EPA-670/2-75-004
April 1975
Environmental Protection Technology Series
FEASIBILITY OF 5 gpm
DYNACTOR/MAGNETIC SEPARATOR SYSTEM
TO TREAT SPILLED HAZARDOUS MATERIALS
National Environmental Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
EPA-670/2-75-004
April 1975
FEASIBILITY OF 5 gpm DYNACTOR/MAGNETIC SEPARATOR SYSTEM TO
TREAT SPILLED HAZARDOUS MATERIALS
By
Robert G. Sanders
Industrial BIO-TEST Laboratories, Inc.
Northbrook, Illinois 60062
Stanley R. Rich
Thomas G. Pantazelos
RP Industries, Inc.
Hudson, Massachusetts 01749
Contract No. 68-01-0123
Program Element No. 1BB041
Project Officer
Ira Wilder
Industrial Waste Treatment Research Laboratory
Edison, New Jersey 08817
NATIONAL ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
REVIEW NOTICE
The National Environmental Research Center, Cincinnati, has reviewed
this report and approved its publication. Approval does not signify
that the contents necessarily reflect the views and policies of the
Environmental Protection Agency, nor does mention of the trade names
or commercial products constitute endorsement or recommendation for
use.
ii
-------
FOREWORD
Man and his environment must be protected from the adverse
effects of pesticides, radiation, noise and other forms of pollution,
and the unwise management of solid waste. Efforts to protect the
environment require a focus that recognizes the interplay between
the components of our physical environment—air, water and land.
The National Environmental Research Centers provide this multi-
disciplinary focus through programs engaged in
- studies on the effects of environmental contaminents
on man and the biosphere, and
- a search for ways to prevent contamination and to
recycle valuable resources.
Pollution resulting from spills of hazardous materials is widely
recognized as very damaging to the water ecosystem and to the public
health and welfare. This report describes new physical-chemical
treatment technology for the cleanup of waters contaminated by this
source of pollution.
A. W. Breidenbach, Ph.D.
Director
National Environmental Research Center
Cincinnati
iii
-------
ABSTRACT
Design and operating details are given for a new type of continu-
ous flow thin-film, gas-liquid-particulate contact device called the
Dynactor. The device is used as a continuous flow short-time contact
reactor to effectively decontaminate water contaminated with spilled
hazardous materials. The decontamination is effectively achieved by one
or more processes involving oxidation, neutralization, precipitation or
adsorption on powdered carbon. Contaminated water is processed by
the pilot plant model Dynactor at 100 psi and at a rate of 5 gpm;
stoichiometric quantities of decontaminating agents in the form of gases,
liquids, slurries or powders are metered into the continuously flowing
liquid configuration. The device is portable, lightweight polypropylene
construction, has no moving parts, requires a pump for liquid motive
power and can be scaled up to process 250 gpm of contaminated water.
Design and operating details are given for continuous flow magnetic
separation to remove flocculated carbon and precipitates from the
Dynactor effluent after decontamination of hazardous materials.
Experimental data on successful decontamination of heavy metals
by precipitation, acids and bases by neutralization, phenol, chlorine
and pesticides by powdered carbon adsorption and other selected hazard-
ous compounds are presented.
iv
-------
CONTENTS
Review Notice
Foreword
Abstract
List of Figures
List of Tables
Section
I
II
III
IV
V
VI
Conclusions
Recommendations
Introduction
Technical Approach
Experimental Studies
References
11
iii
iv
vi
vii
1
2
3
5
19
32
-------
FIGURES
1 Dynactor Diffusion System °
2 Dynactor Model 5 GPM 1°
3 Dynactor Radial Pressure Transformation Section H
4 Laboratory Assembled Dynactor System 12
5 Plenum Chamber and Powder Dispenser 13
6 View Inside Plenum Chamber 14
7 Liquid Feed and Pump Connections 15
8 Continuous Flow Magnetic Thickener/Separator 17
9 Magnetic Separator Model 5 GPM 18
•i Q Phenol Adsorption vs . Initial Phenol Concentration 23
11 Phenol Adsorption as a Function of Initial Carbon 24
to Phenol Ratio
vi
-------
TABLES
No. Page
1 Adsorption of Pesticides in Dynactor System 26
Using Powdered Activated Carbon
2 Adsorption of Chlorine in Dynactor System 27
Using Powdered Activated Carbon
3 Removal of Water Soluble Components — 28
Fuel Oil No. 2 — Carbon Adsorption
Dynactor Processed
4 Removal of Oil in Water Emulsions — 30
Carbon Adsorption — Dynactor Processed
vii
-------
SECTION I
CONCLUSIONS
Under Contract No. 68-01-0123 from the Environmental Protection
Agency, Industrial BIO-TEST Laboratories, Inc., and R P Industries,
Inc. (subcontractor) have proven feasibility of a short contact time
physical chemical treatment system consisting of a Dynactor and a
magnetic separator, to process and decontaminate hazardous materials
spilled in waterways. The feasibility study was directed to the decontami-
nation of the contained spill and the contaminated waterway through
a process of physical and chemical reactions carried out by a continuous
flow thin-film, gas-liquid-particulate contact device (Dynactor). With
this device it was shown possible to meter into the continuously flowing
stream of contaminated water stoichiometric quantities of decontaminating
agents in the physical form of gases, liquids, slurries and powders.
