EPA-R2-72-006
. Environmental Protection Technology Series
A Free Floating Endless
Belt Oil Skimmer
Office of Research and Monitoring
U.S. Environmental Protection Agency
Washington, D.C. 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
U. 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.
-------
EPA-R2-72-006
August 1972
A FREE FLOATING ENDLESS BELT OIL SKIMMER
By
Robert W. Agnew
Contract No. 14-12-908
Project 15080 GBJ
Project Officer
Mr. Kurt Jakobson
Office of Research and Monitoring
Environmental Protection Agency
Washington, B.C. 20460
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 Environmental
Protection Agency and approved for publication.
Approval does not signify that the contents necessarily
reflect the views and policies of the Environmental
Protection Agency, not does mention of trade names
or commercial products constitute endorsement or
recommendation for use.
ii
-------
ABSTRACT
A free floating endless belt oil skimmer was developed as a means of
recovering spilled oil from surface waters. The skimmer utilizes a
unique high efficiency conveyor wringer to power and wring the belt.
The belt is designed to float on the water surface and responds rapidly
to the shape of the waves, thereby maximizing oil-sorbent contact time.
Evaluation of the skimmer was conducted in a 60 foot diameter annular
test tank under the conditions of slightly progressive waves having an
amplitude of two feet. One foot wide neoprene backed polyurethane foams
were utilized as the sorbent material.
The experimental results indicate that the oil pickup rates will vary
with the belt speed, oil slick thickness and belt porosities. Oil pick-
up rates of 8.3 and 3.7 gpm per foot of belt width were attained for
#2 Fuel Oil and Bunker C Oil respectively at a slick thickness of 0.10
inches. The recovered liquid contained approximately 50-70% oil at 0.10
inch slick thickness.
A conceptual design of a five foot wide boat mounted skimmer capable of
harvesting approximately 5 acres per hour of spilled oil is presented.
This report was submitted in fulfillment of Project Number 15080 GBJ,
Contract 14-12-908, under the sponsorship of the Office of Research and
Monitoring, Environmental Protection Agency.
iii
-------
CONTENTS
Section Page No.
I CONCLUSIONS 1
II RECOMMENDATIONS 3
III INTRODUCTION 5
IV LITERATURE SEARCH 7
Characteristics of Oil Spills 7
State-of-the-Art on Oil Removal Devices 9
V APPROACH TO THE PROBLEM 13
VI DESIGN OF EXPERIMENTAL PROTOTYPE SKIMMER 15
VII EXPERIMENTAL PROCEDURES 29
Collection and Analysis of Standard 30
Procedure Experiment Data
Maximum Oil Pickup Rate Tests 33
Wringer Efficiency 36
Results of Standard Procedure Tests 39
Tests with #2 Fuel Oil 41
Tests with #6 Oil (Bunker C) 53
Oklahoma Crude Oil Tests 58
VIII DISCUSSION OF RESULTS 65
Mechanical Wringer and Drive System 65
Belt Integrity 65
Use of Polyurethane Sorbents 66
Studies with Other Sorbent Materials 67
Summary 68
-------
CONTENTS CONT.
Section Page
IX DESIGN CONCEPT FOR BOAT MOUNTED SKIMMER 69
Introduction 69
Hull Requirements 72
Propulsion and Speed 72
Power Unit 73
Outfit Equipment 73
Bow Booms 74
Options 74
X ACKNOWLEDGMENTS 77
XI REFERENCES 79
XII APPENDICES 81
Appendix A - Results of the Bench Scale
Oil Skimmer Investigation 82
Appendix B - Computer Program 91
vi
-------
FIGURES
Page
1 ORIENTATION OF FREE FLOATING BELT SKIMMER
TO WATER SURFACE 16
2 POLYURETHANE BELT DETAILS 17
3 CONVEYOR WRINGER ASSEMBLY 18
4 OIL SKIMMER MECHANISM SCHEMATIC 19
5 REMOVAL OF THE PIN CONNECTOR TO CHANGE A BELT 21
6 EXPERIMENTAL PROTOTYPE TEST SITE 22
7 ANNULAR COLLECTION AND SUPPLY TANKS 23
8 ROTATING PLATFORM ASSEMBLY 24
9 OIL DISTRIBUTION HEADER AND OIL SATURATED
POLYURETHANE BELT 25
10 OIL SEPARATION TANK 26
11 WAVE GENERATOR DRIVE 28
12 OIL PICKUP RATE AND EQUIVALENT SLICK THICKNESS
VS. TIME 34
13 OIL PICKUP RATE VS. BELT SPEED 35
14 LBS. OIL RECOVERED/LB. POLYURETHANE APPLIED
VS. BELT SPEED 37
15 LBS. OIL RECOVERED/LB. POLYURETHANE APPLIED
VS. BELT SPEED 38
16 OIL AVAILABLE FOR SKIMMING VS. FORWARD VELOCITY-
EXPERIMENTAL PROTOTYPE SKIMMER 40
17 OIL PICKUP RATE VS. BELT SPEED 42
18 OIL PICKUP RATE VS. BELT SPEED 43
19 OIL PICKUP RATE VS. SLICK THICKNESS 44
20 OIL PICKUP RATE VS. SLICK THICKNESS 45
21 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK
THICKNESS 47
vii
-------
FIGURES
Page
22 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK 48
THICKNESS
23 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK
THICKNESS 49
24 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK
THICKNESS 50
25 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK
THICKNESS 51
26 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK
THICKNESS 52
27 OIL PICKUP RATE VS. BELT SPEED 54
28 OIL PICKUP RATE VS. BELT SPEED 55
29 OIL PICKUP RATE VS. SLICK THICKNESS 56
30 OIL PICKUP RATE VS. SLICK THICKNESS 57
31 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK
THICKNESS 59
32 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK
THICKNESS 60
33 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK
THICKNESS 61
34 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK
THICKNESS 62
35 PERCENT OIL IN RECOVERED MIXTURE VS. SLICK
THICKNESS 63
36 CONCEPTUAL DESIGN BOAT MOUNTED OIL SKIMMER,
ELEVATION VIEW 70
37 CONCEPTUAL DESIGN BOAT MOUNTED OIL SKIMMER,
PLAN VIEW 71
1-1 OVERALL VIEW OF BENCH SCALE OIL SKIMMER 86
1-2 OIL RECOVERY RATE AS FUNCTION OF OIL LAYER
THICKNESS 89
viii
-------
TABLES
1 MAJOR WORK TASKS ACCOMPLISHED DURING THIS
PROJECT 14
2 OIL BELT SKIMMER DATA 31
3 OUTPUT FROM COMPUTER ANALYSIS OF
EXPERIMENT NO. 35 32
4 MAXIMUM OIL PICKUP RATE TEST CONDITIONS 33
5 OPTIMUM OPERATING CONDITIONS AND RESULTS
FOR #2 FUEL OIL 46
6 RECOMMENDED OPERATING CONDITIONS FOR
RECOVERY OF BUNKER C 58
7 CHARACTERISTICS OF OKLAHOMA CRUDE OIL 58
8 CONDITIONS OF ENDURANCE TEST 45 PPI
POLYURETHANE BELT 66
1-1 COMPARISON OF BELT PERFORMANCE - SUMMARY 88
II-l PROGRAM LISTING PETSKI 91
II-2 DEFINITION OF VARIABLES 93
ix
-------
SECTION I
CONCLUSIONS
The following conclusions can be drawn from information collected during
this study.
1. The free floating endless belt concept developed during this
study is a feasible concept for oil recovery.
2. The conveyor wringer used in this skimmer is a highly efficient
method for wringing and also provides a very effective method
for driving the belt.
3. Utilizing neoprene backed polyurethane foams as a sorbent mater-
ial, the following oil recovery rates can be expected in two
foot waves for the one foot wide belt tested in this study.
a. Light oils - #2 Fuel Oil -8.3 gpm of oil at 0.10 inch
slick thickness. Recovered oil-water mixture contained
approximately 50% water.
b. Heavy oils - #6 Fuel Oil (Bunker C) - 3.7 gpm of oil at 0.10
inch slick thickness. Recovered oil-water mixture contained
approximately 28% water.
4. Oil Recovery rate can be expected to vary almost directly with
slick thickness and belt speed.
5. Optimum performance was attained with a 1 inch thick 80 ppi belt
for light oil (//2 Fuel) and a 3/8 inch thick 45 ppi belt for
heavy oils (Bunker C).
6. Endurance testing of the neoprene backed polyurethane belts indi-
cates that a minimum belt life of 1-2 days can be expected even
in high debris areas. The belt has a minimum susceptibility to
damage from debris because of the free floating configuration.
7. The skimmer is capable of handling almost any sorbent material
which can be fabricated into an endless belt configuration.
8. Polyurethane foams have a tendency for high water pickup,
particularly in thin slicks.
9. It is technically and economically feasible to construct a five
foot wide skimmer for shipboard installation.
-------
SECTION II
RECOMMENDATIONS
Based on observations made during this project, the following recommen-
dations are made.
