EPA 670/2-73-078
September 1973
                           Environmental  Protection Technology Series
                  of
Material Spills
                   ing  Ma          nsfer  Media
                                          of Research and Development

                                             imental Protection Agency
                                                  20460

-------
            RESEARCH REPORTING SERIES
Research reports of the  Office  of  Research  and
Monitoring,  Environmental Protection Agency, have
been grouped into five series.  These  five  broad
categories  were established to facilitate further
development  and  application   of   environmental
technology.   Elimination  of traditional grouping
was  consciously  planned  to  foster   technology
transfer   and  a  maximum  interface  in  related
fields.  The five series are:

   1,  Environmental Health Effects Research
   2.  Environmental Protection Technology
   3.  Ecological Research
   (|.  Environmental Monitoring
   5.  Socioeconomic Environmental Studies

This report has been assigned to the ENVIRONMENTAL
PROTECTION   TECHNOLOGY   series*    This   series
describes   research   performed  to  develop  and
demonstrate   instrumentation,    equipment    and
methodology  to  repair  or  prevent environmental
degradation from point and  non-point  sources  of
pollution.  This work provides the new or improved
technology  required for the control and treatment
of pollution sources to meet environmental quality
standards.

-------
    TREATMENT OF HAZARDOUS MATERIAL SPILLS

       WITH FLOATING MASS TRANSFER MEDIA
                      By

               Basil W. Mercer
               Alan J. Shuckrow
               Gaynor W. Dawson
               Project 15090 HGQ
              Contract 68-01-0124
                Project Officer

                  Ira Wilder
        Environmental Protection Agency
Edison Water Quality Research Laboratory, NERC
           Edison, New Jersey   08817
                 Prepared for

       OFFICE OF RESEARCH AND MONITORING
    U. S. ENVIRONMENTAL PROTECTION AGENCY
           WASHINGTON, D. C.   20460

-------
                   EPA Review Notice

This report has been reviewed by the EPA 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 recommendation for use.
                          11

-------
                       ABSTRACT
An approach for the in situ treatment of spills of soluble
hazardous polluting substances was developed and demon-
strated on a field scale for a static body of water.
Laboratory-scale experimentation showed that floating
sorbents and ion exchange resins could be highly effective
removal agents when applied as small particles beneath
the surface of contaminated waters.

A lightweight commercial activated carbon was found to be
particularly effective for removing organic substances such
as phenol, aromatic hydrocarbons, and organophosphorus
insecticides from water.  The activated carbon can be
pulverized to a small mesh size  (100 x 325) which floats
slowly to the surface of the water.  The small mesh size
enhances both the contact time and the sorption kinetics.

Floating ion exchange resins were also prepared for use
on spills of acid (e.g., sulfuric and hydrochloric), alkalis
(e.g., caustic soda) and toxic salts (e.g., sodium cyanide).
Hollow glass microspheres are incorporated in the resin
granules for buoyancy.

Delivery systems which show promise for field use include:
(1)  ice cakes containing floating media and gravel ballast;
(2)  clay cylinders;  and  (3) weighted plastic bottles.

Field demonstrations were conducted using carbon contained
in weighted plastic gallon bottle.s.  The carbon proved
highly effective in removing an organophosphorus pesticide
spilled in a large basin, and was easily collected through
use of an oil containment boom.  Ice encapsulated floating
anion exchange resin beads were similarly employed to
neutralize a spill of sulfuric acid.

This report was submitted in fulfillment of Project Number
15090 HGQ, Contract 68-01-0124 under sponsorship of the
Office of Research and Monitoring, Environmental Protection
Agency.
                          10.1

-------
                       CONTENTS






Section                                                Page



  I       Conclusions                                    1



  II      Recommendations                                5



  III     Introduction                                   7



  IV      The Concept                                   13



  V       Laboratory Studies                            25



  VI      Field Demonstrations                          61



  VII     Application                                   83



  VIII    Acknowledgments                               87



  IX      References                                    89



  X       Appendix                                      91

-------
                         FIGURES


                                                        Page

 1    Spill of Hazardous Material                        18

 2    Officials Are Notified.  Response Team             19
      Is Set Into Action

 3    Appropriate Media Type Is Selected and             20
      Loaded Onto Delivery Plane

 4    Individual Media Packages Are Air Dropped          21
      Into Contaminant Plume

 5    Packages Sink and Begin to Decompose.  Media       22
      Floats to Surface Removing Contaminant

 6    Standard Oil Skimmers Are Used to Retrieve         23
      Media

 7    Sorption Isotherm for Phenol and Nuchar C-190      28

 8    Comparison of C-190 with WA for Removal of         30
      Phenol

 9    Titration Curves for a Strongly Acidic Cation      36
      Exchange Resin and a Weakly Acidic Cation
      Exchange Resin

10    Polymerized Acrylic Acid Crosslinked with          39
      Divinylbenzene

11    Polymerized Acrylic Acid Crosslinked with          41
      Ethylene Glycol Dimethacrylate

12    Titration Curve for Floating Acrylic Ion           42
      Exchange Resin

13    Example of Structure of an Epoxypolyamine          43
      Anion Exchange Resin

14    Titration Curve for Floating Epoxypolyamine        45
      Anion Exchange Resin

15    Titration Curve for Diamond Shamrock Weak Base     47
      Epoxypolyamine Resin
                             VI

-------
                         FIGURES
                       (Continued)
                                                        Page

16    Acrylic Floating Cationic Exchange Beads           49

17    Epoxypolyamine Floating Anionic Exchange Beads     50

18    Schematic of Carbon Filled Plastic Bottle          54

19    Photograph of Clay Container                       58

20    Location of Facilities for Diazinon Spill          62

21    Aerial Photograph of Diazinon Spill                64

22    Closeup of Diazinon Spill in Sampling Grid         65

23    Results of Phosphate Analysis on Pretreatment      66
      and Posttreatment Samples

24    Additional Results of Phosphate Analysis on        67
      Posttreatment Samples

25    Results of TOG Analysis on Pretreatment and        69
      Posttreatment Samples

26    Additional Results of TOC Analysis on Post-        70
      treatment Samples

27    Photograph of Helicopter with Sling                72

28    Photograph of Air Drop Into Spill Area             73

29    Placement of the Collection Pipe and Boom          75

30    Typical "Media Cake" Employed in Basin             78
      Demonstration

31    Photograph of a Resin Ice Cake                     79

32    Sampling Floating Resin after Treatment            81
                           V1JL

-------
                        TABLES
No
1    Surface Areas and Iodine Numbers of Several
     Commercial Grade Carbons4'5                         25

2    Approximate Floatability of Various Mesh
     Sizes of Nuchar C-190                               26

3    Rise Time for Nuchar C-190 in Four Feet
     of Water                                            27

4    Removal Effectiveness for Nuchar C-190 for
     Various Organic Materials                           29

5    Removal Effectiveness of Nuchar C-190 for
     Organophosphorus Pesticides                         31

6    Percentage Removal of Phenol, Malathion, and
     Diazinon at Various Depths in the Treatment
     Column After Flotation of the Carbon                32

7    Ballast and Package Size Requirements for
     Selected Media                                      51

8    Product Evaluation of Soluble Films                 56

9    Composition of  Emulsifiable Diazinon Solution       61
                         Vlll

-------
                     SECTION  I
                    CONCLUSIONS
The removal  of selected hazardous  materials  from water with
floating mass  transfer  media was successfully demonstrated
in laboratory  and field tests,  using floating activated
carbon for organic removal  and  floating ion  exchange resins
for^lectrolyte removal. The following is a list of con-
clusTons based on these studies.
GENERAL
       The technical feasibility of the concept of
       subsurface injection of floating mass transfer
       media for treatment of hazardous polluting
       substances has been established through field
       demonstration.

       This concept shows great promise as a spill
       response technique.

       Surface collection of spent floating media in
       a static body of water can be accomplished with
       existing oil spill cleanup equipment.

       The quantity of floating media required for a
       desired level of removal of spilled hazardous
       material will vary with the type and concen-
       tration of material being removed.

       The effectiveness of spill treatment decreases
       as the spilled hazardous material becomes more
       dilute.  Thus, rapid response to a spill sit-
       uation is of paramount importance.  For example,
       10 grams of activated carbon will remove 97 per-
       cent of the phenol from a one gram per liter
       solution of this material.  Sixteen grams of
       carbon would be required to obtain the same
       level of removal if the one gram of phenol were
       dissolved in 10 liters of water.

       The use of weighted packages of floating media for
       treating hazardous material spills represents the
       most immediate possible response to these situations
       in remote areas.  Where spill frequency is high
        (e.g., busy harbors) bulk application with infusion
       pumps is thought to be the preferred method of
       application.

-------
FLOATING ACTIVATED CARBON

     ,  Two commercially available activated carbons,
        Nuchar C-190 and Nuchar WA, can be applied at
        the bottom of a water column to effectively
        sorb organic matter as they float to the surface.

     ,  Nuchar C-190 is the carbon of choice based upon
        superior sorptive ability and superior floatability,

     •  Fine carbon particles in the 100 x 325 mesh range
        are capable of approaching equilibrium sorption
        by floating once through the water column.

     •  Both dry and pre-wetted carbons are equally
        effective in removing organics from aqueous
        solution.

     •  Nuchar C-190 is highly effective in sorbing phenol,
        toluene, benzene, styrene, Diazinon, Malathion,
        and, to a lesser extent, acrylonitrile.

     •  A single pass of Nuchar C-190 through a water
        column removes approximately 80 percent of the
        Diazinon spilled in a ten million gallon water
        basin.

     •  The granular grades of Nuchar C-190 used in this
        study are no longer commercially available.  A
        powdered grade, Nuchar C-190-Nr is available but
        this material was not evaluated.

     ,  An impact method of crushing the carbon produced
        a product superior to that obtained by ball
        milling.

     •  The floating activated carbon can easily be herded
        with the aid of a plastic containment boom and
        recovered from the treated water by pumping as a
        slurry from the surface of the water through a
        diaphragm pump.

FLOATING ION EXCHANGE RESINS

     •  The use of weakly acidic and weakly basic ion
        exchange resins for treating spills of acidic or
        basic electrolytes has two major advantages over
        the use of neutralizing chemicals.  These are:
         (1) application of excess resin will not cause pH
        extremes and  (2) the resins are effective for both
        neutralization and removal  of dissolved matter.
        The principal disadvantage is the high cost of
        the resin  (approximately  $290  per  cu.  ft.)

-------
        Floating ion exchange resins were synthesized by
        incorporating hollow glass microspheres in the
        resin matrix to provide the desired buoyancy.

        Bulk polymerization and suspension polymerization
        techniques were found to be successful for prep-
        aration of a floating weakly acidic carboxylic resin
        and a floating weakly basic epoxypolyamine resin,
        respectively.

        Suspension polymerization is the preferred pro-
        duction technique since it should be far more
        economical than bulk polymerization techniques.
        Both cationic and anionic resins can be produced
        by suspension polymerization; however, only the
        preparation of floating anion resin was
        successfully demonstrated by this method.

        The feasibility of commercial production of
        floating resins has been demonstrated by the
        production of several hundred pounds of floating
        epoxypolyamine anionic resin.

        A low removal level of sulfuric acid was observed
        in the field demonstration, but is not considered
        representative of the potential of floating ion
        exchange resins for treating spills of electro-
        lytes.  Rather, the low removal efficiency was the
        result of poor contact between the resin and the
        acid due to a bottom layering effect of the acid.
PACKAGING AND DELIVERY
        Ballast requirements to sink the floating media vary
        from 0.2 pounds of ballast per pound of wet media
        to 12 pounds of ballast per pound of dry media.

        The inclusion of water or ice in the interstices
        of the media significantly reduces the ballast
        requirement.

        Routine packaging techniques which show promise
        for field use include (1) ice cakes containing
        floating media and gravel ballast; (2)  clay
        containers; and (3)  weighted plastic bottles.

        Based on the work to date, the ice cake approach
        appears to be preferable because of lower ballast
        requirements and superior dispersion characteristics
        near the bottom of the water column.

-------
Ice cake packaging is the only identified technique
with no potential for water quality impairment.
This technique involves a large temperature drop
near the ice cake but the average temperature
change in the treatment area will be only 1 to 2 °F.

The feasibility of packaging carbon and ballast
in plastic containers and resin and ballast in
ice cakes has been demonstrated in field tests.

The feasibility of air drops of floating carbon
packaged in plastic containers has been demonstrated
on a large scale.

-------
              SECTION II
            RECOMMENDATIONS
As a result of the success of the laboratory
studies and field demonstrations described in
this report, efforts should proceed to further
develop and implement the floating media concept
for treatment of spills of hazardous polluting
substances.

A slurry injection system should be developed
to allow routine use of floating sorbents in
ports and areas with a history of high spill
frequencies.

Packaging techniques for air transport and delivery
of floating media which were investigated in
this program should be optimized.

A program should be undertaken to extend the
current work to flowing streams.

Further studies should be initiated to establish
the optimum media mesh size with respect to
removal efficiency, cost, and in the case of the
carbon, method of pulverization.

Commercial production techniques for manufacture
of floating ion exchange resins should be refined.

Production techniques for suspension polymerization
of floating cationic resins should be developed.

Techniques for marking spills of various hazardous
materials should be developed, permitting visual
delineation of the spill area and greatly facil-
itating response activities.

Consideration should be given to the development
of a sinking sorption media which can be easily
recovered from the bottom of a watercourse.

-------
                       SECTION  III
                       INTRODUCTION
 Pollution  resulting  from  the  spillage of oil  and hazardous
 materials  has emerged  as  a major national problem.   It
 is presently estimated that some fifteen thousand spills
 involving  oil and hazardous materials occur annually
 in the navigable waters of the  United States^'.
 These spills range in  size from small quantities to
 millions of gallons(2)  and threaten many important
 waterways  in the country.

 Damage from such spills is often extensive.   Massive spills
 can render waterways unfit for  some or  all major beneficial
 water uses and  in some instances may eliminate  large
 segments of the biota  in  the  affected area.   Recognition
 of the severe environmental threat represented  by acci-
 dental spillage of hazardous  polluting  substances has led
 to legislative  and administrative steps designed to
 minimize acute  releases and provide for immediate response '
 activities when necessary.  Of  particular importance is
 the development of methods for  in situ  treatment of spills.
 This report describes  an  EPA  sponsored  program  aimed
 at the development of  one such  method of in situ treatment
 of hazardous materials spills.