These decontaminating agents interface with the contaminated water
in thin-film configuration and effect decontamination during the approxi-
mate 0.2 second residence time in the Dynactor's reaction column.
The decontamination was shown to be effectively achieved by one
or more processes involving oxidation, neutralization, precipitation
or adsorption on powdered carbon. Acids, bases, phenol, chlorine,
aliphatic and aromatic hydrocarbons, cyclic and acyclic pesticides
and miscellaneous hazardous substances were successfully decontaminated
within the scope of the feasibility contract.
In order to separate precipitates and flocculated carbon (resulting
from the decontamination process) in a continuous flow configuration,
a process of magnetic separation was designed and satisfactorily demonstrated.
-------
SECTION II
RECOMMENDATIONS
The feasibility study resulted in sufficient data, equipment
performance and information to meet the design goals of the project
and to recommend a Phase II study. All of the experience and data
documented in this feasibility study were generated with a physical-
chemical treatment system capable of processing contaminated water
at the maximum rate of five (5) gallons per minute (gpm). While
the 5 gpm model is a very convenient and suitably sized unit for
laboratory pilot studies, useful applications in the field will require
a system capable of processing at least 250 gpm. It is therefore
recommended that a Dynactor and separation system be developed
with a processing capacity of 250 gpm engineered in a mobile continuous
flow configuration. This work should include the design, fabrica-
tion, engineering, testing and decontamination performance of the
complete unit. It is also anticipated that some high rate settling
columns will be necessary to initially separate the flocculated carbon
and precipitates from the large volume of processed water. To complete
the separation process, the solids should be further concentrated
and dewatered by a suitably scaled up magnetic separation system.
/
It is recommended that each scaled up component of the system
be tested for mechanical performance and then mounted on a single
self-contained flat bed trailer equipped with a diesel powered electric
generator set.
Finally, it is recommended that the total mobile system be tested
against selected hazardous chemicals under simulated spill conditions.
-------
SECTION III
INTRODUCTION
This report covers work conducted primarily during the period
July 1971 through June 1972 involving a feasibility study of specific
equipment for the mobile processing and decontamination of spilled
hazardous materials in waterways. The work was carried out under
EPA Contract No. 68-01-0123 in response to a Request for Proposal
to develop new and effective methods to "Treat, Control and Monitor
Spilled Hazardous Materials." Industrial BIO-TEST Laboratories, Inc.,
is the prime contractor responsible for program management and the
chemical and analytical aspects of the decontamination studies. R P
Industries, Inc., is the subcontractor responsible for equipment design,
fabrication, modification and engineering evaluation.
Industry is producing and shipping an ever increasing volume
and array of hazardous polluting substances which pose a constant
threat of sudden discharge into the waters of the nation. Accidental
spills will occur through human error or unforeseen or uncontrollable
disasters and circumstances . The spills will cause varying degrees
of hazard and damage in the watercourse, depending on the uses of
the watercourse, type and quantity of materials spilled and their relation
to the size and type of watercourse. Because of the diversity of
potentially hazardous substances and persistence of materials, resulting
in both immediate and long-term effects, the immediate initiation of
proper countermeasures needed for flowing streams, impoundments,
estuaries, and open seas will probably vary, making the type of
response more diverse. In any event, the countermeasures must
be selected to permit rapid application in both congested and remote
areas, light in weight, easily obtainable and transportable, present
limited hazards to handlers and result in no reactivity problems causing
secondary pollution in the watercourses or generation of harmful
sludges.
The Battelle Report (1) on "Control of Spillage of Hazardous
Pollution Substances" clearly documents the fact that techniques for
treating and controlling spilled hazardous materials in the aquatic
environment are inadequate or nonexistent. Methods are available
as a second line of defense for removing almost all water contaminants
under controlled conditions and with fixed water treatment plant
installations. However, most of these techniques are not satisfactory
for application in the aquatic environment. The report recommended
-------
that work must be initiated in the area of control to develop new
methods for containing spills before they reach surface waters, to
contain contaminated waters after a spill, to decontaminate polluted
water areas and to return the water in a restored condition of quality.
Our effort is directed to the decontamination of the contained
spill and the contaminated waterway itself through a process of physical
and chemical reactions carried out by a continuous flow, thin-film
gas-liquid-particulate contact device. With this configuration, it
is possible to meter into the device stoichiometric quantities of decontaminating
agents in the physical form of gases, liquids, slurries and powders.
The agents interface with the contaminated water in the reactor and
effect decontamination during the approximate 0.2 second residence
time in the reaction column. The hazardous material is decontaminated
by chemical reaction processes involving oxidation, neutralization,
precipitation, adsorption on activated carbon or combinations of these.
Powdered carbon containing adsorbed toxic materials and/or toxic
precipitates must be removed from the reactor effluent after decontamination
has been achieved. A new process that renders nonmagnetic solid
materials temporarily magnetic and capable of continuous flow magnetic
separation was successfully evaluated.
Chemicals representing examples of classes of hazardous materials
of high ranking importance were selected for study from the Battelle
Report (1). Acids, bases, phenol, chlorine, heavy metals, pesticides
and water soluble fractions of fuel oil are included in this report as
examples of decontamination reactions carried out with the system described.
This report documents the approach to the problem, the equipment
design, decontamination experiments and results achieved.
