PB82-105230
e«^tof^for\Handling Landbor.n|r- Spiles .of
'"
•• -ta-tt e 11 & Co 1 um bu a
Prepared for
" Lab
Sep 81
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
..
-------
TECHNICAL REPORT DATA
(Pteaseread1nz Qucnonson the reverie before comp1efln PR 2 1 0 5 2 3 0
t REPORT NO 2.
EPA—600/2-8l-_207 ORD_Report
3. RECIPIENrS ACCES$IOr.NO
4, TITLE NQ SUETITLE
Techniques for Handling Landborne Spills of
.
Volatile Hazardous Substances
5. REPORT DATE
September 1981
6. PERFORMING ORGANIZATION CODE
1 AUTNOR S)
0. Brown, R. Craig, M. Edwards, N. Henderson,
T.J. Thomas
8. PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Battelle Columbus Laboratories
Columbus, Ohio 43201
10. PROGRAM ELEMENT NO.
1BB61O, Task No. 52
11.coNrRAc /GRAN-rNo
68—02-1 323
12 SPONSORING AGENCY NAME AND AOORESS -
Municipal Environmental Research Laboratory - Cin.,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincfnn ti._Ohio__45268
13. TYPE OP REPORT AND PERIOD COVERED
Finpl Repnrt
I4. SP 0NS ORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer, John E. Brugger (FIS) 340—6632, (201) 321-6634
.,ntr arP
lb.
Response needs of teams charged with handling spills of hazardous volatile materials
on land are considered by Battelle-Columbus. Items of hardware which could be
adapted or developed to improve spill response capabilities are suggested.
The report examines the available technology (and the lack thereof) being employed
in current spill responses. An assessment of the phenomena that accompany spill
volatilization is provided to determine and justify physical/chemical mechanisms
that could potentially be used to control the hazards arising from volatility.
As a result, approximately 60 items of hardware, which either exist or could be
developed to improve hazardous volatile spill control responses, are discussed.
A set of spill scenarios is developed to compare the new suggested technology
items with current spill response procedures.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IOENTIFIERS/OPEN ENOED rERMS
C.
COSArI FIeLd/Group
Hazardous materials
Vapors
Volatility
Cryogenics
Control equipment
Air pollution control
Emergency response
21
138
20M
18. CISTRIBLITION STATEMENT
(I Tfl DI ID It’
I ELEI 1 SE iv rvuL
19 SECURITY CLASS (Th s eporrj
Unclassi fied
20. SECURITY CLASS (Thz p4 ej
UricJassified
22 ‘RICE
EPA P rm 2220-I ( -73I
-------
NOTICE
TEIS DOCUMENT HAS BEEN REPRODUCED
FROM TEE BEST COPY FURNISHED US BY
THE SPONSORING AGENCY. ALTHOUGH IT
IS RECOGNIZED THA T CERTAIN PORTIONS
ARE ILLEGIBLE, IT IS BEING RELEASED
IN TEE INTEREST OF MAKING AVAILABLE
AS MUCH INFORMATION AS POSSIBLE.
-------
DISCLAINER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publi-
cation. 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.
i -i
-------
FOREWORD
The U.S. Environmental Protection Agency was created because of
increasing public and governmental concern about the dangers of pollu-
tion to the health and welfare of the American people. Noxious air,
foul water, and spoiled land are tragic testimony to the deterioration
of our natural environment. The complexity of that environment and
the interplay between its components requires a concentrated and inte-
grated attack on the problem.
Research and development is that necessary first step in problem
solving, and involves defining the problem, measuring its impact, and
searching for solutions. The Municipal Environmental Research Labora-
tory develops new and improved technology and systems for the preven-
tion, treatment, and management of wastewater and solid and hazardous
waste pollutant discharges from municipal and community sources; for
the preservation and treatment of 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 is a vital communications link between the researcher and the user
community.
This report covers a state—of—the—art survey and extension of tech-
niques for the control of air pollutants from landborne spills of volatile
hazardous materials. Included in the report is a discussion of the poten-
tial for the use of cryogenic techniques for control of volatilization.
In addition, 60 hardware items for better treatment of volatile spills
that could readily be developed from current technology are presented.
Those groups interested in improving current response techniques with re-
spect to both environmental effects and spi 1 i response team safety will find
this report to be of value. Further information on the subject may be ob-
tained by contacting the Oil & Hazardous Materials Spills Branch, MERL—Ci,
U.S. EPA, Edison, New Jersey 08817.
Francis Mayo
Director
Municipal Environmental Research Laboratory
iii
-------
ABSTRACT
Response needs of teams charged with handling spills of hazardous volatile
materials on land are considered by Battelle—Columbus. Items of hardware that
could be adapted or developed to improve spill response capabilities are
suggested.
The report examines the available technology (and the Lack thereof) being
employed in current spill responses. An assessment of the phenomena that ac-
company spill volatilization is provided to determine and justify physical!
chemical mechanisms that could potentially be used to control the hazards aris-
ing from volatility. As a result, approximately 60 items of hardware, which
either exist or could be developed to improve hazardous volatile spill control
responses, are discussed.
A set of spill scenarios is developed to compare the new suggested tech-
nology items with current spill response procedures.
This report was submitted in fulfillment of Contract No. 68—02—1323, Task
52, by Battelle Columbus Laboratories under the sponsorship of the U.S. Envi-
ronmental Protection Agency. This report covers the period January 30, 1976,
to September 30, 1976, and work was completed as of November 1, 1977.
iv
-------
CONTENT S
Foreword . jjj
Abstract iv
Figures vi
Tables vii
Abbreviations and Symbols viii
Conversion Factors ix
1. Introduction 1
2. Conclusions 4
3. Recommendations 6
4. Review of the Causes of arid the Responses to
Hazardous Landborne Volatile Spills 7
5. Phenomena of Hazardous Materials Volatilization 14
6. Proposed Volatile Spill Suppression Concepts 45
7. Scenarios Utilizing New Concepts 82
References 90
V
-------
FIGURES
Number Page
1 Band of expected vapor pressures at dry ice temperatures
of spilled materials versus their boiling temperatures 28
2 Caterpillar D8K crawler tractor 50
3 Remote—controlled bobcat excavator 51
4 Foam dike deployment concept 52
5 Spring—loaded tarpaulin deployment concept . . . . 54
6 Upside—down parachute deployment concept . . 56
7 Commercial line—throwing gun 57
8 Levitating tarpaulin concept 59
9 Commercial hydro—muicher 60
10 Silt stabilization polymer applicator . . . . 61
11 Conceptual technique for forming and deploying
a cover over the spill 62
12 Deployment of a preformed film . . 62
13 Catapult delivery concept 63
14 Four—bar linkage concept 64
15 Commercial stack bale—thrower 66
16 Commercial wheel loader 67
17 Commercial forage blower 68
18 Commercial sand—blasting equipment 71
19 Standard highway centrifugal sander 71
20 Commercial hydro—seeder 72
21 Slurry delivery concept by means of a truck—mounted boom . . . . 73
22 Hot—air balloon delivery concept . . . 75
23 Commercial flame—thrower 77
24 Infrared heating concept . 78
25 Heat exchanger concept . . 79
26 Air curtain destructor 81
vi
-------
TABLES
Number Page
1 Rating of Efficiency of Current Response Techniques
to Control Evaporation S
2 Distribution of Reported Spills of Hazardous Materials . . . . 7
3 Frequency Association of Primary and Secondary
Causes of Hazardous Material Spills 8
Legend for Tables 4 Through 8 32
4 Representative Hazardous Highly Volatile Chemicals 33
5 Representative Hazardous Very Volatile Chemicals 35
6 Representative Hazardous Volatil Chemicals 37
7 Representative Hazardous Chemicals with Intermediate Volatility . . 40
8 Representative Hazardous Chemicals with Low Volatility 43
9 Conceptual Technologies 46
vii
-------
ABBREVIATIONS AND SYMBOLS
d Depth of spill 2
Diffusion coefficient 3 (ctn /sec)
DC Total dosage (g/sec/m )
E Heat capacity of soil (cal/gm—°K)
H Height of plume centerline when it becomes essentially vertical (cm, m)
AH Latent heat of vaporization (cal/mole)
J Flux (g/cm 2 —sec) 2
Heat flux (cal/cm —sec)
k Thermal conductivity (cal/cm—°K—sec)
L Length
P Pressure
Q Emission rate of vapor from a spill (g/sec)
Total vapor release (g)
Heat transfer rate from earth (e) or from the sun (s) (cal/sec)
R ‘ Gas constant
r Characteristic surface dimensions (cm)
S Solar heat flux (cal/cm 2 —sec)
T Temperature (°K)
Tb Boiling temperature ( K)
AT Temperature increment (°K)
t Time
vf Volume, (f) final volume, (i) Lütial volume
v ‘ Wind velocity
x,y,z Distances in a 3—dimensional system
ax,y Dispersion coefficients in x direction, y direction (m)
Dielectric constant
Mean wind speed in the x direction (m/sec)
p Density (glcm 3 ) 3
x Mass concentration (g/ctn )
BTIJ British thermal unit
cfm Cubic feet per minute
tip Horsepower
LN 2 Liquid nitrogen
SAE Society of Automotive Engineers
viii
-------
CONVERSION FACTORS
To convert from to multiply by
British thermal unit (BTTJ) joule 1.05435 x 2 10 3
calorie (cal) 2.52 x 10
atmospheres (atm) pascal (Pa) 1.01325 x
calorie (cal) joule 4.184
foot (ft) meter (in) 3.048 x 10_i
foot 2 (ft 2 ) meter 2 (in 2 ) 9.2903 x io_2
foot 3 /minute (cfm) meter 3 /minute 2
(m 3 /min) 2.832 x 10
degree Fahrenheit (°F) degree Celsius (°C) Tc (T — 32)71.8
degree Fahrenheit (°F) degree Kelvin (°K) TK = (TF+ 4 S 9 . 67 )/l. 8
gallon (gal) liter (1) 3. 7854
meter (in ) 3.7854 x 10
gallon/minute (gpm) liters/sec 6.309 x
meter 3 /min (in /s) 6.309 x 10
horsepower (hp) watt (w) 7.457 x 10
(550 ft lbf/s) cal/sec 1.7822 x 10
inch (in.) centimeter (cm) 2.5400 —2
meter (in) 2.5400 x 10
inch 2 (in 2 ) meter 2 (in 2 ) 6.4516 x lO
knot miles/hr (mph) 1.15078
kilometer/hr
(km/hr) 1.852
mile (mi) kilometer (kin) 1.6093
meter (in) 1.6093 x 10
miles/hr (mph) kilometers/hour
(km/hr) 1.6093
pound—mass (ib) kilogram (kg) 4.5359 x i0 1
pound—force (lbf) newton (N) 4.4482
pound—force/inch (psi) pascal (Pa) 6.89476 x lO
atmospheres (atm) 6.8027 x 10 2
ton (2000 ib) kilogram (kg) 9.07185 x l
yard (yd) meter (in) 9.144 x 10
ix
-------
SECTION 1.
INTRODUCTION
The problem of controlling air pollution from accidental releases of haz-
ardous substances is acute since the response time required to prevent air
pollution from occurring following a spill can be measured in minutes or even
seconds. The situation is further complicated by the dispersed locations at
which accidental spills occur and the need for mobilizing an appropriate con-
trol effort in accordance with geographical and pollution—type demand.
This study was conducted to: (a) examine the state—of—the—art techniques
presently utilized to control and/or mitigate air pollution generated by the
accidental release of hazardous gases or readily volatile hazardous substances
that represent a potentially serious threat to human life and the environment,
and (b) identify potential, practical, and economical technology that can be
developed to reduce and/or eliminate the incidence of serious human health
problems and environmental effects resulting from such air pollution. This
study identifies and describes air pollution control technique concepts (hard-
ware) that, if developed by further research and development efforts, would
possibly result in a reduction or elimination of deficiencies in air pollution
control techniques now utilized.
Concept definitions and inclusions/exclusions to the study are presented
below.
DEFINITIONS
1. State—of—the—Art Technology is defined as commercially available
technology now utilized in land spill responses, as well as tech-
nology classified in the final prototype (or nearly commercial)
stage of development for control of air pollution or for removal
or destruction of an air pollution source.
2. Control/Mitigation Techniques are defined as control technology
hardware and the methodology (or processes) used for hardware
operations in responding to episodes of air pollution generated
from accidental releases on land of gases or readily volatile
hazardous substances.
3. Hazardous Gas and Readily Volatile Hazardous Substances are defined
as solid, liquid, orgas commodities that, upon accidental release
or spillage, represent a threat to human life and the environment
in the form of air pollution, explosions, and fire. The criteria
for defining “readily volatile substances” were determined through
1
-------
comprehensive analyses of responses now utilized to control or
mitigate releases of such substances.
INCLUSIONS/EXCLUSIONS
1. The hazardous gas and readily volatile hazardous materials covered
in the study include only those pertinent materials listed in: (a)
US EPA’s “Designation of Hazardous Substances” list published in the
December 30, 1975, Federal Register (1), and (b) the U.S. Coast
Guard’s “Chemical Hazard Response Information System (CHRIS)” haz-
ardous chemical list (2). A special list, “Hazardous Chemicals That
Produce Vapor,” developed by the U.S. Coast Guard from the CHRIS 400
list, was extensively utilized as a guide for materials included in
the study. This list (developed in report No. CG—D—46—75) of 103
substances (3) is composed of chemicals that produce a significant
amount of vapor in a normal spill situation. The list contains
chemicals shipped as gases and chemicals shipped as liquids but hav-
ing relatively high vapor pressures at ambient temperatures.
2. The study was confined to accidents/spills of materials occurring
on land for inplant, outplant (in transit), and fixed storage facil-
ity environments.
3. Transport spills covered in the study included railroad, pipeline,
and truck modes of material transport (occurring on land).
4. The geographic scope included only spills occurring on land and spe-
cifically excluded accidents resulting in spillage of hazardous
materials in harbors, the inner—continental shelf, cotmnercial inland
waterways, or other inland water bodies. The Federal Water Pollu-
tion Control Act Amendments of 1972 (subsection 3llCl), and the
Magnuson Act (50 Usc 191 and 14 USC 88) have given statutory auth-
ority for response to hazardous chemical spills anywhere within the
waters of the U.S. to the Coast Guard. The U.S. Coast Guard has
sponsored several recent studies on spills in water (4, 5).
5. Radioactive materials were considered outside the scope of this
study.
REPORT DEVELOPMENT
Through interviews with spill control officials, Battelle determined
that less than 10% of the transportation—related spills involve fires of the
materials involved. The incidents involving fire may reduce the hazards from
the spilled materials, while also increasing the likelihood of spillage from
adjacent unruptured containers. Since the control of fire, per Se, is man-
ageable by currently and coionly available technology (water and foam), and
since the control of vapors, once released, is the subject of other ongoing
research, this study did not examine vapor control concepts for spills involv-
ing fire.
2
-------
The text of this report is divided into four sections. A brief discus-
sion of each section follows.
Section 4 . “RevIew of the Causes of and the Responses to Hazardous Land—
borne Volatile Spills.” Spill reports and records are examined to determine
the causes of spills on land of hazardous volatile materials, and the relative
frequencies thereof. A summary of current field response technologies for
controlling hazardous material vaporization is also presented. This su=ary
was compiled from Battelle staff field experience and from spill response
literature.
Section 5 . “Phenomena of Hazardous Materials Volatilization.” An under-
standing of the physical properties at work during a spill and in particular,
the knowledge of the relative importance of the factors contributing to vapor-
ization, must precede considerations of improved vaporization control technol-
ogy. Physicists, chemists, and meteorologists cooperated to produce this sec-
tion, which forms the basis for Section 6.
Section 6 . “Proposed Landborne Hazardous Volatile Spill Vaporization
Rate Control Concepts.” Utilizing the information contained in Section 5, a
number of idea sessions were held to generate concepts for new or improved
volatilization rate control technology. The results of these sessions are
presented.
Section 7 . “Scenarios Utilizing New Vaporization Rate Concepts.” Based
upon actual spills, a number of scenarios involving landborne releases of haz-
ardous volatiles were constructed. The application of existing technology is
then compared with the use of the concepts of Section 6 to determine the pos-
sible utility of these concepts.
3
-------
SECTION 2
CONCLIJS IONS
The current state—of—the—art for control of air pollution from spills of
hazardous materials occurring on land is simplistic; the response being depen-
dent upon local available materials (sorbents such as sand, straw, and flour)
and equipment such as fire trucks (water hoses, pumps, protein—based foams),
bulldozers, etc. Although not within the scope of this program, current re-
sponse techniques and equipment leave much to be desired in terms of personal
safety. The techniques now employed reqire close proxitn.icy to the spill,
while the personnel protective gear provides inadequate protection for the
hazards encountered.
Local fire departments are most frequently the first emergency units to
arrive at spill sites. Because of their training, fire department response to
spills often includes water flooding and, where immediately possible, vapor
source removal. Precautionary population evacuation is frequently initiated
by fire departments.
Where the local fire department is more sophisticated, foam blankets are
sometimes applied to reduce evaporation and flammability hazards. However,
final cleanup activities are generally not begun until emergency spill re-
sponse teams are present. These activities may include sorption and burial.
Ratings of currently used spill response techniques are summarized in
Table 1. The most effective procedures involve vapor source removal (reload
to enclosed vehicles), deep soil burial, water flooding, and air curtain
ignition systems.
There is significant room for improvement in cleanup technology for
spills of hazardous volatile materials on land. This report addresses mechan-
ical, chemical, and physical control, and techniques for applying ameliorative
measures.
An examination of the various mechanical, chemical, and physical means of
control indicates that lowering the temperature of the spill offers the most
effective and universally applicable technique of dealing with volatile haz-
ardous spills.
Solar heating of spills was found to be the predominant source of energy
for the vaporization of spills (rain storms were not included in this analysis).
4
-------
TABLE 1. RATING OF EFFICIENCY OF CURRENT RESPONSE
TECHNIQUES TO CONTROL EVAPORATION
Slight
Reduction
Moderate
Reduction
Highly
Effective
Reloading to Enclosed Vessels
X
Sumping and Trenching
x
Wet Foaming
X
Deep Soil Burial
X
Sorbents (Straw, Mulch, etc.)
x
Water Flooding
X
Dispersants (On Thin Water Layers)
x
Air Curtain Ignition Systems
x
Water Shroud Lines
..X
High—Pressure Water Fog
x
Many spills would benefit from the application of more than one vapor
suppression technique.
More than 60 techniques for delivery and deployment of coolants, tarpaul-
ins, plastic sheets, and foams, by plane, helicopter, parachute, agricultural
devices, skiploaders, cannons, mortars, catapult, line—gun, and crane, have
been reviewed. No effort was made to select those delivery and deployment
techniques deemed most promising.
5
-------
SECTION 3
RECONNENDAT IONS
This study has found there to be a critical absence of vapor suppression
techniques and equipment. Because of this lack, unnecessary damage has re-
sulted from the land spills of volatile hazardous materials. Current tech-
nology (e.g., water flooding), while adequate under some circumstances, has
often produced unwanted hazardous side effects.
Battelle considers the use of cryogenic media to reduce or eliminate vol-
atile emissions to be a most promising avenue of research, notwithstanding the
potential hazards from the applied cryogen and from explosive boiling. The
application of dry ice to reduce spill volatility has been used sporadically
but with considerable success in vapor suppression. Systems that can project
solid—slurries of granulated dry ice need considerable attention from re-
searchers. Because of its very low boiling point, systems using liquid nitro-
gen (LN 2 ) can be used effectively to suppress vaporization of a very broad
range of hazardous substances. Considerable attention must be paid to the
potential problems of explosive boiling of LN 2 and the possible problem of
condensation of free liquid oxygen.
In general, remotely operated systems to deliver the cryogens and adsorb-
ing medium are needed. Systems such as air—driven seed—blower guns could de-
liver sufficient quantities of granulated dry ice to blanket a spill in a few
minutes. Applicator guns and nozzles to deliver LN 2 and dry ice to a target
are also recoomiended for further research.
6
-------
SECTION 4
REVIEW OF THE CAUSES OF AND THE RESPONSES TO
HAZARDOUS LANDBORNE VOLATILE SPILLS
Before exanu.ning the needs for new technology, it is desirable to first
examine the types of landborne spills of hazardous volatiles that are occur-
ring, the nature of their causes, and current field response technology for
controlling volatilization rate.
HAZARDOUS VOLATILES
The types of chemicals that are classified as hazardous volatiles include
all high—vapor—pressure chemicals (those not boiling at ambient temperature
and pressure) and cryogens (those boiling at ambient temperature and pressure).
