WATER POLLUTION CONTROL RESEARCH SERIES • 15080 DJM07/70
**C,
USING VORTEX ASSISTED
AIRLIFT SYSTEM
ENVIRONMENTAL PROTECTION AGENCY • WATER QUALITY OFFICE
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes
the results and progress in the control and abatement
of pollution in our Nation's waters. They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Water Quality
Office, Environmental Protection Agency, through inhouse
research and grants and contracts with Federal, State,
and local agencies, research institutions and industrial
organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Head, Project Reports
System, Office of Research and Development, Water Quality
Office, Environmental Protection Agency, Room 1108,
Washington, B.C. 20242.
-------
RECOVERY OF OIL SPILLS
USING' VORTEX ASSISTED AIRLIFT SYSTEM
PACIFIC NORTHWEST LABORATORIES
a division of
BATTELLE MEMORIAL INSTITUTE
P. 0. Box 999
Richland, Washington 99352
for the
WATER QUALITY OFFICE
ENVIRONMENTAL PROTECTION AGENCY
Project # 15080DJM
Contract # U-12-513
JULY, 1970
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price 50 cents
Stock Number 5501-0069
-------
EPA Review Notice
This report has been reviewed by the Water
Quality Office, 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.
-------
ABSTRACT
Studies were conducted to determine the feasibility of a concept for
recovery of floating oil slicks which utilizes a pump induced vortex
and a vacuum suction or Coanda nozzle. The apparatus used for develop-
mental experimentation comprised a pumping system for vortex production,
a large water basin, a flapper type wave generator, and several config-
urations of the experimental assembly. The range of influence was
smaller than was anticipated. Approximately a 25 foot influence diameter
was achieved for the maximum strength vortex generated in this apparatus.
Extrapolation of measured performance data showed that a 1/8 inch thick
slick could be recovered at the rate of 960 gallons per hour. Experiments
with and without a variety of oils showed that enhanced oil recovery rates
with the vortex was due entirely to the surface current generated by the
vortex. This effect was found to improve oil recovery by a factor of
7.9 above the rates achieved with a suction nozzle alone. The surface
position of the vortex cavity was found to be sensitive to surface waves.
The cavity moved in a circular path within three vortex cavity radii of
the still water cavity location as a wave passed through the assembly.
Tests with a Coanda nozzle (a fluid attachment eductor) showed improved
performance in surface waves. However, the recovered oil-water mixture
was highly emulsified.
This report was submitted in fulfillment of contract 14-12-513 between
Pacific Northwest Laboratories and FWPCA.
Key Words - Vortex, Coanda Nozzle, Wave Suppressor
-------
CONTENTS
Section
I Conclusions
II Recommendations
III Introduction
IV System Investigation
V Description of Experimental Apparatus
VI Characterization of Vortex
VII Oil Collection Experiments
VIII Coanda Nozzle Experiments
IX Acknowledgments
X Glossary
XI Bibliography
Page
1
2
3
5
8
9
11
14
15
16
17
-------
FIGURES
1 SKETCH - DEPLOYED REFERENCE SYSTEM
2 GRAPH OF CRITICAL SUBMERGENCE FOR 18 INCH DIAMETER
INTAKE AS A FUNCTION OF FLOW
3 PHOTOGRAPH SHOWING VORTEX STRUCTURE ASSEMBLY
4 CROSS SECTION OF VORTEX GENERATING SYSTEM
5 SCHEMATIC OF VORTEX AND AIRLIFT SYSTEM
6 PHOTOGRAPHS OF 6000 GPM VORTEX AT THREE LEVELS OF
SUBMERGENCE
7 VORTEX PROFILES AT 2000, 4000, 6000 GPM FOR THREE
LEVELS OF SUBMERGENCE
8 GRAPH OF WAVE SUPPRESSION
9 PHOTOGRAPHS OF VORTEX MOVEMENT
10 GRAPH SHOWING SURFACE VELOCITY VS RADIUS
11 GRAPH SHOWING RANGE OF INFLUENCE AS A FUNCTION OF VORTEX
FLOW WITH THE LOWER CYLINDER BOTH INSTALLED AND REMOVED
12 GRAPH COMPARING MEASURED SURFACE VELOCITY DUE TO THE SUCTION
NOZZLE ONLY WITH THE VALUE FROM EQUATION (1), K = 5
13 GRAPH SHOWING CALCULATED OIL RECOVERY RATES AS A FUNCTION
OF OIL SLICK THICKNESS
14 CALCULATED OIL-WATER RATIO VS SLICK THICKNESS FOR 6000
GPM VORTEX
15 DRAWING OF FOUR INCH SUCTION NOZZLE
16 DRAWING OF THREE INCH COANDA NOZZLE
17 GRAPH SHOWING LIQUID RECOVERY RATE AS A FUNCTION OF AIR
SUPPLY FOR A THREE INCH COANDA NOZZLE
18 GRAPH SHOWING MAXIMUM OIL RECOVERY RATE AS A FUNCTION OF
OIL SLICK THICKNESS FOR A THREE INCH COANDA NOZZLE
-------
SECTION I
CONCLUSIONS
1. Recovery of floating oil slicks by means of a suction nozzle is
technically feasible. The possible rate of recovery is strongly
influenced by oil characteristics and by the precision with which
the suction nozzle position is maintained relative to the liquid
surface. Liquid recovery rates average about one percent of the
air flow rate. The oil recovery rate of suction nozzles is degraded
by the presence of waves. The use of a circular wave suppressor
damped short period two foot waves by about 35 percent. This was
not sufficient to allow efficient oil slick recovery.