 Potential  treatments which conceivably  can be applied iri
 situ to ameliorate the effects  of hazardous polluting
 substances in the aquatic environment are almost as
 numerous as the substances themselves.  Obviously, it
 would be impractical to attempt to develop a  specific
 countermeasure  for each individual hazardous  polluting
 substance.  Moreover,  since the resources which can  be
 invested in developing countermeasures  against  hazardous
 polluting  substances in the aquatic environment are  limited,
 efforts must be directed  toward those areas which promise
 to produce the  maximum return on this investment.

JBattelJLej-Northwest used the following criteria  to evaluate
 potential  countermeasures.

     •  Countermeasures should  be highly effective.

     •  Countermeasures should  be applicable  to a large
        number  of substances.

-------
     •  Counter-measures should be amenable to rapid, easy
        deployment.  Equipment and/or chemicals which
        cannot be rapidly conveyed to the scene of an
        incident are undesirable.

     v  Countermeasures should be free from potentially
        harmful secondary effects in the aquatic environ-
        ment, including noxious sludges,

     •  Countermeasures developed to combat spills of
        hazardous polluting substances should take ad-
        vantage of existing technology, particularly
        that technology developed to combat oil spills,
        to the maximum possible extent.

     •  Countermeasures should provide for easy isolation
        and removal from water of substances added to
        extract or precipitate spilled materials.

Consideration of numerous, potential methods for the in
situ treatment of spills of hazardous polluting substances
led to the following conclusions.

     •  Chemical degradation (transformation, oxidation-
        reduction, etc.)is generally unacceptable since
        in most cases the agents which must be added to
        effect reaction pose an equal or greater potential
        hazard to the aquatic environment than the original
        pollutant.  The situation is further complicated
        by the nature of the spill environment which does
        not allow close control of operating conditions.
        Concentration changes resulting from natural
        dilution processes could cause excess chemical
        addition  and so increase the damage to the
        aquatic environment.

     •  Complex formation and neutralization  suffer from
        difficulties  similar to those of chemical  degrada-
        tion,  in the  lack of control of the reaction
        environment and the inherent threat to water
        quality of the  necessary additives.

     •  Precipitation techniques at first appear to be
        an attractive solution; however, the ultimate
        reaction poses a new type of threat.  Without an
        effective means for removing the precipitate, the
        watercourse is subjected to a concentrated buildup
        of fine particulate matter which could damage
        feeding and spawning grounds, with subsequent
        deleterious effects on aquatic life.  The full

-------
impact of this type of damage has never been
satisfactorily explained, but presumably it
could rival the damages of the original pollutant
to water quality.  Furthermore, the introduction
of colloidal precipitates would threaten gilled
species and would decidedly damage recreational
and aesthetic water quality characteristics.

Solvent extraction and foam separation processes
have several intriguing characteristics.  With
proper choice of solvent, the product stream can
be made to rise to the surface where collection
with existing hardware should be possible.  In
addition, solvent extraction and foam separation
processes are fairly broadbased, with a single
component displaying applicability to many
pollutant releases.  Unfortunately, both processes
require some control over the area to be treated,
and treatment of any large body of water would
necessitate use of several types of feed mech-
anisms.  This implies high capital costs and
storage facilities to stockpile such equipment
regionally.  Further, in the case of solvent
extraction, the solubility of the solvent in water
would cause a considerable pollution problem in
itself due to the large quantity of solvent
required.

Skimming and booming are available techniques for
oil spills.  They could be applied to light in-
soluble materials as well as slightly soluble
organics.  Rapid response aimed at booming and
skimming any undissolved portion of the contaminant
could greatly reduce the severity of a spill.

Flow augmentation can be applied only in certain
circumstances.  Mechanical mixers such as outboard
motors can be used to aid diffusion.  Flow augmen-
tation may be utilized where stored water is
available upstream.  In any event, such a practice
relies in part on good predictive models of the
aquatic environment to determine the extent of
contamination and expected duration of toxic
concentrations.  This information is available for
a number of streams, as is time-of-travel data.
Such data are not gathered in a single source nor
properly referenced to allow for quick retrieval
and application in a practical emergency, however.

-------
      •   Burn-off  techniques, because of  safety  and  air
         quality considerations,  are probably  applicable
         only  in very  limited cases in which the material
         remains confined  in a  small isolated  area.   Many
         hazardous materials emit highly  toxic vapors when
         heated to decomposition.  Consequently, burning
         may  intensify hazardous  conditions and  force
         evacuation of areas surrounding  a spill site.
         The  presence  of heavy  vapors which can  travel
         along the ground  and cause flash-backs  further
         endangers safety  when  burn-off techniques are
         applied.

      •   Oxygen addition may be an attractive  alternative
         for  substances posing  a  BOD problem.  The oxygen
         is added  to the water  to maintain dissolved oxygen
         levels while  a spilled material  is being dispersed
         and  degraded  aerobically.   There has  been recent
         interest  in injecting  air or  pure oxygen into
         water courses to  achieve this  end in  areas  of
         chronic  low dissolved  oxygen.  The required
         apparatus is  simple enough  to allow mobilization
         for  transport to  a spill site.   The problems
         would involve supplying  enough gas to accomplish
         the  task  and  designing equipment capable of
         efficiently transferring the  gas into a variety
         of potential  stream or reservoir configurations.

     •   Biological  degradation,  while attractive in some
         respects,  encounters several difficulties.  In
         order  to  encourage degradation at a rapid rate
         it would  be necessary  to have on hand large
         quantities  of  acclimated cultures.  The problems
         associated with stockpiling many such cultures,
         each specific  to  a particular substance, are
         obvious.   Also, many hazardous materials are
         inherently  resistant to  biological degradation.

     •   Physical  sorption and  ion exchange processes can
         remove substances from solution within  a period of
         a few minutes  and hold the substances in a solid
         form for  long  periods  of time.  The use of ion
         exchange  resins for acid, base, metal,  or toxic
         salt spills and the use of activated  carbon or
         sorption  resins for organic matter could apply to
         a large number of hazardous  polluting substances.

Based on the  above considerations,  it was concluded that
physical sorption and  ion exchange  processes  offer the
greatest promise   for treatment of acute spills  in the aquatic
environment.
                           10

-------
The two major problems associated with the application of
mass transfer media to counteract spills of hazardous
polluting substances in a free water environment are:

     1.  the mechanics of introducing the active media
         into the water (a common problem with all
         treatment agents) and

     2.  the subsequent removal of the media from the
         water.

Traditionally, sorption and ion exchange processes have
been largely restricted to operation within columns.
This method of application is limited by some very real
concerns:  the cost of stockpiling mobile columns for
spill treatment, the problem of rapid deployment of the
large equipment involved, and the practicality of pumping
free waters into a contained environment for treatment.
Some studies have been conducted with powdered activated
carbon in which the sorptive media was dispersed on the
water and allowed to settle(3).  Although this approach
may be satisfactory for selected impoundments, in flowing
streams the aesthetic damage would be extensive and gilled
species would be subjected to the problems inherent in an
environment highly saturated with particulate and colloidal
matter.

An alternative approach is the use of floating mass trans-
fer media wherein the problems of posttreatment residual
effects would be eliminated by collection of the spent
media after use.  Application problems would be reduced
to those of subsurface injection and surface collection.
Subsequent to these conclusions, a program was initiated
by Battelle-Northwest under EPA sponsorship to develop
and demonstrate the concept of floating physical sorbents
and ion exchange resins for iji situ treatment of spills
of hazardous polluting substances.  This report describes
the results of the work accomplished in that program.
                           11

-------
                       SECTION  IV
                       THE  CONCEPT
 Sorption processes  are highly effective  for  removing many
 different  substances  from water.   Their  principal  ad-
 vantages are:   (1)  treatment can be  accomplished without
 the addition of chemical reagents  to the water;  (2) one
 sorbent can be used to remove a large number of different
 substances from water; and  (3) the sorbent collects and
 holds a hazardous material  in a less nocuous (undissolved)
 form for long periods of time.  The  term "sorption", as
 used in this report applies to both  physical sorption
 (e.g., surface attraction as in activated carbon)  and ion
 exchange.

 Physical sorbents act like  chemical  sponges, soaking
 up dissolved pollutants without releasing undesirable
 materials  in their  place.   Ion exchangers, on the  other
 hand, replace a hazardous material in the aqueous  phase
 with a less objectionable material.  For acids and bases
 the H+ and OH" ions are neutralized  to water and the anion
 or cation  corresponding to  the acid  or base  is sorbed by
 the exchanger.  Therefore,  the acid  or base  is essentially
 removed from the water and  a corresponding amount  of
 salt does not remain as would be the case if the acid
 or base were chemically neutralized.  The treatment of
 a salt spill (e.g., CuS04) would most probably involve
 the replacement of  the toxic ion(s)  (e.g., Cu*2) with
 ions (e.g., Na+) that are more compatible with the aquatic
 environment.

 Sorption may be accomplished either  by batch treatment
 or by column treatment.  Batch treatment is  carried out
 by mixing the sorbent with the solution  to be treated
 for a specified period of time after which the sorbent
 is separated from the solution for disposal  or regeneration
 and reuse.   This method is simple  and may be carried out
with relatively low-cost equipment.  Column  treatment is
 accomplished by percolating the solution through a bed
of the sorbent until some level of the sorption capacity
 is utilized.   The sorbent in the column  is usually regen-
erated at this point and the loa   ng cycle repeated.
Column treatment is generally more efficient than batch
treatment and lower effluent concentration values may be
attained.   However, in the case of a spill situation,
column treatment would involve the transport of large
                          13

-------
volumes of water to a treatment facility and then return
of the treated water to the lake or stream.  Another
disadvantage of column treatment in the field is the
substantial time period required to set up and operate
the necessary equipment.  Because response and treatment
time must be rapid to minimize dispersal of the hazardous
material from the site of the spill, batch treatment
appears most feasible for field use.

Recovery of the sorbent after treatment is necessary to
prevent slow desorption of highly toxic material and
for aesthetic reasons.  Most sorbents will sink when
applied to water, which complicates their retrieval.
Moreover, dredging may result in the removal of large
quantities of benthic life and bottom materials with the
media.  Floating sorbents, however, can be recovered by
oil skimming or similar well-established surface collection
techniques.

It is anticipated that floating sorbents would be injected
at the bottom of a watercourse and allowed to float to
the surface through the zone of contamination.  Because
the sorbent must perform its function with only one pass
to the surface/ the kinetics of pollutant removal can pose
a problem.  Kinetic problems in columns are solved by
adjustment of the throughput rate to achieve the residence
time necessary to effect the desired degree of removal.
However, kinetic problems are not so simply solved in a
buoyant sorbent system.

The controlling parameters involved in a free-water
mass-transfer system include temperature, concentration
of solute, particle velocity, and particle surface area.
Obviously, the first two parameters are functions of the
location and nature of the spill and cannot be practi-
cally controlled in a free water environment.  (Indeed,
the purpose of applying a countermeasure is to reduce
the concentration of the solute [hazardous polluting
substance] to some acceptable level).  However, particle
velocity and particle surface area can be controlled
through adjustment of the particle size.

Since mass transfer rates are proportional to surface
area, the division of a large particle into several
smaller ones increases the surface area and thus acceler-
ates the mass transfer rates.  Reduction in particle size
also reduces the rate of rise of a floatable media since
buoyancy is a function of r3 while drag is a function of
r2.  Reduction of the rate of rise increases the residence

-------
time as the media rises to the surface.  Thus, reduction
of particle radius causes an increase in the rate of mass
transfer to occur.  Adjustment of media density can also
be utilized to increase the contact time.

Effective use of floating media requires that (1) buoyant
sorption media be deposited at the bottom of the water
column so that it removes selected dissolved contaminants
(i.e./ toxic ions or organics) as it rises to the surface;
(2) means be provided to uniformly distribute the media
during the deposition step or as it rises to the surface;
and (3) the application rate be properly proportioned to
the amount of pollutant to be removed or neutralized per
unit area of water surface.

Two general methods are envisioned for applying the mass
transfer media to achieve such a distribution: (1) pumping
bulk media as a slurry through a pipe which terminates
near the bottom of the watercourse and which is installed
on a boat or vessel propelled at a speed proportional
to the application rate, and  (2) dropping on the water
surface packages or media capsules weighted with ballast,
the packaging material of which disintegrates or opens
upon exposure to water after sufficient time to allow the
packages or capsules to reach the bottom.

The first method would involve the deployment of a
specialized set of equipment on boats or vessels.  In all
likelihood, the equipment could be made air transportable.
Nevertheless, some delay would result from the time
required for equipment handling and setup and possibly
for training local personnel to operate the equipment.
On the other hand, the package or capsule "bombing" tech-
nique could employ aircraft as well as surface vessels.
Moreover, the distribution of packages would be manually
controlled and no special equipment would be required—
eliminating the associated deployment delays.

Once the media has been applied and allowed to perform its
function, relatively long periods can elapse before recovery
of the spent media, because the potential pollutant will
be sorbed or fixed on a material which is essentially
insoluble.  The physical recovery of the spent media can
be achieved either manually or with mechanized equipment.

When treating a flowing stream, a simple but probably
effective approach would call for positioning of floating
baffles or booms at an angle across the stream and down-
stream of the spill treatment area.  This would serve to
direct the floating media to an accessible shore area.
                          15

-------
Pickup of the media could then be achieved either manually
or by using simple agricultural or earth moving equipment
(e.g., drag lines, back hoes).

In the case of static waters such as lakes, ponds, re-
servoirs, bays, and harbors, a means would be needed for
collecting or concentrating the media and then removing
it from the water surface or from adjacent shore areas.
This could be partially accomplished by techniques
similar to those mentioned for a flowing stream.  How-
ever, there is the added requirement for concentration
and possible removal of the spent media from the static
water surface.  The techniques required for this are
parallel to those necessary for recovering floating
oil slicks.  In fact, application of much of the technology
developed for oil spill countermeasures to the recovery
of spent sorption media appears feasible.

Use in saline waters will have differing effects.  Sorption
processes of materials less soluble in salt water than in
fresh water will be aided by the presence of the salts,
whereas sorption of organics more soluble in salt water
will be hindered.  Treatment of acid and base spills
will be more efficient in salt water than in fresh, while
treatment efficiency of toxic cations and anions will be
reduced.

The economics of such a treatment scheme depend upon the
mass transfer media selected.  In commercial quantities
activated carbon and common resins are relatively in-
expensive.  The opportunities for regeneration and possible
salvage of the spilled contaminant may further improve
the economics of this scheme.