Chemicals that vaporize at a rate sufficient to have toxic or flammable con-
centrations also need to be considered, even though-they are quiescent at
ambient conditions.
A list of 200 chemicals that are transported on land and that present a
hazard from volatilization when spilled was developed from U.S. Coast Guard
and EPA lists of hazardous materials (1, 2, 3). A presentation and discussion
of these chemicals is covered in Section 5.
SPILL CAUSES
The draft of a report was made available to Battelle by the Factory Mut-
ual Research Corporation (F 1RC) (6). With this report it was possible to
examine the records of actual spills and their causes.
F IRC’s breakdown of reported spills by operational area and by Urinary
cause is given in Table 2. Most of the data collected by FMRC (6) on the in-
cidence of spills directly reflects the lack of regulations with regard to
reporting. The distributions given are those of reported spills , and are not
necessarily the true distribution of national spill experience.
Frequency associations of primary and secondary causes, given in terms of
percent of primary cause incidents in which the secondary cause appears (6),
are given in Table 3.
The distribution of reported spills shows that 82% of the incidents are
related to shipping. Specifically, in—transit spills total 57% while loading
and unloading account for 25%. Although it is recognized that the reporting
of spills is incomplete and that the FMRC data reflect this fact and hence nay
be biased, in view of the overwhelming incidence of spills that occur while
7
-------
TABLE 2. DISTRIBUTION OF REPORTED SPILLS OF HAZARDOUS MATERIALS
Operational Area
Primary Cause
to Transit 57%
Non—Tank Rupture, Puncture 43%
Loading—Unloading 25%
Tank Rupture, Puncture 25%
In—Plant Process 10%
Tank Overflows, Leakage 17%
tn—Plant Storage 7%
Hose Transfer System Failures 7%
Other 7%
TABLE 3. FREQUENCY
CAUSES OF
ASSOCIATION OF PRIMARY AND SECONDARY
HAZARDOUS MATERIAL SPILLS
Primary Cause
Secondary Cause
Tank Rupture, Puncture
- Derailment, Collison and/or
Overturn (54.7%)
Tank Rupture, Puncture
Body, Container Failure (33.4%)
Tank Rupture, Puncture
Sharp Object (15.5%)
Tank Overflow, Leakage
Personnel Error (29.6%)
Tank Overflow, Leakage
No Known Secondary Cause (26.4%)
Tank Overflow, Leakage
Mechanical Failure (18.5%)
Hose, Transfer System Failure
Hose or Coupling Failure (75.0%)
Non—Tank Rupture, Puncture
Sharp Object (31.1%)
Non—Tank Rupture, Puncture
Improper Loading (25.5%)
materials are in distribution systems, Battelle has concentrated upon this area
of need.
CURRENT TECHNOLOGY
Based upon interviews with spill response personnel, Battelle staff expe-
rience, and spill literature, a number of currently available techniques
8
-------
suitable for vaporization rate control have been identified. These techniques
are discussed in the following paragraphs.
Because of the lack of sufficient comprehensive spill incidence records,
Battelle interviewed responsible individuals who are involved in immediate
spill response actions. The effort centered upon spill contractors, govern-
ment officials, and industrial environmental response personnel.
After consultation with government officials, corporate environmental
managers, and insurance carriers, it was determined that water flooding is used
on approximately 70% of the chemical spills, while approximately 20% are foamed
over and 10% are treated with organic sorbents. Of the foaming procedures
available, it has been found that the majority of fire control teams use a
protein—based foam to blanket the volatile pools of chemicals. Reduction of
spill area is employed in conjunction with the other techniques.
Vapor Source Removal
By far the most frequent response of spill control personnel to vaporiza-
tion rate control is that of removal of the source of the vapor via offloading
of the material from a damaged container to a more secure container.
Vapor source removal has also been used on hazardous materials which have
already spilled. In these circumstances it is necessary to collect the spilled
Liquid in a contained area, and then pump the liquid into a closed container.
Foam Dikes
Foam diking is a relatively new technology, and thus experience in its use
is limited. Not all surfaces are capable of serving as a base for foam dikes.
There are two types of foam materials currently available.
Polyurethane Foam Systems-—
Polyurethane foam systems appear to possess the necessary characteristics
for field—operable units. The foam is formed by freon dissolved in one of the
two reactants. As the polymerization proceeds, the heat of the reaction and
the changing chemical composition causes the freon to volatilize and create
the foam. The only work necessary is supplied by pressure, which expels and
blends the two components. Other polymer foams require mechanical agitation
and/or are open cell and thus a porous medium.
Polyurethane foams do have some limitations. On dry surfaces such as
cement and asphalt, effective adhesion is easily achieved to contain a water
depth of several feet. The ability to contain liquids behind barriers built
of the foam on soil, however, depends upon the firmness of the soil. On hard—
packed earth, effective control is often achieved, while on loose soil the
fluid eventually works a path in the soil beneath the barrier. The polyure-
thane foam also performs poorly on vegetated ground and wet surfaces. The
foams can also be used to seal storm drains even when moderate quantities of
liquid are flowing through the drain gate.
9
-------
Inorganic Foam Systems——
The major item in this class of foam materials is a light—weight cellular
concrete. Foamed gypsum is also an available material, as well as sodium sil-
icate foam. The latter, however, requires mechanical agitation, is partially
open—cell, and is subject to hydrolysis. All the materials above are formed
by blending already formed, low—expansion foam into the appropriate slurry.
Wet Foams, Blanketing
A technique frequently employed by fire departments to decrease vapor
ization from hazardous chemical spills is the application of a foam blanket
over the liquid. As will be discussed in Section 5, the foams in use today
cannot effectively control vaporization. They can limit evaporation, however,
by reducing heat transfer to the spilled material. Problems of foam usage
include the possibility of foam ignition, the need to continually replenish
the foam, and the susceptibility of foam to “lifting” and drift by winds.
Sumping and Trenching
By construction of sumps and trenches, it has been possible to reduce the
surface area of a spilled material and thus reduce vaporization. Such construc-
tion also has enabled the collection and safe storage of the spilled material.
Deep Soil Burial
On occasion, spilled volatile materials have been buried on site, with
the effect of greatly decreasing the rate of release of hazardous vapors to
the atmosphere. This practice raises the spectre of other problems, such as
groundwater contamination.
Sorption
A response technique that is frequently employed involves the application
of a sorbent to the spilled material. Sorbent application can reduce the
vaporization rate and can often result in the recovery of at least some of the
spilled chemical. This technique is widely used because organic materials
such as straw, charcoal, mulch, flour, and corn cobs are readily available.
However, there are also commercially available reusable sorbents.
If the spent sorbent is recovered, the problem of disposing of the sorbed
hazardous chemical, whether or not the chemical and sorbent can be economic-
ally separated, still remains. When the sorbent and hazardous material are
separated, facilities must be provided for temporary (and perhaps long—range)
storage and final disposal of the chemical. The failure to separate the haz-
ardous chemical and sorbent simply adds to the bulk of the contaminated mate-
rial and thus increases the storage and disposal problems.
Water Flooding
Most chemical spills are treated by water flooding, as a result, primar-
ily, of the abundance of water—pumping equipment and of the fact that emergency
response often involves fire departments. Water flooding of a spill often has
some value:
10
-------
1. The spilled material can enter into solution with the water, thus
decreasing the evaporation rate.
2. If it is not as dense as the spilled material, the water can cover
the spilled material, thus decreasing the evaporation rate.
3. The water can disperse the spilled chemical, decreasing the atmos-
pheric concentrations of the evaporant.
4. The water can be used to herd the spilled material into locations
more suitable for cleanup activities.
Water flooding should not be applied to spills as a matter of course. In many
cases, the application of water flooding has had severe side effects, includ-
ing the pollution of nearby streams and the explosion of contaminants in sewer
systems.
Controlled Ignition
Controlled ignition has been successfully employed on real spills, albeit
infrequently. Flare stacks have been employed by natural gas companies to vent
landfill areas around leaking gas pipes. Continual maintenance of flare stacks
is necessary due to frequent flameouts. A.n air curtain destructor is avail-
able which, in essence, creates incinerator conditions in a trench so that the
substance being burned is completely oxidized. The combustion products will
contain essentially no visible emissions and, more importantly, no unoxidized
hazardous volatiles. The air curtain destructors provide for excess air con-
ditions, more turbulence, and increased retention time compared to simple
open burning.
Burning is generally considered to be one of the most dangerous treatment
operations. It should be attempted only after a careful review of alterna-
tives and consideration of comparative risks. Open burning of spills usually
requires burning agents to initiate and facilitate combustion. There are no
standard, safe methods for igniting open spills.
RATING OF CURRENT TECHNOLOGY
While it is unlikely that all of the techniques that have ever been em-
ployed to control the evaporation of hazardous volatiles from landborrte spills
have been discussed in this section, from Battelle’s experience and from inter-
views, it is believed that the majority (and certainly the most frequently
used) have been addressed.
Table 1 has been prepared as a subjective ranking of the ability of the
above—mentioned technologies to control evaporation. This table does not
address itself to the probability that, given a spill, the technology would be
applicable. Rather, it presents the expected efficiency of evaporation con—
trol, given that the technology is applicable.
11
-------
EVALUATION OF NEEDS
From the information given in Table 1, it appears that there are three
techniques which can be highly effective in controlling vaporization. They
are vapor source removal, water flooding, and controlled ignition. Each of
these techniques has limited applicability. Vapor source removal implies that
it is possible to collect the spilled material and place it in safe containers.
Water flooding, in essence, is simple dilution and can result in other prob-
lems. Controlled ignition is best left to isolated circumstances.
Spills of hazardous volatile materials must be dealt with quickly and
effectively* to limit the impact on public health and the environment. At
present, the arsenal of spill response tools is rather limited in applicabil-
ity. The methods of controlling the vapor release of spills can be general-
ized as belonging to one or more of the following control categories:
1. Control by the use of mechanical means. By placing a barrier
between the hazardous material and the environment, the release of
evaporant to the air can be slowed or stopped. This includes re-
loading to closed containers, and covering the surface with tarpaul-
ins, foam, etc.
2. Control by the use of chemical means. By addition of selected chem-
icals to the spill, the chemical or physical form of the spilled
material is altered to control the release of the hazardous vola-
tile. This includes water flushing, the addition of chemical neu-
tralizers, and controlled ignition (which is the addition of oxygen
at certain temperatures).
3. Control by the use of the physical properties of the spilled sub-
stance. By changing the conditions affecting the spill, it is pos-
sible to utilize the physical properties of the spill to control
vaporization. This includes sorption, the lowering of the tempera-
ture of the spilled chemical in order to decrease vapor pressure, and
the use of insulation to reduce the rate of heat transfer to the
spilled chemical (this, in effect, also lowers the temperature of
the spill).
There are a number of approaches as yet unexplored, by which vaporization
rate control can be attained. The remainder of this report has focused upon
the conceptualization of hardware (based upon existing technology for ease of
development) for the control of vaporization rate. Before proceeding with a
discussion of hardware, a study of the physical processes and their relative
* The question of spill response personnel safety is somewhat peripheral to
this study. It is recognized that current safety gear, even including
experimental suits, is inadequate for some of the potential spills, since
the demand—type regulators will allow infiltration of toxic gases. There-
fore, many of the techniques presented rely upon remoteness for safety,
although personnel safety equipment is not within the study scope.
12
-------
imoortance is presented in Section 3. The study has been used to help deter—
mine the parameters which are likely to have the most effect in hazardous vol-
atile spill vaporization control.
13
-------
SECTION 5
PHENOMENA OF HAZA.RDOUS MATERIALS VOLATILIZATION
The purpose of this section is to examine a spill model and various pos-
sible vapor suppression concepts with respect to the physical and chemical
phenomena involved during the spill and its suppression. The discussion has
been purposely kept simple and the calculations are intended to produce order—
of—magnitude answers. Careful simplification is both necessary and appropri-
ate in order to obtain useful general observations which can be applied to
many spills. More detailed models would, of necessity, involve the use of
additional information specific to the properties and geography associated
with a particular spill.
Battelle has selected 200 industrial compounds from the CHRIS Hazardous
Chemical Data File (2) as representative of materials that present a hazard
from volatilization when spilled. Their approximate physical properties will
be used in numerical estimates and to determine the effects of various control
concepts. The 200 chetnic ls are listed in Tables 4 through S (tables are at
the end of this section), along with their selected physical and chemical
properties, a description of the relative hazards associated with each com-
pound, and information relating to the results of several potential vapor
suppression concepts.
In the following discussion, a very qualitative description is given of
some of the more important phenomena which may be observed during a typical
spill. This will include an identification of physical processes that may be
of importance in understanding the ultimate fate of the spill. A description
of the behavior of the vapor plume formed by the spilled material is also in-
cluded for completeness.
Several plausible approaches to the control of hazardous vapors associ-
ated with spills are examined by utilizing qualitative estimates of the magni-
tude of pertinent physical processes.
As a result of these analyses, Battelle believes that the concept of spill
temperature reduction will reduce the human hazard associated with the widest
variety of possible spill materials. This control technique can immobilize a
spill and retard vaporization of the spilled material during removal and dis-
posal. In addition, no examples were noted among the selected spill chemicals
for which application of the temperature reduction concept as described would,
in fact, worsen the environmental impact.
14
-------
Vapor containment, either with foam, a polymer layer, or water, is a
viable concept for many spill materials. However, since this approach will
not work for some spills and will be counter—productive for others, it is con-
sidered to be less useful than temperature reduction.
Some materials physisorb extremely well and adsorption could be an effec-
tive control procedure for these. For most materials, however, physisorption
would be a marginally effective vapor suppression technique and therefore of
still less potential use.
Finally, chemical neutralization is judged to be of quite limited utility
primarily because only a small fraction of the test compounds can be treated
in this manner, Those few chemicals that can be so treated require a number
of different neutralization agents. In individual cases, neutralization might
turn Out to be the treatment of preference, but generally, this approach is
considered to be of low potential.
SPILL DESCRIPTION
Low—Pressure Material
The chronology of a spill will be examined. The material spilled will be
assumed to have a moderate vapor pressure (less than atmospheric) at ambient
temperature; a material fitting this description might, for instance, be hex—
ane.
As the material spills to the earth, some evaporates on the way down and
a small quantity vaporizes on striking the ground, causing some cooling of the
ground and the remaining liquid.
The liquid actually reaching the ground will soak into the soil and dif-
fuse downward and transversely through the ground.* Some physical sorption
occurs, binding the spilled material to the soil particles. Liquid spilled on
pavement will spread and form very shallow puddles with slight penetration and
some sorption by the pavement.
As the ground becomes saturated with the spilled material, the material
cannot diffuse into the soil as fast as it is spilling. Because of the latent
heat required by the vaporized material, the ground cools slightly, giving up
sensible heat, and heat is conducted in (at a slow rate because the temperature
of the spill differs little from its surroundings) from the surrounding earth.
The formation of poois then begins. Material spilled on pavement will spread
until it reaches soil and then will behave as described above. The flowing
shallow puddle can gain heat rapidly from the air and pavement because the
surface—to—volume ratio is quite large.
After the spill stops or slows, the pool evaporates because of heat radi—
ated from the sun and conducted through the ground or pavement (plus some
* Unless the spilled material is non—polar and the soil is wet — in which
case the spill disperses on top of the soil.
15
-------
convection heat transfer from the air, though this is much smaller*). Either
ground or pavement pools will tend to persist since adsorption increases the
effective “heat of vaporization,” i.e., sorbed spill material will evaporate
more slowly.
After the pooi has evaporated, the ground retains some of the spilled
chemical which must diffuse out of the soil in order to evaporate. Evapora-
tion of the final portions of the spill from the soil is the result of a com-
plicated heat and mass transfer process. The evaporative process continues
at an ever—decreasing rate. When the concentration of the chemical declines
below some lethal level, the residues may be transformed by biological action,
in addition to vaporization.
High—Vapor—Pressure LiquIds
High—vapor-pressure liquids are here defined as those liquids for which
the vapor pressure exceeds one atmosphere at ambient temperature. Thus, this
material is boiling at the beginning of a spill. Therefore, boiling phenomena
are of potential interest to one concerned with spills of high—vapor—pressure
hazardous materials.
There are several types of boiling:
1. Nucleate boiling
2. Film boiling
3. Leidenfrost boiling
4. Explosive boiling
Explosive boiling is applicable only for a spill of a very high—vapor—
pressure fluid (greater than atmospheric pressure) into another liquid; LNG
into water is the classic case. In film boiling, heat transfer from the sur-
roundings to the liquid is through a film of vapor, viz:
Liquid
Vapor
Hot Surroundings
By the same logic as is applied to explain the behavior of foam on vapor (see
discussion below), the liquid—vapor interface will be unstable and boiling
will occur. This results in the familiar rolling boil of water in a pot.
* Convective heat transfer from the air will be much smaller than that con-
ducted through the ground because the spill will be protected from the
warmer air by a layer of cooler vapor, the conductivity of gases is typi-
cally two orders of magnitude smaller than that of solids, and the specific
heat per unit volume of gases is typically three orders of magnitude smaller
than that for solids.
16
-------
Despite the violent appearance of this type of boiling, heat transfer between
the surroundings and the liquid is limited by the poor conductivity of the
vapor film (relative to direct conduction between the liquid and the
surroundings).
In nucleate boiling, the pool is quiescent with very small bubbles form-
ing at nuclei on the surface of the hot surroundings. In this case, the rate
of heat transfer to the liquid can be much larger than for film boiling, since
in nucleate boiling the liquid is in direct contact with the surroundings.
Leidenfrost boiling is a special case of film boiling and occurs when
drops of liquid are completely supported on a film of the vapor which separ-
ates the liquid from a solid surface. These liquid drops (e.g., cryo—fluids
on pavement) are extremely mobile as the “vapor bearing” offers little resis-
tance to movement. Thus, a spill of LNG (for instance) might spread a con-
siderable distance as a liquid (carrying much hazardous material) and fail to
be limited by available heat. Only sufficient heat to maintain the film under
the material is required.
Each of these types of behavior is possible (as are more complicated sit-
uations that occur, e.g., when it rains) with the very high—vapor—pressure
materials such as butane, ammonia, etc., which, when boiling, will lower the
temperature of their surroundings substantially. Thus, a large enough spill
could result in a pool that is in steady state with the ground and insulated
from the atmosphere by cold vapor.
ATMOSPHERIC DISPERSAL
This section would not be complete without a description of the naturally
occurring process by which the evaporated gases are diluted to safe levels
and removed from the spill site. Qualitative discussion of the major features
of transport and diffusion, together with means of estimating concentrations
and dose, are presented. In addition to providing completeness to the discus-
sion of spill chronology, this description gives a feel for the variety of
atmospheric factors which may complicate simple—minded descriptions of plume
phenomena.
Meteorological Factors
At the site of a spill of hazardous material, the principal interest in
meteorological parameters concerns the material removal by atmospheric proc-
esses. Jind speed, wind direction, and atmospheric scabil ty are important
parameters determining the extent of atmospheric dispersion,while temperature,
humidity, and precipitation are of lesser importance.
Dilution and Diffusion
Atmospheric dispersion of a material is controlled by two processes —
dilution and diffusion. The greater the amount of air passing through a given
volume, the more a material within the volume is diluted. Strong winds at a
spill site will produce a lower concentration of the hazardous material down-
wind of the site. Atmospheric turbulence along the axis perpendicular to the
17
-------
wind produces horizontal and vertical diffusion. These turbulent eddies cre-
ate an exchange between the air in the plume and the cleaner air outside,
spreading the material over a more extensive volume. As a result, high con-
centrations of the material within the plume are reduced.
Turbulence has two causes —mechanical (exemplified by the eddies created
when the wind flows past an obstacle) and thermal, which includes the eddies
produced by rising hot air. In meteorology, atmospheric stability describes
the ability of the atmosphere to enhance (unstable atmosphere) or inhibit
(stable atmosphere) turbulent eddies. Under unstable conditions, the plume
will diffuse rapidly. However, unstable conditions can also cause problems
with an elevated plume as the plume may be carried downward momentarily, pro-
ducing high pollutant concentrations at the surface.
Plume Rise
An important consideration in the dispersion of the plume from a spill is
the plume rise. The higher a plume rises above its source, the greater the
distance the material within the plume must travel downward to return to the
surface. This increase in the distance the material remains airborne pro-
vides additional time for horizontal and vertical diffusion with the conse-
quent decrease in atmospheric concentration of the material.
