellants to reduce product flammability. It has been
used in total release foggers in place of TCA. HCFC-152a
(1,1—difluoroethane) is moderately flammable. It has also been
used in total release foggers to reduce the flammability of the
product. HCFC—142b is considered to be almost nonflammable or
t ’ery slightly flammable. The problem with all three of these
..lternatives is that they have very limited water solubility and
re poor solvents. Hydrocarbon solvents are usually used in
onthination with these propellants to improve solubility.
consequently, the VOC content of the final product may be higher
than a comparable product formulated with chlorinated solvents.
Once again, there is a tradeoff. In this case, source reduction
of chlorinated solvents is achieved by increasing the use of
photochemical smog precursors. The new alternative propel].ants
cost much more than propane and butane.
-------
— 73 —
Future A1tern tives HCFC-141b (1, i—dichloro-1-fluoroethane)
and HCFC-123 (l,l,l—triflUoro2,2diChlOrOethafle) are two new
solvents developed by U.S. CFC manufacturers. They have very
low ozone depletion potential and are not regulated under the
Montreal Protocol as are the fully halogenated CFCs. They are
currently undergoing animal toxicity testing, and will not be
comn ercially available until the test results have been
evaluated. HCFC-123 is nonflammable. HCFC-141b is considered
marginally flammable. It is not clear at this time what
poten ia1 these solvents hold for the aerosols industry.
Barrier Packs
Barrier packs are compartmented pressurized containers
in which the product is physically separated from the
propellant. There are several types, but the two major types
are piston containers and bag-in-can containers. In a piston
container, the propellant is stored below the piston and forces
product above it out of the can when the dispensing valve is
depressed. With the bag-in-can arrangement, the product is
retained inside a flexible pouch in the can. Propellant
surrounds the pouch and pushes the product out of the can. The
product in this case does not contact the container at all.
This technology can be used with products that would corrode the
inside of the can or that are highly viscous.
Shaving gels and caulking compounds are two products
most commonly packaged in barrier packs. There are two
advantages to the use of barrier packs. First, viscous products
that cannot be dispensed with traditional aerosol systems can be
packaged in barrier packs. The second advantage is that product
can be dispensed while holding the container in any position.
Barrier packs provide an opportunity for more kinds of products
to be packaged as aerosols. They do not really reduce the use
of CFC propellants. They do not result in solvent usage
-------
— 74 —
reduction. In fact, the adoption of barrier packs is not
recommended as a means to reducing VOC solvent consumption (Lim,
1989). With barrier packs, it may be necessary to use more
solvent in a product because the propellant normally in contact
with components of a blend has been removed, and additional
solvent may be required to keep solid components within an
acceptable range. The use of barrier packs can increase the
container costs by as much as two times.
-------
— 75 —
iv. A AL’ 1SIS OF SOURCE REDUCTION OPTIONS
The aerosols industry annually consumes a little more
than 57 thousand metric tons of chlorinated solvents. In
Section iii, alternative processes and techniques that have the
potential to reduce solvent consumption and releases were
identified. In this section, these options are classified into
three categories. Within each category, the options are
analyzed in terms of implementation costs and source reduction
potential.
CLASSIFICATION OF OPTIONS
Each option from Section III has been classified into
one of the three categories. The classification of options is
summarized in Table 4.1. In column one are listed the options
for which no further analysis is done. There are three reasons
why options are placed in this category. First, in some cases
cost information is insufficient to perform the analysis. A
second reason is that certain options, though technically
feasible, would result in very small reductions in solvent use
or releases if implemented. A third reason is that the
appliCabilitY of an option is very product specific. Analysis
would require a product by product approach. For the hundreds
of product types available, this would involve analysis beyond
the seope of this report.
The second category includes alternatives for which
limited analysis can be carried out. For these options, some
cost information is available, and a rough estimate of source
reduction potential can be made. For the third category of
options, a more detailed analysis in the form of case studies
was carried out.
-------
— 76 —
Table 4.1
CLASSIFICATION OF SOURCE REDUCTION OPTIONS
No Further Analysis Limited Analysis Case Study Analysis
Improved maintenance Liquid solvent Hydrocarbon
Practices Recovery and use of solvents
recycled solvents
Refrigerated Non-aerosol Alternative
condensation packaging propellants
e
Carbon adsorption Correction of over
and under filling
Water—based
formulat ions
Alternative
active ingredients
-------
— 77 —
NOFURTHER j NALYSIS OPTIONS
Four options are included in the no further analysis
category. Improved maintenance practices can potentially reduce
solvent releases by practices such as identifying and
controlling fugitive emission sources. The exact nature of
maintenance improvements adopted and subsequent emission
reductions will vary from plant to plant. Although it is a
relatively simple option to implement, actual source reduction
achieyed will be small. This option was not further analyzed.