-------
SECTION IV
TECHNICAL APPROACH
Two basic items of equipment described herein are essential to the
continuous flow decontamination processing of water contaminated with
hazardous materials.
1. A thin-film gas-liquid-particulate contact device to contact the
water with decontamination chemicals.
2. A moving belt continuous flow magnetic separator to remove
carbon and precipitates yielding a contaminent-free effluent.
Thin-Film Contact Device
Figure 1 shows a cross sectional schematic diagram of the device
TA/f
named the Dynactor1 , a proprietary development of R P Industries,
Inc. Total weight of the unit is less than 40 pounds and stands about
7 feet in height.
The Dynactor can be viewed as a macroscopic diffusion pump
which makes use of diffusion principles in order to aspirate large volumes
of air per volume of motive liquid. Liquid entering the system under
a pressure of 40 to 100 pounds per square inch (typical) is atomized
into thin films and droplets of average thickness or diameter less than
1/64 inch. This liquid discharge diffuses or expands into the reaction
chamber causing air or gas to be aspirated by being trapped within
the moving shower of films and particles. The internal configuration
is constructed to maximize gas-liquid turbulence and contact throughout
the length of the 6-foot long, 12-inch diameter reaction column. The
resulting mixed fluid then continues to travel down the reaction column
with intimate contact maintained between gas and liquid. This causes
physical and chemical equilibria to occur by the time the mixed fluid
exits from the reaction column into the separation reservoir.
The radial pressure transformation section is used to transform
ambient or atmospheric air pressure to the partial vacuum that exists
within the Dynactor. Entering air is accelerated from low velocity
and atmospheric pressure to high velocity and subambient pressure
as it enters the reaction column. By utilizing diffusion the Dynactor
aspirates up to 4,800 standard volumes of gas per volume of motive
liquid. In comparison, venturi eductors will aspirate not more than
100 volumes of gas per volume of motive liquid.
-------
AIR INPUT,
LOW VELOCITY,
AMBIENT
PRESSURE
HIGH
VELOCITY,
SUB- __
AMBIENT
PRESSURE
-LIQUID INPUT, 40 TO 100 PSI
-PLENUM CHAMBER
RADIAL PRESSURE
TRANSFORMATION SECTION
NOZZLE
SHOWER OF THIN FILMS AND
PARTICLES
REACTION COLUMN
TURBULENT MIXED FLUID
GAS OUTPUT
t
BAFFLE
RESERVOIR /SEPARATOR
(LIQUID) ^^_______
LIQUID
LEVEL
DETERMIN-
ING TRAP
T
LIQUID OUTPUT
FIGURE 1. DYNACTOR DIFFUSION SYSTEM CROSS SECTIONAL VIEW
-------
The two-stage Dynactor proposed for this study uses an input
of about 5 gpm of water at 100 pounds per square inch pressure.
Approximately 2,000 standard cubic feet per minute of atmospheric
air is aspirated into the Dynactor and reacted with the liquid. Power
requirements are 1/3 horsepower under these conditions.
The Dynactor's high speed chemical and physical reaction characteristics
are due to its treatment of the flowing liquids essentially in thin-
film form. Its substantially instantaneous gas transfer (oxygen, chlorine,
ozone), mass transfer (reaction with activated carbon, other reagents),
and heat transfer (both evaporative and conductive) are seen from
the following analysis:
Thermal Conductivity
(Fourier's Law)
Q = K A (T? - Tp
h
Q = heat flow/unit time
K = thermal conductivity
Tj = temperature at boundary 1
T2 = temperature at boundary 2
h = layer thickness
A = area
m
t
EL
t
B
Gas Diffusion
(Mass Transfer)
(d2 -
= mass flow/unit time
= coefficient of diffusion
d, = gas concentration at
boundary 2
d} = gas concentration at
boundary 1
h = layer thickness
As a consequence of their similar form, the solutions for both equations
are also similar. There is obtained, therefore, from the solution
of the heat-flow equation, the solution of the mass transfer equation when
the dissolved gas concentration is substituted for the temperature T, and
the diffusion coefficient, B, for thermal conductivity; *"
K.
In the applicable transient heat or mass flow problem in which a layer
of thickness, h, having an initial (a) temperature, or (b) gas concentration,
is subjected at one boundary to a higher or lower (c) new temperature
or (d) new gas concentration, the Fourier Integral analysis and solution
takes the form:
1
constant-
f(h,T) or F(h,d) = constant.
-------
It is seen that the solution involves an exponential term that includes
the square of the thickness of the liquid layer as an exponent of
e. Working through the mathematical details, it can be shown that
a liquid layer of less than .01 inch thick reaches physical and chemical
equilibrium with a contacting gas in less than 0.1 seconds.
With this configuration, it is possible to meter into the continuous
flowing stream of contaminated water stoichiometric quantities of
decontaminating agents in the physical form of gases, liquids, slurries
and powders. The Dynactor is equipped with liquid, gas and powder
metering systems. These enable lime, bicarbonate and powdered
carbon, for example, to be metered by aerosolization directly into
the flowing contaminated liquid and thus effect decontamination by
neutralization, precipitation or adsorption. Liquid agents such as
acetic acid are metered directly into the flow of contaminated water
before it reaches the nozzle.