A plume hotter than the air in its vicinity will either:(l) rise until
it cools to the temperature of the air around it, (2) rise until it reaches an
inversion where plume temperature and ambient temperature are equal, or (3) be
bent into the horizontal by the force of the wind. Strong wind, high ambient
temperature, and the presence of inversions will reduce plume rise. Since
plumes from evaporating liquids will be cool, the effective height of non—
burning vapors will essentially be zero, and the plume will tend to hug the
ground.
Ambient Concentration Prediction
Mathematical equations have been developed for predicting the downwind
concentrations from a source emitting to the atmosphere. The general Gaussian
diffusion equation is
C (x, y, z, H) = 2o e [ _½ (v) 2 ]
(1)
where
C(g tn 3 ) is the mass concentration at a point x, y, z (where 0, 0, 0 are
the source coordinates)
Q(g sec 1 ) is the emission rate
18
-------
(m sec 1 ) is the mean wind speed in the x direction affecting the plume
(in) are dispersion coefficients in the horizontal and vertical
directions, respectively
H(m) is the height of the plume centerline when it becomes essentially
level (zero for nonburning spills).
The dispersion coefficients depict the diffusive characteristics of the
atmosphere. The more unstable the atmosphere, the larger are the dispersion
coefficients. They also are a function of distance from the source, increas-
ing in magnitude with increasing distance. Variations of these general equa-
tions can be used to predict downwind concentrations under specified conditions.
Dosage
The foregoing equation predicts concentrations when the source is emitting
at a continuous rate. Some spills etrat all the pollutants instantaneously in a
puff or cloud. For this case, the dispersion equation is altered to predict
dosage — the integration of concentration over the time of passage of a plume
or puff. The equation for the dose at the surface (z = 0) is
Q 2
-__
D (x,y,O,H) — exp ½ exp ½ (2)
yzx y z
where
Dt = total dosage (g sec m 3 )
Q = total release (g).
t
Field Applications
Several agencies charged with emergency action in response to spills have
developed calculator or computer programs which use these dispersion equations
to predict where the maximum atmospheric concentration will occur and how large
this maximum will be. Extensions of the programs trace out the extent of the
dangerous plume concentrations and which direction the plume will follow.
Observations of parameters needed by the program can be made at the spill
site or interpolated from weather stations in the vicinity. One such plan re-
quires that on—site observations be made of the following meteorological vari-
ables: wind direction, wind speed, percentage of sky covered by clouds, cloud
height, cloud type, temperature, plume type, and precipitation or weather
conditions.
HEAT TRANSFER
The following discussion will refer to spills of chemicals with a vapor
pressure less than one atmosphere at 25°C. At steady state, an energy balance
can be written for a spill which equates the energy lost by the vaporization
process to the energy gained by conduction from the ground, convection from
19
-------
the atmosphere, energy input from rain or snow, and radiation from the sun.
Before examining the energy input from the surroundings in greater
detail, the rate of vaporization from the spill will be considered.
In the absence of wind, a pooi of high—vapor—pressure material will
vaporize until its partial pressure above the pool surface attains the equi—
libriutn vapor pressure of the material at the temperature of the pooi surface.
Equilibrium will not be achieved, however, since there is no container over
the pooi and the vapor will continually diffuse away. As the liquid vapor-
izes, removing energy (the latent heat of vaporization), it will lower the
surface temperature of the pooi. The presence of cooler, more dense material
at the surface of the spill will tend to generate convection currents within
the pool that will serve to reduce temperature gradients within the spilled
fluid.
The steady—state condition will be characterized by a pool temperature
lower than the ambient temperature (heat is entering the spill and being re-
moved in the vapor). The vapor diffusion rate away from the pool will also
play a major role in determining the steady—state condition. Thus, in general,
a high—vapor—pressure material with a larger gas diffusion coefficient and a
larger latent heat of vaporization will also establish a lower steady—state
temperature.
Under wind conditions the vapor will be continually swept away from the
pool. This forced convection process will enhance vaporization and produce a
pool temperature lower than that found under still air conditions. iany other
factors can complicate this simple picture. For example, if the material is
spilled onto a grassy area, each blade of grass may behave as an evaporation
surface so that the effective surface area is magnified. On the other hand,
tall grass nay reduce forced convection mass transfer and solar radiation In-
put by shielding the spill from the wind and sun.
Quantitative Discussion of Evaporative Cooling
The model used for this calculation will be a pool of liquid of character-
istic surface dimension (2r) with a wind velocity (v) passing over it. For
typical spill sizes, uncontaminated air will enter from one side of the spill,
pick up vapor as it passes over the spill, and leave the other side with essen-
tially no large—scale turbulent mixing. Under these conditions, the slug of
air can be regarded as stationary and the diffusion of the vapor into the at-
mosphere viewed as a situation in which the initial condition is air devoid
of vapor. The air immediately above the spill can be regarded as near—
saturated at all times since the maximum evaporation flux,
/ RT ½
J max psat T)
(where psat is the saturated vapor density and R is the gas constant), is very
large (it takes just a few picoseconds for this condition to be established).
Thus, the rate of evaporation is limited by the diffusion or convection of the
vapor away from the spill surface and not by the inherent rate of evaporation;
20
-------
for this reason, diffusion barriers which float on the liquid and in which the
liquid is insoluble,or only slightly soluble, can have a major effect on the
rate of evaporation.
Under the condition and the approximations described above, the diffusion
problem becomes quite simple and the analysis appears in most elementary texts.
The evaporation flux is / 2 \
I 1-x 2
J = osat N exPt L ) 5/cm —sec (4)
where d. 9 is the diffusion coefficient, t the time after which the diffusion
started, and x the distance from the surface. The total flux of material out
of the liquid (at x = 0) is
J = Psat g/cm —sec. (5)
For a slug of air passing over the spill in a time to = 2r/v, the average
evaporative flux into the atmosphere is
t
— 1 o 2
J = - f J(t) dt = 2 sat — g/cm —sec. (6)
A typical value for in gases is about 0.2 cm 2 /sec. If the spill is
about 50 meters long and a wind velocity of 1Q rn/sec (22 mph —a good, stiff
wind) is assumed, i.e., 5 seconds, the average evaporative flux is about
0.2 psat g/ctn 2 —sec.
Since psat - 10 g/crn (for a gas of molecular weight 200 with vapor
pressure of 2 psi at 25°C), then = 2 x i0 g/cm 2 —sec.
Since the latent heats of most of the volatile hazardous materials under
consideration range from 50 to 100 cal/gm, an average net heat flux input of
2 x 10—2 cal/crn 2 —sec 1 cal/crn 2 —min is required to maintain the spill at
constant temperature.
Having determined the approximate magnitude of heat flux needed to main-
tain a moderate vapor pressure spill at constant near—ambient temperature, it
is possible to discuss the possible modes by which the spill can acquire this
heat.
Ground Surface
Heat will transfer to the spill from the ground. At first, it will be
the sensible heat of the soil near the interface (this could cause boiling in
some cases). The heat supplied by conduction through the ground will decrease
with time as described by a transient heat transfer equation.
Heat Transfer to the Spill From the Atmosphere
When there is no wind, the heat transfer from the atmosphere to the spill
will occur via conduction through the vapor—air mixture over the spill. When
21
-------
there is a wind, heat will be carried to the spill by convection. tn either
case, the rate of heat transfer from the atmosphere to the pool is orders of
magnitude less than that from the ground.
Rain
Additional heat can be supplied to a spill by rainwater (or water intro-
duced by attempts at dilution) in the amount of one cal/gm water/°C.
Comparison of Heat Input From the Sun and the Land
It is instructive to compare the energy available to the spill as sensible
heat of the earth to the energy available as a result of direct solar radia-
tion. The specific heat of soils is about 0.2 cal/g/°C and the bulk density is
approximately 1.2 to 1.6 g/cc. When the latent heat of a spilled material is
75 cal/g , then approximately 375 g of earth would have to be cooled an aver-
age of 1°C to evaporate 1 gram of spill. Very roughly, then, a volume of soil
10 in x 10 in x 1 in would have to be cooled an average of 1°C to evaporate 100
gal of spilled material.
Consider a simple geometry (that of an ellipsoid) in which a spill fills
a circular depression of radius r (as seen from above) to a depth d. The
- . 2r
maximum rate of radiant energy transfer from the sun to the spill is
= Sir r 2 cal/mm. (7)
The solar flux, S, will be assumed to be approximately one—half the solar con-
stant, or approximately 1 cal/cin 2 -min.
If it is assumed that the spill is evaporating slowly, is T below the
average soil temperature, and that the soil temperature gradient can be approx-
imated as T/L, then, for soil conductivity of 0.003 cal/(cm—sec—°K), the rate
of heat transfer to the spill from the soil is
2
q = kA =(0.003)(60) L T cal/tnin. (8)
Here, r has been assumed to be much greater than L or d, i.e., the area of the
interface between the spill and the soil is approximately irr 2 . Thus, the ratio
of heat supplied by the earth to that supplied by the sun is
— 0.18 AT
L
22
-------
If a hemispherical depression had been assumed instead of a very flat one,
the ratio would be twice this value. The approach is believed to be a satis-
factory way of estimating the relative importance of soil conduction and solar
radiation. The energy given up by the earth in cooling the shell of thickness
(L)surrounding the spill, by an average of ( T/2) is
- 0.3 rr 2 LAT cal. (10)
This value will be compared with the energy required to evaporate the entire
spill obtained by multiplying the volume of the spill, iid 2 (r—d/3), by an appro-
priate avera e latent heat of vaporization (100—150 cal/gm) and by the density
(0.7—1 gm/cm ); thus
E -lO6vd 2 (r-d/3)
For a large spill that is 1°C lower in temperature* than the surrounding
soil, the rate of heat conducted through the soil is about equal to 36% of the
maximum provided by solar radiant energy when the soil has given up enough
heat to evaporate 0.2/r percent of the total spill. Thus, it can be seen that,
for a large spill, solar radiant energy can be a more important factor in pro-
viding the heat needed for evaporation than thermal conductivity through the
earth. Of course, the existence of an opaque vapor cloud over the spill can
complicate the solar radiation effects.
FACTORS INVOLVED IN SO VAPOR CONTROL CONCEPTS
Many methods have been tried or proposed for the control of the release
rate of vapors from evaporating liquids. It is instructive to examine the
mechanisms by which these proposed methods work, so as to help in the deter-
mination of the relative worth of the various techniques.
Diffusion Barriers
A diffusion barrier reduces the net evaporation rate by hindering or re-
tarding the passage of molecules through it to the vapor phase or by reducing
the spill area available for evaporation. The barriers cannot alter the
equilibrium vapor pressure, but will act to slow the rate of approach toward
the equilibrium. Barriers can be mechanical devices which enclose and seal
off the spill area or floating objects such as “ping—pong” balls, or they can be
an immiscible liquid layer which can be spread over the spill, a surfactant,
or a light foam.
At present, experiments are being conducted on the use of foams on spills
as a means of vapor containment (4). A cross—section of the foam—blanketed
* This number was chosen to represent a steady—state condition (where the pool
temperature is the result of the effects of evaporation, solar radiant energy,
and conduction) and not the initial conditions. The temperature difference
is taken as representative of that of a chemical with a moderate vapor
pressure.
23
-------
spill would appear as shown below.
Atmosphere (
Foam
Vapor
Liquid 3
A goal of the foam—blanketing experiments is to produce this stable lay-
ered configuration. Often, however, the vapor tends to bubble up through the
foam (vapor also diffuses through the foam, but this is a different problem).
This is a problem in classical mechanics called the Taylor instability. The
foam—vapor interface will be stable (i.e., vapor will not tend to break through
the foam) when the mass density of the foam is less than that of the vapor,
i.e., if the relative density of the vapor is 2 (relative to air), then foam
containment will be successful if the density of the foam is less than 2.
Physical Adsorption
Physical adsorption describes an interaction between a molecule and a
substrate surface caused by secondary attractive forces (van der Waals). It
does not involve the transfer of an electron or electrons between the surface
and the molecule. (When electron transfer occurs, the process is called
chemisorption and amounts to a chemical reaction between the surface and the
molecule.) Physical adsorption is reversible with the addition of a modest
amount of thermal energy (usually less than 4000 cal/gtnol); chemisorption
requires the breaking of chemical bonds for reversal and requires correspond-
ingly larger amounts of thermal energy. An example of “physisorption” is the
adsorption of organic molecules on activated charcoal; chemisorpcion might be
exemplified by the rusting of iron.
The forces responsible for physical adsorption are associated with elec-
trical dipoles. There are both first—order and second—order forces. First—
order interaction involves molecules with a permanent dipole moment. When such
a molecule approaches a surface, a t image dipole is induced at the surface,
which results in an attraction between the surface and the molecule. At short
distances, electronic repulsive forces come into play, resulting in a thin
surface layer itt which the molecule may be trapped.
Molecules lacking a permanent dipole moment generally possess some elec-
tronic polarizability. That is, they can display an induced dipole moment.
Spontaneous fluctuations within the molecules result itt transient dipole tno—
tnents,wnich generate image dipoles in the surface and a consequent attractive
force between the surface and the molecule. Because this force depends on
spontaneous charge fluctuations, its magnitude is smaller than for those mole-
cules with a permanent moment. Generally, the strength of the bonding induced
by such forces is proportional to the polarizability of the molecule.
Thus, molecules with a very large dielectric constant (which implies a
permanent dipole moment for the molecule) will physically adsorb well. For
molecules with no permanent dipole moment, those with the largest dielectric
24
-------
constants will be likely to form the strongest bonds.*
With this physical background in mind, the volatile hazardous chemicals
for which dielectric constants ( ) are listed were grouped according to the
value of this parameter. Compounds with a very large dielectric constant
> 30) were judged to have a large dipole moment and probably a large polar—
izability. It was expected that these compounds would adsorb very well and
they were therefore rated as “very good.” Smaller dielectric constants
(10 < < 30) were judged to be due to smaller permanent dipoles, and smaller
polarizabilicy. It was expected that chemicals with in this range would be
moderately attracted to the adsorbing medium. Thus, these materials were
rated as “good.” Still smaller dielectric constants (t 5) were judged to be
a reflection of either large dynamic polarizabilities with no permanent dipole,
or small permanent moments with perhaps small polarizability. Materials with
c of about 3 were rated as “fair.” Dielectric constants of the order of 2 re-
flect the lack of a permanent dipole moment and a relatively weak electronic
polarizability. Materials with t in this range were rated as “poor.’
Dielectric constants could not be found in the literature for all the
materials listed. For some chemicals, estimates of the efficacy of adsorption
as a control mechanism were made by using related criteria. In some cases,
structural similarities were used. As an example, since the cyanide group has
a large permanent dipole moment and is quite polarizable, compounds with a
cyanide group were rated as “very good” candidates for control by adsorption.
On the other hand, since alkanes and alkenes have very small dipole moments
and small polarizabilities, adsorption was assessed to be a “poor” means of
control. Permanent dipole moments have been tabulated for some of the com-
pounds. Although they do not reflect the dynamic electronic polarizability
as does the dielectric constant, they were also used to estimate the relative
effectiveness of adsorption as a method of control.
It must be recognized that ratings based on the size of the dielectric
constant are relative and subjective. That is, it can be estimated on the
basis of dielectric constants that a material with 10 will be better ad-
sorbed on a high—surface—area medium (e.g., activated charcoal) than a mate-
rial with E - 2. Because of other factors involved in the adsorption process,
however, the order might be reversed. The estimated ratings must be checked
experimentally. tn addition, it could well be that even if the material with
- 10 does adsorb much better than that with c - 2, physical adsorption may
be a perfectly adequate control mechanism for the material with - 2. This,
too, would need to be evaluated experimentally.
Effects of Temperature Reduction
Reducing the temperature of a potentially hazardous spill can be advan—
tageous in several ways. It can:
* There is not a linear relationship between the static dielectric constant
and the molecular dipole moment; however, the dielectric constant is a mono—
tonically increasing function of the molecular dipole moment.
25
-------
• Increase the viscosity and thereby effectively help to contain the
spill and to restrict thermal transfer by convection within the spill.
If the material spilled freezes at the lowered temperature, the boun-
dary of the spill can be maintained and, possibly, the rupture from
which the spill is coming might be sealed temporarily.
• Decrease the vapor pressure, thereby reducing exposure to the spill
fumes and probably reducing the explosion and flanunability hazard.
There are numerous coolants which might be used to reduce the temperature
of a spill. Here, two are considered: dry ice and liquid nitrogen (LN 2 ).
These were chosen because they are readily available, provide low temperatures,
are riontoxic, and do not support combustion. (CAUTION: Under no circumstances
should liquid air be used in lieu of liquid nitrogen. Since nitrogen boils at
a slightly lower temperature than oxygen, as the liquid air boils, nitrogen
will boil off preferentially, potentially leaving behind extremely hazardous
concentrations of liquid oxygen and oxidizable material. The concept of the
use of liquid nitrogen must itself be tested experimentally to determine if an
oxygen concentration builds up by condensation from the atmosphere.)
It is easy to determine whether the reduced temperature provided by dry
ice or LN2 could iobilize the spill by comparing the freezing point of the
spilled material with the boiling point of LN 2 or with the sublimation temper-
ature of dry ice. An accurate value for the reduction in vapor pressure re-
sulting from the reduced temperature woqld require accurate vapor pressure
tables for each of these materials. However, it is possible to estimate the
vapor pressure using thermodynamics and some reasonable approximations.
Clapeyron’s relation for a liquid phase in equilibrium with its vapor is
(approxirnately)*
dP P H
dT - RT 2
where P is the vapor pressure, AH is the latent heat of vaporization, R the
gas constant, and T the absolute temperature. This relationship is approxi-
mate in that it assumes that the vapor behaves as a perfect gas and that the
volume of one mole of liquid is negligible with respect to the volume of its
vapor. It is a reasonable approximation to take the latent heat to be con-
stant, in which case Calpeyron’s relation may be integrated to give
* Clapeyron’s relation for any first—order transition is
dP_ H
- T(v _v.)
wherevf andv are the final and initial volumes of material, respectively.
If the volume of the liquid phase is neglected and the vapor treated as an
ideal gas, the approximate Clapeyron’s relation is obtained.
26
-------
P
1n — = — (13)
where P 0 is a constant of integration. P 0 can be determined by observing that,
at the boiling point, P is one atmosphere (14.7 psi). Thus,
1 (atm) = Pe 1 Tb (14)
where Tb is the boiling temperature of the liquid. It turns out that, for
most materials,
20 cal/mole—degree. (15)
This is referred to as Trouton’s rule. Since R = 1.98 cailmole—°K, then
P e 1 ° (atm). (16)
Writing the integrated Clapeyron relation, Eq. 13, for another tetnperature,T’
in (P’/P) = —AH/RT’ (17)
and rearranging and combining Eq. 17 with Eq. 13 yields
(P/P) = (pI/p)(T/T) (18)
If T’ is taken to be the boiling temperature, Tb, then P’ = 1 and
P = P(P) _Tb/T - [ (Tb/T)_1J (19)
Because of the use of Trouton’s rule and other approximations, P has been cal-
culated at the dry ice sublimation temperature and the LN 2 boiling temperature
as a function of Tb for values of P 0 equal to e+ 9 and e+ 1 1. These calcula-
tions should bracket the correct value of P.
The values obtained from Eq. 19 are presented in Figure 1 to illustrate
the possible effectiveness of vapor pressure reduction techniques using dry
ice. Estimates of the vapor pressures obtained in this manner are displayed
in Tables 4 through S.
It is instructive to estimate the quantities of LN 2 or of dry ice needed
to cool a spilled chemical to the control temperatures implied by the applied
coolant. The latent heats of sublimation and vaporization of dry ice and L I 2
are, respectively, 90 cal/gm and 48 cal/gm, so that a given weight of dry ice
is nearly twice as effective a coolant as an equal weight of LN 2 . To cool one
pound of benzene to the dry ice temperature, for example, will require 0.7
pound of dry ice and to cool the benzene to the LN 2 temperature will require
1.8 pounds of LN 2 . (This, of course, does not include the amounts needed to
compensate for the increase in the heat transfer rate to the spill.) If it is
only desired to freeze the benzene, 0.3 pound of dry ice and 0.5 pound of LN 2
will be required.
27
-------
100
10_i
i o 2
w
I
I-
C
I
u_ i
C-
10_a
u _ i
U
>_
I
C
I-
C
C ,
—
tu 10
I
U ,
C I ,
u _ i
I
0 .
I
C
0.
C
> 5
10
Figure 1. Band of expected vapor pressures at dry ice temperature of
spilled materials versus their boiling temperatures.