1. A full scale prototype skimmer should be constructed and mounted
on a suitable vessel for use in oil recovery operations. The
design of the skimmer and vessel would be according to the
conceptual design presented in this report. This skimmer could
be evaluated under actual spill conditions and could serve as
a working laboratory for the evaluation of new sorbent materials.
2. A relatively significant effort needs to be devoted to the
development and evaluation of sorbent materials. It would appear
that some very promising sorbents are being shunted aside in
laboratories because there is not sufficient market potential
to support further work. The EPA and their contractors should
be aware of this situation. Perhaps EPA could act as a sorbent
materials clearinghouse.
-------
SECTION III
INTRODUCTION
Significant spills of oil and petroleum products in protected and
unprotected waters constitute a potential fire hazard and may cause
extensive damage to both marine and aquatic flora and faunae. More-
over, such spills seriously degrade water quality with respect to usage9
particularly for recreational and consumptive purposes. Property
damage resulting from spills near inhabited coastal areas is often
substantial.
The increasing number and size of oil tank ships, coupled with
accelerated off-shore drilling activities, indicate that the potential
for aggravating an already unsatisfactory situation is great. The ideal
solution, of course, would be to prevent any oil spills from occurring,
but experience shows it is not reasonable to expect that preventive
measures will be 100% effective. Defensive measures must be developed
for containing and cleaning up floating oil when accidental and uncon-
trollable spillages occur.
Various measures developed or under development for the control and
prevention of major oil spillage have been reviewed in recent literature
(1, 2). The technical approaches in these measures can be categorized
as; mechanical containment, mechanical removal, physical adsorption,
combustion and biological degradation. These methods vary in effective-
ness depending upon the type of waters, size of spill and the type of
oil. No single combatant method to date is considered sufficiently
developed to pursue at the exclusion of other methods.
The proposed approach during this project was to develop a concept for
a belt type oil harvesting and recovery skimmer, and to design, build,
and test a device large enough to enable the determination of its
feasibility for meeting the stated EPA design goals. The process
objectives for the device were to recover rapidly, efficiently and
without the aid of additives, floating oil from the surface of both
protected and unprotected waters. The proposed device should also be
capable of being rapidly transported to the oil spill site.
A bench scale model of the proposed prototype was constructed and
evaluated in the laboratory. Based on the results of the bench scale
model, an experimental prototype oil skimmer system was designed. This
report documents the design, construction and evaluation of the experi=
mental prototype performance. In addition, a conceptual design for a
full scale, boat-mounted oil belt skimmer is presented.
Initially, a concept was proposed for a belt type skimmer in which a
porous belt was mounted either vertically or at a slight angle with
respect to the water. The belt would be propelled through the oil-water
layer. This concept was later modified to permit the porous belt to
-------
float on the oil surface, thus permitting a longer contact period with
the oil and a minimum of contact with the water. This procedure signif-
icantly improved the skimmer performance and was incorporated into
the design of the experimental prototype system.
The prototype skimmer system was tested under the adverse conditions
of two foot waves and various types of oil including Bunker C,
Oklahoma crude oil and No. 2 fuel oil. Based on the successful
evaluation of the experimental prototype, it is proposed that a full
scale, boat mounted, oil belt skimmer be constructed and evaluated on
an actual oil spill.
-------
SECTION IV
LITERATURE SEARCH
A search of the existing literature was made to determine the charac-
teristics of oil spills, and to obtain pertinent information on oil
removal technology.
Characteristics of Oil Spills
Oil spillages, generally speaking, may be classified as "large" or
"small" depending upon the amount of oil released in the water. Small
spills may vary in volume from less than one hundred barrels to as much
as several thousand barrels. Large spills, by contrast, are usually
measured in tons of oil released. The highly publicized Torrey Canyon
mishap, for example, resulted in the release of about 60,000 tons of
crude oil to the sea (3).
Small oil spills usually occur in protected waters such as harbors and
estuaries, occur more frequently than large spills, and may or may not
be accidental. Because of their proximity to inhabited areas and
shallow-water marine life, they often result in as much damage as larger
but more remote spillages. Common sources of small spills include those
from small vessels, industrial waste discharges, oil transfer operations,
pipeline breaks, and vessel cleaning operations. Large spills are
nearly always accidental and often uncontrollable. Floundering trans-
port tankers (Torrey Canyon) and off shore well "blowouts" (Santa Barbara
incident) are some of the examples of the major oil spills. It has been
reported (4) that an estimated 2 million tons of oil enter the ocean
from tanker cargoes each year. In 1970 alone there were reported to be
10,000 such spills (4).
The floatable oil from spillages may consist of crude oil, refined
products such as gasoline or kerosene, and fish or vegetable oils.
The density and viscosity of these oils vary widely, and may signif-
icantly affect development of countermeasures employed to contain and
clean up spills when they occur. The characteristics of four such
oils used by the U.S. Navy, namely, JP-5 Turbine Fuel, distillate
fuel, Navy special fuel oil and Bunker C fuel oil (also known as #6
fuel oil) have been discussed in detail in a recent study (5).
The edge of an oil slick can move in two ways - the slick can spread
out and cover large areas, and it can move as a unit under the influence
of current or wind. The movement of the edge of the slick would equal
the algebraic sum of the two components. Actually, very little infor-
mation is available in the literature on the spreading of large quanti-
ties of oil because of strong public objections toward such experimen-
tation. Small scale experiments on the rate of spread of crude oils
have been conducted by Berridge et al (6) and Blokker (7). They showed
that the tendency for the oil slick to expand is, in part, a function
-------
of the difference in the densities of the oil and the water. The
spreading rate of a homogeneous oil slick was found to be approx-
imately proportional to the instantaneous mean layer thickness. Other
factors influencing the spreading rate were: viscosity, surface ten-
tion, interfacial tension between water and oil, chemical composition,
pour point of the oil, current, and wind speed. It was concluded that
the value of the pour point of any oil may have a profound influence
on its spreading characteristics. An oil with a. pour point higher than
the temperature of the water, as could be the case with Bunker C, can
form a semisolid mass that would have very little tendency to spread,
particularly if its specific gravity approached that of sea water.
Moreover, the influence of viscosity over spreading velocity was
considered to be relatively small, especially.during the initial stages
of the spill. Blokker (7) noted that the time required for spilled oil
to spread out to a slick of 2 cm thickness was very short, on the order
of one minute for 100 cubic meter of oil spill, with viscosities ranging
from 0.8 to 490 CP at 20°C, while Berridge et al (6) found that the
thickness of the slick tended to keep reducing, and the area increasing,
until the thickness of the slick, for the crude oils with specific
gravities ranging from 0.829 to 0.896, reduced to 0.0008 to 0.0012 inch.
The time required for a spill of 100 cubic meters of oil to reduce to
a slick of that range of thickness was 27.7 hours.
Brockis (9), Smith (9) and Hughes (10) have investigated the effects of
winds and currents on the movement of an oil slick. Although there is
variation in the results of the different investigators, experimental
data coupled with theoretical analysis shows that the speed of movement
of an oil slick as a unit, due to the drag force exerted by a wind
blowing across its surface, would be in the range of 3 and 4% of the
wind speed.
The rate of spreading of an oil slick in protected waters, and its
resulting thickness, can be quite different from those in an open sea.
In a. harbor the water is often contaminated, or becomes contaminated by
surface-active substances in the spreading oil. In these cases the
thickness of the oil slick will tend to be greater than would be the
case on a clean-water surface. In such a case the oil slick may reduce
to 0.04 to 0.08 inch in thickness, and the reduction in thickness may
stop or continue at a slower rate. At the closed end of a harbor,
the wind may cause a considerable increase in the thickness of an oil
slick. An 8-knot wind, for example, may keep a layer of oil that is
trapped at the end of a harbor at a thickness of one inch (7).
In summary, there are two phases of the spread of an oil slick in a
still harbor. The first is the initial rate of spreading, immediately
following the failure of an oil container, and can be assumed analogous
to the sudden failure of a dam. The potential energy of a thick oil
layer is essentially converted into kinetic energy and the effects of
viscosity, evaporation, surface tension, and interfacial tension are
negligible during this first phase. The spread rate in this phase will
-------
take place in accordance with Blokker's findings (7). The slick layer
will then continue to decrease according to the Blokker relationships
and at the end of a day will tend to approach the values reported by
Berridge (6). However8 it should be noted here that the effects of
harbor currents and winds, the amounts of contaminants in water and
the space available for spreading can influence the effects of spread-
ing in the later phase significantly, but none of these can be predicted
accurately in advance,
State-Of-The-Art on Oil Removal Devices
Based upon a literature search (1 to 5, 11 to 14) the following tech-
nological approaches can be defined for the disposal of an oil spillages
containment, chemical treatment of the oil slick, surface treatment
followed by mechanical pickup and/or disposal and the direct mechanical
pick-up.
After an oil spill has occurred, the first line of defense is the
containment of the spill. The required equipment must deal with the"
strong forces of nature. Mechanical booms are commercially available
and have been successfully demonstrated in protected waters. Efforts
are ongoing to develop boom systems for deployment in the open ocean.
The boom concept offers potential for all clean up operations and
reduces the cost of cleanup by minimizing the area to be treated.