2. Recovery of floating oil slicks by use of a low head-high flow sur-
face attachment device (the Coanda nozzle) is of questionable
practicality because the recovered oil is intimately mixed and
emulsified with entrained water. This would greatly complicate
subsequent steps in the recovery process compared to mechanical
recovery methods which do not cause this effect. The emulsion
formation results from atomization of oil and water as very small
droplets during entrainment from the liquid surface.
3. The use of a vortex will expand the area of influence of an oil
recovery suction nozzle up to a factor of 7.9 over a suction nozzle
alone. With a four inch suction nozzle, such a vortex assisted
system collected medium gravity crude oil at a rate of 960 gallons
per hour in calm water conditions.
4. The position of the vortex cavity on a liquid surface is shifted as
a wave passes. This shift in position is a transient circular
displacement of the vortex cavity; the vortex cavity returns to the
original centered position after a wave passes.
-------
SECTION II
RECOMMENDATIONS
This program was limited to the investigation of a vortex
assisted suction nozzle concept for floating oil slick recovery. It
involved pilot scale experiments to test critical components, define
limiting performance' characteristics, and assess the effects of surface
perturbations (waves) with regard to a reference system concept. While
a vortex was found to enhance suction nozzle performance, the influences
of-waves on the overall system suggests that means other than a suction
nozzle be sought for removal of surface oil. Specifically, it is recommended
that a liquid pumping system be utilized whose suction head is in the form
of a floating circular channel. The diameter of the circular head would
be such that it would enclose the possible path of the vortex when it is
subjected to wave action. The liquid pumping system would also be less
vulnerable to malfunction, by flooding during surface disturbances, than
air operated suction devices. Evaluation of this alternative could be
efficiently accomplished by modification of the existing experimental
apparatus.
Construction and sea testing of a full scale operational proto-
type oil recovery system is recommended contingent on successful implemen-
tation of a liquid pumping system. This phase of the work should incorporate
evaluation under representative environmental conditions, spill materials,
and with typical offshore supply vessels and work boats.
The interesting ability of the Coanda nozzle to atomize liquids
as very fine droplets suggests that this might be a useful technique
for efficiently burning oil collected by the system investigated in this
study or by other techniques. While the limiting ratios of water to oil
and oil to air for successful complete combustion are now unknown, a
preliminary investigation of this possibility seems justified. Success-
ful development of this technique would eliminate the need for large
storage tankage for several oil recovery schemes under development and
would thereby improve the economics and efficiency of these methods.
-------
SECTION III
INTRODUCTION
Accidents resulting in massive release of petroleum products
to the sea, such as the TORREY CANYON stranding and Santa Barbara Channel
incidents, have elucidated, among other things: (a) the large expenditure
required for emergency cleanup operations, and (b) the absence of satis-
factory equipment to cope with relatively large volumes of released oil.
Various mechanical and chemical techniques used to clean up small oil spills
in the past were proven incapable of coping with large amounts of oil under
open sea conditions.
The Battelle-Northwest Oil Spillage Study (1), performed for
the U.S. Coast Guard, included a comprehensive review and critical
evaluation of the state-of-the-art in oil spillage prevention, cleanup,
control, effects, and disposal. Techniques were described which offered
promise for further development. This study concluded that mechanical
techniques are likely to be most effective when large amounts of oil are
involved. Furthermore, mechanical techniques do not introduce additional
insult to the environment such as that possible with the use of detergents
or other treating agents. Commercial equipment is presently available or
being developed, both mechanical and chemical, that can satisfactorily
cope with small oil spills (less than a few thousand gallons) in sheltered
waters; therefore the work described in this report was directed toward the
development of equipment to mechanically recover large quantitites of oil
from unprotected waters. Such a system would be applicable to smaller
spills; however, it would not necessarily be the most optimum, economically,
for small spills in sheltered waters.
Numerous mechanical approaches to recovery of oil spills from
open waters have been tried and evaluated. Among these is the technique
of lifting oil from the water surface by use of a suction nozzle. As air
is sucked into a nozzle, a portion of the liquid surface layer in close
proximity is entrained and collected in an air-liquid separator tank. The
concept is analogous to a vacuum cleaner removing dust from a rug.
Recovery rates as high as 5 imperial gallons per minute have been
reported for 3-inch suction nozzles (•*-/ . Most types of oil can be lifted
from the surface provided the oil flows as a continuous film to the suction
nozzle. For heavy viscous oils, such as Bunker C, the rate of oil pickup
is limited by the slow rate of spreading. Too rapid an air flow rate,
necessary to attain high recovery rates, causes the oil slick to "tear"
apart close to the nozzle. To continue recovery, one must wait until the
oil spreads beneath the nozzle again or reposition the nozzle. Recovery
rates for thin, rapidly spreading oils are limited by the thickness of the
slick beneath the nozzle.
-------
Oil recovery operations on the sea face the ever present pertur-
bations of wind and waves. To function properly the nozzle must be main-
tained at a fixed position relative to the surface. Depending on the
operating pressure, the working distance varies from one-half to one
inch above the water surface. Such close proximity to the surface makes
nozzles extremely sensitive to wave motion. Several methods may be used to
circumvent this difficulty. In particular, high flow-low head devices
(the Coanda nozzle is a classical example) capable of immediately shutting
off when a wave washes over the nozzle appears to be a good approach. This
technique was studied during the work reported herein.