10_i
-60 -40 -20 0 20 40 60 80 100 120 140
BOILING TEMPERATURE OF SPILLED MATERIAL 1°c)
28
-------
Effects of Viscosity
Another physical aspect which may enter the problem is that of the vis-
cosity of the spill. An inviscid fluid will spread quickly, gaining surface
sensible heat at the periphery of the spill. A more viscous fluid tends to
spread slowly, limiting heat transfer to the fluid. Thus, vapor control can
be achieved by the addition of “thickeners” or gelling agents to the spill, or
by otherwise constraining the spread of the liquid.
Chemical Neutralization
The reaction of the spilled chemical with other compounds to produce a
less hazardous product or a product with a lower vapor pressure will reduce
the danger of cleaning up spills of certain materials. However, this approach
is fraught with risk of increased environmental insult. The neutralizing
chemical reaction will release heat which could result in vaporization of un—
reacted parts of the spill. Temporarily, at least, the plume will then be more
hazardous than if nothing had been done. Moreover, the added compound and/or
the product (both of which must be low—vapor—pressure compounds of lower tox-
icity for the control process to represent an improvement over doing nothing)
may represent toxic effluents to nearby land, streams, and lakes. For instance,
neutralization of an HCN spill with an aqueous solution of caustic soda will
produce an acueous solution of sodium cyanide and caustic soda, which must be
prevented from entering neighboring waterways. Furthermore, the reacting agent
needed will be very specific to the particular material spilled. Addition of a
dilute acid to a spill (a proper course of action for a spill of basic material)
would clearly be counterproductive if the spilled material were also acidic.
Neutralization of plastic monomers by polymerization requires the addition of
catalytic agents. While a single catalyst might be found that would work for
most of the monomers used today, such a catalyst is as yet unknown.
As evidence of the variety of chemical control agents which might be
needed to effect vapor suppression by chemical neutralization, the following
tabulation is presented:
Chemical Class Reagents Needed For Neutralization
Bases (e.g. , amines) Acids or acidic salts
Acids Bases or basic salts (e.g., caustic soda,
lime, slaked lime, Portland cement, soda ash)
Aldehydes 1
> Amines (e.g., hydroxylamine)
Ketones
Cyanogen chloride
Nitrosyl chloride
Caustic soda
Phosgene
Oxides
Plastic monomers Peroxides
29
-------
CHEMICAL AND PHYSICAL PROPERTIES OF
SELECTED VOLATILE HAZARDOUS SUBSTANCES
A group of 200 industrial organic and inorganic chemicals has been
selected for consideration of possible concepts for vapor control during a
spill. The criteria for choosing the chemicals were that they had an appre-
ciable vapor pressure and that the vapor would be likely to represent a toxic
or explosive threat to human life. The list of 200 is not all—inclusive, but
was deliberately chosen to be sufficiently large that a potential spill con-
trol scheme could be tested to determine the breadth of its applicability.
The 200 chemicals selected are listed in Tables 2 through 6. The chemi-
cals are grouped in these tables by vapor pressure as shown below.
Table 2 — Compounds with boiling points of 25°C or lower
Table 3 — Compounds with boiling points between 26°C and 70°C
Table 4 — Compounds with boiling points between 70°C and 125°C
Table 5 — Compounds with boiling points between 125°C and 2000C
Table 6 — Compounds with boiling points above 200°C.
In these tables, several important physical and chemical properties, haz-
ard ratings, and the effectiveness of potential control procedures are dis-
played, as available for each of the chemicals. The physical properties
listed include the boiling point (in °C), the vapor pressure (at 80° ?), and
the relative density of the vapor (relative to air as unity) at ambient tem-
perature. It is interesting to note that the vapor from some of these mate-
rials is several times more dense than air. For spills of these compounds it
is possible (where the foam density is sufficiently small) that foam can be
used to retard vaporization without vapor breakthrough.
The chemical properties listed include a qualitative description of the
solubility of the chemical in water and a notation of the primary functional
groups represented in the chemical. The first of these properties is quite
important since many spills are first treated by flushing with water. In
addition, the solubility in water is relevant to potential plume control meth-
ods which use an aqueous diffusion barrier. The variety of groups appearing
in the second column of chemical properties illustrates the difficulty of ap-
plying chemical control techniques to spills.
The hazard ratings rank the various chemicals qualitatively on a scale of
1 through 4 with the hazard being greatest for the larger numbers. These rat-
ings are listed for flammability, vapor irritation, and human toxicity.
Finally, the applicability of several potential control schemes is rated.
These schemes are: (a) containment of the vapor through the application of foam
or a water diffusion barrier, (b) reduction of vapor pressure and/or increase
of viscosity through temperature reduction, and (c) prevention of vaporization
through physical adsorption. The applicability of containment is separately
rated for foam application and for use of a water diffusion barrier. The con—
trol scheme for physical adsorption was estimated from static dielectric con-
stants of the chemicals as discussed earlier.
30
-------
The temperature reduction approach to spill control has been evaluated
for dry ice application and liquid nitrogen application. It is seen that
nearly all- the 200 volatile hazardous chemicals freeze at liquid nitrogen
temperature (those chemicals for which freezing points could not be found in
the literature were left blank), and in most cases the vapor pressure is re-
duced to insignificant values. It can be seen that, even at dry ice tempera—
ture, the vapor pressure of many of the chemicals considered is lowered suffi-
ciently that the hazards associated with the vapor may be significantly reduced.
It is for this reason that Battelle believes the temperature reduction approach
has a very strong potential for control of volatile spills.
31
-------
LEGEND FOR TABLES 4 THROUGH 8
PHYSICAL PROPERTIES: MA — Not Available NL — Not Listed
NP — Not Pertinent NF — Not Flammable
HAZARD RATINGS: The hazard is greater the larger the number, e.g.,
acetaldehyde (4) is more flammable than acetone (3).
CONTROL PROCEDURES
Containment: N — The spill will not be contained with a foam, water,
or film overlay because the ambient vapor pressure
is too large
YF — The spill can be contained with a film or foam
overlay
NW — The spill cannot be contained with a water overlay
MW — The spill may be contained with a water overlay
YW — The spill can be contained with a water overlay
Temperature
Reduction: F — Spill freezes at indicated temperature
neg — Vapor pressure is negligible at indicated
temperature
8xl0 2 — Vapor pressure is 8x10 2 psi at indicated
temperature
Adsorption: VG — Very good; it is expected that this material will
adsorb very well relative to the other chemicals
C — Good; it is expected that this material will adsorb
well relative to the other chemicals
F — Fair; it is expected that this material will adsorb,
but not well relative to the other chemicals
P — Poor; it is expected that this material will not
adsorb especially well relative to the other
chemicals
‘ Effectiveness of adsorption control procedures estimated by criteria other
than the static dielectric constant.
Sources: Refs. 2, 7, 8, 9, 10, 11
32
-------
TABLE 4. REPRESENTATiVE HAZARDOUS HIGHLY VOLATILE CHEMICALS
(COffl’OUNDS WiTH BOILING POINTS OF 25°C OR LESS)
No £hcmL . 1 N.ia ,e
PI.yaI .aIVropcrc eu
Bolting Vupur Re) Vapor Pla h
Point Pr .bsiire fists I ty Pol i ,L
(°C) (I’S IA/BO°F) (Air—I) ( (IC, °F)
C I i.mlcaL
I’rornrt lc
Water
Soji,— Pin,i t IuiiaI
bully Gruup
20 19 0 1 5 -58 Yes Aid
—84 (L.a 0 9 Ona No - ne
—33 Gas 0 6 NP Yes Saae
(awl. , ..)
I Acet .ildeliydc (C)
I A . ety1eiie (C)
3 Aaur,.,niu • aultydrona (1.)
6 8 I .LU .1 Icus ,. U.)
s ii tune (C)
6 UutyIu .uie (( .)
2 (IiIu. lute (1,)
8 Lyonug...n ihioride (I .)
9 LLIIai,e (C)
JO RLhyI i.I .Iorlde (C)
It ll.yIcnc U.)
I l I Illyttue lalde (C)
I I II , .,,i Ii , . . (.)
14 lIy.Iro cui c li I .,r I ii ,, (i .)
IS IIydtogi_.i fleoride (C)
16 llydi . . 8 .. .i ‘tultid. (C)
Il IauUot ,e ((.)
IS IaobutyIeu.e (C)
Vapor - — Lout ro I Ptoc,’du,rca
Flurnm. ,— Irritu— Contain— Tem Redaction
bulLy Lion Pui,.ou went IN 2 CO 2 AJ .,o, 1 ,u Ion
4 3 2 N P i0 12 Scull e
4 I — N F< 10 4 — -
I 4 2 N FlU 9 1’ I pal I _4 ,*
— 4 40 I 9 l05 N .,
0 55 2 ( 1 —100 No
— 6 45 1 9 Las flu
—31 100 — NP Yc
Ii 25 — NI ’ No
—89 lOl l I 0 —211 N..
12 25 ii 2 2 —45 No
—1 ) 24 1000 1 II -213 Yea
II 210 IS 0 Yto
— 183 10 1)0 — NP Rea..tu.
—85 lOU — NI’ Ycs
It — — I II . —
-60 IOU I 2 [ ‘IjuwnabI , . Yes
Gas
—12 30 2 0 —117 (to
— 1 27 — —105 (c . .) t O,
— clue
-cu lt
L I
L II, (.1
ci
-tile
i) Ide
A..IJ ( •I
A , .td P
- cii . .
4 1
4 0
4 —
0 4
(Po l sc ,ituiia
Flammable —
2 —
4 0
4 3
I ) —
0 4
0 4
4 —
Flammable -
4 —
0
I
(I
Fncg
I iug
l10
2
1
I ’
I ”
LL.i,.a
4
A)
II
N
F ncg
3x 10 5
i
F i
VLA
—
I
N
F
—
I
I
YF,
N
lies
4
Vl.
Pu , 5
—
4
N
N
1
—
5
—
I
3
4
N
N
fr .2x 10 5
F’i0
-
-
C
VC
— N F iiC5 .4
I II En,.g 3
(coil r I uuu& .tI)
-------
TABLE 4 (continued)
19. Liquelird usiorni gas
(C)
20. L iqorf lnd peiroleom
gas (C)
21 Hetliaiie ((.)
22 fletliyl bromide (C)
£3 PLILyI IiiuiIdc (t.)
£4 Hrnao Iu I orud L II n i to—
motI uni_ (C)
£5 NIL,og..n iLLrunlde (C)
26 Hltrosyl Iilorld (I.)
21. PIi.,sgci c (C)
28 Propane (1.)
29 Propylene (C)
10 SuItor dIoxide (C)
ii 1 richlr,rui loom—
mctliaiit_ (I. .)
.12 Irimr LIyLamI n. (C.)
4 4 N Fiog
- - N Fneg
- 4 Ii Fueg
o o N
o t N Fneg
4 4 N Fncg
Not I Isted YP F neg
F 7z 10 2
3
-2
-2
F I
6x 10 2
Chemical
Boiling
Point
No Urimical Home (°C)
Firystial Properties
Properties
— hazard Ratings —
Control Pruceduces
Vapor
Pressure
(PSiA/80° )
Rd Vapor Flash
0 50bhp Point
(Air—i) (OC. OF)
Warer
Sulu— Fnnrt Ions!
bihIty Groups
Fla na—
bUlLy
Vapor
Irrita—
tion
Poison
Contain—
mCnt
Temp. Reduction
2
GO 2
Adsotprion
—16!
100
—
Fiwemabie
No
—
4
0
0
N
—
gas
- 40
100
-
-156 to
-76
No
-
4
0
0
N
dO 7
-162
100 016 Huusm bIe No
gas
4 33 3 3 PractIcally No
NI ’
—24 iO U ii A .12 (r.c) So!
—41 100 NA N V Ho
2
- 4 1) 0 N P
Br I 3 4 if. YW F neg 2
CI 4 0 2 N Fneg .7
CI•F 0 0 1 N F LU 2
2! 10
—6 50
8 I ll
-42 10 ( 1
—48 lOt)
10 60
—8 ! 17
NA HF
HA HF
- NP
— —156 (cc)
1 5 —161 (cc)
22 HF
36 NP
Reacts
Reac Lu
ReocL
No
No
Sot
No
Id
CI
C l
A, Id
Cl, F
0
0
0
4
4
U
Ca
l .a
Vt . 5
F 5
l ’ s
P
VC
l ’s
3 3 3 1 0 NP Yes AmIne 4 3 3 N lang 2
ii Vinyl chloride (C)
—14 15 2 2 —110 No (.1. —cnn 4 2 2 N F ring .4
F
-------
TABLE 5. REPRESENTATIVE HAZARDOUS VERY VOLATILE CHEMICALS
(COMPOUNDS WITH BOILING POINTS BETWEEN 25 AND 70°C)
I 2 hlhy l._nt_imii.e ( I )
i i iieraait (I .)
14 1—liezeit. (I.)
l5
lb
17
18
(canaL iii . tcd)
56 40 15 1 Yen
69 . 13 30 —1 No
64 4 I i 3 0 —15 Ho
3 1 0 YF. NW
3 3 3 YE, NW
.1 3 3 YE. NW
o 4 4 YE. 11W
4 2 .1 YE. YW
1 2 2 YE. YW
o 0 I 11
3 3 2 YE
4 1 2 YE
AmIne 4 2 2 Yr
AmIne 4 5 4 YE
AmIn o 3 3
- 3 0
—tanc t . R ombom lb Le
I iquid
CN 4 2 4 N
- 3 0 1 YF,NW
- 4 0 I YF, NW
4 1 1 Yr. NW
—
C1 a i a idcaL
Phytaltati Prcp rL1ae
Propertlea
Ilarard Ratinga
Control_P,occ.duros
Boiling
Vapor
Rel.Vapor
Flaeh
Water
Vapor
Point
lb Liatamical Naite (°C)
Pie ’titure D a .nnlty
(PSIA/80°E) (Atr..1)
Point
(OC, °F)
Solo— Fauctionol
butt 7 Groopa
Plait—
bulLy
IrrIta—
Lion
Potoon
Contain—
mont
Tamp Reduction
U I 2
CO 2 Ailnorptluta
I
Aceto,tc (I)
56
4 9
2 LI
4
Yes
Kc .to iaO
2
Acroleln (L)
52
5 I)
1.9
—iS
Yen
Aid ,—eaie
3
Ally1 chloride (I.)
45
7 3
2.6
—20
No
—€uue • CI
6
BromIne (I)
59
4 5
5 5
(IF
No
Nr
5
Carbon dtbuiIlde (I .)
45
7 3
2 6
—22
No
S
6
Utiorotoim (I)
61
4 (5
—
N Y
No
Cl
7
Ulciilt,ruJI Iluoromctliaiac
JO
MA
—
NE
No
Cl. F
8
iilctltylanalite ( I)
56
0 4
2 5
5
Yes
Amine
9
Dletity I tat (icr (I .)
iS
10 0
2 6
—40
SI Iglat
10
I IImciiiyloml, ie (L)
69
.1 3
1 6
2(5
Yea
Ii
I 1—iJlmt.Liayiiiydraz ine
(I)
6J
.1 1
2 0
34
Yes
imeg
1x10 2
C
F tieg
2xIO 2
VI..*
F neg
2x 10 2
Fiteg
Fl u 2
p
F neg
2x10 2
,
Fneg
IIO 2
1
V
F nrg
F ((12
F uacg
4x10 2
F ucg
7510
F org
F /tiIO
(iydnogeii cy ultle (I)
26
1?
0
9
0 (tt..)
Yes
Isoltexatie (L)
60
6 3
—
—20 (cc.)
No
l9opeiilu iie (I)
28
16
—
—70 (cc)
lb
1oo arcnc (I.)
36
ii
2
4
—65 (cc)
No
4 YE
I YE, NW
- Yl, YW
— 4 a a 10 2
F neg 6zl0
F nag halO
F 1010
F neg
F neg
I nag
I’*
pa
VC
p.
P
I ,
F io2
1 .IO
6x 1 0 2
35 io
-------
No ( .hLExlCfll Naa.e
19 tuoprupyl iI r (L)
20. P1 tIiyI a .CtoLo (I.)
21 tielhyI alcohol (I)
22 MLLIy tone .JiIo, I d e (I
DIck lo 1 1 .0 1.0
TABLE 5 (continued)
FI,yalLx1ru p ep —
Botlii .g Vapor Rel.Vapor Flaah
Point Ptexsurc Denelty Point
(°c) (PSIA/80°F) (Au— I) (oc. °F)
Cluethical
Water
Solu— Fonetloi,al
bilicy Croupa
69
60
65
40
— 3 5 —18 SlIght
— 2 8 14 Sol.
2.7 1.1 61 Yes
90 2 9 NP SlIglil.
Hazard KaLIi .ga
Vapor Control Procedures _______
Flacama— irrlta- Contain— Temp. Redu tton
bully lion Poison mont 1.112 CO 2 Adsorplioum
Ester
Al ...
CL
3
3
23 1 . . .,tan.. (I )
24. l—Peuit . .u,e (I)
25. Proplonaldt.Iiyde (I)
26 Pro 1 .y I ..u,e ox [ do (L)
27 SnLI . .m yl 1 . 1ct l Ic (I)
28. 1cLr.il .yd .oLuran (I)
29 VimmyLidene .1.1... ide
— I YF NW
- 1 YP. NW
1 2 IF, NW
2 2 IF, 1W
F n..g
F neg
P uu.g
F umeg
.16 10
31) 12
48 65
34 I I
69 29
66 2)
32 12
F 6x 10 3
l] .1(l
8x10 3
3 I0
2 5 — 51 (cc)
24 0
20 —22
20 —20
- HF
2 5 —4
34 0
No
No
Slight
So 1
Roe.. t a
Sot.
Ho
- 4 0
-one Flaaxnjb [ e
AId 3 2
OxIde 4 3
C i 0 4
1.10cr 3 1
CI ,—eae 3 2
1 IF, NW
IF. NW
2 IF
2 YP,NW
2 IF. NW
2 TV, NW
3 IF
F neg 3x 10 2
F imog F SxIU 2
F umeg F 1x10 2
F neg 3x 10 2
F ueg V 5x 10 3
F neg 8x 10 3
F neg 4xI0
P
F
VC
V Ga
-------
TABLE 6. REPRESENTATiVE HAZARDOUS VOLATILE CHEMICALS (COMPOUNDS
WITH BOILING POINTS BETWEEN 70 AND 125°C)
C I . lea I
bull ing
VOIOL
lb LhowlcaI ibju.... (°C)
lI.ybl ul Irop...rL ILt
I i-opoi L Io u
IIa ar .LK.b
Vapor
Il a . . — hrtca—
Ollity Li . .... I’oioo .
Control Vro . . . .Jiir .u
Vapor Rel Vapor
Pioai.are IhriIolLy
(V$IA/b0° I) (Air1)
Itauli
Point
(t)C °r)
—
Wator
Sulu— Iou..iios,ul
biltry Ci u u 1 . . .
Contain—
.aoni.
R . .d .ntlou
IN 2
CO 2
AduorpLIull
I
A aLI . . mid (I)
jI l l
0
ii
I
I
III
l.a
A . .id
2
1
2
II.
NW
V
2
A . . .tuulLrlIo
(I)
l l2
I
N
I
4
42
l.a
Lyanldo
2
I
2
IV.
NW
I’
n . .
I
A . . .yIonILuIIo
(L)
ii
2
2
I
ll
ii
Y.a
Cya ..td ...
— ( -no
1
.1
1
IV.
NW
V
u.g
4
AIlyl ak.ul.ut
(L)
91
0
94
2
0
90
Y . .a
AL -.,, ..
3
3
2
I I .
NW
F
“°ll 2z 10 2
V
4xl 1J 3
I..
VL
V I.
5
ILn .ne (1)
WO
1.0
2
12 (u-)
No
—
3
1
2
IV.
NW
V i,og
I 3a10 3
I
6
I
IluLyl uI . ..ul,uI (I)
lintyl . .u.In . .. (I)
11W
/8
0 IS
—
2
2
6
S
91
10
l.a
Yea
Al..
Aali .o
3
3
1
—
2
2
IV.
YV,
NW
NW
V nog
1. n..g
S IO ‘
V 4aL0
c
—
N
but yIaId . .hyd .. (I )
13
3 S
2
5
iS
No
AId
3
2
2
YF
NW
F ui..g
4ji10 3
V i. ,
9
La .I,n . L .ImJiIutIh.
(I)
12
2 5
—
NI
No
L I
0
2
4
II.
