600282029
EVALUATION OF FOAMS FOR MITIGATING
AIR POLLUTION FROM HAZARDOUS SPILLS
by
S.S. Gross
R. H. Hiltz
MSA Research Corporation
Division of Mine Safety Appliances Company
Evans City, Pennsylvania 16033
Contract No. 68-03-2478
Project Officer
John E. Brugger
Oil and Hazardous Materials Spills Branch
Municipal Environmental Research Laboratory
Edison, New Jersey 08837
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research Labora-
tory, U.S. Environmental Protection Agency and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or recommendation
for use.
11
-------
FOREWORD
The U.S. Environmental Protection Agency was created because of increas-
ing public and government concern about the dangers of pollution to the health
and welfare of the American people. Noxious air, foul water, and spoiled land
are tragic testimonies to the deterioration of our natural environment. The
complexity of that environment and the interplay of its components require a
concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem solu-
tion; it involves defining the problem, measuring its impact, and searching
for solutions. The Municipal Environmental Research and Laboratory develops
new and improved technology and systems to prevent, treat, and manage waste-
water and solid and hazardous waste pollutant discharges from municipal and
community sources, to preserve and treat public drinking water supplies, and
to minimize the adverse economic, social, health, and aesthetic effects of
pollution. This publication is one of the products of that research and pro-
vides a most vital communications link between the researcher and the user
community.
Aqueous foam blankets have become one of the prime mechanisms of vapor
control with spilled hazardous chemicals. Their use has grown empirically
however, and there are no guidelines for their selection or application. This
report presents an evaluation of fire fighting foams as a mechanism of vapor
control. It contains a use matrix, guidelines on control times and identifies
the important properties of foam with respect to spill control.
Francis T. Mayo, Director
Municipal Environmental
Research Laboratory
111
-------
ABSTRACT
This program has been conducted to evaluate commercially available water
base foams for mitigating the vapors from hazardous chemical spills. Foam
systems were evaluated in the laboratory to define those foam properties which
are important in mitigating hazardous vapors. Larger scale tests were then
conducted in a 3 m by 3 m pan. Polar and nonpolar liquids and liquefied gases
were used as test materials.
Protein, fluoroprotein, alcohol and aqueous film forming foams were
tested at low expansion ratios and surfactant foam agents at low, medium and
high expansion ratios.
The chemicals tested were acetone, n-butyl acetate, di-ethyl ether, n-oc-
tane, triethylamine, benzene, toluene, ethyl benzene, cyclohexane, propane,
ethylene, butadiene, ammonia, chlorine, ethylene oxide, hydrogen fluoride and
sulfur trioxide. It can be concluded that all high quality foams are
effective against nonpolar liquids but only alcohol foams can be recommended
for polar liquids. Generalizations with liquefied gases are difficult because
of their unique characteristics.
This report was submitted in fulfillment of Contract 68-03-2478 by MSA
Research Corporation under the sponsorship of the U.S. Environmental
Protection Agency. This report covers the period from October 29, 1976 to
July 31, 1980.
-------
CONTENTS
Foreword iii
Abstract iv
Figures vi
Tables vii
1. INTRODUCTION 1
2. CONCLUSIONS 3
3. RECOMMENDATIONS 5
4. TECHNICAL DISCUSSION 6
Protein Foams 6
Surfactant Foams 7
Fluorocarbon Surfactants 7
Modified Foams 8
Chemical Foam 8
Foam Drainage and Collapse 9
Foam Equipment 9
Previous History of Foam Application to Chemical Spills 11
Volatile Organic Chemical 12
Liquified Gases 13
Inorganic Water Reactive Chemicals 15
5. SCOPE OF WORK 16
Task I - Delineation of Available System 16
State-of-the-Art Survey 16
Literature Search 17
Matrix Preparation 18
Laboratory Program 18
Matrix Revision 52
Task II - Field Testing 52
Evans City Field Tests 54
Oklahoma Field Tests 56
Nevada Field Tests 61
Foam Matrix 67
Task III - Novel Foam System 67
References 7-^
Bibliography 72
-------
FIGURES
Number Page
1 Benzene Vapor Concentration vs Detector Height 31
2 Benzene Vapor Concentration Percent vs Time for MSA,
Lorcon, Emulsiflame 32
3 Benzene Vapor Concentration vs Time 34
4 High Expansion Foams with Nonpolar Hydrocarbons 35
5 Medium Expansion Foams with Nonpolar Hydrocarbons 36
6 Medium Expansion Foams with Nonpolar Hydrocarbons 37
7 Acetone with Low Expansion Foam 38
8 Benzene with Low Expansion Foam 39
9 Cyclohexane with Low Expansion Foams 40
10 Ethyl Ether with Low Expansion Foams 41
11 Toluene with Low Expansion Foams 43
12 Ethyl Benzene with Low Expansion Foams 44
13 n-Butyl Acetate with Low Expansion Foams 45
14 Triethylamine with Low Expansion Foams 46
15 n-Octane with Low Expansion Foams 47
16 Vapor Build up Over Benzene Surface Comparing Two Types of
Protein Foam 48
17 Effect of Doubling Foam A Thickness for Benzene Spill of
Figure 16 49
VI
-------
18 Vapor Build up of Vapor Concentration Over Benzene Surfaces
Comparing Two High Expansion Foams 50
19 Effect of Foam Additions to Anhydrous Ammonia Spills
on Downwind Vapor Concentrations 60
20 High Expansion Foam on Liquefied Chlorine 62
21 Fluoroprotein Foam on Liquefied Chlorine 63
22 AFFF on Liquefied Chlorine 65
23 High Expansion Foam on Liquefied Chlorine 66
24 Ethylene Oxide Vapor Concentration During Alcohol Foam Test. .68
Vll
-------
TABLES
Number Page
1 Matrix of Foam Capabilities to Suppress or Otherwise
Minimize the Release of Toxic or Flammable Vapors from
Spilled Hazardous Chemicals as Listed 19
2 Properties of Liquefied Gases or Water Reactive Chemicals. . .22
3 Candidate Polar Materials 23
4 Candidate Nonpolar Materials 23
5 Properties of Candidate Foam Agents 25
6 Expansion and Drainage of AFFF Foam Systems 26
7 Properties of High Expansion Foam Systems 27
8 Representative Low Expansion Foams 28
9 Preliminary Matrix of Foam Capabilities on the Spilled
Hazardous Chemicals Listed 53
10 Foam Mitigation Times - Time for Significant Vapor
Breakthrough 55
11 Testing Sequence at Norman, Oklahoma 57
12 Matrix for Polar and Nonpolar Liquid Hydrocarbons 69
13 Test Results for Novel Foam Agents 70
Vlll
-------
M
if •
s ll-i
1 * lit!-? If
• • s .t £ i t S S .
Z*
31
•
J | 5 * 2 »t.
« r * eo «•• - o «C o-c
| » 3 *
e ft '
« « f .i
• : r £ . 1 1 .
s » || s 8 f-
i : 1 1 t s | £ £ i
= : iUil Ili
» «
•5
1 i E E . i Yfc"
' 8* .'..M
: -
ir ili nidi
••
s
«• M
"5 X ***
^ S g^r I S-. 8*^., a
»M ^ o~- e OM^OJR^ ^
."i * *
•§ 5 "
0 O • •
!| il 1 it
!i ili iliiii
12 .5. i V*
li
h
M
e
ii
u
t
K It K 01 (I fl U »I SI H tl (I III 01 C It
Illl Illl Illl Illl Illllllll III! Illl Illl Illl Illl Illl Illllllll Illl Illl III! III! III! III! Illllllll Illl Illll
Illllll IIll Jllll IIll Illl till Jl
f f » t t I
u
u
rri
TIT
"'|TIT|T
T|T
•s
t
•s
I
eel
i = £!
fall
Is
OT
Z.
(W
Z
s
i
s
u
•
1
•
to.
T
•te
£
1 5
i K
ku
#
K
list
lib
- --!
r-8e- «
*
In!
SEE
e e r
Hi
-s.
• o o
fl
»p MA «utnbt
!••! »«nl>t
•upiH DMnbt
M •
« £
!J
• •. '•
•WO S
M
*
•
i
!i
• " s
hi
• W
"• * Z
aoe §
O
.1
III1
HI ii
111 III III
EE£^5J^53
ss«.ss
• ; »
• R I - v
ili .|Ji
* — ^S255s :
* * 9 3 E 3^33
W
H
Z
a
t—
K
MJ
a.
Z
||
U
9
|f
• i -
• Is
j|
8 £
? I = o t S i"r"^
1 T
. i
ix
-------
SECTION 1
INTRODUCTION
In the last few years significant advances have been made in the capabil-
ity to control and contain spills of hazardous materials. These advances
serve to minimize damage to the land and water ecosystems. These spills,
however, can also provide significant air pollution and danger to life and
property in the spill area. The problem is particularly acute for the person-
nel assigned to clean up and dispose of the spilled material.
At the time this study was initiated, water based foams had been known to
be effective in controlling vapor release from some volatile chemicals. A
significant body of data had been developed for hydrocarbon fuels as a result
of the extensive use of foams in the control of fires of such materials.
Extended tests had also been conducted with foam in the control of fires and
spills of liquified natural gas. In addition to these fairly well detailed
tests, restricted testing had been conducted with a number of other materials.
Some results had been obtained with vinyl chloride monomer, ammonia, chlorine
and sulfur trioxide.
Foam blankets act to isolate spills from ignition sources and radiant
energy. Vapor release is also reduced due to the limited permeability of the
foam, its capacity to absorb the chemical, or dilute the surface layer. For
materials which boil below 0°C and whose gas density is less than air, foam
assists in vapor dispersion by raising the temperature of the vapors as they
permeate the foam layer. Water reactive chemical spills can also be mitigated
by foam. They allow the gentle application of water to the surface to effect
dilution and/or conversion to less hazardous chemical species without violent
reaction.
Water based foams are not necessarily applicable to all volatile chemi-
cals. In point of fact, several of the materials already mentioned require
foam formulations other than those commercially available for satisfactory
results. Although the reasons are not very clear, available data shows that
commercially available foams are degraded by materials whose dielectric con-
stant is greater than three.
Prior to this program the available data had indicated some potential for
foam against certain classes of materials. A Coast Guard program, (Greer,
1976) which evaluated control procedures for spilled hazardous materials which
float on water, had reviewed the available data. It found sound basis for
-------
the use of foam to control the vapor hazard from spilled volatile chemicals.
It found further, however, that the classes of materials to which water based
foams might be applicable were not well defined.
The first task of the program was to review the state-of-the-art and eval-
uate the utility of foams as a means of mitigating air pollution from hazard-
ous spills on the basis of the existing technology.
The second part was to assess the applicability of foams and foam gener-
ating equipment against individual chemicals and define the characteristics of
foams which influence their compatibility and control over hazardous spills.
-------
SECTION 2
CONCLUSIONS
The following conclusions can be drawn from the program:
1. High quality foams which have low drainage, i.e., retain their water
content, performed the best in mitigating the vapors during spill
control tests.
2. For nonpolar liquids, any high quality foam cover, regardless of its
chemical type, will provide some degree of mitigation of vapor re-
lease.
3. Only alcohol foam was stable against highly polar low molecular
weight liquids such as acetone. Other types of foam rapidly col-
lapsed.
4. The vapors from polar spills may be mitigated with nonalcohol type
low expansion foams if the volume of foam is several times greater
than the volume of the spill. This reduction is mainly due to water
dilution rather than a foam cover.
5. For liquids which are extremely water reactive, high expansion foam
provides the best control. This is due to the heats of reaction or
solution. The higher the expansion, the slower is the rate of water
addition to the reactive liquid.
6. Foams do not effectively mitigate the vapors for liquefied gases
which are heavier than air and not water reactive. The water drain-
ing into the liquefied gas exaggerates the boil off rate. The re-
leased vapors, being nonbuoyant, are not readily dispersed. There
are reports (1) of foam being effective against butane, but the data
from the tests of this program are not in agreement.
7. For liquified gases such as ethylene which are buoyant at ambient
temperatures, a high expansion foam blanket acts as a heat source.
The vapors rising through the foam increase in temperature and dis-
persion of the released vapor is enhanced.
8. For some reactive liquefied gases, such as ammonia, foam provides
some mitigation of the vapor hazard due to heating of the vapor to
impart buoyancy. High expansion gave the best results, attributable
to a slow rate of water addition to the spill. The high expansion
foam warms and scrubs the buoyant vapors as they pass through the
-------
foam layer. Low expansion foams add water at a faster rate to the
liquefied gas and large vapor clouds are produced. Foam collapse is
exaggerated by reaction with the spilled material and intermittant
foam makeup is required to maintain control.
9. For water reactive liquefied gases heavier than air, such as
chlorine, no commercially available foam has been found to be truly
effective for vapor mitigation. Although some reduction in vapor
release can be achieved, reductions below 1% are difficult to
realize/
10. Laboratory tests can be conducted to predict the behavior of foams
under field conditions.
11. Environmental conditions such as wind and rain are more detrimental
to high expansion foams than to low expansion foams.
12. High expansion foams appear to mitigate a spill by containing the
vapors within the foam mass so a vapor-liquid equilibrium can be
maintained, thus reducing the driving force towards vaporization.
13. Low expansion foams appear to mitigate the vapor by forming a barrier
to prevent vaporization.
14. Foams are an inexpensive, effective method of initial vapor control
during an accidental spill, provided the correct foam made from a
quality agent is used.
15. The foam agent which produced the best results in terms of vapor
mitigation was the metal stearate-based alcohol type foam. This
foam agent was better than the newer forms of polar solvent foams.
The alcohol type foams were more difficult to apply, more was needed
to establish a desired thickness, and the foam cover is fairly stiff.
However, once the foam cover was established, it provided good miti-
gation of vapors.
-------
SECTION 3
RECOMMENDATIONS
The following recommendations are made for future consideration:
1. Water drainage and foam collapse rates, which are extremely important
in predicting foam behavior against a spill, should be standardized
both for high and low expansion foams. Drainage and collapse rate
values vary greatly between manufacturers. These values can change
by several hundred percent depending upon the ambient temperature and
the foam generation equipment. They can also vary as a function of
the drainage test employed. For low expansion foam two different
techniques are listed by NFPA. For high expansion a standard test
has just been incorporated into NFPA 11A.
2. Presently only alcohol type foams are recommended for polar spills.
Other types of foams collapse, providing little or no protection.
Alcohol foams which use a precipitation mechanism are thick and more
difficult to apply. The newest polar solvent foams are more fluid
but the agents are thixotropic by nature, which can be an impediment
in proportioning. Alternate foam agents which are stable against
polar compunds and have better flow properties would be desireable.
3. Acid or alkaline resistant high expansion foams should be developed
for use on reactive liquids and liquefied gases.
5. Foams which have significantly lower drainage rates should be de-
veloped in order to be more effective against spills of liquefied
gases.
6. The foam depth versus mitigation time should be examined to determine
if there is a minimum or optimum thickness which must be applied to
the spill in order to effectively mitigate the vapors.
7. An evaluation should be made to determine minimum application rates
for foam against reactive chemicals. If the application rate is too
low the spill hazard can be exaggerated to exceed the free spill
hazard.
-------
SECTION 4
TECHNICAL DISCUSSION
The application of water-based foams to the control of spills has grown
out of the use of these foams for fire fighting. For the most part, foam
agents and generating equipment developed for fire fighting have been used
whenever foams are applied for spill control. The fact that some control
could be exercised over the release of vapors from volatile chemicals tended
to be incidental. Even the extensive work which has been accomplished in the
last few years with LNG has used commercial foam agents and equipment. Only
recently has any consideration been given to modified or novel foam systems.
Efforts to develop new foams have been made primarily because the avail-
able materials were completely unsuitable. Programs tended to become pre-
occupied with the development of new forms for those specific compounds which
rapidly degraded standard foams. None of the programs made a systematic de-
termination of the important characteristics of foam with respect to spill
control and few determined the specific manner in which they controlled the
vapor release.