The effective range of influence of suction nozzles, while theoreti-
cally very large, is only about five to ten nozzle radii. The term "effective
range" is used here to indicate the furthermost point at which surface
movement is caused by the action of the nozzle. One approach (the one used
in this study) to increase the effective range is to generate a vortex
directly beneath the nozzle. In the presence of a vortex, it was thought
that the surface current generated would extend much further than that
generated by a nozzle alone.
-------
SECTION IV
SYSTEM INVESTIGATION
As a first step in the development program, operational
requirements and system criteria were developed from review of oil spill
histories, oil recovery equipment development and manufacture equipment
specifications. The objective was to delineate the environmental
conditions under which oil recovery systems must operate, physical con-
straints dictated by the characteristics of support vessels available
at most harbors and ports, and the types of material most likely to be
involved in a marine oil spill.
The spectrum of environmental conditions which may prevail
during an oil spill incident is legend. However, environmental conditions
under which oil recovery operations may logically be undertaken are
determined by the need to limit the risk to life during storms; the
fact that most spills from marine vessels occur within the mitigating
influences of close land masses or in harbor or channel entrances, and;
a significant portion of spill sources lie on land which ajoins the sea
or waterways and would not require recovery operations under unlimited
open sea conditions. In spite of these factors, of course, the ideal
oil recovery system would be capable of operation under any conceivable
combination of environmental conditions. However, to realistically
reflect performance characteristics which are practically attainable, ,„ _,
those factors mentioned above were combined with published materials '
which list various typical environmental conditions during marine spill
incidents.
The characteristics of boats and ships available for emergency or
planned use in oil spill cleanup operations spans the range from small
pleasure craft to large ships in the thousands-of-ton category. For this
study, the types and characteristics of offshore work boats available for
charter in ports and harbors was considered as typical. These boats are
in the 90 to 140 foot range length, 80 ton displacement, have deck spaces
on the order of 300 square feet, and are equipped with moderately sized
liquid fuel tanks and pumping capacity. Limitations due to the work boat
characteristics are minimal for the type of system considered in this study
-------
Materials involved in marine spills would be expected to be predominately
crude oils^) . Attention must be given to equipment performance on other
materials such as refined and residual fuel oils, however, to assure maximum
utility of systems to be developed. The results of this phase of the systems
study are summarized as follows:
ENVIRONMENTAL AND OPERATIONAL CHARACTERISTICS
A. Effective Wave Height and Spectrum - Random Sea
1. Height — ^4'
2. Period — 5 sec
3. Current — 0-1 knot
4. Wind Speed ^ 18 mph
5. Debris Present - Kelp, seaweed and small logs and floating drift-
wood
6. Temperature — 50-100°F
B. Characteristics of Oil
1. Types — Crude Oil, fuel and diesel, residual petroleum fuels
2. Density — .85 - .92
C. Slick Thickness — 0 - 1"
PERFORMANCE CRITERIA
A. Rate of Oil Removal
60-300 gpm
B. Suction Nozzle Characteristics
1. Air velocity — 73-290 fps
2. Diameter — 3-9 inch
3. Separation from water surface — 1/4" - 1"
4. Geometry — Fluted conical
C. Vortex Parameters
1. Flow — to 6000 gpm
2. Intake depth — 0-4 ft.
3. Intake diameter — 18" max.
4. Diameter of effect — ^6.8 ft. - 20 ft.
-------
D. Storage Capacity and Separation
1. Rate of oil removal — 60-300 gpm
2. Oil water ratio — (assume 20:1 mixture water; oil)
E. Buoyant Stability
1. Surface heaving <0.5"
The experimental system intended to meet these criteria is outlined below:
Wave Suppressor Diameter - 6 Feet
Intake Piping Diameter - 18 Inches
Lower Cylinder Height - 24 Inches
Vortex Flow - 6000 gpm
Airlift Flow - 5000 scfm
Minimum Airlift Head - 200 Inches H20
Airlift Recovery Hose
Diameter - 9 Inches Maximum
Nozzle Proximity to
Surface - 1.0 Inches +0.5 In.
Nozzle proximity to the water surface can be maintained by adjustment of
the mass volume ratio of the floating portion of the system.
Calculations were made to predict the amount of heaving expected of the
system when operating in rough seas. To insure that a uniform cylindrical floating
section does not heave more than 0.5" in 5 ft. waves, a weight to volume ratio
of 5.7 Ib/ft is required. This ratio was determined by solving the equation of
motion for a floating buoy with an overall dimension small compared to the wave
length.(4).
The reference system envisioned to meet the performance and other criteria
is sketched on Figure 1. It consists of a floating assembly connected to an
offshore type work boat. Vacuum pumps for operation of the suction nozzles and
liquid handling pumps for producing a vortex are mounted on the work boat and
connected to the floating assembly by hoses. Tanks for receiving recovered oil
are mounted on the work boat or towed. The floating assembly consists of a
wave suppressor, the suction nozzle, and the vortex producing assembly. The
latter would be the open end of a suction pipe with or without flow direction
controlling vanes.
-------
SECTION V
DESCRIPTION OF EXPERIMENTAL APPARATUS
Experimental apparatus was assembled for investigation of the
various parameters expected to effect the reference system. A large water
basin equipped with a wave generator was basic to the subsequent experi-
ments .
