WATER POLLUTION CONTROL RESEARCH SERIES  15080DJN 1/71
       CELLING CRUDE OILS TO
      REDUCE MARINE POLLUTION
       FROM TANKER OIL SPILLS
JMENTAL PROTECTION AGENCY  WATER QUALITY OFFICE

-------
          WATER POLLUTION CONTROL RESEARCH SERIES

The Water Pollution Control Research Series describes 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 demon-
stration activities of the Water Quality Office, Environ-
mental Protection Agency, through inhouse research and grants
and contracts with Federal, State, and local agencies, re-
search institutions, and industrial organizations.

Inquiries pertaining to the Water Pollution Control Research
Reports should be directed to the Head, Project Reports
System, Office of Research and Development, Water Quality
Office, Environmental Protection Agency, Washington, B.C. 20242.

-------
         GELLING CRUDE OILS TO REDUCE MARINE
           POLLUTION FROM TANKER OIL SPILLS
                           By

                  THE  WESTERN COMPANY
                    Of North America
                   Research Division
             2201 North Waterview Parkway
               Richardson, Texas   75080
                         for the

                 WATER QUALITY OFFICE

            ENVIRONMENTAL PROTECTION AGENCY
                 Project No. 15080 DJN

                    January, 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1.25

-------
                EPA Review Notice
This report has been reviewed by the Environmental
Protection Agency and approved for publication.
Approval does not signify that the contents necessarily
reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or
commercial products constitute endorsement or recommenda-
tion for use.

-------
                              ABSTRACT
It was expected that techniques similar to those used in the oil well
service industry and certain petroleum end-product uses could produce
gelled crude oil in a damaged tanker, thereby precluding flow of free
oil  into the  sea.

Laboratory evaluations of petroleum solidification, or gelling, techniques
and systems were conducted utilizing foreign and domestic crude oils
over a wide  range of gravities and gelling agent proportions and concen-
trations. Methods for measuring gel strength were evaluated and the
most pertinent utilized in the  screening of candidate gels.  Concepts for
dispensing and mixing the gel agents were devised and tests were
conducted to verify the ability to form laboratory-quality gels in up to
1,000-gallon volumes.

The testing  identified an amine/isocyanate combination as  producing the
strongest gels,  over the  widest range of crude oil types,  and the ability
to form those gels in quantity with  two application systems suitable for
shipboard use.  The results suggest that the gels would stop flow through
a tank rupture under only a limited  range  of potential conditions,  but
that the flow rate might be significantly less than that of free oil
resulting in less pollution over a given time span. The results also show
that the gelled oil tends  to remain floating in  lumps  at the water surface
and will not permit a slick to be formed.   The ability of the gels to be
removed from tanks using conventional pumping apparatus was also
demonstrated.

Economics of applying the gelling technique in operation situations were
explored and system performance was related  to historic spill clean-up
costs indicating the economic feasibility  of the concept.

This report was  submitted in fulfillment of Contract  14- 12-497 under
the sponsorship of the Federal Water Quality Administration.
                                 11

-------
                         CONTENTS





Section                                                   Page
   1     INTRODUCTION	    1




   2     GELLATION TECHNOLOGY	    4




   3     LABORATORY EVALUATION OF GELS	   10




   4     LARGE-TANK TESTS OF THE GELLING SYSTEM	   31




   5     SYSTEM APPLICATION	   47




   6     ACKNOWLEDGEMENTS	   52




   7     REFERENCES	   53




   8     GLOSSARY	   56




   9     APPENDICES	   57
                            iii

-------
                            FIGURES



                                                          Page



1     Rheogram, Gelled Crude Oil	  29



2     Schematic of Recirculation System	  34



3     Schematic of Nozzle Dispensing System	  36



4     Pressure Differentials Cargo Versus Seawater	45
                                IV

-------
                             TABLES


                                                         Page

   I     MEASUREMENT METHODS FOR GEL CRITERIA	  10

  II     CRUDE OIL CLASSIFICATION	  15

 III     PHYSICAL PROPERTIES OF GELLING MATERIALS	  16

  IV     DELAYED GELLING OF CRUDE OIL	  21

  V     PENETROMETER READING OF GELS FOR FIVE-
        PERCENT AGENT CONCENTRATIONS	.  . .  .  24

  VI     SOLUBILITIES OF FIVE-PERCENT CLEARFORK GELS ....  25

 VII     GELLING UNDER VARIOUS TEMPERATURES	  25

VIII     COST TO GEL ONE GALLON OF CRUDE OIL .	  26

  DC    TEN-PERCENT GELS .	  27

   X    VARIOUS MATERIALS GELLED	  28

  XI     RECIRCULATION TESTS (CONCEPT 2)	  37

 XII     RECIRCULATION TESTS (CONCEPT 3)	  38

XIII     NOZZLE TESTS (CONCEPT 1)	  41

XIV     NOZZLE TESTS, GAS AGITATION	  42

 XV     EQUIPMENT LIST, CONCEPT 1 GEL SYSTEM,
        SHIPBOARD INSTALLED	  50
                             v

-------
                            CONCLUSIONS


A system for gelling crude oil has been developed which can result in a
reduction in the marine pollution caused by the accidental discharge of
crude oil from a damaged  tanker.  Both the chemical and mechanical
aspects of the concept are feasible.  However, effectiveness of the system,
in terms of pollution reduction,  is a function of the time lapse to initiate
gelling, rupture size and  rupture location in relation to the waterline.
Complete stoppage of cargo flow through the leak would be accomplished
for only a limited range of expected shipboard distress  conditions. Further
reduction in flow of the gelled cargo, over a more realistic range of
potential operational conditions, would require a significantly stronger
gel.

When pressure and hole-size  conditions exist sufficient to cause  the
gelled oil to flow, the gelled  product will leave the ship at a slower rate
and the total loss to the sea over a given time  span would be less than
for ungelled oil.  The gelled oil that does flow into water floats, tends to
remain in lumps and does  not  form a slick.

The range of domestic  and foreign crude oils tested in this program can be
gelled using a 0.75:1.00  to 1.50:1.00 equivalent weight ratio combination
of amine and isocyanate in concentrations of 5 percent of the crude oil
gelled. The chemicals are available in quantity  and can be handled and
stored without elaborate or unusual precautions.

The optimum gels produced with this system can  be expected to  exhibit
static  shear strengths  in the range of 12, 000 to 25, 000 dynes per  square
centimeter,  as determined by  penetrometer, extrusion rheometer, and static
head tests.  The strength variation is due primarily to the variation in
composition of crude oil.  The gelled oils produced can be unloaded from
the tanker by using conventional positive displacement pumps.

The gels produced are  the strongest possible using five percent of a system
which has only a physical bonding attraction for  the crude oil.  A gel
produced by chemical bonding of the crude oil may be  possible,  but the
state-of-the-art does not permit a speculation of the composition, charac-
teristics or cost of such a system at the present  time.

The program demonstrated that the gels can be produced in 1, 000-gallon
quantities and of a quality equal to those produced in the laboratory using
two different dispensing and mixing concepts.  Such large volume production
can be achieved using a pumped recirculation system with or without air
added, or with a system of multiple nozzle probes traversing the tank to
be gelled.

The chemical storage and dispensing equipment can be designed for ship-
board installations or in appropriate size and weight modules to permit
helicopter or surface transport modes to the point of need.
                                 vi

-------
The cost of having the chemicals and equipment for the gel system installed
on a ship for immediate use is estimated at 1. 13 mils per ton-mile.  This
represents an increase of four percent in average crude oil shipping cost.
The cost of having a gel system capable of being transported by helicopter
or surface vessel to the point of need, exclusive of the transport costs,
is .002 mils  per ton-mile of current United States coastal oil transport.
This represents a 0. 18-percent increase in average shipping cost.

To economically justify the cost of chemicals and equipment in relation to
historic spill cleanup costs,  the total transportable system assumed as
an economic  model must be capable of precluding 23,000 barrels of oil
loss to the sea annually.  Each shipboard-installed system must be capa-
ble of precluding 1, 500 barrels of loss annually.
                                  vii

-------
                        RECOMMENDATIONS
On the basis of the conclusions,  the following recommendations are
offered.  The action and research of the nature herein proposed could
materially reduce the marine pollution caused by accidental oil spills by
tankers.

      1.   The gel system devised should not be implemented in
          the waterborne  oil transport industry at  this time.

      2.   Further work  to develop a stronger or better gel should
          not be undertaken until a more  precise definition is made
          of the gel strength and flow properties required by the
          operational environment.  Additional research should be
          undertaken to determine the conditions under which any
          gel would be expected to perform.  This should include
          an analysis of past disasters causing marine pollution,
          with particular  emphasis on the ship condition, as well
          as the pollution aspects. The  frequency and severity of
          the disaster and the nature of,  and increase in,  damage
          during the period following the disaster should be given
          primary consideration.  A definition of the gel charac-
          teristics necessary to meet various limits within the range
          of these  conditions, together with detailed cost-benefit
          analysis, should be made.   The specific output of this
          work must be a  determination of the operational conditions
          the gel must  satisfy and the resulting gel strength neces-
          sary to make implementation of the concept feasible.

      3.   If a stronger  gel is  required to  justify implementation,
          additional research should  be  performed to develop a gel
          which will meet the pre-identified requirements.   A new
          approach directed toward chemical  (rather than physical)
          bonding should be initiated as  part of such work.

      4.   Large-scale tests should be made to fully develop the
          existing  or improved gelling method and equipment neces-
          sary to implement the working concept.   A full analysis
          of the flow and strength  characteristics of the gel under
          modeled  operational conditions must also be completed.
                               viii

-------
                              SECTION 1

                           INTRODUCTION


                                Scope
This program identified and evaluated the concept of gelling systems for
crude oil as they related to the problem of accidental spills from tankers.
The scope did not include other marine oil pollution problems, such as
offshore production, sunken tankers, natural seepage,  or the overboard
discharge of tank washings and dirty ballast water.
                               Purpose
The program was designed to establish and develop the basis for a working
system which will cause crude oil to be thickened,  or gelled,  sufficiently
to prevent its escape from a damaged tanker.  The gelling materials, when
introduced into leaking tanks,  were to cause gelling of the oil sufficient
to stop its flow from the tanker.  System development required the identi-
fication and evaluation of gelling agents,  methods to introduce the agents
into the oil,  testing of the system on volumes of oil significantly larger
than laboratory scale to verify the design basis of a large-scale working
system, and cost or cost estimates of the  concept.
                              Background

Accidental oil discharges into the seas and waterways are a source of
pollution which is a hazard to marine and bird life.  This pollution also
may be an  actual, or potential,  health hazard to man as well as a threat
to the  natural environment that augments his recreation and aesthetic
enjoyment.  Appendix A, Shipping Characteristics, gives more details on
pollution possibilities.

Petroleum and petroleum products accounted for 37 percent of the  1, 272, 900
tons of United States waterborne commerce in 1965. l  Crude oil imports
are being received in tankers at the rate of 1, 380, 000 barrels per day at
United States docks.2

Recent oil  spills from waterborne transport of crude oil have demonstrated
a need for  prevention of these pollution sources.  The Torrey Canyon
grounding off Lands  End, England in 1967, and the breakup of the Ocean
Eagle near San Juan,  Puerto Rico in 1968 are but two examples of many
such oil spills from  tankers.  These accidental  crude oil spills may be
caused by  collision,  strandings, structural failure, fire and explosion.

After considering several alternate  approaches directed at solving or
reducing oil spill pollution,  it was concluded that the optimum point in

-------
the chain of events at which to attack the problem is when the oil is still
in the tanker.  At this point, it is possible to eliminate the pollution
problem and also deal with the oil most readily.   The wave and current
action has not yet spread and dispersed the oil;  thus,  treating, gathering
or disposing of the oil would be greatly reduced  when accomplished in
the tanker.

The oil field service  companies routinely gel large volumes of crude oil
in their field servicing operations.  These companies own mobile gelling
units which mix chemical agents to create the gels,  often in large tanks
of crude oil.

Based on this experience,  it was  proposed that the spillage of oil from a
distressed tanker could be prevented by rapidly  gelling the oil in a com-
partment in the ship.  The formed gel would block the  flow of oil from, and
the flow of water into, the compartment.   It was postulated that such a
gel could be made using a relatively small gelling agent concentration.
Since only the ruptured  compartment (or those in imminent danger of rupture)
must be  treated,  the  gelling agent and effort required could be small when
compared to materials and effort required to treat the oil after it had been
spilled.


                   Program Description and Approach

The program reported herein was performed in two consecutive phases.
The objectives of the first phase,  or Basic Contract, was to identify and
develop in the laboratory,  a process or chemical agent that would produce
 a suitable gel utilizing the largest  number of crude oil types.  The second
 phase, or Option I,  objective was to devise the mechanical  system neces-
 sary to accomplish gellation and  to demonstrate the ability of the system
 to gel a selected crude oil in larger-than-laboratory volumes (1, 000 gallons),

 Prior to actual development of gelling materials, it was necessary to con-
 duct a literature search and a survey of in-house experience with materials
 and methods used.   This investigation insured that all potential processes
 were  considered and that  the best materials could be selected for screening.

 It was also necessary to search the literature concerning the shipping of
 oil and  ship disasters involving spillage of oil.  Once the nature and
 occurrence of the problem and its limits  were determined, the operational
 criteria which the gelled crude oil  must meet was described.  The experi-
 mental procedure to  be used in the laboratory for measuring the various
 processes' performance against the criteria was then defined.

 The laboratory evaluation and development of the gelling materials pro-
 ceeded  with the wide screening of  candidate materials in a single crude
 oil.   After the candidate gelling materials were  screened, the most promis-
 ing materials  were further tested in various typical crude oils shipped  in
 ocean-going tankers.

-------
Once the best gelling materials were determined in the laboratory, tests
were conducted to demonstrate their ability to be formed in significantly
larger volumes of oil.  As part of these 1, 000-gallon tests, evaluations
were made of various dispensing and mixing methods devised for applying
the gel constituents to the larger volumes of oil.

-------
                              SECTION 2

                       GELLATION TECHNOLOGY


                           Gellation Systems

The literature survey revealed that none of the reports contained infor-
mation directly pertaining to the  gellation or solidification of crude oils;
however,  several detailed reports were obtained containing data concerning
the solidification of fuel oils,  particularly jet fuel.  A comprehensive
listing of reports of this type is given in the Bibliography.  Related reports
and information are found under listings such as hydrocarbons,  solid fuels,
aircraft fires, frac-oils, oil spillage/recovery, gels, and jet engine fuels.
A few indirect listings are urea,  polymers,  urethanes, amides,  metal
stearates and soaps.


Gel Definition

The solid products derived from the action of various agents on liquids
have been described in various ways.  The term "gel" is entirely appropri-
ate, but the diverse macroproperties  of a "gel" make the description a
broad one.  A more limiting description is that the liquid has been "solidi-
fied. "  In any instance, the product desired is one that no longer flows
under low to medium forces.

The phenomena which instigate properties of a gel claim an equally wide
variety of descriptions.  But in all instances,  the  material which is  added
to a liquid (in this case,  crude oil),in some manner  limits  the movement or
flow of the liquid.  An agent which produces a gel also  produces a colloid;
i.e.,  finely divided particles of one  material suspended in another material.
The suspended particles may be insoluble liquid particles or solid particles
but, in order to produce a gel,  the particles must be attracted to one another
or held together in order to form an immobile network.  With the concept of
a colloid in mind, we may define a few categories in which the many
materials tested  for gelling ability may be placed.  These categories are
distinguished by the method by which the  gelling agent  is introduced into
the medium to be gelled.
Gel Categories

All the materials selected for gelling crude oils may be fitted into two
categories;  those which produce gels in situ by the reaction of two
reagents and those which are dispersed physically into the crude oils,
swell and cause a thickening of the oil.

In situ gels are formed when two or more materials  which are dispersible
in a medium to be gelled are added to that medium.   These materials react

-------
to form a product which has limited solubility.  If the colloidal particles
formed, because of the limited solubility of the reaction product,  have
sufficient attraction for one another,  then a gel is produced.  The soap
gel is  one example.   This is a popular and well explored  system used to
gel many refined aliphatic hydrocarbons such as jet fuel.  The mechanism
involves reacting a fatty acid with about 50-percent caustic soda to pro-
duce a soap.  The insoluble soap particles,  in turn,  become suspended  to
form a colloid.  These colloidal particles become attracted to one another
because of the polar properties of the soap,  thus producing a network and
a gel.  The  soap is also capable of acting as an emulsifying agent between
the water (from the caustic  soda) and the hydrocarbon,  yielding an emulsion
gel in  which the  suspended  particles are water, and attraction is again
gained through polarity.

Another type of in situ gel is the amine/isocyanate gel.  Both the soluble
amine  and isocyanate are added to the medium and the  following reaction
yields a urea with limited solubility. A colloid is again produced but the
attractive forces needed to  form an immobile network are  acquired through
hydrogen bonding which we  shall see has significant affects on the strength
of the  gelled system.

A cooling gel is produced by dissolving the gelling agent in the heated
medium.   Upon cooling,  the dissolved gelling agent becomes more insolu-
ble and tends to  precipitate, producing suspended particles needed for a
gel.  Obviously, this type of gel is  more difficult to produce, due to the
necessity of heating and cooling the gelled medium.  An example of a
cooling gel  is the metal salts such as aluminum stearate  or sodium stea-
rate.   These materials are essentially soaps and produce the same gels  as
the soaps that are reacted in situ.

A swelling gel is produced simply by dissolving the gelling agent into the
medium.   Some metal salts  such as aluminum napthenate  will produce a
gel by this technique, but the process is slow,  24 to 72 hours,  and better
results are usually obtained by heating.

Polymeric gel is obtained by the polar attraction or hydrogen bonding
between smaller molecules,  or by chemical bonding,  resulting in a polymer.
Usually,  the gels produced  by polymers are viscoelastic  due to the high
molecular weights and long  chain configurations.  Also, a good compati-
bility with a gelled medium  is difficult to obtain.  Either  the polymer pre-
cipitates from the medium or the medium is insufficiently bound, resulting
in a weak gel.  A polymeric gel may be produced either in situ or by cooling
or swelling.
Constituents
The literature search revealed several materials with a history of gelling
hydrocarbons, such  as jet fuels and some materials which were used for
"thickening" crude oil,  but it could only be speculated that these materials
could "solidify" crude oils.  Some of these materials include:

-------
      Gel In Situ

            1.     Caustic + fatty acid	=>- soap + water
                  NaOH + RCOOH	>  RCOO" Na+ + HOH

            2.     Amine + isocyanate.	>. urea
                  RNH2  + R'NCO	>  R-NHCONH-R'

            3.     Acid chloride 4- amine 	^-  amide + hydrochloric acid
                  RCOC1 + R'NH2	?  RCONHR1  + HC1

      Gel By Other Methods

            1.     Metal soaps
            2.     Polyamides
            3.     Polar inorganic materials
            4.     Various proprietary reagents


Gel Requirements

A survey of ship failures  and associated situations  which arise revealed
that,  in order for the oil aboard a tanker to be advantageously solidified,
the gelled oil should:

      1.     Have  a specific gravity less than that of seawater
            (to. offer the tanker additional buoyancy in the event
            it should begin to  sink;  also,  should any gelled oil
            escape, it will remain on the  surface for easy recovery).

      2.     Be oil insoluble (otherwise, the solubilizing effect
            of the ungelled crude oil will  reduce the structural
            strength).

      3.     Be water insoluble  (should any seawater penetrate
            into the tanker, the structural strength of the gelled
            oil will again be protected).

      4.     Have  low percentage of gelling agent (to reduce the
            cost,  storage area  needed for the gelling agent and
            time needed to disperse  the agent into  the oil).

      5.     Be obtained by fast-acting gelling agents (the quicker
            the oil is gelled,  the less will be spilled).

      6.     Be produced from easy-to-handle gelling agents.

      7.     Have  high structural strength  (the gelled oil should
            exhibit sufficient strength to  seal the rupture).

-------
      8.     Have low toxicity to handling personnel and
            possibly be nonpoisonous to  marine life.

      9.     Use gelling agents that remain  stable after long
            storage  periods (it is possible that the gelling
            agent will be incorporated into  a permanent fixture
            for all tankers).

     10.     Be noncorrosive to the tanker.

     11.     Be removable from the tanker after docking.


Measuring Gel Strength

All of the gel requirements may be easily defined by standard test methods
except for the structural strength of the gelled crude oil.

In fact, the most perplexing issue raised in the conduct of the oil gel
program has been that surrounding the measurement of gel strength.  In
attempting to answer the questions raised;  i.e., how should a gel's
strength  be measured and which possible measures are meaningful,  it was
discovered that no standard or available method existed which could be
directly correlated to actual flow, such as that resulting  from a rupture in
an oil tanker.

In reviewing available standard methods of  measuring viscosities and
physical strengths,  grease technology contained the most useful  infor-
mation pertaining to a gel or a near solid material.  From this technology,
several basic  methods were considered for measurement of the physical
properties  of the gels.

The rotary  vis-cosimeter is suitable for fast relative measurements of lower
viscosity materials,  but gives little insight into yield points or stresses,
or varying  shear rates.

Measurement by penetrometer is  a method by which a cone is allowed,
through free fall,  to penetrate a solid gel until it comes to rest.   The
strength  of the gel is the weight  of the cone divided by the total wetted
area of the cone.  A correction is made for buoyancy.  These strengths
obtained from  penetrometer readings should be a good indication of a gel's
resistance to initial flow (the shear rate approaches zero) but have  limited
usefulness as the shear rate increases.   Appendix  B, Penetrometer  Use,
describes this method.