IN
V u .g
V ;xio
V
10
L .oLoIJ . .I .yU .. (L)
102
0 8
2
4
59
SIlgl.L
Aid
2
.1
3
TI,
NW
F o ..g
Ix 1 1J 3
1’
II
Ly . .IoIa. ...... (I)
WI
2 0
2
9
—4
( ..c)
lb
—
3
1
2
Yb •
NW
V io.
I .bxiO I
II
IblcI.i o.oj, °I”°°- (I)
VI.
I I
3
9
/0
lb
I _I
1
1
3
IV,
IN
F •ic
I 1x 10 1
I —C
I I
IJI..I,IOEOj, 1 O. . . (I)
II
2 I
3
8
93
lb
C I ,—.n..
.1
2
2
IF,
TN
V I .eg
I 4a 10 1
—
14
01 Im.In .Ly I .. . .. (C)
102
0 9
2
8
39
tb
—one
3
0
0
IV
neg
lab
k
IS
I)Ioaano (I)
101
0 1
2
4 )
74
You
—
3
1
3
IF,
14W
F .ieg
V 1a 10 3
P
It.
UIV . . .,t . .. .a (I)
Ni
—
4
1
II I
No
— . .n e
2
—
0
TV
((La
10
I!
LLl . 1 I .u . . . .taL .. (I)
1?
1.9
3
Ii
55
S lIgbL
b.at.c
. 1
1
2
TV.
NW
P nag
4a10 3
V
IN
t.tI.yi .0. ylato (I )
100
0 8
.1
5
44
51 I 5 I .L
Eaton ,—. o
3
3
3
IV,
NW
V .n.g
10
—
19
I tl,yI .,Lol oi (I)
/l l
I 2
1
6
64
You
Al..
3
1
1
YV,
Nil
F ....g
3a 10 3
1.
(. out LI i,ucd )
-------
TABLE 6 (continued)
P P S t0 F-P
— I,
— 2xI0
F l iO
F 7xl0
4 . . io 2
3xI0 3
1 5 10
Cliemiciji
Boiling
Potiit
No Lhi .isicai Nasic (°C)
Physical Fropettie8
Properties
hazard Ratings
Control Pruceduree
Vepor Rct.Vapor
Pressure Density
(PSIA/80°F) (Air—i)
Flash
Point
(OC, °F)
Water
Solu— Functiuiiol
bUlLy Groups
Ftaiuma—
bUlLy
Vapor
LrrIto—
Lion
Poison
Contain—
went
Reduction
UI 2 07 Ad .wrpcion
o i 2 1 150
6 14
I 4 — 34
Ash,ie 3
CI 3
- Iz , sss ablo
20 ELI,yltnt .ilaoii,e (I) Ill
21 Ethiyieiic diehilui idu (I .) 84
22 tthy Iciie gI ycol dimetliyl 85
ethitt (I)
2) Formic acid (I .) 101
24 Ilopiniuc 98
25 i—IIopLc.ie (I) 94
26 hydrogen peroxide 125
27 lu.opro 1 uyt alcohol (I) 82
28 IP—3 (I .) 30 to 160
29. Methyl aerylate (1) 81
30 Methyl ethyl kctouue HI)
(M8K) (L)
31 Methyl Isohutyl ketone 116
(L)
3 3 YV .NIJ
2 3 YI. YW
- YF, NW
O 85
09
13
0 IS
09
41
08
8
Ye
No
Si ighit
Yes
N a
No
Yes
Yes
No
Si iglut
Sal
16 138
:i 5 250(cc)
3 4 25(ce)
—
2 1 65
— 1 10—1 50
26 44
2 5 22
AcId 2 3 3 IF, NW F nog
- 3 0 1 YF .NW Fneg
—cue 3 1 0 IF, NW F neg
— 0 2 I YF ,NW neg
Al ,. 3 1 2 VP, NW Pncg
— 3 1 I YF,NI1 Vneg
— cue 3 3 3 YF. NW F neg
KeLouue 3 I 2 YF, NW F u.eg
12 hleLIuyL mu.tluacrylate (1.)
13 NitrIc acid (L)
34 III tromethiuuue (1)
35 l—0u.Leuue (I )
36 Perch.loroethiyleu .e (L)
Votr lulou , thyLene
101 0 85
89 1 7
101 0 /
IlL 0 03)
121 04
0 42 3 5 75 Slight Ke tono 3 1 1 YF, NW F ueg
—tile
Acid
Vc
P
I.
( , 5
PA
4. ,
Vt.
F,
I .
3
(
50
SlIght
3
.1
3
YF.
NW
F
neg
NA
NP
Sd
0
3
3
VP,
NW
F
uueg
2
1
110
SLight
—
I
—
I
Yr.
MW
F
neg
3
9
/0
No
—cue
1omaobIe
IF,
NW
F
neg
lxi o
2xl0
F i l0
4 I0
HF No -tnt, (A U 1 2 YF, YW F neg F 4 I0
(co u ut l ,u ued)
-------
TABLE 6 (continued)
40 ui-Proi y1 acerae . (L)
41 n—i’ropyl alcohol (I)
42 I’ropylcne dichlorld.
(I) — Olcl I urpcopone
4i Pyc idl,o (L)
41. .Aromothyl IoaJ (I)
45 loluone (I)
4b Ii i(h iOCOoLhanC (I)
41 rrI hiot .ethytene (I )
48 VoIuraI&IIydL (1)
49 VInyl nc tatc (I.)
- 0 No
5 3 NE 0o ocpobes
3
Ph —
3
C I 1
Cl. I
AId 3
Eater, 3
- cue
2 1 YE. NW
- — YI YW
2 YF. NW
2 TV. VU
1 2 YE. VU
I 2 TV, NW
1 2 TV
lie Llicmtcol N.oo..
Phyalcal Pronertlee
Boiling Vupor Rel Vapor
Point P&ca ure Den Lty
(°C) (PSIAI8O°F) (Air—i)
1! Petroleum eLlier (L) 94—140 F
38 Plioaplioioou oxy— 107
chloride (I.)
J9 Phoaphorooa tn—
hlor Edt (I)
Chum I cal
Pronernioc llacard Ratinas
Flash Water
Point Solo— Functional
(OC, °F) bulLy Gloups
NA
NP
TI amma —
Vapor
I rr Ita—
Con Ire I
Cuiitati.- Temp !Uon
bulLy Lion PoIson menu l.N 2
2 6 4 7 NF Decompoaca Cl. P
Adaorpt ion
l s t or
Am
CI
-- 4 — I HI F ,ieg <6x 10 2
LI. P 0 4 1 YE. 11W V hug F
3 - 0 YE. NW V nog NO 4 a 10 i ’s
1 I 2 YE. NW F nog lalO V
I I 2 YE. NW F iieg 2’cIO C
3 1 3 VP. VU uieg lab c
P neg V SalO
97
0
36
2
1
81
Sot
96
I
1
39
10
No
115
04.1
27
68
Sot.
110
—
9
2
—
No
III
06
ii
55
N .
74
NA
No
87
1
5
4
5
9 O(cc) 5
No
103
07
54
No
73
1
4
3
0
23
F neg
F nog
F •,ug
F neg
F neg
7 LO
F 4al0
F 2 l0
t0
V 5xI0
Pr ttI aIIy IlunhI .Immahle
-------
TABLE 7. REPRESENTATIVE HAZARDOUS CHEMICALS WITh INTERMEDIATE VOLATILITY
(COMPOUNDS WITH BOILINC POINTS BETWEEN 125 AND 200°C)
lb. Chemical U une
Boiling Vapor ReI.Vopor
l ’otnt Yre.iaure DniiBity
(°C) (PSL’ ./80°F) (Airl)
Chemical
PI .yalcal Pro,.ertiee F .uperti . .a Hazard Ratlne
Pleat. Water
I’otnt Solu— Fonct (oiiaI
(OC. °F) bulLy Groupe
Vapor
P1a . a— Irrita—
bulLy
Lion Pr ,iaon
Control Proceduree
mont
Contain— Temp Reduction
co 2 AJs . ,rpl to ..
I
A .cLte aul.ydrlde (1)
139
0 73
3 5
1)6
Y a
Anhydrido
2
3
3
IF.
NW
I neg
F l ab 4
2
A ,,yiic acid (L)
[ 41
0 Il
2 S
bib
Yes
Acld,—ono
2
3
2
YF
NW
F . ,eg
I. lab
1
Amyl t .oc (I)
146
013
4 5
9 7(cc)
Slight
PSL ..r
3
—
i
.
liw
nog
jaio
4
Awyi IlL oh ol (I)
biB
0 09
3 0
7l(cc)
Slight
ALohol
3
I
2
YP.
NW
ncg
2aI0
5
AnI I in c (I )
184
0 008
3 2
168
SlIght
Amine
1
1
3
YF
NW
aeg
2xi0
6
Bc ,,z.,Idei .yd. (I)
119
0 1)17
3 7
165
No
——
2
—
2
IF,
MW
neg
2 x 10
I
8cnzuyl ciii,,. tdc (I)
19!
hA
4 9
162
Lkcoapose
(.1
1
4
2
IF,
NW
neg
let0
c—
H
ilc, .nyl &l ,Ioride (1.)
1/9
NA
4 4
15)
No
Cl
2
—
2
IF,
1W
F neg
—S
V /xIO
9
Botyl ,o.etoLe
126
0 28
4 0
88
Slight
.eLer
3
1
2
IF.
119
neg
4a10 4
JO
Ilutyb acryl. .Le (I .)
149
0 12
4.4
118
No
Fater,—ene
2
1
1
IF,
NW
nag
l. .b0
II
hlutyrlc liLid (I)
164
—
.1 0
161
Ye9
A..Id
2
—
2
IF.
NW
I neg
F 5a10 3
12
(.l,lorobeuc,..ne (I)
132
0 28
3 9
9/
No
C l
3
0
2
IF,
I V
F iteg
I 3aI0 4
13
(.reauis (I)
177
0018
3 1
180
No
I ’I,eu,ol
1
2
2
IF,
19
F neg
F ixlO
14
(.uaeu.e (I)
152
0 3)
4 1
iII(Lc)
No
—-
2
1
I
IF,
NW
F iieg
9xl0 5
IS
CycloI .exonou .e (6)
156
0 5
3 i
129
Slight
l’.CtonC
/
3
I
IF.
NW
V neg
F 6zI0 5
16
(.yclol .eayl al .uI ,i .l (6)
161
0.01
3 5
I60
Slight
AI ,
1
1
1
Ii’
.ieg
3d 0
I?
l—Di. n i (I)
ill
0.04
4 8
128
No
—on...
—
—
—
IF
,,eg
F /eIO
18
I)lcI, Ioroi.e,,zo, ,e (I )
181
0 035
5 I
165
No
(.1
1
2
1
II •
IN
P ..og
I. 2x10 5
i9
Dityclo 1 entndlc . ,c (L)
170
0 09
—
90
No
-eec
I
I
2
IF
I . ,eg
F 3x 10
20
l)ietluyl atbonate (I)
12/
0 23
4 I
115
Ho
sL ,r
3
—
2
II’
V neg
F 3z10 6
21
I)ieti .yI glyoi
lb2
0 06
—
158
You
—
-
—
IF
i.eg
Sal0
dimcth.y I ether (I
1.
F
C
6
P
1.
C
la
I—C
I .
(c Out 1iiuic l)
-------
TABLE 7 (continued)
Bolting
Point
No CIteatical N.nac (°C)
Aalotlo . . (4 86 ga
lead/ga I)
ii CloLnriildeliyd .. (1)
14 IICIttnlol (I)
I S Ilextiti.. . I (I )
26 JP—4 (I .)
21 JP—5 (I )
.18 MeolLyl oxide (I)
29 Mineral aplriI (I)
168
I 89
190
I 36
1 (6
(I) Ill
198
1 10
164
Ii i
3 I 2
— — — YE. NW
- -. Y 8 ,NW
- - Y8 .NW
— .1 YE, NW
- 0 -
CI ,e , a lca l
Phxelcal Fronertlee Pros .ertieo hazard Ratlnna
Vapor
Rel.Vapor
Flauh
Water
Vapor
Pr ,aaure
(F ,IA/80°P)
lk.naity
(AIi1)
PoInt
(OC, OF)
Solo—
blilty
E,in LIo iuL
Croops
F1a xn—
bulLy
Irrita—
tio,i Poloon ,aens.
Control Proeo .Iuica
22 Utl ohuLyI ketono (I)
43 t)IoaLhyI but loto (I .)
26 Dodocene (I)
25 I thy II ,i ci .t. (I I
26 Ethyl butinn,1 (I)
2? EIIiylen.. dibromide
28 I tiiyIene glycol (L)
I— 29 l—Etliylexylacrylate (I.)
24 ) Foifurol (1)
ii Eurmuryi .iI._ohoI (I)
24 (,oooh li,uo (I. . )
A . .LoaoLIVc (4 LI ga
lead/go I)
Contain— j ednct Ion
49
o oil
0 0115
0.2
o 03
o 21
o 002
0 05
1112 CO 2 Adaorpttoo
- 140
1. 4 240
5 8 134
09 80
.1 5 128
— 1 W
— 240
- 180
3 3 153
J4 161
No K.t,,,,e 2
Slight —SO 4 I
No —tIle 2
No - 3
No Ale. —
No Br o
Yea Ale. 1
No —cite 2
SlIght AId 2
— Alt 2
— I YE
1. 4 YE. MW
I I YE, NW
2 2 YE. NW
— — YE, NW
I 3 YE, YW
0 1 YE. NW
2 — YE, NW
2 3 YE, NW
- I YF,NW
neg
F nag
F nag
F nag
F nag
V nag
F nag
uieg
F hag
F nag
4x10 5
F 1x 10 5
F halO 5
1x 10 5
h0
V 3aLU 4
F 1x 10 6
Ja10 4
F
Jx 10 5
199
NA
—
-36(cc)
No
-
I ll
NA
.1
4
50(cc)
N..
—
188
NA
3
4
IIA
Yea
Aid
116
0
01
6
0
160
No
Ale.
151
0
5
.1
5
149
SlIght
/tl
1 1 2 Y , NW F ncg 5x 10 6
P
1
(,A
V
L .A
vi ;
L .A
CA
l .A
116 to
28!
116 to
281
I JO
154 to
404
F nag J .tLU
CPI( (,ozbust-
Ible liquid
18 — -lOto
1-4 1. ) (cc)
hA — l40( c)
— 1.4 8?
0 01 3 9 105—140
(i_i )
F nag
V .ueg
F nag
8 lab 5
I lab 5
.3xlU 5
i i ”
No — 1—2
Sllgl,t
No - 2
3 1 1 YE, NW F nag <3x 10 5
— I YE, 14W F ueg F 3x10
F nag 6z1O
( tilil L I ii..cd )
-------
TABLE 7 (contInued)
No Utemical Na .ue
40 MOIm ..LIIJIIUIOIaI,O (L) I 70
eLI .jnol imIuu_
41 PloipliulIuie (L)
42 Non,,iue (I )
43 Nonene (I)
44 0 tano1 L)
45. PI . a .1 (5)
66 Ihuspliorous pei,ta—
chiorid.. (5)
67 lPr 1 IuI Looe (I)
48 Proplonli all (I)
49 Prop Io,uI . aoliydi ide (I)
50 Propylene glywl (L)
SI Styrei .e (I)
52 Sulfur u .o . .IuIorId . (L)
53 lorpent Inc (I)
56 1 —U .d cue (I )
55 Vinyl toluLlIe
HLLIIyI styrene (I)
56 Xylen..s (I)
032 30 100
— 44 88
012 - 78
0 002 4 5 118(c.)
0 016 3.2 185
NA 72 tIP
NA 2 5 165 Sot
009 25 134 Sol
— 45 165 Decomposes
0 002 — 225 Sal
014 36 9) No
0 3 4 7 266 Reocta
015 — 95 No
o 01 S 3 160 No
10) 4 1 1 )7 No
112 to 0 16 3 7 75 to
144 84
Sal t ) 5
I? I, I0
JaIl) 5
L x 10
I0
V 2x 10 4
6a m 5
lu-s
4x 1 11 5
Plivalcal luuec,L1t.a
oiting Vapor
Point Pressure
(°C) (PSIA/80°F)
Ret.Vopor Flash Water
Density Poluit
(Air—i) (o&, °V)
C I a Ic a I
Properties hazard Ratluigu
Solo— Functional
bUSty Cioupa
Vapor
Floama— Irrita—
bUlly Lion Poison went
0 01 1 1 200 S l ii 1,,e,j Ic I 2 2 VI • NW F neg I 4aI0
Control
128
151
140
195
182
162
(Sob)
Contain— Teap Redu t!un
UI , Adsorption
C
1 55
141
I (.9
187
145
1 J8
155
193
I 68
Sot
aislns
S
I
I
YF, 11W
V neg
3 ..l0
F
No
3
—
0
YF.IIW
P iieg
F 9a10
a
No
—cue
3
I
0
IF NW
F ueg
mat0
i’a
No
ak.
2
—
I
YE, NW
F flog
P 9x 10
6
Yea
ie.nol
1
2
3
VP, NW
I’ nog
F 2a 1()
P—G
NA
Ci• P
NA
NA
NA
—
F
V
-
ester
2
—
0
Yr. NW
V nag
acid
2
2
2
YF, NW
P neg
P—F
a.idi
2
—
2
YE. NW
F iu.g
au.
1
0
0
YF, NW
uleg
C
- . .n. .
3
2
2
YF. NW
F nag
I
(1
I
4
6
VI, tiN
F nog
—
3
1
I
YF .NW
neg
-aic
NL
NI
NL
VP. NW
F ncg
- c i . ..
2
2
1
TV, tIN
F uueg
‘ .
lb -
3 I 2 IF, Ni l F nag F03x 10 4 P
-------
TABLE 8. REPRESENTATIVE HAZARDOUS CHEMICALS WITH LOW VOLATILITY (COMPOUNDS
WITH BOILING POINTS ABOVE 200°C) AND MISCELLANEOUS COMPOUNDS
I A,er,, ..c cy..uiotiydi I . . (I) Occ,sn—
post a
2 Ac... C 1 1 1 1t iu ilC (I.)
I AJIpo,,iLrIie (I)
4 Aioso . .los hydroald .. (I.)
5 Cli Lorol .ydr his (I) NP
6 ii-Uetyl alcohols (I) 230
7 DIbotyl philhialote (1.) 335
C.-
8 DlInolos .Iue (1) 268
9 h)IeLh .yle, .o glycol (I) 245
LU l)lLLIlyIeuttriamiuo (I) 207
II Et h iy I tue cyouoh .ydrli . (I) 2.10
12 2 — .th .y I hazy I a I tiihiol (L) 355
I.) Formaidah.yje UP
5OIljttl)fl (I)
16 IIzameLIyien .d lualne (I.) 202
IS hItxo tliyleiic— NP
Letrasine (S)
16 li’—I (keroseut) (I)
17 Ksroo.nc (L)
202 NA 4 1 1110 Flu KnLo, .e
290 NA 3 7 199 No Cyanide
NP NA NP HF Ynu Base
( oaf at)
- 92
5 5 180
NA j55
- 305
— 255
- 210
— 265
4 4 185
[ .0 1 82 (cc)
Slight
Cl
No
Mc.
Ho
.BIer
Yea
A. hi.e
Yes
Ale
Yes
Atahuic
Yes
Lri ,AI-
SlIght
Aic
Yes
Aid.
0.004 NA 160 Yes Aisine
NP NA 482 Ye Aslee
— lOl)(cu) No
— IOO(tc) No
1 1 4 YE. NW V nag VC°
2 — I TV , 1 1W
1 1 3 YE, NW
- - - YE, NW
3 3 4 YE
I 0 0 YE
1 0 0 TV, MW
1 2 2 YE, NW
1 0 1 YE. NW
1 2 2 YE
I 0 2 YE
I I 1 YE. NW
2 3 3 YE, NW
I 1 3 YE, NW P neg F 5xIO
— — — YE. NW F O n g F
2 I I TV, NW V lies UC5
2 I I TV, NW F ring 6
Clicatcal
Boiling
Point
lb U,eeicol House (°C)
Physical EroperLies
I ’nopertheu
— hazard Ratiuugs
Control Procedures
Vapor
Pressure
(PSLA/80°V)
Rel.Vapor FlaoIi
De.islcy PuJiir
(Alr .l) (OC, °F)
Water
Solu— Functional
bulLy Croupa
Vapor
Elauona— unto-
bILIy Lion
Puma..
Contain—
Dent
TeSp. Reduction
2
CO 2 Adsorption
0 (109 2 9 IAS(tt) Yes Cyo,lid
0 16
0 (3001
NA
NA
o ot
0.001
0 006
0.06
I ;
Vt..
