WATER POLLUTION CONTROL RESEARCH SERIES • ORD - 3
              OIL  DISPERSING
CHIMICALS
U.S. DEPARTMENT OF THE INTERIOR • FEDERAL WATER POLLUTION CONTROL ADMINISTRATION

-------
          WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Reports  describe the
results and progress in the control  and  abatement  of  pollu-
tion of our Nation's waters.  They provide a  central  source
of information on the research,  development and  demonstra-
tion activities of the Federal Water Pollution Control
Administration, Department of the Interior, through inhouse
research and grants and contracts with Federal,  State, and
local agencies, research institutions, and industrial organ-
izations.

Water Pollution Control Research Reports will be distributed
to requesters as supplies permit. Requests should be sent to
the Planning and Resources Office, Office  of  Research and
Development, Federal Water Pollution Control  Administration,
Department of the Interior, Washington,  D. C. 20242.

-------
                         u a .
pvRcH\\/e              Headquarters and Clremical Libraries
*                            B'* lA/'est Bld9 Room 3340
                                  Mailcode 3404T
                             1301 Constitution Ave NW
                               Washington DC 20004
                                   202-566-0556
                                 OIL DISPERSING CHEMICALS
                         A Study of the Composition, Properties  and


                         Use of Chemicals for Dispersing Oil  Spills
                                            by


                                   Melvin Z. Poliakoff


                                        Consultant


                                   Tenafly,  New Jersey





                                         for the


                       FEDERAL WATER POLLUTION CONTROL ADMINISTRATION


                                DEPARTMENT OF THE INTERIOR


                             Edison Water Quality Laboratory


                                     Northeast  Region


                                Edison, New Jersey,08817





                              Program Number 15080FHS 05/69


                                Contract Number 14-12-549





                                         May 1969
                               Repository Material
                             Permanent Collection
                                      LIBRARY
                                      iv t  of f-,i nijrtor, FWPCA
                                      f(j'i' 111, ii J oooi.7

-------
           FWPCA Review Notice
This report has been reviewed by  the Federal
Water Pollution Control Administration  and
approved for publication.  Approval does not
signify that the contents  necessarily reflect
the views and policies of  the Federal Water
Pollution Control Administration, nor does
mention of trade names or  commercial products
constitute endorsement or  recommendation for
use.
                      ii

-------
                                              ABSTRACT









                  A  "state-of-the-art" review is provided for chemicals which are




                  used for  dispersing spilled oil.  Among the topics discussed are:




                  the history of the development of oil  spill dispersants; basic




                  emulsion  chemistry, including the nature and properties of surface




                  active agents; chemical  composition of oil spill dispersants,




                  including a description  of generic chemical types; production of




                  dispersants,  properties  of dispersants; chemical analysis of oil




                  spill dispersants; and the measurement of oil  spill dispersing




                  power.









                  KEYWORDS:  Chemical analysis, chemicals, detergents, dispersion,




                              emulsifiers,  oil, surfactants, testing, water pollution




                              control, emuls ions.
I
^-                                              111

-------
                      TABLE OF CONTENTS


                                                             Page

Conclusions ------------------------------------------------  1

Introduction ---------- • --------------------------------------  2

Historical Development of Oil Spill Dispersants -------------  3
Emulsion Chemistry ------------------------------------------  ^
  Surface active agents
  Emulsions
  Emulsion Stability

Chemical Composition of Oil Spill Dispersants --------------  8
  Surf act active agents
  Solvents
  Other additives
Production and Properties of Oil Spill ---------------------- 14
  Dispersants
  Manufacture
  Relationship of physical properties to performance

Chemical Analysis of Oil Spill Dispersants ------------------ 16

Measuring Oil Spill Dispersing Power ------------------------ 19
  A.S.T.M. tests
  Military specification tests
  Laboratory determination of dispersing potential
  Field testing

References ------------------------------------------------- 26
                             IV

-------
                          CONCLUSIONS
The writer has attempted to provide an overview of  the  state-of-the-
art of manufacturing and testing Oil Spill Dispersants.

The chemicals currently available for oil  dispersant  use were devel-
oped out of the existing body of knowledge of  emulsion  technology.
Many of these materials have been used to  "solve" emergency oil  spill
problems under a now badly outmoded concept of pollution control —
which was to eliminate the immediate problem of visual  pollution and
fire hazard, with little or no prior evaluation of  possible ecological
consequences.  This has led to considerable criticism of chemicals
which sink, emulsify or disperse oil slicks (21).   Improved methods of
oil removal from the water, rather than dispersion  in the water, are
under development (22).

Nevertheless, the use of chemical dispersant methods  may continue to
find a place under closely controlled conditions where  removal tech-
niques are not practical.  Such situations exist adjacent to and under
docks and pilings, where the spill presents an immediate fire hazard,
and where removal equipment is not readily available.  Under these
circumstances, the authorities in charge will  weigh the possible
ecological damage which chemical treatment might cause  against the
immediate hazards of no treatment.  Reliable information concerning
toxicity to marine life, as well as other effects  of these chemicals,
is essential before such evaluations can be made.

The promulgation of standardized toxicity test methods  for oil dis-
persant s will aid manufacturers in their own development of  safer
chemicals and methods of treating oil spills.   Developments  in this
field will lead to products which are more effective at lower concen-
trations, less hazardous to handle, and more readily biodegradable, as
well as less harmful to the ecology.
                              - I -

-------
                             INTRODUCTION
This document was prepared by the  author  for the Federal Water
Pollution Control Administration under  Contract Number 14-12-549
in May 1969.

The objective was to furnish a state-of-the-art report on the
composition, properties  and uses of chemical dispersants used for
treating oil spills on water.  The work was specifically confined
to chemical materials which function by virtue of their surface
active, emulsifying, or  dispersing properties.

The information furnished is based on literature search, as well
as the experience and research of  the author in this field.
                              -  2  -

-------
          HISTORICAL DEVELOPMENT OF OIL SPILL DISPERSANTS
In tracing the development of oil spill dispersants we need only go
back as far as the early 1930's when water emulsifiable  degreasers
were in fairly widespread use.   These degreasers were developed to
answer the need for effective methods of cleaning  oily and greasy
surfaces.  They possessed the properties of dissolving or dispersing
in the grease or oil, and making the resultant mixture dispersible
in water so that it could be flushed away with water.