A rheogram is  a graph relating stress and viscosity to shear rate by
extruding the material under pressure from a capillary tube using  a  Burrell-
Severs or Ruska apparatus.  The  process  satisfies the formula:

-------
                      
-------
of shear rates chosen to obtain a best fit of the equation with experimental
data,  we do not consider the yield point to be fundamentally significant.
Therefore,  we are omitting Bingham yield points in this comparison. "  (In
reference to the Buckingham Equation. )  They also conclude that  "...
greases have many yield points,  depending upon the type and rate of
deformation.   Therefore, the conditions under which yield points  are
obtained should be reported with the yield values. "

The Bingham yield point plays an important part in the resistance to flow
of gelled oil.  Thus, a  method is needed to identify the yield point- flow
relationship under the operational conditions  of flow through an orifice
(hole) and  any correlation which may exist with yield points derived from
static and  other dynamic techniques.

As mentioned above, the correlation between  the various  existing methods
of measuring yield points is not good but,  at  least, the methods of measuring
do exist. In relation to outflow of gelled oil through a hole, the link
between the rate of overflow and strength  is also currently undefined.

From the equation
Tw  should be relatable to the conditions existing under a pressure differ-
ential produced  by a hole in a vessel.  If  I~  = 0 (zero  shear rate), then
the stress  under consideration is  1 o   or less.   Also,  P   vary consider-
ably with the method of measurement.  For example,  the Burrell- Severs
apparatus uses a long,  narrow capillary tube with a high length/diameter
ratio, which is a considerable extreme from a hole in a  tank which will
have a low length (hull thickness) /diameter ratio. Also, the shape of the
hole affects  p  .  Although a correlation may exist without empirical  data
from a device simulating operational  conditions to show the correlation,
values obtained through existing techniques  have little  meaning in relation
to operational conditions  and resulting flow characteristics.

To best fit the requirements of this program,  it was decided that the pene-
trometer method would be used to test the relative strengths of the gels
produced.  If strength in relation to large-scale or shipboard conditions
is required, it will be necessary to develop a method for simulating the
flow of oil from  a  rupture.  The experimental data from the test method
should result in an empirical relationship which may permit extrapolation
to larger orders  of magnitude.

-------
                              SECTION 3

                   LABORATORY EVALUATION OF GELS
All possible gelling materials resulting from the literature survey were
screened initially by  application to a single crude oil.  The strength, as
measured by the penetrometer, was the sole criteria for continued testing.
Any material selected as a result of the initial screening was subjected to
testing on all  of the requirements.  Table I,  Measurement Methods  for Gel
Criteria, lists the measurement methods used on the gels produced  during
this  program.
           TABLE I.  MEASUREMENT METHODS FOR GEL CRITERIA
           Criteria
          Method of Measure
   Specific gravity



   Oil insoluble


   Water insoluble


   Percentage gelling agent


   Structural strength

   Gel time
Not measured.  Crude oil plus gelling
agents in percentages considered will
in no way be greater than sea water.

Gelled oil will be submerged in ungelled
oil and tested after one month.

Material will be floated in water for one
month.

Plots of gel strength versus percent will
be obtained.

Penetrometer ASTM D-217.

Time to reach two thirds  maximum strength
as measured by penetrometer.
                           Initial Screening

 All candidates for gelling were tested in  100-gram quantities of crude oil.
 Any material showing possibility was  examined at a higher or lower con-
 centration and/or at varying part weight ratios in the case of multicom-
 ponent systems. Appendix C,  Chemicals and Suppliers, lists the source
 of materials tested.

 Soap gels  at 10 percent prepared from fatty acids ranging  from C12 to  C20
 and sodium hydroxide were reacted in situ  in Clearfork crude oil. These
 materials were known for their ability to  gel jet fuels,  but failed to perform
                                  10

-------
satisfactorily in crude oil.  Dispersants were added to enhance the reaction
and resulted in some improvement but, still there was insufficient gel
strength.  It was felt that the proper solubility range was obtained at about
Cj6 .  Dicarboxylic acids produced similar results.   These materials were
difficult to work with because they were solids at room temperature.
Appendix D,  Soap Gel Screening Data, lists the data obtained from these
screening tests.

Amides  from octanoyl chloride and several of the fatty amines were pro-
duced uisitu_ and a good increase in viscosity occurred, but still with
unsuitable strength.  The by-product,  hydrochloric acid,  made this reaction
undesirable because of the corrosion to the ship.

Metal soaps such as aluminum stearate, calcium stearate and aluminum
octoate were hard to mix into the oils and heating was used.  Stringy
viscoelastic gels formed but, due to the nature of the viscoelastic effect,
these gels had no structural  strength.  See Appendix E, Swelling Gel
Screening Data, for data on these tests.

Initial testing with the isocyanates and amines proved to be encouraging.
Many of the commercially available isocyanates and diisocyanates were
reacted with fatty amines and diamines.  Those contained in the remainder
of this report and hexamethylene  diisocyanate (NCO(CH2)6NCO) showed
good results.  Hexamethylene diisocyanate was discontinued because of
its high toxicity.  Of the diamines tested,  only those with long chain
lengths, such as Jeffamine 400, produced ureas soluble  enough in the
crude oils.  Thus, the combination of aromatic diisocyanates and fatty
amines  produced more solid gels  than any other combination of materials.
Appendix F, Isocyanate/Amine Gel Screening Data,  lists the results of
these screening tests.

Fumed silicas were incorporated into the crude oil and then various polar
additives  (isopropanol, glycerine, etc.) were incorporated to complete the
gel network.  Low strength gels were obtained at 10-percent levels.
Because of high cost and difficult gelling technique,  these  systems were
not feasible.

The fact that the gelling agents must react fast and be easily maneuvered,
placed more emphasis on the in situ type of gel due to the relatively
complicated aspects of obtaining a gel by the cooling or swelling methods.
A gel should be obtainable with a minimum amount of mixing and at ambient
temperatures.  Although some initial testing was performed  on gels of the
swelling or cooling types,  this information was considered  as a  secondary
interest, since no strong gels  or fast gels were obtained by these methods.

With emphasis placed on the in situ gel, the urea gelling system proved
to be far superior.  Many good gels were obtained with this  system.  Only
fair to poor gels were obtained with the other in situ reactions.
                                 11

-------
The following conclusions may be deduced from the data gathered during
initial screening:

      1.    Many materials tested resulted in an increase in
            viscosity, but no materials (except those participating
            in the amine/isocyanate reaction to yield a urea)
            produced "solid" gels.

      2.    There is a relationship between the chemical constitution
            of the gelling agent and the type of gels produced.  Mole-
            cules containing alkyl groups in the chain length ranges
            of C^ to C18 demonstrated the proper solubilities neces-
            sary for the gel condition in both  the soap and urea type
            gels.

       3.    Of all the isocyanates tested, toluene diisocyanate
            proved to be the most promising candidate because of
            low cost, ready availability and performance as a gelling
            component.

       4.    Of the amines tested,  the fatty amines were the most
            promising and were readily available  in the  chain lengths
            from C(, to C18.

 The following combinations of amines and isocyanates produced gels worthy
 of continued testing:

            Armeen L-ll  and TDI 80/20

            Armeen C and MT-40

            Jeffamine 400 and Octadecyl Isocyanate

            Armeen L-15  and TDI 80/20

            Armeen O and TDI 80/20

            Armeen C and TDI 80/20


                       Detail Screening Parameters

  The materials  selected from the initial screening  were subjected to the
  following test scheme to develop necessary comparative data:

        1.    The ratio of the two components  of the gelling agents
             were varied and plotted against the gel strength  to
             determine an optimum ratio.
                                  12

-------
      2.     The percentage of total gelling agent (both parts)
            was varied from one to five percent and plotted
            against gel strength.

      3.     Gelled oil was submerged in salt water and in
            ungelled  oil to test gel retention.

      4.     Oil was gelled and  tested at 32F and 100F to
            determine strength and speed of reaction.

      5.     Various methods of  mixing  the gelling agents  into
            the crude oils were explored to identify characteristics
            and requirements.

      6.     Literature search and/or laboratory tests were conducted
            to obtain more detailed physical and  chemical properties
            of each gelling agent.

      7.     Costs in  terms of cents (of gelling agents) per gallon
            (of crude oil) were derived.

      8.     Strength tests relative to agent storage time were
            developed and conducted.
                     Classification of Crude Oils
Crude oils to be tested were received from various parts of the world.  Each
crude proved to have its own reaction toward the gelling agents.  Gels
produced ranged from pseudoplastic gels to viscoelastic.

Due to the wide variety of combinations of materials composing a crude
oil  (paraffinic and asphaltic materials as well as  various inorganic
compounds),  an exact description was impractical.  The description used
here is the Bureau of Mines system by which crude oils are classified
according to the API gravity of two key distillation fractions.  First, the
crude oil was distilled at atmospheric pressure, collecting fractions at
regular intervals,  usually  25C.  A key fraction was  collected from 250C
to 275C (kerosene range).  Then the distillation was continued (after
cooling) at 40 mm mercury (vacuum).  Fractions were again collected and
key fraction two was collected from 275C to 300C.   From the API gravity
of these two fractions, the crude was classified as follows:
                                   Key Fraction 1     Key Fraction 2

      Paraffin                      40 or lighter     30 or lighter

      Paraffin - intermediate        40 or lighter     20 to 30

      Intermediate - paraffin        33 to 40         30 or lighter


                                 13

-------
                                   Key Fraction 1    Key Fraction 2

      Intermediate                  33 to 40         20 to 30

      Intermediate - naphthene      33 to 40         20 or heavier

      Naphthene - intermediate      33 or heavier    20 or heavier

      Naphthene                    33 or heavier    20 or heavier

      Paraffin  naphthene          40 or lighter     20 or heavier

      Naphthene - paraffin          33 or heavier    30 or lighter
The API gravities were measured with standard API Baume hydrometers.
These values were converted to specific gravity by the equation:
                                  - 131.5
                  .       0    .
                  spgr 60/60 F
Viscosities were measured from a Fann Model 35 Viscosimeter and recorded
in centipoises.  These values were converted by use of a suitable chart
to Saybolt University Seconds (SSU) for reader convenience.  This infor-
mation is contained in Table  II,  Crude Oil Classification.
                    Gel  Properties and Mechanics

Table III,  Physical Properties of Gelling Materials,  lists most of the common
properties of the prime gelling agents.

All of the isocyanates are liquid at reasonable temperatures and could be
used without heating in most instances.  Some of the amines, Armeen O
in particular, have rather high melting points and would require heating
to the liquid state before application to the crude oil in cold climates.
The isocyanates are slightly toxic as a result of the NCO group attached.
However,  adequate protection to operating personnel can be assured  with
reasonable care and standard chemical handling equipment.

The amines are not particularly toxic, their principal hazard being the
slight caustic nature of the chemical.  The flammability of all of these
chemicals is less than the crude oils and add no additional fire hazard in
transporting or storage;  therefore, the shipboard storage and use of any
of these chemicals can be safely accomplished, using the reasonable care
required of any chemical agents.
                                14

-------
TABLE II. CRUDE OIL CLASSIFICATION




Crudes

Clearfork
Westbrook
Howard Glasscock
Sweden Crude
Sun B Mix
Mirando-RHC Mix
Grade A, 46 9 Mix
Solvent Mix
Lybian

Arabian Heavy

Arabian Light

Devonian
Cairo

Alaskan
Bachaquero




Source

Rankin, Texas
Snyder, Texas
Snyder, Texas
Sun Oil Co.
Sun Oil Co.
Sun Oil Co.
Sun Oil Co.
Sun Oil Co.
Standard Oil
of California
Standard Oil
of California
Standard Oil
of California
Atlantic -Richfield
Phillips
Petroleum Co.
Atlantic -Richfield
Atlantic -Richfield
Viscosity



Ctn
MJ
10 a
tin O
17.0
15.0
13.0
2.3
4.0
20.0
57.0
--

47.0

25.5

8.0
2.5

10.0
4.5
300.0


CO
CO
90
80
68
14
43
100
300
--

250

166

60
35

62
45
200


M

s
C3
%
34.0
24.5
32.0
47.0
39.0
23.0
23.5
48.0

28.5

28.5

33.5
46.0

35.0
38.0
17.0
C

3 
CQ
"io 
-------
TABLE III.  PHYSICAL PROPERTIES OF GELLING MATERIALS
Material
( Manufacturer )
TDI 80/ZO
Tonco 90
(Upjohn)
MT-40
(Mobay)
Armeen C
(Armour)
Armeen O
(Armour)
Armeen L- 1 1
(Armour)
Armeen L-15
(Armour)
Jef famine 400
(Jefferson)
Chemical
Composition
80% 2.4 Toluene
20% 2.6 Toluene
Octadecyl
Isocyanate
CH3(CH2)17NCO
Aromatic
Isocyanate
50% C12 ^
18% C14
8% C16 > diamine
Q al r*
8 I" C8 J
16% Other
76% C18 ^
5% Cl6 \ diamine
5% C18 J
14% Other
Primarily
CH3(CH2)10NH2
Primarily
CH3(CH2)14NH2
NH2CH(CH3)CH2-
[OC'H2CH(CH3])5.6 -
NH2
Molecular
Weight
177
295

223
267
170
226
400
Melting
Point
11.5-
13.5C
10 -
20C

24C
+74F
-20F
+50F
-40F
Boiling
Point
120/
10 mm
170/
2 mm






Flash
Point
132C
open cup
185C
open cup
295F
240F
open cup
320F
open cup


347F
open cup
Single Oral
Lethal Dose
(Rats)(g/kg)
4.9 - 6.7
30.0





0.75
Cost/
Ib
$ .45
2.25
.31
.49
.38
.58
.52
.65

-------
Storage and Chemical Strength

It is anticipated that large quantities of gelling agents may be stored
either aboard tankers or at some convenient location for rapid access in
the event a rupture in a tanker occurs.  Since large  quantities of materials
may be  stored,  economical use of these materials is essential.

The first step to preservation is prevention from contact with reactive
materials. The amines react as a base.  The most common deterioration
is  their reaction with carbonic acid (carbon dioxide  and water vapor
found in the air) to form carbonates:

            RNH2 + H2CO3 	>- RNH3+  HCO3

This reaction is usually reversible with heat.  Also,  the amines should be
stored in containers coated with an inert  lining as recommended by the
manufacturer.

The isocyanates are more reactive, particularly toward water.
                                    O

            RNCO + HOH	i RNHC   OH
                O                            00
                ii                             ii       n
            RNHC  - OH + RNCO	?- RNH - C -  O   C   NHR	=s-
                O
                n
            RNHCNHR + CO2  (urea)

The urea is easily detected in the isocyanate by cloudiness. This is not
to be confused with the crystallization of some isocyanates, which can
be removed by heating, and does not interfere with its chemical reactivity.
The production of the urea is  not reversible.

The isocyanate should also be stored in a lined container.   In the event
the containers are opened and exposed to the atmosphere,  purging with
an inert gas such as nitrogen is recommended, depending  on the exposure.
Storage in a cool area is always recommended.  These materials are not
harmed by freezing but, if frozen, must be melted before using.

If  sampling is to be undertaken periodically, then it is advisable to
design storage facilities for minimum exposure during sampling.  Direction
for testing these chemicals are included in Appendic G, Stability in
Storage Tests for Gel Constituents.  The  manufacturer will supply infor-
mation for testing  specific materials.
Gel Mechanics

During laboratory testing of the various classes of gelling agents, the
amine/isocyanate in situ reaction produces far superior gels than any
                                 17

-------
other system tested.  Since the most promising reagents are of insuf-
ficient functionality to produce polymers (both reagents must have a
functionality of two or greater),  there are other forces holding the mole -
cules together.  These other forces are intermolecular forces of which
hydrogen bonding is the most significant.

Ordinary chemical bonds have a strength of 50 to 100 kcal,  compared to
about 5 kcal,  for electrostatic bonds such as hydrogen bonds.   This
explains the thixotropic effects of an electrostatically bonded network.
Shear or mixing supplies sufficient energy to disorder these weaker bonds,
due to the characteristics of the hydrogen bond. After mixing is stopped,
the bonds recover and the gel is restored.  If this is true,  then we would
expect that  the greater a molecule's ability to form hydrogen bonds, the
greater its ability to form gel networks.  This is found to be true with
certain limitations.

Which molecules are  capable of forming hydrogen bonds ?  Only the ele-
ments  O,  N, and F are electronegative  enough to produce hydrogen bonds
of any significant strength.  The hydrogen atom must be both chemically
bonded and  electrostatically bonded to  these elements.  Therefore, mole-
cules which contain these elements may participate in hydrogen bonding
if the geometric arrangement of the molecule permits.  The "molar cohesive
energy"4 (intermolecular forces) of the many chemical  groupings which
were found  to be of interest in the gellation of crude oil.  These include:

                                         Cohesive Energy
                  Group                  kcal/molecules

            -CH2 (hydrocarbon)                0.68

            -COO (ester)                     2.90

            -C6H4 (aromatic)                  3.90

            -CONH (amide)                   8.50

            -OCONH (urethane)               8.74

            -NHCONH (urea)                 higher

The higher the cohesive energy, the higher are the  intermolecular forces.
Then we expect the urea to have the highest potential  for forming inter-
molecular bonds, and this has been found to be true.   The urea  type gels
are the chemical class producing the strongest gels.   The urethane gels
show promise, but none are satisfactory.   The same is true for the amine
materials.

The basic chemical reaction to  produce a urea from an isocyanate  and an
amine is-.
                                 18

-------
R'NCO + RNH2 	*- R1
                                         o   !
                                          H   '
                                      -N-C-N-|R
                                              i
                                       H    Hi
                                    i	i
                                      urea linkage

The hydrogens attached to the nitrogens are  capable of hydrogen bonding
with either the oxygen or nitrogen, probably with the oxygen since it is
more geometrically available.  A typical network would be (dotted lines
indicate hydrogen bonding):

                  R1            R1           R1
                  I             I            I
              H   N       /H -  N      /H -  N
                  1      /      '      /     '
                  c = o'       c = c(      c = o
                       ^x          \     '
              H - N       NH -  N      ^H -  N

                  R            R           R

With this in mind, it is noticeable that decreasing cohesive energy (see
listing above) corresponds to decreasing available hydrogens,  nitrogens,
and/or oxygens for electrostatic bonding.

Now that the ability to form secondary chemical bonds (intermolecular
bonds) has been established, other necessities of the molecular structure
must be explored.  Since the crude oils or other hydrocarbons have little
or no ability to participate in electrostatic bonding, it seems likely that
the gelling agent (urea,  urethane, amide, etc.) becomes incompatible
(insoluble) and separate from the mixture.  This occurs when the gelling
agent is of low molecular weight and the active polar  chemical groups
make up a large percentage of the weight.  If a group  compatible with the
nonpolar hydrocarbons is incorporated into the gelling agent molecule,
partial solubility is gained and  precipitation is avoided.  Fatty amines
were found to  produce such an effect in the reactions  with the isocyanates
such as toluene diisocyanate.   Chain lengths from C7 to  C18 yield a
product of desired solubility. Solubility falls  off as the  chain length
shortens, and the product  becomes too soluble as the chain length
lengthens.  The proper solubility is important in forming  the colloid which
produces the gel.

Difficulty is encountered in obtaining a good solubility balance by using
the urethane reaction. When octyl alcohol reacts with toluene diisocyanate,
the product is  a gummy liquid,  which,  in turn, proves to be soluble in
crude oil.  Viscosity  is increased but no gel formed.  Due to the lower
cohesive energy of the urethane grouping, the  addition of an eight carbon
length hydrocarbon contributes significantly to the solubility of the urethane
in the crude oil.  If the chain length  is shortened,  then insufficient


                                 19

-------
compatibility is obtained to form a structural network with the oil.  If
this is thought of in terms of cohesive energy per unit weight, then the
urethane grouping produces  a less tightly bound system which is more
easily solubilized and as the chain length of the alcohol increased (to
gain compatibility),  the cohesive energy concentration decreases.

All of the chemical classes  discussed produce gels of varying physical
characteristics with no class being limited in the type  of gel they produce.
The physical properties of the gels may  be more closely correlated to the
physical dimensions of the gelling molecule than with the chemical class.

Viscoelastic gels are those  which resist deformation to an applied stress
in a rubbery fashion.  These gels are stringy and will recoil if the gel is
strong enough.  Consequently, a continuous force,  although not very
strong, will slowly displace the  gel and cause it to run.

Pseudoplastic gels are those which exhibit a decrease  in viscosity as the
shear rate is increased.

A Bingham Body has a  finite yield stress which must be exceeded before
flow will occur.  This type of gel holds  its shape until sufficient stress
is applied. Then it yields suddenly, perhaps breaking into parts if the
gel is strong.

Besides exhibiting  the above characteristics,  the gel may also be thixo-
tropic;  i.e.,  viscosity decreases with time under a constant shear stress.
Usually,  the gel regains most of its viscosity when shear is  removed.

The urea-type gels tested under this study produced gels classified as
pseudoplastic and Bingham  Bodies and are thixotropic,  which would be
an aid in the removal of the gelled oil from the tanker.
                    Testing of Gelled Crude Oils

 Preparation

 Several methods of incorporating the gelling agents into the crude oils in
 laboratory testing were evaluated:

       1.    Both amine and isocyanate were injected simultaneously
            through syringes into the oil with no other agitation.
            This resulted in fair gels with some localizing of the
            gelling agents producing lumps.

       2.    One ingredient was added and mixed into the oil then
            the other was injected by syringe with no other agitation.
            This decreased some of the localization but did not
            produce as smooth a gel as Method 3.
                                 20

-------
      3.     Mixing one ingredient thoroughly then adding the
            other ingredient by syringe and further agitating
            with a laboratory stirrer for 15 seconds after addition
            of the second ingredient was most  effective and
            was adopted for comparing gelling  agents.

Methods 2 and  3 were utilized to determine the optimum sequence of the
agents.  Equivalent gels were obtained if there was no  delay (less than
15 minutes) between addition of the two ingredients;  however,  if there
was a delay, addition of the amine first proved far superior.