The vortex generating apparatus consisted of an 18 inch diameter
intake piping section with its open end located beneath the water basin
surface. The depth of the intake was established from:
H - 2.64 D1/2 V 1/4
where
H = Distance between the water surface and intake pipe
D = Intake pipe diameter
V = Mean velocity across the intake pipe
This equation was derived by Springer and Patterson . Figure 2 shows
a plot of this equation. This piping section was routed through a water
tight fitting into the pump house and connected to the suction side of a
6000 gpm, 200 hp centrifugal pump. The pump discharge was fitted with
facilities for pitot tube flow sensing and liquid sampling. The pump
discharge was routed back into the water basin through a water tight
fitting, discharging beneath the basin surface. Eighteen inch diameter
gate valves were provided on both the pump suction and discharge lines for
control of the system flow. Figure No. 3 shows a photograph of the 18-inch
intake piping section, the wave suppressor and the supporting structure.
The wave suppressor was a six foot O.D. circular slotted baffle fitted with
angular plates to impart lateral motion to a particle of water entering on
a radius to the vortex. An additional right cylindrical section with a
closed bottom was provided to direct all or portions of the intake flow
through the wave suppressor. Thus full control of the strength of the
vortex was achieved by raising or lowering this lower cylindrical section
in combination with water flow and water depth. The wave suppressor
assembly was mounted on a four legged structure that provided a small deck
space for personnel access around the unit. Figure 4 shows a sketch of the
vortex generator and wave suppressor. The supporting structure for
the wave suppressor and the vortex generator was located in the water
basin at a distance of sixty feet from a wave generating machine with two
foot high wave capabilities.
-------
SECTION VI
CHARACTERIZATION OF VORTEX
The vortex assisted nozzle system is shown in Figure 5. In
this figure, the main components are identified so that the reader may
have a clear picture in mind as these terms appear frequently throughout
this report. As an additional aid, operational capabilities such as air
flow, water flow, etc. are also given in Figure 5. Flow control was
accomplished by adjusting the gate valve on the discharge side of the pump.
The water flow rate was determined by use of a calibrated pitot tube.
The water level was controlled by pumping water into or out of the basin
and noting the level by observing a marked gage bolted to the side of the
vortex framework. The lower skirt was manually moved up or down to change
the quantity of water entering the vortex from the side thus controlling
the vortex size. Characterization of the vortex was by visual and photo-
graphic means. A grid work (2" x 2" square) was placed along a diameter
of the vortex for the purpose of reconstructing the vortex profile from
the photographs as shown on Figure 6. Such profiles are shown in Figure 7
along with the position of the nozzle with respect to the vortex. Addi-
tional tests were performed by moving the lower skirt down and allowing
more water to be drawn from under the wave suppressor. The vortex size
used for the oil recovery tests was smaller than the maximum capable of the
system and was selected for its compatibility with the nozzle system based
on nozzle diameter and air flow rate.
To this point, all of the vortex measurements were taken in calm
water with little or no wind. The vortex was extremely stable and well
suited as a possible assist to enhance the airlift technique for recovering
oil slicks. To investigate the behavior of the vortex when subjected to
wave motion, the preceeding experiments were repeated while two foot waves
(average height) were present. The upper cylindrical slotted skirt
suppressed the wave motion by about 35 percent, as indicated in Figure 8.
The vortex responded to the dynamic state by meandering about its centered
equilibrium (still water) position. Figure 9 is a sequence from a movie
taken of this motion. In several instances, the vortex collapsed completely
and then reformed at a later time (approx. 1 sec. after collapse). On
the average, the vortex remained within a circle diameter of approximately
3 times the vortex diameter. This meandering motion was prevalent in all
of the tests conducted in the wave motion experiments. It is believed,
at least for this system, that it would be extremely difficult if not
impossible to effectively damp out the motion of the vortex due to waves.
Thus, to keep the nozzle, above the vortex at all times one must employ a
nozzle of sufficient size to cover the expected motion or to develop a
feedback controlled nozzle to "follow" the vortex. The latter approach
is very unrealistic from a practical point of view.
-------
The wave generating machine used for these experiments con-
sisted of a 338 hp hydraulically driven bottom hinged flapboard with two
foot high wave capabilities. Control features are provided for both
period and wave height.
10
-------
SECTION VII
OIL COLLECTION EXPERIMENTS
The purpose of these experiments was to secure quantitative data on the
area of influence of a vortex assisted airlift oil recovery system and on the
efficiency of nozzles as a method for removing oil collected by vortex assistance.
The method used to evaluate the effect of the vortex makes use of a continuity
equation in which it suffices to measure surface current velocities. For
example, the rate at which oil is recovered from a surface is proportional to
the oil slick radial velocity and the thickness, at some point >Vmultiplied by
the circumference through-^. This statement presupposes that the radial velocity
CA) is independent of 9, and that the oil thickness is sufficiently small to
neglect velocity gradients through the thickness.
\
nozzle-
or
where
h = instantaneous slick thickness
Q = rate of oil recovery
K = -Q
2irh
R = nozzle radius
Thus, if>\,is taken to be some point where the surface velocity is small
compared to the oil spreading velocity then the slick thickness (assuming a
large oil slick area) remains relatively constant at least over some period of
time. This "small" velocity will be termed the characteristic velocity Vc which
for these studies was arbitrarily set equal to 0.01 ft/sec. It is apparent there-
fore that for a given oil slick the efficiency, as defined here, between different
nozzles and vortex assisted nozzles can be estimated by merely noting the distances
at which the characteristic velocity (Vc) is found. As an illustration, assume
that (V ) is found to be at 18 feet from the nozzle center for the vortex assisted
nozzle while the same (Vc) is found to be at the 9 foot location for the unassisted
nozzle. Then the vortex enhances oil pickup by a factor of two over the unassisted
nozzle.