Although much still remains to be done to define the full capability of
foams along with their limitations, the prior work has established foam as an
effective control measure. The work with LNG has begun to point out important
properties of foam for cryogens and a Coast Guard program (Greer, 1976) pro-
vided similar data for materials which float. With knowledge of how foams act
with a broad range of chemicals, a base has begun to be established for modi-
f ic at ions and improvement s.
FOAM CHEMICALS
Two basic foam types are currently available — protein-derived materials
and surfactant or detergent based concentrates. A number of novel foams are
available but all are modifications of a hydrocarbon surfactant system. An-
other foam material termed chemical foam exists, but it has ceased to be
available, displaced by protein foam.
Protein Foams
Foam concentrates derived from hydrolyzed protein are widely used in the
control of hydrocarbon fuel fires. Protein foams were probably the first used
in spill control. In their use with fuel fires it was realized that they
effectively blanketed the spill preventing the release of vapors. This
characteristic has been written into specifications by both the federal
government and independent laboratories. These specifications require foams
-------
to provide a barrier to ignition for a minimum period of fifteen minutes.
Fluoroprotein foam agents are a protein base material to which fluorinated
surfactants have been added to improve properties such as fluiditity and sur-
face tension, while reducing the tendency of the protein base to pick up
(absorb) hydocarbon liquids.
In addition to hydrocarbon fuels, protein based systems have been develop-
ed which are applicable to fire control of polar compounds. Marketed as
"alcohol foams" they are of two types. The early systems used a metal stea-
rate to form a protective precipitate. Newer, universal or polar solvent
foams use a polysaccaride to gel the foam on contact with the polar solvent.
Although protein foam has a demonstrated capability in spill control for
hydrocarbon fuels, the extent of its capability, permeation rate, fuel absorp-
tion, etc., has not been adequately defined for spill control other than with
hydrocarbon fuels. Compatibility of foam with a chemical is not necessarily a
measure of its ability to restrict vapor release.
Protein foams are limited to low expansion systems and the concentrates
have a limited shelf life, varying with the manufacturer. They tend to be
messy, difficult to clean up, and odiferous but they are usually standard
items with the fire services. For that reason alone, their capabilities and
limitations for spill control of volatile hazardous chemicals need to be de-
lineated.
Surfactant Foams
Detergent foam concentrates are probably the widest used. They avoid the
problems of other agents. They are suitable for both low and high expansion
and most have an unlimited shelf life if stored correctly. They are not as
well regulated as protein materials and as a consequence there is a wide
variety of surfactant foam concentrates on the market. They represent a broad
range of properties and with the lack of standards the selection of high qual-
ity foam agents can be difficult.
Current commercially available surfactant foams for fire fighting are
predominantly anionic. These materials are readily available, relatively
inexpensive, nontoxic and biodegradeable. The amphoterics tend to be more
expensive and the foaming qualities of the cationics legislate against their
use. There are nonionic agents available. Nonionics have a greater tolerance
for pH variations, but they are not adaptable to modification which minimizes
water drainage from the foam.
Fluorocarbon Surfactants
These materials, termed aqueous film forming foams, are hydrocarbon sur-
factants modified with fluorocarbon surfactants to achieve a low surface ten-
sion. These materials do form aqueous foams, both at high and low expansion
ratios, but the foams have very high collapse rates. In many applications,
the foam is a means of applying water rather than an end in itself.
-------
The fluorocarbon surfactants lower the surface tension of water signifi-
cantly. This makes it possible to float water films on the surface of the
hydrocarbon fuels. In some cases the balance is precarious and the water
must be added gently to achieve floatation. Foam is used to achieve this
gentle application. The basic method of fire control is thus the surface
film.
The resistance to permeation of a thim film is small. Several reports
have been published on vapor permeation through the surface films
(Hederstand, 1924; Meldrum, 1972; White, 1976).
Modified Foams
This terminology covers recent developments but relates only to those
materials which are considered commercially available.
Almost all advances have been with surfactant base materials. The basic
improvement in surfactant type fire fighting foams has been with drainage
and collapse rates. This has been accomplished primarily by the addition of
water soluble polymeric materials to the concentrate. There are numerous
literature and patent references to polymer modified foam concentrate.
There are numerous literature and patent references to polymer modified foam
concentrates but few materials have reached commercial availability. MSAR
has used this approach in the evolution of its Type V foam which has been
shown to have superior properties in fire and spill control, being the only
hydrocarbon surfactant foam to pass the fire tests for both low expansion
and high expansion foams specified by Underwriters Laboratory. Other
complex foam agents are commercially available, which apparently use polymer
additions. National Foam has a foam agent termed "Universal" which is
compatible with a broad range of chemicals and more recently has started
marketing a new polar solvent agent. MSAR also has a foam agent called Type
L which yields a long lasting, slow collapsing foam. It, too, is compatible
with a large number of material classes and has application to spill
control. The 3M Company markets a Light Water Alcohol Type concentrate.
This foam agent has also been advocated for spill control.
Chemical Foam
Although this kind of material is no longer available, the fact that at
one time it was a major commercial product and could easily be produced jus-
tifies discussion in this section. Chemical foam was made by blending
sodium bicarbonate and aluminum sulfate in water. Other chemicals are added
to assist foam formation, with the foaming action provided by the release of
C02 by chemical reaction. Sulfuric acid or similar chemicals were substi-
tuted for aluminum sulfate in special systems such as hand extinguishers.
The advent of protein foam, pressurized extinguishers and other improve-
ments led to the demise of chemical foam as a viable system. Since chemical
foams are no longer commercially available, this type of foam agent was not
examined in the program.
-------
Foam Drainage and Collapse
Although the requirements for spill control may differ from fire, water
drainage is still expected to be the most significant property of a foam. It
controls fluidity, resistance to thermal effects, ability to assimilate
foreign materials and linear collapse rate. The slower the drainage the bet-
ter is each of these characteristics. Foam agents range from simple water
solutions to complex blends of components to achieve the desired character-
istics. With anionic base materials, drainage can be controlled through
chemical modification. As a result, commercial materials cover a range of
drainages. The worst may lose 70% of their water within two minutes after
generation. The best have quarter drainage times (time to drain 25% of the
initial water) of 12 to 15 minutes.
Anionics exhibit a temperature transition property where the drainage
changes from slow to fast over a narrow temperature range. For most this
temperature is below ambient and thus they are fast draining. A few agents
have transition temperatures above ambient and thus have inherently slow
drainage. The best foam systems have transition temperatures in the range 29
to 35°C.
Linear collapse rates of foams are often quoted but they can be mislead-
ing. As a foam drains it leaves a dry skeleton which has little integrity.
Thus the recession of effective foam thickness can be greater than the
measurable collapse rate. Collapse data are significant only for slow
draining foams where the dried layer may be only two to three bubble layers
thicker than the effective foam thickness.
FOAM EQUIPMENT
Equipment to generate foam has been tailored to the needs of the fire
service even more than the foam chemicals. Two basic categories of
generation equipment exist commercially; that which yields low expansion foam
generally in the range 3:1 to 12:1, and that which yields high expansion foam
120:1 to 1500:1. In this latter category, foams of less than 400:1 are now
being referred to as medium expansion foam. Expansions above 1000:1 are now
considered ineffective and the limited work with spills would indicate this
should be reduced to 750:1.
The absence of equipment to fill the gap between 12:1 and 120:1 reflects
the almost exclusive application to fires. High expansion foams have been
primarily directed to three dimensional fires where total flooding of the
fire area can be accomplished. High expansion foams cannot be projected very
far and are affected by winds above 10 mph. Although they make less
effective use of water than high expansion, low expansion foams can be
projected considerable distances, are effective in open areas, and where
structual cooling or coverage of vertical surfaces is required. There are no
data available which show advantage in fire control for an intermediate range
foam.
-------
Other applications have been investigated where both foam stability and
water conservation were important. These have been primarily agricultural
applications or for cellular concrete. For this purpose a mechanical foam
pump was developed which could provide expansions up to 30:1 but the
equipment has not been widely used.
Both low and high expansion equipment comes in fixed and portable units.
Low expansion generators come in several forms — playpipes, applicators, air
nozzles, etc. In each, the basic mechanism is the same — air is
mechanically entrained into a foam solution by agitation within a nozzle
system. Unit sizes range from as low as 5 gallons to 4000 gallons per
minute. The larger equipment is for fixed systems but the lower end of the
range can be used in truck mounted turrets or other similar mobile
installations. The largest sizes are primarily marine monitors for shipboard
use. In each type of system, output and foam projection are a function of
water flow and pressure. There is an optimum range of each for each piece of
equipment, but foam expansion does not vary appreciably over the total range.
The critical factor is to match foam agent input to water flow so that an
acceptable foam is made.
High expansion foam is made by spraying a foam solution against a screen
while applying blowing air from behind the screen. Foam generators range in
size from 100 cfm to 55,000 cfm output. The smaller sizes are usually the
air aspirating type of generator. In this form there are no moving parts.
The foam solution is forced at high pressure against the foam screen. The
force of the water spray aspirates sufficient air to form foam. Since air
aspiration is an inefficient process, expansion ratios are by nature on the
low end of the range. Small units start near 100:1 but even the larger, 1000
cfm, units do not exceed 450:1.
To achieve higher output and expansion ratios, it is necessary to incor-
porate a fan to provide air flow. Four types of fan drives are available —
water motor, electric motor, gasoline engine, or air motor. For reasons of
reliability and safety, water driven units are beginning to predominate. Two
drives are employed, one a water turbine and the second a reaction motor.
The latter is lighter, more efficient but by its design the water used to
drive the motor is not effectively converted into foam and it drains from the
generator as a heavy froth. This has been found to be unacceptable in spill
control for cryogenic or water reactive chemicals since it adversely
aggravates the spill.
The small air aspirating units can produce up to 1000 cfm of foam and can
be run from a hose line as portable equipment. Water and gasoline driven fan
units are also available as portable or wheeled units operating off hose
lines at outputs up to 6000 cfm. A few fire trucks have been built with
12,000 cfm water driven units mounted integrally. Water pressures from 40 to
80 psi are required, with the higher pressures needed for water driven units.
Foam pumps utilize mechnical agitation to make foam. Unlike low expan-
sion units where the force of water flowing through an agitator is the foam-
ing mechanism, the foam pump uses a motor driven agitator. It is this whip-
ping action which provides for the expansion in the 30:1 range. The need for
10
-------
the motor drive makes it possible to project the generated foam considerable
distances as a foam stream.
Generators of this type have been truck mounted and driven from its
power. They are utilized in the field for the generation of foam for use in
cellular concrete. Other units have been tractor mounted to generate foam
for agricultural application.
One additional form of foam generator deserves mention. This is the so-
called flooded plate generator. In this form a perforated metal plate is
mounted midway in a rectangular enclosure. The plate surface is channeled
and foam solution flowed down the channels. Air is introduced underneath the
plate to generate the foam. Output and expansion are controlled by the size
and distribution of the perforations, the rate of solution flow and the air
flow into the generator.
Flooded plate generators have application in that they are able to pro-
duce foam in the range of 100 to 200:1 with solutions too viscous to be suit-
able for normal high expansion generators. Thus, medium expansion foams can
be made from solutions containing high percentages of water soluble polymers.
These result in highly stable slow collapsing, slow draining foams. The
flooded plate can also be used where high percentages of solids have been
slurried into the solution. This makes it an interesting device for con-
sideration in the application of such materials for spill control.
PREVIOUS HISTORY OF FOAM APPLICATION TO CHEMICAL SPILLS
As stated earlier, the first use of foam in spill control was with hydro-
carbon fuels. Since that time there have been numerous investigations and
trials with foams for control of spills of a variety of chemicals. Except
for the extended work with liquefied natural gas these efforts have not been
comprehensive. A significant amount of data has been generated but has never
been consolidated. The Coast Guard program on hazardous chemicals that float
has brought most of this data together.
The data which has been gathered shows that commercial foams can provide
a variety of benefits in the control of hazardous chemical spills. These
benefits fall into two categories — the control of vapor release from the
spill surface and the isolation of the flammable materials from accidental
ignition. The control of vapor release is manifested in several ways. An
integral foam blanket isolates the surface from radiant energy, it acts as a
barrier to vapor passage and it can absorb the vapor to some extent. With
cryogenics whose vapor is buoyant at ambient, the foam acts to warm the vapor
such that is disperses better after permeating through the foam layer. With
aggressive water reactive chemicals, foam offers a mechanism to gently add
water to dilute the spill.
It should be noted that foam has not been found to be applicable to all
materials or classes of materials, nor does simple foam compatibility with a
chemical indicate any measure of spill control. Background knowledge in foam
applicability for spill control is detailed in the following sections.
11
-------
It should be recognized that foams are applicable to contained spills.
They do not seem to be adaptable to running spills. Ground conditions appear
to have a minimal effect on the vaporization of spilled liquids. Wet condi-
tions can aggrevate the initial boil off but it should be a transient condi-
tion. The only significant consideration comes when the spill, although con-
tained, is absorbed or submerged in sand, gravel or similar material. Since
foam can seal against such surfaces and the point of concern is vapor re-
lease, even this set of conditions was not a significant concern. On this
basis it was not felt necessary to modity the laboratory tests to compensate
for topographical conditions.
Volatile Organic Chemical
The largest group of materials in the hazardous chemicals list, to which
foam is applicable, are the volatile organics. With this group foam has two
actions — suppression of vapor release and surface isolation. The data on
the applicability of commercially available fire fighting foams to these ma-
terials has provided further substantiation of an observation made by MSAR in
an earlier EPA program (Friel, 1973). These foams show compatibility only
with those materials whose dielectric constant is less than three. There is
no satisfactory explanation of the influence of the dielectric constant but
the data is fairly conclusive.
By excluding materials above a value of three, polar compounds are ex-
cluded as are most organics with significant water solubilities. For those
nonpolar chemicals which do not degrade the common foams, the data shows that
the capability to restrict vapor release is a function of the solubility of
the chemical, and its vapor pressure. Previous attempts to quantify these
properties were not successful. An order of merit has been established and
it was believed that molecular size was the most important consideration,
followed by water solubility and vapor pressure. What the order of magnitude
of difference may be was not clear. The molecular size feature was compli-
cated somewhat by molecular structure, saturated, unsaturated, or cyclic and
the arrangement of radical groups.
Those materials with low molecular weights, benzene and ethyl acetate for
instance, have a fairly rapid permeation through surfactant foams regardless
of expansion. At the other extreme large molecular sizes, such as isooctane,
show extremely slow permeations and complete suppression may be observed for
more than a day. In general, foams offer some measure of vapor suppression
for all materials with which they are compatible where the vapor hazard is
fire or explosion.
In the case of toxic materials, however, the allowable vapor concentra-
tions are so low that foam can restrict it to less than the TLV for only very
short times. By restricting vapor release it does help the downwind hazard.
Workers in the area would have to use respiratory protection regardless.
Restricting the vapor concentration, even though it is not below the TLV or
even the minimum lethal dose, extends the life of canister protection and
minimizes the hazard due to mask or hose failures.
12
-------
There are at present a large number of foam agents available on the mar-
ket. In absolute terms they should all provide equal protection. The bubble
walls are the principal barrier and these are basically water. The small
amount of surfactant does not have a significant influence. The difference
comes in maintaining a specific foam thickness coupled with a slow drainage.
Low drainage, low collapse foams present a superior barrier and require less
makeup.
In earlier EPA sponsored work, (Friel, 1973) it was theorized that the
degrading action of chemicals such as the polar compounds was due to exag-
geration of drainage and/or precipitation of the surfactant. To combat this
gellation was proposed as a preventive action. A pectin fortified foam was
evolved which did extend compatibility to a broader range of chemicals along
with a capability to control vapor release. The system, however, was not
adaptable to available foam generating equipment and as yet that situation
has not changed.
In more recent work universal foams have been developed and marketed with
conpatability with a wide range of materials. They have not show a substan-
tial improvement in vapor control, however, over that possible with high
quality fire fighting foams. A foam system developed as a long lasting foam
for a nuclear application (Mecca, 1971) has demonstrated extended compati-
bility. This foam derives its life or collapse rate from the concentration
used. As the concentration was increased to extend life, the ability to
restrict vapor permeation also increased. It became clear that the foam
was demonstrating a capacity to absorb as well as block chemical vapors.