The Dynactor, much like oil or mercury diffusion vacuum pumps,
has no moving parts. Since there are no constrictions, there are
no zones or portions of the Dynactor on which solids or liquids tend
to accumulate, thus requiring little maintenance. These units have
been constructed in a wide variety of materials, including polypropylene,
polyvinyl chloride, stainless steel, and mild steel. Nozzle design
is such that these elements tend to be self-cleaning, further reducing
maintenance requirements.
Figure 2 is a photograph of the Dynactor without the plenum
chamber on the top. The liquid input tube, the air intake radial
pressure transformation section and the 6-foot reaction chamber can
be clearly seen. Liquids and slurries of decontaminating chemicals
are metered into the input tube through the connecting valve shown
at the extreme right of the input tube.
Figure 3 shows a close-up of the radial pressure transformation
section which is critical to the air flow through the Dynactor. Figure
4 shows the complete Dynactor system mounted in the laboratory
and includes the plenum chamber enclosing the top, the powder feed
mechanism, the reservoir, and the air exhaust to the fume hood.
Figure 5 is a closeup of the plenum chamber and powder dispenser,
and Figure 6 is a picture taken through the plenum chamber opening
to show where the entrance of the powder dispenser is positioned
in reference to the air intake baffles. Powdered carbon, for example,
is aerosolized by the powder feed mechanism and sucked into the
throat of the Dynactor by the air intake under the baffles of the impedence
section.
8
-------
Figure 7 shows the array of plumbing from the 25-gallon reservoir
through the displacement pump to the nozzle section of the Dynactor.
This is the laboratory 5 gpm model system used to process experimentally
contaminated water from the reservoir in 100 liter batches. The
data reported in Section V were obtained with this model system.
-------
FIGURE 2. DYNACTOR MODEL 5 GPM
10
-------
FIGURE 3.
DYNACTOR RADIAL PRESSURE TRANSFORMATION SECTION
11
-------
FIGURE 4. LABORATORY ASSEMBLED DYNACTOR SYSTEM
12
-------
FIGURE 5. PLENUM CHAMBER AND POWDER DISPENSER
13
-------
FIGURE 6. VIEW INSIDE PLENUM CHAMBER
14
-------
FIGURE 7. LIQUID FEED AND PUMP CONNECTIONS
15
-------
Magnetic Separation Equipment
To provide a complete system for decontamination, it is essential
to remove undesirable end products such as precipitates and carbon
from the effluent of the Dynactor. In a mobile field configuration
the decontamination of spilled hazardous materials in waterways requires
a continuous flow process to remove potentially toxic precipitates,
suspended solids, and spent carbon in suspension. The logistics
of gravity settling tanks and cumbersome filtration devices in the
field preclude their use.
Suspensions of powdered carbon in the effluent of the Dynactor
were impossible to remove by conventional methods. Flocculating
agents were used, and some of these produced a suitable floe within
two seconds after addition of about 1 ppm. However, when the carbon
floe was processed in continuous flow separators or commercial filters,
the shear forces were sufficient to disrupt the lacy structure of the
carbon floe and incomplete separation resulted. However, when small
amounts of inexpensive magnetic oxide were added to the suspended
carbon in combination with a flocculating agent, the resulting floe
became magnetic and could be quantitatively removed in a continuous
flow magnetic separator. The same results have been obtained with
colloidal precipitates.
In order to simultaneously remove the magnetic floe of suspended
solids and dewater the solids, a particular kind of magnetic separator
was designed. Figure 8 shows a diagram of the continuous flow
magnetic thickener/separator used to separate flocculated suspended
solids containing magnetic material. Figure 9 is a photo of the actual
unit capable of handling a flow of about 5 gallons per minute. Liquid
effluent from the Dynactor containing activated carbon, magnetic material
and a polyelectrolyte flocculating agent is allowed to flow by gravity
through the orifice or "slice" of the head box under the moving Mylar
belt suspended below the magnet structure. The magnetic floe of
carbon or other suspended solids is attracted nearly instantaneously
to the bottom side of the moving belt due to the presence of the magnet
structure suspended above this element. Clarified water flows down
to a sump and can be gently released back into the stream. Dewatered
solids material is continuously scraped off the moving belt after the
belt has passed by and away from the magnet structure. Thickened
solids, carbon, and magnetic material are collected in plastic barrels
and held for proper disposal of the hazardous material in question.
Separation of suspended solids in a magnetic floe by the above-
described magnetic separator is rapid, efficient, continuous and produces
a superior water quality effluent while effecting dewatering of the removed
solids.
16
-------
2-4%
SOLIDS
SLUDGE
(TEMPORARILY
MAGNETIC)
V
HEAD-
BOX
"SLICE"
V
TROUGH
MAGNET
-c.
CLARIFIED WATER
•CONTINUOUSLY MOVING BELT
TEMPORARILY MAGNETIC
'THICKENED SLUDGE
\\fe-DOCTOR KNIFE
THICKENED
SLUDGE,20-40%
SOLIDS
FIGURE 8. CONTINUOUS FLOW MAGNETIC THICKENER/SEPARATOR
-------
FIGURE 9. MAGNETIC SEPARATOR MODEL 5 GPM
18
-------
SECTION V
EXPERIMENTAL STUDIES
The Dynactor model test system was housed in a modern chemistry
laboratory equipped with hood, floor drain, safety devices and bench
space for analytical work. Six Dynactors were constructed for engineering
and decontamination testing. These were constructed of polypropylene
and polyvinyl chloride . One complete experimental system was installed
in the laboratory and consisted of (a) one 5 gpm capacity Dynactor,
(b) plenum chamber, (c) sump, (d) exhaust line, (e) 25 gallon capacity
reservoir, (f) a 5 gpm, 100 psi displacement pump, (g) liquid metering
system, (h) turbulent mixing chamber, and (i) powder feed mechanism.