The early products were composed mostly of soaps and solvents.  The
demands of the petroleum shipping industry required products that
could be used effectively aboard ship with seawater.  This led to the
use of materials other than soaps as emulsifying agents  because soap
breaks down in seawater.  Sulfonated petroleum oils and  later more
sophisticated synthetic detergents made their appearance in these
products.

These emulsifying degreasers were widely used aboard ship for engine
room maintenance, as well as for the clean out of  petroleum cargo
tanks prior to welding repairs and prior to upgrading of cargo.

Because of their effectiveness in cleanout of oil  residues  it was
natural that they should be tried for treating oil spills.  In  some
cases they were incorporated in oil slops prior to dumping  overboard
in order to minimize slick formation.

Considerable quantities of these degreasers are used by  industry as
well as by various governmental services.  Some of the  specifications
describing water emulsifiable degreasers established by  the various
government units include the following:

P-C-576       Compound, grease cleaning, solvent  emulsion

P-C-444       Cleaning compound, solvent soluble,  grease emulsifying

MIL-C-7122    Compound, grease cleaning, solvent  emulsion type

MIL-C-20207   Cleaning compound, solvent emulsion, grease removing

MIL-C-22230   Fuel Tank and bilge cleaner

MIL-C-22864   Solvent-emuIsifier, oil slick

MIL-C-25179A  Compound, emulsion cleaning (for aircraft)
                               - 3 -

-------
                      EMULSION  CHEMISTRY
In any study of oil spill  dispersants, familiarity with the surface
active chemicals used,  as  well  as  the mechanisms of emulsion forma-
tion, is a necessary starting point.  A  brief  review of these
subjects will be helpful before examining  the  manufacture and per-
formance of oil spill dispersants.
Surface Active Agents

Surface active agents are  often  defined according to their behavior
in aqueous solutions  (1).   These solutions will usually wet  surfaces
readily, remove dirt, penetrate  porous materials, disperse solid
particles, emulsify oils and grease,  and  produce foam when stirred or
shaken.  All of these properties are  interrelated,  and no surface
active agent possesses only one  of  them to the exclusion of  the rest.

A compound may be called a wetting  agent  rather than a detergent
because its wetting power  is greater  than its detergent power.  Like-
wise a compound may be called an emulsifying agent  rather than a deter-
gent because its ability to emulsify  oils is greater than its cleaning
power.

The molecules of surface active  agents are made up  of two parts:  a
relatively large elongated part  which is  the hydrophobic portion, and
a relatively small solubilizing  polar group which is known as the
hydrophylic portion.   The  antagonism  of these two portions of the
molecule and the balance between them gives the compound its surface
active properties.  When the proper balance between the two  portions of
the molecule exists,  the substance  neither dissolves completely nor
remains completely undissolved,  but rather concentrates at a liquid-
liquid interface.  In an emulsion,  the molecules of surface  active
agent are so oriented that the hydrophylic groups are anchored in the
aqueous phase and the hydrophobic groups  project into the non-aqueous
or oily phase.

Surface active agents are  often  divided  into two broad classes depend-
ing on the character of their colloidal  solutions in water.  The first
class, ionic surface active agents, form  ions  in solution, and like the
soaps are typical colloidal electrolytes. The  second class, the non-
ionic surface active agents, do  not ionize, but owe their solubility
to the combined effect of  a number  of weak solubilizing groups such as
ether linkages or hydroxy  groups in the molecule.

-------
A more detailed discussion of  those surface active agents  used in
oil spill dispersants will be made in a later section  on composition
of these products.
Emulsions (2) (3)

A.S.T.M. (4) defines emulsion as "a suspension of  fine particles  or
globules of one or more liquids in another liquid".   The two  liquids
are of course mutually insoluble, and oil and water  is our  typical
example.  The stability of emulsions ranges from the "shake well  be-
fore using" pharmaceutical preparations of a few years ago  to  such
stable emulsions as homogenized milk.  Surface active agents,  emulsi-
fying agents in particular, are used to stabilize  the dispersion  of
the two insoluble liquids.  The dispersed liquid is  often called  the
internal or discontinuous phase, and the dispersing  medium  is  called
the external or continuous phase.  Emulsions are often classified as
to whether the oil is the dispersed phase — oil in  water emulsions;
or whether water is the dispersed phase — water in  oil emulsions.
The type of emulsion formed is a function of viscosity, dielectric
constant, the concentration of ingredients, specific gravity  and
hydrophile-lipophile balance of the emulsifying agents.

Griffin (5) (6) has established helpful data on hydrophile-lipophile
balance (HLB) of emulsifying agents as effecting emulsion formation and
stability.

A hydrophylic emulsifying agent, one in which the  hydrophylic portion
of the molecule is large with respect to the hydrophobic portion, will
tend to form oil in water emulsions.  Hydrophobic  emulsifying agents
in which the hydrophobic portion of the molecule is  large with respect
to the hydrophylic portion of the molecule will tend to form  water  in
oil emulsions.  Water in oil emulsions can be recognized by the fact
that they do not disperse in water.  They will coalesce when  placed on
the surface of water.  These may well have been the  globs of  "chocolate
mousse" noted in some of the Torrey Canyon references (7).

As an aid in predicting emu Is if ic at ion tendencies  of surface  active
agents, these compounds have been classified as to hydrophile-lipophile
balance.  The HLB of a surface active agent can be determined approxi-
mately by the appearance of its aqueous solutions.  See Table I  (3).
                               - 5 -

-------
                                TABLE I

                 HLB DETERMINATION BY DISPERSIBILITY


Appearance in Aqueous Solution                            HLB Range

No dispersibility                                           1  to   4
Poor dispersion                                             3  to   6
Milky dispersion after vigorous agitation                   6  to   8
Stable milky dispersion                                     8  to  10
Transluscent to clear dispersion                          10  to  13
Clear solution                                            13+
The applications for various surface active agents can be predicted
by the HLB.  From the data in Table II (4) it will be noted that  oil
in water emulsif iers should preferably be selected with an HLB range
above 8.
                                TABLE II

                  HLB RANGES AND THEIR APPLICATIONS


Range                                    Applic at ion

 3 to  6                         Water in oil emulsifier
 7 to  9                         Wetting agent
 8 to 18                         Oil in water emulsifier
13 to 15                         Detergent
15 to 18                         Solublizer
Becher (4) has pointed out that spreading phenomena,  which are sus-
ceptible to direct observation, may be used as a rapid index for
determining the approximate HLB required for stability of emulsions.
In this procedure, a drop of the oil to be emulsified is placed on
the surface of aqueous solutions of varying HLB contained in petri
dishes.  At the highest HLB, complete spreading is observed.  As the
HLB decreases so does the spreading.  The point where the system goes
from spreading to non-spreading will be the optimum HLB for most
emulsion stability.