The mechanism of vapor passage through foam is one of slow permeation of
vapor through the bubble walls. Gas concentrations in the foam bubbles can
ultimately exceed the lower explosive limit. Thus, the foam could be ig-
nited. High expansion foam is more susceptible to this possibility than low
expansion.
If ignition does occur the burn back rate will be lower than over an un-
covered surface. Should the vapor concentration be in the explosive range,
the foam structure suppresses flame propagation and only burn back will
occur.
Liquefied Gases
Studies have been made with a number of cryogenic materials. They show
that the vapor hazard associated with spills of these materials could be
mitigated by foam application.
Liquefied Natural Gas—
Of the liquified gases which have been studied, liquefied natural gas has
received the majority of the attention. In worked sponsored by the American
Gas Association extensive testing was carried out with commercial high expan-
sion foams and equipment. The results have been quite positive.
13
-------
The AGA sponsored work showed that foam aided both vapor dispersion and
fire control. At the boiling point of liquefied natural gas, -161°C, the re-
leased vapor is very cold and denser than air. It moves downwind at or near
ground level where it could intersect an ignition source. Tests have shown
that if the vapors are forced to pass through a foam blanket a foot or more
thick, they warm sufficiently to be buoyant upon release to the atmosphere.
The water draining from the foam into the LNG exxaggerates the boil off rate.
The boil offrate is not too critical as long as the gas is warmed sufficient-
ly to be buoyant, and the effectiveness of the foam blanket is maintained.
Fast draining foams lose their fluidity and form openings in the foam layer
through which the vapors can pass rapidly with a minimum of warming. Thus
low draining, slow collapsing foams are desired.
These characteristics become much more important when fire is considered.
Foam is not able to extinguish LNG fires; the boiling action prevents sealing
of the surface. Due to the extremely low boiling point of LNG, spills rapid-
ly freeze the spill substrate. This acts as an insulating barrier and long
term thermal energy imput is mainly from radiation. In a fire, foam acts as
a barrier to radiant energy including reradiation from the fire. By means of
energy blockage, and reductions in fire intensity as high as 95% can be
achieved. Maintencance of fire control allows the spill to be burned off
under controlled conditions.
Data on LNG fire control is sufficiently well developed that NFPA 11A now
provides guidelines for high expansion foam system design for LNG facili-
ties.
Liquefied Petroleum Gas—
Liquefied petroleum gas is similar in some respects to LNG. Fire control
can be exercised by a foam blanket. The benefit of foam to the vapor hazard
of LNG with foam is strongly influenced by the fact that the vapor can be
made buoyant. This is not true for LPG, whose vapor is heavier than air at
all temperatures. The question of vapor mitigation with foam for propane is
still under discussion and needs further investigation.
Vinyl Chloride Monomer—
In the case of vinyl chloride monomer, there can be benefits from using
foam for vapor control. In tests with VCM, no measurements were made of
vapor dispersion per se. The measurements taken were of boil off rate, and
significant reductions were observed. Due to its much higher boiling point,
-14°C, the influence of drainage is considerably less than with LNG.
Further, the VCM gas density is greater than air. As a consequence the vapor
rises through a foam mass at a slow rate. It was concluded that each of
these, plus the blockage of radiant energy, contributes to the reduced boil
off rate.
Ammonia—
Due to the exceptionally large heat of solution, treatment of ammonia
spills with water sprays is not practical even though ammonia is infinitely
soluble. The early investigations with standard foams were not encouraging.
Foams were rapidly degraded. Some success was realized by incorporating
14
-------
materials into the foam which caused it to gel when neutralized. A system
was evolved in which the foam generated gelled upon application to ammonia.
Low expansion foam blankets were developed in laboratory tests which con-
trolled vapor releases from ammonia for hours.
Attempts to utilize this system on a larger scale were unsuccessful, how-
ever. Ammonia vapor in the air caused gelling of the foam at the point of
generation and plugging of the generator. Foam could be generated at some
distance from the spill and then pumped to the spill. At present, there are
no further efforts in this direction.
Chlorine—
Chlorine has been reported to be controllable with foam (Buschman, 1975).
The data is very limited however and the degree of vapor control is not well
defined.
Inorganic Water Reactive Chemicals
This heading treates materials which are normally liquids. Two chemicals
in this class have seen tests with foam. A formal test program was conducted
with spills of 863 (Pfann, 1976), while cursory tests were conducted with
SiCl4.
Sulfur Trioxide—
Under MCA sponsorship the manufacturers of 863, oleum and related com-
pounds, conducted extensive testing to find acceptable means of handling
spills of these materials. A few very effective techniques were found which
could be implemented in fixed installations. These were not acceptable for
field use, however. Although foam left much to be desired it was the only
technique of those available which had any merit.
Foam application to spills of 803, oleum, etc., results in rapid de-
struction of the foam due to the aggressive water reaction. The reactivity
with foam is considerably less aggressive than with water. However, the
lessening of reaction is desireable since dilution is a basic method of miti-
gating a spill of this type of material.
Reaction with water converts 863 to l^SO^., a hazardous material of
itself but not of the same order as 803. Additionally, it is a liquid
aerosol, rather than a vapor, and its downwind travel is restricted. Aggres-
sive water reactions force evolution of 863 along with 112804. Foam
acts to limit the release of 803.
Silicon Tetrachloride—
SiCl4 reacts with water to form SiC>2 and HC1. As with 803, the
acid aerosol is less hazardous then the tetrachloride.
In an actual spill situation foam was tried and abandoned (Hampson, 1976).
It caused what was considered excessive fuming. No data were obtained to de-
termine any change in downwind characteristics between a free spill and foam
treatment.
15
-------
SECTION 5
SCOPE OF WORK
The program to evaluate and develop foams for mitigating air pollution
from hazardous material spills was divided into three tasks. Task I was the
delineation of available systems, Task II involved field testing of available
systems and Task III was the examination of novel foam systems.
TASK I - DELINEATION OF AVAILABLE SYSTEMS
The emphasis in the first task was to define the capabilities and limita-
tions of existing foam technology, and to arrange this information in a
manual form for use by both f irst-on-the-scene emergency personnel and on—
site coordinators. This information would also serve to point out gaps in
the data and indicate areas of need for improved or novel foam systems.
State-of-the Art Survey
To initiate the program a state-of-the-art survey was made. In this, a
complete listing was made of all of the foam agents, both low and high expan-
sion, which are currently commercially available. They were characterized as
to their intended use and their physical and chemical properties. Per-
formance data was solicited as to foaming ability, drainage, collapse rate
and other pertinent data.
Simultaneously with this, solicitations were made to collect all avail-
able data on the use of foam for spill control, both field and in-plant use.
A general literature survey was conducted. Data were solicited from those
companies involved in foam systems, both manufacturers and system designers.
Government agencies, principally EPA and the Coast Guard, but including
NIOSH, OSHA and the Bureau of Mines, were surveyed. In addition, state agen-
cies and major city fire departments were solicited for any information they
had available.
The survey covered the chemical industry, primarily through appropriate
trade organizations, the Manufacturing Chemists Association, American
Petroleum Institute, The Chlorine Institute, Society of Plastic Engineers,
American Gas Association and others. Those research organizations which have
been active in hazard chemical spill work were covered, although most of the
data from this source was also available through the sponsoring agencies.
16
-------
Results of the Survey—
Thirty inquiries were sent to foam and foaming equipment manufacturers.
Seventeen replied, with some declining to participate in the program. Those
companies which responded are listed below:
1. Akron Brass Company
2. Allison Control, Inc.
3. American LaFrance
4. Bowman Distribution
5. Elkhart Brass Mfg. Co. Inc.
6. Feecon Corporation
7. Fire Boss
8. Fire-Chem Mfg. Co. Inc.
9. Huntington Laboratories, Inc.
10. Walter Kidde and Co. Inc.
11. Laurentian Concentrates, Ltd.
12. Mine Safety Appliances Co.
13. National Foam Systems, Inc.
14. Rockwood Systems Corporation
15. Santa Rosa Mfg. Co.
16. The Mearl Corporation
17. 3M Company
Of these seventeen, Akron Brass Company and Bowman Distribution stated that
their products had no capabilities in this area.
Literature Search
The literature search included sources such as Chemical Abstracts, Dis-
sertation Abstracts, Fire Research Abstracts and Reviews, and Industrial and
Engineering Chemistry. Most of the references obtained from these sources
pertained to the use of foams against burning materials.
More information was found on the use of aqueous film forming foams for
spill control than any other type of foam agent. The basic hypothesis for
vapor transfer resistance by thim films was intiated by Hederstrand (1924).
Various investigators have contributed to the research and development of
this approach. Tuve et al (1964) provided the first discussion with refer-
ence to aqueous film-forming foam.
Meldrum (1972) prepared a practical evaluation of aqueous film-forming
foams and their utilization. White (1976) expanded the data of evaporative
control by surfactant films by incorporating the temperature factor. The en-
vironmental problems posed by fluorochemicals were reported by Kroop and
Martin (1974), but there has not been much further public data in this area.
Archer (1955) investigated the effects of discontinuties in the film or vapor
release. All of these relate to nonflowing liquid surfaces.
A listing of pertinent documents reviewed is given in the Bibliography
appended to this report.
17
-------
Matrix Preparation
The data gathered was compiled and analyzed. A matrix was prepared which
listed those classes of materials for which there was data on vapor control
by foam applications. The matrix assesses each type of foam for each materi-
al and establishes a priority for the foam systems with each applicable ma-
terial.
As was expected, the data exhibited numerous gaps and in many cases was
not incontrovertible. Theoretically, we might consider extrapolating the
existing data but this could lead to marginal or even unacceptable recommen-
dations. It appeared better to develop data through experimental work and
then update the matrix.
The preliminary matrix is shown in Table 1.
Laboratory Program
Materials Selection—
The needs of the matrix after its initial development were clear. Defini-
tive data was needed for those basic categories of chemicals which pose a
vapor hazard to the environment or the on-the-scene personnel.
It was decided that the valid determination of the capability of foam can only
be made with real materials. Thus, selected hazardous materials (representa-
tive of each class) were used for testing. These can only be safely handled
in the laboratory. Thus the development of the basic data was done on a
laboratory scale. Subsequent field testing was used to verify these data
using simulant materials which could be released to the atmosphere.
Test materials were taken from liquefied gases, materials which have aggres-
sive water reaction and volatile organics. Some materials can fall in two
categories.
The initial liquefied gas selection came from the nonwater-reactive materials.
The important properties were the boiling point and the vapor density. To
span the range of both, three materials were selected. These were liquefied
petroleum gas (B.P. -42°C, vapor specific gravity at 0°C - 1.5), ethylene
(B.P. - 104° C, vapor specific gravity at 0°C - 1.0) and 1,3 butadiene (B.P.
-4°C, vapor specific gravity at 0°C - 1.9). By adding existing data for LNG
(B.P. - 161°C, vapor specific gravity at 0°C - 0.55) to these three, a range
of boiling points from -161 to -4°C is provided, covering a specific gravity
range from 0.55 to 1.9.
Three additional cryogenic materials were selected, ethylene oxide,
chlorine, and ammonia. Ethylene oxide (B.P. 11°C, vapor specific gravity -
1.5) provided a liquified gas with measurable water solubility but with a low
heat of solution. Chlorine (B.P. -34°C, vapor specific gravity - 2.4) served
to represent a material with an acid reaction. Ammonia (B.P. -33°C, vapor
specific gravity - 0.6) represents materials with an alkaline reaction. It
also has an aggressive water reaction.
18
-------
TABLE I. MATRIX OF FOAM CAPABILITIES TO SUPPRESS OR OTHERWISE MINIMIZE THE
RELEASE OF TOXIC OR FLAMMABLE VAPORS FROM SPILLED HAZARDOUS
CHEMICALS AS LISTED
Surfactant Fluoro-
Recommend. Low Ex. High Ex. ProteIn ProteIn AIcohoI AFFF Mech.
i-octane
n-heptane
Cyclohexane
Hexane
Benzene
Toluene
Gasoline
Kerosene
Naphtha
Methanol
l-propanol
ButanoI
ButyI Ce11osoIve
Acetone
Methyl Ethyl Ketone
Methyl i-butyl Ketone
Lacquer Thinner
Paint Thinner
SiCI4
i-amyI AIcohoI
Ethyl Acetate
ButyI Acetate
i-propyI Acetane
Methyl Aery I ate
MethyI MethacryI ate
S03
Acetic Acid
Caproic Acid
Methyl Bromide
Butyl Bromide
N,N'-dimethyI Formamide
TetrachIoroethane
n-Octanol
TetrahydronaphthaIene
LNG
Chlorine
Ammonia
R
R
R
R
R
R
R
R
R
R
R
R
ND
R
ND
R
ND
ND
R
ND
ND
ND
ND
R
R
R
R
ND
ND
R
ND
R
R
R
R
R
R
B+
B+
B+
B+
B+
B+
B+
B+
B+
E-
E-
E-
ND
E-
U
B+
U
U
C+
U
U
U
U
B+
B+
C+
A+
U
U
A+
U
B+
B+
B+
E-
B+
A+
A+
A+
A+
A+
A+
A+
A+
A+
A+
E-
E-
E-
ND
E-
U
A+
ND
ND
C+
U
-
-
-
A+
A+
C+
B+
ND
U
U
E-
A+
A+
A+
A+
A+
A+
B+
B+
B+
B+
-
-
B+
B+
B+
E-
E-
E-
U
E-
U
U
ND
ND
C+
U
-
-
-
ND
ND
C+
ND
ND
ND
ND
ND
ND
B+
E-
F-
B-
B+
B+
B+
B+
B+
-
-
B+
B+
B+
E-
E-
E-
ND
E-
ND
ND
ND
ND
C+
ND
U
U
U
U
U
C+
ND
ND
ND
ND
ND
ND
+
ND
F-
B-
B+
E-
E-
E-
E-
E-
E-
E-
E-
E-
A+
A+
A+
ND
A+
ND
ND
ND
ND
ND
ND
to
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
-
ND
F-
ND
ND
C+
C+
C+
C+
ND
ND
C+
C+
C+
E-
E-
E-
ND
E-
ND
ND
U
U
F-
ND
ND
ND
ND
ND
ND
F-
ND
ND
ND
ND
ND
ND
ND
U
F-
E-
E-
U
U
U
U
ND
ND
U
C+
U
-
-
-
U
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
U
U
ND
ND
ND
ND
U
ND
ND
ND
ND
19
-------
IDENTIFICATION OF SYMBOLS
+ Data available indicating vapor pollution control
- Data available indicating no vapor pollution control
U Limited data available-capabilities uncertain
ND No Data
R Foam use recommended over spill
ND Insufficient Data
A Best foam formulation
B Next best foam formulation
C Acceptable in some situations; see text
E Unsuitable foam formulation
F Deleterious foam formulation
Foams are best when ambient is above freezing, winds are less than
15 knots and there is no precipitation. Foam can be employed as
long as it can be generated. Temperature effects will be seen
first in generator operation. Rain will exaggerate collapse but
will not impair foam generation or flow. Wind above 10 mph will
lift foam. Downwind barriers will assist in holding foam in place.
Low expansion is more tolerant of weather than high expansion but
where high expansion is shown to be the best choice the inability
to build thickness should be the only factor considered in
switching.
20
-------
To include aggressive water reactants which are acidic, two materials —
sulfur trioxide and anhydrous hydrogen fluoride — were selected. A listing
of the properties for the liquefied gases or water reactive chemicals used in
the program is given in Table 2.
Although the liquefied gases and water reactive liquids offer interest-
ing materials from the standpoint of foam application, the large number of
volatile chemicals do not exhibit either property. These materials are main-
ly volatile organics. In the work prior to this study, three properties have
been shown to be important — molecular weight, water solubility and vapor
pressure. There were indications also that the degree of saturation might be
a significant variable, along with the chemical arrangement straight chain or
cyclic.
The selecting of compounds to cover the volatile organics was restricted
to those materials whose vapor pressure at 20°C is greater than 0.5cm. Polar
compounds were given token representation since it had already been es-
tablished that only alcohol type foams were applicable.