The process is expected to be applicable to all types of
impounded or flowing waters, fresh or saline (however, this
work involved application to impounded waters only).  A
wide range of hazardous materials can thus be treated
with a minimum of equipment and elapsed time.  Stock-
piling of materials could be reduced to one or two
national centers.  Oil collection equipment will in fact
be used and thus should already be deployed through the
National Contingency Plan.  In many cases, the removal
efficiency should be sufficient to return receiving
waters to an acceptable water quality.  In the remaining
cases, treatment will substantially minimize the amount of
dilution required to reduce the pollutant concentration
below its damage threshold.
                         16

-------
The simplicity of application and ultimate effectiveness
of floating sorption media can best be illustrated with
a hypothetical spill incident.  Consider a barge which
founders while carrying 50,000 gallons of benzene.

Assuming the nature and location of the spill is reported
promptly and accurately, countermeasures could be taken
to minimize the damage to marine fisheries in the area.
Upon identification of the spilled material as an aromatic
compound, transport aircraft would be loaded with packaged
floating sorbents and dispatched to the release site.

Personnel at the spill site would maintain radio contact
with the aircraft and direct their drops to cover the
contaminant plume.1  If the aircraft arrived approximately
six hours after the spill occurred and the stream had
a one knot current and high diffusivities, the spill
would have moved approximately 11,000 meters downstream
and would be 580 meters long and the width of the stream.
While making low passes over the stream, the aircraft
would release a pattern of media packages over the affected
area.

Individual packages would sink to the bottom, disinte-
grate, and release the floatable carbon.  Particles of
carbon would disperse with the eddy currents and slowly
rise to the surface, sorbing benzene as it passed.

A screen skimmer or boom could be deployed downstream
to guide the spent carbon to a surface pumping device.
The collected media could then be stored for drying
and subsequent disposal or regeneration and reuse.  While
retrieval is best accomplished as soon after deployment
as possible, the carbon could be allowed to drift with
the current for several days while skimming devices
were located and put in place.

The sequential steps in treating a spill are illustrated
in Figures 1-6.
                           17

-------
                                                                * N.  —,
                  •A,
oo
                                            -
                                            FIGURE 1




                                   SPILL OF HAZARDOUS MATERIAL

-------
                                              . • •-
             FIGURE 2
     OFFICIALS ARE NOTIFIED.
RESPONSE TEAM IS SET INTO ACTION

-------
(0
                                           FIGURE 3

                             APPROPRIATE MEDIA  TYPE IS SELECTED
                               AND  LOADED ONTO  DELIVERY PLANE

-------
                        -






                                    •


                               •
   '
      -
/

                                                   '



                                       -
•
 .
         ,

                                                 •

                              .
                         FIGURE 4


        INDIVIDUAL MEDIA-PACKAGES ARE  AIR DROPPED
                  INTO CONTAMINANT PLUME

-------
to
to
                                               .-      -
                                        FIGURE 5
                           PACKAGES SINK AND BEGIN TO DECOMPOSE

                      MEDIA FLOATS TO SURFACE REMOVING CONTAMINANT

-------
U)
                                          FIGURE  6

                              STANDARD OIL SKIMMERS ARE USED
                                      TO RETRIEVE MEDIA

-------
                       SECTION V



                  LABORATORY STUDIES


FLOATING CARBON SORBENTS

Physical sorption results from the attractive forces that
are present at the interface of two phases.  The surface
of a solid will attract and hold molecules present in
either gases or liquids.  The amount held per unit area
of the surface is relatively small but it becomes a
significant fraction of the total mass of the solid
phase when the surface area per unit of mass is very
large.  The surface area of a good activated carbon, for
example, will be on the order of 1000 square meters or
more per gram of carbon.  Surface area values and iodine
numbers are listed for several commercial grade carbons
in Table 1.  The iodine number gives a general indication
of the efficiency of the carbon in adsorbing small mole-
cules .

                        TABLE 1

      Surface Areas and Iodine Numbers of Several
              Commercial Grade Carbons*4'5'


                                                 Iodine
                                 Surface         Number,
  Carbon        Supplier     Area (BET) m /g  mg^ 12/g Carbon

Darco        Atlas Chemical      600-650           650
             Industries

Piltrasorb   Calgon              950-1050          900
300          Corporation

Filtrasorb   Calgon             1000-1200         1000
400          Corporation

Nuchar WV-W  Westvaco              850             850
             Corporation

Nuchar WV-L  Westvaco             1000             950
             Corporation

Nuchar C-190 Westvaco              900         700-800
             Corporation
                            25

-------
A number of physical sorbents are available for removing
non-polar materials such as organic substances from water
but the best known and most widely used is activated
carbon.  An investigation of available activated carbons
with the desired floatability disclosed two commercial
granular carbons which resist wetting and remain floating
for long periods of time.  These are Nuchar C-190 and
Nuchar WA, both products of Westvaco.

Early in the program, some qualitative observations on
the floatability of various mesh sizes of Nuchar C-190
were made.  Samples of this carbon were slurried in water
and shaken for various time intervals.  After shaking,
the samples were allowed to stand for five minutes and
visual observations were used to estimate the quantity
which remained floating.  Table 2 summarizes the obser-
vations, which show that the finer mesh sizes are more
susceptible to wetting and sinking than the larger.  Due
to the vigorous mixing in these tests, the percentage of
floating carbon is probably lower than that which would
normally be anticipated in actual spill cleanup operations,
                        TABLE 2

          Approximate Floatability of Various
              Mesh Sizes of Nuchar C-190
Shaking Time
(hours)
Mesh Size
>20
20-30
30-50
50-100
100-200
200-325
1

100
95
95
95
90
80
2
Approximate
100
95
95
90
90
80
4
Percent
95
95
95
90
85
75
6
Floating
95
90
90
90
80
70
24

85
85
85
85
80
70
                           26

-------
Batch contact experiments were conducted to determine the
sorption isotherms for phenol using Nuchar C-190 activated
carbon.  The procedure involved mixing various quantities
of carbon with a standard solution of phenol (183 mg/1)
buffered at pH 6.5^6'.   Phenol is more strongly sorbed below
pH 7 and buffering is required to prevent a significant
change in pH.  Samples of 100 x 325 mesh carbon were
mixed with the phenol solution for 16 hours (to assure
equilibrium conditions)  after which the solution was
filtered and analyzed for total organic carbon.  Reagent
blanks with and without  carbon but containing no phenol
were run simultaneously  to correct for very small amounts
of extraneous organic matter or carbon which may have
penetrated the filter.   The quantity of phenol sorbed
by the carbon was determined by the difference in phenol
concentration before and after contact with the carbon.
The results of the experiments are illustrated as a
sorption isotherm in Figure 7.  The sorption isotherm
compares favorably with  data reported in the literature.

The relationship between mesh size and rise time was
investigated using a two inch diameter glass column and
a water depth of four feet.  Carbon was placed on the
water surface and the column was inverted.   The time
required for the bulk (-95 percent)  of the carbon to
rise through the water column to the surface was
recorded.  The results of these experiments are pre-
sented in Table 3.

                         TABLE  3

   Rise  Time for Nuchar  C-190 in Four Feet of Water
                                         Time  for  ~95 Percent
                                        of Carbon  to Surface
Mesh Size                                 (minutes)	

30-50                                            5

50-100                                          10

100-200                                         15

200-300                                         15
Various mesh sizes of Nuchar C-190 and Nuchar WA in both
wetted and dry  form were evaluated for their uptake of
phenol by floating the carbons through a  five foot
column of phenol solution  (185 mg/1) buffered at pH
                           27

-------
            1000
0
2
X
Q.
CT
O
CO
O£
•<
O
E
             100
00
              10
                            J	L
                                          10
100
1000
                                               SOLUTION PHENOL, mg/l
                                              FIGURE  7



                         SORPTION ISOTHERM  FOR PHENOL AND  NUCHAR  C-190

-------
 6.5.   The  carbon was  released  from a  bottle  at  the
 bottom of  the  column  at  an  application  ratio of 2000
 mg of  carbon per liter of solution.   The  results of
 these  experiments  are illustrated  in  Figure  8.   The dry
 carbon was prepared by oven drying at 100°C  for 16 hours
 and the wet carbon was contacted with water  for two
 hours  and  then filtered  to  remove  interstitial  water
 before use.  The time required  to  float approximately  95
 percent of the carbon through the  five  foot  column varied
 from 3 to  12 minutes  for the dry and  wet  30  x 50 mesh  carbon,
 respectively to 20 to 25 minutes for  tne dry and wet
 200 x  325 mesh carbon, respectively.  Nuchar C-190
 was selected for further study  on  the basis  of  its
 superior sorptive  ability and floatability.   The Nuchar
 C-190  was found to contain  a smaller  percentage of
 sinking granules than the Nuchar WA.  A mesh size of
 100 x  325 was  selected for  further study because phenol
 removal is near maximum with carbon particles in this
 size range.

 In order to assess the effectiveness  of Nuchar C-190 in
 sorbing other  hazardous organic materials, a  set of
 batch  removal  experiments was conducted.  Toluene, benzene,
 styrene and acrylonitrile were  selected for  study.  Sam-
 ples containing 50 microliters of each of these materials
 in 200 ml of water were contacted with 400 mg of 100 x
 325 mesh carbon and were shaken vigorously for one hour.
Each sample was allowed to stand overnight in a closed
container and  the aqueous phase was then analyzed for
TOC with the results given in Table 4.

                         TABLE  4

          Removal Effectiveness  for Nuchar C-190
              for Various Organic Materials
                       Initial          Final
                       TOC             TOC             Removal
 Material               (mg/1)           (mg/1)          Percent

 Toluene                 184              8               96

 Benzene                 198             12               94

 Styrene                 174              8.5             96

 Acrylonitrile            146             90               38
 Nuchar C-190 Dose = 1000 mg/1
                           29

-------
  100

o
   80
   60
   40
   20
    0
                                                            O DRY  CARBON

                                                            D WET  CARBON
                                                                        C-190
                                                 WA
                    DOSE                 2000 mg/lCARBON

                    PHENOL CONCENTRATION 185 mg/1

                    TEMPERATURE
                    pH                   6.5
                                                                      I
              30-50
              MESH
50-100
 MESH
100-200
 MESH
200-325
 MESH
                                   FIGURE 8
               COMPARISON OF C-190 WITH WA FOR REMOVAL OF PHENOL

-------
Two organophosphorus pesticides were similarly examined
with the results shown in Table 5.  In addition to TOC
measurements, phosphate analyses were conducted using
the perchloric acid digestion technique for total phos-
phate as described in Standard Methods(7).
                        TABLE 5

         Removal Effectiveness of Nuchar C-190
            for Organophosphorus Pesticides
Carbon
Dose
Material (g/1)
Diazinon


Mai a th ion


0
1
2
0
1
2
TOC
TOC
Concentration Removal
(mg/1) Percent
114
8
6
120
8
9
___
93.5
94.5
___
93.0
92.5
Total
P04
(mg/1)
21.4
2.8
1.4
20.0
0.6
1.0
PO4
Removal
Percent
___
87.0
93.0
_....
97.0
95.0
Diazinon

  Active  Ingredients

    0,0 diethyl  0-(2-isopropyl-4 methyl-
    6-pyrimidinyl) phosphorothioate

    Aromatic  Petroleum Derivative Solvents

  Inert Ingredients

Malathion

  Active  Ingredients

    0,0-dimethyl dithiophosphate of diethyl
    mercaptosuccinate

    Aromatic  Petroleum Derivative Solvents

  Inert Ingredients
16.75%

68.875%

14.375%
57%

33%

10%
                          31

-------
Larger scale experiments were conducted in a six inch
diameter, six foot deep column of tap water containing
200 mg/1 of (1) phenol, (2) an oil emulsion of Malathion,
and (3) an oil emulsion of Diazinon.  Removals of 86
percent, 82 percent, and 86 percent, respectively, were
obtained by floating 1000 mg/1 of 100 x 325 mesh carbon
through the water.  The percent removal of phenol and
pesticide did not vary significantly at various depths
through the column after flotation of the carbon  (see
Table 6).  Carbon containing sorbed phenol was observed
to remain floating for at least two days following
application.
                        TABLE 6

     Percentage Removal of Phenol, Malathion, and
           Diazinon at Various Depths in the
           Treatment Column After Flotation
                     of the Carbon
 Depth,
  ft.

   1

   2

   3

   4

Composite
Sample
Phenol
Removal,
Percent

  82

  86

  88

  87

  86
Malathion
Removal,
Percent

   79

   76

   82

   82

   82
Diazinon
Removal,
Percent

   90

   84
   86
Column experiments were also conducted with a more con-
centrated Diazinon mixture  (48 percent in xylene) added to the
six inch diameter column containing 28 liters of tap water.
In the first experiment, 4 ml of 48 percent Diazinon oil
emulsion were rapidly mixed with 8 ml of tap water to form
a stable emulsion.  The emulsion was then added to the
column followed by the introduction of a weighted bottle
containing 40 g of 40 x 325 mesh Nuchar C-190 which floated
through the diluted emulsion.  Diazinon removal in this
first experiment was 89 percent as determined on a total
phosphate basis.  A second experiment was conducted wherein
                            32

-------
both the Diazinon and carbon were  reduced  to  25 percent
of that used in the  first experiment.  The Diazinon
removal in the second experiment was 85 percent which
illustrates the higher removal efficiency  obtained at
higher concentrations by the same  proportion  of carbon.

Final laboratory work was aimed at identifying the best
means of producing large quantities of floating carbon
in the appropriate mesh sizes.  A  comparison  was made
between a sample of  ball-milled Nuchar C-190  and a
sample of Nuchar C-190 that had been hand  crushed on a
100 mesh screen.  When floated through a two  inch diameter,
four foot deep column of 200 mg/1  phenol solution, the
hand crushed carbon  removed 62 percent of  the phenol
compared to 57 percent removal by  the ball-milled carbon.
Both carbons were applied at the 1000 mg/1 level.  The
ball-milled carbon appeared to flocculate  to  a greater
degree than the hand crushed carbon when released from
the bottle at the bottom of the column of  phenol solution.
The greater degree of flocculation would tend to reduce
contact of the carbon with the phenol solution and thereby
reduce the amount of phenol sorbed from solution.

Two mesh sizes of Nuchar C-190 were commercially avail-
able:  (1) +30 mesh which is essentially 12 x 30 mesh,
and (2)  "unground" material which  is largely  14 x 325
mesh.  In view of the lower uptake of phenol by ball-
milled carbon it was decided to employ an  impact method
of crushing a large quantity of Nuchar C-190  for use in
a field demonstration.  A jet pulverizer, marketed by
Majac, Inc., Pittsburgh, Pennsylvania, was initially
considered for crushing the carbon.  This  device pulver-
izes granular material by impacting the particles in
opposing jet streams.  The +30 mesh Nuchar C-190 was
selected for crushing on the basis that the narrow mesh
size range would provide a higher  anticipated yield of
a 100 x 325 mesh size using the jet pulverizer.  On the
basis of prior performance with similar materials, a yield
of about 50 percent was anticipated.