V nag
nag
V utg
F neg
F uicg
F 114.5
P neg
neg
F lutlO 6
41108
106
F nag
F 2xi0
I 7z10 7
5 1 . 10_ S
11106
260
200 LO
260
(is
05
C.
— 1mg
F — vc
IN N. .phLi.u 14.110 (S)
19 tbun,.noi (I)
20 U 0 1 , 7 I pi,o ..u I (1)
1 18 NA
113 NA
304 NA
ii OIt,,e (I) 1)errouspoaa.. NP
44 190 No
50 210 No
— .300 SlIght
Al..
I Iituio I
I I I Sill Acid
I 2 2 YE, YW
1 0 0 YV,NW
1 I 1 YE, NW
1) 4 3 YE, NW
F neg
F i,eg
F aug
F 3sL0
F 3zI0
31108
P
C c , . , . I ,.u,cd)
-------
TABLE 8 (continued)
22 I’orulore .ildeliydc (5) Uoco puscs
23 Piio plioi un I, it a— 514
uuUide (5)
24 loll oil (I) NA
25 Ietradcioiioi (I ) 26J
2 1 i I— ieLr.l ,ie (I) 251
27 let ol.ydr mopliL ha 1 iie (1.) 208
28 Toloeno 2 4— 250
.111 aOCyOioite (I
29 it iijiiorol,eiiziiic (I ) 213
JO Irtde. eiie (L) 23 )
31. Tridciyl alcohol (I.) 274
— lildeci mnol
Tr1ctIiunoIawlii (I.) lk.coaiposes
Ii lethy lenc—
letro ioli,e (L)
- 199 Slight Aid
- Flaismobie geacts P
- 6 3 210 Ue uinposes CI
001 - 115 No
NA 69 250 No
NA 5 2 375 Sol Awlmie
NA 5 0 290 Yes A [ ne
— — — Nc,L I Isted — — — — VI P F
2 YE, NW I. ntg F n g
NA NA
- 0
— 0
3 6
— — — Not L1ste d — - —
1 0 0
a i v. Nw r neg F 1x10 2
1 2 I YE Etiog 2 10
Chiesilta I
Boil lag
Point
No Cliesical None (°C)
Physi cal Pr per tiL s
Propert1e __
— Ila,.iid Rnt u
Control PFOLCd ,. __
T doctlon
IN 7 CO 2 Adsorption
Vapor
Pressure
(PSIA/80°I.)
Rd V.mpor
I)enslty
(Air ’.))
Flash
Point
(OL. Op)
Water
Solo- Fonctiommul
bulLy Croops
llonssa—
bility
Vapor
Irrito-
Lion Potsoii
Contain-
meat
N I ’
NI’
NA
NA
NA
0 012
.4)01
NA
14 2115
6 8 130(cc)
46 190
6 2 270
AlL
—tile
NA
No
No
Ho
No
NA
32
33
—eli”
AL
YE,
NW
P
neg
I. islO
Yl,
11W
F
mii
I. SalO
YE,
NW
F
n g
F 4xI0
VI.
F
neg
8x 10’
YF,
YW
F
ui g
F mug
F-i’
YE,
NW
F
neg
2xIO
P
YE,
NW
I.
mug
F 2uulO
Ck
2/7
-------
SECTION 6
PROPOSED VOLATILE SPILL SUPPRESSION CONCEPTS
In Section 4, current field methods for controlling the volatilization
from landborne spills of hazardous materials were discussed. The conclusions
of Section 4 were that vapor source removal, water flooding, and controlled
ignition were effective means of vapor control, when circumstances were favor-
able. Section 5 presented a brief technical discussion of the liquid vapori-
zation process and an analysis of a simple model for vaporization from a spill.
It was determined that temperature reduction, sorption, and vapor barriers were
three possible means by which the spill vaporization rate could be controlled.
This section discusses possible alternative approaches to the application of
promising vapor control techniques. The concepts described herein were de-
vised from existing technology in other areas of engineering.
There are basic issues not explicitly addressed in these concepts,
including: -
1. Personnel Safety — — host of the systems described allow for person-
nel to be at some distance from the spilled material, thus decreas-
ing their exposure risks. The use of safety equipment by personnel
who may be exposed to hazardous materials is certainly recommended,
but it was not within the scope of this program to examine personnel
safety per se.
2. Availability —— The problem of equipment availability will always
exist in a resource—limited world. Battelle did not study the dis-
tribution of equipment needed across the U.S. , since the goal of
this program was to examine technology, not economics.
3. Time —— The concepts presented cover a range of time scales required
for application, from several hours (for CO 2 application) to several
days (energy transfer concepts).
4. Equipment Loss —— 1hen dealing with spills of hazardous materials,
the equipment used should be considered expendable. If the spill
does present a hazard, the attitude of spill control personnel is
that the sacrifice of equipment is a better choice than the sacri-
fice of humans. For this reason, Battelle did not examine the ef-
fects of corrosion, etc., on the conceptual equipment.
Table 9 contains a list of the plausible control methods, many of which
were generated in idea sessions. The methods have been organized under major
4)
-------
headings or concepts which emphasize the similarities in the overall approach.
The table also serves as a general outline for this section.
TABLE 9. CONCEPTUAL TECHNOLOGIES
DIKING/EARTHMOVING CONCEPTS
a. Manual Construction of Dikes and Ditches
b. Earthmoving Equipment (Bulldozers, etc.) for Construction of Dikes
or Trench—Digging
c. Standard Earthmoving, Agricultural, or Specially Constructed Plows
to Mix the Hazardous Chemicals With the Underlying Soil
d. Remote—Control or Wire—Guided Systems to Construct Dikes or Dig
Trenches
e. Oil Booms as Dikes
f. Foam Barriers as Dikes
g. Explosives to Form Dikes
PREFORMED COVERS DEPLOYMENT CONCEPTS
a. Parachutes as Preformed Covers for Spills
b. Deployment of Blankets or Tarpaulins on the Spill by Means of
Parachutes
c. Spring—Loaded Tarpaulins That Would Unfold on Landing in the Spills
d. Tarpaulins or Blankets Deployed While Falling by Means of Rockets
e. Deployment of Parachutes in an Upside—Down Fashion by Means of Heli-
copter Dowuwash
E. Cannon Arrangement to Fire a Folded Cover, Weighted Along its Edges
in Such a Manner That it Would Rotate in Flight and Centrifugally
Deploy to its Full Size
g. Drag Paract-iutes or Tarpaulins Over the Spill Either Manually or with
Manipulators or Cranes
h. Fire Navy Line Guns Over the Spill to Provide a Means of Pulling a
Tarpaulin or Cover Across a Spill
i. Tarpaulin Deployment by Fixing One Side to the Ground and Dragging it
Over the Spill by Means of Rockets or Helicopters
j. Parachutes or Tarpaulin Placement Over Spills While Directing Air
Blasts Under or Over the Parachutes or Tarpaulins to Levitate Them
k. Geodesic Dome Constructed Over the Spill
1. Inflate a Balloon Over the Spill
(continued)
46
-------
TABLE 9 (continued)
SKIN-TYPE COVERS DEPLOY €NT CONCEPTS
a. Spray Chemicals That Would Polymerize and Form Covers on the Spill
b. Spray Foam Onto the Spill
c. Form a Cover (e.g., Nylon) at the Site
d. Preformed Film Overlay System
LARGE OBJECT DELIVERY CONCEPTS
a. Manual Deployment
b. Catapult to Deliver Large Objects Into the Spill
c. Four—Bar Linkage to Deliver Large Objects
d. Air Cannon to Deliver Large Objects
e. Explosive Cannon to Deliver Large Objects
f. Portable Conveyor to Deliver Large Objects
g. Helicopter or Airplane to Drop Large Objects Onto a Spill
h. Device Similar to an Agricultural Bale Thrower to Deliver Large
Objects to Spills
i. Earthmoving Eguipment Eor Dumping Large Objects on a Spill
SMALL OBJECT DELIVERY CONCEPTS
a. Standard Straw or Silage Blower to Disperse Small Objects
b. Small—Scale Blower Modeled After a Straw or Silage Blower (with Less
Range and a Smaller Delivery Rate)
c. Air Cannon Delivery of a Quantity of Small Objects or Particles Con-
tained in a Canister Which Will Release the Objects at a Certain Dis-
tance or Time
d. Explosive Cannon Delivery of a Quantity of Small Objects or Particles
Contained in a Canister Which Will Release the Objects at a Certain
Distance or Time
e. Catapult Delivery of a Quantity of Small Objects or Particles Con-
tained in a Canister Which Will Release the Objects at a Certain Dis-
tance or Time
f. Conventional Crop Dusting Airplanes to Dispense Small Objects or
Particles
g. Dispersal of Small Objects with a Centrifugal Pump
h. Dispersal of Small Objects or Particles with a Sandblasting—Type Pump
Arrangement
(continued)
47
-------
TABLE 9 (continued)
i. Dispersal of Small Objects or Particles With a Centrifugal Slinger
Similar to the Type Used on Sanding Trucks
SLURRY TRANSFER CONCEPTS
a. A Standard Hydro—Seeder to Disperse a Slurry
b. Use of “Slippery Water” Component in the Slurry to Increase the Spray
Range of Conventional Water—Hose—Based Pumping Systems
c. Utilize Existing Fire Truck Pumps and Hoses to Pump Slurries
d. Deploy a Slurry Transfer Hose Over the Spill with a Long Boom, Either
a Crane or a Specially Constructed Boom
LIQUID TRANSFER CONCEPTS
a. Pumps and Liquids From Available Fire Trucks for Pumping Liquids
b. Slippery Water to Extend the Range of Conventional Water—Based Pump-
ing Systems
c. Deployment of a Liquid—Dispensing Hose Nozzle with a Balloon
d. Deployment of a Hose and Nozzle with a Supporting Boom, Either a Com-
mercial Crane or with a Specially Constructed Boom
ENERGY TRANSFER CONCEPTS
a. Flame—Throwers to Heat or Ignite a Spill
b. Infrared Lights with Parabolic Reflectors to Heat a Spill
c. Solar Energy Lenses to Increase the Temperature of the Chemical Spill
d. Sonic Disturbance or Blast
e. Increase the Temperature of the Spill Via a Standard Heat Exchanger
System
f. 4icrowave Transmitters
g. Incendiary Bombs to Increase the Temperature of the Spill
h. Decrease the Temperature of the Spill Via Standard Cooling Equipment
j. Air Curtain Destructor Devices Used to Burn the Hazardous Chemical
in a Controlled Situation
DIKING/EARTHMOVLNG CONCEPTS
One of the more effective techniques for controlling volatilization is to
capture the spilled material in pools and pump it into closed containers.
Since spills will generally spread over a wide area, tools to alter the contour
48
-------
of the terrain would be useful for spill containment. As an alternative to
capture, burying the spill in the soil is also effective in decreasing the
vaporization rate.
Manual Construction of Dikes and Ditches
This operation is a simple and common—sense treatment for the less dan-
gerous spills. Workers would dike the spill or dig drainage ditches with
shovels in order to contain the spill. Their tools might possibly be made of
metals or of plastics which would reduce the risk of causing sparks when hit-
ting obstructions.
It is estimated that the development of a set of prototype tools would
cost approximately $20,000 over a period of four months.* Protective gear for
the exposed workers would be necessary.
Earthmoving Equipment (Bulldozers, etc.) for Dike and Trench Forming
Earthmoving equipment could be used to construct dikes or dig trenches.
The equipment’s blades could be fabricated of a metal or plastic which would
reduce the sparking hazard. The engines could be equipped with spark—
arresting mufflers. The use of diesel engines (compared to gasoline engines)
is recommended because of their lower operating temperatures. The operator’s
cab should be fitted with an auxiliary breathing system such as is presently
used by fire and rescue squads.
A representative vehicle is the Caterpillar 08K crawler tractor shown in
Figure 2. Equipped with a stock bulldozer blade and adjustable ripper, this
machine costs $160,000. It is estimated that the total cost for a model modi-
fied as described above would be $200,000.
Spill Burial or Mixture With Underlying Soil
Standard earthmoving or agricultural rippers, plows, discs, or harrows
might be used to bury the hazardous chemical mixtures within the underlying
soil. The working depth for such tools may range from about 4 in. for the
case of an agricultural harrow to 72 in. for the case of a single shank rip-
per mounted on a large crawler tractor.
The blades of such devices could be fitted with wearing surfaces made of
nonsparking materials, such as beryllium copper, to replace the stock steel
parts. This would give safer operation during spill coverage operations.
* Each cost estimate for a prototype system was calculated by obtaining five
independent cost estimates from experienced researchers, eliminating the
highest and lowest estimate, and averaging the remaining three estimates
(rounded to the nearest $1000).
49
-------
Figure 2. Caterpillar D8K crawler tractor.
- (Source: Caterpillar advertisement)
The cost of a typical 11—ft—wide harrow is $2600.* The cost of a 40—hp
diesel tractor required to pull such a harrow is $8200. The price of a large
ripper—mounted crawler is $160,000.
Remote—Control or Wire—Guided Systems
Because of possible hazards of a toxic or explosive atmosphere at a spill
site, it may be desirable to operate diking or trenching equipment by remote
control. This would obviously lessen the risk to the equipment operator.
In the last several years the remote control of increasingly larger and
more complex equipment has been achieved. The objective has been to remove
the operator from a hostile environment, while still retaining most of the
machine’s performance. A typical system is shown in Figure 3, which is a
remote—controlled Bobcat excavator.
It is estimated that a crawler such as the Caterpillar D8K could be pur-
chased and fitted with a Moog radio control for a total cost of $260,000.
Oil Booms as Dikes
A number of firms have developed inflatable booms for containing oil
spills on waterways. Such booms could be used at land spill sites. Water
might be used instead of air to inflate the booms in order to form a sturdier
barrier and to help conform to terrain irregularities.
* Available commercial equipment is currently described in English units of
measurement; where applicable,this practice will be followed in this report.
r :•. -:
50
-------
Figure 3. Remote—controlled Bobcat excavator.
(Reproduced by permission from SAE)
I.
r, •.
I- .
(f _-
.; ,
51
-------
A typical oil boom is made up of sections of inflatable rubber bladders.
The bladders may also be filled with water. A section of oil boom is typically
50 ft long.
Foam Barriers as Dikes
A foam dispenser might be used to quickly erect a barrier around a pollu-
tion source as depicted in Figure 4. The foam would rapidly stiffen and form
a barrier or dike which would contain the spill.
A stock Hydro_Seeder* might be used as—is or be modified to suit this
purpose. The cost of such a unit with a 3000—gal capacity is approximately
$12,000.
Explosives to Form Dikes
An explosive charge might be used under certain conditions to form a dike
to contain the spill. This would be possible when nearby personnel and Struc-
tures would not be affected by the consequences of the blast. The barrier
erected by the charge would not be impervious, but would slow the spread and
hence the evaporation of the spill.
PRE-FORNED COVERS DEPLOYNENT CONCEPTS
In this group, and in the group that follows, the principal concept in-
volves the placement of a vapor barrier -over the spill so as to reduce volatil-
ity and introduction of the hazardous vapors into the atmosphere.
These concepts are viewed as rapid response techniques, since air deliv-
ery to even the most remote areas can occur quickly. Rapid setup of the covers
is also envisioned.
* Trade name, Bowie Industries, Inc.
Figure 4. Foam dike deployment concept.
52
-------
Parachutes as Spill Covers
Parachutes could be dropped directly onto the pollution source to serve
as covers.
Parachutes used as pollution covers may have several disadvantages.
Stock parachutes are made of thin nylon material with about one—third of the
weight used in denim jeans. The value of this fabric as a pollution cover is
questionable. The use of heavier nonporous fabrics may be possible, but air
vents would probably need to be provided in the canooy for proper air flow dur-
ing descent, thus degrading the covering ability of the parachute when in po-
sition. Parachutes need to fall at a rate of about 20 ft/sec to font properly.
When the parachute’s canopy tends to collapse as the tension goes out of the
lines, which are suspended from the canopy, the uniform shape of the canopy
disappears and the canopy may land somewhat unpredictably and in a relatively
small heap.
Parachutes are seldom built in diameters over 100 ft. Such parachutes
require a 3000—lb ballast weight to form properly. The cost for an ordinary
100—ft—diameter nylon parachute is typically about $300.
Blankets or Tarpaulins Deployed by Parachutes
Blankets or tarpaulins could be dropped on the spill by means of para-
chutes. While delivery could be made in remote locations in an expeditious
manner, it would probably be necessary to at least partially spread out the
tarpaulins after landing, either manually or by means of devices such as
grapple hooks, or by spring—loaded mechanisms contained in the tarpaulin
package.
A typical 20—ft x 20—ft chemical—resistant tarpaulin will cost about $400
and weigh about 35 lbs. A suitable parachute would probably be about 16 ft in
diameter and cost about $800. -
Spring—Loaded Covers
Tarpaulins could be designed to unfold automatically after being dropped
or thrown onto the spill, as shown in Figure 5. A typical scheme would use
steel springs fastened diagonally across the tarpaulin. The springs would
unfold the tarpaulin when a series of restraining straps or bands for holding
the package together were released. Similar schemes have been used success-
fully for the compact storage and deployment of underwater nets.
A typical prototype chemical—resistance, spring—loaded 20—ft x 20—ft tar-
paulin would cost about $8000 to develop and $1000 to produce in quantity.
Rocket Deployment of Falling Covers
A tarpaulin or blanket could be dropped from a height and deployed by
means of rockets while falling.
53
-------
Figure 3. Spring—loaded tarpaulin deployment concept.
Inasmuch as this concept is rather novel, it would not be easy to trans-
fer existing technology to make it workable. Significant problems to be
solved would include the identification of the most expeditious way of using
rockets to open the tarpaulin or cover, the point of descent at which the
rockets are to be fired, the avoidance of fire hazards due to the use of rock-
ets near volatile spills, and a means o ensuring that the tarpaulin or blanket
would be deployed in a safe and predictable manner.
54
-------
Spill—Cover Deployment by Helicopter Downwash
Parachutes might be deployed in an upside—down fashion by means of a hel-
icopter’s downuash. The downwash would serve to open the parachute and main-
tain its shape. An air velocity on the order of 20 ft/sec would be required
to deploy a conventional parachute. A full—size parachute for a UH—l heli-
copter would be about 100 ft in diameter and would be suspended about 200 ft
below the helicopter, as shown in Figure 6.
As stated previously, conventional parachutes are made of thin nylon
material, which might not be effective for covering spills. If a special par-
achute were designed for this purpose, vent holes would probably need to be
cut in the canopy to allow air to be bled through for proper inflation of the
canopy.
A suitably—sized parachute of this type would cost approximately $6000.
Spill—Cover Deployment by Special Cannon
A folded cover fired from a cannon may present several problems. The
desired range of such a system would probably not exceed 500 ft. There are
no known commercially available cannon arrangements suitable for use with
this concept. The most applicable mortar currently in the inventory of the
U.S. Army, the 4.2—in., has a minimum effective range of 2750 ft. The prob-
lem is therefore to develop both a folded cover assembly that deploys in
flight and a special launcher for it. Two of the more promising concepts
might be based upon a pneumatic cannon used in circus Stunts or an impulse—
rocket such as the U.S. Army’s M72 Light Anti—Tank Weapon. The folded cover
could be weighted along its edges and ejected from the cannon in such a manner
that it would rotate in flight and centrifugally deploy to its full size.
Development costs for a prototype system have been estimated at approxi-
mately $200,000 over a period of 12 months.
Manual and Mechanical Manipulator Deployment of Spill Covers
Paracnutes or tarpaulins that have been placed near the spill could be
dragged across the spill either manually or by using devices such as cranes.
It migr t be desirable to attach a simple claw at the end of a hydraulic crane
mast in order to place the tarpaulin in position.
A typical 35—ton—capacity hydraulic crane has a horizontal reach of 150
ft and sells for approximately $170,000. A simple manipulator claw could be
added to the boom for approximately $20,000.
Spill—Cover Deployment With Navy Line Guns
A line gun could be used to throw lines across the spill which would
enable a tarpaulin or cover to be pulled across the spill. The U.S. Navy and
Coast Guard use line guns in naval operations. There are small commercial
versions also available, which can throw lines distances of about 300 ft, as
shown in Figure 7.
55
-------
Figure 6. Upside—down parachute deployment concept.
56
-------
LINE-PAK’
Such small line guns cost about $300.