Gorman and Hall (8) have observed a quite reasonable  correlation
between dielectric constant of oil phase and the required HLB number
for suitable emulsif ic at ion.
                               - 6 -

-------
Emulsion Stability

The particle size or droplet size of the dispersed oil in an emulsion
is one of its most important characteristics.   It will determine
appearance as well as stability.  Particles larger than one mu  will
produce milky white emulsions.  Particles between one  mu and 0.1 mu
will produce blue-white emulsions.  Smaller than this  size produce
from gray semi-transparent emulsions to transparent emulsion.

Particle size is a function of the composition of the  ingredients,
quantity and type of surface active agent, and the method and forces
applied in forming the emulsion.  Generally speaking larger quanti-
ties of surface active agents produce smaller  particle size. Order
of mixing also plays an important part.  More  effective dispersion
and smaller particle size usually results when the surface active
agent is incorporated in the oil phase prior to mixing with water.
The surface active agents will tend to orient  themselves at the oil-
water interface.  The strength and compactness of this interfacial
film of surface active agent is important from the standpoint of
stability.

The two major examples of instability in emulsions are creaming and
breaking.  An emulsion is considered to break  if it separates into
two immiscible phases.  These phases may be redispersed upon agitation
if the surface active agent is still present.   Creaming is not  de-
emu Is if ic at ion, but rather separation of the emulsion into two
emulsions.  In one of the two emulsions the concentration of the dis-
persed phase is higher than in the other.  Rate of creaming is  usually
proportional to the difference in density between the phases.

Some factors which influence the stability of  emulsions are:

1.  Composition and quantity of emulsifying agent.

2.  Presence of electrolytes in the aqueous phase.  Large quantities
of electrolytes interfere with ease of emuIsification and also  cause
breaking of an emulsion.  This becomes significant in seawater.

3.  Mode of addition of surface active agent.   Thorough mixing  of the
emulsifying agent with the oil prior to agitation is more likely to
produce a stable emulsion.

4.  Type and force of agitation.  Mixing with  high shear tends  to break
the particles down to smaller size.  The greater the shear the  smaller
the particles and the likelihood of greater emulsion stability.
                               - 7 -

-------
5.  Temperature.   At higher temperatures  emulsions  usually tend to
form more readily and break more readily.  High temperatures are
rarely a factor in oil spill dispersion.   However,  at  extremely low
temperatures,  oils generally have a very  high viscosity  and  it
becomes very difficult to get sufficient  agitation  to  break the mass
of oil into tiny  droplets.   If the oil  has become waxlike, it is
almost impossible to produce satisfactory emulsion  stability.

6.  Decomposition of emulsifying agent.  Biodegradation  or other
forms of decomposition will lead to breaking of an  emulsion.  Hope-
fully, when this  occurs,  the oil has become widely  dispersed and the
biodegradation of the oil is occurring  simultaneously  with the
emu1s ify ing agen t.

Item 3 above is very important in the preparation of stable emulsions
and it is instructive to  discuss the various possibilities in the mode
of addition of the components of an emulsion.

1.  Emulsifying agent is  mixed with water and then  oil is  stirred  into
the water.  Such a procedure usually results in a coarse unstable
emulsion with large oil droplet size.

2.  Emulsifying agent is  mixed with oil and this mixture is added to
water with stirring.  This procedure produces an oil in  water emulsion
which  is likely to be stable and uniform.  Droplet  size  may well be
in the colloidal  range producing good stability.

3.  Emulsifying agent is mixed with oil,  and water  is  added to this
mixture while stirring.  This procedure is likely to produce a very
high viscosity water in oil emulsion.  Viscosity may become  so high
that mixing becomes difficult.  The stability of this  emulsion is good
but since it is a water in oil emulsion,  it cannot  be  further dis-
persed in water without very high power requirements for agitation.

4.  Nascent soap is formed in the emulsion leading  to  good stability.
If free fatty acids exist in the oil, and alkali is present  in the
water, then soap will form as the two phases are mixed together.  This
in-situ formation of the emulsifying agent (soap) leads  to good
dispersion.
            CHEMICAL COMPOSITION OF OIL SPILL DISPERSANTS



Oil spill dispersants typically contain three types of ingredients.

1.  Surface active agent.  This is the major active component.
                              - 8 -

-------
2.  Solvents.  Solvents may or may not be present for the purpose
of diluting the active  ingredient, to modify viscosity, adjust
freezing point, and to  enable more rapid blending or miscibility with
oil.

3.  Additives.  Additional ingredients may be present for the purposes
of modifying pH, corrosion inhibition, color and appearance,  to
increase hard water stability, and as aids in dispersion properties.

Since the surface active  agent is such a vital component of an oil
spill dispersant we will  examine the chemical composition of  these
substances at some length.
Surface Active Agents Used in Oil Spill Dispersants

1.  Soaps.   The oldest and best known of anionic emulsifying agents
are the soaps.  Soaps are  the salts of the long chain fatty acids
derived from naturally occurring fats and oils.  The fatty acids
present in  a soap will vary from 8 carbons to 22 carbons in chain  length
and may possess varying degrees of unsaturation depending on the fat  or
oil employed.  Typical soaps are made by reacting an alkali metal,
ammonia or  amine with fatty acids derived from coconut oil, rosin  and
tall oil.  Soaps are highly effective emulsifying agents when used in
soft water.  High concentrations of dissolved salts and electrolytes,
particularly heavy metal ions, will precipitate soap.  Because of  this
sensitivity to electrolytes, soaps are not normally the major emulsi-
fying component in oil spill dispersants designed for use in seawater.

2.  Sulfonated organics.  In this group we include the alkali metal,
ammonia or  amine salts of  sulfonated petroleum oils (petroleum sulfon-
ates or mahogany acids), and the much larger and more commonly used
group of alkyl aryl sulfonates.

The earlier forms of these compounds were prepared by sulfonation  of
alkyl benzene with fuming su If uric acid and later with sulfur trioxide.
These compounds are often referred to as ABS.  The branched chain
alkyl structure of these sulfonates produced very stable molecules
incapable of rapid biodegradation (so called hard detergents).  In
1965 the detergent industry switched over to the use of straight chain
or linear alkyl benzene which produced much more readily biodegradable
'sulfonates  known as IAS.  The most widely used of these sulfonates is
the linear  dodecyl benzene sulfonate.
„ <>
                 12            -            Na
                               -  9 -

-------
The formula illustrates the anionic  character of  the  sodium  salt of
dodecyl benzene sulfonic acid.   The  sulfonate may be  neutralized
with any of the alkali metals,  ammonia or amines  to produce  the neu-
tral salt.  The amine salts are known to  possess  greater solubility
in oils and are more widely used as  emulsifying agents.