A 4000 pound order of +30 mesh Nuchar C-190 was shipped
to a pilot plant facility for crushing.  Subsequent
testing with a jet pulverizer showed a yield of only 20
percent in the 100 x 325 mesh range.  Further testing
indicated that a similar mesh range could be obtained at
a higher proct ssing rate and lower cost with an air impact
pulverizer.  The bulk of the 4000 pounds of carbon was
then subjected to the impact pulverizer and air classi-
fied to give 2387 pounds of +325 mesh material which was
reported to contain  38 percent of  the desired 100 x 325
                           33

-------
mesh size.  Subsequent analysis of a sample of the +325
mesh fraction showed only 17 percent 100 x 325 mesh
carbon.  The mesh size used for the field demonstration
was therefore broadened to 40 x 325 mesh to give a
maximum recovery of 49 percent of the +325 mesh fraction
or 29 percent of the total carbon that was pulverized.
The actual recovery was 23 percent of the total prior
to breakdown of the sieving apparatus, which was used
to classify the carbon.  A total of 675 pounds of 40 x
325 mesh carbon was obtained which was supplemented with
170 pounds of 12 x 325 mesh carbon for the field demon-
stration.  As a result of the low yield of the desired
mesh size from processing +30 mesh Nuchar C-190, it is
believed that the "xmground" Nuchar C-190 may have been a
more desirable starting material.  Alternate sources of
floating activated carbon will have to be investigated for
future work because neither the unground nor +30 mesh
Nuchar C-190 are commercially available any longer.  Since
the completion of these studies, Westvaco has discontinued
sales of unground and +30 mesh Nuchar C-190 and will soon
discontinue sales of the same grades of Nuchar WA.  West-
vaco is continuing to market a powdered grade of Nuchar
C-190 which has a significant fraction of the carbon
particles in the +325 mesh size.  This and other potential
sources of floating activated carbon are discussed in
the Appendix.
FLOATING  ION EXCHANGE RESINS

Ion exchange resins are effective for removing and/or
neutralizing a broad spectrum of hazardous ionic substances
from water.  These substances include acids, bases, and
toxic salts where either the cation or the anion or both
are toxic to some degree.

The degree of sorption for a particular ion under equil-
ibrium conditions is largely a function of:  (1) the
concentration of that ion; <2) the concentration and
types of competing ions; and  (3) temperature.  A number
of different types of ion exchange resins are available
under the categories of cation exchangers or anion exchang-
ers.  The cation exchangers can be divided into different
types depending on the acid strength of the groups.
Strong acid groups such as -SO3 remain ionized even at
low pH and can function in acid solutions.

In contrast, weak acid groups such as -COO" are ionized
only at high pH.  At low pH they combine with H+ to form
undissociated -COOH and no longer act as fixed charges
                            34

-------
where an exchange of ions can take place.  For example,
a weakly acidic cation exchange resin would not be
effective  for removing ammonium ion from a solution at
pH 3 because of the resin's strong affinity for hydrogen
ion.  Under neutral to basic conditions, as in the case
of an ammonia spill, a weakly acidic cation exchange
resin could be effective for both neutralizing the basic-
ity and sorbing the ammonia.  The titration curve of a
weakly acidic resin is illustrated in Figure 9
along with that of a strongly acidic resin.  Using a
weakly acidic ion exchanger for treating a caustic
spill would have an advantage in that an overdose would
not cause an overreaction which would produce acidic
instead of neutral conditions.  Anion exchangers which
vary in base strength and act in a manner similar to
cation exchangers are also available.

           (8)
Boyd, e_t a_l    note that for organic zeolites in the
60 to 70 mesh range, up to 90 percent of the theoretical
capacity can be achieved within 30 seconds.  Kressman
and Kitchener(9) found phenolsulphonic resins reach
90 percent of capacity in two to nine minutes with
simple cations.  Helfferich(1Q) makes similar obser-
vations for strong acid and base exchanges.  At the
beginning of this program it was postulated that
adjusting the rise rate of the selected media to provide
the several minutes required for utilizing the major
portion of the media's capacity should be possible.

The first attempts to produce floating ion exchange beads
employed suspension polymerization techniques.  Preliminary
experimentation centered around the inclusion of hollow
glass microspheres (Eccospheres ® manufactured by Emerson
and Gumming Inc., Canton, Massachusetts) in the resin matrix.
Standard styrene-divinylbenzene ion exchange resin beads
are generally prepared by polymerizing a mixture of
styrene and divinylbenzene in a stirred aqueous suspension
containing a suspension stabilizer (e.g., starch, gelatin,
polyvinyl alcohol).  The size of the resin beads produced
is determined by the stirring rate.  Polymerization
usually requires several hours after the addition of a
catalyst (e.g., 1 percent benzoyl peroxide).  Ion exchange
groups are normally inserted into the resin matrix after
preparation of the beads.

Initial attempts to include the hollow microspheres in
the monomer phase during aqueous-suspension polymeriza-
tion with styrene-divinylbenzene mixtures were not
successful.  Though mixed with the monomer phase prior
to suspension, the glass micros-pheres quickly transferred
                           35

-------
                         12
                         10  -
                          8  -
                                                               WEAKLY ACIDIC
                                                                  RESIN

-------
to the aqueous phase when the phases were mixed.  This can
be attributed to hydrophilic characteristics on the
surfaces of the microspheres.  Treatment of the beads
with silicone formulations failed to alter the hydro-
philic characteristics sufficiently to favor the organic
phase.

Parallel efforts were directed to formulation of resins
via bulk polymerization techniques.  Here, the glass
ndcrospheres are stirred into the catalyzed monomer
as it begins to set, resulting in the formation of a
solid mass of resin with individual microspheres trapped
throughout the resin matrix.  The resin mass is then
fractured and sieved to obtain the mesh size desired.
This technique showed promise with soft aqueous gels of
weak acid acrylic polymer resins, but did not produce
satisfactory yields of small mesh sizes with hard resin
masses.  Excessive shattering of the glass microspheres
occurred when crushing the hard resin masses to produce
small grain sizes.

An initial attempt was made to produce a resin with acry-
lamide crosslinked with N, N'-methylenebisacrylamide.
These two organic monomers are water soluble and can be
readily polymerized at room temperature.  A floating ion
exchange resin was prepared by mixing hdllow glass micro-
spheres in the monomers prior to polymerization.  (Hy-
drolysis converts the amide group in the resin structure
to carboxyl groups with ion exchange capability).
Unfortunately, the product beads were found to be
structurally weak and consequently unsatisfactory.

Two types of resin were successfully prepared and
selected for further study in preparation and evaluation
of floating ion exchangers.  These were (1) a weak acidic
carboxylic resin prepared from acrylic acid and a suit-
able crosslinking agent using bulk polymerization tech-
niques which included a soft gel intermediate, and (2)
a weakly basic epoxypolyamine resin prepared from
epichlorohydrin and polyethyleneimine using suspension
polymerization.

The acrylic resin granules were produced in a modified
bulk polymerization technique from acrylic acid cross-
linked with ten percent ethylene glycol dimethyacrylate.
The hollow glass microspheres were incorporated into the
resin matrix by trapping an aqueous mixture of the monomers
and microspheres in the interstices of a packed bed of
small glass beads or other granular material (e.g., gypsum)
prior to polymerization.  After the monomers polymerized,
                           37

-------
 the  solid mass  of resin was crushed and the usable resin
 granules  separated by filtration with water.   The grain
 size and  shape  of the bed material  dictates the size of
 the  resin granule produced.  An alternative method of
 preparation  was used  which employed a soluble salt (e.g.,
 NaCl)  rather than insoluble granules to trap  resin monomers
 and  hollow microspheres together for polymerization.
 This method  is  more adaptable to production of small
 resin particles.   The salt can be removed  to  a large degree
 by dissolution,  thus  avoiding excessive crushing which
 tends to  fracture the glass microspheres.   Some crushing,
 however,  is  still required which increases  the number of
 hollow microspheres required.   Microspheres constitute
 the  greatest single cost  for production of  floating
 resins.

 Weak acid acrylic ion exchange resins have  very high
 exchange  capacities due to the relatively  large number
 of carboxylic  (-COOH)  groups on the  resin matrix.   An
 example of the  resin  structure of polymerized  methacrylic
 acid crosslinked  with divinylbenzene is illustrated in
 Figure 10.   The  H+' on a COOH group  is exchangeable  to a
 large  degree  with other cations providing that the  solution
 pH does not  fall  below a minimum level  (5-7).   This
 ability to act  as  a buffer is  desirable since  an excess
 of the resin  applied  to counteract a caustic spill  will
 not  result in "salt splitting" to produce acids which may
 cause  as  much difficulty as  the spilled caustic.  For
 example,  a carboxylic exchanger has  a high  affinity  for
 hydrogen  ion with  the result that this  resin displaces
 sodium ion from  a  solution of  sodium chloride  with  great
 difficulty due  to  the formation of  a strong acid, HC1,
 as one of the reaction products.  Thus,  the equilibrium
 is shifted far  to  the left as  illustrated by the  following
 reaction:
          RCOOH + NaCl      	  RCOONa + HC1

For the case of reaction with sodium hydroxide, the
equilibrium shifts far to the right since water is the
reaction product rather than acid:
          RCOOH + NaOH  	      RCOONa + H20

The resin illustrated in Figure 10 is the standard weak-
acid acrylic resin of commerce  (Amberlite IRC-50), which
is made from methacrylic ester and divinylbenzene by the
aqueous suspension polymerization process.  This resin has
an ultimate exchange capacity  (at high pH) of 10 milli-
equivalents per gram of dry resin.  At pH 7, the capacity
is about eight milliequivalents per gram of dry resin.
                           38

-------
\D
        COOH


         CH — CH.
        COOH
COOH
          COOH
        COOH


         CH—CH;
   CH — CH—CH	CH — CH — CH
     COOH


•CH?	CH—CH2-
 COOH


-CH—CH2—CH-
        COOH

        \
—CH.,	CH—CH;
 COOH


-CH—CH2—CH
                •CH.
                      COOH


                     -CH	CH;
        COOH       COOH       COOH

        \          \         \
         CH—CH2	CH — CH-	CH
 COOH


-CH — CH
                COOH


                -CH — CH2—CH
                                        COOH


                                       -CH—CH;
                                COOH      COOH

                                \        \
                                -CH—CH—CH
                                                       COOH


                                                        CH
                       COOH


                        CH
                                             FIGURE  10


                  POLYMERIZED  ACRYLIC ACID CROSSLINKED WITH DIVINYLBENZENE

-------
The ultimate exchange capacity of the floating acrylic
resin, prepared in the laboratory from acrylic acid
crosslinked with 10 percent ethylene glycol dimethacrylate
was found to be eight milliequivalents per dry gram
(including the weight of the hollow microspheres).  The
structure of this resin is illustrated in Figure  11.  Two
different batches of floating resin, 20 x 50 mesh and
25 x 70 mesh, were used to determine the uptake of alkali
from a 200 mg/1 solution of NaOH in distilled water.
Six grams of resin of each mesh size were released at the
bottom of a four foot by two inch column of 200 mg/1 NaOH
solution; the uptake of alkali by the 25 x 70 mesh resin
was determined to be 82 percent as compared to 66 percent
removal by the coarser 20 x 50 mesh resin.  The 82 percent
removal represents about 25 percent utilization of the
ultimate exchange capacity of the resin.  Uptake of alkali
from a dilute distilled water solution is expected to be
somewhat slower than that for a solution containing
neutral salts.  Natural waters found in lakes and streams
will contain varying concentrations of neutral salts
which should increase the efficiency of alkali removal by
this resin.  A titration curve for the floating acrylic
resin is given in Figure 12.

Samples of floating epoxyamine anion exchange resin were
prepared for evaluation in the laboratory.  The formula
used consisted of a 3:1 ratio of epichlorohydrin to
tetraethylene pentamine.  Epichlorohydrin was partially
reacted with tetraethylene pentamine in an ice bath to
form an aqueous syrup containing 49 percent water.  This
syrup was then allowed to reach room temperature and
remain there for 45 minutes or more.  A dilute sodium
hydroxide solution was added as a catalyst and hollow
glass microspheres were mixed into the syrup at a ratio
of two cubic centimeters of microspheres to one gram of
epichlorohydrin and tetraethylene pentamine.  The mixture
was then added to rapidly mixed hot mineral oil to
complete the polymerization and produce resin beads.
The mineral oil contained a small amount  (1 percent) of
turkey red oil (sulfated castor oil) to aid in reducing
the size of the aqueous globules prior to polymerization.
Initial batches of resin beads were generally larger in
mesh size than desired.  Subsequent batches were prepared
in smaller mesh sizes by increasing both the mixing
speed and the oil to syrup ratio from about 10:1 to about
50:1.  The structure of an epoxypolyamine anion exchange
resin is shown in Figure 13.
                            40

-------
COOH      CH

 CH— CH2 - C

           C
           t
COOH
\
     COOH       COOH     COOH

CH? - CH— CH2 - CH— C^— CH -
                                                   COOH
                                                   COOH
                                                  CH-
                             — CH — CH — CH— CH— C
           r*
           0
COOH       C = 0     COOH
\          I         \
 CH^^~ CH^^^~ C ~^~™ CH»" "  Cn^^~ Cn^
           CH,
          COOH
          \
                              C\H>
           CH — CH2—C
                                                   COOH

                                                    CH — CH;
                    1
                                = 0
                                                                                    COOH

                                                                               CH - CH
COOH
\
COOH
\
                               COOH
                               \
                                                                   COOH

                                                                   -CH
 CH — CH2 - CH - CHg- - CH — O^ - CH — CHj—

                                         CH3

                                    FIGURE 11
r*
0
1
c=o
1
C 	
1


COOH
\
CH2 	 CH — C


COOH
\
H2 	 CH 	 C


COOH
\
Hg— CH 	


COOH

— CHj 	 CH
POLYMERIZED ACRYLIC  ACID CROSSLINKED  WITH  ETHYLENE GLYCOL  DIMETHACRYLATE

-------
10
                     10
                                I         I        1         1	I         I
                       024        6        8        10       12
                                  HILLIEQUIVALENTS  OF ALKALI ADDED/GRAM OF RESIN

                                             FIGURE  12

                   TITRATION  CURVE FOR FLOATING ACRYLIC  ION EXCHANGE  RESIN

                                     Titration  in 0.2 M NaCl

-------

1
N 	 CH2 	
CH — OH
1

N — CH2—
CH0
I2
CH — OH
j
f2
N— CH2—
CH,
I2
CH — OH
1

-CH — CH2— N
CH,
I2
OH CH — OH

CH
                          FIGURE 13




EXAMPLE OF STRUCTURE OF AN EPOXYPOLYAMINE ANION EXCHANGE RESIN

-------
 The  floating anion  exchange  resin  prepared in the manner
 described  above  was found  to have  an  ion  exchange capacity
 of  four  milliequivalents per gram  of  resin.   A titration
 curve  for  the resin is  given in  Figure  14.   The resin
 beads  used for evaluation  were largely  in the 30 to  50
 mesh size  with 12 percent  +30 mesh and  9  percent -50
 mesh.  When released from  an ice cake to  float up through
 a four foot column  of water  containing  183  mg/1 HC1  and
 200  mg/1 NaCl, the  resin removed 61 percent of the acid
 at an  application ratio of two grams  resin  per liter of
 water.   This removal is somewhat less than  the removal of
 NaOH by  floating acrylic cation  exchange  resin under
 comparable conditions;  this  is believed due  to the smaller
 ion  exchange capacity of the epoxyamine resin (4  meq/g vs
 8 meq/g).   In both  cases,  however,  only a small fraction
 of the available ion exchange capacity  is utilized.   A
 smaller  mesh size is considered  to be more  important in
 increasing the removal efficiency  than  adjusting the
 resin  bead density  to a level just under  that of  water.
 Resin bead density  cannot  be made  too nearly  that of
 water since the  density of the bead will  increase as it
 adsorbs  ions  from the water  to replace  either H+  or  OH-
 ions.  Thus,  the resin bead  rise rate will  decrease  as
 the  exchange  process proceeds.   If the  resin  bead density
were initially very  close  to that  of  water,  the bead
 might initially  rise and then sink as heavier ions are
 exchanged  for H+ or  OH~.