The more reactive isocyanate material deteriorates through its reaction
with water and  other impurities in the  crude oil.  Table IV,  Delayed
Gelling of Crude Oil, shows the variation obtained by varying the agent
addition sequence.
            TABLE IV.  DELAYED GELLING OF CRUDE OIL
Added 1st
Armeen C
TDI 80/20
Armeen O
TDI 80/20
Armeen C
MT-40
Armeen L- 15
TDI 80/20
Armeen L- 1 1
Added 2nd
TDI 80/20
Armeen C
TDI 80/20
Armeen O
MT-40
Armeen C
TDI 80/20
Armeen L- 15
TDI 80/20
Penetrometer Reading (2nd
Ingredient Added 24 Hours
Later) Gel Stand 1 Hour
283 (Br)
Weak Gel
360 (Br)
Weak Gel
251 (Br)
Weak Gel
211 (Br)
Weak Gel
196 (Br)









To conclude, the gels were prepared for laboratory testing by preparing
300 grams of gelled crude oil in a 600 milliliter beaker.  The crude oils
were weighed into the beaker (285 grams in the case of a 5-percent gel),
the selected amine was added with agitation from a laboratory stirrer,
then the selected isocyanate was added under the same agitation.  The
gels began to form immediately and were stirred for an additional 15
                                 21

-------
seconds.  Then the gelled crude oil was allowed to stand for one hour
before testing on the penetrometer.


Optimum Amine and Isocyanate Ratio

During initial screening, the amines were mixed with the isocyanates in
varying proportions to determine if a better gel could be obtained by
deviation from the stoichiometric ratio.  The results of this would deter-
mine if   1)  any of the reagents were reacting with the crude,   2) any of
the reagents  were of a different strength than that posted in the literature,
3)  steric effects were hindering the mobility of any reagent, or 4)  any
unknown factors were impairing the reagents' gelling ability which would
be reflected in the mixing ratio.  The mixing ratio (F) describes these
ratios and is defined in the  following equation:

                  EWA  F      WA
                   EWj        Wj


            EW,   =  equivalent weight of amine
               \

            EWj   =  equivalent weight of isocyanate

            W    =  weight of amine added
              f\

            W    =  weight of isocyanate added

            F     =  stoichiometric ratio factor

The ordinate of these graphs is the penetrometer reading in tenths of
millimeters.  It should be remembered that a lower penetrometer reading
corresponds to a  stronger gel.

From this data (see Appendix H, Optimum Ratio  Studies),  the optimum
ratio of the amine to the isocyanate for the various systems is seen to be
between 0.75:1.00  to 1.50:1.00 equivalent weight ratio.


Gel Strength Versus Percent Gelling Agent

Once the optimum weight ratios  were  established,  these ratios were held
constant and the  percentage of total gelling agent was varied. Gels were
formed at 5,  4, 3, 2 percent (and intervening half percentages) until the
gel formed was too  weak to record a penetrometer  reading.  Appendix I,
Gels Strength Versus Percent Gelling  Agent,  contains a display of the
data obtained from these tests.  Data on the graphs was not extrapolated.
The lowest percentage indicated by the curve drawn was the lowest which
could be read on  the penetrometer with each  respective gelling system
and crude oil.  The wide variety of constituents of crude oils  was equally
                                 22

-------
reflected in the variety of responses of the crude oils to the gelling agents.
If data can be gained during larger scale testing concerning gel strength
needed to seal a prescribed rupture,  these graphs would point out the
least and most economical percentages which would produce the desired
gel.  A consolidation of the penetrometer readings are given in Table V,
Penetrometer Reading of Gels for Five-Percent Agent Concentrations.

Weighed portions of gelled  Clearfork crude were placed in weighed portions
of synthetic seawater (ASTM D-1141) and ungelled Clearfork crude and
allowed to stand 30 days.  Then these gelled lumps were strained out of
the respective  liquid medium and the medium was reweighed and percentage
weight loss (-) or weight gain (+)'  of the medium was  recorded  (Table VI,
Solubilities of  Five-Percent  Clearfork Gels).


Gelling Under Temperature Variations

The six selected systems were tested for gelling ability at  100F and 32F
in Clearfork crude and Solvent Mix crude. To accomplish this, all ingredi-
ents (crude, amine and isocyanate) were adjusted to  the specified temper-
ature. In the event the gelling  agents were  solid at 32F,  they were warmed
until  they first became liquids (the melting points of gelling ingredients
are given in Table III, Physical Properties of Gelling Materials).  The
amine was then added and mixed into  the oil.  Next the isocyanate was
added to gel the  oil.  These gels were aged  one hour (still  at specified
temperature) and the penetrometer reading was recorded.  The  results are
in Table VII, Gelling  Under Various  Temperatures.
Gelling Cost

Table VIII, Cost to Gel One Gallon of Crude Oil, was prepared to outline
the raw material costs of the gelling agent.  The costs of the raw materials
were average, medium quantity figures.   Bulk quantities would be less
expensive.


Effects of Salt Water and Ungelled Oil

In the event the oil gelled aboard a  tanker is spilled into the ocean,  the
gel must remain intact to form lumps which may be harvested. Also, in
the event it is not necessary to gel  all the oil contained in a tanker, then
the gelled oil must retain its strength while  in contact with ungelled oil.

Since none of the ingredients producing the urea gel are soluble  in water
(fatty amines become water soluble  if the alkyl  portion is shortened to C&
or lower), and the oil itself is  not soluble,  then it is  expected that no
effects will be observed.

Since the fatty acid/caustic gel is of an ionic nature (acting  as an emulsi-
fier),  oil gelled by this method turns cloudy upon contact with water.


                                 23

-------
                            TABLE V.  PENETROMETER READING OF GELS FOR
                                     FIVE-PERCENT AGENT CONCENTRATIONS
Gel Agents
Alaskan (Nikiski)
Arabian (Heavy)
Arabian (Light)
Bachaquero
Cairo (Egypt)
Clearfork
Devonian
Grade A 469 Mix
Howard Glasscock
Lybian
Mirando
Solvent Mix
Sun B Mix
Sweden
Westbrook
Gel Agents
Armeen C
TDI 80/20
294
241
'243
409
261
280
268
281
237
258
262
311
283
280
230
Armeen O
TDI 80/20
348
380
255
failed
270
235
323
222
286
failed
338
327
305
327
326
Armeen L- 15
TDI 80/20
357
430
failed
362
242
204
300
310
275
293
failed
failed
297
351
304
Armeen C
MT-40
435
276
258
367
250
257
275
250
276
270
260
237
239
232
285
POPDA 400
Tonco 90
403
217
284
349
218
212
257
262
219
260
222
255
208
288
195
Armeen L- 1 1
TDI 80/20
201
358
220
269
236
195
225
206
221
390
211
228
220
241
255
tv

-------
      TABLE VI.  SOLUBILITIES OF FIVE-PERCENT
               CLEARFORK GELS
Gelling
Jeffamine 400
Armeen C
Armeen L- 15
Armeen C
Armeen O
Armeen L- 1 1
Agents
Octadecyl Isocyanate
TDI 80/20
TDI 80/20
MT 40
TDI 80/20
TDI
Water* (%)
-1.0
-1.3
-0.96

-0.85
-0.89
Oil*(%)
-.88
0
-0.83
-1. 17
-3.33
-3.4
-H-
Soaking gel for 30 days caused this weight change on respective
solvent (water or oil)
TABLE VII.  GELLING UNDER VARIOUS TEMPERATURES
Gelling System
Armeen C and TDI 80/20
Armeen O and TDI 80/20
Armeen L-15 and
TDI 80/20
Armeen C and MT 40
POPDA 400 and
Tonco 90
Armeen L- 1 1 and
TDI 80/20
Clearfork Crude
Penetrometer
Reading (Br)
100F
407
324
217
377
Gel too
thin
192
32F
Gel too thin
247
267
	
325
216
Solvent Mix
Penetrometer
Reading (Br)
100F
330
372
	
390
219
259
32F
386
322
No gel
formed
258
Gel too
thin
206
                       25

-------
      TABLE VIII.  COST TO GEL ONE GALLON OF CRUDE OIL
System
Armeen O
TDI 80/20
Jef famine 400
Octadocyl
Isocyanate
Armeen C
TDI 80/20
Armeen L- 1 1
TDI 80/20
Armeen L-15
TDI 80/20
Armeen C
MT-40
Gelling
Ingredients
Cost/lb
$.378
$.45
$.65
$2. 20
$.488
$.45
$. 57
$.45
$.52
$.45
$.487
$ . 31
Stoichiometric
Ratio Factor
1.0
1.0
.75
1.25
1.50
1.0
Percentage of
Gelling Agent
1%
$.0309
$.119
$.036
$.039
$.038
$.033
5%
$.156
$.59
$. 180
$. 197
$. 191
$. 263
It could be expected that ungelled oil would present the most problems
since it could possibly act as a solvent for the gelled oil.  This was
found to be true if the gelled oils and ungelled oils were blended with
agitation but,  if the gel was  allowed to stand unagitated in the oil for
30 days,  only slight swelling effects were observed.  Only urea gels
were strong enough to test in this  manner.
Ten-Percent Gels

Preliminary calculations had indicated that for reason of cost for materials,
the upper limit of amine/isocyanate used to get the crude oil would be
held at five  percent.  For the purposes of gaining an insight into the
results of applying a much higher concentration of gel constituents, a
limited test  was run at 10-percent concentration.

Two crude oils were selected, a high gravity Sweden crude of API 47,  and
a low gravity Bachaquero crude of API 17.  Two gel constituent systems
were selected, the lowest and highest cost of the systems  selected for
further testing.  Table DC,  Ten-Percent Gels,  shows the results of this
testing,  indicating that the increase in strength gained would not seem to
offset the doubling in cost of materials,  or the increased storage space
required  for  the additional chemicals.
                                 26

-------
                   TABLE IX.  TEN-PERCENT GELS
System
Armeen L- 1 1
TDI 80/20
Jeffamine 400
Octadecyl Isocyanate
Armeen L-ll
TDI 80/20
Jeffamine 400
Octadecyl Isocyanate
Weight
Ratio
_igm)
21.26
8.74
12. 12
17.88
21.26
8.74
12. 12
17.88
Crude Used
And Amount
(270 gm)
Sweden
Sweden
Bachaquero
Bachaquero
Penetrometer
Reading
(Br)
240
205
185
125
75 after
standing
24 hours
Other Materials Gelled

As an auxiliary interest, some common solvents and fuels were gelled with
the urea systems.  Most aliphatic materials can be gelled with these
materials.  Aroma tics tend to be weaker,  stringy gels.  Polar solvents
produce weak particulate gels due to the limited solubility of the alkyl
portions of urea gelling molecule.  See Table X,  Various Materials Gelled.
Flow Properties

A high gravity crude oil (API 38) was gelled using 5-percent Armeen C-
MT40 gelling agents.  This gel was tested in a Burrell-Severs extrusion
rheometer.  Two graphs (Figure 1,  Rheograin Gelled Crude Oil) were
obtained relating shear stress  and  viscosity to shear rate.  From the
rheogram, it was concluded that the gel was shear thinning and had a
finite yield  point.  The yield shear stress of 11, 000 dynes per square
centimeter obtained corresponds closely to  the 12, 000 dynes  per square
centimeter obtained with the penetrometer.  The difference in curves
obtained from the worked and unworked gel  shows the thixotropic nature
of the gelled oil.
                              Summary

The laboratory evaluation of gelling systems resulted in gels which were
far superior to any found in previous work or identified in the literature
                                 27

-------
                                    TABLE X.  VARIOUS MATERIALS GELLED
Gelled Material
JP-5
JP-5
Toluene
Toluene
Regular Gasoline
Regular Gasoline
Motor O il
SAE 30
Motor Oil
SAE 30
Isopropanol
Gelling Agent
Armeen O
TDI 80/20
Armeen L- 1 1
Armeen O
TDI 80/20
Armeen L- 1 1
TDI 80/20
Armeen L-l 1
TDI 80/20
Armeen O
TDI 80/20
Armeen O
TDI 80/20
Armeen L- 1 1
TDI 80/20
Armeen L- 1 1
TDI 80/20
%
Gelling Agent
5
5
5
5
5
5
5
5
5
Penetrometer
Reading
331 (Br)
244 (Br) (8/15/69)
170(Br) (8/18/69)
Failed (Br)
No gel after 1 hr.
251 penetrometer
reading after 4 hrs .
292 (Al)
379 (Al)
Failed Al
350 (Br)
377 (Al)
Remarks
Good gel was formed.
Good gel formed with in-
crease in strength after
several days.
Very weak gel. Failed the
penetrometer.
Good gel after 4 hours.
Viscoelastic gel. Forms a
gellatinous blanket on
surface.
Viscoelastic gel. Stringy
and thin.
Thin and stringy gel.
Weak stringy gel.
Weak particulate gel.
Gelling agent formed
suspended particles.
N)
oo

-------
 O
 ii


N6
 u

 w
 CD
20  . .
      10 . .
 w
 to
 CD
 
 CO
                                     _ settled gel

                                     .   worked gel
 QJ
 w
      600 -.
100         200
                                            300
                                                   400
                          Shear Rate  r Sec.
     400 ..
                                               settled gel

                                        	worked gel
 w
 o
 o
 Ul
200
                                             4-
                    100
                           200


                     Shear Rate
                        300


                        ~i
                                                        400
                Figure l.  Rheogram, Gelled Crude Oil
                                  29

-------
survey.  All systems investigated resulted in a physical bond,  such as
hydrogen bonds or other electrostatic attraction,  to form the gel.  There-
fore, from the physical bond  standpoint, the urea-formed gel is the
strongest gel possible at the percentages used.

It may be postulated that gels might be formed  by chemical bonding,
rather than physical bonding,  using the unsaturation of the crude oil
components or the naturally occurring impurities  as one component with
a catalyst or crosslinking agent as  the additive.  As far as can be deter-
mined, little,  if any, work has ever been conducted in this area which
might offer an entirely different approach to the problem.   Any  such  investi-
gation would require a major basic  research effort.  Whether a gel could be
produced through this approach, or if such results could be routinely
reproduced, or if such a  system could be universal in  its ability to gel
all classes of crude,  and what the  character and  strength of such a  gel
might be, is unknown and subject to only speculation at this time.

The laboratory evaluation of the amine/isocyanate gel systems shows
that it is possible to produce gels  in all domestic and foreign oils tested
which are unaffected by salt water  or crude oil.  Gelling of crude oil is
possible at the extremes of cold or hot climates which may be encountered.
Many other finished petroleum products which may be transported can also
be  gelled by this same system.

The best overall system from all standpoints was Armeen L-ll and TDI
80/20.   It was found, however, that the manufacturer, Armour Chemical,
was not  producing the chemical in larger than laboratory quantities.   It
was,  therefore,  eliminated from consideration in  larger volume tests.  The
Jeffamine 400 and octadecyl isocyanate system produced the next strongest
gel at room temperature  but encountered some difficulty at  climatic
extremes.  The cost was the highest of any system and large-scale  pro-
duction of the isocyanate,  as would be required by wide usage, was
questionable.  This system was also eliminated from large-scale tests.
The Armeen C and MT 40 system produced gels  in all oils tested, performed
well at climatic extremes and was moderate in  cost and available in large
quantities.  This system was, therefore, selected to be used in the  large-
scale tests.
                                30

-------
                            SECTION 4

             LARGE TANK TESTS OF THE GELLING SYSTEM


                          Program Design

The  1, 000-gallon tests were designed to verify that the chemical agents
selected as a result of laboratory evaluation could produce gels of com-
parable characteristics in significantly larger volumes.  The key to
successful performance of these tests was the proper functioning of the
dispensing and mixing concepts.


                        Dispersion Concepts

The  best gelling  agent combinations resulting from the laboratory study
all require thorough,  uniform and rapid distribution throughout the oil to
be gelled. The initial task in dealing with volumes significantly larger-
than-laboratory quantities was  to devise means of introducing and dis-
persing the gel agents into the oil.

A total of 16  dispersing concepts were formulated,  including preliminary
assessments of their advantages and  disadvantages. Each of these are
detailed in Appendix J,  Gel Agent Dispensing Concepts.

A relative evaluation of each of the 16 concepts was made,  based on the
preliminary assessments.  This resulted in the immediate elimination of
Concepts 8 through 16.  The remaining concepts were then considered in
more detail and the preliminary assessments extended.

The  submersible  pump concept  (Concept 6) consists of lowering a pump
into  the oil to be gelled through which the chemical agents, oil and air
can  be pumped for gelling.  The gelled oil containing air floats to the
surface and suction and discharge of the submersible pump may be  kept
relatively free of the  gelled oil.  On the other hand, maneuvering the
pumps and suction is cumbersome.  It is difficult to keep the suction free
of gelled, or partially gelled, oil.  As viscosity increases,  horsepower
requirements are  excessive and overall pump effectiveness is low for a
large tank.  This concept was not recommended to be included  in the  test
program.

The  propeller mixer concept (Concept 7) consists of a portable  or permanent
propeller mixer being positioned in the tank. The gelling agents are dis-
persed into the vortex formed by the propeller. This concept is similar
to the use of a laboratory propeller stirrer in a beaker.  In this case,  it
has  been determined that  horsepower  requirements become excessive  with
increasing viscosity as the oil thickens.  Poor mixing also results  in
portions of the tank due to maneuvering constraints.  This concept was
not recommended to be included in the test program.
                                31

-------
The following concepts were recommended to be included in the test program.
On a relative basis, they are the most practical in meeting overall system
requirements.


Rotating Nozzles (Concept 1)

The rotating nozzle concept consists of using multiple openings in a tool
in such a manner that when the fluids are  pumped from the nozzle openings
the tool rotates.  The gelling chemicals  are dispersed through the same
nozzle openings.  By varying the pressure and sequence of injection, the
differing amounts of agents can be injected.  The nozzle is raised and
lowered through the oil to assure the dispersion of the chemicals through-
out the oil to be gelled.  Turbulence caused by the injection creates mixing.

There are some shipboard operating difficulties which seem solvable, pro-
viding the dispersion concept is successful.  Some of these are:  multiple
baffles, web stiffeners and other obstructions  in the  tank;  means of tra-
versing the nozzles vertically through the tank, and  methods  of flushing
the lines between agent  application.


Recirculation (Concept 2)

The recirculation concept consists of pumping  the oil from the tank  through
a manifold, where the two gelling chemicals are  proportionally added and
then returning the mixture to the original tank.  The pump suction can be
from the top or the bottom of the tank.  The discharge of the oil-chemical
mixture can also be on the top or bottom of the tank.   Two proportioning
pumps are required, one for each of the chemicals  injected.  The varying
pump suction requirement aboard ship can be overcome by using a turbine
pump (deep well type) immersed in the oil with a flexible suction attached
with a floating intake  port.


Recirculation With Air (Concept 3)

It was postulated that through the addition of a gas or air into the oil-
chemical mixture at the time the gelling is occurring, the density of the
gelled oil can be reduced.  The lighter gelled oil floats on the ungelled oil
and thus  simplifies the possible problem of the gelled oil getting into the
suction of the circulation pump and plugging it.  In addition,  it is believed
this technique  can assist in obtaining a uniform gel throughout the tank.
This is a variation of Concept 2, requiring minimum equipment changes.


Vertical Sparger (Concept 4)

Locating  sparger pipes adjacent and parallel to the outer hull might allow
oil in the immediate area of likely damage or rupture to be gelled first.
Turbulence caused by  nozzle injection creates mixing, particularly near  the


                                   32

-------
hull.  Localized gelling might also be accomplished with Concepts  1 and
5; therefore, it is recognized that Concept 4 might be deferred until the
others have been tested.
Horizontal Sparger (Concept 5)

The horizontal sparger concept is similar to the vertical sparger,  except
that this installation is appropriately located across the bottom of the
compartment or tank.  The attempt here is to gel the total volume of the
tank.   It is postulated that air is required to assist in the dispersion and
mixing of the chemicals into the  oil.  The air also permits the gelled oil to
rise to the top such  that the ungelled oil would be in proximity to the sparger
pipes for better mixing.

The list of concepts generated and the analysis and recommendations were
submitted to the Contracting Office for approval.  Concepts  1, 2 and 3
were approved for testing.
                           Facilities Design

Test Tank

The tank for the first scale-up from laboratory gelling tests was designed
with a square cross section.  This was done to  start bridging the  gap
between laboratory container and actual ship tank shape.  The fabricated
steel tank dimensions were 4.5 feet wide, 4.5 feet long and 8 feet high.
This allowed a 2.25-foot freeboard above the 1, 000- gallon oil level.  A
steel ladder allowed test progress to be observed from above.  A vertical
sliding gate was placed on the bottom of one side of the tank.  This,  gate
was  4 feet long by 1 foot high and provided  a variable opening to test the
ability of the gelled oil to resist flow from the tank.  A two-inch diameter
valve opening at tank-bottom was also installed.  No provisions were made
at this time to measure flow properties against differential heads  of sea-
water.
Recirculation Dispersion System

Figure 2,  Schematic of Recirculation System,  is a schematic drawing of this
dispensing method.  The recirculation of the crude oil is accomplished using
a centrifugal pump so that experimental recirculation rates can be varied by
throttling the pump discharge.   The  gel agents are metered into the circu-
lating crude oil by a duplex piston pump.  The recirculation system
(Concept  2) takes suction  from the top of the oil and discharge the mixture
of crude and gelling agents at the bottom of the tank. The recirculation
system with air (Concept 3) takes suction from the bottom of the tank and
discharges the gelling mixture on top of the ungelled oil.  Air  or gas
injection  can easily be provided in the recirculation line.
                                   33

-------
                                  Isocyanate
  Oil Level
                                                Amine
                                           Duplex Piston
                                              Pump
Figure 2.  Schematic of Recirculation System

-------
jNlozzle Dispersion System

The rotating nozzle dispensing system was designed to operate with a
single line and a single pump to move both gelling agents with a crude oil
flush between agents. The nozzle selected is a NCR-15, 1 1/2-inch
revolving cellar nozzle, as sold by the Fyr-Fyter Company.  The six nozzle
holes are threaded for installing the 1/8-inch solid stream nozzles required
by the gel agent volumes  and rate of discharge.  A cable and winch,
attached to an  "A" frame,  is used to move the nozzle vertically through
the crude oil.  Figure 3,  Schematic of Nozzle Dispensing System, is  a
schematic drawing of the  rorating nozzle system.
                         Large-Scale Testing

A single chemical agent combination and a single crude oil were used in
all tests to provide consistent data and to keep the number of tests and
cost at a reasonable level.  The agents were Armeen C and MT 40.  The
oil was  a  Wise County (Texas) conglomerate from the L. A. Henry Lease
having an API gravity of 38.