Time-distance data were taken on the movement of small floating particles
for nozzle operation both with and without the vortex. These data taken with
11
-------
a variety of oil types and without oil show the range of influence to be due
entirely to the surface current generated by the vortex. Typical values are
presented in Figure 10. The oils used for these tests were 30 wt. motor oil
36°-38° API Canadian crude, 40°-42° API Canadian crude and diesel oil. To
determine that the suction nozzle was capable of complete recovery of the oil
gathered by the vortex system,samples of the discharge of the 6000 gpm pump
were taken while the system was processing approximately a one-half inch thick
slick of a 36°-38° API Canadian crude. Figure 11 shows the range of influence
for three of the flow conditions investigated.
Comparison of the surface velocity data taken with that predicted by
equation 1(K =-5.0) is shown in Figure 12. From this correlation, calculated
values for oil recovery rates can be predicted and are presented in Figure 13.
From these curves, the rate of oil recovery with the vortex assist.and the lower
cylinder installed is 7.9 times that of the airlift nozzle alone.
For the suction oil recovery experiments, the lower cylinder was removed
to provide a forced vortex strength compatible with the 861 SCFM suction system.
In these experiments, measured amounts of oil were administered over a specific
area and the time for complete recovery recorded. Based on the nozzle recovery
characteristics and these data, the oil-water ratio can be calculated as a function
of slick thickness. Figure 14 presents these results. Although no tests were
conducted in the presence of currents, it is believed that the presence of the
vortex will still augment the operation of the suction nozzle.
The nozzle used for these tests was basically a truncated cone with a
horizontal skirt added to the lower periphery (Figure 15). The addition of
the horizontal skirt provided a longer residence time for the liquid surface
to be acted upon by the negative pressure gradient in this region, facilitating
a smoother removal of the surface liquid. Vertical positioning of the nozzle
over the vortex was quite critical to achieve the objective - sufficient liquid
recovery to insure 100 percent oil recovery in even relatively thick oil
slicks and yet not over load the air hose with liquid during minor surface
disturbances. This separation was one inch for the 861 SCFM air flow rate used
for these tests and provided a 60 gpm liquid recovery rate. This is equivalent
to approximately one percent by volume liquid. Higher recovery rates were
attempted but resulted in a significant loss of oil passing into the vortex
generating system due to the nozzle periodically filling the suction line with
water. This condition resulted in a pulsating type recovery flow.
Experiments were performed to determine if the range of influence of the
vortex, both with and without oil was affected by the action of waves. Waves
to two foot maximum height were generated during operation with a variety of
vortex flow conditions and no change in the total area of influence could be
observed. Although the area of influence during these tests did not appear to
differ from the corresponding calm water test, it was observed that the center
of the area moved toward the wave generating machine by some small amount. Due
to the difficulty in measuring the velocity of a surface particle in the action
of waves no quantitative data on this effect was gathered.
12
-------
During both the calm sea and wave tests even relatively light winds
(2-4 knots), had a strong influence on the migration of the oils being
tested. Because of the low surface current velocities outside of the wave
suppressor the pollutants would be blown down wind out of the range of
influence of the vortex. The surface oil collected by the vortex was then
generally limited to the areas immediately adjacent to the wave suppressor
and up wind of it. During oil recovery operations these local breezes
generated wind waves that were measured at up to two inches. No detrimental
effect was experienced on the smooth flow of liquid from the surface from these
small, short period waves.
13
-------
SECTION VIII
COANDA NOZZLE EXPERIMENTS
The purpose of these experiments was to evaluate a Coanda
eductor for use as an airlift oil recovery device.
One of the major faults of a conventional vacuum airlift system
for recovery of oil from a water surface is its sensitivity to even
small wave motion. The proximity to the liquid surface is very critical.
If the airlift nozzle is submerged the system will fill with liquid to a
height equal to its negative pressure. When the nozzle is again above
the surface the distribution hoses, now filled with fluid, can be cleared.
This natural process is relatively slow. A Coanda eductor, when subjected
to the same wave motion will not fill its connecting hoses or piping with
fluid, but will merely stop operating until the unit is above the surface
again at which time it will instantly return to service. It was for this
feature that it was selected for evaluation as an oil recovery device.
A three inch diameter, six inch long Coanda eductor was designed
and fabricated for testing (Figure 16). The first experiments were
performed to determine the optimum proximity of the nozzle to the liquid
surface. For the three inch diameter unit this was 0.25 inches. The
next series of experiments were performed to give the total liquid
recovery as a function of air supply rate. Figure 17 presents these
data. The final experiments established the oil-water recovery ratio as
a function of oil slick thickness. It was found that the maximum oil
recovery rate in an oil slick greater than 0.40 inches thick is 55.5 gal/hr.
Figure 18 shows the maximum oil recovery rate as a function of oil slick
thickness. Two oils were used for these tests, an SAE 30 motor oil and
a Canadian crude (36° - 38° API). No measurable difference in recovery
rates between these two materials was observed.
During the Coanda eductor tests, one feature was revealed that
was considered objectionable from the standpoint of oil recovery. This
feature was the emulsification of the recovered oil and water which
required unusually long periods of time for coalescence to occur. It was
observed during the testing that the performance of the Coanda eductor
was somewhat sensitive to back pressure on the discharge hose. It is
assumed that this feature would limit the length of the discharge hose
transporting the recovered oil to a storage tank.
14
-------
SECTION IX
ACKNOWLEDGMENTS
This report summarizes research conducted by Battelle-Northwest
for the Federal Water Quality Administration, Department of the Interior
under Contract 14-12-513.