 Following  the  laboratory development  phase, Diamond
 Shamrock Chemical Company was engaged to  investigate the
 possibility of preparing several hundred  pounds of floating
 epoxyamine  resin in  their  facilities  at Redwood City,
 California.   Diamond Shamrock manufactures both a weak
base epoxyamine  resin (Duolite A30T)  and  an intermediate
base resin  (Duolite  A30B).   These  anion exchange  resins
are reported  to  have ion exchange  capacities  of  8-9  meq/
gram.  It  was  considered that except  for  the  addition of
the hollow  glass microspheres, the  technique  of manufacture
of the epoxyamine resin would not  be  drastically  altered
 from standard production techniques.  The major problem
lay in the  mixing of the microspheres in  the  syrup just
prior to introduction into the oil.

Diamond Shamrock was successful  in  preparing  a  floating
epoxypolyamine resin both  in  their  laboratory and sub-
sequently with their commercial production equipment.  A
sample of  laboratory prepared epoxypolyamine  floating beads
was delivered  to Battelle  for evaluation  prior  to pro-
duction of  a  large batch.   Figure  15  shows the  titration
                           44

-------
12


10


 8


 6


 4


 2


 0
I
I
I
_L
             1234

        MILLIEQUIVALENTS HC1 ADDED/GRAM DRY RESIN


                      FIGURE 14

         TITRATION CURVE FOR FLOATING EPOXY-
            POLYAMINE ANION  EXCHANGE  RESIN

                Titration in 0.2  M NaCl

-------
curve for this sample.  The exchange capacity for this
resin was approximately 70 percent greater than that for
the resin previously produced.  The beads also appeared
to have greater structural integrity.

Once the acceptability of the floating resin was substan-
tiated, Battelle contracted with Diamond Shamrock to
produce a single lot of 14 cubic feet for use in a field
demonstration.  The resulting product had the same physical
appearance as the earlier sample but showed a greater mesh
size dispersion favoring the fraction of larger particle
size.  The larger mesh size of the resin caused a reduced
rate of exchange such that only 37 percent of the acid in
a 0.005 N HC1 solution was removed at 2000 mg/1 resin
application.  This compares with 53 percent removal under
the same conditions with a -30 mesh fraction of resin.
It is felt that these size deviations will not be a
problem in commercial production once preparation tech-
niques are refined.

Since it is probable that higher concentrations of acid
than 0.005 N will be encountered under actual field
conditions, an experiment was conducted to measure the
uptake of sulfuric acid from a0.025N solution with 10
g/1 of applied floating epoxypolyamine resin.  The re-
moval of acid at this higher level was 94 percent, which
is believed to be due to the more rapid diffusion of
acid from the solution to the surface of the resin bead
since both the acid and the resin were increased by a
factor of five.  Film diffusion is usually rate limiting
for removal of constituents from a. dilute solution by ion
exchange, whereas particle diffusion becomes rate limiting
at the higher concentrations.  The higher rate of uptake
from concentrated solutions again emphasizes the import-
ance of rapid response to treatment of spills.

While both the bulk polymerized acrylic resin and the
suspension polymerized epoxypolyamine resin were success-
fully produced and demonstrated to be effective for
neutralizing contaminated waters when allowed to float
up through a water column, suspension polymerization
is the preferred production technique.
                           46

-------
PH
    10
                   MILLIEQUIVALENTS H2$04 ADDED/GRAM DRY RESIN




                              FIGURE  15



TITRATION CURVE FOR DIAMOND SHAMROCK WEAK BASE EPOXYPOLYAMINE RESIN

-------
Bulk techniques prove to be far more expensive because
of the associated losses of microspheres.  Although no
suspension polymerized cationic resins were produced
during the program, there is no reason to believe that
acrylic resins could not be so produced if acrylic acid-
dissolved ethylene glycol dimethacrylate or water soluble
sodium acrylate were employed as reagents.  Results in
the laboratory have demonstrated that the microspheres
will remain in the monomer phase, but production techniques
were not investigated for the length of time needed to
derive a procedure which would produce beads with sufficient
strength.  The approach still shows great promise.

Regardless of the production technique selected, two
considerations remain paramount: 1) small bead sizes are
always preferred because of the accelerated exchange
kinetics; and 2) resins perform much better in concentrated
spills than in dilute ones.

Photomicrographs which show the hollow glass microspheres
incorporated in the resin matrices of a floating acrylic
resin and a floating epoxypolyamine resin are presented
in Figures 16 and 17, respectively.

APPLICATION METHODS

Mechanical subsurface injection by means of slurry pumps
installed on ships is probably the most desirable method
of delivery in areas such as ports and harbors where
equipment of this nature is readily available.  However,
in remote areas rapid transport of subsurface injection
equipment may not be feasible.  Therefore, techniques
which lend themselves to air transport and injection of
the floating sorbents were selected for development in
this program.

Air delivery would involve "bombing" the affected area
with packages containing the sorption media.  In addition
to the active media each package unit must contain an
appropriate amount of ballast so that it will sink to the
bottom of a waterway prior to release of the media.  The
optimal size of an individual package is dependent upon
two separate considerations:  1) achieving uniform dis-
tribution of the packages  (and thus the media) over a
spill area, and 2) overall economics of the concept.
While the former consideration tends to favor smaller
packages, the latter biases the decision toward larger
units.
                           48

-------
      Microspheres
           4
                 FIGURE  16
ACRYLIC FLOATING CATIONIC  EXCHANGE BEADS

-------
01
o
                                           FIGURE  17

                       EPOXYPOLYAMINE FLOATING ANION1C  EXCHANGE BEADS

-------
Ballast requirements can be determined from the bulk
density of the media employed.  Media which must be
delivered in a dry state will obviously require both
larger packages and greater amounts of ballast than media
which can be delivered wet.  Table 7 contains data on the
measured densities of various media and subsequent ballast
and package size requirements.  Cement ballast with a
bulk density of 180 pounds per cubic foot was assumed for
the calculation of package size.
                             TABLE  7
    Ballast and Package Size Requirements for Selected Media
Media

Furnace
Dried Carbon

Air Dried
Carbon

Wet Carbon

Dry Epoxy
Resin

Wet Epoxy
Resin
           Required
Density    Ballast
(Ib/ft3)   (Ib/lb Media)
   8


  12

  58.7


  15.7


  56.0
11.5


 7.9

 0.21


 5.1


 0.26
            Size of Package
            Containing
            One Pound of Media,
                (ft3)
0.188


0.127

0.018


0.092


0.019
                            51

-------
It is evident that potential packaging techniques  for
the floating media and ballast for aerial delivery fall
into two categories:   (1) methods employing retrievable
containers, and  (2) methods employing non-retrievable
containers or containers which will decompose in water
over a period of time.  Of prime importance in the program
has been the identification of a packaging technique which
would not result in significant degradation of water quality
if the packages were used in spill response activities.
Packaging techniques which were considered include plastic
containers, soluble synthetic films, unfired clay containers,
and ice cakes.
Plastic Bottles
Plastic bottles are an inexpensive/ standardized container
form potentially useful for packaging floating media.

Plastic containers were employed in column tests to
evaluate the effects of various bottle mouth widths on
media release rate and release patterns.  It was observed
that media release from wide mouthed bottles was accomp-
anied by simultaneous release of quantities of large air
bubbles.  These air bubbles tended to convey clusters of
media rapidly to the surface, thus preventing dispersion
of the media in the water column.  Further tests indicated
that this problem could be virtually eliminated by using
containers with walls which sloped upward to a mouth of
about one inch diameter.  Using this type of container,
there was a short period of media release followed by a
longer period in which small air bubbles escaped but
little media was emitted.  This venting period preceded a
sustained release of clouds of media from the bottle which
distributed the media throughout the water column.

Either cement or sand and gravel proved satisfactory
as ballast in partially filled plastic bottles.  One
gallon polyethylene bottles half-filled with cement were
found to withstand the impact of a fall from a fifty foot
height to a gravel bed.  Similarly, these containers
withstood drops from a 100 foot height into water.

It was determined that the mouths of the plastic bottles
could be sealed with a water soluble material, Quicksol
A, to prevent the media from falling from the containers
during the air drop.  However, in field trials with
carbon filled containers, it was observed that no signif-
icant quantity of media was spilled during a drop from
bottles with open mouths.
                           52

-------
Figure  18 is  a schematic diagram of a plastic bottle of
the type used in the field demonstration containing
ballast and floating carbon.

The major drawback associated with plastic containers is
the potential aesthetic and/or ecological damage which
might result  if weighted plastic bottles were allowed to
remain permanently in a waterway after treatment of a
spill.

Retrieval of plastic containers could be accomplished
either manually  (e.g., by divers or with grappling hooks)
or by use of containers which will surface and float
some time after release of the media, thus permitting
surface collection.  A limited laboratory study showed
that cement ballast could be attached to polyethylene
bottles with water soluble connections.  After several
hours' exposure to water these connections dissolved,
causing the ballast to separate and allowing the bottles
to rise to the surface.  However, this or some other tech-
nique for producing containers amenable to surface
collection would involve significant fabrication costs
for the containers.  In fact, use of commercially available
plastic containers and subsequent manual retrieval from
the bottom of a waterway probably would represent a cheaper
approach.  If plastic containers are to be given further
consideration as a potential packaging technique, a
thorough economic evaluation will be required in order
to determine the most viable alternative from a cost
standpoint.

Another approach which was considered but which was not
examined in great detail in this program is the use of
biodegradable containers.  If a container could be
identified which would eventually degrade without creating
a significant oxygen demand or other pollution problems,
it might offer an attractive alternative.  Since such
materials are still in the experimental development
stage, an adequate technical and economic assessment
of the feasibility of this approach cannot be made at
this time.

Soluble Films

Contacts with several major pharmaceutical firms indi-
cated that both the technology and equipment are presently
available for large scale encapsulation of a mixture of
ballast and floating media in water soluble capsules.
                          53

-------
             CONCRETE BALLAST
                FIGURE 18




SCHEMATIC OF CARBON FILLED PLASTIC BOTTLE





                   54

-------
Several samples of water soluble films and bottles were
obtained from vendors and subjected to a cursory exam-
ination to determine their suitability for containing
floating media.  Individual packets with volumes of
approximately four cubic inches were prepared from each
material and loaded with sand ballast and polystyrene
chips to simulate media "bombs".  Each packet was then
immersed in four gallons of tap water (@25°C) to determine
the time required for release of the'polystyrene chips.
Results of this test and pertinent product data are
given in Table 8.

Most of the films dissolved uniformly and released the
polystyrene chips cleanly with little or no partially
dissolved film adhering to them.  There was some tendency
for the chips to stick to the gelatin capsules and Klucel
bottles which dissolved much more slowly.  The immersion
time required for release could be adjusted by using
thicker films or multiple layers of film.

One pound of polymer will yield about 7500 square inches
of film with a thickness of three mils.   If individual
packets with volumes of four cubic inches each were fab-
ricated, a pound of polymer would provide containment
for a volume of approximately 1200 cubic inches.  The
weight of dry carbon thus packaged would be approximately
five pounds or a 5:1 ratio of carbon to soluble film.
This quantity of soluble film would be equal to or perhaps
greater than the spilled organic material being removed
from the water.

Experiments were conducted to study the effects resulting
from dissolution of one of the films,  Quicksol A.  It
was determined that 160 mg of Quicksol A dissolved in water
yielded a total organic carbon concentration of 88 mg/1.
Activated carbon at 2000 mg/1 removed 47 percent of this
TOC.  The same dose of activated carbon removed 91 percent
of the 171 mg/1 phenol solution, but could adsorb only 75
percent of a phenol-Quicksol A mixture.   The dissolved
film showed a definite tendency to use up available carbon
sorption capacity.  Biochemical oxygen demand (BOD) tests
indicate that a 1000 mg/1 solution of Quicksol P has a
five day BOD of only 8.2 mg/1.  Although the use of these
films should not cause excessive oxygen depletion at a
spill site, the addition of relatively large quantities
of soluble organic matter to the water is a questionable
approach.  In addition, the tendency of the dissolved
films to use up a significant portion of the available
sorption capacity of activated carbon is not a desirable
characteristic.
                          55

-------
in
                                           TABLE  8

                             Product Evaluation of Soluble Films

Trade
Name
Quicksol A(a)

Quicksol Pfa)

Quicksol P-U(aJ
Edisol-M(a)


Klucellb)





Polymer
Type
polyvinyl
alcohol
polyethylene
oxide
urethane
hydroxypropyl
cellulose
(food grade)
hydroxypropyl
cellulose
(food grade)
PH a Y"fTia f*& 11+" 1 (^fl 1
* net .L iiiaw^ u i. ^^cu.
gelatin capsule

Thickness
(mil)
1.25

2.0

2.25
3.0


10


12
^> &
Release
Time
(minutes)
0.5

0.25

0.75
1.25


30-50


7 Af\
£*t\J

Cost
($/lb in 1000 Ib lots)
1.94

1.94

2.00
2.22


1.75
-------
Clay Containers

In an attempt to find a packaging agent which could also
serve as the ballast material, an effort was initiated
to study the feasibility of unfired clay containers.  A
local pottery shop was engaged to produce various clay
container shapes and sizes through a slip molding technique.
The resulting products were dried but were not fired so
that they displayed an affinity for water which soon
caused them to collapse when immersed.