Recirculation System (Concept 2)

Four tests  were conducted using this concept at 5-percent agent concen-
tration.  The oil was recirculated with a Marlow-type HAIEC self-priming,
centrifugal petroleum pump.  The suction was taken just below the top of
the oil level  in the tank,  and  the mixed oil and gelling agents were dis-
charged on the bottom of the tank.  The gelling agents were applied to the
recirculated oil,  using a Neptune duplex chemical feed  pump.  The amine
was  injected in the pump  suction and the isocyanate in the pump discharge.
In theory,  the mixture of oil and gelling agents was to be just ready to gel
when discharged  from the pipe in the bottom of the  tank, gel on discharge
with little  or no dilution from  the crude oil in the tank and gradually dis-
place the ungelled oil toward  the top of the tank.  In  practice,  the gelled
oil had such  a neutral buoyancy in the crude oil that it moved with any
currents in the tank rather than stay on the bottom. Therefore, instead of
getting a uniform homogeneous gel throughout the tank,  a mixture of gelled
lumps in ungelled crude always resulted.

During these tests it was learned that the gel agent pumps and piping,
particularly the isocyanate system, must be clean and dry before using to
pump the agent.  Particulars on this series of tests are  included in Table
XI, Recirculation Tests (Concept 2).


Recirculation With Air System (Concept 3)

Four tests  were conducted using this concept.  Experimental observations
are included  in Table XII,  Recirculation Tests (Concept 3).  The equipment
for this test series was the  same as that for Concept  2  with the addition


                                 35

-------
                                 Isocyanate
                                         Amine
Figure 3.  Schematic of Nozzle Dispensing System
            36

-------
             TABLE XI.  RECIRCULATION TESTS (CONCEPT 2)
     Test
  Date               1/25/70      1/29/70       4/9/70       4/15/70

  Oil  (gal.)            450         1,000           500        1,000

  Recirculation
  Rate,  Oil(gpm)       8.2           8.2           8.0         8.0

  Amine Rate (gph)     17.9          17.9          17.9        13.5

  Isocyanate
  Rate (gph)           6.6           6.6           6.6         5.0

  Percent Gel
  Agents by  Weight    5.0           5.0           5.0         3.0

  Penetrometer     Floating
  Reading             Lump           255           313         300
                      285
  Remarks:
      Test 1   Lumps of gelled oil suspended in ungelled crude.  Flowed
              freely from 2-inch opening in bottom of tank.

      Test 2   Same as Test 1.

      Test 3 - Extreme difficulty in keeping isocyanate pump lines open.
              Frequent cleaning necessary.  Resultant gel poorer than
              previous tests.

      Test 4 - Changed discharge to reduce mixing of gelled and ungelled
              oil.  Results same as Test 1.
of an eductor placed in the discharge line.  The suction-line was changed
from the top of the oil level to the bottom of the tank,  and the discharge
line placed at the top level of the oil.  The discharge line included an
elbow which made the mixture discharge parallel to the oil surface. The
eductor was placed about three feet ahead of  the discharge point and was
fitted with a low pressure nitrogen line into the throat.  The  nitrogen was
used in lieu of a compressed air source for convenience.

The first test in this series (Test 5) used too  much nitrogen.   An agent con-
centration of 3. 5 percent was used.  The gelled oil increased in volume by
                                 37

-------
          TABLE XII.  RECIRCULATION TESTS (CONCEPT 3)
    Test            5             6	   7            8
Date             4/22/70      5/4/70      5/11/70       6/2/70

Oil (gal.)           1,000        950         1,000          1,000

Re circulation
Rate (gpm)          7.8          7.8          6.5           6.3

Amine Rate (gph)     13.5         13.5         13.5           13.5

Isocyanate
Rate (gph)          5.0          5.0          5.0           5.0

Nitrogen Rate
(cfh)               100           30          10             0

Percent Gel
Agents by Weight    3.5          3.5          5.0           5.0

Penetrometer
Reading            313          392          246           209
Remarks:

   Test 5 - Trouble with isocyanate pump lines clogging.  Resultant
            gel very foamy.  Volume increased 20 percent.  Lumps of
            gelled oil and ungelled oil flowed freely out of 2-inch
            opening for about 100 gallons, then flow slowed as
            ungelled oil decreased.

   Test 6 - Gel somewhat thinner than Test 5, but much less foamy.
            Oil appears to be uniformly gelled except for bottom part
            of tank.  Oil flowed freely until all  poorly gelled oil
            exhausted.  Flow then slowed.

   Test 7 - Uniform gel except for bottom amount in tank.  Oil  flowed
            freely out of  2-inch valve on bottom of tank until mixture
            of gelled and ungelled oil exhausted.   Gelled oil then
            flowed slowly out of valve.

   Test 8   Uniform gel except for about 100 gallons on bottom of tank.
            Gelled oil exuded out of 2-inch valve until 4.75 head of
            gelled oil reached.   Gate  opened one inch, gelled  oil
            exuded until one foot head reached.
                                38

-------
20 percent which was assumed to be intolerable aboard ship.  Also, the
increased foaming of the gel reduced its strength to resist flowing.

As in all tests in this series,  the centrifugal pump used in recirculation
lacked sufficient lift to remove the bottom few inches of crude oil which
had been thickened slightly by dilution of the gel  mixture first returned to
the tank prior to complete gelling.  In an actual shipboard situation,  this
pump would be  a  positive displacement pump, eliminating the difficulty.

The next test (Test 6) was a repeat of the previous test, using less nitrogen.
The resulting gelled oil was less foamy but still lacked strength to resist
flowing out of the 2-inch valve at the bottom of the tank with 6.75 feet of
gelled oil above the valve.

Test 7 increased  the gel agent/oil ratio to 5 percent and reduced the
flotation nitrogen.  The gel  formed well about five  seconds after discharge,
floated on  top of  crude in the tank and, as the test progressed,  gradually
lowered the ungelled oil level.  The resulting gelled oil flowed  out of the
2-inch valve opening under 6.75 feet of gelled oil  head.  No attempt was
made to determine the limits on this extrusion.

Tests made during the recirculation of oil alone showed that  sufficient air
would be introduced by the normal vacuum produced by  the eductor.   The
next test (Test 8) was performed in this manner.  The gelled oil floated as
before with the resulting gel the stiffest of any test and fully equal to any
produced in the laboratory.  After the gelling was  completed,  the 2-inch
valve on the bottom of the tank was opened, allowing the gelled oil to
flow.  The extruded oil was continually  removed to keep the external
resistance to flow uniform.  The gelled oil flowed until the head inside
the tank was reduced from 6.75  feet to 4.75 feet,  measured from the  top
of the opening.  The sliding gate was then opened  one inch,  providing a
. 33-square-foot opening.  The gelled oil slowly extruded until a 1-foot
head was reached as measured from the top of the opening.  The gate was
then opened to two inches and the remainder of the gelled oil flowed  out.


Nozzle Systems (Concept 1)

The concept being tested in this series originally  included a series of
nozzles which would revolve around a central shaft, distributing the
gelling agents.  It was  found in practice, however, that for small-scale
tests there are  no commercial devices suitable. One such device which
was investigated was a fire fighting cellar nozzle.  The six nozzle openings
were much too large and were modified to suit the volumes of agents  to be
used on these tests.  The 5/16-inch and  1/4-inch  nozzle openings were
tapped and fitted with 1/8-inch nozzles.  It was found  that the  discharge
from the nozzles  produced insufficient torque  to provide much rotational
movement when submerged in a liquid.  Also,  the  close clearances of the
bearing surfaces  were easily fouled by small particles,  stopping the
rotation.
                                39

-------
In search for suitable nozzles, a tank cleaning nozzle manufactured by
Spraying Systems Company was found which had a 360 degrees coverage
approximately normal to the delivery pipe. The deflector cap was machined
so as to make the spray normal to the supply pipe instead of the upward
deflection.

Liquid-in-liquid spray patterns were observed prior to gelling tests in an
open top water tank with colored  particles to indicate flows.  It was found
that the mixing zone of each nozzle hole included a  sector of about 25
degrees.  The practical radius of the mixing zone appeared to be about
250 nozzle diameters. Some mixing was observed beyond that point but
appeared to be very minor.  In view of this, the nozzle was modified to
have eight holes  .093 inch in diameter.  This would result in almost com-
plete mixing coverage to the nearest tank wall.

The first test of the series (Test 9) was made delivering both the amine
and isocyanate through the same  pipe and nozzle with a crude oil  flush of
the lines between deliveries.  The crude oil flush did not remove all of
the amines.  The reaction between the full strength isocyanate and residual
amine was sufficient  to partly clog the nozzles and reduce the mixing
effectiveness.  Subsequent tests were performed using  separate pumps,
lines and nozzles for each gelling agent.

The next three tests in the series were essentially the same in procedure
and end results.  Nozzle size changes were made to attempt to improve
mixing,  and filters were installed later to protect the nozzle openings.
These tests all appeared to mix the amine satisfactorily into the crude oil,
but very shortly (15 to 30 seconds) after starting the isocyanate injection,
the crude oil began to thicken rapidly, reducing the radius of the mixing
zone  and eventually allowing mixing only in the immediate vicinity of the
nozzles.  Table XIII,  Nozzle Tests (Concept 1), lists particulars on this
series.


Nozzle With Gas Agitation System

The mixing of the gelling agents into the crude oil appeared to need
improving.  The nozzle test setup was changed to include an inert gas
sparger placed on the bottom of the tank.  A series of 16 nozzles placed
one foot apart and three inches from the outer wall of the  tank were
installed in the tank.  A nitrogen bottle with regulator and rotameter were
connected by hose to the sparger. It was believed that the gas bubbles
rising along the walls of the tank would improve the mixing by increasing
the circulation within the tank.  Table XIV, Nozzle Tests,  Gas Agitation,
details these tests.

The first test in this series  (Test 13) resulted in a gel only slightly  thicker
than the crude oil.  The low pressure  of the isocyanate injection was
thought to be responsible for the  poor gel. The isocyanate injection
pressure was increased for the next test (Test 14)  by reducing the nozzle
size. The injection pressure increased and the gel consistency thickened


                                 40

-------
               TABLE XIII.  NOZZLE TESTS (CONCEPT 1)
Test
10
11
12
Date                6/16/70     6/22/70

Oil  (gal.)            500         500

Amine   (gal.)          23          23

Isocyanate  (gal.)       9          g

Percent Gel Agent
by Weight             6.5         6.5

Number and Size                 6-. 081
Nozzles (in.)          8-.093     8-.106

Injection
Pressure (psi)         80          50

Injection Time (min)    4.0         3.0
          6/24/70

            500

            23

            9


            6.5

           6-.093
           8-.125
         6/25/70

           500

            23

            9


           6.5

         6-.093
         8-.125
      80 Isocyanate  48 Isocyanate
          40 Amine   40 Amine
            4.5
           2.25
Remarks:

   Test  9 - Thin,  mushy gel.  Localized gelling in center of tank.

   Test  10- Thin gel with localized heavier gelling in center of tank.

   Test  11- Thin gel with localized heavier gelling around center of
             tank.  Isocyanate nozzles plugged.

   Test 12 - Filters ahead of nozzles.  Thin gel with localized heavier
             gelling in center.
 considerably, but still thinner than desired.  Since the limit of volume
 versus pressure available from the turbine injection pump had been
 reached,  a pressure shooting tank was installed which would use nitrogen
 pressure to inject the isocyanate.

 The nozzle arrangement for both amine and isocyanate  were  changed at
 this time. A 1-inch diameter pipe coupling was drilled and  tapped to
 accept a commercial solid stream nozzle.  The amine nozzle head con-
 tained 12 Spraying System Company nozzles (number T0010)  with 0.086-
 inch openings.  The isocyanate nozzle head contained  6  equally spaced
 nozzles (number TT0008) with 0.061-inch openings.
                                 41

-------
             TABLE XIV.  NOZZLE TESTS,  GAS AQUATIC:7
Test
Date
Oil (gal.)
Amine (gal. )
Isocyanate (gal.)
Percent Gel Agents
by Weight
Size and Number
Nozzles, Amine
(in.)
Size and Number
Nozzles,
Isocyanate (in. )
13 14 15
7/2/70 7/9/70 7/14/70
500 500 500
23 23 23
999
6.5 6.5 6.5
8-. 125 8-. 125 12-. 086
6-. 093 6-. 086 6-. 061
Injection Pressure,
Amine 26 35 34
Injection Pressure,
Isocyanate 22 43 100
Gas Flow (cfh)
Remarks :
Test 13 Oil
40 35 30

only slightly thicker.
16
7/16/70
500
23
9
6.5
12-. 086
6-. 125
33
100
30


Test 14 Uniform gel penetrometer 325.
Test 15 - Uniform gel penetrometer 295.
Test 16 - Uniform gel penetrometer 254.
The next test (Test 15) and the confirming test (Test 16) both resulted in
gels which were equal to any of the laboratory gels.
                                 42

-------
                               Summary

Mechanical Systems

The recirculation system is a successful and simplified method to gel
oil. All the oil in a tank was treated uniformly and gelled. The deter-
mination of the pump rate and proportioning of the chemicals into the
circulation line were accomplished by conventional equipment and good
engineering practices. The mixing of the chemicals in the circulation
line was sufficient to result in a gel which was  comparable to gels
produced under laboraotry  conditions.

The method, however,  is relatively slow and would require approximately
four hours to gel an average center tank using a 100 barrel/minute
recirculation pump.

Based on program results at this point, the recirculation dispersion
system required no further test to determine operating limitations.
Actual equipment size and power requirements associated with particular
ship installations,  however, must be determined for each installation.
This conclusion is applicable to both the  shipboard installed and
helicopter transported systems.

Non-rotating nozzles of proper size and application pressure will
produce 360 degrees coverage with sufficient agitation to adequately
mix and disperse the agents in a 1,000-gallon tank. A separate  line
and nozzle must be used for each agent and both chemicals could be
dispersed simultaneously. The resulting gel is comparable to gels
produced under laboratory  conditions.

This method is relatively fast. It took 21/2 minutes to complete
dispersion in the 1,000-gallon tank with a delay between agent injection
injections. In fact, total dispersion must be rapid because of gel set-
up time.

The program results provided the necessary information from which the
parameters determining  the operating  limitations and equipment require-
ments for larger volume nozzle dispersion  system tests can be
identified and planned.  The critical factors related to the nozzle
dispersion radius as related to the  dimensional volume to be gelled.
The 1,000-gallon tests  provided one limited volume related set of
variables.
Gel Characteristics

Program tests have shown that gels comparable to those produced in
the laboratory can be produced in large volumes with the mechanical
systems  devised.
                                43

-------
The tests produced two data points for the gel relative to its ability to
resist flow through a rupture with  differential pressure. These two
points  indicate that the best gel developed will not keep the cargo from
flowing through the rupture under all conditions.

In all of the 1,000-gallon tests the gelled oil would flow out of a
two-inch valve opening under a head of 6.75 feet of gelled oil. This
translates to a differential pressure of 2. 5 psi.  An attempt was made
to define the upper limit of the head of gelled oil under which no flow
would occur. It was  found that for a gel  which was comparable to the
best laboratory gel,  a  head of 4.75 feet  of gelled oil produced no flow
through the two-inch valve. This  relates to a pressure differential
of 1.75 psi and a shear stress of  12,000  dynes  per square centimeter.
Referring to the rheogram on the gelled oil Figure 1, this  indicates an
extremely high shear rate for the small opening.

Similarly, the gate tests defined a no flow pressure differential for the
best gel of . 35 psi or a shear stress of 24,000 dynes per square centi-
meter with gap dimensions of 4 feet wide by  1 inch high.  The penetrom-
eter reading on this gel was 209 which translates to a yield shear stress
of 25,000 dynes per square centimeter.

The tank tests indicated that the gel,  as produced in the laboratory
and reproduced in the tests,  lacks the strength  to withstand differential
pressures greater than 1.75 psi  (heads of 4.75 feet), when bridging
a 2-inch round hole or pressures of .35 psi (heads of one foot), when
bridging holes 4 feet long by 1 inch high.

In the normal fully loaded shipboard containment configuration (plotted
on Figure 4, Pressure Differentials Cargo Versus Seawater), differential
pressures at the bottom of a tank will range from .7 psi to 1.0 psi
outward. At the waterline,  this range becomes 3.4 psi to 6.0 psi. When
a rupture occurs below the waterline, oil will flow through the rupture
until the pressure is equalized or  to the  lowest  point of the rupture  if
above the waterline. When the rupture passes through the waterline,
water will enter the tank and eventually  displace all the oil.  In summary,
the deeper the rupture  is located the more favorable the situation.
                    *
Under the worst conditions, (a rupture above, at, or through the
waterline) the gel might be effective in holding  cargo a minimum of
one foot above the waterline (Figure 4, Pressure Differentials  Cargo
Versus Seawater). For  a rupture  completely below the waterline oil
loss prior to the gel becoming effective would be less than the above
and a function of the maximum dimension rupture area and its depth
below the waterline. The minimum loss would be experienced with the
rupture at the bottom.

It can readily be  seen  that the limits of effectiveness of the best gel
produced, in terms of cargo loss to be avoided,   is a complicated
                                44

-------
                                           Deck
      Western Sun
        30,000 DWT
               Esso Manhattan
                115,000 DWT
20    10
10    20  j  20    10    0

   Static Pressure,  psi
30
 Figure 4.  Pressure Differentials Cargo Versus Seawater
                          45

-------
function of the rupture dimensions, its location in relation to the water-
line,  the size of the tank and the time lapse between rupture and gelling.
Furthermore,  the performance  criteria becomes more severe if the water-
line is lowered as a result of  the ship running aground.

While the observations of gel performance in  the tank tests were  limited,
it seems apparent that the strength range of the gels produced was
insufficient to preclude flow from a rupture of significant size over all
but a  limited range of pressure differentials.  The value of the gellation
system  developed would seem to rest primarily on the reduced flow
rate of the gelled cargo and decreased pollution in any  given time span.
The strength  ranges for the gels produced, derived from various test
data is  quite consistent. Based on this consistency, some generalizations
from the rheogram can  be made. Analysis of the rheogram indicates that
the flow rate of the gelled cargo will decrease as  flow occurs and
terminate at  some positive pressure differential (critical shear stress).
Because of this, the gel would also be expected to resist flushing and
intrusion of water resulting in even smaller rates of loss from the tanker
than ungelled oil. Establishing these performance characteristics
precisely for the gellation system developed requires extensive testing
under more varied and  controlled conditions than those  required for the
development  of the system to this point.

In addition to strength, the handling characteristics of  the gel were
observed. Following each test,  the gelled oil was transferred by a
positive displacement  pump to a larger waste  pit where it was subjected
to local climatic extremes  for several months. The pumped material
reformed into large globs of gelled oil, which floated on the water in
the pit. At no time was there an oil slick or loose oil observed on the
surface of the water.
                                46

-------
                             SECTION 5

                        SYSTEM APPLICATION


Most United States tankers built prior to the  1960s have relatively
short cargo tanks usually 40 feet or less in length.  The cargo section
is divided by two longitudinal bulkheads into port, center and starboard
tanks.  The center tanks are usually approximately square and the wing
tank width usually about half the length. The majority of the ships in
United States waters are of this  configuration. Two classes of ships
now in service were selected as application examples.
                             Requirements

The 30,000 DWT class tanker represents the most frequently used ship
in United States service.  The dimensions of an average wing tank in
such ships (The Western Sun, Sun Oil Company owner) are 40 feet long,
24 feet wide and 45 feet deep. The nominal capacity, loaded to 4-foot
ullage, is 7, 500 barrels of crude oil.

If the nozzle method was to be used  to gel the oil, extrapolation from
the 1,000-gallon test would indicate the need for 2-nozzle probes
in this tank, each to gel half the tank or a plan area of 20 feet by
24 feet.  To achieve this,  each nozzle probe would require the following:

                Isocyanate nozzles            10
                Amine nozzles                20
                Nozzles  size                0.75  in.
                Isocyanate pump size         55 bbl per min
                Amine pump size              100 bbl per min
                Quantity isocyanate          2, 325 gal-
                Quantity amine              7, 250 gal-

If the recirculation method were used to gel the oil in the tank, the
following would be required:

                Recirculation pump            200 bbl per min
                Isocyanate pump              55 gal-per min
                Amine pump                  150 gal. per min
                Quantity isocyanate          5,560 gal.
                Quantity amine               14, 500 gal.

The 250,000 DWT class ship represents one of the largest ships now
in worldwide service. The dimensions of a forward wing tank on the Esso
Scotia, a 249,952 DWT tanker launched in 1969, are 200 feet long,
47 feet wide and 84 feet deep. The nominal capacity of this  tank (98%
full) is 144,000 barrels of crude oil.