The research team comprised:
J. D. Smith Principal Investigator
(Development Engineer - Systems Design
and Development)
E. R. Simonson Analytical Analysis
(Development Engineer - Systems Design
and Development)
P. L. Peterson Scope and Principal Proposal Author
(Development Engineer - Systems Design
and Development)
P. C. Walkup Project Director
(Manager, Systems Design and Development)
The Mobile Oil Co., Ferndale, Washington is sincerely thanked
for providing crude oil for these experiments.
The Texaco Oil Co., Anacortes, Washington is acknowledged with
sincere thanks for providing crude oil for these experiments. The support
of the project by the Federal Water Quality Administration, J. C. Willmann,
project officer is acknowledged with the thanks of the project team.
-------
SECTION X
GLOSSARY
1. Vortex - A fluid having a whirling or circular motion tending to
form a central cavity.
2. Vortex generator - Submerged intake piping connected to the 6000 gpm
pump.
3. Wave suppressor - Circular-slotted baffel fitted with plates to induce
circular motion to incoming water.
4. Suction nozzle - A truncated cone used for vacuum removal of oil.
5. Coanda nozzle - An eductor based on the fluid attachment principal.
6. Range of influence - That distance where surface motion attributable
to nozzle or vortex action is just perceptible.
7. Basin - 209 ft x 432 ft, 16 ft maximum depth rectangular water pond.
8. Vortex assistance - Increasing the efficiency of nozzle by supplying
a vortex directly beneath it.
9. Pollutant - Diesel oil, 36° - 38° API Canadian crude, 40° - 42° API
Canadian crude, 30 wt. motor oil.
16-
-------
SECTION XI
BIBLIOGRAPHY
1. Oil Spillage Study Literature Search and Critical Evaluation for
Selection of Promising Techniques to Control and Prevent Damage —
Battelle-Northwest, Richland, Washington, November 20, 1967 for the
U.S. Coast Guard AD6662B9.
2. P. C. Walkup, et al, Study of Equipment and Methods for Removing Oil
from Harbor Waters, Report No. CR. 70.001 by Battelle-Northwest for
U.S. Navy, August, 1969.
3. The Dillingham Corporation for the American Petroleum Institute,
"Systems Study of Oil Spill Cleanup Procedures, Volume 1, Analysis of
Spills and Control Materials", API Publication 4024.
4. Principle of Naval Architecture, editor John P. Comstock, Published
by Society of Naval Architecture and Marine Engineers, 74 Trinity Place,
New York, New York, 1967 Revised Edition.
5. E. Kent Springer, F. M. Patterson, "Experimental Investigation of
Critical Submergence for Vortexing in a Vertical Cylindrical Tank.
6. J. W. Smith, "Problems in Dealing with Oil Pollution on Sea and Land,"
Scientific Aspects of Pollution of the Sea by Oil, Proceedings of
a symposium held on 2 October, 1968, Reprinted from the Journal of the
Institute of Petroleum, November 1968, p. 60.
7. D. P. Hoult, ed., "Containment and Collection Devices for Oil Slicks,"
Oil on the Sea, proceedings of a symposium on the scientific and
engineering aspects of oil pollution of the sea, sponsored by
Massachusetts Institute of technology and Woods Hole Oceanographic
Institute and held at Cambridge, Massachusetts, May 16, 1969.
8. C. L. Breitschneider, "Wave Forecasting," Handbook of Ocean and
Underwater Engineering, J. J. Myers (Editor-in-Chief) prepared under
the auspices of North American Rockwell Corporation.
9. T. H. Gaines, "Notes on Pollution Control, Santa Barbara," Appendix
III, undated.
10. M. P. Holdsworth, "Control of Accidental Oil Spillage at Sea,"
Proceedings of the Institute of Petroleum Summer Meeting,
Brighton,1968.
11. Kundo, G., Hayaski, M. and Murakami, Y., "Studies on the Method
of Collecting Oil from the Surface of the Water", Osaha Industrial
and Technological Laboratory Quarterly, Vol. 18, No. 1, May 1967.
1Z
-------
12. P. C. Walkup, et al, Study of Equipment and Methods for Removing
Oil from Harbor Waters, Report No. CR. 70.001 by Battelle-Northwest
for the U.S. Navy, August, 1969.
13. Sir Geoffrey Taylor, "The Dispersion of Matter in Turbulent Flow
Through a Pipe," Proc. Roy. Soc._(London) Ser A, Vol. 223, no. 1155,
pp. 446-448, May 20, 1954.
14. C. E. Zo Bell, "The Occurrence, Effects, and Fate of Oil Polluting
the Sea," Intern. J. Air Water Pollution, vol. 7, pp. 173-198, 1963.
15. P. Hughes, "A Determination of the Relation Between Wind and Sea-
Surface Drift," Quart. J. Royal Meteorol. Soc, vol. 82, pp. 494-502,
1956.
16. S. A. Berridge, R. A. Dean, R. G. Fallows, and A. Fish, "The
Properties of Persistent Oils at Sea," J. Inst. Petrol, vol. 54,
no. 539, November, 1968.
17. P. C. Blokker, "Spreading and Evaporation of Petroleum Products on
Water," Paper presented to the Fourth International Harbor Conference,
Antwerp, June 22-27, 1964.