It was found that sand could be added to the clay in
varying amounts to increase the speed with which the
package lost integrity.  Inclusion of approximately 30
percent sand resulted in a decomposition time of 2 to 5
minutes.  Pyramidal and cylindrical shapes displayed the
cleanest release characteristics, trapping little of the
floating media under the collapsed clay.

An impact evaluation was made to determine whether clay
containers could withstand the forces encountered when
dropped into water.  Vessels shaped somewhat like artillery
shells  (Figure 19) with a one quart capacity (approximately
1 ft. high x 3 in. O.D.) were dropped from a height of
100 feet into the Columbia River.  Each clay container was
attached to 185 feet of light cord so that it could be
retrieved and inspected.  Vessels with a wall thickness of
0.25 inch or greater withstood the impact.  Those having wall
thicknesses of 0.125 inch or less collapsed.  Several align-
ment modifications could potentially increase the effective-
ness of the clay cylinders.  Walls could be tapered so that
heavy sections would absorb the impact while thinner walls
at the top would rapidly disintegrate when immersed in water.
Alternatively, lids could be loosely molded into place so
that soon after coming to rest on the bottom of the water
body they would fall off, allowing the media to escape.  It
might also be desirable to add fins to the design since the
models tested continued to tumble when initially given
angular momentum.

Clay containers would require use of dry media and would
thus have a low payload to ballast ratio.  A further dis-
advantage of unfired clay is the turbidity caused by
suspension of the fine clay particles in water.  The mixing
                         57

-------
          FIGURE 19




PHOTOGRAPH OF CLAY CONTAINER
             58

-------
action caused by the release of air and media suspends
a significant amount of clay in the water.  Use of a
fired clay container with an unfired clay seal is a
possible alternative to minimize the amount of suspended
clay formed.
Ice Cakes

The search for a completely innocuous encapsulation agent
led to an attractive possibility—the use of ice as a
binding agent.  Ballast and media could either be layered
and frozen into discrete packages for aerial distribution
or intimately mixed and frozen into blocks which could
be shattered into any size desired at the time of release.

Advance production and stockpiling of media ice cakes
would have associated high storage costs since refrig-
erated storage would be required.  In addition, sub-
limation might be a problem associated with long term
storage of ice cakes.  One possible solution to these
drawbacks would be dry storage of components (media and
ballast) and subsequent quick freezing when use of the
media is required.  Transport of the ice cakes might not
involve too many difficulties since low temperatures
prevail in the higher altitude flying lanes.

Column evaluations showed media release from ice packages
to have several excellent characteristics.  Media release
begins almost immediately after the ice cake reaches the
bottom of the water column and the media soon distributes
itself evenly throughout the water column.  Release of the
media is spread out over the period of time required for
the ice to melt and thus the release is smooth.  Problems
with air bubbles are avoided and ballast requirements
are much less than when dry media is utilized.

Freezing the media in an ice matrix was found to have no
debilitating effects on the performance of floating ion
exchange resins.  Packaging in ice did, however, appear
to result in some loss of capacity in the case of the
floating carbon.  Parallel column removal tests run with
a standard phenol solution showed that unfrozen carbon
removed 65 percent of the phenol while an equal quantity
of carbon frozen in an ice cake achieved only 51 percent
phenol removal.  Removal differences fluctuated in sub-
sequent tests, but the carbon ice cakes were always found
to be somewhat less effective than unfrozen carbon.
                           59

-------
Parallel column runs with unfrozen carbon at solution
temperatures of 0°C and 17°C showed little difference
in phenol removal efficiencies  (75 percent at the lower
temperature and 77 percent at the higher) .  This ob-
servation .led to the conclusion that the reduced
efficiency -of the carbon frozen in ice cakes was not a
function of temperature so much as the result of inter-
actions between the ice and the carbon.  It was noted
that finer carbon particles appeared to floe when
released from the ice cake.  Flocculation of carbon
particles was not observed in release from plastic
bottles.  This flocculation effect could reduce the
actual coverage of the media and hence reduce removal
efficiency.  In addition, it is possible that residual
ice crystals on the carbon may interfere with surface
forces and hence reduce sorption.

In deference to the major objectives of the program,
these apparent interactions were not fully explored.
Because of the lower ballast requirements and superior
dispersion of the carbon near the bottom of the stream
or lake, the carbon ice cake approach appears preferable
in spite of the somewhat lower sorption capacity or rate.
However, further work is needed to establish the feasi-
bility of the carbon ice cake packaging concept.  Clearly,
there is great promise-in the use of ice with floating
ion exchange resins.
                           60

-------
                       SECTION  VI
                 FIELD DEMONSTRATIONS
 In order to  test  the effectiveness of  the  floating media
 concept on the  scale and under  the conditions of  actual
 hazardous materials spills,  field demonstrations  were
 conducted using both the floating epoxypolyamine  anion
 exchange resin  and the  floating Nuchar C-190 carbon.
 These demonstrations concerned  application of the floating
 media concept to  static or impounded water only.  Treat-
 ment of flowing streams may  involve different techniques
 because of the  turbulence of the water.
DIAZINON SPILL

A 10 million gallon water storage basin was selected for
the spill treatment demonstration.  The basin was filled with
Columbia River water to a depth of 12 feet and calibrated
ropes were placed across the western end of the 200 x 400
foot structure as illustrated in Figure 20.

A commercial grade of the organophosphorus pesticide,
Diazinon, was selected for the spill.  Diazinon is highly
toxic to aquatic life forms and hence causes major concern
when spilled.  It is not persistent beyond several weeks,
however, so immediate complete cleanup by the floating
carbon was not required in the event discharge of the basin
water was required at some future date.  The composition
of the 48 percent Diazinon in an emulsifiable solution is
given in Table 9.  Seventy-eight pounds of the organic
solution emulsified in 160 pounds of water was employed
for the spill.

                        TABLE  9

     Composition of Emulsifiable Diazinon Solution


Ingredient                                      Percent

0,0-diethyl 0-(2-isopropy1-4-methyl-               48
  6-pyrimidinyl) phosphorothioate

Xylene                                            36

Inert  Ingredients                                 16
                           61

-------
10
                                                             400'-
                   I
                   -N-
 SPENT CARBON
COLLECTION TANK
                 CALIBRATED
                   ROPES
                              INITIAL SPILL  ZONE
                                 400 SO.  FT.
                              COLLECTION
                                 PIPE
                         DIAPHRAGM PUMP
                                                    BUOY
                                                     DEPLOYED
                                                       BOOM
                                                         HATER STORAGE BASIN
                                                                                                       200'
                                                    FIGURE  20

                                 LOCATION OP FACILITIES  FOR DIAZINON SPILL

-------
The Diazinon was spilled at 0700 by spreading the 25
gallons of emulsion from a rowboat in a 20 foot by 20
foot square as illustrated in Figure 20.  Aerial photo-
graphs of the spill taken at 0715 are given in Figures
21-22.  Sampling of the spill area began at 0730.  Samples
were taken with ten foot aluminum tubes connected by
flexible tubing to peristaltic pumps.  One liter glass
bottles were filled with water withdrawn from 1, 5, and
10 foot depths.  Sample lines were thoroughly flushed
between sampling points.

A total of 845 pounds of Nuchar C-190 activated carbon was
employed in treating the spill which is about a 10:1 ratio
of carbon to Diazinon.  The carbon included 675 pounds of
40 x 325 mesh carbon and 170 pounds of 12 x 325 mesh
carbon packaged in 1,128 one gallon polyethylene bottles
each of which contained 1/2 gallon of carbon (3/4 pound)
and 1/2 gallon of concrete (8 pounds).

Bottles of carbon and ballast were dropped into the spill
area from a helicopter equipped with a platform suspended
beneath the helicopter with a sling.  Servo-motors mounted
on a drop gate of the platform were wired for control from
within the helicopter.  The slings were fixed with an offset
harness so that under normal flight conditions they would
hang at an oblique angle with 60 carbon-filled bottles
(540 pounds) resting against the drop gate.  Over the drop
site the gate was opened and the bottles were allowed to
slide down into the water.  Figure 23 shows the platform
in place beneath the helicopter.  Two platforms were employed
so that one could be loaded while the other was in use.

The actual treatment phase of the demonstration was begun
at 0810 when the first of 17 loads of carbon were dropped
into the spill area as shown in Figure 24.  Flags were used
to mark the drop lane and a flagman was employed to signal
the helicopter pilot to release the gate.  In this manner
drops were distributed evenly across the contaminant plume.
Due to a failure in the gate control of both platforms, the
18th and final load could not be dropped from the helicopter.
These bottles were subsequently distributed by hand from the
side of the basin and from a rowboat.  The application work
was completed by 1100; however more than half the carbon was
dropped by 0915.  Intermittent failure of the tailgate
release on the platforms increased the time between subse-
quent drops after 0915.  Three of the 1128 bottles contained
insufficient ballast and consequently floated to the surface
without releasing their carbon.
                            63

-------


             FIGURE 21
AERIAL  PHOTOGRAPH OF  DIAZINON SPILL

-------
en
Ul
                                            FIGURE  22
                          CLOSEUP  OF  DIAZINON  SPILL  IN SAMPLING GRID

-------
cr>

                                             FIGURE 23
                               PHOTOGRAPH OF HELICOPTER WITH SLING

-------



                 FIGURE  24




PHOTOGRAPH OF AIR  DROP INTO SPILL AREA

-------
By 1000 carbon was very evident in the surface waters around
the spill.  Many portions of the contained area were too dark
to allow observance of objects more than six inches beneath
the surface.  A thin oil slick was also visible on the water
since, as previously mentioned, prior to this field demonstra-
tion the basin was used for numerous oil spill contaminant
demonstrations and residue oils remained behind after this
work.  As the bottles disturbed the water and moved down
through the weeds on the bottom of the pond, globules of
this residual oil were freed to rise and form a slick.  The
carbon appeared fairly effective in removing this oil as
well as the Diazinon.

At 1400 two sampling boats set out to scoop random carbon
samples from the surface of the spill area.  These were
composited for evaluation at a later time.  Post spill
water sampling was then initiated using the same procedure
as before but this time with a greater number of sampling
points.

The samples were analyzed for both phosphate and total
organic carbon (TOC).  The latter determination may be
subject to some error due to the presence of oil slicks
from previous spill studies in the basin water.  Phosphate
was determined by digesting an aliquot of the sample with
a sulfuric acid-nitric acid mixture prior to colorimetric
measurement of the phosphate concentration by the ascorbic
acid-phosphomolybdate method(7).  Analysis of basin water
with known concentrations of Diazinon gave 98 percent
recovery of the phosphate by this procedure.  The TOC
analyses were performed with a Beckman Model 915 carbon
analyzer.

Results of the phosphate analyses of the pretreatment and
posttreatment samples are given in Figure 25, with the
locations on the sample grid.  Several increases in phos-
phate concentrations are noted between pretreatment and
posttreatment samples which are believed to be largely
caused by analytical variations.  Additional posttreatment
phosphate analyses are given in Figure 26.  The data illus-
trate that the emulsion was relatively dense and the bulk
of the spilled Diazinon formed a layer near the bottom of
the basin.  Laboratory studies conducted with the six inch
diameter column did not indicate bulk movement of a fine
emulsion to the bottom of the column.  The density of the
Diazinon is greater than water while the density of the
xylene in the emulsifiable solution is less than water.
Large droplets of the emulsifiable solution will float when
initially dropped in water but will slowly sink after a few
minutes.  It is believed that the lighter xylene is either
evaporated or extracted into the water causing the density
of the droplet to increase.
                           68

-------
10 FT II
* (I 8-0 0) 100
B I* 9-0 2) *t
C |4 0-0.9) 78




ZO FT U
I (1 1-1 2) «(.
B (12 4-0 0) 100
C (1} 6-0 9) fS




10 FT H
A (1 9-0 2) 89
B (7 6-0 0) 100
C (13 0-0 4) 97
                              is FT H
                            • (0 0-0.0)   0
                            B ID 0-0 0)   0
                            C (0 0-0.1)  ---
                               5 FT H
                           K (0 2-0.2)   0
                           B (1 4-0 2)   16
                           c t) 1-0.6)   ai
                             •  CENTER
                           A (2 0-0 2)   90
                           I (4 «-Q 4)   91
                           C (10 3-0 9]  91
      I FT DEPTH
      S FT DEPTH
      10 FT DEPTH
      DISTANCE AflO DIRECTION FBOM CENTO
      mm»lHEM1 W>4 COHCUimTlON. rag/1
      POSttREATHEHI P04 CONCf»TB*TIOI1, aq/l
      PERCENT REOUC1IOH
    S FT C
A (O.S-O.Z)  60
a (i 1-0 o) 100
C (10 9-0 2) 98
    15 FT E
A (0.7-0.3) 57
B (1 6-0 it 18
C (10 5-0 1) 97
   Zi FT C
A (O.B-0.2)  7S
B (I 0-0 II  60
C (7 1-0.4)  95
                       FIGURE  25.
     RESULTS  OF  PHOSPHATE  ANALYSIS  ON
PRETREATMENT  AND POSTTREATMENT  SAMPLES
                            69

-------
39 FT SH
 A  0.2
 B  0.2
 C  0.9
            25 FT NW
             A  0.2
             B  0.0
             C  0.4
            25 FT SW
             A  0.3
             B  0.2
             C  0.3
                           A =  1 FT DEPTH
                           B =  5 FT DEPTH
                           C =  10 FT DEPTH
                           (P04 IN mg/1}
            39 FT HE
             A  0.0
             B  0.0
             C  0.4
 25 FT NE
  A  0.0
  B  0.0
  C  0.2
11 FT NW
A 0.2
e o.o
C 0.5

11 FT SH
A 0.0
B 0.2
C 0.6




X^




11 FT NE
A ---
B 0.0
C 0.0

n FT SE
A 0.2
B 0.2
C 0.4
25  FT SE
 A  0.3
 B  0.5
 C  0.3
            39 FT SE
            A  0.2
            B  0.0
            C  0.3
                            FIGURE 26
            ADDITIONAL RESULTS  OF PHOSPHATE ANALYSIS
                   ON POSTTREATMENT SAMPLES
                                70

-------
Observations indicated that the wind drifted the emulsion
briefly to the south and east while the emulsion was near
the surface, but once the emulsion settled several inches
into the water it appeared to remain stationary.  The bulk
movement to the west as indicated in the data of Figure  25
and in the aerial photograph, Figure 21, was not observed
visually at ground level following the spill.