These components are identifiable in Figures 4 and 7.
In operation, the 25 gallon reservoir contains water contaminated
with a predetermined concentration of the selected hazardous material
to simulate a spill in confined waterways. The contaminated water
is then pumped through the nozzle at the top of the Dynactor at a pressure
of about 100 psi. The liquid pressure can be varied from 40 to 140
psi with the model equipment producing a liquid flow rate of about
3 to 5 gpm. Since the Dynactor is equipped with liquid, gas and powder
metering systems, decontaminating agents such as lime, carbonate,
bicarbonate, ozone, chlorine, acetic acid, powdered carbon and other
adsorbents can be metered in stoichiometric quantities into the flow
of contaminated water through the Dynactor.
Insoluble precipitates formed during decontamination reactions
and powdered carbon containing adsorbed toxic materials must be removed
from the reactor effluent after decontamination has been achieved.
Therefore, the new process, described in Section IV, that renders non-
magnetic solid materials temporarily magnetic and capable of continuous
magnetic separation was evaluated.
The types of decontamination reactions studied in the Dynactor
system were aeration, ozonation, neutralization, precipitation and carbon
adsorption. It is also possible to carry out combination reactions such
as aeration, neutralization and carbon adsorption essentially simultaneously
in one passage of contaminated water through the system.
19
-------
Since the purpose of this project is to demonstrate feasibility
of the system, only a few hazardous materials were selected that could
be easily handled in the laboratory and analyzed by standard procedures.
Acids, alkalies, chlorine, cyanide, phenol, lead salts, various pesticides
and water saturated with fuel oil were selected for study. The major
emphasis was placed on defining the types of reactions that could be
carried out in the system in a continuous flow-through configuration.
Aeration
Since the Dynactor is a thin film aeration device capable of aspirating
large volumes of air per volume of motive liquid, a typical flow-through
experiment results in an effluent completely saturated with dissolved
oxygen. Repeated experiments for oxygenation capability raised the
dissolved oxygen concentration of water from initial levels of 1 ppm
to 10 ppm in a single pass through the Dynactor. Supersaturation
of the effluent with dissolved oxygen was always achieved. Field
samples of Des Plaines, Illinois, River water containing 5 mg/liter
dissolved oxygen were aerated to 9 mg/liter (20°C) by passage through
the Dynactor.
Simple, effective, continuous flow, high speed aeration of a
waterway is in itself a significant contribution to water quality by
increasing the dissolved oxygen content of streams, lakes, or lagoons.
Many substances are toxic because they can threaten water quality
by reducing the dissolved oxygen content. This can adversely affect
aquatic life. There is general agreement that it is desirable to maintain
high dissolved oxygen levels in water while the spilled material
is being dispersed, decontaminated, and degraded aerobically.
Ozonation and Chlorination
The Dynactor used for gas experiments contains a plenum surrounding
the top air intake section. The plenum has a 16-inch intake port which can
be baffled to meter measured quantities of gases from compressed
cylinders. The unit can also be used in a completely closed circulation
system by connecting the gas outlet from the reservoir to the plenum intake
port. Thus, ozone, for example, can be recirculated for optimum uptake
and utilization by the thin film of water in the reaction column of the
Dynactor.
Ozone was delivered to the sealed plenum of the Dynactor from
a Weisbach Ozonator through a Tygon tube at rates of approximately 7
liters per minute containing approximately 15 mg Oo per liter. Maximum
absorption of ozone was demonstrated in the effluent. The conversion
20
-------
of potassium iodide to iodine was studied by processing 25 gallons of
KI solutions through the Dynactor while bleeding measured quantities
of ozone into the closed plenum. The iodine was quantitated by sodium
thiosulfate titration. Complete conversion of KI to 1^ was achieved in
a single pass through the Dynactor with significantly less than stoichiometric
quantities of ozone.
Chlorine and oxygen can also be metered into the Dynactor for
thin film gas liquid contact where needed for oxidative decontamination
of hazardous materials.
Neutralization
Since the Dynactor is equipped with both a liquid metering system
and a powder feed system, saturated sodium bicarbonate solutions and
powdered sodium bicarbonate were individually studied as neutralizing
agents for acetic and hydrochloric acids. Each form of bicarbonate effectively
neutralized the acid representing the spilled hazardous material. For
example, stoichiometric quantities of powdered sodium bicarbonate were
aerosolized into the plenum of the Dynactor and instantly neutralized
0.0IN acetic acid at a flow rate of 5 gpm at a head pressure of 100 psi.
pH, titratable acidity and odor measurements were recorded on the effluent
from the Dynactor to substantiate complete neutralization. Repeated experiments
have been made with higher concentrations of acetic and hydrochloric
acids with effective neutralization.