Sulfated organic compounds comprise  another large group  of surface
active agents, but they are not as widely used for oil dispersing.
These differ from the sulfonated compounds in that the sulfur atom is
linked to the parent molecule through an  oxygen atom, whereas the
sulfonated sulfur atom is joined to  a carbon atom of  the parent mole-
cule.

3.  Phosphated esters.  This group  of anionic surface active agents
is prepared by reaction of the hydroxy group of a long chain alcohol,
alkyl phenol, or linear ethoxylate with phosphorus pentoxide or  some
other source of the phosphate radical.  Conditions of reaction may be
controlled to produce varying proportions of mono or  di-ester in the
end product.

                                   0    .   +•
                                   H _ —o   TI       R =  Allcvl
Monoester          R(OCH9 CH2 ) -0-P^ .  %      *   AiKyL
                        *   *  n    ^-0   H              phenol
                                                   n =  1 to 20
 Diester           [R(OCH2 CH2 )nO]2P-CT  H
 These free acid phosphate esters may be neutralized with alkali metals,
 amines or ammonia to produce extremely stable compounds with very good
 tolerance for dissolved salts.

 4.  Carboxylic Acid Esters of Polyhydroxy Compounds.  This large group
 of nonionic  surface active agents are produced by the reaction of fatty
 acids with polyhydric alcohols or by direct addition of ethylene
 oxide to the carboxylic hydroxy group.  By varying the chain length of
 the fatty acid as well as the amount of ethylene oxide used in the
 reaction, the HLB may be controlled.  As an example, the monostearic
 acid ester of triethylene glycol (3 moles of ethylene oxide) is illus-
 trated below.
                               CO (OCH2CH2 ) 3OH
                               - 10 -

-------
5.  Ethoxylated alkyl phenols and alcohols.   This group  of nonionic
surface active agents is derived by adding ethylene  oxide in varying
quantities to alkyl phenols or long chain  alcohols to  produce very
stable ether structures.  With the trend toward more readily biode-
gradable surface active agents, straight chain alcohols  are more and
more replacing the branched chain ethoxylated nonyl  and  octyl phenols.
Eleven to 15 carbon chain alcohols are typically reacted with from 5
to 10 moles of ethylene oxide.  Higher ethylene oxide  content provides
greater hydrophylic character, and lower ethylene oxide  content
provides greater hydrophobic properties or oil solubility.


R\/ 0(CH2CH2°^n-lCH2CH2OH     n  = Average number  moles of ethylene
                                     oxide
Alkyl phenol ethoxylate APE     R  = Alkyl-Octyl,  nonyl


R -0(CH2CH20)n_1CH2CH2OH        R* = C11H23 to

Linear Alcohol ethoxylate LAE        C1^1
This series of nonionic adducts exhibits inverse solubility properties
in water.  The temperature at which an aqueous 1% solution of the com-
pound becomes cloudy when heated is known as the cloud point (23).

6.  Block Polymers.  These nonionics are typically prepared by adding
polyoxyethylene to both ends of a polyoxypropylene chain, with both
ends of the resulting molecule termination in hydroxy groups.


            HO(CH2-CH2-0)a(CH-CH2-0)b(CH2-CH20)cH
                           CH3


By varying the numerical value of a, b and c, a wide variety of mole-
cules may be synthesized to produce the optimum HLB for a particular
use.

7.  Alkanolamides.  These nonionic compounds were first reported by
Kritchevsky in 1957 and are often referred to as Kritchevsky conden-
sates (9) (10).  Those most widely used in dispersant formulation are
                                - 11 -

-------
prepared by reacting two  moles  of  diethanolamine with  one mole of
fatty acid according to the reaction


       RC02H  +  HN(CH2CH2OH)2  	*•  RCON(CH2CH2OH)2 +  H20


R = Fatty acid-frequently coconut  fatty acid or lauric acid.  Commer-
cial products produced by the above reaction will usually be a mixture
containing:

                 Diethanolamine -  29.0%
                 Fatty Acid     -   4. %
                 Ester Amide    -   3.5%
                 Diethanolamide -  63.5%

The members of this group of compounds  are usually  quite viscous and
may solidify at room temperature.
Solvents.  Since many of the surface active agents  applicable  to  oil
spill dispersant compounding are viscous  or solid materials, some form
of solvent is often necessary in order to reduce viscosity  for ease of
handling.  In addition, the solvent  may act to dilute the compound for
economic reasons, to depress the freezing point for low  temperature
usage, to enable more rapid solubility in oil, and  to achieve  optimum
concentration of surface active agent for performance reasons. The
presence of a suitable solvent also  serves to  thin  the oil  to  be  dis-
persed, reducing viscosity and making it  more  easily emulsif iable.

The three general classes of solvents used in  oil spill  dispersants are
petroleum hydrocarbons, alcohols or  other hydroxy compounds, and  water.
Petroleum hydrocarbons provide good solvency to aid in penetrating the
oil.  They reduce viscosity and generally have low freezing points.
They are also low in cost.  Usually petroleum fractions with boiling
points above 300°F are used, and these may produce finished dispersants
with flash points as low as 110°F.  The proportion of aromaticity is
significant, since this effects solubility and emulsification proper-
ties as well as toxicity.  It is significant that Smith (7) in describ-
ing the Torrey Canyon aftermath reports that the aromatic  solvents used
were 10 times as toxic to marine life as were the surface  active agents.
Some typical fractions of applicable petroleum solvents include mineral
spirits, kerosene, #2 fuel oil, and heavy aromatic napthas which contain
significant quantities of higher alkylated benzenes.
                                - 12 -

-------
The alcoholic or hydroxy group of solvents  include alcohols, glycols
and glycol ethers.  These solvents also  lower the viscosity as well
as the freezing points of finished dispersants.  In addition, they
furnish a co-solvent effect, often needed to mutually  dissolve the
various ingredients in a dispersant for  stability of the compound in
storage.  This group of solvents may be  used in  conjunction with
petroleum hydrocarbons as well as with aqueous solvent  systems.  Some
of the more frequently encountered chemicals in  this group include
ethyl alcohol, isopropyl alcohol, ethylene  glycol, propylene glycol,
ethylene glycol mono methyl ether, ethylene glycol mono butyl ether,
and diethylene glycol mono methyl ether.  The more volatile members of
the group are quite flammable.