                                47

-------
If the nozzle method were used to gel the oil,  extrapolation from the
1,000-gallon test would indicate the need for  5 nozzle probes in this
tank, each to gel the oil over a planned area of 40 feet by 47 feet. To
achieve this, each nozzle probe would require the following:

          Isocyanate nozzles         20
          Amine nozzles              36
          Nozzle size                1.5 in.
          Isocyanate pump            350 bbl per min
          Amine pump                750 bbl per min
          Quantity isocyanate         375 bbl
          Quantity amine             1,075 bbl

If the recirculation method were used on  this ship to gel the oil in one
tank, the following would be required:

          Recirculation pump         2,000 bbl per min
          Isocyanate pump            550 gal.  per  min
          Amine pump                1,500 gal. per min
          Quantity isocyanate         2,000 bbl
          Quantity amine             5,500 bbl

Previous experience with gelled hydrocarbons has shown that they can
be pumped satisfactorily with a positive  displacement pump.  Experience
gained in removing the gelled oil from the 1,000-gallon tank has
verified this. After each test, the gelled oil was pumped out of the
tank using a Worthington gear pump, 7 GAUM, through a 2-inch suction
hose and a 2 I/2-inch discharge hose. Extremely stiff or well set  gels
can be thinned if necessary  with heat using  steam or by applying  fresh
crude. In almost every case unloading would be accomplished in port
where a variety of acces sable pumps and other equipment would be
available.

                       System Economic Analysis

In most cases, the availability of an oil  gelling system should be
considered an insurance policy against certain sources of water pollution.
The response capability of the system, in time, is the critical factor
determining the pollution risk assumed. Lower risk indicates a high
system cost resulting from a more expensive and immediately employable
system.

Two types of systems can be considered, one protecting each ship
plying United States contiguous water, another protecting  geographic
areas available to any ship  in extremities or with a probability of
being placed in extremities.
                                48

-------
The shipboard installed system will have the capability to commence
gelling the tank in jeopardy in the shortest possible time,  thus reducing
the free oil lost before gelling starts.  If it is assumed that the ship is
the Sun Oil Company's Texas  Sun with 51,700 DWT capacity,  the ship-
board installed system will cost $150,000 as listed in Table XV,
Equipment List,  Concept 1 Gel System, Shipboard Installed. If it is
further assumed that four tanks on the ship will need protection and
that the chemical  life is three years,  the chemical cost will be $135,666
per year.  The equipment cost, amortized over 20 years, will be $7, 500
per year,  making a chemical and equipment cost of $ 143, 166  per year.
A routine  trip for this ship is from the Texas Gulf to Marcus Hook,
Pennsylvania, a distance of 1,825 miles. Thirty trips a year carrying
51,700 tons  per trip will amount to 2,830 million ton-miles per year.
This will  result  in a ton-mile  cost of  0.051 mils. Compared with a
1. 13 mils/ton-mile cost fora 2,000-mile haul in a 48,000 DWT tanker,
this represents an increased shipping cost of approximately four percent
for this protection.

For ships  in United States  territorial waters which do not have the ship-
board installed system, a helicopter transportable system complete with
pumps, hoses, and chemicals can be devised which will provide
assistance in times of distress. These helicopter systems, if placed
every 150 miles along the United States  coast,  could be in convenient
reach of any ship  in jeopardy  in a  relatively short time. At each base,
there could be stored sufficient chemicals to gel one 25, 000-barrel
tank, equivalent to a center tank on the  Esso Manhattan.  Additional
chemicals and equipment,  if needed,  could be transported from the
adjacent helicopter bases. Assuming the dispensing equipment cost is
the same  as  that for a shipboard installed system, equipment  stored
at the bases with  a 20-year expected life would result in a yearly  charge
of $185,000. The  cost of chemicals stored at these sites will  total
$6, 300,000. A three-year chemical life would result in a yearly charge of
of $2, 100,000 for chemicals. Approximately  100 billion ton-miles  of
United States waterborne oil transport are traveled each year. This
would result in a ton-mile  cost of .002 mils and, when compared to the
average 1. 08 mils per ton-mile cost for a 3, 000-mile haul in a 48, 000
DWT tanker, represents a shipping cost  increase of only 0. 18 percent
for this protection.

While at first glance the helicopter transported  system appears to  be a
much less costly  system,  it must be remembered that the time lag
between the  disaster and the arrival of the equipment and chemicals
has allowed  oil to freely flow from any rupture.  An additional  cost will
be incurred  in the  capital costs for the helicopters and the expenses
incurred in manning and maintaining the  systems.

While some comprehensive studies of the occurrence, nature,  and
characteristics of  tanker spills have recently been made available,
most published information tends to dwell heavily on the pollution
aspects and  results. Important details concerning the location,  size


                                 49

-------
         TABLE XV.  EQUIPMENT LIST CONCEPT 1 GEL SYSTEM,
                   SHIPBOARD INSTALLED
              Item                                Estimated Cost
Isocyanate tank,  30, 000-gallon capacity             $  9,000

Amine tank, 100, 000-gallon capacity                   20,000

Isocyanate pump, 400 gpm at 100 psi                    2, 500

Amine pump,  1, 000 gpm at 100 psi                      5,000

Motor generator,  100 kw diesel driven                  12,000

Distributor piping and hose                             3, 500

Mixing nozzles                                        2,000

Flexible hose,  100 feet                                1,000

Hoist and derrick, portable,  extendable                 15,000

Electric control panel,  starters,  etc.                    5,000

Electric wiring and cables                              3,000

Nitrogen blanket system                           	2, OOP

          SUB TOTAL                               $80,000

Marine engineering                                    20,000

Installation                                          50, OOP

           TOTAL ONE TIME COST TO INSTALL       $ 150 , OPP
                                 50

-------
and history of the actual leakage for the documented occurrences  seem
to be generally unavailable.

Likewise,  information concerning the historical costs of eliminating the
spills and their environmental effects  is sketchy and not very precise.
Many costs are not included in the published figures, and in fact are
not known. Similarly, the volumes  of oil spilled and removed are  often
not known, or appear as very rough estimates. Regardless of the
precision of the aata, however, there  is a need to establish an economic
framework in which to evaluate the cost-benefit relationship of a  pro-
posed oil spill control system.

Available information indicates that the identified cost to remove  spills
has ranged from $10 to $50 per barrel of oil spilled. This range does
not include many important  costs,  such as wildlife care and beach
rehabilitation,  gratis labor and many other costs  currently classified  as
intangibles,  necessary to completely restore the environment.  It is
possible that a more realistic total cost might approach $ 150 to $200
per barrel.

While the  reduction in flow rate under large  pressure differentials and
rupture  sizes,  and resistance to water displacement has not yet been
accurately defined for the oil gel,  it is possible,  within the above cost
framework, to identify the reduction in pollution required for the system
to be economically appealing. The  annual cost in chemicals and equip-
ment should be. offset by an expected reduction  in oil exiting to the
general environment, thereby eliminating the clean-up costs for that
increment.

Based on an estimated total clean-up cost of $ 100 per barrel, the ship-
board installed and transportable systems should  be capable of
eliminating the flow of approximately  1, 500  and 23,000 barrels  per
year respectively into the sea to justify the  chemical and equipment
costs.

Observation of the waste pit during tests  indicated that the gelled oil
which does leave the ship will tend to remain in lumps and will not
form slicks.  These characteristics  provide other intangible savings in
reducing the  costs associated with wildlife, marinelife and shoreline
damage.
                                 51

-------
                            SECTION 6

                      ACKNOWLEDGEMENTS
We wish to express our appreciation to the Agricultural and Marine
Pollution Control Branch of the Federal Water Quality Administration
for their guidance and assistance in the conduct of this study. In
particular, we wish to thank Mr. Harold Bernard,  Chief of the Branch,
and Mr. George Putnicki, Project Officer, for their assistance which
has aided immeasurably  in the compilation,  interpretation and
presentation of the information in the report.

We would also like to express our thanks to the many  chemical
suppliers  who furnished gelling chemicals during the laboratory
screening phase, supplied us with brochures describing the available
chemicals and responded to our specific questions.

Finally, we wish to thank the Atlantic Richfield Company, Phillips
Petroleum Company,  Standard Oil of California and the Sun Oil
Company for their invaluable help by  supplying crude oils typically
carried in their tankers,  and the Sun Oil Company for their
cooperation in allowing us to inspect their ships and shipping
facilities, and furnishing us drawings and data on their ships.
                              52

-------
                             SECTION 7

                             REFERENCES


                           Literature Cited

1.  Anonymous, Waterbore Commerce of the United States,  Calendar
    Year 1965, Part 5,  National Summaries;  U.  S. Army Corps of Engineers,
    Vicksburg, Mississippi, 1965.

2.  Anonymous,  "Crude Imports by Origin,  "The Oil and Gas Journal,
    p.  131, July 28, 1968.

3.  Griddle, D. W. and Dreher,  J. L. , "Yield Points of Lubricating Grease",
    in Proceedings; 26th Annual  Meeting of National Lubricating Grease
    Institute.

4.  Saunders,  J. H. and Frisch,  K.  S. ,  Polyurethanes, Chemistry and
    Technology,  Part I, Chemistry,  John Wiley and Sons, Inc. ,  1962


                         Selected Bibliography

Battelle Memorial Institute,  Oil  Spillage Study Literature Search and
Critical Evaluation for Selection of Promising Techniques to Control and
Prevent Damage, November 20, 1967.

Anonymous, "Chemical Ocean Shipping Costs,  Chemical Week, p. 75,
October 29, 1969.

Gilmore, G. A. , et al,  Systems Study of Oil Cleanup Procedures, 2 vols,
Dillingham Corporation,  February 1970.

Anonymous, 1969 World Tanker Fleet Review, John I. Jacobs & Co. , Ltd.

Anonymous, "U. S. Tanker Rates Sail to Record Highs, "  The Oil and Gas
Journal,  pp. 56,58, February 16, 1970.

Anonymous, "American Bureau of Shipping Reports Another Record Year,  "
Maritime Reporter/Engineering News, p.  21,  April 15, .1970.

Anonymous, "Modest Gains Forecast for  1970,  " The Oil and Gas Journal,
p. 116, January 26, 1970.

Ackers, H. G. , Review of Welded Ship Failures National Academy of
Sciences, National Research Council, December 15, 1970.
                                 53

-------
Smith,  J. E. , Torrey Canyon Pollution and Marine Life. Cambridge Univer-
sity Press,  1968.

USATRECOM Technical Report 64-66,  A Study of Rapid Solidification of
Hydrocarbon Fuels, Southwest Research Institute,  1965.

Beerbower, A. and Philippoff, W. ,  History of Gelled Fuels -  Their
Chemistry and Rheology, Aircraft Fluids Fire Hazard Symposium, Cornell
University,  1966.

Yee, J.  E. ,  Oil Pollution of Marine Waters,  Department Of the Interior,
1967.

Bates,  M. ,  Water Pollution by Oil Spillage,  Southern Methodist University,
1969.

Blade,  O. C. , Bibliography of Reports Containing Analyses of Crude Oils
by the Bureau of Mines  Method, Department  of the Interior, 1959.

Brockis, G. J. ,  Preventing Oil Pollution of the Sea, Paper presented at
an Informal Meeting Biologische Anstalt,  Helgoland,  September 22 and
23, 1967.

William, M. L.  and Ellinger, G. A. , Investigation of Fractured Steel
Plates Removed from Welded Ships, National Bureau of Standards,
February 25, 1949.

Williams,  M. L. ,  Meyerson, M. R. , Kluge,  G. L. and Dale,  L. R. ,
Examinations and Tests of Fractured Steel Plates Removed from Welded
Ships, National Bureau  of Standards,  September 22, 1949.

Williams,  M. L. ,  Meyerson, M. R. , Kluge,  G. L. and Dale,  L. R. ,
Investigation of Fractured Stell Plates Removed from Welded Ships,
National Bureau of Standards, June 1,  1951.

Williams,  M. L. ,  Examinations and Tests of Fractured Steel Plates
Removed from Welded Ships. National Bureau of Standards, April 2,
1953.

Williams, M. L. ,  Investigation of Fractured Steel Plates Removed  from
Welded Ships, National Bureau of Standards,  September 17,  1957.

Williams, M. L. ,  Correlation of Metallurgical Properties, V-Notch
Charpy Energy Criteria,  and  Service Performance of Steel Plates from
Fractured Ships, National Bureau of Standards, November 25,  1957.

Technical Progress Report of the Ship Structure Committee,  March  1,
1948.
                                 54

-------
Second Technical Progress Report of the Ship Structure Committee,
July 1, 1950.

Third Technical Progress Report of the Ship Structure Committee, August
1,  1953.

Fourth Technical Progress Report of the Ship Structure Committee,
April 1, 1959.

Annual Report of the Ship Structure Committee,  May 1, 1962.

Biennial Report of the Ship Structure Committee, October 30,  1964.

Stevens, D.R. and  Gruse, W.A. , The Chemical Technology of
Petroleum, Mellon Institute of Industrial Research, June, 1942.

Morrison, R.T. and Boyd, R.N.,  Organic Chemistry,  Allyn and Bacon,
Inc. , July,  1959.

Implications of the  Gelled State of Matter for Naval Applications, The
Western Company of North America, February,  1968.

ASTM Standards,  Petroleum  Products,  Part 17, January,  1967.
                                55

-------
                             SECTION 8

                             GLOSSARY
    - American Petroleum Institute
ABS - American Bureau of Shipping

DWT - Dead Weight Tons,  the carrying capacity of a ship

Wing tank - Tanks along the sides of a ship divided into  port (left)
and starboard (right) tanks

_C enter tank -  A line of tanks in the center of the ship between port
and starboard wing tanks

Ullage - The void above the liquid in a tank to allow for expansion, etc.

Stoichiometric ratio - The theoretical chemical combining ratio

The Bingham yield point - The point at which a solid begins to flow
when subjected to constantly increasing stress

Penetrometer reading - A measure of gel strength outlined in ASTM-217.
                                56

-------
                             SECTION 9

                            APPENDICES
A.  Shipping Characteristics	  59

    Table A-1:     Typical United States Built Tankers	  61
    Table A-II:    Tanker Fleet of Leading Countries	  62

B.  Penetrometer Use	  66

C.  Chemicals and Suppliers	  71

D.  Soap Gel Screening Data	  74

    Table D-I:    Soap Gel Screening Data	  75
    Table D-II:   Soap Gel Screening Data	  79

E.  Swelling Gel Screening Data	  81

    Table E-I:     Swelling Gel Screening Data	  82

F.  Isocyanate/Amine Gel Screening Data	  83

    Table F-I:     Amine Tonco 90 Gel Screening Data	  84
    Table F-II:    Amine HMD Gel Screening Data	  87
    Table F-III:   Amine MT-40 Gel Screening Data	  90
    Table F-IV:   Amine MRS Gel Screening Data	  93
    Table F-V:    Amine MDI Gel Screening Data	  96
    Table F-VI:   Amine Isonate 136-T Gel  Screening
                  Data  	  97
    Table F-VII:   Amine TDI 65/35 Gel Screening Data	  98
    Table F-VIII:  Amine NGO-10 Gel Screening Data	  99
    Table F-IX:   Amine TDI 80/20 Gel Screening Data	104
    Table F-X:    Amine T.P.M.T. Gel Screening Data	108
    Table F-XI:   Amine TMD- 1  Gel Screening Data	 . 109

G.  Stability in Storage Tests for Gel Constituents	112

H.  Optimum Ratio Studies	114

    Figure H-l:   Armeen C and MT-40	115
    Figure H-2:   Jeffamine 400  and Tonco 90	115
    Figure H-3:   Armeen O and TDI 80/20	116
    Figure H-4:   Armeen C and TDI 80/20	116
    Figure H-5:   Armeen L- 15 and TDI 80/20	117
                                57

-------
    Figure H-6:    Armeen L-11 and TDI 80/20	117

I.   Gel Strength Versus Percent Gelling Agent	118

    Figure 1-1:     Armeen L-ll and TDI 80/20	119
    Figure 1-2:     Armeen L-ll and TDI 80/20	120
    Figure 1-3:     Armeen C and MT-40 	121
    Figure 1-4:     Armeen C and MT-40	122
    Figure 1-5:     Jeffamine 400 and Octadecyl Isocyanate. .  .  123
    Figure 1-6:     Jeffamine 400 and Octadecyl Isocyanate  . .  .  *24
    Figure 1-7:     Armeen L- 15 and TDI 80/20	125
    Figure 1-8:     Armeen L-15 and TDI 80/20	  126
    Figure 1-9:     Armeen O and TDI 80/20	127
    Figure I-10:    Armeen O and TDI 80/20	128
    Figure I-11:    Armeen C and TDI 80/20	129
    Figure 1-12:    Armeen C and TDI 80/20	130

J.   Gel Agent Dispensing Concepts	131

    Table J-l:     Gel Agent Dispensing Concepts	132
                               58

-------
       APPENDIX A



SHIPPING CHARACTERISTICS

-------
                             APPENDIX A

                     SHIPPING CHARACTERISTICS


There are approximately 3, 500 tankers in  use worldwide in transporting
crude oil and other petroleum products. Of this number,  352 are American
flag ships over 10,000 DWT. The United States  coastal fleet consists
of about 250 tankers with an average carrying capacity of 22, 500 DWT
or 190,000 barrels of crude  oil.  Table A-l,  Typical United States Built
Tankers, is a  listing of the  tankers which regularly visit United States
ports. The tankers unloading at  United States ports, whether foreign or
domestic, are limited to 100,000 DWT or  less by the generally shallow
Western hemisphere harbors. The large ships, 150,000 DWT to
350,000 DWT, are mainly used in the Persian Gulf-Western Europe or
Persian Gulf-Japan runs.

Prior to  1956,  and subsequent to the closing of the Suez Canal, tankers
were generally limited to 100,000 DWT, or  less, the maximum size which
can travel the canal.  As a direct result of the canal closing, very
large tankers were developed which could operate from the Persian Gulf
to Western Europe at  less cost than smaller ships using the canal. Now
under contract to class in 33 different countries, according to the
American Bureau of Shipping Standards, are 1,795 merchant ships of
all types with a gross total  of 15,700,000 tons or an average of 8,700
DWT per ship. In 1969, orders were placed for 88 tankers in the  20,000
to 30,000 DWT range. As of the end of 1969, the ABS had classed only
46  tankers over 100,000 DWT. While the  trend toward larger ships  is
apparent (estimated at approximately 110  plus ships in service or on
order) for both the Persian Gulf-Japan and Persian Gulf-Western Europe
runs, the major portion of world tanker fleet additions are small ships.

Only American built,  manned and registered ships can be used in United
States coastal traffic, i.e. , both load and unload from United States
ports. Foreign registered ships can either load or unload at United
States ports,  not both. This factor  coupled  with shallow port restrictions
promotes the use of relatively smaller United States built ships in
United States  commerce. In the  future, larger tankers may be unloaded
offshore into pipelines .or smaller shuttle  tankers, or use harbors
deepened extensively  by dredging.

The 1965 waterborne commerce of petroleum products in the United
States amounted to 97 billion ton miles, with an estimated 68. 8
million tons of crude  oil imported via tanker during 1969. The 1969
imports were 40-percent from Canada, 26-percent from Latin America,
15-percent from Africa, 12-percent from the Middle East, and the
remainder from the Far East. This indicates that two-thirds  of the
imports  are transported in the same coastal vessels as used in domestic
trade. Table A-II,  Tanker Fleet of Leading Countries,  lists  the tankers
registered in the top nine countries in 1969.

                                60

-------
                                             TABLE A-I.  TYPICAL UNITED STATES BUILT TANKERS
                       TZ-SE-A1
             Olympic
              Games
 Esso
Dallas
Western      Esso
  Sun      Gettysburg
Sinclair
 Texas
Texas       Esso      Princess       Esso
 Sun    New Orleans    Sophie    Manhattan
Nominal DWT


Draft (ft)
Cargo Capacity
   (bbl)
Number of
  Wing Tanks
Number of
  Center Tanks
Size of Wing
  Tanks (ft)
Size of Center
  Tanks (ft)
Cargo Capacity
  Wing Tanks (bbl)
Cargo Capacity
  Center Tanks (bbl)
    Loaded to 4' Ullagi
 16,700      18,000     28,000
   30
                          18
               30
                          33
             30,000     35,000      47,000       51,700     67,800        70,000     115,000


               35         37          38          40         40            44         49
141,000     15Z.200    225,000       251,100    317,700     395,000     418,000    583,000       586,000    1,000,000
                                      18
                                                 20
                                                 10
                                                              20
                                                               10
                                                                         20
                                                                          10
                                                                                      14
                                                                                      11
                                                                                                  22
                                                                                                  11
                                                                                                             12
                                                                                                                          28
                                                                                                                           14
                                                                                                                                      32
                                                                                                                                      16
36x17x38    36x18x38   40x22x42      40x24x45   40x27x48    30x80x49    45x30x50   102x35x54     40x35x60   40x40x67



36x36x38    36x36x38   40x40x42      40x40x45   40x40x48    40x40x49    45x42x50   102x45x54     40x45x60   40x52x67



  4,000       4,200      6,500         7, 666*     9,300       22,000      11,150*    34,500        15,000     19,300




  8'500       8'500     10>000        11,250*     12,500       13,500      16, 000*    44,200        19,200     25,000

-------
TABLE A-II.  TANKER FLEET OF LEADING COUNTRIES
Country
Liberia
Great Britain
Norway
Japan
United States
Greece
Panama
France
Russia
Ships
Under
10, 000
DWT



127
59

30
41
158
Ships
Over
10, 000
DWT
674
400
303
161
352
172
140
135
167
Total Tonnage
Over 10, 000
DWT
(Million)
30.8
18.7
15.6
13. 3
8.9
5.7
4.8
4. 2
3.9
Average
DWT
(Over
10, 000)
45,000
46,000
51,000
82,000
25,000
33,000
34,000
31,000
23,000
                      62

-------
The situations under which it will be necessary to employ the gelling
system are varied and serve to identify the operating and employment
characteristics which  must be designed into the system.
                              Collision

The greatest potential for accidental oil spills is from collisions. The
Liverpool Underwriters Association reported  196 collisions  involving
238 tankers occurred during the period of June 1964 to April 1967. Oil
spills resulted from 22 of these collisions. The spills resulting from
a collision may be minor (a small amount of  oil lost from one tank) to
major (a large amount of oil lost from more than one tank). There is,
however, little likelihood of additional leakage over that caused by the
collision. Therefore, if the leaking oil is to  be stopped by gelling the
oil in the leaking tank, it must be done  as soon as possible after the
collision. This will require shipboard dispersing and mixing equipment
and a supply of gelling chemicals on board.  In this type failure, it  is
likely that only one or two tanks will be involved.  Once these are
gelled,  no further tanks  may need to be  protected.
                              Stranding s

While the frequency of strandings is considerably less than that of
collisions,  the percentage and severity of the oil spills is  greater. The
Liverpool Underwriters Association reported 91 stranding incidents
occurred between June 1964 and April 1967, resulting  in 17  cases of
oil spillage. The oil spills resulting from strandings,  like those from
collisions,  may be major spills or minor ones. The stranding-caused
spills are very likely to increase with time stranded due to the
instability of the ship partially out of water and at the mercy of wind
and waves. The Torrey Canyon disaster, for instance,  started with
six of 18 tanks torn open on impact and the remainder of the tanks
eventually broken by heavy seas.  In another incident,  the tanker,
Arrow, ran aground.on a Wednesday,  northeast of Halifax,  spilling some
Bunker C grade oil. The leakage continued until it split in two the
following Sunday, snapping as it balanced on the small reef.  The
stopping of the oil flow  in this type accident will be a two-part
operation.  First, immediate gelling of the oil in the leaking tanks  must
be accomplished. The immediate response requirement will place a high
priority on a shipboard-carried system. Gelling of the unruptured tanks
of oil will prevent their  loss in the event of further breakup. This may
be done in a non-emergency made using additional chemicals supplied
by barge, boat or helicopter.
                                63

-------
                          Structural Failures

The occurrence of Group I  structural failures in United States ships
since World War II is well documented.  A Group I fracture is one that
is at least 10 feet long and has weakened the main hull structure
sufficiently to either sink  the ship or place it in a dangerous condition
until repaired. During the  period of 1943 through 1963, there were 116
recorded Group I casualties of United States tankers, mostly World
War II built tankers.  Improper loading,  poor structural details and
notch sensitive  steel were principal causes of failure. While this is
a low probability type failure, there is clearly a potential source of
oil spills.  Often there is a total fracture of the hull, separating the
power  source and pumping  equipment from part of the cargo. Once the
casualty has occurred,  there  is little chance that other tanks will begin
leaking oil.