A variety of materials and compounds are available which may be used to
absorb, sink, or disperse floating oil in a slick. Three types of
absorbents appear to have potential for slick treatment (1, 12):
a. Floating absorbents (straw, sawdust, etc.)
b« Plastics and other polymeric absorbents (polyurethane foam)
c. Gelling agents
Floating absorbents are relatively inexpensive, and the resultant oil/
absorbent combination may be disposed of by burning or burial. However
recovery of oil from the absorbent is a difficult task. Plastics and
polymeric materials are expensive, but allow the recovery of relatively
high percentages of absorbed oil. Gelling agents for solidifying
floating oil show promise, and are currently in the development stage.
Biological degradation by natural processes is the ultimate fate of
spilled oil not otherwise removed, but it is a relatively slow process.
Such natural purification requires the proper species of microbiota
in the presence of such appropriate environmental conditions as suffi-
cient oxygen the proper nutrients, and favorable water temperatures.
Seeding oil slicks with selected microorganisms to accelerate this
process is currently under investigation.
9
-------
It has been indicated in the literature (1, 2, 13, 14) that mechanical
and physical collection of spilled oil should be employed xirtierever
possible. Mechanical devices for collecting spilled oil from the
surface of waters include rotating cylinders, moving belts, suction
pumps, centrifuges and hydrocyclones. Drum and belt skimming systems
generally have relatively lower oil recovery rates but higher oil to
water ratios compared to pumps, centrifuges and hydrocyclones. The
storage requirement of the later group of devices are therefore
generally significantly higher and normally require auxiliary oil/
water separation equipment.
Suction devices, in general, are only effective on relatively thick
slicks (15). Another operational difficulty encountered is the forma-
tion of water-in-oil emulsions because of the passage of the oil and
water through a pump impeller. These emulsions are sometimes very
stable and difficult to break up.
Numerous devices that employ some configuration of the rotating drum
or endless belt are either currently commercially available or are
being developed. One of the important requirements of the material
utilized for oil pickup is that the material chosen should be wetted
easily by the 'oil. The oil that adheres to the moving surfaces is
then either scraped off and/or squeezed out and deposited in a collec-
tion vessel. Another type of unit akin to an endless belt system
employs long rolls or sorbent materials, such as felt, which retain
the oil for subsequent disposal.
An oleophilic belt "oil scrubber" employs a very large loop absorbent
material such as polypropylene wool. This device is operated by moving
this continuous absorbent belt through an oil slick between two pulleys
and squeezing the oil from the belt using wringers mounted on a ship
or at a shore facility (13, 14, 15). The results indicate that the
maixmum recovery rate of an oleophilic belt oil recovery system is
generally limited by the rate at which the oil may be transferred to
the belt surface and interior.
A rotating disk oil removal device developed by Lockheed Corporation
and called "Clean Sweep" utilizes rotating drums to pick up oil on
both sides of a number of closely packed vertical disks half-immersed
in the sea (14, 16). The rotation of the drum is fast enough to hold
the oil on the disks where the oil is removed by acetal resin-edged
aluminum scrapers, directed to a central channel, and pumped to storage
bags. The entire drum is mounted on flotation devices, making the whole
unit resemble a catamaran. Oil pickup is said to be about 70 to 75%
complete (16).
Another rotating disk design for the removal of oil has been made by
Atlantic Research Corporation. The disks resemble the circular blade
of a power saw, without the serrated edge. Recent modifications
include a triangular frame supported at each angle by a globe shaped
buoy. Within the frame are two angled rows of disks. Hoses vacuum
10
-------
up the oil scraped from the disks and carry it to a storage barge.
The system is claimed to have an oil recovery efficiency of about
80 to 85% (14, 16).
A mechanical skimming device under development employs a free vortex
concept (14, 16). A rotating, submerged impeller assembly produces a
vortex flow (circular and down, like water draining out of a bath tub)
in a subsurface column of water. Axial flow through the impeller
produces an inward funneling flow in the overlying water. Under the
action of these flow fields, an oil slick will migrate toward the
vortex axis, submerge and concentrate in a central pocket. An over-
head pump can then remove the oil from the pocket, with relatively
little water. Preliminary data indicates oil removal capacities in
the range of 100 gal/min.
Another patented skimming system from Ocean Pollution Control Inc.
utilizes an adaptation of the shrimp-trawl fishing principle (14). The
unit is a tapered, flattened funnel made of flexible sheet material
reinforced with netting. The leading edge of the flexible material
is held above the water's surface by floats to insure that all oil that
passes beneath is channeled into the collecting funnel. A skimmer picks
up and pumps the oil and water mixture to gravity tanks or other systems
for final separation and storage (16).
A recently developed skimmer is based upon the concept of collecting
the oil under the surface of the water, thus reducing the effect of
waves (17). As the skimmer moves through the water, or the water
moves past the skimmer, the oil is forced to follow the surface of an
inclined plane to a collection well underneath the skimmer. Buoyant
forces cause the oil to surface in the well, forcing water out the
bottom. As the oil collects, it is pumped off to storage tanks. It
is claimed (17) that separation occurs automatically and virtually no
water is collected. Another modification in this system utilizes a
moving inclined plane, rotating into the water. Oil and sorbents are
held against the plane by a combination of hydrodynamic, buoyant and
cohesive forces. The oil and sorbents are collected in the well and
any residual oil is scraped off the belt as it passes through the
well volume (17).
From the information published in the literature (2,5, 15) on belt type
skimmer devices, it is indicated that:
1. Among the major variables expected to determine the capacity
and efficiency of a belt-type skimmer oil recovery device
is the belt material. To meet the recovery rate objectives,
a belt material with high selective capability to absorb and/
or adsorb oil from an oil water mixture is required. Such
a material must have at least the following four qualities:
a. It must be hydrophobia (or conversely oleophilic)
11
-------
bo It must have a high capacity for ad-or absorbing and
retaining oil
c. It must release oil either under pressure or in some
other practical way.
d. It must be resistant to the effects of oils and water
including sea water
Most polymeric (plastic)materials and natural organic materials
meet the first requirement.
2. The capacity of a material to absorb and retain oil is directly
related to its surface area. Porous materials having large
areas available in internal pores open to the surface (retic-
ulated structure) are expected to have relatively the largest
capacity to absorb and/or adsorb and retain oil. The ability
to release oil is expected to vary widely with the porosity
(pore size) of the material and the viscosity of the oil.
3. The resistance of materials to oils and water varies widely
depending on the material and the oil involved.
Based on these requirements it appears that belt material for the pickup
of oil should be selected from the following two categories of materials:
1. Plastic or elastomaric foams
2. Felts of natural or synthetic fibers
12
-------
SECTION V
APPROACH TO THE PROBLEM
The development of a skimmer to recover floating oil from water consis-
tent with the specifications outlined by the Environmental Protection
Agency RFP Number WA 70-23 was approached by a number of simultaneous
and progressive work phases. These phases are listed in Table 1 in
order to provide an understanding of the manner in which the project
was undertaken.
The initial work conducted for this project involved laboratory studies
and bench scale tests to screen and select suitable sorbent materials
and to evaluate the basic skimming concept. The details of these studies
are given in Appendix I.
Based on the results of these initial studies, an experimental prototype
skimmer and the appropriate support test equipment was designed,
fabricated and erected. A detailed description of this equipment will
be presented in a subsequent section of this report.
The skimmer was evaluated to determine oil pickup rates at various oil
thicknesses and with various sorbent materials. Tests were conducted
with two foot high, slightly progressive waves. Approximately 130
tests were conducted under these conditions. Additional tests were
conducted to evaluate maximum pickup capacities and wringer efficiencies.
The final phase of the project was devoted to the development of a
conceptual design for a full scale boat mounted skimmer.
13
-------
Table I
MAJOR WORK TASKS ACCOMPLISHED DURING THIS PROJECT
Phase I - Planning, Design, and Fabrication of Experimental Prototype
Task 1 - Preliminary Bench Scale Investigations
Task 2 - Design of Experimental Prototype
Task 3 - Test Plan Design
Task 4 - Fabrication and Erection of Experimental Prototype Skimmer
Phase II - Operation and Evaluation of Experimental Prototype
Task 1 - System Operation and Testing
Task 2 - Data Accumulation and Evaluation
Task 3 - Final Report, Recommendations and Design Criteria
for Boat Mounted Skimmer
-------
SECTION VI
DESIGN OF EXPERIMENTAL PROTOTYPE SKIMMER
The experimental prototype skimmer was designed to incorporate those
concepts determined essential to effective oil spill pickup by the
literature search and subsequent bench scale tests. The most important
concept to minimize water pickup is the flotation of the absorbent belt
material on the spilled oil. Concomitantly, to prevent oil recycling
and low pickup capacity, the concept of high wringing efficiency neces-
sitated development of a long retention time, uniform pressure mechanism
to allow a sufficient drainage interval for the viscous oil. In order
to test the concepts under anticipated conditions, the design and con-
struction of a unique experimental facility permitted the simulation of
two foot waves and up to five knot pickup craft velocities. Schematic
diagrams and photographs to illustrate the prototype skimmer and exper-
imental facility are shown in Figure 1 to 11.