Refrigerated condensation and carbon adsorption are two
vapor collection and recovery techniques that are widely used in
other industries, but they are not appropriate for the aerosols
industry. The amount of solvent lost to the atmosphere during
product filling is much less than the amount emitted during
product use. The solvent concentration in filling plant exhaust
streams is very low. Neither vapor recovery technique would be
cost-effective under these conditions. Also, very little
solvent could actually be recovered.
There are two reformulation strategies that can
potentially reduce chlorinated solvent requirements. The first
option is to replace solvent with water as the carrier. The
second option is to reformulate the active ingredient component
of a product. Both of these reformulation strategies are
complex, because they impact propellant and hardware selection
and can significantly change the way a product is used. In
addition, product reformulation is very product specific. In
order to estimate the source reduction potential of these
options, it would be necessary to analyze in detail each type of
product that contains chlorinated solvents, identify alternative
formulations, and develop product efficacy, toxicity and
consumer preference information.
-------
— 78 —
This analysis is not attempted in this report. A more detailed
discussion of reformulation costs is, however, provided in a
subsection to follow. One of the options evaluated is a water
based formulation, combined with DME propellant.
LIMITED ANALYSIS OPTIONS
There are two options for which we have enough
information to perform a limited cost analysis and develop a
rough estimate of source reduction potential. The options
analyzed in this subsection are the use of recycled solvent, and
liquid solvent recovery.
Recycled Solvent
As noted in Section III, most aerosol fillers do not
use recycled chlorinated solvents in their products with the
exception of some paints, automotive and industrial products.
At least one filler surveyed also uses recycled solvents in
household products. In general, though, there is reluctance
throughout the industry to replace virgin solvent with
recycled. Aerosol fillers require a high purity grade of
solvent. In order to provide this, solvent recyclers must
process recovered solvent more than they need to in order to
sell the same solvent into the metal cleaning industry. It
costs-more for them to do the additional processing. As a
result, the recycled solvent that aerosol manufacturers purchase
costs about the same as, or a little less than, virgin solvent.
There is little economic incentive for them to choose recycled
solvent over virgin. Also, because of their concerns about
solvent purity, some fillers feel that it is necessary to test
the quality of every incoming batch of recycled solvent before
it can be used in product formulations. When this is done, any
cost savings that might result from purchasing recycled material
are offset by laboratory costs to conduct the analyses.
-------
— 79
It is known that some of the virgin chlorinated solvent
used in aerosols has already been replaced by recycled, to what
extent this has occurred is difficult to estimate. In the SRRP
staff assessment, 91 percent of the total virgin TCA and 90
percent of the total virgin NETH consumed in the aerosols
industry could potentially be replaced by recycled solvent. The
details of this estimation are provided in Appendix A. This
replacement could occur without any additional cost to the
industry (not considering added costs for laboratory analysis).
It would reduce virgin solvent requirements by 18 thousand mt
for METH and 30.9 thousand mt for TCA. This represents a
combined savings of 48.9 thousand nit, or 85 percent of the
virgin chlorinated solvent demand in the aerosols industry.
While this reduction in virgin solvent demand is
technically achievable, and could cost the industry nothing to
implement, the switch to recycled solvent is not likely to
occur. The largest markets for recycled chlorinated solvents
are the metal cleaning and electronics industries. If recycled
solvents are diverted away from these markets via purchases by
aerosol fillers, these industries will be forced to increase
their consumption of virgin solvents. It would cost them more
to use virgin material than to use recycled. Also, there would
be a net increase in virgin solvent production to meet demand of
these two large markets.
J iauid Solvent Recovery
If aerosol fillers do not use recycled chlorinated
solvents in their products, there is still an opportunity to
achieve on-site source reduction. Fillers can evacuate unusable
cans on-site in a self-designed unit or in a purchased unit.
Liquid contents are collected and reused on-site, or sent
off—site to a cement kiln, incinerator or solvent recycler. In
-------
— 80 —
this analysis, we will determine the costs to crush waste cans
on-site with a purchased unit, and either reuse the solvent or
send it off-site. We use two plants to evaluate the
cost—effectiveness of this option. Plant A is a very large
facility that generates three hundred unusable cans per day.
Plant B is a much smaller facility that generates ten unusable
cans per day.
The cost to purchase a fully contained can evacuator
ranges from $55,000 to $60,000 for a smaller unit and $75,000 to
$80,000 for the larger unit. The smaller unit processes 600 to
1,000 cans per hour. Plant B only generates 250 cans per year.