Here again, a quick gelling operation must be done on the leaking tanks,
along with any in jeopardy, with shipboard equipment and chemicals.
                          Explosion and Fire

Fire is probably the most common disaster which may confront a
petroleum tanker. If a tank is ruptured as  a result of the fire, the
petroleum cargo probably will burn,  thus preventing widespread water
pollution. The gelling of the adjacent tanks, or any tank in jeopardy,
with shipboard equipment and chemicals would result in a petroleum
product with a reduced vapor pressure, thus reducing its flammability.


                              Summary

The holes in ships which will require the gelling technique to stop the
outflow of oil are as varied as the types of accidents which may befall
the ship. In the event of a collision, the hole will be expected to occur
through the waterline, be fairly large, and may not get larger with the
passage of time.

In the event of a stranding or grounding, the hole will be expected to
occur below the waterline, principally on the bottom of the ship.  The
character and size of the hole will be dependent  on the obstruction
struck.  Also, the hole will likely get worse with the passage of time,
due to the instability of  the ship  and the unusual stresses produced.
The differential  pressures between the oil  inside the tank and seawater
outside at the rupture will not be great, with equilibrium being reached
in short order,  except when significant tidal action occurs.

The holes formed by  structural failures are likely  to be large and  quickly
formed.  The crack will usually stop  at some crack-arresting discontinuity,
and not propogate unless additional  stress is placed on the area.


                                64

-------
Summarizing,  the holes which will require plugging are likely to be
large;  i.e., several square feet,  above,  at or below the waterline
and quickly formed.

In order to stop flow through such openings, both from an external tank
rupture into the water or via  an external tank from a subsequently
ruptured  interior tank,  the gelled oil must have the ability to bridge the
gap:

                   over a potential maximum rupture
                   dimension  of several  feet,

                   under a variety of possible  pressure
                   differentials,  and

                   without losing its  integrity  by itself
                   rupturing and flowing.

These  requirements indicate the need for  a gelled oil with high
elasticity, shear, and tensile  strength in bending.

While  the potential pressure conditions under which the gel will be
placed in practice can be readily reproduced in laboratory-scale tests,
the physical dimensions of the rupture, or gap over which it must
operate,  cannot be readily reproduced. These dimensions can only be
achieved in large-scale tests. The numbers and character of the variables
involved make a theoretical  approach impossible. Prior work in these
phenomena were determined  in the Basic  Contract to be insufficient
and inappropriate.  Further work in this area was beyond the scope
and intent of this program.
                                65

-------
    APPENDIX B



PENETROMETER USE

-------
                           APPENDIX B

                       PENETROMETER USE
The test procedure given by ASTM-217 was followed to measure gel
strength by use of a penetrometer. A summary of this method follows:

             1.   The unworked gel (or grease) was adjusted
                 to 77F and placed in position for test.

             2.   The tip of the cone was brought to the top
                 (level surface) of the gel.

             3.   The cone was released and allowed to fall
                 or reach  equilibrium  for five seconds.

             4.   The cone travel was  measured in tenths
                 of millimeters read direct from the dial.

Two cones were used during testing, a heavier brass  cone (102. 5
grams) and a lighter aluminum cone (35.0 grams). The holding movable
attachment to the penetrometer  weighed 47. 5 grams. Thus,  the total
weight exacted by each cone was:

             Brass              150. 0 grams
             Aluminum            82.5 grams

The cone used in each test is indicated by  (Br) or (Al). All measurements
are reported as tenths of millimeters of cone travel, with notation given
as to which cone was used.

Cone travel may be converted to various pressure units due to the fact
that the wetted surface area of  the cone is  constantly  increasing during
its fall; however, the general formula for the calculation of pressure
is:

             Pressure =  "Weight cone -  buoyancy
                        Total wetted surface area

The buoyancy must be determined from volume  displacement multiplied
by the specific gravity of the medium. The  volume of a cone is given
by:

             V = TT/ 3 r2 h

and the  surface area by:
             A = TT r ~\/ r2 + h2


                                67

-------
For any cone drop greater than 15 mm (150 mm/10) the general formula
for pressure may be resolved to:
Dynes
              wt cone in gms-sp gr (1. 047)(H- 1. 081)3+. 1987)  980.7
                         (4. 4429)(H- 1. 1081) 2 +  1. 2714(
                                                          I*
Grams /    _  wt cone in gms-sp gr (1. 047)(H- 1. 081)3 + . 1937
       in2   ~             (4. 4429)(H-1.081)2 + 1. 2714/6.452

where H is  cone travel in centimeters (l centimeter = 100 mm/10).
For conversion 69, 109 dynes/cm2 = 453.6 gm/in2  = 1 pound/in2.

When using the Bingham Body and pseudoplastic type gels, a
penetrometer cone drop greater than about 275 mm/10 (Br) is equivalent
to a gel that will  slump when erected in a column 3 inches high and
31/2 inches in diameter. A gel reading greater than 325 mm/10 will
pour from a beaker.  Due to the wide variety of physical properties of
gels,  these numbers can only be  considered as a rule of thumb.
                              68

-------
 CHART      15:54   D2 THU 05/22/69
  WHAT  IS  WEIGHT 0F C0NE IN GRAMS
 ?  150.
 WHAT IS SPECIFIC GRAVITY 0F MEASURED MEDIUM          BRASS
 ?  .892
 CHART  0F 10TH MM CONE DROP VS DYNES/SQ.CM.  DIVIDED  BY 100
        0123456789
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
717.7
596.3
494.6
411.9
345.3
291.8
248.7
213.6
185.0
161.3
141*6
125.0
11 1.0
99.1
88.8
79.9
72.1
65.3
59-3
54.0
49.3
45.0
41.2
37.8
34.7
31.8
29.2
26.9
24.7
22.7
704.8
585.2
485.5
404.5
339.4
287.1
244.8
210.5
182.4
159-2
139.8
123-5
109.7
98.0
87.8
79.0
71.4
64-7
58.8
53.5
48.8
44.6
40.9
37-5
34.4
31.6
29.0
26.6
24-5
22.5
692.0
574.3
476.6
397.3
333.6
282.4
241.1
207.5
179.9
157.1
138.1
122.1
108.5
96.9
86-9
78.2
70.7
64.0
58.2
53-0
48.4
44.2
40-5
37.1
34.1
31.3
28.7
26-4
24-3
22.3
679.3
563.7
467.9
390.3
328.0
277.9
237.4
204.4
177.4
155.0
136.3
120.6
107.2
95.8
86.0
77.4
70.0
63.4
57.6
52.5
47.9
43.8
40.1
36.8
33-8
31-0
28.5
26.2
24.1
22.1
666.
553.
459.
383.
322.
273.
233.
201.
175.
153.
134.
119.
106.
94.
85.
76.
69.
62.
57.
52.
47.
43.
39.
36.
33.
30-
28.
26.
23.
21.
9
2
4
5
5
4
8
5
0
0
6
2
0
8
1
6
3
8
1
0
5
5
8
5
5
8
3
0
9
9
654.6
543.0
451.0
376*7
317.1
269.1
230.3
198.6
172.6
151.0
133.0
117.8
104.8
93.7
84.2
75.9
68.6
62.2
56.6
51.6
47-1
43.1
39.5
36.2
33.2
30.5
28.0
25.7
23-7
21.7
642.6
532.9
442 .8
370.2
311.8
264.8
226.8
195.8
170.3
149.1
131.3
116.4
103.6
92.7
83*3
75.1
67.9
61.6
56.0
51.1
46.7
42.7
39.1
35.9
32.9
30.2
27.8
25.5
23.5
21.5
630.7
523.0
434.8
363.7
306-6
260.7
223.4
193-0
168.0
147.2
129.7
115.0
102.5
91-7
82.4
74.3
67.3
61.0
55.5
50.6
46.2
42.3
38.8
35.6
32.6
30.0
27.6
25.3
23.3
21-3
619.0
513.4
427.0
357.5
301.6
256.6
220.1
190.3
165.7
145.3
128.1
113.7
101.3
90.7
81.6
73.6
66.6
60.5
55.0
50.2
45.8
41.9
38.4
35.3
32.4
29.7
27.3
25.1
23.1
21.2
607.6
503.9
419.4
351.3
296.6
252.6
216.8
187.6
163.5
143.4
126.6
112.3
100.2
89.7
80.7
72.8
65.9
59.9
54.5
49.7
45.4
41.6
38.1
35.0
32.1
29.5
27.1
24.9
22.9
21.0
EXAMPLES  246 MM/10 C0NE TRAVEL INDICATES  14*910  DYNES/SQ  CM



                              69

-------
CHART     14:36   D2 FRI 05/23/69
 WHAT IS WEIGHT 0F C0NE  IN  GRAMS
? 82.5
WHAT IS SPECIFIC GRAVITY 0F MEASURED MEDIUM       ALUMINUM
? .892
CHART 0F 10TH MM C0NE DR0P VS  DYNES/SQ.CM.  DIVIDED BY 100
      012345678

15 395.0 387.9 380.8 373-9 367.0  360.3 353.6 347.1 3406 334.3
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
328.1
272.0
226.3
189.6
160.0
136.2
1 16.8
100.9
87.8
76.8
67.6
59.8
53.1
47.3
42.3
38.0
34.1
30.7
27.7
25.0
22.6
20.4
18.4
16.6
14.9
13.4
12.0
10.7
9.5
322.0
267.0
222.3
186.3
157.4
134.1
1 15.1
99.5
86.6
75.8
66.8
59.1
52.5
46.8
41 .9
37.5
33.8
30.4
27.4
24.7
22.3
20.2
18.2
16.4
14.7
13.2
1 1 .8
10.6
9.4
316.0
262.1
218.3
183.1
154.8
132.0
113.4
98.1
85.4
74.8
65.9
58.4
51 .9
46.3
41.4
37.1
33.4
30. 1
27.1
24.5
22.1
20.0
18.0
16.2
14.6
13.1
1 1*7
10.4
9.3
310.1
257.3
214.5
180.0
152.3
129.9
111.7
96.7
84.3
73.9
65.1
57.7
51 .3
45.8
41.0
36.8
33.0
29.8
26.8
24.2
21.9
19.7
17.8
16.1
14.4
13.0
1 1.6
10.3
9.1
304.3
252.6
210.7
177.0
149.9
127.9
1 10.1
95.4
83.2
72.9
64.3
57.0
50.7
45.2
40.5
36.4
32.7
29.5
26.6
24.0
21 .7
19.5
17.6
15.9
14.3
12.8
1 1.5
10.2
9.0
298.7
248.0
207.0
174.0
147.5
126.0
108.5
94.0
82.1
72.0
63.5
56.3
50-1
44.7
40.1
36.0
32.4
29.2
26.3
23.7
21.4
19.3
17.4
15.7
14.1
12.7
11 .3
10.1
8.9
293.1
243.4
203.3
171.1
145.1
124.1
106.9
92.7
81 .0
71.1
62.8
55.6
49.5
44.2
39-6
35.6
32.0
28.9
26.0
23.5
21 .2
19.1
17.3
15.6
14.0
12.5
11.2
10.0
8.8
287-7
239-0
199.8
168.2
142.8
122.2
105.3
91 .5
79.9
70.2
62.0
55.0
49-0
43.8
39.2
35.2
31 ;7
28.6
25.8
23.3
21.0
19.0
17.1
15.4
13.8
12.4
11.1
9.8
8.7
282.3
234.7
196.3
165.4
140.5
120.4
103.8
90.2
78.9
69.3
61.2
54.3
48.4
43.3
38.8
34.8
31 .4
28.3
25.5
23.0
20.8
18.8
16.9
15.2
13.7
12.3
10.9
9.7
8.6
277
230
192
162
138
1 18
102
89
77
68
60
53
47
42
38
34
31
28
25
22
20
18
16
15
13
12
10
9
8
.1
.5
.9
.7
.3
.6
.4
.0
.8
.5
.5
.7
.9
.8
.4
.5
.0
.0
.2
.8
.6
.6
.7
.1
 5
.1
.8
.6
.5
EXAMPLE:   492  MM/10 C0NE TRAVEL INDICATES 440 DYNES/SG CM
                              70

-------
       APPENDIX C



CHEMICALS AND SUPPLIERS

-------
                            APPENDIX C

                     CHEMICALS AND  SUPPLIERS
N-3-APCHA
Alamine 3
Alamine 4D
Alamine 15D
Actinol D29LR
Actinol F.A. 2
Actinol F.A. 3
Armeen C
Armeen O
Armeen L-9
Armeen L-7
Armeen L-ll
Armeen L-15
2-(2-Aminothalamine)
Diam 26
DMAPA
Duomeen C
3,3-DADPA
Dicyclohexylamine
DETA
EDA
1,6-Hexanediamine
N-Hexylamine
Neofat 140
Pamak 1
Pamak 4
Pamak 6
Pamak 10
Pamak 546A
Pamak S1418
Pamak DLE
P & G L-3IO
P & G S-210
POPDA* 190
POP DA* 230
POPDA* 400
POPDA* 1000
POPDA* 2000
POPTA* 403
     Type

Primary amine
Primary amine
Primary amine
Primary amine
Primary amine
Primary amine
Primary amine
Primary amine
Primary amine
Primary amine
Primary amine
Primary amine
Primary amine
Diamine
Diamine
Diamine
Diamine
Primary amine
Cyclic amine
Triamine
Diamine
Diamine
Cyclic amine
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Diamine
Diamine
Diamine
Diamine
Diamine
Triamine
       Supplier

Abbott Chemicals
General Mills
General Mills
General Mills
Arizona Chemical
Arizona Chemical
Arizona Chemical
Armour Chemical
Armour Chemical
Armour Chemical
Armour Chemical
Armour Chemical
Armour Chemical
Armour Chemical
General Mills
Generic
Armour Chemical
Matheson & Coleman
Matheson & Coleman
Matheson & Coleman
Eastman Organic Chem.
Eastman Organic Chem.
Eastman Organic Chem.
Armour Chemical
Hercules, Inc.
Hercules, Inc.
Hercules, Inc.
Hercules, Inc.
Hercules, Inc.
Hercules, Inc.
Hercules, Inc.
Procter & Gamble
Procter & Gamble
Jefferson Chemical Co.
Jefferson Chemical Co.
Jefferson Chemical Co.
Jefferson Chemical Co.
Jefferson Chemical Co.
Jefferson Chemical Co.
-H-
 POPDA and POPTA are now Jeffamine
                               72

-------
      Name
     Type
      Supplier
Emersol 310
Emersol 305
Emersol 233
Emery 3162
Emersol 315
RD-3826
S-1598
S-1597
TC 1678
TC 1857
Versadyme  216
G-17
HMD

Isonate 136T
MRS
MT-40
MDI
NCO-10
TDI 65/35

TDI 80/20

TMD-1
Tonco 90
T.P.M.T.

TP 2540
TP 4040
Mondur O
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Fatty acid
Hexylmethyl-
  diisocyanate
Bitolylene diisocyanate
Polyisocyanate
Organic isocyanate
Methane diisocyanate
Polyisocyanate
Toluene diisocyanate
  65/35
Toluene diisocyanate
  80/20
Diisocyanate
Octadecyl isocyanate
Triphenyl methane
  triisocyanate
Emery Industries, Inc.
Emery Industries, Inc.
Emery Industries, Inc .
Emery Industries, Inc.
Emery Industries , Inc .
Armour Chemical
Hercules, Inc.
Hercules, Inc.
Hercules, Inc.
Hercules, Inc.
General Mills
General Mills

Mobay Chemical
Upj ohn
Mobay Chemical
Mobay Chemical
Mobay Chemical
Kaiser
DuPont Chemical

DuPont Chemical

Thorson Chemical Corp,
Upj ohn
Octadecyl isocyanate   Mobay Chemical
                                73

-------
      APPENDIX D




SOAP GEL SCREENING DATA

-------
TABLE D-I. SOAP GEL SCREENING DATA
Components

-H-
Actinol D29LR
NaOH
Actinol F. A. 2

NaOH
Actinol F. A. 3

NaOH
Emersol 305 -H-

NaOH
Emersol 310
NaOH

Emersol 315

NaOH
Gelling Components
(% by volume)
Run 1 Run 2 Run 3

10
5
10

5
10

5
10

5
10
5

10

5

5
2
5

2
5

2
5

2
5
2

5

2

2
1
2

1
Z

1
Z

1
Z
1

Z

1
Remarks (i. e. , viscosity,
penetrometer reading, etc. )


Shear-sensitive gel.

Weak gel formed in Run 1
with viscosity increase to
186 cps.
Weak and thin gel formed
having a viscosity of 130 cps
(Run 1)
Run 1 formed a thin gel
of 148 cps. Others were not
successful.
First run yielded a weak
dilatent gel. Viscosity
increased to 200 cps. Other
runs not important.
Slight viscosity increase in
first run, no gels were
formed.
-H- Two component systems comprised of NaOH and a proprietary fatty acid.

-------
Components
Emersol 233 -tf
NaOH
Emery 3162
NaOH
G-17
NaOH
Neofat 140
NaOH
Pamak 1
NaOH
Pamak 4 -H- -H-
NaOH
Gelling Components
(% by volume)
Run 1 Run 2 Run3
10
5
10
5
10
5
10
5
10
5
10
5
5
2
5
2
5
2
5
2
5
2
5
2
2
1
2
1
2
1
2
1
2
1
2
1
Remarks (i. e. , viscosity,
penetrometer reading, etc. )
Weak thixotropic gel formed
in first run. Others unimpor-
tant.
Waxlike precipitate formed
upon standing.
Forms a heavy precipitate -
No gels were formed. Oil
remained thin. A thin brown
crust formed on top of the oil
upon standing.
Viscosity increases but no
gels were formed.
Thin gel in first run with 200
cps viscosity. Others not
important.
-H- Two component systems comprised of NaOH and a proprietary fatty acid.
-H- -H- Gellation systems showed similar results using CR-280, an ethoxylated dehydroabietylamine as
an additive to enhance gellation. Fatty acid was mixed with CR-280 in the ratio of 80 ml fatty
acid to 20 ml CR-280.

-------
Components
Pamak 6 -H-
NaOH
Pamak 10
NaOH
Pamak 546A
NaOH
Pamak S1418
NaOH
Pamak DIE -tt- +
NaOH
P & G L-310
NaOH
Galling Components
(% by volume)
Run 1 Run 2 Run 3
10
5
10
5
10
5
10
5
10
5
10
5
b
2
5
2
5
2
5
2
5
2
5
2
2
1
2
1
2
1
2
1
2
1
2
1
Remarks (i. e. , viscosity,
penetrometer reading, etc. )
Thin gel formed in first run
with 182 cps viscosity. No
other runs successful.
In first run a moderately
thick gel was formed. Weak-
er gel formed in Run 2 and
no crel in Run 3.
Weak, dilatent gels formed
in first run, none in Run 3.
Very thin gel formed in first
run. Other runs were un-
successful.
Medium thin gel with some
viscosity increase.
Thin gel with some viscosity
increase.
-fl- Gellation systems showed similar results using CR-280, an ethoxylated dehydroabietylamine as
an additive to enhance gellation. Fatty acid was mixed with CR-280 in the ratio of 80 ml fatty
acid to 20 ml CR-280.