Results of the TOC analyses presented in Figures 27 and  28
correlate reasonably well with the phosphate'analyses.  As
in the case of the phosphate results, several increases
occurred between pretreatment TOC concentrations and post-
treatment TOC concentrations.  Analytical variations and
oil contamination of the samples are believed to have
caused these increases.  Interpolation of the TOC data over
the affected water column accounts for 66 pounds of the
Diazinon solution or 85 percent of the 78 pounds originally
spilled.  Similarly, the phosphate data accounts for 68
pounds or 87 percent of the original solution spilled.

Based upon the water sampling data,  the total quantity of
Diazinon remaining in the water column after treatment was
calculated in the same manner as had been the spill account-
ability numbers from the pretreatment data.

Phosphate analyses indicated a total of 4.2 pounds of Diazinon
solution left in the water,  or five percent of the original
78 pounds.  Total organic carbon analyses indicated a residual
of five pounds or six percent of the original spill.   This
suggests an average of 94.5  percent removal from combined
dilution and treatment effects.  However, examination of
the post treatment data revealed that low concentrations of
Diazinon might occur for some distance outside the sampling
grid.   Since most of the carbon was applied rapidly to the
major dispersion of the spill in the southwesterly direction
it is believed that good contact was made between the carbon
and the bulk of the spill.  Eddy currents set up by bottles
descending to the bottom of the basin may have persisted for
some time after the treatment and may have caused dispersion
of the treated water containing low concentrations of the
Diazinon.  As a worst case estimate, a volume six times that
of the sample grid could be assumed to have the same average
posttreatment Diazinon concentration as that found within
the sample grid.  In such a case the removal would be 67
percent.  The volume of water would then include both that
in the sample grid and all of the water contained outside the
grid in the basin west of the eastern edge of the sample grid.
Thus,  the actual removal of Diazinon should be between 67
and 94 percent.

Composite samples of carbon collected after treatment of the
Diazinon spill were analyzed in an effort to determine the
amount of Diazinon recovered by the carbon.  The results of


                           71

-------
                       is rr *
                    » (7.1-1.5)  79
                    I (1.7-0.1)  SI
                    C (1.8-4 1) -131
                      IS FT »
                    » (7 3-1.5) -1
                    6 (I.I-I.4) -17
                    C (J.6-I.I) ii
JO rr v
I (H 7-2 1) SI
e (id 5-2 6) 87
C (IS 3-1 i) 7)

.


20 Fl U
« (12 7-1 4) 89
B (39 6-0 2) 49
C (43 2-3 0} 41

•


10 FT H
» 17 J-1.4) 81
B (2« 0-0 7} 97
C (47 8-J 6] »

—


Cdrrti
* (a s-o si 94
B (17 8-2 1) 88
C (IS 6-5 0) 86

•


S Fl E
A (4 0-2 3) 4)
B (11.0-1 0) 91
C (33.0-0 1} 98

m


IS Fl (
« (4 6-4.7) -2
8 (6 8-1.6) 76
C (37 1-2.2) 94

M


15 FI c
* (4 5-1 8) «0
8 (8 0-2 2) 73
C (22.0-1 2) 94












0
• (©-Q)i ©
8
C














» i n OIPIN
10 FT S
* (IS 2-1 0)
B (16 7-1 8)
C (SI 4-2 S)

93
89
9i
I
25 FT S
1 (3 2-1 1)
1 (14 3-1 4)
C (30 S-? <)

6(
90
91
1
40 FT S
(4 0-1 OJ
(4 6-0 8)
(2 5-3 0) .

t 10 n DIPIH
Q OISTIIICE >ND OlXCIlOn ritOM CEJIHH
© pRtrRCurHEnr 100 ccncciirMrioii. ««/)
Q POSTixiiHiM ioc co»ctmii«iioii, 19/1
© PEBCtNJ «EOUCIION

75
83
20






                FIGURE  27
RESULTS  OF  TOC  ANALYSIS  ON PRETREATMENT
       AND POSTTREATMENT  SAMPLES
                     72

-------
            25  FT NW
             A   3.2
             B   2.0
             C   5.0
                           A = 1  FT DEPTH
                           B = 5  FT DEPTH
                           C = 10 FT DEPTH
                           (TOC IN mg/1)
            25 FT  SH
             A  1.7
             B  1.8
             C  4.6
25 FT NE
 A  1.9
 B  1.2
 C  1.5
11 FT NW
A 0.4
B 0 6
C 2.7

11 FT SU
A 0.4
B 0.9
C 2.9




y




11 FT NE
A
B 1 5
C 1.0

11 FT SE
A 1 4
B 7 1
C 11 .9
25  FT SE
 A   1.8
 B   1.4
 C   1.8
39  FT SH
 A  1.8
 B  1.4
 C  3.0
            39 FT  SE
             A  2.6
             B  1.3
             C  1.8
                            FIGURE  28
               ADDITIONAL RESULTS  OF TOC  ANALYSIS
                    ON  POSTTREATMENT SAMPLES
                                 73

-------
these analyses were inconclusive, however, due to considerable
variation between analytical results.  Insufficient informa-
tion was found in the literature concerning the extraction of
Diazinon from carbon for subsequent analyses.  The develop-
ment of adequate analytical procedures for analyzing the
carbon for Diazinon content would require a level of effort
for which time and funding were not available.

Sediment samples were analyzed to verify that the pesticide
had not settled and associated itself with the bottom sedi-
ments.  No change in sediment phosphate levels was detected.
Sediment supernatant samples showed a slight increase in
phosphate level.  This increase, however, when extrapolated
to the total spill area could account for less than 50 grams
of Diazinon.  It is clear that the floating carbon proved
highly effective in removing the spilled Diazinon emulsion.
Even greater efficiencies could be expected for light organics
which would remain near the surface.  In such cases, the
rising carbon would simulate a countercurrent contacting pro-
cess.  Carbon resting on the surface or gently circulating
nearby with wave action would remain in contact for greater
periods of time and thus remove larger amounts of the organic
contaminant.

Booming operations were begun at 1415.  A 200 foot spliced
section of a plastic Rodeorm Oil Containment boom manufactured
by Trelledorg, Inc. was used to herd the carbon into the
southeast corner of the basin.

The boom was pulled through the water manually with the aid
of rope attached to the end of the boom.  The northern end
was first brought into the western wall, and then the tri-
angle thus formed was diminished by moving both ends of the
boom towards the apex.  It was found that the carbon herded
easily with no underwash evidenced.

At the beginning of the booming operation the wind had driven
most of the carbon to the collection corner.  During the
operation the wind shifted causing the carbon to drift out
against the boom.  Some carbon escaped by moving out through
the gap left between the end of the boom and the basin wall.
Most of the carbon, however, appeared to pile up against the
boom.

When the boom was well into the corner, carbon appeared to
form a blanket up to two inches deep in places.  A "J" shaped
pipe was lowered into the corner of the basin and connected
to a gasoline powered diaphragm pump.  Figure 29 illustrates
the placement of the pipe.  The suction from the pump drew
down a vortex from-the surface of the basin and thus removed
the carbon, which was then pumped to a 15 foot diameter
portable swimming pool.  The booming operation was completed
by 1645.  Water in the pond appeared to have returned to its
original clarity with no noticeable carbon residues.
                           74

-------
 PUMP TO
COLLECTION
  BASIN
                                                          BOOM
                                   2" x 4" BELL REDUCER
      RETAINER
        WALL
                           FIGURE 29

          PLACEMENT OF  THE COLLECTION  PIPE AND BOOM


                              75

-------
Water was subsequently drained from the carbon storage tank
and the remaining carbon was allowed to dry.  Measurements
of the total volume and bulk density of the recovered carbon
indicated that 90 percent of the original volume of carbon
was recovered.  Laboratory tests show that two percent of
the carbon employed in the demonstration does not float and
an additional one percent remains in the plastic container.
A substantial portion of the remaining seven percent lost in
the demonstration may have been trapped on weeds in the
bottom of the basin and may not have floated to the surface.
The 90 percent recovery was considered to be good.

No alternative skimming or booming devices were tested for
comparison of effectiveness.

SULFURIC ACID SPILL
An abandoned sedimentation basin on the Hanford Reservation
was selected for evaluation of the floating epoxypolyamine
anion exchange resin.  The 30 x 50 foot concrete lined basin
was filled to a depth of six feet with water from a nearby
fire hydrant.

Sulfuric acid was selected for the field demonstration since
it is the chemical produced in the greatest quantity in the
United States and it represents one of the major hazardous
spill threats.  During predemonstration tests it was found
that concentrated solutions of sulfuric acid would rapidly
sink and spread out over the bottom of the test basin when
introduced at the surface of the water.

It was also observed that lateral movement of a dilute acid
spill was too rapid in the basin to allow effective spill
treatment with the quantity (13.7 cu. ft.) of floating ion
exchange resin available for the demonstration.  Water
currents in the basin appeared to be largely responsible for
the movement rather than dispersion of the acid in the water.
It was therefore decided to conduct the demonstration spil'l
of sulfuric acid in a small submerged open tank (five feet
in diameter by two feet high) placed in the basin.  The tank
served to simulate the kind of natural depression found in
most river beds or lake bottoms.  Further predemonstration
testing with a submerged tank showed some loss of sulfuric
acid when the spill occurred at the surface of the water
directly over the center of the submerged tank.  The concen-
trated acid descended rapidly, thereby gaining sufficient
momentum to travel across the bottom of the tank and up the
wall to spill a portion of the acid over the wall of the
tank.  In order to avoid any loss of acid from the simulated
depression, a subsurface spill of concentrated sulfuric acid
was planned for the spill treatment demonstration.
                           76

-------
The acid spill treatment demonstration was conducted with 45
pounds of concentrated sulfuric acid containing one ounce of
m-cresol purple indicator which imparted a red color to the
acid, thus allowing visual observation of the acid beneath
the water surface.  Sulfuric acid was transferred to the
submerged tank in the basin via a one inch diameter PVC pipe.
At the completion of the transfer, the acid blanketed the
floor of the tank in a thin layer less than two inches deep.

Pretreatment sampling was performed using a six foot length
of 1/4 inch diameter stainless steel tubing connected to
flexible tubing and a peristaltic pump.  Analysis of the pre-
treatment samples showed acid to be virtually completely
contained in the thin layer on the bottom of the tank.  A pH
reading of 3.8 was recorded at the top wall (two feet above
the bottom) of the tank.  Directly above the tank and one
foot below the water surface the pH was determined to be 6.9
as opposed to a prespill pH of 7.2.

Fifty media ice blocks containing epoxypolyamine anion resin
and gravel ballast were used in the spill treatment operations.
The media ice blocks were prepared by covering a layer of
gravel in a mold with a layer of 30 percent resin slurried in
water and freezing in a home freezer unit.  Each media ice
block was packaged in a plastic bag and stored in a freezer
until use.  A schematic diagram and a photograph of the media
ice blocks used in the demonstration are given in Figures 30
and 31, respectively.

During the spill treatment operation, the media ice blocks
were removed from the plastic bags and were dropped into the
spill area approximately 30 minutes after the sulfuric acid
was spilled.  Nearly half of the media was dropped around the
periphery of the submerged tank and the remainder was fairly
evenly distributed on the bottom of the tank.  It was planned
that a substantial portion of the resin ice cakes should  fall
outside of the tank to provide resin  for  acid sorption above
tne periphery of the submerged tank.  Laboratory studies had
suggested that resin rising from the  tank would cause vertical
mixing and dispersion of  the acid above the tank.  However,
little of the anticipated mixing actually occurred, because
transferring the acid directly to the bottom of the tank
caused formation of a thin, dense acid layer.  It is believed
that the  layer of gravel ballast on the bottom of the resin
ice cakes held the resin  above the acid layer thus virtually
eliminating contact between the resin and the acid.

Analysis  of samples taken after the spill confirmed that only
a  small portion of the acid was removed from the water by the
resin.  Post spill sampling was carried out in the  same  manner
as the pretreatment sampling.  The only real change in acidity
appears to have occurred  in the center of the tank near  the
top, where the pH rose from 3.8 to 6.6.   Apparently a small
amount of vertical mixing occurred at this point.
                           77

-------
-o
to
              FLOATING EPOXYPOLYAMINE

              ANION EXCHANGE RES IN
                      2" RES IN-ICE LAYER
                                                                     1.25" GRAVEL BALLAST LAYER
                                               FIGURE  30


                      TYPICAL  "MEDIA  CAKE" EMPLOYED IN BASIN DEMONSTRATION

-------
           FIGURE 31




PHOTOGRAPH OF RESIN ICE  CAKE

-------
Within  two hours,  all of the  resin had reached the surface  (Fig-
ure  32).  Samples  of the floating resin were continuously taken
and  composited  as  the media surfaced.  The resin samples were
thoroughly mixed and a composite sample was then analyzed to
determine the capacity utilization and sulfate sorption.  The
unused  ion exchange capacity of the spent resin was determined
by titrating samples of the resin with standard acid.  The
results showed  a difference of 0.405 milliequivalents per gram
when compared to the capacity of virgin floating resin.  This
represents 4.43 pounds of sulfuric acid when extrapolated to
the total of 13.7  cubic feet of resin dropped, or ten percent
of the original 45 pounds of concentrated acid spilled.

Similarly, analysis of the spent resin for sorbed sulfate
ion revealed 22.1 mg more sulfate per gram of resin than in
the virgin floating resin.  This accounts for 5.3 pounds of
sulfuric acid or 12 percent removal.  The sorbed sulfate on
the spent resin was determined by eluting samples of the
resin with 1 N NaOH and measuring the eluted sulfate by a
turbidimetric 83804 procedure.

This low level of removal is not considered representative of
the potential of the floating ion exchange resins for treat-
ing spills of electrolytes.  There is no reason to believe
that the resin would not be effective for a diffuse or diluted
spill.  A demonstration of this type to treat an unconfined
spill would require considerably more resin than could be
feasibly produced in this program.

The acid spill demonstration does, however, point out the
need for sufficient contact of the sorbent with the material
to be removed since densely layered spills are certainly
within -the realm of possibility.  The sorbent-ice cake repre-
sents the most logical approach to attaining good bottom
contact.  Neither bottles nor clay containers would approach
the same degree of bottom contact.  Placement of the ballast
in the center of the resin-ice cake rather than on the bottom
would be expected to improve bottom contact.   To be fully
effective, however, the material to be removed must be
dispersed above the resin-ice cake for good contact with
the main plume of  resin rising above the cake.