Clearly, the can crushing unit would not be appropriate for
Plant B. Plant A does not generate enough waste cans to operate
the unit for one hour a day. In a five day workweek, Plant A
generates 1,500 reject cans, enough for two cycles. We assume
that the unit is operated one day a week for, at most, two
hours. In one year the unit will process 75,000 cans. Assuming
a ten year equipment life, a 10 percent cost of capital, and a
unit costing $57,500 installed the annual capital cost of the
unit amounts to $9,315. The per can cost is $0.l24,
significantly less than the cost of off-site can disposal.
This cost does not include the cost of a pressurized
storage vessel for the liquid concentrate and propellant
colleeted from the cans. Maintenance and operating costs are
not included either. Operating costs would include electricity
and a source of air pressure. Maintenance would entail periodic
replacement of seals that can be damaged by repeated contact
with propellants and solvents. The equipment supplier we spoke
with did not have detailed operating and maintenance costs
available, since they were still evaluating the details of their
systems.
-------
— 81 —
The liquid concentrate and propellant drain from the
crusher and are stored in a pressurized vessel. The material is
likely to contain many constituents unless some effort is made
to separate and crush similar products in each cycle. In either
case, it is unlikely that the filler will want to reuse this
material. One reason is that the mixture contains flammable
liquified propellarits. If the material is open to the
atmosphere, some of the propellant will evaporate, resulting in
a potentially hazardous work environment and the release of
VOC’sto the atmosphere. Most likely, Plant A will send the
material to an off-site treatment storage and disposal
facility. The solvent portion can be recovered, the combustible
and flammable portion can be used as supplemental fuel. The
remaining material consisting of miscellaneous product active
ingredients will be incinerated. Many waste treatment companies
charge generators for the waste they take, but give a credit for
the recoverable solvent portion of the waste.
It is difficult to estimate how much it would cost to
send the waste liquid off—site because the waste will vary in
ccmposition depending on what products are processed, and will
contain many different, components. To illustrate how a
charge/credit arrangement might work, we assume a waste stream
of fixed composition. Not all of the waste cans generated will
contain chlorinated solvents. Economically, it would be better
for the filler to segregate the waste that does not contain
chlorinated solvents from the waste that does. If half the cans
are filled with chlorinated solvent based products, that amounts
to 150 cans per day. We assume an average can size of ten
ounces, twenty percent solvent concentration and an average
liquid density of 8 pounds per gallon. Recyclers charge and
credit schemes vary, but for this case we assume a charge of
$1.65 per gallon of nonrecoverable material and a credit of
$1.25 per gallon for recoverable solvent. Considering only that
-------
— 82
portion of waste that contains chlorinated solvents (that is,
150 units per day) it costs Plant A $7,809 per year to dispose
of the waste as summarized in Table 4.2. This includes a credit
for recoverable solvent. As noted above, it does not include
maintenance costs for operating the can evacuator, nor does it
include the capital cost of two storage tanks for collection and
segregation of the recovered liquid or for disposing of the
crushed cans as solid waste. It also does not include the cost
for disposing of the waste cans that do not contain chlorinated
solvents.
Different waste disposal costs have been reported by
others. Waste treatment companies that collect intact waste
cans charge as much as one dollar per can. Another filler has
estimated a cost of $0.83 per can for incineration and solvent
recycling (Flanner, 1988). Our example contains a number of
assumptions that may not be accurate for a given plant, but it
illustrates that it may be possible to recover chlorinated
solvents and reduce disposal costs at the same time.
Non-aerosol Packaging
Many types of aerosol products are available in
non-aerosol form. Some of the alternate products and packages
were discussed in Section III. The aerosol can and other
hardware account for approximately 3 5—50 percent of the cost of
an aerosol product (Geigel and Miller, 1985; ICF, 1987a). When
a product is packaged in non-aerosol form, these costs can be
greatly reduced. There is no need for a valve system in a
liquid, solid or paste product. Non-aerosol spray pumps are
much simpler than the aerosol delivery systems. Container and
valve costs for non—aerosol sprays are significantly reduced
when packaging in non-aerosol form.
-------
— 83 —
Table 4.2
ANNUAL COSTS TO COLLECT LIQUID AND SEND OFF-SITE
Cost
Can Crushing (@ $1.24/can) $4,650
NonrecoVerable to Incineration 3,878
(Q $1.65/gal)
solvent Recovery (credit @ $1.25/gal) (719)
$7,809
AssUn tjofls :
Average can size = 10 oz.
Average liquid density = 8 lb/gal.
150 units per day containing chlorinated solvents.
250 operating days per year.
-------
— 84 —
Switching to a non-aerosol package can change a
product’s solvent requirements. There may be a wider choice of
solvents that can be used than when the same product is packaged
as an aerosol. It is difficult to estimate the impact this
change might have on product cost and the source reduction
potential. Once again, the analysis is very product specific.
The only conclusion that can be drawn is that non-aerosol
products are, in general, less expensive than their aerosol
counterparts. It should be noted that this is not a universally
accepted conclusion, however. Whether the aerosol productsare
more expensive or not, their continued growth reflects the fact
that consumers like to use aerosol products.