Due to the large number of chemicals falling in this category, the se-
lection was made to cover the three variables deemed most significant by the
dielectric constant, vapor pressure and water solubility. Final selection
was made after the initial matrix was developed.
For the polar group the dielectric constant was the prime variable, fol-
lowed by vapor pressure and water solubility. The range of variables was
covered with four materials. Three materials were selected to cover a range
of dielectric constants from about 2.5 to 5.01. Available data indicated
that below 2, materials would behave the same as nonpolar compounds. Above a
dielectric constant of 4, non alcohol types of foam are seriously degraded.
One compound of high dielectoric constant and high water solubility was se-
lected to complete this category. The selected polar materials are listed in
Table 3. They cover a range of molecular weights from 58 to 116, vapor pres-
sure varies from 1 cm to 55 cm, and water solubility from 0.5 g/cc to com-
plete miscibility.
With nonpolar compounds, the dielectric constant is less significant
since most are below a value of 3. Vapor pressure and water solubility were
felt to be the principle variables.
Table 4 identifies the selected nonpolar candidates. The first three on
the list are of the same family, differing in vapor pressure and molecular
weight. Cyclohexane has properties similar to benzene except for its water
solubility and the fact that it is a five membered ring.
N-Octane was selected as a representative of long chain aliphatic ma-
terial of low water solubility.
Selection of Foam Chemicals—
The selection of the foams used in the laboratory tests came from those
in wide use by the five service. To insure objective selection, foam charac-
terization was performed by University Engineers on independent laboratory.
21
-------
CO
^
Q_2
M
33
w
B
w
>
1— 1
H
O
Cd
W
H
IS
cd
0
CO
w
CO
^
e>
Q
W
W
o-
J
["T!
O
CO
W
H
Cd
Cd
CM
|
•
CM
Cd
M
^J
H
01
^1 *Q
3 -H
14-1 ^
i-l 0
D -H
co 1-1
H
C 01
*» '^
J= X
4J O
w
0>
c
0)
-0
cd
4-)
3
03
CO
r-H
\o m
• •
O 00
-H CO
,-H
sr
•
si- 0
1 0
i — i
QJ
cd
a
o
}^
CM
-H 00
• *
CM -H
st O
1 ^H
01
C
0)
1— 1
>".
r^ st
• •
co m
O -H
1
1
1
1
vO
vO
CO
0
F-4
CO
CM
O
00
vO
vO
oo
CM
CM
CN
es
CN
00
vO
ON
vO
oo
CM
vO
o
CT>
m
vO
m
o
in
vO
•
-*
ro
r^
I
CN
fl
I
vO
O
o
oo
O
CM
CO
o
O
•
m
co
CM
4_i
0>
a
o
CM
c
o
00
c
•H
f-H **"v
•H O
O o
« *~s
o
4-1
cd
0)
4-1
c
Ol
4-1
CB
!-4
C
O
• H
4-1
N
• H
l-l
O
a
cd
^-^
00
f_t
cd
0
4J
cd
0) x~>
IB 0
CM
o 5d
•H
M-I DO
O i— 1
o> cd
a o
co ^
• 3
O •
• i— i
co
I
O V-i
CX-r-l
cd cd
E> •— '
O
CO
•o
• H
3
cr
•H
i-J
o
i
)»i
01
4-1
cd
^
*^
|
4J
3
i— i
O
CO
M-4
O
4-1
cd
01
X
s~^
00
*^^.
1—4
cd
o
4-1
si
00
• H
01
cd
i— i
3
u
0)
1— i
o
X
22
-------
TABLE 3. CANDIDATE POLAR MATERIALS
Name
Dielectric
Constant
Vapor
Pressure
(mm)
Water
Solubility
(g/cc)
Molecular
Weight
Acetone
n-butyl acetate
Ethyl ether
Triethylamine
20.7
5.01
4.41
2.42
180
10
550
77
Miscible
0.5
7.5
1.5
58
116
74
101
TABLE 4. CANDIDATE NONPOLAR MATERIALS
Name
Dielectric
Constant
Vapor
Pressure
(mm)
Water
Solubility
(g/cc)
Molecular
Weight
Cyclohexane
Benzene
Toluene
Ethyl benzene
n-octane
2.05
2.28
2.38
2.41
1.95
100
95
35
10
14
0.0001
0.080
0.047
0.014
0.0015
84
78
92
106
114
23
-------
It was their task to obtain samples of all major commercially available foam
agents of each class — protein, fluoroprotein, AFFF, alcohol and surfac-
tants, and for each applicable expansion ratio, establish the important
properties of drainage, collapse, pH and surface tension. For each property,
standard tests were used if available. For low expansion foams, tests are
established and detailed in NFPA 11 and 11B, and appropriate federal and
Underwriters Laboratories specifications. High expansion, however, did not
have a standard test for drainage and collapse. This has been corrected
since this program was conducted. For this program, the test in use at the
time by MSAR was used. This test measures drainage and collapse of a column
of 500:1 expansion foam 61 cm in diameter by 91 cm in height. Based upon
these data, representative agents were selected for use in the laboratory
test program.
Commercial fire fighting foams were ordered for testing by University
Engineers from authorized distributors of each manufacturer. Foam agents
were purchased of the following manufacturers:
1. Rockwood - protein, alcohol and surfactant
2. Mearl - protein, fluoroprotein and surfactant
3. National - protein, fluoroprotein, alcohol, AFFF and surfactant
4. Huntington - surfactant
5. Laurentain (Lorcon) - protein, fluoroprotein, alcohol, AFFF and
surfactant
6. 3M - AFFF
7. Kidde - surfactant
8. Elkhart - surfactant
9. MSA - surfactant
Akron Brass, a manufacturer of surfactant foams, declined to participate and
our inquiries to Jefferson Chemical (Firesorb) and Western Fire (Firechem)
were not answered.
The results for low expansion foams except AFFF's are shown in Table 5.
Data was obtained using a 6 gpm nozzle designed according to a joint Army,
Navy and Air Force standard (JANAF). The drainage data was obtained using
the test described in NFPA 11.
The results for AFFF's are given in Table 6. The same low expansion
foam equipment and tests were used for the AFFF tests as were used for the
other low expansion agents.
Table 7 presents the results for high expansion foams. It should be
noted that in some cases the reported values for each category are not con-
sistant with those reported by the manufacturers. The data has been derived
using standard equipment and accepted techniques. It is presented here as
reported to MSAR by the independent testing laboratory.
Two foam agents were chosen for the laboratory test program from each of
the protein, fluoroprotein, alcohol and AFFF manufacturers. Due to effective
regulation of the industry, most agents are somewhat comparable.
24
-------
CO
H
q
28
£
w
H
M
O
3
O
w
i— i
H
04
OH
O
04
OH
•
in
w
,.j
pa
H
0)
d) cfl c
4J C -H
U 'co S
3 M
S
Q) C y
o o ^
CO '"^ CO
14-1 CO CO
V4 C C
3 ,
CO H T3
•
c
1— 1
o
T3
fii —f
PL CO
c
CO
4-1
CO
c
o
•H
S*$ 4J
• r4
•o
5
O
•r4
^J
CO
OS1 T3
C QJ
01
•H
eg
C C
CO TH
a
X
w
c
o
&*S 4-)
•H
•o
5
o
u
cfl
OS
C CO
0 hJ
•r4
co c
C -r-l
cfl
OH
X
£5
cO
O
voomco oo CM in *H r^vt t^co m CM
VOOSOSCM c^ovr^oo
in vo in *d* ^f ^ *n vo *d~ *d~ co m in o co
o
CO
•1-1
^
I
I
4J
O
OOOO OOOO OO OO OSO
CO CO CO CO vO vO vO vO CO CO vO vO vO Cl) vO
4-1
o
c
•o
r-4
3
0
o
-HCOf^^ CMOSVOCO COCO CMr-. 1^ -H
• ••• •••• •• •• • •
*^f «d" CO ^^ ~oo
inoooor^moovovo
oo-d-ovovovo
oovoooinmoo
vooo
-------
cu c e
o o o
Cd -H •>>.
y-i co co
rJ C 0)
3 cu c
tn t-< >
o
CM
ON
oo
CM
ON
OO
•sj
CM
•
00
ON
ON
03
H
en
CU C8
4J C
^ -H
CQ CO
3 VJ
o- o
CM
ro
o
O
ON
CM
vO
•
o
fa
1
£EH
o
W
o
z
r-l
^
05
O
Q
O
r-
v£>
vO
o
on
o
ro
C5
on
o
vO
4-1
B
cu
60
<
B
cd
0
O
Ui
0)
I—I
td
C
0
•H
4-1
cd
Z
m
CO
3
r-l
o.
B
cd
o
fa
B
0
O
^4
0
rJ
s-s
CO
0)
4J
cd
^
4J
r^
60
'rJ
rS
CO
CO
O
fe
i— I vO
cQ
C ta
O fe
• H ptj
^j ^ri
cd
Z
1-1
cu
4J
cd
4J
60
rJ
s
m
26
-------
cu
to ^
a. a) 1-1
CO 4J .C
i-l
OJ |_*
M 3
<« f- 0|
C • H CM
o
o
o
o
o
o
o
o
O r-
-H ON
m
oo
ao
vO
ON
in
vO
m
m
oo
m
o
es
CO
m
oo
CO
oo
W
35
O
M
lurface
'ens ion
B
o
co
C
O -H
• •
r^ CN
n en
•
m
CM
in
CM
m
CM
m
CM
CM
CM
CM
vO
CM
WHO
CO
w
w
a,
s
oo
ON
•
m
ON es
m oo
oo
oo
PQ
IS
c ^
O 5-S
T3 4J
•0 cfl
CM
C
0)
60
C
o
4J t>0
S
n)
-------
TABLE 8. REPRESENTATIVE LOW EXPANSION FOAMS
Foam Type
Protein 3%
Protein 3%
Protein 6%
Protein 6%
Fluoroprotein 3%
Fluoroprotein 3%
Fluoroprotein 6%
Fluoroprotein 6%
Alcohol
Alcohol
Surfactant
Surfactant
Surfactant
Surfactant
AFFF 3%
AFFF 6%
Foam Agent
Rockwood
National
Lorcon
National
Lorcon
National
Mearl
National
National
Rockwood
Lorcon
Huntington
Mearl
MSA
National
3M
Surface Tension
(dynes/cm)
50
40
47.5
46.4
29.4
30.2
27.6
27.7
38.2
51.1
26.5
37.0
25.3
25.6
20.3
18.6
Quarter
Drainage
Time (min)
.66
1.23
.45
1.41
.37
.54
.37
.63
.42
1.25
.53
.69
1.25
1.69
1.42
0.76
28
-------
The hydrocarbon surfactants are essentially unregulated and a broad
spectrum of materials are available. The existing data indicates that
only slow draining materials should be selected. It appeared important,
however, to fully define capability as a function of drainage and collapse
for the information of potential users. Four surfactant foam agents were
chosen for laboratory testing to encompass the available foam properties.
All the foam agents chosen are listed in Table 8. In each case, they
reflect the range of drainage and surface tensions in each category.
Test Program—
With the selection of the test chemical and the foams to be evalu-
ated, the laboratory test program was initiated. Three separate test
sequences were used, one for low expansion foam, one for medium expansion
foam and one for high expansion foam. For low expansion materials, expan-
sions of 2.5:1 were evaluated. The low expansion foams were produced
using a laboratory blender. Medium expansion foams of 250:1 were produced
with an air aspirating generator. High expansion foams at 500:1 were pro-
duced with an induced air generator.
Laboratory Test Procedures for Volatile Organics—The test sequence for this
class of materials paralleled one successfully employed in previous work. A
250 ml volume of the chemical to be evaluated was placed in a 30 cm diameter
glass vessel. A foam layer was applied to the surface. The thickness was 5
cm for low expansion foams, 30 cm for medium expansion and 50 cm for high
expansion. The vapor concentration above the foam was continuously moni-
tored and recorded.
Sensors were placed in the laboratory test vessel to measure vapor
permeation through the foam. The sensors consisted of the MSA Series 500
Combustible Gas Detection System and chart recorders to continuously monitor
vapor concentration versus time above the foam layer. Additional gas detec-
tors were provided to measure the vapor concentration versus time within the
foam. This is an equilibrated test sequence that defined the rate of perme-
ation through the foam and the time period for which a given thickness of
foam provided protection.
By using a specific thickness with no make up, the influence of drain-
age and collapse rate on the length of protection was defined. Relating the
rate of permeation for a wide variety of materials to their properties, such
as water solubility and vapor pressure, permits the interpolation of the
data to at least all of the materials in each subgroup if not the complete
class.
Laboratory test procedures for liquefied gases—The test sequences for these
materials were similar to those for the volatile materials. Vaporization
rate and dispersion are the important factors. Since foam can neither de-
velop pressure over the spill or completely exclude radiant energy, spills
of these materials boil even with a foam cover. The lack of water solubili-
ty means a continuous rapid release of gas. Foam isolates the surface,
reduces the temperature differential and warms the gases.
29
-------
The laboratory test sequence for insoluble liquefied gases was similar
to that used for volatile organics. A glass vessel 30 cm in diameter with a
volume of 33 liters was used. The glass vessel was insulated and the chemical
to be tested was added until a 1.3 cm liquid layer existed in the bottom of
the vessel. Foam was added to the depths specified previously. Because of
problems of weighing or measuring the volume of off gasing from the lique-
fied gas, only visual observations could be made. Notes were taken on the
intensity of boiling at the foam-liquid interface as well as the behavior of
the foam layer. For reactive liquids the extent of foam degradation was
qualitatively noted.
During the test program, conditions which could be encountered in the
field, wind, rain, temperature, etc., were not factored in. The laboratory
tests were based upon reproducible conditions to obtain results which can be
evaluated, analyzed and compared.
Laboratory test results for volatile organics—The test method used involved
an uncovered test chamber and continuous monitoring of the vapor concentra-
tion just above the collapsing foam column. This test method simulated the
real conditions of unrestricted vertical dissipation and the dynamic be-
havior involved during foam collapse. In previous studies the test proce-
dure for evaluating the vapor mitigation properties of foams involved a
closed test chamber in which samples were taken at a constant height above
the spill surface independent of the collapsing foam level. This method
gave abnormal readings since vapor dissipation was restricted in the verti-
cal direction. The vapor concentration just above the foam layer was not
obtained. One real condition which is simulated by the uncovered test cham-
ber is that of swirling; small eddys found at the liquid-air interface.
Swirling effects can change the vapor concentration above a spill by several
hundred percent; an important consideration especially near the lower ex-
plosive limit (LEL).
One data point of importance was the height above a spill corresponding
to the LEL. Figure 1 is a graph of the benzene vapor concentration versus
height above a benzene spill surface with no foam cover. The test is done
in a 30 cm diameter by 45 cm high glass battery jar. As the detector head
is lowered into the test chamber the LEL for benzene is found to occur bet-
ween 9 to 18 cm above the spill surface. The vapor concentration falls off
very quickly above 18 cm. A covered test chamber increased the height where
the LEL is reached to approximately 30 cm. The graph indicates that even
when considering swirling effects, the benzene has a maximum height at which
the LEL can occur.
By following the collapsing foam column, the vapor concentration
leaving the foam surface can be recorded. It can be considered that when
the residual foam thickness is below the critical LEL height, the foam has
lost its effectiveness.
Figure 2 shows a graph of the benzene vapor concentration for a medium
expansion foam column at 10 cm and 20 cm above the spill, as well as the
vapor concentration above the collapsing foam column. The initial foam
thickness was 30 cm. At 10 cm all the foams (Emulsiflame, Lorcon, MSA) had
30
-------
2.6 .
2.4
2.2-
2.0 .
^•s
" 1.8
e
o
•^ ic.
4-1 1.6
cd
0)
g 0.8
N
I 0.6
0.4
0.2
0.0
Benzene liquid surface
0
Graph shows the upper and lower concen-
tration percent due to swirling and changes
in ambient conditions
46 8 10 12 14 16 18 20
Height of Detector Head (inches)
LEL
Figure 1. Benzene vapor concentration vs detector height
31
-------
Emulsiflame - 4 in.