Two days after the acid spill, operations to recover the ion
exchange resin were initiated.  The resin was easily boomed
and herded to a corner of the water basin where it was
skimmed from the surface and dewatered by screening through
a 100 mesh screen.  Virtually complete recovery of the total
quantity of floating resin used in the spill treatment demon-
stration was achieved in this manner.
                         80

-------
00
                                                                           -
                                                                	




                                             g


                                          FIGURE 32
                          SAMPLING FLOATING RESIN AFTER TREATMENT

-------
                      SECTION VII
                      APPLICATION
Although the basic concept of utilizing floating media for
spill treatment has been successfully dfcmonstrated/ it is
emphasized that neither media production methods nor
application techniques have been fully optimized.  The
following discussion will summarize the status of the
technology developed for treating hazardous material
spills with floating mass transfer media/ however/ and
provide some understanding of its use.

Floating activated carbon represents thfi most versatile
media evaluated in this work because of its ability to
remove a broad range of hazardous materials/ principally
organics, from water.  Although only a few hazardous
materials (phenol/ benzene, toluene, styrene, Diazinon,
Malathion, and acrylonitrile) were investigated in this
study, activated carbon could be effectively used for
adsorbing many other organic substances which have limited
solubility in water and/or consist of long-chained
molecular structures.  Included in this category are a
wide variety of hydrocarbons; organic halogen compounds
(e.g., pesticides); long-chained monofunctional alcohols,
aldehydes, ketones, and fatty acids; low solubility esters;
and many others.  Examples of common organic compounds
which are not readily adsorbed from aqueous solutions are
methanol, ethanol, urea, sugars, and ethylene glycol.

Effective treatment of hazardous organic material spills
with floating activated carbon will depend on the following
factors.

     1.  The material must be readily adsorbed by the
         carbon from aqueous solutions.

     2.  The carbon grain size should be small, prefer-
         ably in the 50 x 325 mesh size range, with a
         significant fraction passing a 100 mesh sieve.

     3.  The carbon must be pretreated by soaking in
         water and drying at a low temperature  (100°C
         or less)  to assure proper wetting and disper-
         sion when released from packages beneath the
         spill zone.
                           83

-------
      4.  The  spill  zone  must  be  accurately defined  and
         the  carbon applied accordingly.  On-the-spot
         judgments  may necessitate  application  of a
         larger  amount of  carbon where  higher concen-
         trations of spilled  material are believed  to
         exist.  A 10:1 ratio  of  carbon  to spilled
         material is recommended for application.

      5.  Response to the spill must be  rapid since  dis-
         persion of the  spilled  material  over a wide
         area will  decrease the  effectiveness of the
         carbon  and may prevent  accurate  location of
         the material.

      6.  The floating carbon  containing the sorbed  haz-
         ardous  material must be  collected and  removed to
         prevent desorption of the material into uncon-
         taminated  water (desorption will be slow unless
         wave action or other surface turbulence disperses
         the carbon in the water).

Release of the floating carbon from weighted plastic
bottles in a hazardous material spill zone has  been  demon-
strated, but only in static or impounded waters.  Application
to flowing streams  may require a  different technique.
Other methods of application  such as subsurface injection
by infusion pumps and release from ice cakes or unfired
clay containers  appear practical  but have not been demon-
strated on a large  scale.

The use of a floating epoxypolyamine ion exchange resin
for removing hazardous substances has also proved successful.
The resin was prepared by incorporating buoyant hollow
glass microspheres  in the resin matrix.   As is  the case
with activated carbon, a small resin particle size is
required to maximize the effectiveness of this medium.   The
floating epoxypolyamine ion exchange resin obtained  for
the field demonstration had a particle size range of 16 x
100 mesh with the bulk of the resin particles in the 16 x
40 mesh range.   A particle size passing a 50 mesh screen
would be more desirable;  however, production methods must
be refined to accomplish this.

The floating epoxypolyamine resin evaluated during this
program was found to be quite effective in removing acid
from water.  The field demonstration pointed out, however,
that spills of high density acids may be difficult to
treat because the acid forms a dense layer at the bottom
                           84

-------
which cannot- be readily contacted by the resin.  The ice
cake method of applying the resin represents the most favor-
able packaging concept investigated for use in treating
hazardous materials concentrated near the bottom of a body
of water.  Although the floating epoxypolyamine resin has
been evaluated for use in treating only acids in this study
it is anticipated that the resin will be useful for
removing toxic anions  (e.g., cyanide) also.

The effective use of the floating ion exchange resin in
treating spills will be governed by the same conditions
as those for carbon listed above with the exception of
pretreatment (item 3}.  The ion exchange resin is maintained
in a moist condition and should not be allowed to dry out.

The feasibility of commercial production of floating
activated carbon and floating epoxypolyamine resin has been
demonstrated.  Although production methods have not been
optimized, .sizable quantities of these materials can be
obtained for further studies or for application in the
field.  The sources of these materials and costs are
given in the Appendix.
                            85

-------
                     SECTION VIII
                    ACKNOWLEDGMENTS
The assistance and advice provided by Mr. Ira Wilder, EPA
Project Officer, are gratefully acknowledged.  The authors
also wish to express their appreciation to Westvaco for
samples of activated carbon provided for experimental
studies; to Dr. Irving M. Abrams, of the Diamond Shamrock
Chemical Company, who provided information and suggestions
concerning the preparation of ion exchange resins; and to
Battelle-Northwest personnel, Mr. James Coates, Mr. Marvin
Mason, Mr. Richard Parkhurst, Mr. Greg Swank, Mr. Robert
Upchurch, Mr. Gary Schiefelbein, and Mr. Terry Brix, who
assisted in the carrying out of the experimental work.
                           87

-------
                       SECTION IX
                       REFERENCES
 1.   Thompson,  C.  H.,  and P.  R.  Heitzenrater.   "The  Environ-
     mental Protection Agency's  Hazardous  Material Spill
     Program."   Presented at  the American  Institute  of
     Chemical Engineers Workshop, Charleston,  West Virginia,
     October 27-29,  1971.

 2.   Thompson,  C.  H.,  and K.  E.  Biglane.   "Oil and Hazardous
     Materials—The  Chemical  Industry's Liability or Asset.11
     Presented  to  Chemical Markets Research Association in
     Chicago, 111.,  Feb. 24,  1971.

 3.   Hyndshaw,  A.  Y.  "Use of Activated Carbon to Prevent
     Water Supply  Contamination."  Water and Waste Engineering,
     February,  1969.

 4.   Swindell-Dressier Company.   "Process  Design Manual
     for Carbon Adsorption."   Environmental Protection Agency,
     Technology Transfer, Program 117020 GNR,  Contract
     #14-12-928, p.  4-4, October, 1971.

 5.   Product Data  Bulletin, Nuchar Granular Activated
     Carbon, Grade WV-L 8 x 30,  Westvaco Chemical Division,
     West Virginia Pulp and Paper Company.

 6.   Hassler, J. W.    Activated  Carbon.   Chemical Publish-
     ing Company,  New  York, NY,  p. 330, 1963.

 7.   Standard Methods  for the Examination  of Water and Waste-
     water .  13th  Edition, APHA, AWWA, WPCF, 1971.

 8.   Boyd, G. E.,  A. W. Anderson, and L. S. Meyers,  Jr.
     "The Exchange Adsorption of Ions from Aqueous Solutions
     by Organic Zeolites II.   Kinetics."  Journal ACS, Vol.
     69, November, 1947.

 9.   Kressman,  T.  R. E. , and J.  A. Kitchener.   "Cation^
     Exchange with a Synthetic Phenolsulphonate Resin."
     Faraday Society Discussion, No.  7, 1949.

10.   Helfferich, F.   Ion Exchange. McGraw  Hill, San  Francisco,
     California, 1962.
                            89

-------
                       SECTION X
                       APPENDIX
          SOURCES, PREPARATION, AND COSTS OF
             FLOATING MASS TRANSFER MEDIA
ACTIVATED CARBON

The floating activated carbon used in the field demonstration
was prepared from Nuchar C-190, +30 mesh carbon manufactured
by Westvaco of Covington, Virginia.  Four thousand pounds
of this carbon were purchased at a cost of $1,693 delivered
to Majac, Inc., of Pittsburgh, Pennsylvania.  The carbon
was pulverized and air classified to remove the fines
(-325 mesh) by Majac at a cost of $1,200.  The carbon
was then shipped to Battelle-Northwest for the remaining
processing steps which consisted of dry sieving to separate
the 40 x 325 mesh fraction and washing and drying to condition
the carbon to wet and disperse properly when applied for
spill treatment.

The production rate of the continuous sieving apparatus
used to separate the desired mesh fraction was limited to
about four pounds of 40 x 325 mesh carbon per hour.  Wash-
ing was accomplished by mixing the carbon with water in a
50 gallon tank to thoroughly wet the carbon particles.  The
carbon was then allowed to rise to the surface, removed
from the tank by skimming, and placed in canvas bags for
draining.

Drying was accomplished in about one week at ambient air
temperatures by spreading the damp carbon to a depth of
1-2 inches on diamond mesh sisal paper in well ventilated
rooms.  The average temperature and relative humidity
during the drying period were 70°F and 48 percent, res-
pectively .

Seventy-one percent or 1700 pounds of the carbon received
from Majac was processed throuah the sieving apparatus.
The final yield of treated 40 x 325 mesh carbon was 675
pounds which cost $1539 to process through the sieving,
washing, and drying steps.  The total cost of the 675
pounds of 40 x 325 mesh carbon was $3,593, which includes
the $1,539 above and 71 percent of the purchase and
pulverizing/air classifying costs.
                           91

-------
 Since  the  completion  of  this  work  Westvaco  has  discontinued
 sales  of the  +30  mesh and  unground grades of  Nuchar C-190
 and will soon discontinue  sales  of the  same grades  of
 Nuchar WA.  Westvaco  will  continue to market  a  fine mesh
 size of Nuchar C-190  (Nuchar  C-190 N) with  about  one-third
 of the carbon in  the  100 x 325 mesh range.  This  material
 costs  approximately 17 cents  per pound  (excluding delivery
 costs) in  quantities  of  30,000 pounds or more,  which is
 about  half the cost of the +30 mesh grades  formerly
 marketed.   The Nuchar C-190 N carbon has not  been evaluated
 as a floating mass transfer media;  however, there is no
 reason to  expect  that it would not function as  well as
 the other  grades  tested.   In  addition to the  lower  purchase
 price, the  Nuchar C-190 N  could  be processed  to give the
 desired mesh  size at  a lower  cost  since no  pulverization
 would  be required.

Westvaco also markets another grade of powdered carbon,
Nuchar C-115 N, which has about one-third of  the  carbon
 in the 100 x  325 mesh range. A cursory examination  indicates
a high percentage of  floating particles in  the desired
mesh size range.


 EPOXYPOLYAMINE RESIN

The floating epoxypolyamine ion exchange resin was obtained
by special order  from the Diamond  Shamrock  Chemical Company,
Nopco Chemical Division, Duolite Ion Exchange Resins,
Redwood City,  California.  Five hundred pounds  (14 cu. ft.)
of this resin were purchased for $4,000.

The floating anion exchange resin prepared  in the Diamond
Shamrock Chemical Company's development laboratories was
characterized as  follows:

     Reference number              1,219,  67

     Free base form                100 percent floaters

     Chloride form                 100 percent floaters

     Sulfate form                   95 percent floaters

     Total acid absorption         7.2 mg/gm
     capacity  (dry weight basis)

     Moisture retention capacity   59 percent

     Granule strength              Fairly hard

     Particle size                 30 to 70 mesh
                           92

-------
  The 500 pounds of resin  produced in plant facilities had
  essentially the same characteristics  except that  the
  particles were largely in the 16 to 100 mesh size range.
•OS GOVERNMENT PRINTING OFFICE 1973 546-312/162 1-3    93

-------
  SELECTED WATER
  RESOURCES ABSTRACTS
  INPUT TRANSACTION FORM
                  /. Report No
                                                          3.  Accession No
                                    w
  4  Title
    TREATMENT  OF HAZARDOUS MATERIAL SPILLS
    WITH FLOATING MASS TRANSFER MEDIA
  7  Author(s)

    Mercer, B.  W.,  Shuckrow, A.  J.,  and Dawson,  G.  W.
  9 Organization

    Battelle Memorial Institute
    Pacific Northwest Laboratories
    Richland,  WA  99352

 12. Sponsoring Organization

 15 Supplementary Notes
    Environmental Protection Agency report number,
    EPA-670/2-73-078, September 1973.
                                   5. Report Date

                                   6.

                                   8 Performing Organization
                                     Report No.

                                   W Protect No

                                     15090 HGQ
                                   11  Contract/Grant No

                                   Contract  68-01-0124
                                   13  Type of Report and
                                      Period Covered
  16 Attract  An approach for  the in situ treatment of spills of  soluble haz-
ardous polluting substances  was developed  and demonstrated  on a field scale
for a static body of water.   Laboratory scale experimentation showed that
floating  sorbents and ion  exchange resins  could be highly effective removal
agents when applied as small particles beneath the surface  of contaminated
waters.
           A lightweight commercial activated  carbon was found to be partic-
ularly effective for removing organic substances such as phenol, aromatic
lydrocarbons, and organophosphorus insecticides from water.
           Floating ion exchange resins were also prepared for use on spills
jf acid,  alkalis, and toxic  salts.  Hollow glass microspheres are incorpor-
ated in the resin granules for buoyancy.
           Field demonstrations were conducted using carbon  contained in
weighted  plastic gallon bottles.  The carbon  proved highly  effective in
removing  an organophosphorus pesticide spilled in a large basin, and was
sasily collected through use of an oil containment boom.  Ice encapsulated
floating  anion exchange resin beads were similarly employed to neutralize
i spill of sulfuric acid.
  17a Descriptors
"Activated Carbon, *Ion  Exchange,*Resins,*Water Pollution  Treatment, Water
 Pollution Control
  776 Identifiers

 "Spills, *Hazardous Materials, -"Hazardous  Chemicals ,*Mass Transfer Media
  17c COWRR Field & Croup  05G
  18. Availability
19. Security Class.
   (Report)
                      20. Security Class,
                         (Page)
21. No of
   Pages

22. Price
Send To:
                         WATER RESOURCES SCIENTIFIC INFORM ATION CENTER
                         US DEPARTMENT OF THE INTERIOR
                         WASHINGTON. D C 20240
  Abstractor  A. j. Shuckrow
           {institution  Battelle Memorial  Institute
WRSICI02(REV JUNE 1171)
                    Pacific Northwest Laboratories u.a«t

-------