Earlier studies have compared the costs of aerosol and
non-aerosol products. A study published in 1984 examined the
economic effects of banning CFC propellants in aerosols. In
most product categories evaluated in that report, aerosol
products cost more than non-aerosol products, on the basis of
cost per ounce of active ingredient (ICF, 1984). The average
aerosol premium ranged from 10 percent for women’s hairsprays to
60 percent for spray starch. An exception to this trend was
aerosol insect repellant, which was more expensive in
non-aerosol form by an average of 67 percent. It should be
noted that this study made no attempt to look at chlorinated
versus nonchiorinated solvent usage. For that reason, and
because the study is more than five years old, it is of limited
use in assessing product costs today from the perspective of
chlorinated solvent use reduction.
A more recent study was conducted by Westat in 1987
(Westat, 1987). The objective of this study was to gather data
on the concentration and incidence of chlorinated solvents in
household products. Aerosol products were analyzed along with
liquids, pastes, pump sprays and powders. The cost of each
-------
— 85 —
product, where available, was reported along with container size
and solvent concentration. This study is also of limited use
for our purposes. There is insufficient information to
determine whether the aerosol and non-aerosol products have
equivalent amounts of active ingredients, or whether they are
designed to perform the same function. Also, since the focus of
the study was chlorinated solvent exposure, there was no attempt
to compare chlorinated solvent based aerosol products with
nonchlorinated solvent, non-aerosol products, which is the ideal
sourc reduction option for our study.
In terms of source reduction potential, a switch to
non-aerosol products could significantly reduce solvent
consumption. The potential is greatest for non—aerosol products
that do not contain chlorinated solvents at all, such as solids,
pastes, solventless liquids or water based products. For
aerosol products that contain chlorinated solvents as active
ingredients, a switch to non-aerosol products may not result in
solvent savings. The non—aerosol versions on the market are
likely to contain chlorinated solvents as well. Because of the
product specific nature of the aerosols industry, it is very
difficult to estimate the source reduction potential that might
be achieved.
CASE STUDY ANALYSIS
In this subsection, two reformulation strategies are
examined that might be implemented to reduce chlorinated solvent
consumption in aerosols. First, general cost considerations
related to product reformulation are discussed, and then a
hypothetical case is examined in which a flammable hydrocarbon
solvent blend replaces chlorinated solvent in a spray paint.
secondly, a DME propellant/water based spray paint is
evaluated. Also evaluated in this section is the option of
correcting over- and under-filled cans.
-------
— 86 —
General Cost Considerations
Product reformulation costs fall into three
areas——research and development (R&D) costs, manufacturing
capital costs, and marketing/distribution costs. R&D efforts
include laboratory work to develop and evaluate alternative
formulations, product efficacy testing, stability testing to
determine if can corrosion or leakage will occur,
hardware/container, stability testing and product toxicity
testing. Prior to full scale manufacturing of a new product
blend, it might be necessary to do some market testing to assess
consumer acceptance of the product. Manufacturing costs may
include retrofitting existing process and storage equipment or
new equipment purchases. Depending on what materials are
handled by a facility, the introduction of new or more flammable
chemicals might require the installation of additional vapor
detection and/or fire control equipment. Pumps, gaskets, seals
and hoses may have to be replaced if they are incompatible with
new materials, such as stronger solvents that may not have been
previously handled. If water is going to be added to product,
it may become necessary to replace existing transfer and storage
equipment with corrosion resistant equipment.
If the reformulated product is more flammable than
before, distribution and warehousing costs increase. Fire codes
stipulate how flammable aerosol products can be stored and
require certain levels of fire suppression equipment for
different classes of flammable and combustible products.
One of the advantages to using chlorinated solvents in
blends is that their higher relative densities add weight to
aerosol packages. When other solvents are used in place of
chlorinated solvents, the final product will likely weigh less.
Consumers choose among products on the basis of product weight
(among other factors). In response to this aerosol
-------
— 87 —
manufacturers often increase the container size, adding more
product than before, so that the relative product weight remains
the same.
When different size cans are used, or if the
reformulated product has different flammability properties, the
artwork and labelling on the container may have to be
redesigned. The average cost to change the label for an aerosol
can has been estimated at $2,250 (ICF, 1988). This includes the
fixed cost to change lithographic or paper printing plates. It
does not include cost of paper or labels or other expenses
related to the actual labeling process, as these costs are not
expected to change. When formulations are changed, the spray
pattern is affected. To obtain the desired spray pattern with a
new formulation it may be necessary to change the valve system.
This can add additional costs. Even more costs will be incurred
if products must be reregistered with federal or state
regulatory agencies.