Spill Cover Deployment by Rocket or Helicopter
A tarpaulin could be placed in position over the spill by means of heli-
copters or rockets. One side of the tarpaulin would be fastened to the ground
along one side of the spill to stabilize it during positioning. Once in place,
helicopters or rockets would be used to drag the tarpaulin over the spill.
The use of helicopters is probably the more practical approach since they
would probably be more controllable and stable in positioning the tarpaulin
than rockets would be. Contact with commercial fireworks manufacturers indi-
cates that suitable rockets would need to be specially developed for such a
task. It is not known to what extent a developmental program for such a task
would need to be pursued.
Since helicopters are normally in the inventory of local and state auth-
orities, the major cost for this system would be only for that of a suitable
tarpaulin. The cost is typically $800 for a 20—ft 2 chemically—resistant
tarpaulin.
LINE
CUTTER
FIRING KNOB
Figure 7. Commercial line—throwing gun.
57
-------
Use of Air Blasts to Aid in Deployment of Spill Covers
Parachutes or tarpaulins could be dragged into position with the help of
air blasts directed over or underneath them in order to levitate them, as
shown in Figure 8. Tarpaulins equipped with pull ropes would be required,
along with fans to levitate the tarpaulins. If four 40,000—cfm fans, each
powered by a 120—hp diesel engine, are required, the total cost of the system
would be about $85,000.
Spill Containment by Structural Enclosure
A geodesic dome or similar prefabricated building might be erected over a
spill. A commonly available 22—ft x 60—ft inflatable plastic greenhouse is
available. It consists of plastic sheeting placed over a supporting frame-
work. The cost of this simple enclosure is about $2500.
Inflatable Spill Cover
A balloon—like structure could be inflated over a spill. An inflatable
plastic cover measuring 22 ft x 60 ft is available from commercial sources.
It requires a small air blower running continuously to keep it inflated and
costs about $2500.
SKIN—TYPE COVERS DEPLOY NT CONCEPTS
Skin—type covers consist of a thin’layer or sheet of relatively nonporous
material that can be placed over or formed on top of the spill to reduce vola-
tility. These concepts are probably in the intermediate time range, upwards
of one day after the spill.
In—Situ Formation of Polymer Film
A spraying system could be devised to deliver a poLymerizing film—forming
cover over the spill. A Likely candidate for such a spraying system is the
Hydro_Mulcher* system which has been used for functions as diverse as mulch-
ing, tree seeding, as demonstrated in Figure 9, and flame—throwing. It can
deliver up to 3000 gal of liquid at a range of up to 200 ft in about 20
minutes.
A 3000—gal stock Hydro— 1ulcher system suitable for mounting on a truck
would cost about $12,000.
Spill Coverage by Foam
Foam could be sprayed directly onto the spill. It would need to be
cohesive and light enough to form a solid layer on top of the spill. In addi-
tion, the foam would need to be capable of resisting permeation by the spill
vapor over a period of time.
* Trade name, Bowie Industries, Inc.
58
-------
FOLD F I)
rARP
SPILL
U i
Figure 8. Levitating tarpaulin concept.
-------
Figure 9. Commercial Hydro-Muicher,
(Source: Advertisement)
A typical foam—spraying apparatus, the Bliss Jet—X—2 system, can deliver
up to 15,000 gal of water—based foam per minute and requires 42 gpm of water
at 100 psi. The unit weighs 50 lbs and is 2.2 in. x 22 in. x 34 in. A Bliss
Jet—X—2 system costs approximately $630.
On—Site Spill Cover Formation
A cover could be formed at the site of the spill. The most expeditious
method of doing this would probably be to form the cover directly over the
spill. A similar system has been developed for the stabilization of silt on
lake and ocean beds at diver work sites, as shown in Figure 10. A similar but
longer system suitable for deployment by a telescoping crane or similar de-
vice, as shown in Figure 11, could probably be developed for about $100,000.
Use of Preformed Films
A spool of preformed film could be unrolled over a spill area by a suit-
able mechanical device. A typical deployment scheme might involve use of a
standard earthmoving backhoe modified to accept rolls of preformed film simi-
lar to that shown in Figure 12.
A prototype system could be developed in about 8 months at a cost of ap-
proximately $75,000.
60
-------
Figure 10. Silt stabilization polymer applicator.
(Source: Battelle)
61
-
l a
-------
FILM FORMING NOZZLE
Figure 11.
Conceptual technique for forming and
deploying a cover over the spill.
Figure 12.
RESERVOIRS
PUMPS
HYDRAULIC BOOM
SPILL
A$ —
Deployment of a preformed film.
62
-------
LARGE OBJECT DEL LVERY CONCEPTS
One of the more promising vaporization control concepts is that of spill
temperature reduction. The temperature reduction could be accomplished by
adding liquid nitrogen or dry ice to the spill.
Dry ice is commercially available in 50—lb blocks. It is desirable to
examine means to safely emplace these blocks in the spilled material. Other
forms of dry ice frequently commercially available are 5—lb disks and pellets.
Dry ice production facilities can be found in almost every city
Similarly, bales of straw (80 lb each) and bags of charcoal (25 ib)
could be lofted into the spill as sorbents.
? tanual Deployment
A large object such as a cake of dry ice could be simply thrown into the
spill manually. However, the object would have to be safe for manual handling
and be of manageable size and weight. It is also assumed that it would be
safe for the individual to get close enough to the spill to deliver the object.
Since there should be personnel at the spill site, the cost of this con-
cept would be minimal.
Catapult Devices
A catapult could be used to deliver large objects into spills. Catap 4ts
are typically used at shooting ranges for hurling skeet targets. Typical tar-
gets weigh about 1 lb.
An investigation did not find a longer, more suitable catapult. A com-
pressed-gas-powered catapult,as shown in Figure 13, could probably be developed
for large object delivery. Spring—powered catapults could also be developed.
The developmental costs for a prototype system have been estimated to be ap-
proximately S58,000 over a period of 6 months.
Figure 13.
0
-‘o
. _._O
0
I • 0 0
AIR TANK
AIR ACTUATED
CATAPULT SPILL
Catapult delivery concept.
63
-------
Four-Bar Linkage Devices
A mechanism using a four—bar linkage arrangement could be used for con-
tinuous delivery of moderately large objects into spills. Figure 14 shows
how the four—bar Linkage could be arranged for significant ranges. Investiga-
tion has shoc n that such four—bar linkage mechanisms are not generally commer-
cially available.
A prototype system could be developed in about 6 months at a cost of ap-
proximately $53,000.
Air Cannon Devices
An air cannon could be used to deliver large objects onto spills. Air
cannons are not generally available on the commercial market. The same prin-
ciple was used in World War I for short—range mortars and is occasionally used
today in “human cannonball” acts at circuses.
Developmental costs for a prototype system have been estimated at approx-
imately $73,000 over a period of 8 months.
Figure 14. Four—bar linkage concept.
ELECTRIC
MOTOR
//
64
-------
Explosive Cannon Devices
-An explosive cannon could be used to deliver large objects onto spills.
The largest mortar generally used by the U.S. Arniy has a 4.2—in, bore. Simi-
larly, the largest cannon has a 175—mm (6.9—in.) bore. It is doubtful that
these bore sizes are large enough to warrant a serious consideration for this
proposed use. Both of these weapons normally are used at ranges of several
miles.
A prototype short—range explosive cannon for firing large objects could
probably be developed in about 8 months at a cost of approximately $97,000.
Portable Conveyors
A portable conveyor, similar to agricultural or industrial types used to
transport grain, crushed rock, or hay bales, could be used to deliver large
objects into spills. A typical 56—ft conveyor can have a reach of 35 ft be-
yond its towing wheels and be self—propelled.
Typical cost for such a unit is $11,000.
Helicopters or Airplanes
Large objects could be dropped on spills by means of helicopters or air-
planes. Helicopters or airplanes would probably be available to pollution
authorities. The U1-i—l “Huey” Iroquois helicopter, in the inventory of many
National Guard units, has a payload capability of 4000 lb. A helicopter could
carry onboard a dumpable container full of large objects that could be dropped
onto the spill.
Developmental costs for a prototype system attachable to the helicopter
have been estimated at approximately $48,000 over a period of 6 months.
Devices Similar to Agricultural Bale—Throwers
A device similar to a bale—thrower, used on agricultural hay balers to
throw hay bales onto wagons, could be used to hurl large objects onto chemical
spills (Figure 15). A stack bale thrower can toss 80—lb hay bales a distance
of 20 ft or more. A similar device could be developed for increased range.
A stack bale—thrower costs $1100. Some modifications would be required
for mounting it on a skid or truck bed.
Earthtnoving Equipment
Large objects could be dropped on spills by using earthmoving equipment.
A number of earthmoving machines might be used for this purpose. Probably the
best choice would be a wheel loader. Such a machine could travel into spills
and dump its load where desired. It should be suitably equipped with a spark—
arresting muffler, tires than can withstand contact with the spilled liquid,
and an operator life support system (Figure 16).
65
-------
L’1
I
‘ ‘...- V
: ‘ 1 St •; •
4 r S IP: P• ‘ . . 1 ’s
Figure 15. Commercial stack bale—thrower.
(Source: A.dvertisement)
66
-------
Figure 16. Commercial wheel loader.
Source: Advertisement)
The price of a typical wheel loader with a 2.5—yd 3 (23,000—ib) capacity
bucket is about $60,000.
SMALL OBJECT DELIVERY CONCEPTS
These concepts were defined for use in spreading dry ice pellets, sand,
straw, charcoal, and other small items onto the spill.
Standard Straw or Silage Blowers
A centrifugal fan, such as a forage blower used for filling agricultural
silos, could be used to disperse small objects onto spills. A typical large—
capacity agricultural forage blower can blow material at rates up to 150 tons!
hr (5000 lb/mm). Particle velocities would approach 130 ft/s. Relatively
dense particles could probably travel 100 or 200 ft from the blower’s outlet
(Figure 17).
j.t....
..
:
67
-------
Figure 17. Cotmnercial forage blower.
(Source: Advertisement)
The cost for a typical blower would be $1500. A base for it and a sta-
tionary diesel engine to power it would cost approximately $10,000. The total
cost would be about $11,500.
Small—Scale Blowers
A small—scale version of a forage blower may be a feasible method of dis—
tributirtg small objects over a spill.
A small stock blower or fan might be modified for use as a material hand-
ling blower. However, such units will probably need to be constructed more
-_--- ‘
68
-------
sturdily than blowers or fans which move only air. A power source such as a
small air—cooled diesel engine would also be required.
Such a system could probably be developed in about 6 months at a cost of
about $40,000, if a suitable stock blower can be found.
Air Cannon — Disintegrating Canisters
An air cannon could be used to propel a disintegrating canister full of
small objects over a spill. The small objects would be distributed over the
spill as the canister disintegrated.
Neither the air cannon nor the canisters are commercially available. The
canister to be used would need to be compatible with the cannon. It is doubt—
ful that any standard military ammunition would be suitable for use in an air
cannon.
A prototype system could be developed in about 9 months at a cost of ap-
proximately $100,000.
Explosive Cannon — Disintegrating Canister
An explosive cannon could fire a disintegrating canister full of small
objects over a spill. The small objects would be distributed over the spill
as the canister disintegrated.
The major problems with this approach are that cannons in the U.S. Army
typically have bores no larger than 175 mm (6.9 inches) and they are usually
weapons designed for ranges of several miles, rather than several hundred
feet. These problems might limit the particle—carrying capacity of special
ammunition and could cause aiming and deployment problems.
Developmental costs for a prototype system have been estimated at approx-
imately $115,000 over a period of 9 months.
Catapult — Disintegrating Canister
A gas-oven oil accumulator could hurl a canister containing a quantity of
small objects over a spill. The small objects would then be distributed over
the spill as the canister disintegrated.
The major problem with this approach is that neither suitably—sized cata-
pults nor canisters are commercially available. Both components would have to
be designed and built to order.
A prototype system could be developed in about 9 months at a cost of ap-
proximately $80,000.
Crop—Dusting Airplanes
Conventional crop—dusting airplanes could be used to disperse small ob-
jects or particles over spills. Typical crop—dusting airplanes have delivery
69
-------
payloads between 100 and 2000 ib, depending on the size of the airplane.
Their range from their point of origin is on the order of 40 mi.
Crop—dusting airplanes typically charge $75 for a 25—acre dusting job.
Centrifugal Pumps
Small objects could be spread over a spill with a centrifugal pump. This
concept might be attempted with the particles either wet or dry. In the case
of wet particles, the pump would be handling a slurry, which is a standard in-
dustrial practice. For dry particles, the pump would act essentially as a
centrifugal slinger. Provisions would have to be made for feeding the dry
particles into the eye oE the impeller. The particles exiting from the peri-
phery of the impeller would tend to impinge upon any diffusers that might be
in the vicinity of the impeller and would tend to either shatter and/or lose
velocity.
The concept could probably be attempted with existing centrifugal pumps
on emergency equipment that is presently available.
Sandblasting—Type Pump Arrangement
A commercially available sandblaster might be used to strew small par-
ticles over a spill area. A typical commercial sandblaster can deliver par-
ticles up to 1/8—in, diameter at rates ip to 1500 lb/hr (Figure 18).
Such a sandblaster is typically mounted on a truck bed and costs about
$20,000.
Centrifugal Sanders
Small objects or particles could be spread over spills using highway cen-
trifugal sanders. Typical centrifugal sanders are mounted on the rear of dump
truck boxes. They consist of an auger mounted transversely across the rear of
the dump box, which feeds sand to a centrifugal slinger. The entire mechanism
can be hydraulically driven. A typical spreader assembly suitable for mount-
ing on an 8—ft—long dump truck box has a capacity of 4.5 yd 3 , which can be
unloaded in about 10 minutes and scattered 20 ft (Figure 19).
Such a unit typically costs $5000.
SLLTRRY TRANSFER CONCEPTS
The concepts considered here were devised for the transfer of slurries,
such as lime/water, solid/liquid nitrogen, or rapid—setting inorganic foams,
to the spill.
70
-------
Figure 18.
Commercial sand—blasting equipment.
(Source: Advertisement)
Figure 19. Standard highway centrifugal sander.
(Source: Advertisement)
qq.
- S
— . ,.C (
- _:‘_•
71
-------
Standard Hydro _ Seeder*
A standard Hydro—Seeder could be used to deliver a slurry to a spill.
The Hydro—Seeder is a device that is used to deliver seeds to freshly prepared
highway road banks. Essentially, a centrifugal pump forces a mixture of seeds
and water through a hose nozzle aimed at the desired location for the seeds.
Solid particles up to 3/4—in, diameter can be sprayed wi h the system. T1
mixture can be sprayed up to 200 ft from the nozzle. The larger Hydro—
Seeders can spray 3000 gal of mixtures in about 20 minutes (Figure 20).
A 3000—gal—capacity Hydro—Seeder, ready for mounting on a truck, costs
about $12,000.
The Use of “Slippery Water ”
Slippery water is water in which shear—reducing additive has been mixed.
These additives have been used by fire departments in order to increase the
range of fire hoses. The range of a hose can be nearly doubled, since the
pressure drop arising from shear loss is markedly decreased.
- r. ,.
— - . .--- — . -
Figure 20.
Commercial hydra—seeder.
(Source: Advertisement)
* Trade name, Bowie Industries, Inc.
72
-------
The use of slippery water requires the installation of a metering system
to accurately mix the shear—reducing additive with the water. Such systems
are typically installed directly onto fire trucks.
The cost of installing a typical wet water system on a vehicle such as a
fire truck is approximately $6000.
Slurry Pumping With Existing Fire Truck Pumps and Hoses
Chemical slurries could be delivered to spills by using existing fire
truck pumps, hoses, and nozzles. Fire truck pumps are usually of the centri-
fugal type and are intended to pump water. Pumps can usually handle liquids
of viscosities up to 5000 SSV (SAE 40 motor oil at room temperature). Liquids
much thicker than this usually degrade the performance of any type of pump
when compared to the performance of the pump moving water. The presence of
intake screens on fire truck pumps should be taken into account in considering
this system. Another consideration is the possibility of fouling or corroding
the fire truck pumps during or after pumping slurries.
Since this concept uses existing equipment, the only major cost would be
any extra maintenance cost incurred by using this concept on the equipment.
Long Boom Deployment of Slurry Transfer Hoses
A boom arrangement could be used to deliver foam over a spill some dis-
tance from an operator. A typical truck—mounted crane equipped with a tele-
scoping hydraulic boom could be used to position a slurry transfer hose over
a spill. The slurry would be forced through the hose by a slurry pump posi-
tioned near a reservoir of slurry (Figure 21).
A telescoping truck crane with a 150—ft horizontal reach costs approxi-
mately $170,000. A suitable slurry pumping system capable of pumping up to
600 gal/mm would cost approximately $35,000.
SLURP V
RESERVOIR
Figure 21.
Slurry delivery concept by means of a truck—mounted boom.
SLURRY
PUMP
HYDRAULIC BOOM
NOZZLE
SPILL
73
-------
LIQUID TRANSFER CONCEPTS
Concepts involving liquid transfer processes include water flooding and
the addition of chemical neutralizers to the spill. Delivery of liquid nitro-
gen is a possibility for some approaches.
Pumps and Liquids From Available Fire Trucks
An obvious solution to the problem of liquid delivery to spills is the
use of existing fire trucks and their liquids to cover the spill. Additives
might be traxed with the liquid to enhance its “pumpability” or improve its
effectiveness after it is delivered to the spill, but the main concept is to
use as much standard equipment as possible.
Since existing equipment would be used, the only extra cost would be for
special liquids and additives (if required) and any extra maintenance costs
their use would incur.
Slippery Water Usage
The range of fire hoses can be increased by the use of shear—reducing
additives. Fire departments have made use of these additives. This involves
fitting a fire hose system with an additive injection/mixing device which
combines the desired amount of shear—reducing additives with ordinary water.
The cost of such a system installed on a typical fire truck is $6000.
Balloon Deployment of a Liquid—Dispensing Hose Vehicle
A hose could be supported over a spill by a small balloon and the nozzle
directed at the spill. Balloons have been used as support platforms in many
situations. They are occasionally used to transport large objects over ter-
rain where conventional ground—type equipment is forbidden or is not practical.
Examples of this are logging operations or 6ff—loading bulky containers from
ships near shore. However, balloons have several disadvantages. They may be
difficult to move from place to place. They are vulnerable in strong winds,
especially the hot—air variety, and they may require several hours to set up.
Reaction forces from the nozzle could be severe.
A typical helium balloon at 1000 lb lifting capacity is 33 ft in dia-
meter and is 84 ft long when inflated. The balloon and ground support equip—
merit cost approximately $90,000 (Figure 22).
Deployment of Hoses and Nozzles With Supporting Booms
A boom arrangement could be used to deliver liquid over a spill some dis-
tance from an operator. A truck—mounted crane equipped with a telescoping
hydraulic boom could be used to position a liquid transfer hose and nozzle
over a spill. The pump and hose would be furnished by fire trucks at the
scene.
74
-------
HOT AIR
BALLOON
NOZZLE
GUY LINES
U i
HOSE
Figure 22. Hot—air balloon delivery concept.
-------
A typical telescoping truck crane with a 130—ft horizontal reach costs
approximately $170,000.
ENERGY TRANSFER CONCEPTS
The ideas grouped under this heading explore the use of energy transfer
to accelerate evaporation from spills, to ignite spills, to extinguish spill
fires, and to oxidize the volatiles of spills. The purpose of increasing the
spill evaporation rate is to remove any residual hazardous material, under
controlled conditions, after the oool of liquid has been removed.
Flame—Throwers for Heating or Ignition of Spills
A flame—thrower might be used to heat or ignite a spill. The U.S. Army
has portable flame—throwers in its inventory. The units are carried by an
individual and contain 4½ gal of fuel. They weigh about 50 lb when loaded
and 22 lb when empty. The maximum effective range of such devices is about
150 ft with a flame duration of 5 to 8 sec (Figure 23). Civilian flame—
throwers having half the capacity and one—sixth the range oE the U.S. Army
units cost about $260.
Spill Heating With Infrared Radiation
Infrared lights, focused with a parabolic reflector, could be used to
heat a spill site and increase the rate of evaporation. Both gas and elec-
trically powered portable infrared heaters are available commercially. A gas—
powered 32,000—BTU—per—hour gas heater costs approximately $80. A 2l,000—BTTJ—
per hour electrical heater costs aporoximately $160. A suitable reflector
could be fabricated. Developmental costs for a prototype system have been es-
timated at approximately $28,000 over a period of 4 months (Figure 24).