Water is the least toxic, least hazardous and most economical of the
solvents.  It does suffer, however, from a lack  of solubility or
miscibility with oils.  Where water is used as the solvent, special
problems exist in the choice of surface  active agents  and other addi-
tives in order to provide the necessary  miscibility with oils.  Glycols
and alcohols are used to aid in miscibility as well as freezing point
depression when water is used.

Other Additives

Alkalis                       - to modify pH

Sodium phosphates             - dispersant aid and water softening

Sodium silicates              - corrosion inhibition and dispersing  aid

Ethylenediamine tetraacetic
  acid                        - water softening, heavy metal  chelation

Lignin sulfonates             - dispersing aid

Polymerized alkyl napthaline
  sulphonates                 - dispersant aids

Sodium nitrite                - corrosion inhibitor

Dyestuff                      - color identification

Aromatic oils                 - covering odor

Because oil  spill problems have been spotlighted in the news  in  recent
years, the number of commercially available dispersants has grown
rapidly.  The imaginative trade names and prices of  these compounds
sometimes bear little relationship either to efficiency in dispersing
oil spills or possible toxic effects.  Studies of  the available  products
                              - 13 -

-------
indicate that large segments of  them can be described through  the
following classes:
Surface Active Agent
 Solvent
Ethoxylated alcohols & alkyl phenols  - Petroleum
Ethoxylated alcohols & alkyl phenols  - Aqueous  & Additives
Polyhydric Fatty Acid Esters          - Aqueous  & Additives
Alkanolamide                          - Aqueous  & Additives
Sulfonated An ionics                   - Petroleum
Combinations of above                 - Aqueous  or Petroleum

The aqueous solvent may include such water soluble solvents as  alco-
hols, glycols and glycol ethers.
         PRODUCTION AND PROPERTIES OF OIL SPILL DISPERSANTS
The manufacture of oil spill dispersants  usually involves  a  simple
blending or compounding operation.   The various  ingredients  may be
manufactured elsewhere to dependable reproducible standards.  The
compounder must exert considerable  care  in developing the  final formu-
lation as well as in developing analytical methods for quality control
purposes.  The following standardized tests are  useful (11):
Test

Color


pH

Free Alkalinity or free acid

Total Alkalinity

Water Content

Non volatile matter

Volatile hydrocarbons


Flash point
 Method Reference

 A.S.T.M. Color D1500-64
 Saybolt Color D156-64

A.S.T.M. E-70

 A.S.T.M. D820-58

 A.S.T.M. D800-58

 A.S.T.M. D800-58

 A.S.T.M. D820-58

 A.S.T.M. D800-58 also
 MIL-C-20207C (12)

 Cleveland open cup A.S.T.M.
 D92-66.Pensky-Martens closed
 tester A.S.T.M. D-93-66
 Tag closed tester A.S.T.M.
 D56-64

-------
Test

Spec if ic gravity

Foaming Properties

Loss on ignition (Ash)

Pour Point

Aromatic it y of Petroleum
  Hydrocarbon Solvent
Method Reference

A.S.T.M. D1298-67

A.S.T.M. D1173-53

A.S.T.M. D 800-58

A.S.T.M. D  97-66

A.S.T.M. D1019, in conjunction
  with A.S.T.M. D875
The reputable supplier will perform a sufficient number of  these
tests or others of his own design to confirm the desired composition
and performance properties.  He will also retain a sample from each
batch for periods of six months to two years.
Relationship of Physical Properties to Performance

Viscosity, as stated previously, will effect ease of application.
Viscous materials require more powerful pumping equipment.   Lower
viscosity compounds are likely to penetrate oil faster.  The Bureau
of Ships Specification MIL-S-22864C13) for oil spill dispersants has
established a viscosity maximum of 400 centistokes at 20°F.  Such  a
viscosity insures ease of application, but it could well rule out
somewhat more viscous products which may have other advantageous
properties.

Flash point information will indicate the flammability hazard associ-
ated with the use of a dispersant.  The Bureau of Ships Specification
MIL-S-22864 has established a 150°F minimum.  In view of the fact  that
oil and fuel spills often represent severe fire hazards, one may well
question the advisability of using a dispersant which has a fire hazard
of its own.

Composition data will provide information concerning toxicity problems
in the handling and use of the dispersant.  The presence of highly
volatile ingredients may prevent usage in confined areas.

Uniformity and stability of the dispersant in storage is essential, so
that stratification or changes in properties do not occur over long
periods of time in widely varying storage and temperature conditions.
                               - 15 -

-------
Dispersants are supplied ready for use  or as a concentrate which  is
diluted with water or petroleum solvent  just prior  to  use.   Since
oil spills invariably occur without warning, dilution  prior  to use
is often impractical, and the compound  is best forwarded to  the site
of application ready for use.

Biodegradability information concerning  the dispersant, as well as
the individual surface active agents  present in the compound, will be
significant in predicting long term contamination caused by  the use
of the dispersant.
             CHEMICAL ANALYSIS OF  OIL  SPILL DISPERSANTS
Detailed analytic techniques are beyond the scope  of this report, but
it will be helpful to set down  several  procedures  which will establish
the general chemical nature of  unknown  samples.

Volatile solvents may be determined by  a conventional  distillation.
If the solvents are not water soluble,  the  steam distillation technique
outlined in military specification  MIL-C-2027C(12) is  useful.

Procedure:  "The solvent content shall  be determined by weighing  100 +
or -0.1 gram of compound into a 1,000 mis.  round bottom flask.  Add 20
grams of anhydrous barium chloride  dissolved in 100 mis. of water to
the flask and steam distill the mixture.  Collect  the  distillate  in
250 mis. graduated cylinders.  Continue the distillation until  not more
than 1 ml. of solvent comes over with 250 mis. of  distillate.   Note the
total volume of non-aqueous layers.  Combine the solvent portions from
the several collecting cylinders, dry over  anhydrous sodium sulphate,
and filter through a #12 fluted Whatman filter paper.  Determine  the
specific gravity at 25°C using  a pycnometer; calculate weight % of solvent
as follows:

            vol. non-aqueous steam  distillate x specific gr.
% solvent =                 weight  of sample                x  100
The solvent shall be reserved for mixed aniline and chlorinated solvent
analysis.