Costs of Chemical. Substitution: Replacing METH Based Spray
Formulations with Hydrocarbon Solvent Blends and Alternative
Propel lants
In this subsection, we compare the costs of two METH
based spray paints with a flarnable TCA acetone solvent based
formula and with a flammable TCA/acetone solvent formula and
with a water/DME propellant system. Products I and II in Table
4.3 represent typical spray paint formulations based on METH.
In Product III, METH is replaced with a TCA/acetone solvent
blend. Product IV consists of a water based spray paint
formulated with DME propellant. The composition of each product
is noted in terms of weight percent of the various ingredients.
For each case, the cost to formulate the product is determined
by multiplying the ingredient cost in dollars per kilogram by
the concentration of the ingredient in the blend. Ingredient
costs are listed in Table 4.4. In order to calculate a cost per
-------
Table 4.3
I II III IV
(1) (2) (2) (1)
S/kg wt% $Lkg wt% S/kg wt% S/kg wt% S/kg
(3)
Solids (resins, 0.66 12 .079 12 .079 12 .079 12 .079
pigments, other
additives
METH 0.64 28 .179 35 .224
TCA 0.89 40 .356
(4)
Water 0.005 21 .001
Acetone 0.66 32 .211 20 .132
Xylene 0.27 25 .068
To luen& 0.26 25 .065 6 .016
MEK 0.66 5 .033
HC propellant 0.275 28 .077 28 .077 28 .077
DME 0.95 31 .295
Total 100% 5.546/kg 100% 5.445/kg 100% 5.644/kg 100% $.492/kg
Notes:
(1) Formulation taken from ICF (1988).
(2) Formulation based on personal communication with industry representative.
(3) Cost of solids from ICF (1988), not adjusted to 1989 dollars. Since the same weight perc
each formulation, the relative cost of the solids in bach product will remain the same.
(4) SRRP estimate.
-------
— 89 —
can of spray paint, it is necessary to estimate an average can
size. We use an average size of 8 ounces, or 0.23 kilograms.
This number has been used in earlier studies (e.g. ICF, 1988;
ICF, 1987a)
From the calculations, Product Il—-a METH based
formulation is the least expensive to produce. The TCA/acetone
based paint is 40 percent more expensive than Product II.
Product I, which contains both METH and TCA, costs more than the
METH based paint without acetone. The DME/water based paint
costs less than both the TCA/acetone and METH/acetone systems.
The cost figures in Table 4.3 represent only the costs
to formulate a new product. They do not include R&D costs,
manufacturing capital expenditures, or distribution costs
related to the development, production and marketing of a new
product. Earlier studies have estimated these costs for various
types of aerosol products.
Research and development (including stability and
toxicity testing) and marketing costs related to the
reformulation of one spray paint has been estimated at $7,500 in
1987 dollars (ICF, 1987b) per product. That is, each different
color reformulated would represent an additional $7,500 in
development costs. In 1989 dollars, then, each new product
costs-$7,970 in R&D. To assess this cost on a per can basis, we
assume that an aerosol product has a five year life. We use the
production rate of 43,200,000 units per year from Plant A
described earlier in this section, and assume that 25 percent of
Plant A’s product line of paints and related products contain
METH, and will therefore be replaced with other products.
Therefore, the development cost per can of spray paint is
$0. 008.
-------
— 90 —
Table 4.4
SOLVENT AND PROPELLANT COSTS
(1)
Solvents S/lb ______ S/kg
TCA 0.405 0.89
METH 0.29 0.64
PERC 0.31 0.68
CFC—113 1.22 2.69
Acetone 0.30 0.66
MEK 0.30 0.66
Xylene 0.90 0.27
Toluene 0.84 0.26
NMP 3.51
Propel lants ____ ____
DME
HC Propellants
(Average)
Sources:
(1) CMR, 1989d.
(2) Personal Communication with Industry Representative.
(1)
S/gal
1.59
S/lb
(1)
0.43
0. 10—0. 15
0.125
(2)
S / kg
0.95
0.22—0.33
0.275
-------
— 91 —
Earlier studies (Geigel and Miller, 1987; ICF, 1987b)
have reported that no capital costs are incurred from switching
from METH to another hydrocarbon solvent. We talked with one
filler, however, who replaced METH with a TCA/acetone blend in a
product line. They were required by local fire codes and their
insurance carrier to install sprinkler heads, separate feed
lines and filling heads for the acetone because of its
flammability. To allow for flexibility in the future, the
company decided to upgrade three tanks, not just one, and to
install the extra filling head and auxiliary equipment to three
lines instead of only the one currently used for paints.