Lorcon - 4 in.
MSA - 4 in.
Medium expansion - 12 in. initial foam height
Detector at 4 in., 8 in., and must above
collapsing foam column (adjustable)
Emulsiflame Lorcon
Adjustable (5 in.)* Adjustable (6 in.)
MSA Adjustable (5 in.)
LEL
in.
- 8 in.
*height of adjustable head
when LEL is reached in
inches
0 10 20 30 40 50 60 70
Time (minutes)
80 90 100
Figure 2. Benzene vapor concentration percent vs time for
MSA, Loron, Emulsiflame
32
-------
the same rate of increase in the benzene vapor concentration. At 20 cm the
height at which the LEL is reached above the collapsing foam column is approx-
imately the same.
The mitigation properties of medium expansion foams can be related to the
collapse rate, rather than the drainage rate. It should be noted, however,
that a foam with a dry cellular structure (fast drainage, slow collapse) is
less resistant to environmental effects and these influences could material-
ly alter the capability of the foam.
The aqueous film-forming foams (AFFF) results are shown in Figure 3. A
5 cm cover of low expansion and 30 cm of medium expansion foam has about the
same vapor mitigation time. The 3% National Foam which had improved vapor
mitigation time due to its slower collapse rate. When comparing the mitiga-
tion time of the low expansion with a 5 cm foam cover to that of medium
expansion at 10 cm above the spill, the low expansion controlled the vapor
permeation through the foam better than medium expansion. The low expansion
foam layer can become saturated, wherein the LEL will be above the 5 cm foam
thickness. In the medium expansion foam, as long as the foam height remain-
ed above the critical LEL height of 9 to 18 cm, the LEL concentration will
occur within the foam mass.
Figure 4 is a graph of the vapor concentration of the cyclic hydrocar-
bons (cyclohexane, benzene, toluene, and ethyle benzene) versus time for the
high expansion surfactant foams (MSA Ultra, Lorcon Full-Ex, Emulsiflame).
Octane behaved like ethyl benzene in terms of its vapor mitigation time.
The graph shows the mitigation time provided by a 50 cm foam blanket. The
polar hydrocarbons (acetone, n-butyl acetate, ethyl ether, and triethy-
lamine) caused all the high expansion foams to collapse, rapidly. From
Figure 4 the slow draining, slow collapsing MSA Ultrafoam had the best miti-
gatin time, the fast draining, slow collapsing Lorcon Full-Ex foam had an
intermediate mitigation time and the fast draining, fast collapsing Emulsi-
flame had the shortest mitigation time.
The medium expansion data for the surfactant and Aqueous Film Forming
Foams (National 3% and Light Water 6%) are given in Figures 5 and 6. Once
again, only the nonpolar hydrocarbons are shown since the polar hydrocarbons
caused rapid foam collapse. In general, the mitigation times follow the
collapse and drainage rate.
The data for low expansion foams is shown in Figures 7-10. The expan-
sion rate was approximately 2.5:1 and the initial foam height was 5 cm. The
foam categories are: 3% regular protein, 6% regular protein, 3% fluoropro-
tein, 6% fluoroprotein, 6% alcohol, 6% surfactant, 3% AFFF and 6% AFFF. In
Figure 7 the mitigation times of the various foams above a simulated acetone
spill are shown. The Rockwood 6% alcohol foams has a very long mitigation
time. The National Aerofoam 99-6% (alcohol), 3% National regular protein
and 6% National regular protein have intermediate mitigation times while the
remaining foam agents show short mitigation times. In Figure 8 the various
foams are shown with benzene. In this case the 6% surfactant, 3% AFFF and
6% AFFF show the shortest mitigation times, while the 3% National fluoropro-
tein and 6% Lorcon regular protein show intermediate times and the remaining
33
-------
AFFF-medium and low expansion
Foam height 2 in. - low expansion
Foam height 18 in. - medium expansion
3% National Foam
6% Light Water
low exp.
0.0
20
LEL
3% med. exp.
height 7 in.
reached
adjustable
when LEL
6% med.
exp. adjustable
detector/height
6.5 in. when
LEL reached
40 50 60 70 80
Time (minutes)
90 100 110
Figure 3. Benzene vapor concentration vs time
34
-------
2.8
2.6
2.4
2.2
2.0
1.8
o 1.6
•H
4J
g
§
1.4
A. MSA Ultrafoam 2% B. Lorcon Full-Ex 2% C. Emulsiflame 2%
1. Cyclohexane 1. Cyclohexane 1. Cyclohexane
2. Benzene 2. Benzene 2. Benzene
3. Toluene 3. Toluene 3. Toluene
4. Ethyl Benzene 4. Ethyl Benzene 4. Ethyl Benzene
Cl B
*room temperature approximately
10°F below normal
C2 K3/ B2
3 1-0
a
0.8
0.6
0.4
0.2
0.0
0 10 20 30 40 50 60 70 80 90 100 110 120 130
Time (minutes)
Figure 4. High expansion foams with nonpolar hydrocarbons
(initial foam height was 20 in.)
A4
35
-------
A. MSA Ultrafoam 2% B. Lorcon Hi-Ex 2% C. Emulsiflame 2%
1. Cyclohexane 1. Cyclohexane 1. Cyclohexane
2. Benzene 2. Benzene 2. Benzene
3. Toluene 3. Toluene 3. Toluene
A3
0.0
0 10 20 30 40 50 60 70 80 90 100 110 120 130
Time (minutes)
Figure 5. Medium expansion foams with nonpolar hydrocarbons
(initial foam height was 12 in.)
36
-------
D. AFFF Light Water
1. Cyclohexane
2. Benzene
3. Toluene
4. Ethyl Benzene
E. AFFF National Foam 3%
1. Cyclohexane
2. Benzene
3. Toluene
4. Ethyl Benzene
E2
0.0
0 10 20 30 40 50 60 70 80 90 100 110 120 130
Time (minues)
Figure 6. Medium expansion foams with nonpolar hydrocarbons
(initial foam height was 12 in.)
37
-------
1. National 3% Regular
2. Rockwood 3% Regular
3. National 6% Regular
4. Lorcon 6% Regular
5. Lorcon 3% FP
6. National XL-3% FP
7. Mearl 6% FP
8. National XL-6% FP
9. Rockwood Alcohol 6%
10. National Aerofoam 99-6%
11. Flame-Out 6%
12. MSA Ultrafoam 6%
13. Rockwood Jet-X 6%
14. Lorcon Full Ex 6%
15. National 3% AFFF
16. 3M Light Water 6% AFFF
3.0-.
10 20 30 40 50 60 70 80 90 100 110 120 ]30 140
Time (minutes)
Figure 7. Acetone with low expansion foam
(initial foam height was 2 in.)
38
-------
1.
2.
3.
4.
5.
6.
7.
8.
National 3% Regular
Rockwood 3% Regular
National 6% Regular
Lorcon 6% Regular
Lorcon 3% FP
National XL-3% FP
Mearl 6% FP
National XL-6% FP
9. Rockwood Alcohol 6%
10. National Aerofoam 99-6%
11. Flame-Out 6%
12. MSA Ultrafoam 6%
13. Rockwood Jet-X 6%
14. Lorcon Full Ex 6%
15. National 3% AFFF
16. 3M Light Water 6% AFFF
10 20 30 40 50 60 70 80 90 100 110 120 130 140
Time (minutes)
Figure 8. Benzene with low expansion foam
(initial foam height was 2 in.)
39
-------
1. National 3% Regular 9.
2. Rockwood 3% Regular 10.
3. National 6% Regular 11.
4. Lorcon 6% Regular 12.
5. Lorcon 3% FP 13.
6. National XL-3% FP 14.
7. Mearl 6% FP 15.
8. National XL-6% FP 16.
Rockwood Alcohol 6%
National Aerofoam 99-6%
Flame-Out 6%
MSA Ultrafoam 6%
Rockwood Jet-X 6%
Lorcon Full Ex 6%
National 3% AFFF
3M Light Water 6% AFFF
3.O..
16
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
Time (minutes)
1,J,b,
459
Figure 9. Cyclohexane with low expansion foams ">•>->•>?
(initial foam height was 2 in.)
40
-------
1. National 3% Regular
)4 2. Rockwood 3% Regular
2 3. National 6% Regular
|£ 4 4. Lorcon 6% Regular
-7 ,0 5. Lorcon 3% FP
/ 1 X
3.0
2O
H
• w
2.6 •
2 A
• H -
J
2.2 /
02.0.
c 1.8
o
4-1
td T /•
M 1.0
4-1
C
Q) i /
o 1.4
d
o
o
n 1-2
o
1* i.o
0.8
0.6 •
0.4
0.2
f I •»
58
13 61 2
Ith
c*il
1
y
•H
6. National XL-3% FP
m 7. Mearl 6% FP
IU 8. National XL-6% FP
9. Rockwood Alcohol 6%
1 10. National Aerofoam 99-6%
11. Flame-Out 6%
12. MSA Ultrafoam 6%
13. Rockwood Jet-X 6%
14. Lorcon Full Ex 6%
15. National 3% AFFF
16. 3M Light Water 6% AFFF
'
1 9
0 10 20 30 40
50 60 70 86 90
Time (minutes)
100 110 120 130 140
Figure 10. Ethyl ether with low expansion foams
(initial foam height was 2 in.)
41
-------
foams all have long mitigation times. In Figure 9, cyclohexane is the
hydrocarbon tested with the various foams. The 6% surfactants again show
the shortest mitigation times. The 3% and 6% AFFF's are intermediate and
the remaining foams are all very effective in mitigating the vapors from
cyclohexane. Figure 10 shows the foam agents with ethyl ether. Only the
Rockwood 6% alcohol foam shows a long vapor mitigation time. The remaining
foam agents have short mitigation times.
The data for the tests with the remaining volatile chemicals are given
in Figure 11 through 15. Figure 11 gives the data for low expansion foams
with toluene, Figures 12 through 15 give similar data for ethyl benzene, n-
butyl acetate, triethylamine and n-octane, respectively.
The low expansion foams were resistant to collapse with all polar
hydrocarbons except triethylamine. Only the fluoroprotein and Rockwood
alcohol foams did not break down with triethylamine.
In general, the relative mitigation times of the low expansion foam
categories are: 6% alcohol>3% or 6% AFFF's>6% surfactant. Of all the
properties examined, drainage was determined to be the most important. In
Figure 16, the rate of vapor release is plotted for a spill of benzene under
essentially stag-nant conditions. Curve is for a free spill. Curves 2 and
3 compare the effectiveness of two protein foams at the same expansion and
cover thickness. Clearly Protein B is the superior material.
The only difference between protein foams A and B is drainage. The
point is presented more clearly in Figure 17 where the thickness of foam A
has been doubled in the respect to foam B and it is still not equivalent.
This feature was found to be true of all foam types. Figures 18 shows
the vapor control capabilities of high expansion foam of two surfactant
agents. Here also the slower draining foam provides the superior perform-
ance.
Laboratory test results for liquefied gases—The result of foam application
on a liquefied gas appears to be dependent on the drainage characteristics
of the foam. The water in the foam supplies heat to the liquefied gas in
three possible ways: the latent heat of fusion as the water freezes, the
heat capacity due to temperature changes and the heat of solution for water
reactive chemicals. The heat input by water increases the evaporation rate.
In many cases the heat input from the freezing water is greater than the
heat of vaporization of the liquefied gas. A very low rate of water addi-
tion is desirable; this can best be accomplished by using a high expansion
foam with a low drainage rate. However, high expansion foam (500:1) could
not be tested in the laboratory due to the large quanity of liquefied gas
needed for the high expansion foam tests. More specific information on the
performance of high expansion foam had to await field testing. Only repre-
sentative foams from the medium and low expansion categories were tested.
The following foams were used in the laboratory liquefied gas study:
42
-------
3.0 •
2.8
2.6 f
2.4
2.2 -
2.0 ..
1.8
1.6 -
(3
O
t-l
4-1
cfl
M
4J
0)
O
e
O
0 1.2
n
O
p.
£ !
1.4 •-
0 -•
0.8 -.
0.6
0.4
0.2
0.0
1. National 3% Regular
2. Rockwood 3% Regular
3. National 6% Regular
4. Lorcon 6% Regular
5. Lorcon 3% FP
6. National XL-3% FP
7. Mearl 6% FP
8. National XL-6% FP
9. Rockwood Alcohol 6%
10. National Aerofoam 99-6%
11. Flame-Out 6%
12. MSA Ultrafoam
13. Rockwood Jet-X
14. Lorcon Full-Ex
15. National 3% AFFF
16. 3M Light Water 6% AFFF
0 10 20 30
40 50 60 70 80
Time (minutes)
90 100 110 120 130
Figure 11. Toluene with low expansion foams
(initial foam height was 2 in.)
43
-------
£s
fi
o
4J
ca
Concent
M
O
>
3.
2.
2.
2.
2.
2.
1.
1.
1.
1.
1.
0.
0.
0.
0.
0.
1.
2.
3.
4.
5.
6.
7.
8.
o.
8
6
4
2
o.
8-
6.
4
2
0
8
6
4
2
0
9.
10.
11.
12.
13.
14.
15.
16.
•
v--'
1 '^^f
0 10 20
National 3% Regular
Rockwood 3% Regular
National 6% Regular
Lorcon 6% Regular
Lorcon 3% FP
National XL-3% FP
Mearl 6% FP
National XL-1% FP
Rockwood Alcohol 6%
National Aerofoam 99-6%
Flame-Out 6%
MSA Ultrafoam
Rockwood Jet-X
Lorcon Full-Ex
National 3% AFFF
3M Light Water 6% AFFF
-^^^
^- , ^-' -^ •
30 40 50 60 70 80
Time (minutes)
11
^— --~ / q12
90 100 110 120 130 4,5
fi'Q
Figure 12. Ethyl benzene with low expansion foams
(initial foam height was 2 in.)
44
-------
^
a
o
•H
4J
to
1-1
4J
01
a
c
q
u
M
o
cd
3.
2.
2.
2.
2.
2.
1.
1.
1.
1.
1.
0.
0.
0.
.
.
1.
2.
3.
4.
5.
6.
0-
8-
6-
4.
2.
0-
8-
6-
4-
2-
0-
8-
6-
4-
•
7.
8.
9.
• 10.
11.
. 12.
13.
14.
15.
16.
,x—
'LL
. — -7 y
n i r\
National
Rockwood
National
Lorcon 6%
Lorcon 3%
National
Mearl 6%
National
Rockwood
National
Flame-Out
3% Regular
3% Regular
6% Regular
Regular
FP
XL-3% FP
FP
XL-6% FP
Alcohol 6%
Aerofoam 99-6%
6%
MSA Ultrafoam
Rockwood
Jet-X
Lorcon Full-Ex
National
3M Light
x
-------
Q
a
0
•H
4J
n)
i-i
•J-J
C
0)
0
e
o
o
0
(X
>
*1
*2
*3
*4
5
6
7
8
9
3.0 .
28 ,
• \j
2.6 •
2.4 -
2.2 -
2.0 .
1.8
1.6-
1.4
1.2
1.0-
0.8
0.6.
0.4
0.2
0.0
0
L *10
*11
*12
*13
*14
*15
*16
7 5
•
/
|6 f/
ft / /
//
II II
II If
If //
II //
if //
II u
II I
II I
II /
/
i
1
ll //
JnT 1
•/\/\
10 20 30 40
. National 3% Regular
. Rockwood 3% Regular
. National 6% Regular
. Lorcon 6% Regular
. Lorcon 3% FP
. National XL-3% FP
. Mearl 6% FP
. National XL-6% FP
. Rockwood Alcohol 6%
. National Aerofoam 99-6%
. Flame-Out 6%
. MSA Ultrafoam
. Rockwood Jet-X
. Lorcon Full-Ex
. National 3% AFFF
. 3M Light Water 6% AFFF
*foam collapses
a
50 60 70 80 90 l6o lio 120 130
Time (minutes)
Figure 14. Triethylamine with low expansion foams
(initial foam height was 2 in.)