Ash test or loss on ignition will determine  the presence  of  inorganic
ingredients.
                                - 16 -

-------
Analysis for the surface active agent is of most interest
Where detailed specific information is required it would be well  to
start with the method for "Separation of Active Ingredient  from Sur-
factant; Syndet Compositions A.S.T.M. D2358-65T" (11).   Once the
surfactant has been separated, A.S.T.M. method D-2357-65T may be
used for "Qualitative Classification of Surfactants by  Infra red
Absorption" (11).

In the absence of instruments, qualitative analysis of  the  surface
active agent may be performed as outlined below.
Separation of Surface Active Agent

Dry sample in an oven at 105°C.  Reflux sample in 95% isopropyl  alco-
hol for one hour.  Filter hot through Whatman #1 filter paper.   The
filtrate will contain the surface active agent.  The residue,  if any,
will contain inorganic salt such as sodium phosphates or silicates.

Evaporate the filtrate over steam bath and save for qualitative  tests.
Ionic activity tests (Weatherburn Test) (15)

Reagents:     1.  Dye solution:  0.03 grams methylene blue, 12 grams
                  concentrated sulfuric acid, 50 grams anhydrous
                  sodium sulphate dissolved in water to make a total
                  of one liter solution.

              2.  Anionic surfactant solution-0.05% Aerosol OT
                  (Sodium dioctyl sulfo succinate).

              3.  Chloroform
Ionic activity tests  (Weatherburn Test) (15)

Procedure:    1.  Into a 25 mis. test tube place 8 mis. of dye solu-
                  tion and 5 mis. chloroform.  Add anionic surfactants
                  solution drop by drop, shaking vigorously between
                  drops and allowing phases to separate.  Continue
                  adding dropwise until the two layers are equal in
                  color and intensity viewed in reflected light.  Usu-
                  ally 10 to 12 drops of anionic solution are required.

              2.  Now add 2 mis. of 0.1% solution of the unknown and
                  shake vigorously.
                               - 17 -

-------
Results  :     1.  Chloroform phase (lower)  is deeper in color
                  and aqueous phase is mostly colorless —
                  anionic is positive.

              2.  Water phase (upper) is deeper in  color than
                  the chloroform phase — cat ionic  is positive.

              3.  Both phases are more or less the  same color —
                  probably a nonionic.

              4.  If the aqueous phase has  become milky and
                  hence slightly lighter in color,  it may still
                  be nonionic.  Soaps do not react  in this pro-
                  cedure.  If both anionics and nonionics are
        ,          present, the reaction of  this test will be
                  an ion ic po s it ive.
Test for Nonionic (Cationic  Negative) (16)

Reagents :     1.  Cobaltothiocyanate solution:   dissolve 24.0
                  gms.  of  potassium thiocyanate  (KCNS)  and 1.0
                  gms.  of  cobalt  nitrate (CoNOj  .  6H20) in
                  distilled  water and dilute to  100 ml. volume.

              2.  0.2 N sulfuric  acid:   dilute 5.6 mis. of cone.
                  H2SO^ (sp. gr.  1.84)  to 1 liter  with  distilled
                  water.

Procedure:     1.  To 5  mis.  of active ingredient solution add
                  0.2 N H2SO^ dropwise  until the pH is  about  4
                  (to indicator paper).

              2.  Add 1 ml.  of the Cobaltothiocyanate solution.
                  A blue color is positive for nonionic material
                  providing  no cat ionic material is present.

                  This  test  is negative for diethanolamides.
Ignition Test

Ignite .about 0.1 grms. of surface active agent  on a platinum dish
or piece of platinum foil.  Heat gently at  first and then  more
                            - 18 -

-------
strongly.  Note:  (1)  Ease with which compound burns  and decom-
poses.  (2)  The odor of gases or vapors evolved,  and  their
acidity or alkalinity to litmus.

Ignite strongly until almost all organic  matter has  burned off.
Cool, add a few drops of 30% hydrogen peroxide, and  reheat.   If
unburned carbon still remains after strong ignition, cool and add
several drops of concentrated nitric acid.  After  standing a  few
minutes, heat gently until acid has vaporized and  then ignite
strongly.  Repeat the nitric acid addition if necessary until
carbon has burned off.
Discussion;

A residue after ignition indicates the presence of a metallic  ele-
ment .  If the residue is less than about 1% it is probably contam-
ination and may be disregarded.  The metal is likely an  alkali
metal and may be confirmed by standard procedures.

Alkaline vapors during the ignition indicate the presence of
ammonia or amines.  Odor of buring sugar indicates glucose, sorbi-
tol or carbohydrate derivatives.  Proteins char with a characteris-
tic odor.

Acidic fumes are given off by salts of volatile organic  acid and
by organic sulphates and sulphonates.

If ash is present, run a standard sodium fusion and use  the result-
ing solution for analytical tests for the presence of  sulfur,
nitrogen, halogen, phosphorous.  If no ash is present, the surface
active agent is likely to be nonionic.

If nonionic, and the aqueous solution shows no cloud point on
heating, it could be a diethanolamide — nitrogen should be confirmed.

The reader is referred to "Systematic Analysis of Surface Active
Agents" by Rosen-Goldsmith (14) for more complete data on analytical
procedures.
                 MEASURING OIL SPILL DISPERSING POWER
Thus far we have discussed physical and chemical procedures which
are fairly well documented.  Our most vital property — ability to
                            - 19 -

-------
disperse an oil spill under a wide variety of conditions  —  is
the most difficult property to evaluate.   Very  little  has been
published on this  subject.   The problem  is complicated by the
large number of variables present  under  field conditions; to
enumerate a few:

      Composition  of water
      Composition  of oil
      Viscosity of oil
      Temperature  of water and air
      Length of time oil  has been  exposed to atmosphere
      Wind and wave conditions
      Tidal conditions
      Amount of agitation possible
      Method of application of dispersant

In reporting on the Torrey Canyon  oil  spill, Snith  (7) writes that
the oil was Kuwait crude  oil which contains about 25%  volatile
ingredients.  As evaporation took  place  the oil became progressive-
ly more viscous.  It was  theorized that  after about  three weeks  at
sea, approximately 15% of the original amount would  remain as a
black tarry asphaltic residue.  Evaporation, photo oxidation, as
well as bacterial  degradation, all contribute to these changes.