The successful use of powdered sodium bicarbonate, sodium carbonate
and calcium hydroxide as dry neutralizing agents in the continuous
flow configuration has proven the feasibility of this system for neutralizing
acids and alkalies by processing the contaminated water with dry neutralizing
agents.
Precipitation
Stoichiometric quantities of heavy metal precipitating chemicals
such as sodium carbonate, sodium bicarbonate, lime, and sodium sulfi.de
were metered into the system to precipitate toxic heavy metals such
as lead, mercury, cadmium and zinc. The precipitating chemical agent
was used in the powdered form whenever possible. For example,
water containing 100 ppm of lead nitrate was pumped through the Dynactor
while powdered sodium bicarbonate was aerosolized into the 5 gpm
continuous flowing system. Lead carbonate precipitate was immediately
observed in the effluent and a filtered sample showed 2 ppm of lead
by atomic absorption analysis. Most of the toxic heavy metals can
be reduced to safe levels by precipitation and removal of a highly insoluble
salt.
21
-------
Carbon Adsorption
Because powdered carbon treatment of contaminated water is
a recognized effective countermeasure for the removal of about 80 percent
of the hazardous chemicals of high ranking importance in the Battelle
Report (1), significant effort has been devoted to the perfection of
a process for continuous flow carbon treatment of hazardous material
contaminated waters. The powdered activated carbon studied was Nuchar
190-N produced by West Virginia Paper and Pulp Company. This carbon
was chosen for the initial studies because extensive data on pesticide
adsorption by Nuchar 190-N was available from scientific literature(Z),
Powdered activated carbon was metered into the Dynactor as a
10 percent aqueous suspension and as a dry powder. The dry powder
is the preferred method of treatment for a mobile field application.
Using the powder feed mechanism designed for the Dynactor, the activated
carbon is aerosolized directly into the throat of the unit making intimate
contact with the thin film of flowing liquid ejected from the nozzle.
The wetting and dispersion of the carbon by the turbulent thin film
contact is excellent at a liquid flow rate of 4 gallons per minute and
a carbon metering rate of about 15 to 20 grams per minute. The adsorption
of phenol, Ethion, DDT, Toxaphene and chlorine has been extensively
studied and will serve as examples to demonstrate the feasibility of
the Dynactor as a continuous flow carbon treatment unit for the removal
of hazardous materials from contaminated water. Initial and final concentra-
tions of phenol were assayed by gas chromatography and the pesticides
by electron capture gas chromatography after concentration by extraction
in hexane. Chlorine was measured by the iodometric method. The
effluent from the Dynactor was immediately filtered through a 0.45
micron Millipore filter to remove the carbon and analyzed.
The results on the adsorption of phenol are shown in Figure
10. These data were obtained using a 10 percent suspension of Nuchar
190-N metered into the Dynactor; comparable data have also been obtained
using powdered Nuchar 190-N aerosolized into the unit. Adsorption
of phenol from the contaminated water is achieved in less than 2 minutes
and greater than 90 percent removal of phenol is obtained with carbon
to phenol weight ratios of about 20 to 1. Phenol adsorption as a function
of initial carbon to phenol ratio is shown in Figure 11.
The adsorption of representative pesticides by Nuchar 190-N was
determined experimentally in the Dynactor. Ethion, DDT, and Toxaphene
were selected for study and quantitatively analyzed by electron capture
gas chromatography. Multiple experiments were conducted varying
the concentration of pesticide and the amount of powdered carbon.
22
-------
Experimental Conditions:
3 liters 10% Nuchar 190-N metered into
5 gal/min flow rate of 25 gallons of phenol
solutions processed by Liquids Dynactor
o
IQ
o\9
100
90
80
70
60
50
40
30
20
10
0
100 200 300 400
Initial Phenol Concentration (ppm)
500
600
FIGURE 10. PHENOL ADSORPTION VS. INITIAL PHENOL CONCENTRATION
23
-------
Carbon = Nuchar 190-N processed as a
10% slurry through Liquids
Dynactor
10 20 30 40 50 60 70 80 90 100
Ratio: Carbon/Phenol
FIGURE 11. PHENOL ADSORPTION AS A FUNCTION OF INITIAL
CARBON TO PHENOL RATIO
24
-------
The carbon was aerosolized into the Dynactor by the powder feed mechanism
and the pesticide contaminated water was pumped through the system
at about 100 psi and a flow rate of 4 gpm. The carbon was separated
from the Dynactor effluent by Millipore filtration immediately upon collection.
Table 1 shows the adsorption data obtained for each of the pesticides.
The system worked well in removing greater than 99 percent of the
initial pesticide concentration, leaving only parts per billion concentration
in the effluent.
Pesticides in general present particular problems when spilled
in waterways. Initial concentrations of soluble pesticides are usually
very low (1 to 10 ppm) , but toxicity to fish and other organisms can
be substantial at 0.05 ppm and less (3). It appears advisable to use
excessive ratios of carbon to pesticide in order to obtain an effluent
having a safe pesticide level. Since initial concentrations of pesticides
are likely to be in the 1 to 10 ppm range because of limited solubility,
the total quantity of carbon required is not great even at excessive
ratios.
Several chlorine removal experiments were conducted using Nuchar
190-N processed by the Dynactor system. The chlorine concentration
before and after removal was measured by the iodometric method.