Figures 1 to 4 illustrate the prototype oil skimmer readied for the
oil pickup operation. The front view in Figure 1 shows the mounting
of the skimmer mechanism above the water surface. The endless oil
pickup belt is fed down onto the oil surface where it floats without
submergence and follows closely the motion of the waves. Sufficient
length of belt is spread on the oil surface so that the belt is always
in contact with the oil, even during the deepest troughs. The oil
belt is composed of a thick section of porous and resilient absorbent
material bonded to a thin, strong backing of neoprene conveyor belting
(Figure 2). The belt lies on the spill surface with the absorbent
material facing downward. The mechanism of absorption occurs because
of the attraction of the oil to the oleophilic belt material as well
as to the upward buoyant pressure which the oil exerts in reaction
to the downward force of the belt weight. The variable speed skimmer
drive feeds the belt at a rate fast enough to prevent drag turbulence,
yet not too fast so that the effective pickup capacity exceeds the
available pickup volume of oil in contact with the belt.
An important and practical advantage of the floating belt is its
resistance to damage from debris. Because the belt has no rigid guide
mechanisms at the water surface, there are no impact points to encourage
debris attachment. In its free floating configuration, the belt will
slide over the top of floating debris. Attachment of debris to the belt
is highly unlikely. Accidental pickup of a piece of debris is not
likely to jam the spring-loaded wringer and stop belt feed because of
the large pressure contact area (432 square inches) which the conveyor
wringer utilizes to pull the belt through.
The path of travel of the belt through the wringer mechanism is shown
in Figure 3 and A. Efficient oil removal is achieved as the belt passes
through the wringer with the absorbent material facing downward and
is squeezed between the upper and lower sections of a perforated con-
veyor wringer. Using this type of wringer, a substantially greater
15
-------
FIGURE 1
ORIENTATION OF FREE FLOATING BELT SKIMMER TO WATER SURFACE
16
-------
FIGURE 2
POLYURETHANE BELT DETAILS
17
-------
FIGURE 3
CONVEYOR WRINGER ASSEMBLY
18
-------
Deflector Trough
o
t—I
f
CO
I—I
2
-rj
33
t~*
O
Spring Loaded Conveyor
Porous Conveyor Belt
Fixed Conveyor
Take-Up
Perforated
Pressure
Plate
Shock Absorbing
Guide Roller
-------
portion of the absorbed oil can be removed from the belt than with
the conventional "clothes wringer" apparatus consisting of two spring-
loaded pressure rollers located one above the other. The conveyor
wringer provides the much needed time under pressure which is necessary
to remove viscous oil from the absorbent belt material at the required
belt feed rates. For the belt feed conditions tested during the
prototype study, the wringing time was 1.9 to 9.0 seconds for the
conveyor wringer, which corresponds to 0.05 to 0.25 seconds for the
roller wringer. Bench scale studies of the roller wringer demonstrated
that for these feed rate conditions the pressure roller could not
prevent excessive recycle of oil, even at elevated wringer pressures.
Visual as well as quantitative observations indicate that the conveyor
wringer consistently provides a much drier and oil free belt than the
roller wringer.
The skimmer and floating belt were designed to provide a quick and easy
method to change belts. Belt changes on the experimental prototype
skimmer could be accomplished by three men in approximately 10 minutes.
Two men were assigned to change the belt and the third was required
to operate the controls. A belt change was accomplished by stopping
the belt when the pin connector was at the top of the skimmer and
easily accessible (Figure 5). The connector pin was then pulled and
the new belt to be installed was connected to the tail end of the old
belt. With the skimmer drive turned on, the new belt was "threaded"
through the wringer by the old belt. When the old belt has cleared
the entire path of travel, the old and new belts were disconnected and
the head and tail ends of the newly installed belt were connected.
The skimmer was then ready for operation.
A photograph of the prototype oil skimmer test site is shown in Figure 6.
The test tank is 60 feet in diameter with a 2 ft wide by 6 ft high
annular water section filled to a 4 ft depth. The oil skimmer is
mounted from a frame extending above the annular tank on a pivoted
platform which was continuously rotated during the pickup operation at
speeds up to 5 knots (Figure 7 and 8). Beneath the center pivot in
Figure 7 are stationary, annular supply and collection tanks into and
out of which oil could be transferred while the platform was in motion.
A uniform layer of oil could be spread on the water surface through
the distribution header (Figure 9). During the run another pump was
utilized to drain the oil-water mixture from the skimmer trough to the
collection tank. Manual diversion valves on the collection line were
used for sampling and flow measurement. A multi-partitioned, calibrated
sampling chamber was utilized to collect samples at intervals during the
test run. Before reuse, the collected oil-water was pumped to a holding
tank (Figure 10) for separation.
In order to simulate actual pickup conditions in protected waters, 2 ft
wave conditions were created for the oil skimming tests by a wave
generator using the submerged piston technique. This device was designed
under the direction of a project consultant, Dr. Clifford Mortimer,
20
-------
FIGURE 5
REMOVAL OF THE PIN CONNECTOR TO CHANGE A BELT
21
-------
FIGURE 6
EXPERIMENTAL PROTOTYPE TEST SITE
22
-------
FIGURE 7
ANNULAR COLLECTION AND SUPPLY TANKS
23
-------
FIGURED 8
ROTATING PLATFORM ASSEMBLY
24
-------
FIGURE 9
OIL DISTRIBUTION HEADER AND OIL SATURATED POLYURETHANE BELT
25
-------
FIGURE 10
OIL SEPARATION TANK
26
-------
Director of the Center for Great Lakes Studies, University of
Wisconsin-Milwaukee. In all ways but one, the problem of wave
generation in channels has been extensively studied. The unknown
area involved the efficient refraction of waves to make them
travel around a circular annual tank. The underwater piston
technique, driven by a motorized cable transmission system (Figure
11) produced a very uniform wave of approximately 17 ft. wave
length and comparable to the short waves found in harbors and
protected waters.
27
-------
FIGURE 11
WAVE GENERATOR DRIVE
28
-------
SECTION VII
EXPERIMENTAL PROCEDURES
The experimental prototype skimmer and test tank were located outside
and adjacent to the Rex Chainbelt Technical Center. The test tank was
filled to a four foot depth with City of Milwaukee tap water from a
nearby fire hydrant. Each tankful of water was used for approximately
30 tests before it was drained and refilled.
The recovery of three types of oils (Oklahoma Crude, #2 Fuel Oil, and
Bunker C) was evaluated during the test program. These oils were stored
outside the test tanks in 55 gallon shipment drums until they were
needed for tests. They were then pumped to the oil supply tank located
in the center of the annular test tank (Figure 7).
Prior to the initiation of an experiment, the particular belt which was
to be evaluated was installed according to the procedure outlined in
the Design of Experimental Prototype Section.
With the exception of two groups of tests, the maximum oil pickup tests
and the wringer efficiency tests, all experiments were conducted using
the standard procedure outlined below. Hereafter, all experiments using
these procedures will be referred to as standard procedure tests (SPT).
1. A calculated amount of oil to provide the desired initial oil-
layer thickness was spread uniformly over the entire water
surface of the test tank. The amount of oil spread was moni-
tored by means of a calibrated speedometer on the pump drive.
In addition, the change in oil level in the supply tank was
measured during each experiment. The change in oil level was
determined to more accurately reflect the amount of oil spread
and was therefore used in all calculations. No oil was added
to the tank surface during cleanup experiments.
2. The wringer was adjusted to the desired pressure.
3. The wave generator was started and adjusted to provide two
foot waves. The time required to attain the desired wave
condition was approximately 5 minutes.
4. The rotating platform was started and adjusted to the desired
forward velocity.
5. The skimmer was turned on and adjusted to the proper speed.
In general, Steps 4 and 5 were accomplished in less than 1/4
of a tank revolution.
6. Samples of the oil-water mixture were obtained by diverting
the entire flow from the oil collection pump to the 5-gallon
29
-------
sample containers at various time intervals during the test
run. Initially, samples were taken at very close time
intervals (on the order of 30 seconds). The time interval
between samples was gradually increased as the amount of oil
left on the surface diminished with time. The volumetric rate
of flow was determined by measuring the time required to collect
each sample. Flow which was not collected in the sample con-
tainers was pumped to the collection tank in the center of the
annular tank.
7. The skimming was continued until, by operator's judgment* a
substantial portion of the spilled oil had been collected.
8. The volume of oil-water mixture in the center collection tank
as well as in each sample container was measured and recorded.
9. The oil-water mixture in the sample tanks was allowed to separate
for one hour and the volume of oil and water was then recorded.
During the tests using Oklahoma Crude it was necessary to add
3000-4000 ppm of Nalco 7713 to break the emulsion which had
formed.
10. The oil-water mixture in the collection tank was then transferred
to the oil separation tank. The oil in the separation tank
was pumped back to the supply tank for reuse and the water was
returned to the annular test tank.
11. Prior to the initiation of a new experiment the water surface
was visually checked for residual oil. If any substantial
amount of oil (- 2 gallons or more) was present additional
skimming was conducted prior to spreading more oil.