-------
oo
Components
P & G S-210 -tt-
NaOH
S-1597
NaOH
S-1598 -H--H-
NaOH
TC 1678 -W--H-
NaOH
TC 1857
NaOH
Versadyme 216
NaOH
Gelling Components
(% by volume)
Run 1 Run 2 Run 3
10
5
10
5
10
5
10
5
10
5
10
5
5
2
5
2
5
2
5
2
5
2
5
2
2
1
2
1
2
1
2
1
2
1
2
1
Remarks (i. e. , viscosity,
penetrometer reading, etc. )
Thin gel was formed with
some viscosity increase.
Thin gel was formed with
some viscosity increase.
Thin gel was formed with
some viscosity increase.
Thin gel was formed with
some viscosity increase.
Thin gel was formed with
some viscosity increase.
Precipitate formed, no gel.
-tt- Gellation systems showed similar results using CR-280, an ethoxylated dehydroabietylamine as
an additive to enhance gellation. Fatty acid was mixed with CR-280 in the ratio of 80 ml fatty
acid to 20 ml CR-280.

-------
                              TABLE D-II.  SOAP GEL SCREENING DATA


Components
Actinol F. A. 2

NaOH
Actinol F. A. 3

NaOH
Emersol 310

NaOH
Emery 3162

NaOH
Neofat 140

NaOH
Pamak 1

NaOH
Gelling Components
(% by volume)

Run 1 Run 2 Run 3
10

5
10

5
10

5
10

5
10

5
10

5
5

2
5

2
5

2
5

2
5

2
5

2
2

1
2

1
2

1
2

1
2

1
2

1

Remarks (i. e. , viscosity,
penetrometer reading, etc. )
Gel formed in Run 1.
Other runs unsuccessful.

Thin gel in all except Run
3.

Waxlike precipitate formed
upon standing.

Thin gel with high viscosity
in first run, others unsuc-
cessful.
Thin gel with high viscosity
in first run, other unsuc-
cessful.
Weak gels formed in Runs 1
and 2. Noticeable viscosity
increase.
sO

-------
oo
o
Components
Pamak 10

NaOH
Pamak S1418

NaOH

Pamak 546A

NaOH
P & G L-310
NaOH
S-1597

NaOH
Gelling Components
(% by volume)
Run 1 Run 2 Run 3
10

5
10 ~

5

10

5
10
5
10

5
5

2
5

2

5

2
5
2
5

2
2

1
2

1

2

1
2
1
2

1
Remarks (i. e. , viscosity,
penetrometer reading, etc. )
No gel formation. Only
slight viscosity increase.

First run yielded a thick
greaselike gel. Second run
was a thin gel with slight
viscosity increase.
First run yielded a thick
greaselike gel. Second run
was a thin gel with slight
viscosity increase.
First run yielded a thick
greaselike gel. Second run
was a thin gel with slight
viscosity increase.
First run showed thick vis co-
elastic gel. Other runs were
unsuccessful.

-------
         APPENDIX E



SWELLING GEL SCREENING DATA

-------
                           TABLE E-I.  SWELLING GEL SCREENING DATA
Gelling Agent
Aluminum Octoate
Special Lithium Isostearate
Alumagel
Aluminum Stearate
Zinc Soap
Calcium Stearate
Percentage
> 20
> 20
> 20
>20
>20
r>20
Remarks
No gel
No gel
Weak gel
No gel
No gel
No gel
00
IV

-------
             APPENDIX F



ISOCYANATE/AMINE GEL SCREENING DATA

-------
TABLE F-I.  AMINE TONCO 90 GEL SCREENING DATA
Components
Alamine 3
Tonco 90
Alamine 4D
Tonco 90
Alamine 15D
Tonco 90
Armeen L-7
Tonco 90
Armeen L-9
Tonco 90
Armeen L-9
Tonco 90
Armeen L-l 1
Tonco 90
Armeen L-l 5
Tonco 90
Armeen O
Tonco 90
Gelling
Components
(% BY Weight)
1.50
3. 50
1.90
3. 10
2.40
2.60
1.40
3.60
3.02
1.98
1.65
3.40
1.80
3. 20
2. 20
2.80
2.35
2.60
Remarks (i.e. .viscosity,
penetrometer reading, etc.^
No gel, slight viscosity
increase.
No gel, slight viscosity
increase.
No gel, slight viscosity
increase.
Shear- sensitive, thin gel.
Shear- sensitive, thin gel.
Thick & firm gel formed.
No gel, no viscosity
increase.
No gel, no viscosity
increase.
Thin, shear-sensitive gel.
                    84

-------
  Gelling
Components
Remarks (i. e. viscosity,
Armeen C
Tonco 90
Diam 26
Tonco 90
DMAPA
Tonco 90
Dicyclo hexylamine
Tonco 90
Duomeen C
Tonco 90
N- Hexylamine
Tonco 90
POPDA 190
Tonco 90
POPDA 190
Tonco 90
POPDA 230
Tonco 90
POPDA 1000
Tonco 90
1.95
3. 10
1.80
3. 20
1.30
3. 70
1.90
3. 10
1.40
3.60
1. 30
3.70
1.20
3. 70
0.60
4.40
1.40
3.60
3. 15
1.85
Thin, shear- sensitive gel.
Thin gel formed.
Firm, homogeneous gel
formed.
No gel and no viscosity
increase.
No gel, slight viscosity
increase.
Shear-sensitive, thin gel.
Shear- sensitive, thin gel.
No gel, increase in
viscosity.
No gel, increase in
viscosity.
Shear- sensitive, thin gel.
         85

-------
                        Gelling
                      Components      Remarks  (i. e. , viscosity,
  Components	(% By Weight)	penetrometer reading, etc.)
   POPDA ZOOO              3.90         No gel, no viscosity
   Tonco 90                1.15         increase.
   POPTA 403               1.55         No gel, slight viscosity
   Tonco 90                3.40         increase.
Two component systems a proprietary amine and a proprietary
isocyanate.
                              86

-------
TABLE F-II.  AMINE HMD GEL SCREENING DATA
Components
Alamine 3
HMD
Alamine 4D
HMD
Alamine 4D
HMD
Alamine 15D
HMD
Alamine 15D
HMD
Alamine 15D
HMD
Diam 26
HMD
DMAPA
HMD
Armeen L-l 1
HMD
Gelling
Components
(% By Weight)
3.00
2.00
3.40
1.60
4. 10
0. 90
3.80
1. 20
4.60
0.40
3.60
1.40
3.30
1. 70
2.60
2.40
4. 10
0.90
Remarks (i. e. , viscosity,
penetrometer reading, etc. )
Spontaneous formation of
good gel.
Spontaneous formation of
good gel.
No gel, increase in
viscosity.
Spontaneous formation of
good gel.
Thin gel with viscosity
increase.
A shear-sensitive precipi-
tate gel formed.
Formation of a grainy preci-
pitate gel. Shear sensitive.
Formation of a shear-
sensitive, weak gel.
Firm, thixotropic gel.
                   87

-------
Components
    Gelling
 Components
(% By Weight)
Remarks  (i.e., viscosity,
penetrometer reading, etc.)
 Duomeen
 HMD
       3.00
       2.00
Formation of shear-
sensitive, firm gel.
 Ethylenediamine
 HMD
       1. 30
       3. 70
No gel, increase in
viscosity.
  1, 6-Hexanediamine      2. 00
  HMD                   3.00
                   No gel,  increase in
                   viscosity.
 Armeen L-11
 HMD
       3.40
       1. 60
Formation of a strong,
firm gel.
 Armeen L-l 1
 HMD
       3. 10
       1.90
No gel,  increase in
viscosity.
 Armeen L-l5
 HMD
       4.40
       0.60
Good gel formation.
Thick and homogeneous.
 Armeen L-l5
 HMD
       3. 10
       1.90
No gel,  noticeable
increase in viscosity.
 Armeen L-l5
 HMD
       3. 70
       1.30
Formation of strong, firm
gel.
 N-Hexylamine
 HMD
       2. 70
       2. 30
Spontaneous formation of
tough gel.
 N- Hexylamine
 HMD
       1. 10
       0.90
Formation of non shear-
sensitive gel.  (2%)
                              88

-------
                       Gelling
                      Components       Remarks  (i. e. , viscosity,
  Components	(% By Weight)     penetrometer reading,  etc. )
   N-Hexylamine           3.20        Formed good,thick gel.
   HMD                    1.80
   N-Hexylamine           2.20        Good,firm gel.  Highly
   HMD                    2.80        thucotropic.
   RD 3826                 3.50        Formed good,firm gel.
   HMD                    1.50        Not shear sensitive.
Two component systems a proprietary amine and a proprietary
isocyanate.
                              89

-------
        TABLE F-III.  AMINE MT-40 GEL SCREENING DATA
Components
 Alamine 4D
 MT-40
   Gelling
Components
(% By Weight)
     3. 10
     1.85
Remarks  (i.e., viscosity,
penetrometer reading, etc. )
Thin gel, shear sensitive.
 Alamine 15D
 MT-40
     3.60
     1.45
Thin gel, shear sensitive.
 Armeen C
 MT-40
     3. 20
     1.80
Firm and thixotropic gel.
 Armeen O
 MT-40
     3.60
     1.40
No gel.  Thin liquid.
 3,3-DADPA
 MT-40
     1.55
     3.45
Thin gel. Shear  sensitive.
 Diam  26
 MT-40
     3. 10
     1.95
Thin gel. Shear  sensitive.
 Dicyclohexylamine
 MT-40
     3.00
     2.00
No  gel.  Thin liquid.
 DMAPA
 MT-40
     2.45
     2.50
No gel.  Thin liquid.
 Duomeen C
 MT-40
     3.40
     1.60
Thick,  greaselike  gel.
Shear sensitive.
                            90

-------
Components
EDA
MT-40
1, 6-Hexylamine
MT-40
L-7
MT-40
L-9
MT-40
L-ll
MT-40
L-15
MT-40
N-3-APCHA
MT-40
N-Hexylamine
MT-40
POPDA 190
MT-40
POPDA 230
MT-40
Gelling
Components
(% By Weight)
3. 70
1.30
1.80
3. ZO
2.60
2.40
2.90
2. 10
3. 10
1.90
3.40
1. 60
3. 20
1.85
2.45
2. 55
2.40
2.65
2. 30
2. 65
Remarks (i.e., viscosity,
penetrometer reading, etc. )
Thin liquid, no viscosity
increase.
No gel. Slight viscosity
increase.
Thin, shear- sensitive gel.
Viscosity increase.
Thin, shear sensitive gel.
Viscosity increase.
Thin, shear- sensitive gel.
Viscosity increase.
Thin liquid only slight
viscosity increase.
Very thin liquid. Granular
precipitate was formed.
Shear sensitive, thin gel
formed.
Shear-sensitive, thin gel
formed.
No gel. Some viscosity
increase.
91

-------
                        Gelling
                      Components       Remarks  (i .e. .viscosity,
  Components	(%_ By Weight)	penetrometer reading, etc.)

   POPDA 400               3.00        Large mass of precipitate
   MT-40                  2.00        forms.
   POPDA 1000             3.95        Large mass of precipitate
   MT-40                  1.05        forms.
   POPDA 2000             4.36        Large mass of precipitate
   MT-40                  0.60        forms.
   POPTA 403               2.50        Globular percipitate forms.
   MT-40                   2.50        Increase in viscosity.
Two component systems a proprietary amine and a proprietary
isocyanate.
                              92

-------
         TABLE F-IV.  AMINE MRS GEL SCREENING DATA
Components
   Gelling
 Components
(% By Weight)
Remarks  (i. e. .viscosity,
penetrometer reading,  etc. )
 Alamine 4D
 MRS
     2.90
     2. 10
No gel,  some viscosity
increase.
 Alamine 15D
 MRS
     3.30
     1.65
Very thin gel.  Noticeable
increase in viscosity.
 Armeen C
 MRS
     2.90
     2. 10
Shear-sensitive,weak gel,
100  cps.
 Armeen O
 MRS
     3.35
     1.65
Shear-sensitive  weak gel.
Viscosity increase to 200
cps.
 Diam  26
 MRS
     2.80
     2. 20
Grainy precipitate settled
upon standing.
 Dicyclohexylamine
 MRS
     2. 70
     2. 30
No gel, slight viscosity
increase.
 DMAPA
 MRS
     2. 20
     2.80
Insoluable mass  settles  out
as precipitate.
 3,3-DADPA
 MRS
     3.90
     1. 10
No gel, no viscosity
increase.
 Duomeen C
 MRS
     3. 15
     1.85
No gel,  some increase in
viscosity.
                             93

-------
Components
1 , 6-Hexanediamine
MRS
L-7
MRS
L-9
MRS
L-ll
MRS
L-15
MRS
POPDA 190
MRS
POPDA 230
MRS
POPDA 400
MRS
POPDA 1000
MRS
POPDA 2000
MRS
Gelling
Components
(% By Weight)
1.50
3.45
2.30
2. 70
2.60
2.40
2.80
2. 20
3. 15
1.85
2. 10
2.90
2.30
2.65
3.00
2.00
3.95
1.05
4. 35
0.60
Remarks (i. e. , viscosity,
penetrometer reading, etc. )
Shear-sensitive, grainy gel.
Some viscosity increase.
Increase in viscosity but no
gel.
Increase in viscosity but
no gel.
Increase in viscosity but
no gel.
Increase in viscosity but
no gel.
Forms granular precipitate
upon standing.
No gel. Slight increase in
viscosity.
Forms large mass of spongy
precipitate.
Forms large mass of spongy
precipitate.
Forms large mass of spongy
precipitate.
94

-------
                        Gelling
                      Components       Remarks  (i.e., viscosity,
 Components         (% By Weight)     penetrometer reading, etc.)
   POPTA403               2.50        Globular precipitate  formed
   MRS                     2.50
   N-Hexylamine           2. 15        Only slight increase in
   MRS                     2.85        viscosity.
Two component systems a proprietary amine and a proprietary
isocyanate.
                               95

-------
TABLE F-V. AMINE MDI GEL SCREENING DATA
Components
Armeen C
MDI
Armeen C
MDI
L-7
MDI
L-7
MDI
Armeen O
MDI
Armeen O
MDI
L-9
MDI
L-9
MDI
Two component
isocyanate.
Gelling
Components
(% By Weight)
2. 20
2.80
1.64
3.36
1.50
3.50
1. 22
3. 78
2.65
2. 30
2. 24
2. 76
1.80
3. 25
1. 10
3.90
systems a proprietary
Remarks (i. e. , viscosity
penetrometer reading, etc. )
Thin, shear-sensitive gel.
Thin, shear- sensitive gel.
Thin, shear- sensitive gel.
Large amount of precipitate
settled out.
Large amount of precipitate
settled out.
Large amount of precipitate
settled out.
Large amount of precipitate
settled out.
Waxlike precipitate formed
upon standing.
amine and a proprietary
                  96

-------
TABLE F-VI.  AMINE ISONATE 136-T GEL SCREENING DATA
Gelling
Components Remarks (i.e., viscosity,
Components (% By Weight) penetrometer reading, etc.)
Armeen C
Isonate 136-T
Armeen C
Isonate 136-T
L-7
Isonate 136-T
L-7
Isonate 136-T
L-9
Isonate 136-T
L-9
Isonate 136-T
Armeen O
Isonate 136-T
Armeen O
Isonate 136-T
Two component systems
isocyanate.
2.05 Thin, shear-sensitive gel.
2.95
1.46 Thin, shear- sensitive gel.
3. 54
1.50 Thin, shear- sensitive gel.
3.5Q
0.80 No gel, slight viscosity
4. 20 increase.
1.80 Thin grainy gel. Shear
3. 20 sensitive.
1.16 Weak, shear-sensitive gel.
3.84
2.50 Thin gel. Shear sensitive.
2. 50
2. 0 0 Thin gel. Shear sensitive.
3.00
a proprietary amine and a proprietary
                        97

-------
          TABLE F-VII.  AMINE TDI 65/35 GEL SCREENING DATA
  Components
  Gelling
Components
(% By Weight)
Remarks  (i.e., viscosity,
penetrometer reading, etc. )
   Duomeen C
   TDI65/35
     3.00
     2.00
Spontaneous formation of
shear-sensitive gel.
   Duomeen C
   TDI 65/35
     1. 60
     3. 40
Spontaneous formation of
non-globular type of slurry.
   EDA
   TDI 65/35
     1. 30
     3. 70
No gel,  only viscosity
increase.
   EDA
   TDI 65/35
    0.50
    4.50
No gel, only viscosity
increase.
   1, 6- Hexanediamine      2.00
   TDI 65/35               3.00
                 No gel,  only viscosity
                 increase.
   1,6-Hexanediamine      1.00
   TDI 65/35               4.00
                 No gel,  only viscosity
                 increase.
Two component systems a proprietary amine and a proprietary
isocyanate.
                              98

-------
TABLE F-VIII.  AMINE NCO- 10 GEL SCREENING DATA
Components
Alamine 3
NCO-10
Alamine 4D
NCO-10
Alamine 15D
NCO-10
Armeen C
NCO-10
Armeen C
NCO-10
Armeen C
NCO-10
Armeen O
NCO-10
Armeen O
NCO-10
Armeen O
NCO-10
Gelling
Components.
(% By Weight)
2. 50
2. 50
2.90
2. 1 0
3.30
1. 70
2. 90
2. 10
3.48
1.52
2.48
2. 52
1.65
3. 30
1.98
3.00
3. 30
1.65
Remarks (i.e., viscosity,
penetrometer reading, etc. )
Formation of globular
precipitate. Increase in
viscosity.
Heavy precipitate formed.
Slight viscosity increase.
Spontaneous formation of
grainy gel.
Shear-sensitive weak gel.
Thin liquid, some increase
in viscosity.
No gel, only viscosity
increase.
Grainy gel. Thick and
greaselike.
Firm, thixotropic gel
formed.
Thick and firm gel formed.
                     99

-------
Gelling
Components
Components (% By Weight)
DADPA
NCO-10
DETA
NCO-10
Diam 26
NCO-10
DMAPA
NCO-10
Duomeen C
NCO-10
Duomeen C
NCO-10
1 , 6-Hexanediamine
NCO-10
N-Hexylamine
NCO-10
L-7
NCO-10
L-9
NCO-10
1.65
3.35
1.39
3.61
2.80
2. 20
2.00
3.00
2.40
2.60
2.40
2.60
1.50
3.50
2. 20
2.80
2.65
2. 35
2.60
2.40
Remarks (i.e., viscosity,
penetrometer reading, etc. )
Thin, runny liquid.
Grainy precipitate formed.
Slow formation of non-
globular precipitate.
No gel, some viscosity
increase.
No gel, no viscosity
increase.
No gel, no viscosity
increase.
Slow formation of a weak
gel.
Formed a precipitate.
No gel only viscosity
increase.
No gel only viscosity
increase.
100

-------
Components
L-9
NCO-10
L-ll
NCO-10
L-ll
NCO-10
L-ll
NCO-10
L-ll
NCO-10
L-15
NCO-10
L-15
NCO-10
L-15
NCO-10
POPDA 190
NCO-10
POPDA 230
NCO-10
Gelling
Components v
(% By Weight)
2. 12
2. 88
2.80
2. 20
5.60
4.40
3.40
1.60
2.40
2.60
1. 90
3.10
2. 30
2.70
1.30
3. 70
2.06
2.94
2.30
2. 70
Remarks, (i.e., viscosity,
penetrometer reading, etc. )
Smooth, thin gel formed.
Weak gel formed.
No gel was formed, only
large amount of precipitate.
Shear-sensitive, weak gel.
Shear- sensitive, weak gel.
Good gel formed.
Fairly good gel.
No gel, no viscosity
increase.
Increase in viscosity, no
gel.
Increase in viscosity, no
gel.
101

-------
Components .
PC-PDA 230
NCO-10
POPDA 230
NCO-10
POPDA 400
NCO-10
POPDA 400
NCO-10
POPDA 400
NCO-10
POPDA
NCO-10
POPDA
NCO-10
POPTA 403
NCO-10
POPTA 403
NCO-10
POPTA 403
NCO-10
Gelling
Components
(% By Weight)
1.80
3. 20
2.80
2. 20
3.00
2.00
2.6
2.4
3.60
1.40
4.00
1.00
4.40
0.60
2. 50
2. 50
2.00
3.00
3.00
2.00
Remarks (i. e. , viscosity,
penetrometer reading, etc. )
Increase in viscosity, no
gel.
Large amount of precipitate.
formed.
Large amount of precipitate
formed.
Mushy gel with globular
precipitate.
Globular precipitate settled
out.
Precipitate settled out on
standing.
Precipitate settled out on
standing.
Grainy gel formed.
Shear-sensitive, grainy gel.
No gel. Some viscosity
increase.
102

-------
                        Gelling
                      Components      Remarks  (i. e. , viscosity
 Components          (% By Weight)    penetrometer reading,  etc.)
   RD-3826                 3.00        Shear-sensitive gel
   NCO-10                 2.00        formed.
   TETA                    1.77        No gel, slight viscosity
   NCO-10                 3.23        increase.
   EDA                     0-90        No gel, slight viscosity
   NCO-10                 4.10         increase.
Two component systems a proprietary amine and a proprietary
isocyanate.
                               103

-------
   TABLE F-IX.  AMINE TDI 80/20 GEL SCREENING DATA
Components
Ala mine 3
TDI 80/20
  Gelling
Components
(% By Weight)
Remarks (i.e.,  viscosity,
penetrometer reading, etc. )
    1. 70
    3.30
A weak, thixotropic gel
formed.
Alamine 3
TDI 80/20
    1.40
    3.60
No gel, only viscosity
increase.
Alamine 4D
TDI 80/20
    2. 30
    2. 70
Spontaneous formation of
good, thixotropic gel.
Alamine 4D
TDI 80/20
    2.80
    2. 20
Thin, shear-sensitive gel.
Alamine 4D
TDI 80/20
    1.70
    3. 30
Thin, shear-sensitive gel.
Alamine 15D
TDI 80/20
    2. 70
    2.30
 Shear-sensitive gel
 formed.
 2-2(Aminothalamine)     2.50
 TDI 80/20               2. 50
                No gel, very slight increase
                in viscosity.
Armeen L-7
TDI80/20
    2.80
    2. 15
Shear-sensitive gel.  Gel
thick, but not hard.
Armeen L-9
TDI 80/20
    2.80
    2. 20
Shear-sensitive, thin gel.
                            104