Solar Energy Lenses
Large concentrating lenses could be used to heat the chemical spill with
solar energy, and thus increase the rate of evaporation. A typical scheme
might involve several concentrating lenses which would only be able to focus
the sun’s rays upon a few small local areas at any particular time. The tem-
perature at these points might be raised several hundred degrees, depending
upon the heat transfer characteristics of the surroundings and the boiling
point of the chemical spilled. The lenses could be panned about the spill so
that the entire surface of the spill would have been heated, although it
would not be uniform. Once the concentrated rays were removed from a particu-
lar area, however, that area could be expected to cool rapidly.
A standard type of concentrating lens that might be used in this type of
situation is a Fresnel lens. They are manufactured in sizes up to about 3 ft
x 15 ft. The cost of the lenses would be about $3.50/ft 2 .
Sonic Disturbance or Blast
In about 10 percent of the spills, fire is involved. While the fire
consumes much of the released vapor, it also causes damage to nearby cargos
76
-------
UNLOADED WEIGHT AS PICTURED: 42 lb
LOADED WEIGHT (DEPENDING ON TYPE OF FUEL USED): 60.70 lb
TYPE OF FEED: FUEL PROPELLED BY GAS UNDER PRESSURE.
METHOD OF OPERATION: MANUAL.
DURATION OF FIRE: CONTINUOUS DISCHARGE 8.9 seconds.
MAXIMUM EFFECTIVE RANGE: 20.25 meters UNTHICKENED FUEL
40.50 meters THICKENED FUEL
REMARKS: OPERATING CAPACITY 4-1/2- 4-3/4 gallons OF FUEL
WEIGHING 2510 lb. THIS FLAME THROWER IS
STANDARD B. MAY BE FIRED EFFECTIVELY AT
TEMPERATURES AS LOW AS-26 0 F.
AN ALTERNATIVE TO THIS IS THE M202/M74 FLAME SYSTEM
WITH A RANGE OF 200 yards.
Figure 23. Commercial flame—thrower.
(Source: Dept. Army Manual)
77
-------
INFRARED LIGHTS
WITH PARABOLIC
REP LECTORS
Figure 24. Infrared heating concept.
and may cause the release of more chemicals. Thus, under many spill situa-
tions, it is desirable to extinguish the fire.
The affect of an explosion or the sonic boom of a jet might be utilized
to blow out a burning spill. An explosive charge could be deployed directly
above the chemical spill by lowering it with a crane. The charge could then
be detonated to either jar the spill or, in cases of burning spills, to put out
the flame. This procedure would be similar to one used in putting out fires
in oil fields.
Material and equipment costs for putting out a fire in this manner should
be negligible compared to labor costs.
Standard Heat Exchanger System
A standard heat exchanger system could be developed for increasing the
temperature of a spill, thus increasing the evaporation rate. A typical sys-
tem might consist of a pump, a boiler, a heat exchanger, and some tubing. The
spilled liquid would be heated up in the heat exchanger and then recirculated
into the spill to raise the overall temperature of the fluid. This type of
system has been studied for possible use in the off—loading of oil tankers.
The materials for the pump, the heat exchanger,and the tubing must be selected
to withstand the corrosive nature of some of the chemicals.
Costs for this type of a system have been estimated at approximately
$lO—$l2 per 1000 BTIJ/hour. These estimated rates are based on standard sys-
tems with capacities in the range of 100,000 to 1,000,000 BTU/hour. Fuel
costs should be relatively insignificant (Figure 25).
Microwave Transmitters
Microwave energy could be used to heat and evaporate the spill. Micro-
wave energy is quickly dissipated unless it is projected into an area with a
high degree of reflection. For this reason, infrared projectors would appear
.1
SUPPLY
SPILL
78
-------
Figure 25. Heat exchanger concept.
“C
-------
to be much superior to microwaves. (Use of infrared lights was the second
concept listed under energy transfer.)
Incendiary Bombs
A standard, commercially—available incendiary mixture, such as thertnite,
could be spread over the spill and ignited. The combustion of the thermite
would achieve temperatures of approximately 3300°F for a short period of time.
About half of the incendiary metal (by weight) would be left after ignition.
A typical chermite reaction has an approximate specific energy ratio of
50 BTtJfpound. The cost of thertnite is about 55 /lb, resulting in an approxi-
mate cost of l /BTU.
Standard Cooling Equipment
A standard heat exchanger system could be adapted for use in decreasing
the temperature of a chemical spill. Such a system might involve a pump, a
refrigeration unit, a heat exchanger, and some cubing. The liquid chemical
would be pumped through the heat exchanger, where it would be cooled, and then
sprayed back into the pool to lower the overall temperature of the spill.
This scheme is similar to the one discussed for raising the temperature of
the spills. The materials for the pump, the heat exchanger, and the tubing
would have to withstand the corrosive nature of some of the chemicals.
Costs for this type of system have”been estimated at approximately $50
per 1000 BTU/hr.
Air Curtain Destructor Devices
The hazardous chemical spill could be drawn off into trenches at a con-
trolled rate. Air curtain destructor devices could then be emplaced over the
trenches to ignite the chemical vapors. A super—charging air curtain causes
the vapors to burn rapidly with little or no smoke emission.
Portable air curtain destructors are commercially available in 21— or
42—ft lengths. The cost of a 21—ft unit with a diesel engine drive would be
313,600 (Figure 26).
80
-------
Figure 26. Air curtain destructor.
(Source: Advertisement)
-------
SECTION 7
SCENARIOS UTILIZING NEW CONCEPTS
The situations described in this section were compiled from actual field
experiences of the Battelle—Columbus staff. The intention here is to assemble
and compare the currently employed procedures for combatting volatile hazard-
ous spills on land with potential control concepts that could be “the better
way ” if further developed.
The spill scenarios that are included in this section occur frequently in
the heavily industrialized states and their occurrence in the future is highly
probable. The scenarios are exemplary and should not be construed as present-
ing the only spill control techniques that were actually attempted and/or were
successful.
It has been found that in practice, response teams have utilized combina-
tions of vapor suppression techniques to fit the variations of the physical
field setting. In all scenarios selected, the safety of the cleanup crews was
of paramount concern.
LPG*_DIFFUSION FROM TJNDERGROUND PIPE
Situation
A report has been received from a volunteer fire department that they are
obtaining explosive values on their monitoring equipment in basements of homes
in a semi—rural area. It was rapidly determined that the gas was diffusing to
the surface along a ditch in front of the homes. An LPG pipeline marker is
nearby. A call to the company operating the pipeline reveals that they had an
automatic shutdown during the night before and they are walking the line to
find the break. Fumes are flowing heavily from the ditch.
Ambient Conditions
The surface air is calm and a conduction fog is blanketing the area. The
terrain is flat, with repeating surfaces of sand, gravel, and alluvium making
up the site soils. The wells in the area are venting gas.
* Liquefied Petroleum Gases
82
-------
Problem
The highway has been closed and the neighborhood evacuated for an unknown
period of time. Explosions have occurred in some basements.
Current Response
Using explosion—proof core drilling rigs, numerous shallow shafts have
been drilled and fitted with flaring ignition systems to burn off, at height,
the permeating gas with the logic of ‘WICKIMG” the gas to the flames.
Proposed Mew Technology
Option I——
After the gas main has been shut down, upstream trenches should be dug
into the lens where the sand and gravel surface. A slightly modified sand—
blasting pump is utilized to blanket the bottom of the trench with 1/8—in.—
diameter dry ice pellets. The flashpoint should drop considerably. The CO 2
sublimes and fills the trench bottom. Since the CO 2 is a gaseous blanket,
the natural gas can vent slowly but under greatly reduced ambient air release
rates.
Option II——
After determining the subsurface piezometric gradient, several shallow
injection wells are drilled up the geologic gradient and away from the per-
meating natural gas. Fresh water is me ered into the receiving formation.
The advancing permeation of the water will increase the rate of methane gas
migration into the wicking flare stacks as described in the Current Response
description.
Option III——
Increase the rate of gas emission from the deep trenches by utilizing
infrared lights (explosion—proof) with parabolic reflectors to heat the trench
walls to further increase the volatility. Above these trenches, large fans
(commonly used to abate smoke damage in buildings that have had a fire) are
directed to provide dilution and direction to the plume. Convection currents
are established in the trench to create a “chimney” draft effect.
STREANINC—VOLAT ILIZ 1MG ACRYLONITRILE TANI( CA.R (RAILROAD)
Situation
Because of a humping yard accident, the dome cover and valves of a 5000—
gal tank car have been sheared and acrylonitrile is streaming into the rail-
road bed ballast and drainage channels adjacent to the facility.
Ambient Conditions
It is an extremely hot day and the humping yard temperature is about
110°F. The fumes are being blown at a velocity of 1.5 to 5 in/s (3 to 10 knots)
directly toward a grain—processing warehouse facility. The spill is dammed in
a concrete—lined ditch.
83
-------
Current Response
Even -though ignition has not occurred, the local fire brigade of the rail-
road is directing 2 deluge (6—inch) hoses at the emitting opening of the tank
car, which is nearly upside down. Another fire—fighting unit is covering the
pools of acrylonitrile with sand. The earthmoving equipment operators are
being subjected to extreme danger.
Proposed New Technology
A large helicopter drops a large parachute of Rypalon material to cover
the tank car. The bottom of the parachute is equipped with 6—in.—diameter
light canvas hose. Through this hose system liquid nitrogen is pumped into
the chute and inundates the plumbing system of the tank car. Eventually, with
the added insulation of the chute, the tank car plumbing will be frozen and
the acrylonitrile will be frozen in place. Volatilization is greatly reduced.
The response time for helicopter—aided deployment of the covers in spill
response circumstances has been less than 2 hours.
MASSIVE GASOLINE SPILL — BULK FARM
Situation
Due to a T ’computer” error, an additional 80,000 gal of gasoline were
piped into a full tank. The subsequent overfill, which occurred during the
night, nearly filled the dikes surrounding the tank and volatilized rapidly.
Ambient Conditions
An air stagnation condition covers the local area. Explosive mixtures
of gasoline—air are gathering in many pockets and buildings. Sunlight is
intense and the volatilization is rapid. Several trenches and sumps are gath-
ering previous gasoitne runoff that created the dikes.
Current Response
A concerted effort is being made to re—load the spilled gasoline into
waste—oil tanks. The local fire department is applying various depths of
protein—based foams to the surface of the pool of gasoline. The perimeter of
the spill zone is being flushed with open hydrants and a sheen of gasoline can
be seen flowing into a storm drain and into a creek.
Proposed New Technology
Option I——
Several vacuum—siphon lines are placed in the gasoline by using booms of
aerial fire ladders. The dike zone is treated with large blocks of dry ice
delivered by a four—bar linkage device. The gasoline spill is then blanketed
* DuPont trade name.
84
-------
with dry ice pellets delivered by a continuous air cannon device. Tarpaulins
are deployed over the dike zone utilizing remotely operated aerial ladder
equipment. Once a vapor shield of CO 2 has formed under the tarpaulin, the air
vacuum pumps can begin operation to remove the cold gasoline. The time neces-
sary to complete the above option is estimated at less than one hour.
Option II——
Utilizing two or more Silage blowers with a mulching machine, layers of
pulverized straw or peat could blanket the spill and insulate as well as
absorb a significant percentage of the spilled gasoline. Rates of application
could approximate 5000 lb/mm.
Option III——
To limit the surface area of the gasoline spilled, the gasoline could be
encircled by quick—setting semi—solid foams. Such efforts could be accom-
plished in less than 20 minutes if the equipment were readily available to the
spill contractor or fire departments.
VENTING INDUSTRIAL SUMP - HYDROGEN SULFIDE
Situation
At a local rendering plant located in a congested industrial park, a
large outdoor sump is generating repeating pulses of hydrogen sulfide gas (H 2 S).
Anaerobic decomposition is evident in the mechanically—mixed sump tank. Sev-
eral firemen have been injured.
Ambient Conditions
The sump is partially enclosed by the roof of an open—air receiving dock.
However, a gentle breeze of 3 to 5 knots is blowing the I-1 2 S toward several
nearby industries. The sump has concrete walls.
Current Response
The remaining firemen evacuate the premises and direct several forced—air—
blowing and exhausting fans over the sump. Water was initially added to the
bubbling fluid in the pit, but soon afterward it seemed to increase the rate
of H 7 S emission.
Proposed New Technology
The fluid in the sump has been depleted of oxygen and the system has gone
anaerobic. Large quantities of air or oxygen should be bubbled through the
sump in an attempt to re—establish an aerobic process.
BURNING PESTICIDE WAREHOUSE, HCN, PHOSGENE, ETC.
Situation
After a fire fed by tanks of xylene and toluene, a warehouse containing
340 tons of mixed pesticides has been reduced to a smoldering ruin. Fumes
85
-------
from the destroyed warehouse are being blown across the street into a resi-
dential neighborhood and are also enveloping a small hospital. The winds are
variable and the plume trails and loops along the surface. Approximately 140
people have been treated at surrounding hospitals for the toxic fume inhala-
tion. Runoff has contaminated a water supply reservoir. Massive fishkills
are in progress. The city’s water intake has been shut off. The walls and
roof of the warehouse have totally collapsed.
Ambient Condition
It is a cool summer night and the smoldering debris has been determined
to lie upon an intact concrete floor area. Runoff waters have flooded the
streets and curbs of the formulating and packaging plant. The terrain is
generally flat.
Current Response
The several fire departments have been spraying the debris with even more
water and are maintaining the evacuation status.
Proposed New Technology
Option I——
Utilizing large helicopters, several fireproof tarpaulins are deployed
over the debris pile in an upside—down parachute attitude. The chutes are
lined radially with canvas hose lines with gas defuser webbing near the center
of the circular tarpaulins. High—pressure induction fans blow liquid nitro-
gen through the tarpaulins. The cryogen filters through and chills the once—
smoldering pesticide—laden ashes. The toxic fumes are no longer generated and
the liquid pesticide—laden runoff is frozen. After several hours of treat-
ment, the solid frozen mass could be loaded into covered waste disposal trucks
for additional treatment, encapsulation, etc.
Option II——
Several large industrial sandblaster units can be loaded with dry ice
pellets crushed to about 1/8—in, diameter and used to blanket the smoldering
pesticide—laden debris with several inches of dry ice pellets. Fireproof
tarpaulins are then pulled over the rubble to provide insulation. A marked
reduction in fume generation should result.
Option III——
The pesticide—laden rubble could be smothered with earth carried in by
large payloaders. Soil from nearby sites could be used to bury the piles of
smoldering chemicals. It is estimated that approximately 20 tons of soil
could be moved and dumped on the rubble in less than 2 hours.
NATURAL GAS — PIPELINE RUPTURE —ABOVE GROUND
Situation
Because of a hillside slump, a 6—in.—diameter natural gas pipeline has
ruptured and is exposed on a slope above a secondary road in the small valley
86
-------
below. The road is closed to traffic. Some birds and other small wildlife
have died in the plume.
Ambient Conditions
There is a downhill convection of the methane and the gas is flowing
under a canopy of trees.
Current Response
Attempts are made to find the upstream shutoff valves while the fire de-
partment plays a remote deluge hose upon the leak.
Proposed New Technology
A cyano—acrylate film is applied over the spill with a 3—in. —diameter
pressure hose fitted with a small acid atomizer providing a slight acid medium
for the film to form upon. The ruptured pipe end is simply covered by a mound
of film that sets up in less than 45 seconds. The methane is held in place
by the pipe until back pressure can be applied to vacuum the methane out of
the line. The material is applied with a piece of equipment similar to Figure
11.
PHOSGENE GAS EMISSION — INDUSTRIAL SOURCE
Situation
En a basement storage room a 1—ton cylinder of phosgene gas is venting
phosgene from a defective seal. The chamber’s ventilating system is pumping
phosgene gas out of several ports. The light yellow gas is rolling down a
streetside curb and into storm sewer drains.
Ambient Conditions
The humidity is high and the emission has occurred during a light rain.
Current Response
A mist nozzle is used to spray water on the cylinder dome cap but the
phosgene is still moving out of the room.
Proposed New Technology
Option I——
The phosgene cylinder is packed in dry ice to lower the temperature and
pressure of the phosgene. A large spray aerosol can of cyano—acrylate is
sprayed on the leaking seals to form a solid film cover to retard the leakage
of phosgene gas.
Option II——
A high—pressure slurry of 30 percent lime is directed on the leaking
cylinder to adsorb and react with the phosgene. The phosgene is in effect
37
-------
scrubbed from the air and partially neutralized (Figure 22).
O tion III——
To reduce the leak rate and the concentration of phosgene, the cylinder
could be placed on its side and the room flooded with a water slurry of acti-
vated carbon. With hose attachments the water—carbon slurry could be pumped
by a modification of the design of a commercial hydro—seeder. (Figure 21 )
SILICON TETRACHLORIDE BULK TANK LEAK
Situation
In a heavy industrial area an 800,000—gal bulk tank of silicon tetrachior—
ide (SiC1 4 ) began venting deadly clouds of HC1 vapors.
Ambient Conditions
The meteorological conditions were stagnant with 2— to 4—mph winds. A
large pool of hydrochloric acid accumulated in diked areas around the tank’s
offloading zone.
Current Response
The diked area was blanketed with sand, then with lime, and then with
#6 heating oil. As a last resort, a quick—setting concrete was dumped on the
plumbing system in an attempt to seal the pipe leak.
Proposed New Technology
Option I——
Trenches can be dug inside the diked area to gather the SiC1 4 in sumps.
This will greatly reduce the area of volatilization. With a modified snow
blower, several inches of pellecized dry ice are spread on the chemical. At
the source of the spill, the pipe gallery, a hood or tent is lowered over the
spill by the boom of an aerial fire truck. Styrofoam blocks are added to the
bottom of this makeshift tent or hood. Liquid nitrogen is then fed over the
aerial ladder and into the hood area over the leaking pipe. The SiC1 4 will
rapidly freeze solid and plug any further emission. The remaining volumes
could be offloaded into appropriate tank trucks, railroad cars, etc. It will
be necessary to bleed the refrigerant liquid nitrogen over the frozen pipe
gallery to maintain its cryogenic property. The rate of emission over the
ponded chemical will greatly reduce the escape of the hazardous gas.
Option II——
High—speed conveyor belts can be utilized to load dry ice into the tank
through the roof vents of the tank to reduce the temperature of the SiC1 4 .
This will lower the rate of vaporization of the SLC1 4 .
Option III——
Construct a prefabricated acid—resistant dome. This acid—resistant mate-
rial could be very plastic and flexible in nature. Erection of the dome could
88
-------
be assisted by forced—air pumps and fans. Once this dome or tunnel corifigura—
don is placed over and against the pipe gallery, this structure would insu-
late and provide a limited exposure area for both the S1C1 4 and the coolant
(LN 2 , dry ice, etc.).
89
-------
REFERENCES
1. U.S. EPA’s Designation of Hazardous Substances, December 30, 1975, Fed-
eral Register, Vol. 40, pp 59960—60017.
2. Dept. of Transportation, U.S. Coast Guard, Chemical Hazards Response In-
formation System (CHRIS) — Hazardous Chemical Data, CG—446—2 (January
1974).
3. Dept. of Transportation, U.S. Coast Guard, Hazardous Chemicals That Pro-
duce Vapor, CG—D—46—75.
4. Greer, J.S., “Feasibility Study of Response Techniques for Discharge of
Hazardous Chemicals That Float on Water,” Contract No. DOT—CG-5l870—A,
Rept. No. CG—D—56—77 (October 1976).
5. Whiting, L.D., R.E. Shaffer, R.H. Frickel, “Feasibility Study of Response
Techniques for Discharge of Hazardous Chemicals That Vaporize.” Chemi-
cal Systems Lab, USA ARRADCOM, Aberdeen Proving Ground, MD. To be
published.
6. Buckley, J.L. and S.A. Wiener, “Hazardous Material Spills: A Documenta-
tion and Analysis of Historical Data,” EPA Contract No. 68—03—0317 (April
1976), EPA—600/2—78—066 (April 1978).
7. National Fire Protection Association, Fire Protection Guide on Hazardous
Materials, 6th Ed.
8. Chemical Rubber Publishing Co. Handbook of Chemistry and Physics, 47th
Ed. -
9. Dow Chemical, Physical Properties of Chemical Substances (1952).
10. Tinimernans, J., Physico—Chemical Constants of Pure Organic Compounds,Vol. 1,
Elsevier Publ.Co., New York (1950).
11. Timmermans, J., Physico—Chemical Constants of Pure Organic Compounds,
Vol. 2, Elsevier Pubi. Co., New York (1965).
90
------- |