Collection and Analysis of Standard Procedure Experiment Data
As explained in the previous pages, periodic samples of the oil-water
mixture were collected and recorded during the duration of an experiment.
A sample data sheet from Experiment No. 35 is shown in Table 2 to
Illustrate the means by which the data was collected and results were
determined.
A computer program (PETSKI) was developed to calculate the various
parameters of interest. These parameters included:
1. Percent oil in the recovered oil-water mixture
2. Oil-water mixture pickup rate (gpm).
3. Oil pickup rate (gpm)
4. Cumulative oil volume collected
5. Cumulative percentage of oil recovered
6. Equivalent oil slick thickness remaining at the time of
the sample.
30
-------
Operated by: MS, TR^ MG Experiment No. 35 Date: 7/13/71 Duration of Experiment 12 min 39 sec.
Oil Type & Viscosity #2 Fuel Test Plan Code # Oil Temperature: 88°F
Belt Code No. 3/8" x 45 ppi Oil Thickness 0.125 in. Wringer Pressure 4.0 psi
Wringer Setting
Platform Setting
Oil Supply Tank Level
Platform Velocity 2.5 knot
Start
End
% Oil at Surface —
Skimmer Velocity 55 fpm Skimmer Setting
1.772 Oil Collection Tank Level
27
1.600
Amount of Oil Spread =»28 gallons Oil Supply Pump Setting 51
Time of Oil Spread — Platform Velocity
No. of Revolutions
Start
End
Oil Supply Rate _
Platform Setting _
0.616
0.962
14 gpm
32
o
M
f
w
F
H
cn
Sampling
Time
Min Sec
0
1
2
2
3
5
6
8
10
12
15
10
0
50
45
0
45
0
0
0
Sample
Collection
Time sec.
29.9
24.9
25.0
25.8
26.
30.
30.0
29.4
27.9
29.8
,3
,1
Sample
Volume
gal.
4.25
4.47
4.29
4.32
4.40
4.45
4.10
4.0
4.24
4.23
Settling
Time
hrs.
1 hr.
1 hr.
1 hr.
1 hr.
1 hr.
1 hr.
1 hr.
1 hr.
1 hr.
1 hr.
Volume
of oil in
Sample gal.
re
N>
,74
,35
2.20
1.90
1.35
1.07
0.65
0.30
0.20
0.16
CO
sc
W
W
H
-------
TABLE 3
OUTPUT FROM COMPUTER ANALYSIS OF EXPERIMENT NO. 35
RUN NUMBER 35
AMOUNT OF OIL SPREAD 27.04 GAL
INITIAL OIL DEPTH .1190 INCHES
TIME VOLUME VOLUME PERCENT COLLECTION SAMPLE CUMULATIVE EQUIV
INTO OF OF OIL RATE MIDPOINT OIL PCT OIL
RUN SAMPLE OIL MIX OIL VOL REC DEPTH
(MIN) (GAL) (GAL) (PCT) (GPM) (GPM) (MIN) (GAL) (PCT) (IN)
.3
1.2
2.0
2.8
3.8
5.0
6.8
8.0
10.0
12.0
4.25
4.47
4.29
4.32
4.40
4.45
4.10
4.00
4.24
4.23
1.74
2.35
2.20
1.90
1.35
1.07
.65
.30
.20
.16
40.94
52.57
51.28
43.98
30.68
24.04
15.85
7.50
4.72
3.78
8.53
10.77
10.30
10.05
10.04
8.87
8.20
8.16
9.28
9.13
3.49
5.66
5.28
4.42
3.08
2.13
1.30
.61
.44
.35
.5
1.4
2.2
3.0
4.0
5.3
7.0
8.2
10.2
12.2
.9
4.9
9.4
13.5
17.0
20.3
23.3
24.5
25.5
26.3
3.2
18.0
34.9
50.0
62.7
75.1
86.2
90.6
94.5
97.4
.115
.098
.077
.060
.044
.030
.016
.011
.007
.003
TOTAL MIXTURE COLLECTED 109.6 GAL
PERCENT OIL IN THE MIXTURE 24.02 PERCENT
AVERAGE COLLECTION RATES:
MIXTURE
OIL
8.66 GPM
2.08 GPM
32
-------
The computer program was run OH a General Electric Co, timesharing
computer with input and output being handled through a teletype terminalc
A program listing for PETSKI together with definitions of the variables
The computer output from experiment No0 35 is shown in Table 3,, Output
information from the computer program was used to evaluate the skimmer
performance under the various operating conditions.
The results of these calculations permitted observation of a number of
effects. For example„ instantaneous oil pickup rates and oil percentages
could be determined for various equivalent oil slick thicknesses„ It
was also possible to obtain average collection rates and oil percentages
duriag an entire slick pickup operation. For experiment No0 35S the
average oil pickup rate was 2,08 gpra during the time when the slick thick-
ness varied from 00119 inches to 0,003 inches,, Also, the total oil-water
mixture collected during this experiment contained 24 = 0 percent oil,,
To facilitate interpretation and analysis of the datas the results of
each experiment were plotted in a manner similar to that shown in
Figure 12,
Maximum Oil Pickup Rate Tests
A series of experiments were conducted to evaluate the maximum low
viscosity oil pickup capacity of the skimmer with 1 and 2 inch thick
polyurethane belts„ The tests were conducted in quiescent water with
approximately a 6 inch layer of Oklahoma crude oil. The rotating
platform was maintained in a fixed position (zero forward velocity),
The tests were conducted at four skimmer (belt) speeds» 20, 559 90 and
140 feet per minute for each of the 6 belts listed below in Table 4,
TABLE 4
MAXIMUM OIL PICKUP RATE TEST CONDITIONS
Belt Thickness s, inches Pores/Lineal inch (ppi)
1 1 20
2 1 45
3 1 80
4 2 10
5 2 20
6 2 45
Oil pickup rate as a function of belt speed is shown in Figure 13 for
the various belts tested. The general trend is for an increase in pick-
up rate with an increase in belt speed up to a maximum level.
33
-------
u>
Equivalent
Slick Thick
ness
03
0)
«J
c
m
CO
I
o
•H
H
B
9)
r-l
a
cr
H
Time (Minutes)
-------
C/3
•
a
!-i
en
Pi
MAXIMUM OIL PICKUP RATE RESULTS
Belt Thickness Belt Porosity
Inches pni
60 80
Belt Speed (Feet/Min)
100
120
140
-------
Two Inch thick belts of 10 and 20 pores per lineal inch showed the
highest pickup rates. This observation is consistent with theory since
these belts presented the maximum amount of sorbtion volume and had an
open surface which permitted the oil to penetrate into the foam.
In order to compare these results with Tthose of Schatzberg and Nagy (18),
the oil pickup rate was converted to pounds of oil per pound of sorbent
and plotted versus belt speed (Figures 14 and 15). Although there is
significant variation in sorption capacity with belt porosity the
maximum pickup capacity was observed to be 22 pounds of oil per pound
of sorbent. Schatzberg reported 30.6 pounds of oil/pound of sorbent
for a light crude oil. Although there are a number of possible reasons
for this discrepancy in results, it is felt that the primary reason is
one of sorption contact time. All of the data, with the exception of
the 1" thick 20 ppi belt indicates a decreasing unit pickup rate with
increasing belt speed (decreasing belt contact time). Schatzberg'
data was collected in a series of laboratory tests utilizing 15 minutes
oil-sorbent contact time.
This information is significant and indicates that maximum sorbtion
efficiency can probably not be obtained with a continuous belt concept,
due to the difficulty encountered in obtaining adequate contact time.
Wringer Efficiency
Utilization of an endless belt oil skimming concept requires a highly
efficient wringer in order to be able to return a clean belt to the oil/
water interface. Throughout the design phase, significant effort was
devoted to the development of a high efficiency wringer. The wringer
used in the experimental prototype skimmer has proven to meet these needs.
During skimming operations it was possible to wipe your hand across the
surface of a wrung belt and detect only a very minimal amount of oil.
Initial tests indicated that no significant improvement in wringer
efficiency at pressures over 4 psi. Accordingly it was decided to
conduct all tests at a wringer pressure of 4.0 psi.
A series of tests were conducted to quantitatively determine the effi-
ciency of the wringer. Twenty inch long sections of oil saturated belt
(polyurethane foam and neoprene rubber backing) were put through the
wringer and weighed to determine the amount of oil removed from the
belt by the wringer. These tests indicate that approximately 91% of //2
fuel oil and 80% of Bunker C could be removed from 3/8 and 1 inch thick
polyurethane belts by the wringer. For the purpose of comparison, tests
conducted with the bench scale roller wringer indicated wringer efficien-
cies of approximately 40-60% depending on belt and oil type.
It is felt that a substantial portion of the oil remaining on and in
the belts is being reabsorbed from the perforated conveyor belts. This
oil could be removed by applying a vacuum to the final section of the
wringer to assist in removing oil from the conveyor belts. Although
36
-------
25
u>
r1
CS
CO
w
n
o
o
r-1
o
tr1
50
tr. '
t-1
I-H
d
CO
H
Crt
^
8
20
P-
D.