18
-------
Cyclone
Separator
Vortex Pump
Floating Wave
Suppressor
Vortex Hose
Vortex Pipe
DEPLOYED REFERENCE SYSTEM
FIGURE 1
-------
AN EMPIRICAL RELATIONSHIP FOR VORTEX FORMATION SHOWING
THE CRITICAL SUBMERGENCE DEPTH VS FLOWRATE FOR AN 18
INCH DIAMETER DRAIN
100
90
80
70
60
50
40
§
w
o
o
Pi
U
30
20
10
FLOW - GPM X 10
-3
8 9 10
FIGURE NO. 2
-------
VORTEX STRUCTURE ASSEMBLY
-• '•:<• M >-** ,^,
FIGURE 3
-------
18 Inch
Diameter
Intake
Pipe
Wave Suppressor
Basin Surface
I
Lower Cylindrical
Section
VORTEX GENERATOR
FIGURE NO. 4
-------
861 SCFM
Blower
1000 Gal. Decanting Tank
SCHEMATIC
VORTEX AND AIRLIFT
SYSTEMS
6000 GPM Pump
FIGURE NO.5
-------
6000 GPM VORTEX
SUBMERGENCE
44 IN.
SUBMERGENCE
38 IN.
LOWER CYLINDER
REMOVED
SUBMERGENCE
38 IN.
FIGURE 6
-------
VORTEX PROFILES
(one inch minor increments)
6000 GPM
4000 GPM
200 GPM
Submergence - 32 In.
Submergence - 38 In.
Submergence - 44 In.
FIGURE NO. 7
-------
I
H
Q
W
CO
CO
WAVE SUPPRESSION
(SIX FOOT DIAMETER SLOTTED BAFFLE)
12
10
10
GENERATED WAVE HEIGHT - IN.
20
FIGURE NO.8
-------
VORTEX MOVEMENT
ONE SECOND INTERVALS
j
FIGURE 9
-------
RADIAL SURFACE VELOCITY
36 - 38° API CANADIAN CRUDE
With 6000 GPM Vortex Assist
o
n
M
H
I
H
S!
M
n
Suction Nozzle Only
(861 SCFM)
I
J_
12 24 36 48 60
DISTANCE FROM NOZZLE EDGE - IN.
72
84
96
FIGURE NO. 10
-------
EFFECTIVE RADIUS OF INFLUENCE
30 I-
20 -
w
u
§
&
J
fe
M 10 .»
o
CO
Lower Cylinder Installed
Lower Cylinder Removed
2000 4000
VORTEX FLOW - GPM
6000
FIGURE NO. 11
-------
RADIAL VELOCITY
£
M
o
w
0.6
0.5
0.4
0.3
0.2
0.1
I I
Equilibrium Thickness
36 - 38° Canadian Crude
Experimental Data 0
I I I
6 12 18 24
RADIAL DISTANCE IN.
30 36 42 48
FIGURE NO. 12
-------
CALCULATED OIL RECOVERY RATES FOR THE SUCTION
NOZZLE ALONE AND WITH THE VORTEX ASSISTANCE
FOR 36 - 38° API CANADIAN CRUDE OIL
40
£
o
o
w
CO
CO
W
u
§
p-l
O
Pn
O
30
With Vortex Assist
Lower Cylinder Installed
20
O1
10 —
With Vortex Assist
Lower Cylinder Removed
Suction Nozzle Onl
0.1 0.2 0.3
SLICK THICKNESS - IN.
0.4
0.5
FIGURE NO. 13
-------
CALCULATED OIL WATER RATIO VS SLICK THICKNESS
6000 GPM VORTEX - LOWER CYLINDER REMOVED
30 |_
20
H
W
PM
I
O
W
I
10
I
.25 .5
SLICK THICKNESS - IN.
.75
1.0
FIGURE JNO. 14
-------
4"
V
\
I
2"
8"
.2".
AIRLIFT NOZZLE
FIGURE NO. 15
-------
COANDA NOZZLE
1
Deflection
Cone
FIGURE NO.16
-------
LIQUID RECOVERY VS AIR SUPPLY
THREE INCH COANDA NOZZLE
60
54.0
' 51.5
50
47.5
5C
PM
42.8
a
o
40
CTi
oo
n
30
30
40
AIR SUPPLY - SCFM
50
FIGURE NO. 17
-------
OIL RECOVERY RATE VS SLICK THICKNESS
(THREE INCH COANDA NOZZLE)
60
50
40
0s
a,
o
w
I
w
o
hJ
M
o
30
20
10
55.1
55.5
43.6
I
0 0.062 0.125 0.187
SLICK THICKNESS - IN.
0.250
0.312
0.375
FIGURE NO.18
-------
Bl BLIOGRAPHIC:
Battelle Northwest, Recovery of Large Marine Oil Spills
by Use of a Vortex Assisted Airlift System, Final Report
FWPCA Contract No. 14-12-513, July, 1970
ABSTRACT
Studies were conducted to determine the feasibility of a
concept for recovery of floating oil slicks which utilizes a
pump induced vortex and a vacuum suction or Coanda nozzle.
The apparatus used for developmental experimentation com-
prised a pumping system for vortex production, a large water
basin, a flapper type wave generator, and several configura-
tions of the experimental assembly. The range of influence
was smaller than was anticipated. Approximately a 25 foot
influence diameter was achieved for the maximum strength
vortex generated 1n this apparatus. Extrapolation'of measured
performance data showed that a 1/8 Inch thick slick could be
recovered at the rate of 960 gallons per hour. Experiments
with and without a variety of oils showed that enhanced oil
recovery rates with the vortex was due entirely to the surface
current generated by the vortex. This effect was found to
improve oil recovery by a factor of 7.9 above the rates
achieved with a suction nozzle alone. The surface position
of the vortex cavity was found to be sensitive to surface
waves. The cavity moved in a circular path within three
vortex cavity radii of the still water cavity location as a
wave passed through the assembly.