Table 4.5 lists the equipment modifications that must
be considered for converting to a DME propellant to ensure
compliance with the standard. Because of its flammability,
substantial costs can be incurred in switching to DME
propellant. Many plants, in order to handle DME safely, must
upgrade portions of their facilities to the National Electrical
Code (NEC) Class I, Group C equipment standards. Costs
associated with this equipment conversion are estimated to be
$15,000 per line in 1988 dollars, or very close to $15,950 in
1989. Therefore, it would cost about $63,780 for Plant A to
convert all four lines. Annualized over a fifteen year
equipment life, with a 10 percent cost of capital, the annual
capital cost for DME conversion is $8,473. This cost should be
averaged over the annual production volume and added to the
manufacturing costs of the DME/water based spray paint. If all
of the METH based paint is replaced with the DME/water blend,
the added cost would be an additional $O.008 per can. For the
DME/water paint, total development costs amount to $O.016 per
can.
-------
— 92 —
Table 4.5
EQUIPMENT MODIFICATIONS FOR DME STORAGE AND HANDLING
A. Electrical equipment that must conform to requirements of
NFPA Class I, Group C:
Meters, relays
Instruments
Wiring methods
Conduit and cable system sealing
Liquid or condensed vapor trap drainage
Arcing devices such as switches, circuit breakers,
motor controllers and fuses
Motors
Lighting fixtures
Flexible cards
Spare parts for all of the above
B. Materials of construction for the following must be
compatible with DME:
Tank flanges
Valve seats, packing and seals
Pump seals
Hoses
Pressure relief devices
Filling machine gaskets and seals
C. Other considerations
Recalibrate gas detection equipment for DME
Check tankage, piping and other equipment for proper
marking and labelling
Ensure proper grounding of all equipment
Check equipment against current codes (ASME, etc.)
Source: ICF, 1987a.
-------
— 93 —
There are still other costs that have not been
incorporated into the above. When products are reformulated
and, as a result, are more flammable, warehousing and
distribution costs may increase. It costs $2,000 to add
sprinklers to each storage tank and $3,000 more to add a
separate line and filler head for each filling line. If we
assume that Plant A must make the same changes, and chooses to
do so for each of four filling lines and three bulk storage
tanks, then the total cost would be $18,000. If this cost is
spread over a ten year equipment life, the cost per can of
product would be negligible. We do not add this cost to the
total product reformulation costs, but note that these kinds of
expenses must be considered when evaluating formulation changes.
Correction of Over and Under FillincT
This option was discussed in Section III as a means of
reducing the number of filled containers that must be disposed
of due to defects. The approach is based on an article by
Flanrier (1988). In order to evaluate this options, we introduce
two hypothetical plants to use as models in the cost analysis.
Characteristics of these hypothetical plants are summarized in
Table 4.6. Plant A is a large facility that fills 72 million
units a year of paints, automotive and industrial products and
insecticides. It has four high speed filling lines that each
process 150 cans per minute. Plant B is a smaller facility that
fills 25 cans per minute of a special use industrial product.
Plant B produces 3 million units a year.
Based on discussions with industry sources and a review
of survey responses, we can assume that one can in every
thousand cans processed through a high speed line is unusable.
(The rate for slower lines may differ, but for this analysis we
use the same number). At this rate, Plant A generates 288
unusable cans a day, and Plant B generates 12 a day.
-------
— 94 —
Table 4.6
CORRECTION OF PROPELLANT FILLING -
COST SAVINGS ACHIEVABLE
A) Annual Production (units/yr)
B) Number of Filling Lines
C) Units/Mm/Line
D) Units/Day
E) Unusable Units Generated Per Day
F) Total Disposal Cost ($/yr)
G) Cans Corrected Per Day
H) Disposal Costs Avoided Per Year
I) Added Labor Cost Per Year
3) Direct Cost Savings, (H) — (I)
K) Savings in Product Value
L) Total Cost Savings, (J) + (K)
Plant A
72,000,000
4
150
288, 000
288
$64,800
173
$38,925
$14, 418
$24,507
$21,625
$46, 132
Plant B
3,000,000
1
25
12, 000
12
$2,700
7
$1,575
583
992
$ 875
1,867
-------
— 95 —
Quotes of costs to send waste cans off-site for
incineration, solvent recovery or other treatment range from
$.83 to $1.00 per can. If we take $.90 per can as a mid-range
cost and calculate the daily cost for disposing of the total
volume of waste cans generated at each site, the result is
$250.20 for Plant A and $10.80 for Plant B. In yearly costs,
Plant A would spend $64,800 and Plant B would spend $2,700 to
dispose of all rejected cans.