CO
V
kl
O
a,
fl>
0)
O
C£
rH
•H
O
10
15
10
0
1 Inch Thick Belt
80 ppi
45
Q 20 pp3
20
40
60 80
Belt Speed (Feet/Min)
100
120
140
-------
o
1-1
F4
PO
O
O
3
8 «
P3
O
CM
W
f
H
CO
60 80
Belt Speed (Feet/Min)
100
120
140
-------
the experimental prototype skimmer was designed to utilize vacuum in
the final one foot of the wringer, a vacuum pump was not purchased
because initial tests indicated a high efficiency in the wringer.
A vacuum section should probably be included in any future wringers,
especially if recovery of very thin oil layers (sheens) are to be
attempted.
Results of Standard Procedure Tests
The standard procedure tests (SPT) were conducted according to the methods
outlined in a previous section. The objective of these tests was to
determine the oil pickup rate and percentage of water in the oil-water
mixture for various polyurethane belt types (thickness and pore size),
belt speeds and forward skimmer velocities. As previously mentioned,
all tests were conducted under the condition of a slightly progressive
wave with an amplitude of two feet.
Nearly all oil skimmers presently in service make use of deflector booms
attached to the front of the recovery vessel to sweep a wide area ahead
of the boat and funnel the oil to the skimming device. This technique
has significant advantages in that it permits a vessel to cover a some-
what wider area with a single pass as well as presenting a thicker oil
layer to the skimmer than exists on the water surface ahead of the vessel.
Attempts were made to utilize a deflector boom concept of the experimen-
tal prototype skimmer. However, due to the geometry of the test tank,
it was found that oil recovery rates were decreased when the deflector
boom was installed. This effect was attributed to the tangential wave
action in the tank, and the subsequent "pumping" of water through the
boom entrance. Consequently this technique was abandoned. It is
therefore important to realize that the data presented in this report
reflect the actual equivalent oil slick thickness and that no thickening
in the approach area to the skimmer was accomplished. Further implica-
tions of this fact will be discussed later.
The amount of oil to be encountered by an oil slimmer and consequently
available for recovery is a function of the oil slick thickness, the
skimmer width and the forward velocity of the skimmer. Figure 16 shows
the amount of oil available to the experimental prototype skimmer tested
in this project for various forward velocities and slick thickness.
This figure therefore indicates the maximum oil pickup rates which could
be attained under the specific conditions of a particular test.
The oil pickup capacity of the polyurethane belts can be defined by the
following equation;
Qn = K x Vn x A x 7.48
O D
where: QB = belt pickup capacity in gpm
39
-------
2.5
Forward Velocity (Knots)
FIGURE 16
OIL AVAILABLE FOR SKIMMING VS. FORWARD VELOCITY-
EXPERIMENTAL PROTOTYPE SKIMMER
40
-------
Vg = Belt velocity in feet/minute
A = cross sectional area of belt in square feet
7.48 = conversion from cubic feet to gallons
K = Dimensionless sorption coefficient
Based on the results of the maximum pickup rate tests it would appear
that K and VB are related, in that sorption is a function of contact
time at least over the range of belt speeds (contact times) studied.
It should be noted that this equation is applicable only for oils which
are readily absorbed into the foam material. The equation should not,
for example, be used with Bunker C oil for which absorbtion is minimal
due to its viscous nature. It can, however, be said that in general,
the oil pickup rate for low viscosity oils can be increased by increas-
ing either the belt speed and/or the cross sectional area of the belt.
Tests with #2 Fuel Oil
Thirty-six SPT tests were conducted with #2 Fuel Oil. The results were
processed using the time sharing computer program PETSKI and subjected
to an analysis of variance to define the significant variables affecting
oil pickup rate and percentage of oil in the oil-water mixture which
was recovered.
Based on the results of the analysis of variance, it was determined that
the forward platform velocity did not affect either the oil pickup rate
or the percentage of oil in the recovered mixture at the 95% level.
Consequently all of the data which follows has been calculated on the
average of the results obtained at 1 and 2.5 knots forward velocity.
Variables which significantly influence oil pickup rate and the percent
of oil in the recovered mixture include belt speed, belt thickness, pore
size of the polyurethane and slick thickness.
For a given thickness of belt material, the volume of sorbent material
presented to the oil surface can be increased by increasing the speed
of the belt. Figures 17 and 18 show the oil pickup rate as a function
of belt speed for the 20, 45 and 80 ppi belts at both 3/8 and 1 inch
belt thicknesses. Maximum oil pickup rates were obtained at a belt
speed of 90 feet per minute.
Oil pickup rate is also influenced by the porosity of the belts. For
the three porosities tested, the maximum oil pickup rates were observed
to exist with 80 ppi belts. It is felt that for low viscosity oils,
80 and even 100 ppi belts should be used since they present more resis-
tance to outflow or drainback of oil than the higher porosity belts as
well as being less susceptible to water pickup.
The oil recovery rates for the belts investigated are shown at various
slick thicknesses up to 0.10 inches in Figures 19 and 20.
41
-------
12
10
8
#2 Fuel Oil
3/8" Thick Polyurethane Belts
0.1" Slick Thickness
0
a
20ppl
45 ppi
80 ppl
B.
60
•8
20
40 60
Belt Speed ft/min
FIGURE 17
80
100
OIL PICKUP RATE VS. BELT SPEED
42
-------
#2 Fuel Oil
1" Thick Polyurethane Belts
0.1" Slick Thickness
40 60
Belt Speed, Ft./Min.
FIGURE 18
OIL PICKUP RATE VS. BELT SPEED
100
43
-------
92 Fuel Oil
3/8" Thick Polyurethane Belt
Belt Speed = 90 fpm
.02 .04 .06 .08
Slick Thickness, Inches
FIGURE 19
OIL PICKUP RATE VS. SLICK THICKNESS
.10
44
-------
#2 Fuel Oil
1" Thick Polyurethane Belt
Belt Speed = 90 fpm
.04 ,,06
Slick Thickness
.10
FIGURE 20
OIL PICKUP RATE VS. SLICK THICKNESS
-------
Although there is significant variation among the various porosities and
belt thicknesses, an approximately four fold decrease in pickup rate
was observed for the 1" 80 ppi belt as the slick thickness decreased
from 0.10 to 0.04 inches. This information confirms early comments
regarding slick thickness and indicates the need for immediate response
to oil spills as well as the use of deflector booms to thicken the oil
immediately in front of the skimmer.
The percentage of oil associated with the recovered oil-water mixture
was determined for each experiment which was conducted. These data are
presented in Figure 21 through 26 and indicate how the oil content
varied with slick thickness, belt velocity and belt porosity. The
highest percentage of oil was obtained with the 80 ppi belts at a slick
thickness of 0.10 inches. Although belt speed exerts some influence on
oil content of the recovered oil-water mixture, the effect is minor
compared to belt porosity and slick thickness. Based on these data,
it can be concluded that for //2 Fuel Oil the maximum percentage of oil
which can be obtained with this skimmer at slick thicknesses of 0.10
inches and less is about 50%.
An apparent discrepancy exists for the optimum foam as determined for
the Maximum Oil Pickup Rate tests (Figure 13) and the #2 Fuel Oil tests.
Maximum oil pickup rate was observed in thick oil layers (5-6 inches)
with a 20 ppi belt. However, when the SPT tests were conducted on thin
layers (0.125 inches and less) the 80 ppi belt was optimum. There is
little doubt that the open pore 20 ppi belt has the greatest oil pickup
capacity. However, in thin slicks, the resistance of a belt to water
becomes at least as important as the pickup capacity of the belt. The
affinity of the 20 ppi belts for water appears to impede the pickup of
oil and hence for thin slicks an 80 ppi belt which has less liquid
recovery capacity but more resistance to water pickup was found to be
optimum.
The optimum operating conditions and results observed during these tests
when skimming #2 Fuel Oil are summarized in Table 5. It is felt that
operation of the skimmer under these conditions will yield the best
overall results.
TABLE 5
OPTIMUM OPERATING CONDITIONS AND RESULTS FOR #2 FUEL OIL
Wringer Pressure 4.0 psi
Belt Speed 90 fpm .