Chlorine was diffused into the 25 gallon reservoir from a compressed
cylinder of chlorine. Table 2 shows the results of the chlorine adsorption
experiments. Chlorine is easily removed by the short time contact
with powdered carbon in the Dynactor. Relatively small quantities of
carbon are needed to effect 99 percent removal of chlorine from contaminated
water.
Very preliminary experiments were made on removal of water
soluble components of No. 2 fuel oil by treatment in the Dynactor with
powdered carbon (Nuchar 190-N). Water was saturated with fuel oil
by mixing overnight and allowed to separate. The water soluble fraction
was drawn off and this was processed in the Dynactor using powdered
carbon in the powder feed unit. Initial and final COD measurements
were used as a semiquantitative guide to removal of water soluble hydrocarbons
Perception of fuel oil odor after treatment was also used as a sensitive
index of removal. Results obtained are shown in Table 3. The data
indicate that about 200 mg/liter of carbon will remove essentially all
of the water soluble hydrocarbons from fuel oil. More quantitative
work will be required to establish carbon quantities for less than saturated
solutions.
Some preliminary work was also completed on carbon removal
of oil in water emulsions. Oil in water emulsions was produced from
No. 2 fuel oil in a mechanical homogenizer; 5, 2.5, 1.0, 0.5 and 0.1
percent oil to water ratios were prepared. Varying amounts of powdered
25
-------
TABLE 1
SJ
Adsorption of Pesticides in Dynactor System
Using Powdered Activated Carbon
Pesticides
Ethion
Ethion
Ethion
Ethion
Ethion
DDT
DDT
DDT
DDT
DDT
Init. Pest.
Cone . (ppm)
18
13
8.5
5
1.6
2.00
1.00
0.45
0.19
0.098
Final Pest.
Cone . (ppb)
11
28
4
4
6
17
0.65
0.49
0.75
0.40
Carbon
Cmg/D
1600
100
1000
750
200
500
700
1000
600
600
% Removal
99.9
99.8
99.9
99.9
99.6
99.2
99.9
99.9
99.6
99.6
C/Pest.
Ratio
89
8
118
150
125
250
700
2200
3200
6000
Toxaphene
1.1
2.5
700
99.8
650
-------
TABLE 2
Adsorption of Chlorine in Dynactor System
Using Powdered Activated Carbon
Initial
C\2 Cone.
mg/1
123
162
214
278
Final
Effluent
Cone, mg/1
0.01
0.01
0.57
6.57
Carbon
Used mg/1
1500
1300
800
700
%
Removal
99.99
99.99
99.7
97.6
C/C12
Ratio
12.0
8.0
3.7
2.5
0.61 Tap Water
27
-------
TABLE 3
Removal of Water Soluble Components
Fuel Oil No. 2 - Carbon Adsorption
Dynactor Processed
Initial
C.O.D.
296
43
11
C.O.D.
15
3.5
0
Carbon Used
mg/1
1500
1500
1500
Tap Water C.O.D. - 5.9
Mg Carbon/1 Removal of Odor
1500 Yes
1000 Yes
500 Yes
250 Yes
200 Yes
100 Faint Trace
28
-------
powdered carbon were used in the Dynactor to establish an effective
ratio of carbon to oil emulsion for the removal of emulsified oil. Turbidity
and odor measurements were made before and after carbon treatment
as indices of removal. The results are shown in Table 4. It appears
that a carbon to oil ratio of about 2 to 1 will remove essentially all
of the fuel oil and water soluble components from emulsions up to 5
percent. Higher concentrations were not tested and quantitative analytical
studies not conducted within the scope of this program. These experiments
with oil are preliminary in nature and are intended to examine the
feasibility of carbon removal of components of oil in water by Dynactor
processing.
Magnetic Separation of Carbon
To provide a complete system for decontamination of hazardous
materials, it is necessary to remove undesirable end products from
the effluent of the Dynactor such as precipitates and spent carbon.
The program originally proposed, in Contract 68-01-0123, the development
of a static sealed centrifuge designed as a Solids Statifuge to remove
solids and a Gas Statifuge to remove gases. Both of these units were
designed, fabricated and tested in accordance with the contract. The
Gas Statifuge failed to effectively remove dissolved gases from decontami-
nation reactions and was therefore not further tested.
Exhaustive testing of the Solids Statifuge yielded only about
75 percent removal of the flocculated carbon from the aqueous effluent
of the Dynactor after activated carbon had been used to adsorb toxic
materials processed in the system. The reason for this limited carbon
removal by the Statifuge is the use of finely divided powdered activated
carbon which must be flocculated from suspension prior to removal.
The flocculated carbon is lacy in physical structure, and hence, very
fragile and easily redispersed. The shear forces of the Statifuge break
up part of the flocculated carbon and some of the dispersed carbon
particles escape removal and appear in the effluent from the Statifuge.
Such an effluent would not meet water quality standards, and therefore,
another method of carbon removal was considered essential to the program.
It is important to realize that for the Dynactor system to perform in
a mobile field configuration, the entire decontamination process, including
removal of the carbon, must perform on a continuous flow basis without
the need for large settling tanks or columns.