46
-------
3.0 •-
2.8 -
2.6 -
2.4 "
2.2
2.0
§
2
•u
c
o
CJ
1.4
o
I* 1.0
0.8
0.6
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
National 3% Regular
Rockwood 3% Regular
National 6% Regular
Lorcon 6% Regular
Lorcon 3% FP
National XL-3% FP
Mearl 6% FP
National XL-6% FP
Rockwood Alcohol 6%
National Aerofoam 99-6%
Flame-Out 5%
MSA Ultrafoam
Rockwood Jet-X
Lorcon Full-Ex
National 3% AFFF
3M Light Water 6% AFFF
1,2,3,4,5,7,
8,10,11,12,
13,14,16
10 20 30
40 50 60 70 80
Time (minutes)
90 100 110 120 130
Figure 15. n-Octane with low expansion foams
(initial foam height was 2 in.)
47
-------
100,000
e
O4
c
o
•H
e
-------
100,000—.
10,000 —
6
ex
ex
o
•H
4.)
nJ
S-i
4J
a>
u
e
o
a
1,000
100
Lower Explosive Limit
1. Free surface
2. 5:1 Foam 1" Thick Type "A"
3. 5:1 Foam 1" Thick Type "B"
4. 5:1 Foam 2" Thick Type "A"
0 20 40 60
80 100 120 140
Time (minutes)
Figure 17. Effect of doubling foam "A" thickness for benzene
spill of figure 16
49
-------
100,000 —i
10,000 —
S
o
a
o
CJ
1,000 _
100
- Lower Explosive
Limit
Legend;
1. Free surface
2. 200:1 Foam 10" Thick Type "X"
3. 200:1 Foam 10" Thick Type "Y"
—T
20
— i - 1 - 1 - 1 - 1 - 1
40 60 80
Time (minutes)
1
1
1 T
100 120 140
Figure 18. Vapor buildup of vapor concentration over benzene surfaces
comparing two high expansion foams
50
-------
Medium Expansion: MSA Ultrafoam
National AFFF 3%
Emulsiflame
Protein: National 6% Regular
Fluoroprotein: National 3%
Alcohols: National Aerofoam 99-6%
Rockwood Alcohol 6%
Low Expansion
Surfactants: MSA Ultrafoam
Low Expansion AFFF: National 3%
For liquefied ethylene the heat input by water is greater than the heat
of vaporization. As soon as any foam system was applied to liquefied ethyl—
ene the boil off rate increased. Since the specific gravity of liquid
ethylene is 0.569, any water draining from the foam became frozen and sank,
exaggerating the boil off. Low expansion foams increased the vaporization
rate more than the medium expansion foam because of their higher water con-
tent. Protein foams which were applied as a 5 cm cover also became frozen
at the foam-liquid interface. Channels formed through the frozen foam
layer, permitting the liquefied ethylene to evaporate. The remaining un-
frozen foam layer frothed 2 to 3 times its original 5 cm thickness, perman-
ent channels through which ethylene vapors continued to evaporate were form-
ed even through the froth.
The low expansion surfactant and AFFF also frothed after application to
4 to 5 times their original thickness. Since the bubbles are filled with
ethylene, the frothing is probably not beneficial. Medium expansion foams
also exaggerated boil off but not as much as the low expansion foams. After
application of a medium expansion foam cover, large bubbles of ethylene
passed through the foam column due to an increased boiling rate. Wet medium
expansion foams resealed themselves after the large bubbles passed through
the foam column. As the foam drained and dried out the ethylene bubbles
formed permanent channels as the walls froze.
Application of foams to liquefied propane produced about the same re-
sults as with ethylene, except the boil off was not as intense.
Liquefied 1,3-butadiene showed results similar to ethylene. The boil-
off intensity due to foam application was less for the medium expansion foam
than for low expansion foam.
Ethylene oxide, because of its higher boiling point, did not cause the
water in the foams to freeze but the heat of solution was sufficient to
exaggerate boil off upon foam application. Protein foams formed crevices in
a short period of time due to foam degradation, exposing the liquid ethylene
oxide-water solution. Of the protein based foam agents, only Rockwood alco-
hol foam formed a stable cover. The low expansion surfactant and AFFF foams
formed stable covers after some initial frothing. Medium expansion foams
all collapsed when generated onto the ethylene oxide surface.
Liquid ammonia has a high heat of solution, which along with the other
heat inputs produced intense boiling upon foam application. Protein based
51
-------
foams exaggerated the boil off more than the other types of foams. Rockwood
alcohol foam was the only protein based foam that did not form a stable cover.
It collapsed in several minutes. Low expansion surfactant and AFFF foams
frothed to 4 times their original thickness but were stable. Medium expan-
sion, in relation to the other foams, showed the smallest increase in the boil
off rate. The medium expansion foams also frothed to 1.5 to 2 times their
original height of 30 cm, but the collapse rates were normal.
Chlorine has a low heat of vaporization with limited water solubility.
Even so, the heat input of water caused boil off to be intensified. Medium
expansion foam collapsed in 5 minutes leaving only the frozen interface com-
posed of ice and chlorine hydrate over the liquid surface. Protein foams
thickened upon application. Chlorine bubbling through the foam produced a
very porous cover. Alcohol foams were quickly destroyed by liquefied
chlorine. The low expansion surfactant and AFFF foams frothed 7 to 8 times
their original thickness and formed a relatively stable foam. The bubbles
were filled with chlorine, however, and foam collapse released chlorine.
During all of the chlorine tests a green vapor cloud was clearly visible
over the foam cover. Water has a lower density with respect to liquefied
chlorine water. Drainage from the foam forms an ice cover over the liquefied
chlorine, but the cover at the foam-liquid chlorine interface is porous and
allow significant quantities of chlorine vapors to be released.
Hydrogen fluoride, because of its extremely high heat of solution, boiled
and fumed violently upon application of foam. A medium expansion foam cover
could not be developed until the hydrogen fluoride was somewhat diluted. Low
expansion protein, surfactant and AFFF foams were stable over liquid HF but
fuming continued above the foam cover. Alcohol foams were destroyed. Hydro-
gen fluoride vapor, because of its low specific gravity, may dissipate fairly
readily without foam cover. Foam does not appear to help.
Sulfur trioxide produced more violent reaction than hydrogen fluoride
when foam was applied. Medium expansion foams collapsed, with splattering and
the release of black fumes. After dilution by continued application, medium
expansion foams formed a short lived cover above the sulfur trioxide. All the
low expansion foams also produced black fumes, intense splattering and boiling
upon application but they eventually formed a relatively stable foam cover.
Matrix Revision
Upon completion of the laboratory test program, the data obtained was
compiled and evaluated. Using this data, the matrix prepared in the beginning
of this task was revised and annotated. The matrix is shown in Table 9.
TASK II - FIELD TESTING
The laboratory tests provided a basic definition of capabilities of the
currently available foam systems to control spills of volatile hazardous
chemicals.
52
-------
TABLE 9. PRELIMINARY MATRIX OF FOAM CAPABILITIES
ON THE SPILLED HAZARDOUS CHEMICALS LISTED
Organics-AIIphatic
Acids
Alcohols
Aldehydes &
Ketones
Esters
Halogenated
Hydrocarbons
Acetic Acid
Caproic Acid
Amy I Alcohol
Butanol
Butyl Cellosolve
Methanol
Octanol
Propanol
Acetone
Methyl Butyl Ketone
Methyl Ethyl Ketone
Butyl Acetate
Ethyl Acetate
Methyl Aery Iate
Methyl Methacrylate
Propyl Acetate
• Butyl Bromide
Methyl Bromide
TetrachIoroethane
Heptane
Hexane
Octane
Nitrogen Bearing - Dimethyl Formamide
Organ ics-Aromatic
Hydrocarbons
Organ ics-AIicyclIcs
Organics-Industrie I
Organics -Cryogens
Inorganics
- Benzene
- Tetrahydronaphthalene
- To Iuene
- Cyclohexane
- Gasoltne
- Kerosene
- Naptha
- Paint Thinner
- Liquefied Natural Gas
- Silicon TetrachI oride
- Sulfur Trioxide
Inorganics-Cryogens - Ammonia
- Chlorine
Recommend 1
ND
ND
ND
R
ND
R
R
R
R
R
R
ND
ND
ND
ND
ND
ND
ND
ND
R
R
R
ND
R
R
R
R
R
R
R
R
R
R
R
R
R
ND
U
U
E-
ND
E-
U
E-
E-
E-
U
U
U
U
U
U
U
U
U
C+
C+
C+
U
C+
U
C+
B+
C+
C+
C+
C+
C-
E-
E-
C+
C+
ND
ND
U
E-
ND
E-
U
E-
E-
E-
U
U
U
U
U
U
U
U
U
8+
B+
B+
E-
B+
U
B+
A+
B+
B+
B+
B+
A+
A+
A+
A+
C+
ND
ND
ND
E-
U
E-
U
E-
E-
U
U
U
U
ND
ND
U
ND
ND
ND
B+
B+
B+
ND
B+
U
B+
B+
B+
B+
B+
B+
E-
£• *m
E-
C+
C+
ND
ND
ND
E-
ND
E-
U
E-
E-
ND
ND
U
U
U
U
U
ND
ND
ND
B+
B+
B+
ND
B+
ND
B+
B+
B+
B+
B+
B+
E-
E-
E-
C+
C+
ND
ND
ND
A+
ND
A+
U
A+
A+
A+
A+
ND
ND
ND
ND
ND
ND
ND
ND
A+
A+
A+
ND
A+
ND
A+
C+
A+
A+
A+
A+
E-
E-
E-
E-
E-
IDENTIFCATION OF SYMBOLS
Type of Foam:
1
Surfactant Low Ex Surfactant High Ex
3
Prote i n
Fluoroprotein AI
U Limited data available - capabilities uncertain
ND No data
R Foam use recommended over spiI I
A+ Best foam formulation
B+ Next best foam formulation
C+ Acceptable in some situations
E- Unsuitable foam formulation
53
-------
Based upon the results from the laboratory tests a selection was made of
candidate materials for field testing. The selection excluded any chemical
which has a high toxicity (low TLV). Thus, for instance, HF and 863 were not
considered. Such exclusion did not hurt the verification of the laboratory
data. Data from other materials were combined and translated to approximate
the conditions these materials would present.
The foam agents chosen for a majority of the field tests were those from
each foam class that had the best drainage characteristics, i.e., retained
their water content. Several foam agents with poor drainage were field tested
to substantiate the laboratory results.
Three separate field test sites were used. The first tests were conduct-
ed at the Evans City plant of MSAR. These tests were for the hazardous
liquids such as benzene. The second site was a remote location near Norman,
Oklahoma, where the liquefied gases such as propane were tested. To test
chlorine and ethylene oxide, a very remote area with security was needed.
Thus, the third test site was at the DOE facility at Mercury, Nevada.
Evans City Field Tests
The results of the field tests with volatile organics are shown in Table
10. These field tests were conducted at Evans City where 220 liters of
hazardous chemical were placed into a 3 m by 3 m pan. Gas sensors were placed
at 5 locations just above the foam cover to determine the length of time a
particular foam reduced the vapors from the pan. For most of the low expan-
sion foam tests, 75 liters of water were used to produce foam for each test.
The foam thickness varied from 5.7 cm for the protein based foams to 10 cm for
the surfactant foams. The Rockwood Alcohol foam, because of its thicker
nature, used 115 liters of water to completely cover the solvent in the pan.
The high expansion foams used less than 2 liters of water to provide a foam
cover 30 to 45 cm in depth. The values shown in the table indicate the length
of time for a significant vapor breakthrough (normally 1% vapor concentra-
tion). In most cases, once vapor breakthrough began the vapor concentration
rose very rapidly above the foam cover.
The results in Table 10 are generally similar to the earlier laboratory
tests if the foam agents are compared on a relative basis. The notable
difference between the field and laboratory test results was the stability of
the foams against polar compounds. In the laboratory the liquid volume ratio
of foam solution to solvent was 4 to 1. The large amount of water from the
foam solution diluted the polar solvents. In the field tests the ratio of
foam solution to solvent was 1 to 2.75. The effect of water dilution was
greatly reduced. No foam agent except Rockwood alcohol was resistant to col-
lapse with the polar compounds which have a high ambient vapor pressure, ace-
tone and ethylether. N-Butyl acetate (with a low vapor pressure) caused all
the low expansion foams except fluoroprotein, alcohol, surfactant and fluoro-
surfactant foams to collapse.
Triethylamine is not included in the table because of large variations
which occur in solubility around 20°C. This property of triethylamine results
54
-------
rC
O
S3
O
OJ
1
^4
S
«
ftj
O
O-i
^
H
2
d
M
b
M
e
M
W
Oi
o
b
fVl
§
r-l
H
1
^
C/3
fVI
§
M
H
2
O
r-l
^
O
1 — 1
>.
4-1
3
«
i
c
i— i
>-,
J2
4-1
U
Ol
c
• H
1— 1
>,
x:
4-1
01
• rl
l-l
H
1 1 1 1
1 1 1 1
01
C
cd
4-1
y
O
1
C
Oil)
vO 1 1 I
TJ
0)
CO
01
4-1
cd
4J
0)
3
a
cd i i o
r-l 1 1 VO
v— 1
O
o
•o
01
01
c
o
4-1
01
y
<3
ca
a
CO
rH
r-i r. r r
o
o
TJ
M
01
J=
4-1
U
01
ca
cx
co = r r
r-i
r-l
O
01
c
01
3
i— I
O
H
CJ
0101
vO 1 vo 1
0)
C
01
N
C
0)
rH rH rH
O O O O
vo en en en
1
1
O
vO
0
V0
O
vO
O
CM
O
vO
O
vO
1 1 1
1 1 1
I 1 1
I m i
1 CM 1
T)
(U
co
CX
CO
rH 1
r-l S 1
o
o
T)
OI
CX
cd r. -
r- 1
i— i
o
o
I m i
I m i
—H
m m m
rH r- 1
|
1
1
1
1
TJ
a>
CO
CX
cd
1—4
r«H
O
u
~
i_H
o
i— «
m
en
i I I
I I I
O 1 1
vO 1 1
O 1 1
vO 1 1
TJ
01
ca
CX
cd
1— 1
r r r-i
8
TJ
01
CM CM CO
m vo cx
— H rH Cd
1— 1
1— 1
o
o
O 1 1
vO 1 1
en vo CM
rH rH
in
TJ
0)
ca
CX
cd
o
u
C
01
I—I
o
ca
m
m
CM
en
CM
vO
ai
oo o>
oi g 4-1
4-1 Cd
3 TJ C
n oi
•rl TJ OI
e c *
•rl 4J
n s oi
C C TJ
oi cd 01
> u E
•H >H M
00 >W O
•H M-l
§g
00 CX
• H .H CO
O
b
01
c
CO
X
0)
01 I O O I O
vO I I vO vO I vO
O I I
vO I I
O
en
o
vO
PQ
<
H
y
4-1
C!
01
00
o
b
1 6% Regular Protein
6% Regular Protein
CO
C
O
•rl
4J
CO
&
c
0
y
i_i
3
3% Fluoroprotein
e
o
y
i-i
o
1 XL-3 Fluoroprotein
CO
C
o
4-1
CO
vO
CT> b
C7\ b
b
S < J=
CO S S 00
OS O O B^S -i-l
i— i b O p-3 i-4 bvoPS
O 1 >-3 b
JC O X X fa M X
0 1 S 1 U < -5 Oi— i oenS1-)
CO -rl -1-13 -rl
TJ i— 1 IH COTJ COb CO i— 1 4J T>
Ocd4JCOC CCO.CO
OnrHCOOcOCedCOOO
SODCXSCXOCXO-HS
Jd -H X^XyX-HnJ^
OJ-J
-------
in significantly different responses depending upon the ambient temperature,
%,»,. making a generalization difficult without further investigation.
N-Octane did not cause any of the foam agents to collapse.
For the nonpolar compounds shown in Table 10, all of the low expansion
foam agents with low water drainage rates form an effective mitigation barrier
when using the lower explosive limit as the point of measure. Foam agents
with high water drainage rates showed poor results. The loss of water results
in a light, easily permeated foam layer which quickly loses its ability to
mitigate the vapor. In general, the protein based foam agents provide a long-
er mitigation time than the surfactant and fluorosurfactant based foam
agents.