Belt Thickness 1.0 inches
Belt Porosity 80 pores/lineal inch
Forward Skimmer Velocity 1-2.5 knots
Expected Operating Results at 0.10 in. Slick Thickness
Oil Pickup Rate 8.3 gal/min
% Oil in Recovered Oil-Water Mixture 49
46
-------
3/8" Thick Belt
20
#2 Fuel Oil
• 20 fpm
0 55 fpm
0 90 fpm
.04 .06 .08
Slick Thickness, Inches
FIGURE 21
PERCENT OIL IN RECOVERED MIXTURE VS. SLICK THICKNESS
47
-------
70-
60
50
3/8" Thick Belt
45 ppi
n Fuel Oil
Belt Speed
o
D
20 fpm
55 fpm
90 fpm
X
•H
* 40
•o
4)
30
04
c
20
10
0
0
.02
.04 .06 .08
Slick Thickness, Inches
FIGURE 22
.10
PERCENT OIL IN RECOVERED MIXTURE VS. SLICK THICKNESS
48
-------
3/8" Thick Belt
80 ppi
#2 Fuel Oil
20 fpm
55 fpm
90 fpm
0
.04 .06 .08
Slick Thickness, Inches
FIGURE 23
PERCENT OIL IN RECOVERED MIXTURE VS SLICK THICKNESS
49
-------
1" Thick Belt
20 ppi
.04 .06 .08
Slick Thickness, Inches
FIGURE 24
PERCENT OIL IN RECOVERED MIXTURE VS SLICK THICKNESS
50
-------
70
60
«
S 50
x
4)
40
o
•o
-------
1" Thick Belt
80 ppi
Fuel Oil
• 20 fpm
55 fpm
0 90 fpm
.04 .06 .08
Slick Thickness, Inches
FIGURE 26
PERCENT OIL IN RECOVERED MIXTURE VS. SLICK THICKNESS
52
-------
TABLE 5 (CONT.)
Operating Results at 0.04 Inch Slick Thickness
Oil Pickup Rate 2.6 gal/min
% Oil in Recovered Oil-Water Mixture 22
Tests with #6 Oil (Bunker C)
Twenty-two separate tests were conducted using Bunker C oil to evaluate
the oil recovery rate and oil content in the recovered mixture. The
tests were conducted and analyzed in the same manner as was reported
for the #2 Fuel Oil tests.
Because of the highly viscous nature of this oil certain difficulties
were experienced with these tests which were not encountered in the #2
Fuel Oil tests. These difficulties are enumerated below.
1. Because of its physical properties, Bunker C did not spread
uniformly over the water surface. The general tendency was
for the formation of oil patches and non-uniform slicks at
various locations in the tank. As a result of the physical
constraints of the test system it was not possible to move
directly into an oil slick as would normally be done on an
open water oil spill. The data are therefore somewhat less
sensitive to the slick thickness than was observed in //2
Fuel Oil tests.
2. A significant amount of oil was alternately deposited and
washed from the side walls of the annular tank as a result
of wave action. This oil was unavailable for skimming
while it was on the tank wall and therefore reduced both
the percentage of oil recovered and the oil pickup rates.
3. On cold days the Bunker C tended to form large oil balls due
to the wave action in the tank. The skimmer was unable to
remove these balls some of which were estimated to weigh in
excess of 50 pounds. An oil ball of this size is equivalent
to approximately 25% of the oil which was spread for a
particular test.
For these reasons a less extensive test program was conducted with
Bunker C than with the #2 Fuel Oil.
The oil pickup rates for the various belt speeds studied are shown in
Figures 27 and 28. The maximum pickup rates at 0.1" equivalent slick
thickness are 3.7 gpm for a 45 ppi 3/8" thick belt and 4.2 gpm for a
20 ppi 1" thick belt. Figures 29 and 30 illustrate how these pickup
rates change with slick thickness. It can be noted that the change in
pickup rate is not as significant as was observed in the #2 Fuel Oil
tests. It is felt that this is a result of the viscous nature of
53
-------
5 _
o
M
f
*v
M
n
CO
w
H
"0
w
s
ro
•vj
o.
60
I
u
1F6 Oil - Bunker C
3/8" Thick Polyurethane Belts
0.1" Slick Thickness
20 ppi
45 ppi
80 ppi
60 80
Belt Speed, ft/min
100
120
140
-------
v_n
o
M
t-
o
3
n
C/J
ta
r*
H
g
o
ro
oo
E
CL.
00
ce
P.
o
0
O
T
#6 Oil - Bunker C
1" Thick Polyurethane Belts
0.1" Slick Thickness
20 ppi
45
/
20
40
60 80
Belt Speed, ft/min
100
120
140
-------
#6 Oil - Bunker C
3/8" Thick Polyurethane Belt
Belt Speed - 140 fpm
.04 .06
Slick Thickness, Inches
FIGURE 29
OIL PICKUP RATE VS. SLICK THICKNESS
.10
56
-------
#6 Oil - Bunker C
1" Thick Polyurethane Belt
Belt Speed - 55 fpm
.04 .06
Slick Thickness, Inches
FIGURE 30
OIL PICKUP RATE VS. SLICK THICKNESS
57
-------
Bunker C and reflects a tendency for belt surface pickup rather than
absorption into the polyurethane foam.
The change in percentage of oil in the recovered oil-water mixture as
a function of slick thickness is shown in Figures 31 through 35. The
highest oil percentages over the entire range of slick thickness was
attained with a 3/8" thick 45 ppi polyurethane belt.
Based on the results of the Bunker C tests it is recommended that the
skimmer be operated under the conditions shown in Table 6 to obtain
best results in a Bunker C recovery operation.
TABLE 6
RECOMMENDED OPERATING CONDITIONS FOR RECOVERY OF BUNKER C
Wringer Pressure 4.0
Belt Speed 140 ft./min
Belt Thickness 3/8" polyurethane
Belt Porosity 45 ppi
Forward Skimmer Velocity 1-2.5 knots
Expected Operating Results
At 0.10 inch slick thickness
Oil Pickup Rate 3.7 gpm
% Oil in Recovered Oil-Water Mixture 72
At 0.04 inch slick thickness
Oil Pickup Rate 1.9 gpm
% Oil in Recovered Oil-Water Mixture 35
Oklahoma Crude Oil Tests
The initial testing of the experimental prototype oil skimmer was
performed using an Oklahoma crude oil which was selected as being
representative of low viscosity oils. This oil had the characteristics
shown in Table 7.
TABLE 7
CHARACTERISTICS OF OKLAHOMA CRUDE OIL
Density 0.83 gms/cm^
Temperature 26.7°C
Viscosity @ 100°F 2.9 centipoise (Ostwald)
58
-------
70
60
«,50
Vi
w
X
•H
£
Q)
4J
•H
o
•c
0)
8 3d
3/8" Belt
20 ppi
#6 Fuel Oil - Bunker C
Belt Speed Vg
O
Q
55 fpm
90 fpm
1AO fpm
2(
10
.02 .04 .06
Slick Thickness, Inches
FIGURE 31
.08
.10
PERCENT OIL IN RECOVERED MIXTURE VS. SLICK THICKNESS
59
-------
70
3/8" Belt
45 ppi
#6 Fuel Oil Bunker C
0)
3
x 50
-------
70 -
3/8" belt
80 ppi
#6 Fuel Oil - Bunker C
• 55 fpm
O 90 fpm
0 140 fpm
.04 .06 .08
Slick Thickness, Inches
FIGURE 33
PERCENT OIL IN RECOVERED MIXTURE VS. SLICK THICKNESS
61
-------
70
60
01
n
4J
x
•H
33
50
40
cu
i-i
0)
o
830
oi
T
1" Belt
20 ppi
//6 Fuel Oil - Bunker C
Belt Speed Vfi
•
0
55 fpm
90 fpm
20
10
.02 .04 .06
Slick Thicknesst Inches
FIGURE 34
.08
.10
PERCENT OIL IN RECOVERED MIXTURE VS. SLICK THICKNESS
62
-------
70
1" Belt
45 Ppi
#6 Fuel Oil - Bunker C
60
Belt Speed VB
• 55 fpm
QJ 50
a
0)
•u
40
i
/
•o
0)
20
10
.02
.04 .06 .08
Slick Thickness, Inches
FIGURE 35
.10
PERCENT OIL IN RECOVERED MIXTURE VS. SLICK THICKNESS
63
-------
This oil tended to emulsify very easily and required the use of 3000-4000
ppm of Nalco 7713 to break the emulsion. After the addition of the
Nalco 7713 the emulsion separated by gravity with the oil containing
only 1-2% water.
During the entire duration of the Oklahoma crude experiments the per-
cent oil in the recovered mixture were quite low. Subsequent investi-
gations demonstrated that the oil was in fact emulsifying on the surface
of the test tank. The reason for this emulsification is thought to be
due to the oil characteristics in combination with the extremely turbu-
lent conditions created by the wave action.
Samples collected from the surface of the oil slick after the wave
generator had been in operation for approximately 10 minutes indicated
that the oil contained 45% water. Although the skimmer was able to
effectively recover the emulsified oil, it was not possible to accurately
evaluate the skimmer performance unless frequent analysis of the water
content of the emulsion was made. Consequently, it was decided to use
the information collected for #2 Fuel Oil as being representative of the
low viscosity oils.
As a result of this experience, it is recommended that in the case of
spills of easily emulsified oils either shipboard or shore aide facili-
ties will be required to de-emulsify the recovered oils. This facility
would require either chemical treatment equipment or some other
feasible means for emulsion breaking.
64
-------
SECTION VIII
DISCUSSION OF RESULTS
The results obtained during the evaluation of the experimental prototype
skimmer have been presented in the previous section. In addition to
the quantitative data developed in these tests, certain qualitative
information and engineering observations were made which provide addi-
tional insight into the overall project objectives.