Tests with a Coanda nozzle (a fluid attachment eductor)
showed improved performance in surface waves. However, the
recovered oil-water mixture was highly emulsified.
ACCESSION NO.
KEY WORDS:
Vortex
Coanda Nozzle
Wave Suppressor
BIBLIOGRAPHIC:
Battelle Northwest, Recovery of Large Marine Oil Spills
by Use of a Vortex Assisted Airlift System, Final Report
FWPCA Contract No. 14-12-513, July, 1970
ABSTRACT
Studies were conduc
concept for recovery of
pump induced vortex and
The apparatus used for
prised a pumping system
basin, a flapper type w
tions of the experiment
was smaller than was an
influence diameter was
vortex generated in thi
performance data showed
recovered at the rate o
with and without a vari
recovery rates with the
current generated by th
improve oil recovery by
achieved with a suction
of the vortex cavity wa
waves. The cavity move
vortex cavity radii of
wave passed through the
Tests wi th a Coanda
showed improved perform
recovered oil-water mix
ted to determine the feasibility of a
floating oil slicks which utilizes a
a vacuum suction or Coanda nozzle.
developmental experimentation com-
for vortex production, a large water
ave generator, and several configura-
al assembly. The range of influence
ticipated. Approximately a 25 foot
achieved for the maximum strength
s apparatus. Extrapolation of measured
that a 1/8 inch thick slick could be
f 960 gallons per hour. Experiments
ety of oils showed that enhanced oil
vortex was due entirely to the surface
e vortex. This effect was found to
a factor of 7.9 above the rates
nozzle alone. The surface position
s found to be sensitive to surface
d in a circular path within three
the still water cavity location as a
assembly.
nozzle (a fluid attachment eductor)
ance in surface waves. However, the
ture was highly emulsified.
ACCESSION NO.
KEY WORDS:
Vortex
Coanda Nozzle
Wave Suppressor
BIBLIOGRAPHIC:
Battelle Northwest, Recovery of Large Marine Oil Spills
by Use of a Vortex Assisted Airlift System, Final Report
FWPCA Contract No. 14-12-513, July, 1970
ABSTRACT
Studies were conducted to determine the feasibility of a
concept for recovery of floating oil slicks which utilizes a
pump Induced vortex and a vacuum suction or Coanda nozzle.
The apparatus used for developmental experimentation com-
prised a pumping system for vortex production, a large water
basin, a flapper type wave generator, and several configura-
tions of the experimental assembly. The range of influence
was smaller than was anticipated. Approximately a 25 foot
Influence diameter was achieved for the maximum strength
vortex generated 1n this apparatus. Extrapolation of measured
performance data showed that a 1/8 Inch thick slick could be
recovered at the rate of 960 gallons per hour. Experiments
with and without a variety of oils showed that enhanced oil
recovery rates with the vortex was due entirely to the surface
current generated by the vortex. This effect was found to
improve oil recovery by a factor of 7.9 above the rates
achieved with a suction nozzle alone. The surface position
of the vortex cavity was found to be sensitive to surface
waves. The cavity moved in a circular path within three
vortex cavity radii of the still water cavity location as a
wave passed through the assembly.
Tests with i Coanda nozzle (a fluid attachment eductor)
showed improved performance in surface waves. However, the
recovered oil-water mixture was highly emulsified.
ACCESSION NO.
KEY WORDS:
Vortex
Coanda Nozzle
Wave Suppressor
-------
1
5
Access/on Number
2
Subject Field & Group
Q5G,08C
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
Battelle Northwest Laboratories, Richland, Washington
Systems Design-Development Section
Title
Recovery of Oil Spills Using Vortex Assisted Airlift System
1 Q Authors)
P. C.
J. D.
E. R.
Walkup
Smith
Simon son
16
21
Project Designation
15080DJM07/70
Note
22
Citation
Battelle Northwest Research Report, July 1970, 40 p, 18 Fig., 17 Ref.
23
Descriptors (Starred First)
* Research and Development, *Water Pollution, *Water Pollution Control
*Vortices, *0ily Water
25
Identifiers (Starred First)
*Vortex, *Coanda
27
Abstract
Studies were conducted to determine the feasibility of a concept for recovery of
floating oil slicks which utilizes a pump induced vortex and a vacuum suction or
Coanda nozzle. The apparatus used for developmental experimentation comprised
a pumping system for vortex production, a large water basin, a flapper type wave
generator, and several configurations of the experimental assembly. The range of
influence was smaller than was anticipated. Approximately a 25 foot influence
diameter was achieved for the maximum strength vortex generated in this apparatus.
Extrapolation of measured performance data showed that a 1/8 inch thick slick
could be recovered at the rate of 960 gallons per hour. Experiments with and
without a variety of oils showed that enhanced oil recovery rates with the vortex
was due entirely to the surface current generated by the vortex. This effect was
found to improve oil recovery by a factor of 7.9 above the rates achieved with
a suction nozzle alone. The surface position of the vortex cavity was found to
be sensitive to surface waves. The cavity moved in a circular path within three
vortex cavity radii of the still water cavity location as a wave passed through the
assembly.
Tests with a Coanda nozzle (a fluid attachment eductor) showed improved performance
-in giirfar.p- wavpg. Hnwp.yp.r, fhp rp.r.owrpri nil-wafpr irHvl-urp wag highly pmiil.eif led.
Abstractor , _ . .
John D. Smith
Institution
Battelle Northwest
(Smith-Battelle)
WR:102 (REV. JULY 1969)
WRSI C
SEND TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 20240
ft GPO: 1969—359-339
------- |