High expansion foams provide less protection time than low expansion
foams, due primarily to wind effects. However, the high expansion foam can be
applied much more rapidly and with about one-sixtieth of the liquid foam so-
lution volume. Repeated applications of high expansion foams could provide
the same mitigation time as low expansion foams with much less water.
Oklahoma Field Tests
Field testing of liquefied ethylene, propane, butadiene and ammonia was
conducted at the test site of Energy Analysts in Norman, Oklahoma. The actual
testing sequence is shown in Table 11.
In each test, a 3 m x 3 m concrete pit was filled with liquefied gas to a
depth of 7.5 to 10 cm. Gas sensors were located downwind of the pit to
measure the vapor concentration before and after foam application. Time was
allowed for the liquefied gas placed into the pit to achieve steady state boil
off. Foam was applied until the surface was covered. The foam depth was 5-8
cm for low expansiion foam and 15 to 30 cm for medium and high expansion
foam.
Additional tests were run with propane in the 3 m x 3 m pan to compare a
medium vs low expansion surfactant (MSA Ultrafoam) and two regular protein
foam agents (Rockwood and National). Because the odor was causing some con-
cern to earby inhabitants the final test with ammonia using the 6% Light Water
AFFF was done in a smaller 1.5m x 1.5m pan.
The following data points were recorded on magnetic tape for computer
analysis:
wind speed - 2 locations
wind direction - 2 locations
wet bulb temperature - 2 locations
dry bulb temperature - 2 locations
relative humidity - 2 locations
incident solar radiation - 1 location
atmospheric pressure - 1 location
downwind combustible vapor concentration - 10 locations
v, downwind ammonia vapor concentration - 7 locations.
56
-------
C
0)
4P
§
• 1-1
CO
C
CO
a
x
0)
• H fc
^3 |TI
-t O
Cfl 14-1
C cfl
O W
60
z10
en B~S B-S cfl
S vD vO S
o
• H
CO
c
cfl
a.
X
01
o
I—I I—I
•^ o
§ °
CO O
O ^H
y-i <;
co
M T3
4J O
i-l O
P »
l-i C
CO -H
i-< 0)
3 4J
60 O
0) t-i
C l-i
•H CO
0) i—<
4J 3
O 60
U 0)
SS
^ -a
CO O
c o
•21
4J O
CO o
Z ai
c
o
•1-1
ca
C
cfl
a
g
S
3
fti *O
60 0)
01 4-1
g os o
C O. b
O &4
_ oi i-i b
B 4-1 O <;
-^03
l-i 1-1 |j
g PL, b 0)
to 4-1
O ,-1 r-< cfl
M-l efl CO &
CO C C
)-i O O 4-1
4J -r^ -H J2
i—I 4-1 4J 60
•-> CO co .H
Z Z hJ
•O -H
C
o
• 1-1
co
C )-i
CO CO
Cu,-<
X 3
0) 60
•H C
T3 -H
Ol 0)
S 4J
v^ O
l-i
B-S fr« CO
vO s£3 S
ss
£
0)
4-1
cfl
vO
O r-l
IM CO S
Cfl (-
U O 4-1
4-1 •!-! J3
i-4 4-1 60
P CO -H
B^S S-S
vO vO
VH
01
4-1
01
e
3
cr
C
01
w
0)
c
co :
ex
o
CO
4-1
I
CO
57
-------
The data showed that the benefit of foam in mitigating the vapor hazard
in terms of downwind gas concentration is a function of the gas density. The
initial application of foam resulted in a sharp increase in the boil off rate.
The magnitude of the increase varied inversely with the expansion of the foam
and directly as a function of drainage — the lower the expansion and/or the
faster the drainage, the larger the magnitude of boil off pulse.
After the initial pulse, the boil off rate decreased as did the downwind
gas concentrations, but not always below the values seen before foam applica-
tion. For a spill of liquefied ethylene, where the vapor density at ambient
conditions is equal to air, medium and high expansion surfactant foams afford-
ed reductions in the downwind vapor concentration. This reduction was in the
range of 10%. With low expansion foams, protein-based or AFFF, the downwind
vapor concentration persisted at a level greater than that from a free spill.
It was noted that with ethylene, even under the best conditions the downwind
gas concentration remained above the LEL for a considerable distance from the
spill center.
The results for propane, where the vapor density is greater than air,
followed closely those for ethylene, except the initial vapor cloud given off
during foam application is smaller than in the case of ethylene. The propane
vapor cloud tended to remain close to the ground. A medium surfactant and low
expansion AFFF, protein, alcohol, and surfactant foams were tested. High
expansion foam was not tested with propane. The medium expansion foam did not
cause as large a vapor cloud as compared to low expansion when applied to a
liquefied propane spill. No distinct change in the downwind concentration was
distinguished with any of the foams after passage of the initial pulse.
Butadiene, whose vapor density is greater than propane, showed no reduc-
tion in the downwind vapor concentration when a foam cover of medium expansion
surfactant, low expansion protein, AFFF or fluoroprotein was applied. This
cloud remained closer to the ground than propane and appeared to take longer
to dissipate.
In summary, for the gases ethylene, propane and butadiene the test
results showed the influence of foam on downwind gas concentrations to be
essentially controlled by gas density. The existing data on liquefield
natural gas agree ith this observation. The foam blanket acts as a heat sink
for the liberated gas, warming it as it rises through the blanket. With LNG,
which is considerably lighter than air at ambient temperatures, the effect of
foam is pronounced. With ethylene, whose density is less than air at ambient
temperatures, there is a small positive effect. Propane and butadiene, which
are heavier than air, show no benefit from the warming action of the foam.
The exaggeration of the boil off rate by the water draining from the foam may
actually elevate downwind concentrations above that existing with a free
spill. This behavior should carry over to other similar liquefied gases.
Changes in downwind gas concentrations are not the only consideration in
assessing the ability of foam to control the vapor hazard of volatile chemi-
cals. Benefits in controlling fires are also possible. Fire control was by
definition beyond the scope of this contract. However, the most practical
58
-------
method of dissipating residual spill material after completion of the vapor
dispersion measurements was to burn it off. Observations made during these
actions showed that in all cases the burn back (rate of fire spread) across
the spill surface was considerably slower with a foam blanket in place than
for a free liquid surface (Brown, 1979).
Ammonia showed a significant reduction in the downwind vapor concentra-
tion with the use of medium expansion surfactant foam, with the foam layer
lasting for approximately 10 minutes. Until the foam began to break up, the
vapor concentration remained below the vapor concentration before the applica-
tion of the foam. The low expansion foams caused large vapor clouds to be
given off upon application and no significant reduction occurred in the down-
wind vapor concentration after the vapor initial pulse dissipated. The pro-
tein foam formed channels through which ammonia continued to vaporize. The
AFFF produced a great deal of frothing and bubbling. The success of the
medium expansion foam in mitigating the ammonia vapors was probably due to the
low water content and foam depth which facilitated scrubbing of the water
reactive ammonia vapors. The only problem with the use of a medium expansion
surfactant foam on an ammonia spill was the nedd for periodic make-up.
It is interesting to note that little or no ice was formed when the foam
was applied to the ammonia spill. This is due, apparently, to a combination
of solubility and heat of solution. In the cases of ethylene, propane and
butadiene, a porous ice layer did form on the surface of the liquefied gas.
The results for ammonia are shown in Figure 19. The test conditions were
as follows:
Low expansion (8:1) protein foam:
Pool size 3 m x 3 m
Foam depth 10 cm
Sensor #1 1.16 m from pool edge,
0.62 m elevation
Sensor #2 3.87 m from pool edge,
0.62 m elevation
Low expansion (8:1) AFFF:
Pool size 1.5 m x 1.5 m
Foam depth 10 cm
Sensor #1 0.7 m from pool edge,
0.62 m elevation
Sensor #2 2.94 m from pool edge,
0.64 m elevation
Medium expansion (250:1) surfactant foam:
Pool size 3 m x 3 m
Foam depth 30 cm
Sensor #1 1.24 m from pool edge,
1 0.62 m elevation
Sensor #2 3.1 m from pool edge,
0.62 m elevation
59
-------
Data averaged for wind direction
but not for velocity or temperature
C
O
•H
•u
c
0)
u
c
O
u
f-l
O
a.
6 -
5 -
4 -
3
2 -
1 -
0
8
-Foam on
9 sees.
10
12
14 16 18
AFFF
2.25 ft
9.5 ft
0
68
Foam applied
30 sees.
10
12
14
16
18
Water
5 '
4 '
3 •
2.
1 •
n .
Medium ex-
pansion
4
15
f t
ft
ft \
\\ ,
H /
4
"~i/:
\r
I
re
ft
8 10 12
Time (minutes)
14
16
18
Figure 19. Effect of foam additions to anhydrous ammonia spills on
downwind vapor concentrations
60
-------
Protein foam showed no benefit and aggravates the situation upon initial
application. Chimneys or voids formed in the blanket through which the gas
was continuously released.
Due to the boiling action, no aqueous film was formed with the AFFF and
the foam layer was strongly agitated by the gas release.
At the end of the medium expansion foam test, water was added to the
spill. The chart readings show the dramatic increase in the downwind ammonia
vapor concentration which occurred. The effects of water dilution alone
should reduce the downwind vapor concentration. However, the heat of solution
results in a higher vaporization rate which offsets the dilution effects.
High expansion foam behaved similarly to medium expansion except that it de-
grades faster and showed somewhat less efficiency in controlling vapor re-
lease.
The reduction in the downwind vapor concentration after the application
of a medium expansion surfactant foam is evident. In the test using the AFFF,
the wind was shifting direction rapidly and large changes occurred in the
sensor output. However, it is still evident that little reduction in the
downwind vapor concentration was found.
Nevada Field Tests
At the Nevada test site, foam tests with liquefied chlorine were intially
conducted using a weighed, insulated 0.93 m by 1.24 m pan to determine the
chlorine weight loss before and after foam application. After the pan was
filled to approximately 7.5 cm in depth with liquefied chlorine, the foam was
applied. The amount of foam agent and water was also measured to establish a
chlorine weight loss during and shortly after foam applicatin. In the past,
chlorine weight loss data was only noted before and after foam application.
The weight of foam agent and water applied to the pan was not measured, and no
weight loss could be determined as the foam was applied. It was certain that
some chlorine was vaporized due to freezing of the foam.
Chlorine detectors were used to measure the downwind vapor concentration.
However, the wind direction during the tests varied so widely that the vapor
concentration data was not meaningful.
The results from the first foam tests are shown in Figure 20. High ex-
pansion surfactant foam (300:1) was used. Several minutes after application
the high expansion foam collapsed. Note the increase in the chlorine vapori-
zation rate after the first foam application. This increase is due to the
heat of fusion of the water which drains from the collapsing foam. The water
freezes in contact with the liquefied chlorine (-33°C) to form a porous ice
layer which remain after the foam collapses. A second foam application did
not produce the large increase in vaporization. However, this foam cover
lasted only slightly longer than the initial foam cover.
In Figure 21 the results of a fluoroprotein foam application to liquefied
chlorine are shown. The vaporization rate increased rapidly as the foam was
applied and for sometime thereafter. There was some problem in determining
61
-------
'O
co
.jn
G
•H
•m
vO
o ••:
in
CO
m
o
•ro
m
'CM
o
CM
-O
m
O
01
•H
14-1
01
cr
c
o
•H
01
ni
o
CM
60
O
00
o o
(tmn/qT)
o
•
CM
62
-------
oo
o
oo
.. m
-- o
. .m
o
vO
..o
. .in
J
c
)
)
lilt
-O
-m
o
O O O O O
oo %o
-------
the volume of water added, so the possible lower and upper weight losses are
shown. The actual weight loss was most likely near the upper limit because
large chlorine vapor clouds were generated during foam application. The fluo-
roprotein foam thickened several minutes after application and a stiff, porous
foam layer overlying an ice cover resulted. Beneath the ice layer there were
several inches of liquefied chlorine. The volume of water needed to apply the
low expansion fluoroprotein foam was much greater than the high expansion
foam. The large quantity of water draining from the fluoroprotein foam pro-
duced the resultant increase in the vaporization of liquefied chlorine.
In Figure 22 the results for an aqueous film forming foam are shown. The
gap between 30 and 50 minutes is due to difficultes with the foam generator.
The difficulty was overcome and no further problems were experienced during
the AFFF test. A increase in chlorine vaporization was also noted as the AFFF
was applied to the liquefied chlorine surface. The AFFF collapsed several
minutes after application and only a porous ice layer remained on the lique-
fied chlorine surface.
The problem with foam application to liquefied chlorine is primarily the
water draining from the foam. The water content of the low expansion AFFF and
fluouroprotein foams is approximately 1 liter of water per 8 liters of foam.
With quarter drainage times of 1 to 2 minutes, assuming no abnormal collapse,
the low expansion foams have lost 25% of their water content soon after appli-
cation. The high expansion foam contains 0.016 liters of water per 8 liters
of foam. Assuming no abnormal collapse and a quarter drainage time of 10
minutes, the high expansion foam would be a more desirable cover when water is
a problem.
Since the high expansion foam appeared to produce the smallest increase
in vaporization, it was then used in a larger test. A pan, 3 m by 3 m in
size, was filled to a depth of approximately 5 cm with liquefied chlorine.
The chlorine vaporization rate was calculated using depth measurements. A
high expansion foam cover was applied and the change in depth was noted. The
high expansion foams, as in the earlier case, collapsed shortly after applica-
tion with the resultant increase in vaporization. The results for the 3 m by
3 m pan are shown in Figure 23.
All of the foams tested against a liquefied chlorine spill react with
chlorine resulting in rapid degradation of the foam. The foam drainage,
coupled with the rapid collapse, resulted in exaggerated boil off of the
chlorine and a sharp increase in the downwind vapor concentration. Once an
ice layer was established, successive foam applications were not as severe in
exaggerating boil off.
This data does not show foam to provide any benefit in reducing the vapor
hazard from a chlorine spill. It reflects the capability of those foam
systems currently available to the fire service. The fact that successive
foam applications do not create a large gas release pulse would indicate that
foams which would be resistant to collapse upon contact with chlorine would be
successful in reducing the vapor release rate. Since foams are known which
are resistant to hydrochloric acid, development of a foam which is stable on
chlorine should be possible.
64
-------
oo
co
•1C
. .o
- o
. .m
in
-------
--0
.o
-d-
LO /~\
ro to
..o
..in
O E-i
CM
• -o
o
o
o
•
o
o
CN
o
o
a
•a
-------
The ethylene oxide test was done in a 0.9 m x 1.2 m pan. After a 5 cm
layer of liquefied ethylene oxide was established in the pan a foam cover was
applied. Combustion gas sensors were used to determine the vapor concentra-
tion before, during and after foam application. The sensors were located
over the downwind side of the pan, 30 cm from the ethylene oxide surface.
Since the polar ethylene oxide causes all but alcohol based foams to col-
lapse, only one mitigation test was conducted using the alcohol foam which
had shown the properties in the laboratory (Rockwood Alcohol). The results
of the alcohol based foam application are shown in Figure 24. Immediately
upon addition of the alcohol foam, the vapor concentration of ethylene oxide
increased rapidly to a level above the LEL. This result is probably due to
the displacement of ethylene oxide vapors above its liquid by the foam.
After the foam application was ended the vapor concentration decreased to
approximately 30-40% of its uncovered value. Some of the reduction in
ethylene oxide vapor concentration may be due to dilution by the water from
the foam. After approximately 15 minutes, the foam cover was broken beneath
one of the sensors to determine the vapor concentration of the uncovered
ethylene oxide-water mixture. The vapor concentration increased from 25% to
showing that foam was effective in mitigating the vapors.
Foam Matrix
From these field tests a matrix has been established for the application
of foam for use by first-on-the-scene personnel. This matrix is shown in
Table 12. It provides a guideline for use of foam on a hazardous spill. The
matrix shows that for polar compounds only a high quality alcohol foam is
recommended for vapor mitigation. Even though other classes of foam agents
may be stable against polar liquids with low ambient vapor pressures, it is
difficult in a general matrix to include such exceptions.