Substantial efforts were expended to find a satisfactory continuous
flow carbon separation method suitable for field use. Separation was
studied in commercial cyclone separators, commercial continuous flow
centrifuges, commercial filters and finally a magnetic separator. Cyclones,
centrifuges and filters yielded only partial removal, but magnetic separation
gave essentially 99 percent removal of the flocculated carbon. Twenty-
five percent magnetic material in the form of finely divided iron oxide
29
-------
TABLE 4
Removal of Oil in Water Emulsions
Carbon Adsorption - Dynactor Processed
Carbon/Oil % Odor % Turbidity
Ratio Removed Removed
0.1 50 10
0.2 50 20
0.5 80 30
0.7 85 60
1.0 100 85
1.5 100 95
2.0 100 100
30
-------
is added to the powdered carbon used to feed into the Dynactor for the
adsorption of toxic materials; 2 ppm of a polyelectrolyte flocculating
agent (Nalco 8173) is added to the suspended carbon in the effluent
from the Dynactor, and the magnetic particles are trapped in the lacy
floe.
In order to simultaneously remove the magnetic floe of suspended
solids and dewater the solids, the magnetic separator described in
Section IV was used.
The decontamination of hazardous material spills often results
in the formation of fine colloidal precipitates such as heavy metal
carbonates and sulfides. These fine precipitates are difficult to remove
by settling or filtration. Magnetic separation studies were applied
to the separation of lead carbonate. Magnetic iron oxide can be added
to the sodium bicarbonate prior to the formation of lead carbonate precipitate
or after the colloidal lead carbonate precipitate has been formed. Amounts
of magnetic material required are about three times the amount of lead
carbonate precipitate. A polelectrolyte flocculating agent such as Nalco
8173 is added to the suspended precipitate at a concentration of 2 to
3 ppm and mixed for 10 seconds; the magnetic floe formed is then readily
removed by passing through the magnetic separator.
Although magnetic separation studies were limited to carbon and
lead carbonate, it is anticipated that all types of suspended solids can
be removed from water by the magnetic separation process described.
Optimum proportions of flocculating agent and magnetic material may
vary with the amount and type of suspended solids present, but the
basic principle appears applicable.
31
-------
SECTION VI
REFERENCES
1. Dawson, G. "W. , A. J. Shuckrow, and W. H. Swift, Control
of Spillage of Hazardous Polluting Substances, Battelle Memorial
Institute, FWQA 15090 FOZ, October 1970.
2. Schwartz, Jr., Henry G., Adsorption of Selected Pesticides on
Activated Carbon and Mineral Surfaces, Environmental Science
and Toxicology 1. 332-337, 1967.
3. McKee, J. E., H. W. Wolf, Water Quality Criteria, 2nd Ed.,
Resources Agency of California, 1963.
32
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
I. REPORT NO.
EPA-67Q/2-75-QQ4
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
FEASIBILITY OF 5 gpm DYNACTOR/MA6NETIC SEPARATOR
SYSTEM TO TREAT SPILLED HAZARDOUS MATERIALS
5. REPORT DATE
April 1975; Issuing Date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S),
Robert G. Sanders, Industrial Bio-Test Lab.,Inc
Stanley R. Rich and Thomas G. Pantazelos, RP Industries,
8. PERFORMING ORGANIZATION REPORT NO.
Inc.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Industrial Bio-Test Laboratories, Inc., Northbrook,
111. 60062
Through subcontract with RP Industries, Inc.,
Hudson, Mass. 01749
10. PROGRAM ELEMENT NO.
1BB041; ROAP 21AVN; Task 022
11. CONTRACT/GRANT NO.
68-01-0123
12. SPONSORING AGENCY NAME AND ADDRESS
National Environmental Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14.
Una I Rep
SPONSORING^
AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Design and operating details are given for a new type of continuous flow thin-
film, gas-liquid-particulate contact device called the Dynactor. The device is used
as a continuous flow short-time contact reactor to effectively decontaminate water
contaminated with spilled hazardous materials. The decontamination is effectively
achieved by one or more processes involving oxidation, neutralization, precipitation
or adsorption on powdered carbon. Contaminated water is processed by the pilot plant
model Dynactor at 100 psi and at a rate of 5 gpm; stoichiometric quantities of decon-
taminating agents in the form of gases, liquids, slurries or powders are metered into
the continuously flowing liquid configuration. The device is portable, lightweight
polypropylene construction, has no moving parts, requires a pump for liquid motive
power and can be scaled up to process 250 gpm of contaminated water.
Design and operating details are given for continuous flow magnetic separation
to remove flocculated carbon and precipitates from the Dynactor effluent after decon-
Itamination of hazardous materials. . . . . . .. ..
Experimental data on successful decontamination of heavy metals by precipitation,
acids aKd bases by neutralization, phenol, chlorine and pesticides by powdered carbon
adsorption and other selected hazardous compounds are presented.
17.
KEY WORDS AND DOCUMENT ANALYSIS
b.lDENTIFIERS/OPEN ENDED TERMS |c. COSATI Field/Group
DESCRIPTORS
*Water treatment, *Decontamination,
*Magnetic separators, *Activated
carbon treatment, Water pollution,
Chemical removal (water treatment),
Separation, Neutralizing,
Precipitation (chemistry)
Hazardous materials spill
cleanup, Hazardous materials
spill control, Hazardous
polluting substance spills,
Hazardous chemical spills,
Dynamic reactor, Dynactor
13B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
41
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
£ U. 5. 60YKMIWT PMHTIK6 0FFK£: 1975-657-591/53*7 R«glon No. 5-11
------- |