-------
Components
DAD PA
TDI 80/20
DETA
TDI 80/20
Diam 26
TDI 80/20
DMAPA
TDI 80 /20
Duomeen C
TDI 80 /20
Duomeen C
TDI 80/20
Duomeen C
TDI 80/20
Duomeen C
TDI 80/20
EDA
TDI 80/20
EDA
TDI 80 /20
Duomeen C
TDI 80/20
Gelling
Components
(% By Weight)
1.90
3. 10
1.63
3. 37
2. 20
2.80
2.80
2.20
2. 50
2. 50
3. 20
1. 80
2.00
3.00
1.80
3. 20
0.60
4.40
1. 50
3. 50
3.00
2.00
Remarks (i.e., viscosity,
penetrometer reading etc. )
No gel. Only slight
viscosity increase.
Precipitate formed upon
standing.
Precipitate formed upon
standing
Weak gel formed.
Spontaneous formation of
globular precipitate.
Spontaneous formation of
grainy gel.
Precipitate is formed upon
standing.
No gel. Only slight increase
in viscosity.
No gel. Only slight increase
in viscosity.
Precipitate present with only
a slight increase in viscosity
A weak, precipitate gel
formed.
105

-------
Components
    Gelling
  Components
(% By Weight)
Remarks  (i. e. ,  viscosity,
penetrometer reading, etc.)
   EDA
   TDI 80/20
     1. 30
     3. 70
Spontaneous formation of
grainy gel.
   1,6-Hexanediamine    2.00
   TDI 80/20            3.00
                  No gel.  Only slight
                  viscosity increase.
   N-Hexylamine
   TDI 80/20
     2.40
     2.60
Spontaneous formation of
shear-sensitive gel.
   N-Hexylamine
   TDI 80/20
     6.60
     5.40
Spontaneous formation of
firm gel.
   POPDA 190
   TDI 80/20
     2. 30
     2. 70
Only slight viscosity
                  increase.
   POPDA 230
   TDI 80/20
     2. 50
     2. 50
Only slight viscosity
increase.
   POPDA 400
   TDI 80/20
     3. 20
     1.80
Globular precipitate formed.
   POPDA 400
   TDI 80/20
     2. 80
     2. 20
Only slight viscosity
increase.
   POPDA 400
   TDI 80/20
     1. 20
     0. 67
Weak semi-solid gel formed.
   POPDA
   TDI 80/20
     4. 10
     0. 90
Formation of globular
precipitate..
                            106

-------
Components
POPDA 2000
TDI 80/20
POPTA 403
TDI 80/20
POPTA 403
TDI 80 /20
RD-3826
TDI 80/20
TEPA
TDI 80/20
TETA
TDI 80/20
Gelling
Components
(% By Weight)
4. 50
0. 50
2. 70
2. 25
3. 30
1. 70
2.40
2.60
2.40
2.60
2.03
2.97
Two component systems a proprietary
Lsocyanate.
Remarks (i.e., viscosity,
penetrometer reading, etc. )
Formation of globular
precipitate.
Formation of a thin gel.
Only slight increase in
viscosity.
No gel. Precipitate formed.
No gel. Precipitate formed.
No gel, only slight
viscosity increase.
amine and a proprietary
107

-------
       TABLE F-X.  AMINE T.P.M.T. GEL SCREENING DATA
   Components
    Gelling
  Components
(% By Weight)
Remarks  (i. e. ,  viscosity,
penetrometer reading, etc.)
     EDA
     T.P.M.T.
     1.00
     4.00
Spontaneous formation of a
precipitate.
      1,6- Hexanediamine   1. 60
      T.P.M.T.            3.40
                  Spontaneous formation of a
                  precipitate.
     Duomeen C
     T.P.M.T.
     2. 50
     2. 50
Spontaneous formation of a
precipitate.
     Armeen L-ll          0.70
     TP 2540 and TDI      4. 30
                  No gel and no viscosity
                  increase.
      1, 6-Hexanediamine   0.30
      TP 2540 and TDI      4. 70
                  Formed a precipitate.
     Armeen L-ll          0. 50
     TP 4040 and TDI      4. 50
                  No gel, no viscosity
                  increase.
      1, 6-Hexanediamine   0.20
      TP 4040 and TDI      4.80
                  Formed a precipitate.
Two component systems a proprietary amine and a proprietary
isocyanate.
                              108

-------
TABLE F-XI.  AMINE TMD- 1 GEL SCREENING DATA
Components
Alamine 4D
TMD-1
Alamine 15D
TMD-1
N-3-APCHA
TMD-1
Armeen C
TMD-1
Armeen O
TMD-1
Armeen L-7
TMD-1
Armeen L-9
TMD-1
Armeen L-ll
TMD-1
Armeen L- 15
TMD-1
Gelling
Components
(% Bv Weiaht)
3. 10
1.85
3. 60
1.45
3. 20
1.80
3. 20
1.80
3. 60
1. 40
2.60
2.40
2. 90
2. 10
3. 10
1. 90
3.40
1. 60
Remarks (i.e., viscosity,
penetrometer reading, etc. )
Shear-sensitive, weak gel
forms upon standing.
No gel, no viscosity
increase.
Insoluable mass settles out
upon standing.
Formation of firm gel. Shear-
sensitive to viscosity check.
No gel and no viscosity
increase.
Formation of a precipitate
upon standing.
Formation of a precipitate
upon standing.
No gel, only slight viscosity
increase.
No gel, only slight viscosity
increase.
                    109

-------
Gelling
Components
Components (% By Weight)
Dicyclohexylamine
TMD-1
DMAPA
TMD-1
Duomeen C
TMD-1
Diam 26
TMD-1
3,3-DADPA
TMD-1
POPDA 190
TMD-1
POPDA 230
TMD-1
POPDA 400
TMD-1
POPDA 1000
TMD-1
POPDA 2000
TMD-1
3.00
2.00
2.45
2. 50
3.4Q
1.6o
3. 10
1.95
1. 55
3.45
2. 40
2. 65
2. 60
2.40
3.30
1. 70
4. 10
0. 90
4. 60
0. 40
Remarks (i.e., viscosity,
penetrometer reading, etc. )
No gel, no viscosity
increase.
Precipitate settles out upon
standing.
Formation of homogeneous anc
thick gel.
Globular precipitate forms
upon standing.
No gel, slight viscosity
increase.
Thin, globular precipitate
forms upon standing.
Mass of precipitate formed
upon standing.
Mass of precipitate formed
upon standing.
Mass of precipitate formed
upon standing.
No gel, no viscosity
increase.
110

-------
                         Gelling
                       Components       Remarks (i.e., viscosity,
   Components	(% By Weight)     penetrometer reading, etc.)_
      POPTA 403            2.80         Globular precipitate forms
      TMD-1               2.20         upon standing.
Two component systems a proprietary amine and a proprietary
isocyanate.
                                Ill

-------
         APPENDIX G

STABILITY IN STORAGE TESTS FOR
      GEL CONSTITUENTS

-------
                            APPENDIX G

                  STABILITY IN STORAGE TESTS FOR
                         GEL CONSTITUENTS
Procedure for determining equivalent weight of amines (total amines):

        1.   Accurately weigh three grams of amine.

        2.   Place amine into Erlenmeyer flask. Add 40 milliliters
            of dry isopropanol  (99%).

        3.   Add one milliliter of bromcresol green indicator and
            with agitation,  titrate to yellow end point with about
            . 5N HC1.

        4.   Calculate equivalent weight of the amine as follows:

            equiv wt amine =   (1.000 x (wt amine added)
                            (ml of acid added) x (N of acid)

Procedure for determining equivalent weight of isocyanates:

        1.   Accurately weigh three grams of isocyanates to
            be  checked.

        2.   Place isocyanate into Erlenmeyer flask and add 25
            milliliters  dry toluene.

        3.   Accurately weigh nine grams of dibutylamine.

        4.   Add amine to flask  containing isocyanate and toluene
            and warm.  Let stand  for 15 minutes before titrating.

        5.   Add 100 milliliters  of dry isopropanol  (99%).

        6.   Add one milliliter bromcresol green indicator and
            titrate to yellow end  point with about  . 5N HC1.

        7.   Calculate equivalent weight of isocyanate as follows:

                       	wt isocyanate added	
  equiv wt isocyanate =  wt amine added - (ml of HC1) x (N of HCT"
                       equiv wt  of amine       1, 000

The equivalent weights  should be determined once at the beginning of
storage period for reference and then at any doubtful interval.
                                113

-------
       APPENDIX H



OPTIMUM RATIO STUDIES

-------
   200--
 e
 s
 w
 +J
 o
 B1
 (0
 0
   250--
   300--
 CD
 C
 0
P-,
   350
              0.50    0.75    1.0    1.25    1.50
                     Stoichiometric Ratio Factor

                  Figure H-l.  Armeen C and MT-40
                                                 __l	^_
                                                  1.75    2.00
S
S
w
   150--
(0
   200..
t-,
CD
+->
0


S~i
-t-J
CD
G
0
PL,
250--
   300
                              -I-
                                   4-
-4-
              0.50   0.75    1.0     1.25    1.50
                     Stoichiometric Ratio Factor

              Figure H-2.  Jeffamine 400 and Tonco 90
                                                  1.75   2.00
                                115

-------
  200--
s
a
w
  250 ._
-o
(0
~
s
2
-t->
CD
C
0
PL,
  350
                     4-
4-
                1
            0.50     0.75    1.0    1.25
                 Stoichiometric Ratio Factor
1.50    1.75
                             2.00
                FigureH-3.  Armeen O and TDI 80/20
  200 '-
S
S
w
  250 --
T3
fO
0)
CD
S
O
s-,
4->
CD
C
0
OH
  300 --
  350
            0.50     0.75    1.0    1.25
                 Stoichiometric Ratio Factor
               1.50   1.75
               2.00
               Figure H-4.  Armeen C and TDI 80/20

                               116

-------
 s


 CO
   200  --
 (0
   250  -
 S

 2
 +J
 a)
 G
 0
 CU
                      Failed
   300  ._
   350
          -H	1	1	1	1	
           0.50    0.75      1.0   1.25     1.50

                   Stoichiometric Ratio Factor


            Figure H-5.  Armeen L-15 and TDI 80/20
                                                     1.75    2.00
   200  --
S
S
to
x;
-rH

T3
1-1
0)
4J

-------
                APPENDIX I



GEL STRENGTH VERSUS PERCENT GELLING AGENT

-------
  200
                      Clearfork

                      Westbrook
  250
               	Howard Glass cock
               	  Solvent Mix(Sun Oil Co.)
K
u
0)
j->
CD


O
0)
CU
   300
   350
  400
                                         Percent Gelling Agent


                                  Figure I -1. Armeen L-ll and TDI 80/20

-------
tM
O
          200
        S
        6
          250
fi  300
--i
T3
(0

-------
INJ
         20C
       6 Z5C
       en
       c
       M

       g 300
       CD
       -t-
       CD
       S
       o
       0)

       CD
       PL,
350
         400
                             Clearfork
                             Westbrook
                     _.	 Howard Glasscock
                     	Solvent Mix (Sun Oil Co)
                      -RV
                      i
                                                 Percent Gelling Agent

                                            Figure I -3.  Armeen C and  MT-40

-------
CsJ
         200
       E 250
      ,c
      4->
      o
       tn
       C
       a)
         300
         350.
      CD
      PL.
        400
                           .Sun B Mix
                            469 Mix
                           "Mirando-RHC Mix
                            Sweden Crude
                                              Percent Gelling Agent

                                      Figure I -4.  Armeen C and MT  - 40

-------
       200
N>

OO
      s
      S250
      o
      II


      tn
      S300
      (1)
      M
      Q)

      S
      O
      i-,


      1350

      (D

      P-,
        400
                                Sun B Mix

                                Mirando-RHC Mix

                                469 Mix

                                Sweden Crude
                                             Percent Gelling Agent


                                Figure I- 5. Jeffamine 400 and Octadecyl Isocyanate

-------
   200
I 250
Cn
-i-i
To 300
->u |

(0
PL,
  400
Clearfork
Westbrook
Howard Glass cock
Solvent Mix (Sun Oil Co. )
                        234
                                        Percent Gelling Agent
                           Figure 1-6. Jeffamine 400 and Octadecyl Isocyanate

-------
Ul
         200
       1 25
       tn
io 300
(D
K
S-j
Q)
4-3
CD
S
o
       CD
       C
       Q)
         350
         400
                              Clearfork
                              Westbrook
                              Howard Glass cock
                   (Did not gel) Solvent Mix (Sun Oil Co.)
                                                      3                     4

                                                Percent Gelling Agent

                                         Figure I -7. Armeen L-15 and TDI 80/20

-------
N)
      CD
      S-.
      CD
g

g
c
cu
D-,
         200
      125.
      D)
         300
        350
        400. _
                             Sun B Mix
                             469 Mix
                             Sweden Crude
                (Did not'Gel) Mirando-RHC Mic
                                                    3                     4
                                            Percent Gelling Agent
                                      Figure I-8.  Armeen L-15 and TDI 80/20

-------
  200
I 25
"g 300
    +
CD
s
o
  350
CD
OL,
  400
            	  Clearfork
             	  Westbrook
            	Howard Glass cock
            ____.!  Solvent Mix(Sun Oil Co. )
                                         Percent Gelling Agent
                                 Figure I -9. Armeen O and TDI 80/20

-------
Cs)
00
        ZOO
6
S 250
      Cn
       300
      CD
      4->
      CD
      g
      O



      I  350
      CD
      P-i
       400
                            Sun B Mix
                            469 Mix
                            Mirando- RHC Mix
                            Sweden Crude
                                                   4-
                                                                    4-
                                                    3                       4

                                             Percent Gelling Agent


                                      Figure I-10.  Armeen O and TDI 80/20

-------
vD
         ZOO
       a
       s
       tn
       C
       iH
       -a
       
       CD

       g 35
       0-,
         400
Clearfork
Westbrook
Howard Glass cock
Solvent Mix (Sun Oil Co.)
                                                     3                       4
                                                Percent Gelling Agent
                                            Figure I -11. Armeen C and TDI 80/20

-------
        200
UO
O
      S
      fi
      i-H
      T3
      (0
      S-i
      CD
      4-J
      0)
      S
      O
      i-1
      -t->
      Q)

      g
      P-,
        25
        400
	Sun B Mix
	469 Mix
	Mirando - RHC Mix
      Sweden Crude
                                                      3                       4

                                                Percent Gelling Agent

                                       Figure 1-12. Armeen C and TDI 80/20

-------
          APPENDIX J



GEL AGENT DISPENSING CONCEPTS

-------
                                   TABLE J-I.  GEL AGENT DISPENSING CONCEPTS
                 Concept
                                        Advantages
         Disadvantages
OO
1.  Rotating Nozzle

Use of injection nozzles which
are canted and designed to rotate
horizontally on the discharge
line. One agent is discharged
from the nozzles during travel
vertically from bottom to top of
tank.  Crude from tank is used to
flush agent from the discharge
system. Second agent is
discharged from nozzles during
travel from bottom to top of tank-

2.  Recirculation
        Pump oil out of tank. Mix in
        gel agents and return to tank.
        3. Recirculation With Air

       Pump oil out of tank,  mix in
        gel agents and air and return to
        tank. Gelled oil floats on oil.
                                           Can be portable or permanent
                                           ship equipment. Oil gelled in
                                           bulk process.
                                  All oil treated discretely and
                                  uniformly. Can be permanent
                                  or portable system.  Simple
                                  installation. No obstructions
                                  in tank.
                                  All oil treated uniformly. Could
                                  be permanent or portable
                                  system.  Simple installation.
                                  No obstructions in tank.
                                  Minimum dilution  between
                                  gelled and ungelled  oil.
Theoretical analysis of
dispersion limits impossible.
Testing series is required to
define equipment parameters.
Operating limits are defined
by gelling time.
High lift for oil removal pump.
Pump suction needs to vary
for different oil levels. Some
dilution between treated and
untreated oil.
High lift for oil removal pump.
Auxiliary power is needed.
Pump suction level needs to
be variable for different oil
levels.  Is  not known if gelled
oil with air floats in oil.

-------
               Concept
                                        Advantages
                                    Disadvantages
00
OJ
       4. Vertical Diffuser

       Diffuser pad,  plate or pipes
       adjacent to outer hull - to
       distribute gel agents next to
       possible holes.
5. Horizontal Diffuser

Lattice work of pipes with
set nozzles are placed at bottom
of tank. Air injected with agents
to foam or lighten gelled crude
to make it float.
       6. Submersible Pump

       Use submersible pump
       lowered in tank of crude oil,
       inject air and gelling agents
       into pump to gel and float crude
       oil.
                                   Near possible leak. Gel
                                   minimum crude oil.  No major
                                   obstruction added to tank.
No interference between gelled
oil and gelling process. No
major obstruction added to
tank. May possibly be able
to add portable system to
tanks as needed.
                                   No installation in tanks is
                                   required. Minimum mixing of
                                   gelled and ungelled oil.
                                Permanent installation only.
                                Can not be used for portable.
                                Needed in every tank.
                                Possibility of damage in
                                collision since near outer
                                shell. Cannot distribute gel
                                agents to entire tank.
Difficult to adopt as portable
system.  Lattice may not be
in oil layer after leak started.
Floating gel may not seal leak
in bottom of tank until much
loss has occurred. Auxiliary
power is needed to pump
agents.  It is not known if
air mixed with gel agents
floats on oil.
                                Very high horsepower is
                                required to agitate entire
                                tank. Air may not float gel.

-------
               Concept
      Advantages
        Disadvantages
O0
       7.  Propeller Mixer

       Use portable or permanent
       propeller type mixer in tank. Gel
       agents dispensing into vortex by
       rotating  blades.
       8. High Velocity Jets

      High velocity jets, powered
       by gas  generator (propellant, gun
       or rocket) or explosive . Locate
       tanks of chemical on deck for
       movement to leaking tank.
       9. Honeycomb

      Honeycomb with gel agents in
       the cells are placed at bottom of
       tank. Agents are released at
       emergency by ultrasonic generator
       or other means.
Simple installation. Portable or
permanent system possible.
No external power source
necessary. Possible to direct
toward leak.
Always in tank ready to go. No
movement of materials  or equip-
ment during emergency.
High horsepower to achieve
good mix. Installation
difficult.
Tank very heavy and difficult
to move in a possible no
power condition. Possible
hazard with explosives and
crude oil. Good mixing of
agents and crude not possible
from single application point.
Five percent of ship capacity
permanently lost by carrying
agents to gel all tanks. No
way to mix gel agents into
crude. Danger of rupturing
cells  during tank cleaning. No
way to check for deterioration
of gel agents. Special equip-
ment  necessary to break
Honeycomb.

-------
                  Concept
                                         Advantages
                                      Disadvantages
UO
Ul
          10. Encapsulated Agents

          Gel agents are encapsulated
          in beads of wax, polyethylene,
          gelatin, etc.  Pour into tank at
          time of emergency. Agents  are
          released by dissolving capsule,
          Destroy capsule by ultrasonic
          or crushing.
          11.
Tapes made of little bags of
gelling agent. Tapes are dis-
pensed into tank in proportion
to capacity of tank. Tapes  are
dissolved or run through rollers
to expel gelling agents.
          12. Horizontal Diffuser -
              Top of Tank

          Lattice work of pipes with
          jet nozzles are placed at top
          of tank.
                                 No handling of liquids. Simple
                                 application method.  No ship
                                 modification is required.
No liquids to handle. Very
minor ship tank modifications.
                                 Always out of water layer. Can
                                 be permanent simple install-
                                 ation. No major obstruction
                                 added to tank. Could be
                                 portable.
                                Not possible to get uniform
                                distribution in crude oil tank
                                density of crude oil varies.
                                No way to control sinking
                                of beads  in fluid.
No positive or complete mix
possible.
                                Adequate mixing in all parts
                                of tank probably not attain-
                                able.  Too much gelling
                                possible in same area.

-------
              Concept
       Advantages
         Disadvantages
bo
         13.  Discharge Overboard

         Pump oil out of leaking
         tanks. Gel oil and discharge
         overboard.
         14.  Endless Belt

         Endless moving belt to
         carry gelling chemicals down
         tank wall.
         15.  Diaphragm.

        Construct loose vertical
        diaphragm in tank,  pump oil
        from one side of diaphragm,
        add gelling agents and return
        to other side.
Oil is removed from leaking
tank. Good mixing of oil and
gel agent. All oil is treated.
No dilution of gelled oil after
mix. Gelled oil easy to
harvest.  Gelled oil on  water
is not spread over large area.
Could use as permanent or
portable  system.  Simple
installation.  No obstructions
in tank.
Complete isolation of crude
oil and gelled crude. Prevents
loss of oil  in itself.
High lift for oil removal pump.
Need special equipment to
harvest.  Pump suction needs
change for changing oil levels
Hoist is needed to place pump
in place.

If we could pump  out of tank,
we could just as easily put
ungelled oil into a barge for
salvage rather than putting
gelled oil on water.
                                No way to mix gel agents
                                with all the oil in the tank.
                                Expensive installation.
Obstructions in tank prevents
this method. Probably
expensive.

-------
       Concept
      Advantages
       Disadvantages
16. Random Cylinders

Chemicals are stored in
alternate ends of cylinder with
offset nozzle on each end.
Compressed gas is stored to
expel gel agents and at same
time impart rotational or random
movement of cylinder.  Many
small cylinders are used  to gel
oil.
No alternate power source
needed.
Large number of devices are
needed to gel large tank.
Device needs to be extra
heavy to permanently store
compressed gas. Device is
presently not in state-of-art
of chemical dispensing, much
development is needed.

-------