According to one source, more than 50 percent of the
unusable cans are rejected because they had too much or too
little propellant (Flanner, 1988). If we take 60 percent as our
estimate, this means that Plant A has 173 cans a day and Plant B
has about 7 cans a day that can be corrected. We assume that
damaged cans can be identified by a quick visual inspection. If
it then takes an employee two minutes to weight check the other
rejected cans, add propellant to underfilled cans and release
propellant from overfilled cans, Plant A requires 346 minutes of
labor, and Plant B requires 14 minutes of labor each day. At a
labor rate of $10 an hour, the rework amounts to $57.67 a day
for Plant A and $2.33 a day for Plant B. During an operating
year, the labor costs to correct over and underfilling total
$14,418 for Plant and $583 for Plant B.
The direct cost savings to correct over and
underfilling consists of the disposal costs avoided minus the
labor costs incurred, Subtracting labor costs from these totals
yields a direct savings of $24,507 annually for Plant A and $992
for Plant B. It is assumed that both plants operate one eight
hour shift a day, 250 days a year.
In addition to the direct disposal costs avoided, the
analysis should account for the value of each can of product
that can now be sold instead of shipped out as waste. This
value will, of course, vary greatly depending on the type of
-------
— 96 —
product, the concentrate, container and associated hardware. To
keep the analysis simple, an average value of $.50 per can is
used, in which case Plant A saves an additional $21,625 and
Plant B saves $875 more. Total cost savings (avoidance of
disposal costs and recovery of lost product minus labor costs)
are $46,132 for Plant A and $1,867 for Plant B.
The source reduction potential of this option is more
difficult to determine. In Plant A, chlorinated solvents are
used in a wide variety of products. Plant A also fills a number
of products that do not contain chlorinated solvents. The
reduction in off—site releases from Plant A can be estimated by
assuming a product mix and the concentration of chlorinated
solvents in each product. The details of this calculation are
provided in Appendix B. If Plant A adopts the practice of
manually correcting over and under filled cans, 2,309 kg of
chlorinated solvents can be prevented from going to an off-site
facility for incineration, solvent recovery or use as
supplemental fuel.
-------
— 97 —
REFERENCES
Aerosol Age, “South Coast Air Quality Director Puts Issues
into Perspective”, Vol. 33, No. 4, p.22, April 1988
(Aerosol Age, 1988a).
Aerosol Age, Vol. 33, No. 8, p.6, August 1988 (Aerosol Age,
1988b)
Aerosol Age, Vol. 34, No. 9, p.6, September 1989, (Aer2sol
Age, l989b).
Aerosol Age, “1989-1990 Buyer’s Guide”, Vol. 34, No. 3.0,
p.44, October 1988 (Aerosol Age, 1989)
Bartlett, Philip L., “New Propellant Performance Reduces
the Risk Factors in Aerosol Paints”, Modern Paint and
Coatings, Vol. 78, No. 7, p. , April 1988, (Bartlett,
1988)
Bartlett, Philip L., “Reducing Dropout of Aqueous DME Total
Release Insect Sprays”, Aerosol Age, Vol. 34, No. 10, p.20,
October, 1989 (Bartlett, 1989).
Beacon Engineering Co., “EVAC I and EVAC II Product
Brochures”, undated (Beacon Engineering, undated).
Bohnenn, Leo J.M., “Recent European Developments in DME
Technology”, Aerosol Age, Vol. 33, No 11, p.40, November
1988 (Bohnenn, 1988).
City of Los Angeles HTM Project, Board of Public Works.
Fact Sheet. “Aerosol Containers: What Should I Do With Them
When They’re Empty” 1990.
-------
— 98 —
Chemical Marketing Reporter, September 27, 1982 (CMR,
1982).
Chemical Marketing Reporter, March 29, 1983 (CMR, 1983).
Chemical Marketing Reporter, January 20, 1986 (CMR, 1986).
Chemical Marketing Reporter, September 23, 1987, p. 9 (CMR,
1987).
chemical Marketing Reporter, Vol. 235, No. 5, p.46, JaQuary
30, 1989 (CMR, 1989a)
Chemical Marketing Reporter, Vol. 235, No. 8, p.54,
February 20, 1989 (CMR, 1989b).
Chemical Marketing Reporter, “Aerosols Fueled by Personal
Care”, Vol. 235, No. 17, p. SR29, April 24, 1989 (CMR,
1989c)
Chemical Marketing Reporter, “Chemical Prices”, Vol. 236,
No. 20, p.42, November 13, 1989 (CMR, 1989d).
Chemical Specialties Manufacturers Association, Inc.,
“Pressurized Products Survey 1984”, 1985 (CSMA, 1985).
Chemical Specialties Manufacturers Association, Inc.
“Aerosol Pressurized Products Survey United States 1988”,
1989, (CSMA, 1989a).
Chemical Specialties Manufacturers Association, Inc. “Re:
Protection of Stratospheric Ozone; Advance Notice of
Proposed Rulemaking, 54FR15228’ (April 17, 1989)”, letter to
U.S. EPA, June 30, 1989 (CSMA, 1989b).