The Plymouth Laboratories (7) used an  unsophisticated  method of
determining stability of  detergent —  oil emulsions.   Two mis. of
detergent, 2 mis.  of Kuwait crude  oil, and 96 mis. of  sea water
were shaken up in  a 100 ml. graduate cylinder and allowed to stand.
Observations were  made of the appearance of the resulting emulsion.

A few emulsion stability  tests reported  in the  literature are
worthy of note.
Emulsion Stability of Soluble Cutting Oils.   A.S.T.M.  D1479-64 (17)

Soluble oil is dispersed at a suitable concentration in  an  appropri-
ate test water.  The emulsion is stored for  24 hours after  which
the bottom fifth is separated for determination of  oil content.
Comparing oil concentrations in stored and freshly  prepared emulsion
gives "the percentage of oil depletion".

In the procedure, 125 mis. quantities of emulsion at various dilu-
tions are stirred in 250 mis. beakers with a four bladed beater at
500 + or -50 rpm for one minute.  Emulsion is poured into a 125 mis.
                             - 20 -

-------
separatory funnel and allowed to stand for 24 hours, after which
the bottom 20% is drawn off and transferred to  a  Babcock milk-
test bottle.  Sulfuric acid is added to break the emulsion and
the bottle is centrifuged 10 minutes or until separation is com-
plete.  Stability is calculated in terms of "percentage of oil
depletion" at a specified dilution in a given test water.
Bnulsion Characteristics of Petroleum Oils  and Synthetic Fluids.
A.S.T.M. D1401-67 (17)

This method describes a procedure for measuring the ability of
petroleum oils or synthetic fluids to separate from water after
an emulsion has been formed.  A 40 mis.  sample of  oil and 40 mis.
of distilled water are stirred for 5 minutes  at 130°F.  in a gradu-
ated cylinder.  Stirring is done using a 3/4  inch  wide  by 4 3/4
inch long paddle at 1500 + or -15 rpm.  The time required for the
separation of the emulsion thus formed is recorded.  If complete
separation does not occur after 1 hour,  the volumes of  oil, water,
and emulsion remaining at the time are reported.
Emulsion Test from Specification MIL-C-22230 (Ships).   (18)

While this test was designed to measure cleaning ability,  it  is of
interest as a basis for developing a laboratory procedure  for
evaluating oil spill dispersants.

The method calls for preparing a standard soiled metal panel  by
coating it with Navy Special Fuel Oil.  The panel is immersed in  jars
containing cleaning compound solutions and subjected to agitation on
the Fisher oscillating hot plate.  The solution is then poured out
and allowed to settle for 24 hours.  Benzol is poured on the  surface
and gently stirred to dissolve the supernatant separated oil  phase.
Using a syringe, a quantity of the benzol-oil solution is  removed
and the oil content determined photometrically.
Specification MIL-C-22230 also calls for a test of miscibility with
sea water.  This may well be a legitimate test for screening out
dispersants normally used in fresh water which might break down
quickly in sea water.  In this test one part  of compound is mixed
with 100 parts of sea water.  The solution is heated and agitated  for
1 hour and no visible separation of compound  should be noted after
this period of agitation.  For oil spill dispersants the sample
should not be heated, but the test run at room temperature.
                              - 21 -

-------
Sea water as used in MIL-C-22230 is  3  1/2  percent  sodium chloride
solution.  The presence of magnesium and calcium  in  sea water
could be significant in affecting the  performance  of  oil spill
dispersants.  For laboratory tests it  is suggested that the  syn-
thetic sea water be prepared according to  specification MIL-C-
7907A (19).

         Sodium Chloride                   150.0 grams
         Magnesium Chloride, hexahydrate    66.0 grams
         Calcium Chloride, dihydrate         9.6 grams
         Sodium Sulfate, anhydrous          24.0 grams
         Water to make a total of            6   liters

In developing military specification MIL-S-22864 Solvent-Eknulsify-
ing, Oil Slick, the Navy has done the  most significant work  in
establishing a performance test for  these  compounds.  This procedure
is useable as a basis for developing more  general  procedures
applicable to a wider variety of conditions.  Some approaches which
bear further investigation are discussed below:

(a)  Simulation of wave action should  be explored.  Changes  in design
of the pumping mechanism or possible baffling of the  tank may be
useful in this connection.

(b)  The test should be conducted using water of the  types encountered
in the field.  Products for use at sea should be tested using
synthetic sea water.  Products offered for use on  inland waterways
should be tested using water approximating the composition of these
waters.  In the Navy tank test initial agitation is obtained by hosing
the surface with fresh water.  While this  is  sometimes possible when
treating an actual oil spill, it is  more likely that  agitation will
be accomplished using the available  water  at  the site.  It is recom-
mended that the entire test be done  using  the same kind of water so as
to more closely approximate field conditions.

(c)  The nature of the oil can have  substantial influence on the
results obtained in this test.  Crude  petroleum from  different sources
varies significantly in content of sulfur, nitrogen,  and oxygen.  Those
crudes containing large proportions  of napthenic and  other carboxylic
acids, phenols and sulfides would be much  more sensitive to  alkaline
dispersants.  The test should be broadened to include a representative
group of crude oils as well as other fuels and vegetable oils.

(d)  The Navy tank test utilizes a one to  one ratio  of dispersant com-
pound to oil.  This high ratio of dispersant  to oil  is unrealistic  in
                              - 22 -

-------
terms of usual field conditions.  Those dispersants high  in  solvent
content will tend to lower the viscosity of  the  oil significantly —
to a much larger extent than would be the case  if  the  dispersant
were used at more usual proportions of one to five parts  oil.

A high specific gravity dispersant which is  soluble in the oil would
aid the oil in remaining below the surface of the  water during the
test, even if it were not an effective dispersant.

(e)  The method of adding the dispersant, by pouring from a  graduated
cylinder, needs improvement.  The oil tends  to  spread  around on the
surface in a very non-uniform manner, and it is  up to  the operator to
try to pour from the cylinder contacting as  much of the oil  as possi-
ble with the dispersant.  It would be well to develop  some sort of
ring device which would tend to confine the  oil  approximately at its
limits of spreadability.  The oil spill dispersant may then  be gently
flowed on or sprayed through a sieve-like device so that  it  will
spread more evenly over the surface.  In this way  lesser  quantities of
dispersant may be evaluated.  A dwell time,  allowing the  dispersant to
permeate through the oil before agitation, is also significant.

(f)  Temperature control is important.  Certain of the nonionics are
very sensitive to temperature.  Oil viscosities vary  substantially with
temperature.  The temperature of the test should approximate typical
field conditions, and it would be helpful to develop  data which would
reflect subfreezing weather conditions.