For nonpolar liquids, a high quality low or high expansion foam could be
used. Alcohol foams are not always the best foam agent to use if another
class of foam is suitable. Alcohol foams do not flow as well as other types
and therefore it would take longer to blanket a spill. If the addition of
large volumes of water will further compound the problems of the hazardous
spill, high expansion foam is recommmended.
Liquefied gases were not included in the matrix because of the poten-
tially harmful results if foam is used incorrectly. Given the very specific
behavior of foams with liquefied gases, its use is recommended at this time
only for gases whose vapor density at ambient temperatures is less than air
at the same temperature.
TASK III - NOVEL FOAM SYSTEM
At the conclusion of Tasks I and II, additional foam agents were examin-
ed for their mitigation potential. These foams are available commercially
but not in universal use by the five services.
67
-------
CU CO
o
co
e
0)
en
cu
J3
60
CO
a
0)
co
ca
a)
a)
a)
t>o
co
a)
co
•u
co
n)
cu
co
0)
4-1
c
01
a
- -O
co
CU
o
u
bO
C
•H
M
T)
•H (3
•U 03
nj P.
M
4J 4-1
C
CU
T3
•H
CU
g
r-l
w
•
csi
cu
oo
•H
Pn
M-l
o
•
-------
en
Z
O
a
&
oc
4-1
II
4J
a
cd
cd cd
cd
w w w w w w
§
Q
33
Q
M
O*
M
i-J
Pi
o
PH
§
z
CM
W
PH
3
|
cj
cd
3
c/>
O
cj
•S
cu
4J
O
p.
o
o
3
•H
CU
4-1
O
t-l
p-l
W w w W w W
o
v*X
o
cj cj u u
H
w
CJ CJ CJ CJ CJ CJ CJ
CJ CJ CJ CJ CJ CJ CJ
a
w
w w
co
cu
4J
co
CO
rH
O
43
0
O
rH
CO
>, cu
43 S3
CU O
TJ 4J
rH CU
CO
r-|
O
a
cu
43
PJ
CU
4J CO
cd T^
c a
CU 3
00 0
cd o
33 CJ
co
cu
a
5
H
<4
CO
cu
a
rH
<3
CO
cu
£
rH
<4
co
CU
13
CU
•H
Q
iphatic
rH
cd
O
a
r*1
CJ
o
•H
4J
cd
o
^
<1
mph; also minimizes amount of water
u-i
H
V
13
CU
CU
CX
co
13
C
•H
4-4
M
"cd
CO
cd
rH
rH
•H
CX
CO
I-l
cu
o
S
o
rH
<4-l
4-1
O
CO
cu
o
13
CU
co
cd
o
cu
43
>,
M
O
4J
O
cd
m
CO
•H
4J
cd
CO
>>
rH
g
2*
s
0
<4-l
U-l
O
CO
CU
a
>*
4-1
M
cu
43
4->
O
CO
cd
rH
rH
CU
mitigation times are not as good as
13
c
cd
cu
00
cd
C
•H
cd
J_(
T3
B
cd
o
o
CO
cd
o
tM
T3
CU
CO
cd
43
c
•H
cu
4-1
O
V-i
CX
T3
CU
13
a
CU
S
o
o
cu
Pi
i
<3
^
^4
o
4J
a
cd
MH
CO
•H
4J
cd
w
1
U
13
CU
T3
(2
cu
g
o
o
cu
Pi
4-1
O
1
w
69
-------
1. 3M Light Water Alcohol
2. National Universal Foam Liquid
3. National PSL Foam Liquid
4. MSA Long Life Foam
These foam agents were tested in the laboratory. The laboratory tests
were similar to those used in Task I. The exception was the amount of vola-
tile liquid used in the test and only low expansions were evaluated. A com-
parison was made with Rockwood All-Purose Alcohol foam since it had shown the
best mitigation potential for both polar and nonpolar compounds.
The volatile chemicals used to test the novel foam systems as well as the
test results are, shown in Table 13.
TABLE 13 - TEST RESULTS FOR NOVEL FOAM AGENTS
(minutes to reach 50% of the LEL)
Rockwood 3M National MSA
Benzene 90 90 90 90
Cyclohexane 90 90 90 90
Acetone 90 30 80 Collapse
Ethyl Ether 90 16 36 Collapse
The four chemicals are representative of the materials which during Task
I seemed to provide a reliable indication of the foam behavior on a variety of
chemicals.
The results show that for nonpolar compounds the novel foam agents per-
form as well as Rockwood Alcohol. However other conventional protein and
fluoroprotein foam agents from Task I also performed well against nonpolar
compounds. The real difference in performance between the novel foam agents
and Rockwood is seen in the polar compounds. None of the novel foam agents
were as effective against acetone nor ethyl ether as the Rockwood Alcohol
foam. The 3M and National foam agents were stable against polar compounds,
but the MSA Long Life Foam collapsed. The results show that these foam agents
are not as effective as Rockwood All-Purpose foam against polar compounds.
Nor do they provide any improvement over the widely distributed protein and
fluoroprotein foam agents against nonpolar compounds.
70'
-------
REFERENCES
Archer, R.J. and LaMer, V.K., "The Rate of Evaporation of Water Through Fatty
Acid Monolayers", J. Phys. Chem. 59, 200 (1955).
Brown, L.E., and Romine, L.M., "Fighting Liquefied Gas Fires with Foam",
Hydrocarbon Processing, September 1979.
Buschman, C.H., "Experiments on the Dispersion of Heavy Gases and Abatement of
Chlorine Clouds", Fourth International Sysmposium on Transport of
Hazardous Cargoes By Sea and Inland Waterways, National Academy of
Sciences, 1975.
Friel, J.V., et al, "Control of Hazardous Chemical Spills by Physical
Barriers", PB-221 493, U.S. Dept. of Commerce, March 1973.
Greer, John S., "Feasibility Study of Response Techniques for Discharges of
Hazardous Chemicals that Float on Water", Final Report for U.S. Dept. of
Transporation, U.S. Coast Guard, October 1976 (Contract DOT-CG-51870-A).
Hampson, Thomas R., "Chemical Leak at a Bulk Terminals Tank Farm", Proceedings
of 1976 National Conference on Control of Hazardous Material Spills.
Hederstand, G., The Influence of Thim Surface Films on the Evaporation of
Water", J. Phys. Chem. 28, 1245 (1924).
Kroop, R.K., and Martin, J.E., "Treatability of Aqueous Film-forming Foams
Used for Fire Fighting", Report No. AFWL-TR-73-279 (1974).
Mecca, J.E., et al, "The Development of a Long-Lived, High-Expansion Foam for
Entrapping Air-Bearing Noble Gas", Final Report FY-70, DUN-7221, February
1971.
Meldrum, D.N., "Aqueous Film-forming Foams: Facts and Fallacies", Fire Jour.
^6_(1), 57 (1972).
Pfann, J.R., and C.A. Sumner, "Sulfur Tioxide Spill Control", Proceedings of
1976 National Conference on Control of Hazardous Materials Spills.
Tuve, R. L., Peterson, H.B., Jablonski, E.J., and Neill, R.R.,"A New Vapor-
securing Agent for Flammable Liquid Fire Extinguishment, NRL Report No.
6057 (1964).
White, I., "Effect of Surfactants on the Evaporation of Water Close to 100°C",
Indust. Eng. Chem. Fundam. 15(1), 53 (1976).
71
-------
BIBLIOGRAPHY
Battelle Memorial Institute, Dawson, G.W., Control of Spillage of Hazardous
Polluting Substances, PB 197 596 (Nov. 1970).
Baumann, H., Baumann-Graf Apparatus for the "Determination of Foam Stability
and of Maximum Foam for Solutions of Foaming Materials", Chemilun-Ztg.
£0(13), 449 (1966), cf. C.A. 65:12883e.
Barthauer, G.L., High Expansion Foam Producing Materials, U.S. Patent No.
3,479,285 (Nov. 1969), cf. C.A. 72:23045w.
Benson, S.P., K. Morris and J.G. Corrie, "An Improved Method for Measuring the
Drainage Rate of Fire Fighting Foams", Fire Research Note, Number 972
(May 1973).
Bikerman, J.J., "Foam Fractionation and Drainage", Separation Science 7(6),
647 (1972).
Bikerman, J.J., Applied Physics and Engineering, No. 10; Foams
[Physiochemical Aspects], Springer (1973).
Borlon, 0. and E. Nagy, "Problems of Gas-Liquid Mass Transfer in Bubbles and
Dynamic Foam Columns", Intl. Cong. Chem. Eng. Chem. Equip. Des. Autom.
Proc. 5th, 1975, CHISA, Prague, Czeck, cf. C.A. 85:126540z (1976).
Buschmann, C.H., "Experiments on the Dispersion of Heavy Gases and Abatement
of Chlorine Clouds", 4th Symp on Transp. of Haz. Cargoes by Sea and
Island Waterways (Oct. 1975).
Chono, M., K. Toyomoto, S. Nagomi and N. Matsuo, Powdery Foam Useful in Oil
Adsorbent, Japanese Patent No. 60,223 (1974).
DiMaio, L.R., P.J. Chiesa and M.S. Ott, "Advances in the Protection of Polar-
type Flammable Liquid Hazards", Fire Technology 11(3), 164(1974).
Faintsimmer, R.Z., N.A. Bedritskii, V.G. Geshelin and R.N. Raskin, "Control of
Vaporization and Oxidation of Petroleum Hydrocarbons", USSR Patent No.
12,921 (1958), cf. C.A. 53:3679g.
Fittes, D.W. and D.D. Richardson, "The Influence of Albohol Concentration on
the Effectiveness of Protein Foams in Extinguishing Fires in Gasoline-
Alcohol Mixtures", Jour. Inst. Petrol. 49, 150 (1963).
72
-------
French, R.J. and P.L. Hinkley, "Resistance of Fire Fighting Foams to
Destruction by Petrol", Jour. Appl. Chem. (London) 4, 513 (1954).
_«w^_^_^^^^^^__^^_^«««_« v^_«^^^^^_^_^^^^aa<_«
General Aniline Film Corp., Oil-Absorbent Foamed Silicate For Oil Pollution
Control, U.S. Patent No. 3,728,208 (1973).
Haas, P.A. and H.F. Johnson, "A Model and Experimental Results for Drainage of
Solution Between Foam Bubbles", Ind. Eng. Chem. Fundam. 106 (2), 225
(1967).
Juchniewicz, R., "Influence of the Temperature on the Stability of Fire
Fighting Foams", Nafta (Poland) 13, 138 (1957) cf. C.A. 52:12272g.
Langmiur, I. and D.B. Langmiur, "The Effect of Monomolecular Films on the
Evaporation of Ether Solutions", Jour. Phys. Chem. 31 1719 (1927).
Mecca, J.E., H.F. Jensen and J.D. Ludwick, Nobel Gas Confinement Study. I.
Development of a Long-lived, High Expansion Foam for Entrapping Air-
bearing Noble Gases, USAEC, DUN-7221 (1971).
Meldrum, D.N., "Aqueous Film-forming Foams: Facts and fallacies", Fire
Journal 66(1), 57 (1972).
Moran, HJ.E., J.C. Burnett and J.T. Leonard, Suppression of Fuel Evaporation
by Aqueous Films of Fluorochemical Surfactant Solutions, NRL Report No.
7247 (April 1971).
Mutrikor, A. Ya, A.I. Kozlov and O.V. Maminov, "Mass Transfer in Foams", Tr.
Kazan. Khim-Tekhnol. Inst. 53, 82 (1974), C.A. 83:62571j.
National Institute for Occupational Safety and Health, The Toxic Substances
List, 1974 Edition, HEW Publication No. (NIOSH) 74-134 (June 1974).
Prigorodor, V.N., "Stability Criterion of Foams and Methods of Its
Determination", Kolloid. Zh. 33(3), 459 (1971).
Shih, F.S. and R. Lemlich, "Continuous Foam Drainage and Overflow", Indus.
Eng. Chem. Fundam. 10(2), 254 (1971).
Tevlarides, L.L., C.A. Coulaloglou, M.A. Zeitlin, G.E. Klinzing and B. Gal-or,
"Bubble and Drop Phenomena", Ind. & Eng. Chem. 62 (11), 6 (1970).
Thompson, W., "Measurement of Drainage Rates of High Expansion Foams", J_._
Appl. Chem. (london) 12, 12 (1962).
Tuve, R.L. and E.G. Jablonski, Compositions and Methods for Fire
Extinguishment and Prevention of Flammable Vapor Release, U.S. Patent No.
3,258,423 (June 1966).
73
-------
U.S. Coast Guard, Chemical Hazards Response Information Service, Hazardous
Chemical Pats, Report No. CG-446-2 (Jan. 1974).
Veatch, F., and E.G. Hughes, Control of the Evaporation of Crude Oil, U.S.
Patent Nos. 2,797,138 and 2,797.140 (1957), cf. C.A. 51:14250i~and
14251c.
Vijayendran, B.R., "Bursting of Soap Films. VI. Effects of Electrolytes and
Desorption on the Shape of Aureoles", J. Phys. Chem. 79(23), 2501
(1975).
Vismanadham, C.R., S. Singh and V. Ranganathan, "Laboratory Test Methods for
Foam-making Compounds Used in Fire Fighting", J. Sci. Inc. Research
(India), 19A, 498 (1960), cf. C.A. 55;7841i.
Welker, J.R., H.R. Wesson and L.E. Brown, "Use of Foam to Disperse LNG
Vapors", Hydrocarbon Processing 53(2), 119, (1974).
Zuiderweg, F.J. "The Effect of Surface Tension Pehenomena on Mass Transfer",
Ingeniew 73(2), Ch. 1 (1961).
74
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
EVALUATION OF FOAMS FOR MITIGATING AIR POLLUTION
FROM HAZARDOUS SPILLS
5. REPORT DATE
September 1980
6. PERFORMING ORGANIZATION CODE
AC-952
7. AUTHOR(S)
S.S. Gross; R.H. Hiltz
8. PERFORMING ORGANIZATION REPORT NO
MSAR 80-152
9. PERFORMING ORGANIZATION NAME AND ADDRESS
MSA Research Corporation
Division of Research and Development
Evans City, PA 16033
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-03-2478
12. SPONSORING AGENCY NAME AND ADDRESS . ,
Municipal Environmental Reserch Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. -
• f f so
14. SPONSORING AGENCY CODE
13. SUPPLEMENTARY NOTES
Project Officer - Dr. John E. Brugger
16. ABSTRACT
This program has been conducted to evaluate commercially available water base foams
for mitigating the vapors from hazardous chemical spills. Foam systems were eval-
uated in the laboratory to define those foam properties which are important in miti-
gating hazardous vapors. Larger scale tests were then conducted in a 3 m x 3 m pan.
Polar and nonpolar liquids and liquefied gases were used as test materials.
Protein, fluoroprotein, alcohol and aqueous film forming foams were tested at low
expansion ratios and surfactant foam agents at low, medium and high expansion ratios.
The chemicals tested were acetone, n-butyl acetate, di-ethyl ether, n-octane, tri-
ethylamine, benzene, toluene, ethyl benzene, cyclohexane, propane, ethylene, but-
adience, ammonia, chlorine, ethylene oxide, hydorgen fluoride and sulfur trioxide.
It can be concluded that all high quality foams are effective against nonpolar
liquids but only alcohol foams can be recommended for polar liquids. Generali-
zations with liquefied gases are difficult because of their unique characteristics.
This report was submitted in fulfillment of Contract 68-03-2478 by MSA Research
Corporation under the sponsorship of the U.S. Environmental Protection Agency.
This report covers the period from October 29, 1976 to July 31,1980.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
a. DISTRIBUTION STATEMENT
Release to public
19. SECURITY CLASS (This Re port I
Unclassified
21. NO. OF PAGES
84
20. SECURITY CLASS (This page/
Unclassified
22. PRICE
EPA Fofin 2220-1 (R«». 4-77) PREVIOUS EDITION is OBSOUSTE
74
------- |