Chemical Week, September 23, 1987, p. 9 (Chemical Week,
1987).
-------
— 99 —
Cuzic, John E., “Choosing Chlorinated Solvents for Aerosol
Products”, Aerosol Age, Vol. 34, No. 7, p.26, July 1989,
(Cuzic, 1989).
Dow Chemical Co., Personal Communication of M. Reardon with
Pamela Yates (SRRP), , (Dow, 1989a).
Dunn, D.P., “Dyine]. Propellants — One Company’s Innovative
Response to Regulations”, reprinted from Aerosol Report,
Vol. 27, No. 1—88, January 1988 (Dunn, 1988).
Flanner, Lloyd, “Hazardous Waste Minimization”, Aerosol
Age, Vol. 33, No. 2, p.22, February, 1988 (Flanner, 1988).
Ceigel, M.A. and A. Miller, “Integrated Solvents Analysis:
Aerosols”, Draft Report, SRI International, July 1985,
(Geigel and Miller, 1985).
Gibson, S., “McClellan Seeks New Ways to do Dirty Jobs”,
Sacramento Bee, June 11, 1989 (Gibson, 1989).
Halogenated Solvents Industry Alliance, Inc., “White Paper
on Methylene Chloride”, April 1985 (HSIA, 1985).
Halogenated Solvents Industry Alliance, Inc., referenced in
unidentified document, (HSIA, 1986).
Halogenated Solvents Industry Alliance, Inc., “White Paper
on 1,1,1—Trichloroethane”, June 1987 (HSIA, 1987).
Halogenated Solvents Industry Alliance, Inc., “White Paper
on Methylene Chloride”, February 1989 (HSIA, 1989).
-------
— 100 —
ICF, Inc., “Aerosol Working Paper Series, Paper 1”,
October, 1984 (ICF, 1984)
ICF, Inc., “Use and Substitutes Analysis of Chlorinated
Solvents in Aerosols”, Draft Report, November 20, 1987,
(ICF, 1987a)
ICF, Inc., “Market Analysis of Chlorinated Solvents in
eroso1s”, Draft Report, September 22, 1987 (ICF, 1987k).
ICF, Inc., “Options for Regulating Methylene Chloride and
Perchioroethylene in Aerosol Products: A Cost Benefit
Analysis”, Draft Report, August 30, 1988 (ICF, 1988).
ICF, Inc., “Use and Substitutes Analysis of Methyl
Chloroform (1,1,1-trichioroethane)”, Preliminary Draft
Report, February 14, 1989 (ICF, 1989).
Industrial Economics, Inc., “A Structural Approach for
Evaluating Pollution Prevention Opportunities”, October,
1988 (IEI, 1988).
Johnsen, Montfort A., “Disposal of Flammable Aerosols”,
Aerosol Age, Vol. 34, No. 1, p.20, January 1989 (Johnsen,
1989).
Johnsen, Montfort A., “Flammability Aspects of
Dimethylether, P-22, P-142b, P-152a”, Aerosol Age, Vol. 33,
No. 8, p.32, August, 1988 (Johnsen, 1988).
Kawain, A. and J.B. Flynn, “Aerosols”, in Kirk-Othiner
Encyclopedia of Chemical Technology, Vol. 1, p.582, 1979.
-------
— 101 —
Lim, Walter, “Filling and Formulating for Barrier Packs”,
Aerosol Age, Vol. 34, No. 7, p.22, July, 1989 (Liin, 1989).
Martinez, J., R. Wassel and G. Bockol, “Emission Estimates
and Controls Memorandum for Emissions from those Aerosol
Packaging Facilities Responding to Section 114
Questionnaires”, Memorandum, October 13, 1987, Radian
Corporation (Martinez et al, 1987).
Pfetzing, E., “Estimation of Worker Exposure for Aerosol
Packing”, Memorandum, PEI Associates, Inc., December 8,
1987 (Pfetzing, 1987).
Rouse, W.M. and F. Nowak, “The Development of
Hydrocarbon-Tolerant Hairspray Polymers”, Aerosol Age, Vol.
33, No. 12, p.48, December 1988 (Rouse and Nowak, 1988).
Smart, M.M and S.P. Landels, “Integrated Solvents
Analysis: Pesticide Uses”, Draft Report, SRI
International, July 1985 (Smart and Landels, 1985).
Shields, S., Beacon Engineering Co., Telephone Conversation
with P. Yates of SRRP, November, 2.989 (Shields, 1989).
Strobach, D.R., “Alternatives to CFCs”, Aerosol Age, Vol.
33, No. 7, p.33, July 1988 (Strobach, 1988).
Westat, Inc. and Midwest Research Institute, “Household
Solvent Products: A “Shelf” Survey with Laboratory
Analysis”, July 1987 (Westat, 1987).
------- |