(g)  The condition of any undispersed oil at the conclusion  of this
test is significant, and should be examined. After the  pump is turned
off, allow tank to settle for 30 minutes.  Now,  half  submerge the out-
side of a one liter beaker through the surface  at  the center of the
tank.  Hold it stationary in this position for  10  seconds, and then
remove it.  Using a wash bottle containing the  test water, determine
whether or not oil clinging to the outside of the  beaker is  readily
rinseable.
Laboratory Determination of Dispersing Potential;

The following procedure is suggested as a basis for development  of  a
laboratory evaluation of dispersants.  It is designed to estimate
dispersing potential under ideal conditions.

Varying quantities of dispersant are mixed with a constant quantity of
oil in order to establish the minimum quantity of dispersant  required
for effective dispersion.
                            - 23 -

-------
Weigh out four separate 5 gram portions  of  oil.   Mix  thoroughly
with 4 gram, 2 gram, 1 gram and 0.5 gram quantities of  dispersant
respectively.

Pour 500 mis. of specific test water into each of four  1000 mis.
beakers.  Place a metal panel in each beaker of  dimensions approx-
imately 1" by 7" by 1/16".  Transfer the treated oil  samples to the
beakers and agitate on the Fisher oscillating hot plate for one
hour.

Let stand one hour.  Now carefully remove the metal panel.  Holding
it vertically reimmerse it in the beaker, keeping it  there a few
seconds.  Now remove it and gently swirl it in a beaker of clean
test water.  Remove and swirl again in a second  beaker  of clean test
water.

Examine the metal panels for presence of coalesced oil. The minimum
concentration of dispersant required to  leave a  negligible residue
of coalesced oil on the panel may be called its  dispersing potential.

Other observations may be found useful.  For example  the aqueous
residue may be transferred to a separatory  funnel, and  dispersed oil
content determined after specified periods  of standing. The coalesced
oil clinging to the walls of the test beakers should  be noted  — and
may be determined photometrically by dissolving  in a  fixed quantity of
benzo1.

Multiple tests using synthetic sea water as well as fresh water are
suggested.  Temperature control may be facilitated through the use of
a constant temperature bath placed on the oscillating hot plate.
Field Testing:

The development of a meaningful large  scale  field test  is  an extremely
complex problem because it is difficult  to control  the  many variables.

Prior experience and discussions with  Marine and FWPCA  personnel  lead
to these thoughts:

(a)  Comparative tests must be done in similar tidal  and climactic
environment.

(b)  A specific quantity of test oil may be  retained  within a circular
boom.

-------
(c)  A weighed quantity of oil  spill  dispersant may be dispensed
using a pressurized fire extinguisher.  Alternately, smaller
quantities may be dispensed from garden type pressurized sprayers.
Special larger units may be constructed to handle  specific volumes
at specific pressures and with  fixed  spray nozzles.

(d)  Agitation may be accomplished with a fire pumper using care-
fully controlled pumping pressure, flow rate, and  time.

(e)  Effectiveness may be determined  by sampling within the retain-
ing boom 5 minutes and 30 minutes after agitation.  Tidal flow will
substantially affect these results.

In this discussion no mention has been made of treatment of oil
spills by eduction of the dispersant  into a  stream of water.  While
this method is effective for flushing an oil spill from one area to
another, it is rarely capable of producing effective dispersion of
large quantities of oil.  (See  the section on Emulsion Stability —
mode of addition.)  Oily materials which have spread to their ulti-
mate limits and which exist as  microscopically thin slicks may be
successfully treated with this  technique, using  the same chemical
dispersants.
                                - 25 -

-------
                           References
 1.  Kirk,  R.  E.,  Orthmer,  D. F.   Encyclopedia of Chemical Tech-
 nology,  Inter Science,  1963-1967,  2nd  ed.

 2.  Bennett,  H.,  et  al.  Practical Emulsions.  Chemical Publish-
 ing Co., 1968.

 3.  Becher, P. Emulsions; Theory and  Practice.  Reinhold,  1965.

 4.  Amer.  Soc. Testing  Mats.  D  459-64; Definitions Relating to
 Soaps and Other Detergents.

 5.  Griffin,  W. C.,  J.  Soc. Cosmetic Chemists.  1:311  (1949).

 6.  Griffin,  W. C.,  J.  Soc.Cosmetic Chemists.  5:249  (1954).

 7.  Smith, J. .E., (ed.) Torrey Canyon, Pollution and Marine jjjfe.
 Cambridge U.  Press,  1968.

 8.  Gorman &  Hall, J. Pharmaceut.  Sci.  52:442 (1963).

 9.  Kritchevsky,  W., J. Am. Oil  Chem.  Soc.   34:178 (1957).

10.  U. S.  Patents 2,089,212 and  2,173,058.

11.  Anier.  Soc. Testing Mats.  Standards.  Part 22.

12.  Military  Specification MIL-C-20207C - Cleaning Compound, Sol-
vent Emulsion, Grease Removing.

13.  Military  Specification MIL-C-22864A (Ships) - Solvent-Emulsi-
fier, Oil-Slick.

14.  Rosen, M. J.  and Goldsmith,  H. A.  Systematic Analysis  of
Surface Active Agents.   Inter Science,  1960.

15.  Weatherbum,  Can. Text. J. 71(16):45-6  (1954).

16.  Amer.  Soc. Testing Mats.  Committee D-12 files.

17.  Amer.  Soc. Testing  Mats.  Standards.  Part 17.
                              -  26  -

-------
18.  Military Specification MIL-C-22230 (Ships).   Fuel Tank and
Bilge Cleaner.

19.  Military Specification MIL-C-7907A (WP) - Cleaning Compound,
Dec ontaminat ing.

20.  Fisher Oscillating Hot Plate.  Fisher Scientific  Co.  Catalog.
Fairlawn, N. J.

21.  International Conference on the Pollution of the  Sea  by Oil.
Proceedings, 1962.

22.  Oil Spillage Study; Literature Search and Critical Evaluation
for Selection of Promising Techniques to Control and Prevent Damage.
Prepared by Pacific Northwest Laboratories (1967).  U. S.  Dept. of
Commerce Clearing House Document AD666289.

23.  Standard Methods of Chemical Analysis, Vol. Ill B, Edited by
F. J. Welcher, 6th Edition, 1966, D. Van Nostrand Co., Inc., Prince-
ton, p. 1839.
                                                    •ft GPO 963-404
                               - 27 -

-------