Mechanical Wringer and Drive System
The efficient wringing of sorbent materials is essential to reuse of
the sorbents. As was indicated in the previous sections, the conveyor
wringer used in this skimmer has proven to be a highly effective method
for removing oil from compressible, resilient sorbents. The wringer
assembly, although somewhat complex in appearance, proved to be
essentially maintenance and trouble free during the duration of the
project.
The conveyor wringer also provides a very effective means of powering
belts and should have application with other commercial concepts such
as the Shell Pipeline Corporation "Oil Scrubber". Very little belt
slippage occurs even with a heavily oil laden belt because of the large
contact area available for driving the belt.
There are two items on the wringer assembly which should be modified in
future designs to provide improved performance.
1. The number of powered rollers can be reduced with no loss in
drive capabilities. This will simplify the apparatus and help
to reduce the cost of manufacturing.
2. The size of the rollers on the conveyor wringer should be
increased to enable the installation of a large oil collection
pan under the pressure plate. This will prevent oil overflow
at high oil pickup rates.
If sorbent materials are used which are strong enough so that a nonporous
backing material is not required, the bottom pass of the belt, which is
used to invert the belt surface prior to passage through the wringer,
could be eliminated.
Belt Integrity
During the bench scale testing phase, questions were raised as to the
strength of the polyurethane materials and the bond of the polyurethane
to the neoprene rubber backing. An endurance test was conducted to
evaluated the useful life of a 45 ppi neoprene backed polyurethane
belt. The endurance test was conducted on the bench scale skimmer under
65
-------
the conditions shown in Table 8. Although the belt was severely slashed
by the foreign materials intentionally inserted in the belt, it
retained its integrity through 100 hours of operation. At that time,
the endurance test was terminated since it was felt that no additional
information would be derived from further testing.
TABLE 8
CONDITIONS OF ENDURANCE TEST 45 PPI POLYURETHANE BELT
Hours
0-6 Run on bench scale skimmer in tap water with Mobil Naprex
#920 - viscosity 60 SUS/100°F oil layer
6-13 Run on bench scale skimmer in tap water with Bunker C -
viscosity approximately 3000 SUS/100°F oil layer
13-35.5 The belt was impregnated with sticks, approximately 10
2-inch long nails, large chunks of broken glass and 5
triangular shaped steel wedges. The belt was run in
water covered with a layer of Bunker C oil.
35.5-100 The intentionally "tortured" belt was run in oily, sea
water obtained from the hold of a tanker to determine
whether sea water would exert an acceleration on rate of
hydrolysis.
100 Endurance test terminated
During the experimental prototype test phase, close attention was paid
to belt integrity. At no time during the tests was any belt deterio-
ration noted even though some belts were operated for up to 30 hours
and remained oil wetted for periods up to 5 months.
Due to the free floating orientation of the belt, it is extremely
unlikely that the belt will pick up or be severely damaged by the debris
which is normally encountered in spill cleanup operations.
Based on these results, it is felt that neoprene backed polyurethane
foams will provide a minimum of 1-2 days serviceable life even under
adverse spill conditions.
Use of Polyurethane Sorbents
As a result of the extensive testing done with the polyurethane foams
(Scott Industrial Foam, Reticulated) during this project certain conclu-
sions can be drawn regarding the suitability of this material as an
oil sorbent medium.
1. When backed with a suitable material, polyurethane foams are
66
-------
durable and can be expected to provide a minimum of 1-2 days
of service in oil recovery operations.
2. Although highly touted in both commercial and technical
literature as a hydrophobic, oleophilic material, polyure-
thane foams absorb significant quantities of water (=50% of
total liquid sorbed) when used to absorb oil on thin slicks
(0.10 in. or less).
3. Because of the short sorption times available, polyurethane
foams when used in an endless belt concept are incapable of
sorbing the same unit volumes of oil that are reported in the
literature.
4. Utilizing the free floating belt concept developed in this
project the polyurethane foams are more effective in recovering
light oils (#2 Fuel) than heavy oils (//6 Fuel).
Studies with Other Sorbent Materials
An attempt was made to evaluate the sorption of oils with materials
other than polyurethane foam. Three materials were investigated,
Acrylonitrile bonded polyester mats, zero twist polypropylene fiber and
polypropylene microbatt.
The Acrylonitrile polyester mats were bonded to neoprene rubber backing
using Carboline adhesive. This belt proved unsatisfactory in that it
was not resilient enough to permit continued wringing. Belt deteriora-
tion was noted almost immediately and belt failure occurred after
approximately 15 minutes of use. Because of these difficulties, no
further tests were conducted with this material.
The Eastman Chemical Company provided samples of zero twist polypro-
pylene fiber. This material was hand woven into a pigtail type braid
and tested on the bench scale skimmer. Although the material does not
release oil very well in the braided form, it is felt that If the
material could be woven into a belt material and tufted, it would
provide a satisfactory sorbent material. One of the real advantages
of polypropylene is that it is extremely hydrophobic and should enable
the recovery of essentially water free oil.
Rex is currently pursuing the development of a polypropylene woven belt
for testing on future oil skimming applications.
Some effort was devoted to evaluating a microbatt polypropylene belt.
Polypropylene in this form is extremely weak and must be encased In some
support media to provide strength and resiliency. A belt of this
material was provided by the Hercules Corporation for testing on the
experimental prototype skimmer. The microbatt was encased in Delnet
GQ16 and a crimped fibrlllated web was mixed with the microbatt to
provide resiliency. Although laboratory tests had shown the microbatt
67
-------
polypropylene to be extremely hydrophobia the field tests indicated
that approximately 50% of the recovered liquid was water. Further
investigation revealed that water was being mechanically trapped in the
Delnet casing which was the reason for the high water pickup.
Future attempts to evaluate the polypropylene microbatt should utilize
a more open casing material to facilitate oil penetration as well as
free water drainage.
Summary
In general the results of this project have been encouraging, partic-
ularly the performance of the conveyor wringer and the free floating
belt concept. Accordingly a conceptual design of a boat and the
neceasary support equipment for a five foot wide free floating belt
skimmer is presented in the following section.
It is felt that significant improvements can be made in the overall
results and performance of this proposed skimmer with the development
of new sorbent materials and/or the incorporation of existing materials
(polypropylene) into a suitable belt structure. The general concept
for skimming and wringing which was developed in this project is
applicable to future developments in sorbent materials with little or
no mechanical alterations.
68
-------
SECTION IX
DESIGN CONCEPT FOR BOAT MOUNTED SKIMMER
A conceptual design of a full scale floating belt oil skimmer and
support vessel has been prepared based on the results obtained in this
study. The design of the vessel was performed by R. A. Stearn Inc.
of Sturgeon Bay, Wisconsin. The following description of the vessel
is taken almost in its entirety from a report submitted to Rex Chainbelt
Inc. by R. A. Stearn Inc.
Introduction
The craft described below and illustrated in Figures 36 and 37 is
designed to accommodate a single five (5) foot wide free floating oil
belt skimmer. Based on results obtained in this study, it is antici-
pated that this skimmer could recover 5000 gallons of liquid/hour with
an oil content of 50% for #2 Fuel Oil, 2 foot waves and a slick thick-
ness of 0.10 inch. For Bunker C oil the recovery would be approximately
2500 gallons/hour of liquid containing 50% oil under similar wave
height and slick thicknesses.
This is not an attempt to prepare a final design, but is rather a
description of a concept of boat design which will best perform the
assigned functions. Some refinements will become apparent during the
preparation of the final design since certain components such as the
propulsion units have not been fully researched. The successful use
of water jet propulsion will, in part, depend upon a detailed analysis
by the jet unit manufacturer of the maneuverability characteristics at
slow speeds. Alternative types of propulsion, such as inboard-outboard
units, are feasible but felt to be more vulnerable to damage from
floating debris, and less able to cope with draft.
The dimensions (length and width) of the craft described are felt to
be the minimum necessary to provide a stable platform for the oil belt
skimmer and to obtain sea-keeping characteristics required for operation
in two-foot seas. Reduced hull depth, resulting in reduced weight and
cost, is possible for the non-cargo carrying model described as one of
the options, without loss of sea-keeping characteristics.
The cost of the unit described is estimated to be $35,000, not including
cost of the oil belt skimmer, if built in quantities of 5 to 10 units.
Larger quantities should result in savings by application of production
techniques, whereas smaller quantities will result in higher costs to
offset production plan preparation and costs incidental to pattern
preparation and low quantity purchasing.
69
-------
FLOATING OIL BELT
CONCEPTUAL DESIGN
FIGURE36-BOAT MOUNTED OIL SKIMMER,ELEVATION VIEW
I....-.J i
i i
0.51 2345
SCALE-FEET
10
-------
45 FT
OIL DEFLECTION BOOMS
COLLECTED OIL
/ TRANSFER PUMP
COLLECTED OIL STORAGE
COLLECTED OIL STORAGE
t
w
CONCEPTUAL DESIGN
FIGURE37-BOAT MOUNTED OIL SKIMMER.PLAN VIEW
O.5 I 2 3 4 5
SCALE- FEET
-------
Hull Requirements
The following approximate parameters are established to best serve the
functional requirements of the oil belt skimmer:
1. The craft is to be readable. Dimensions have been established at
12' overall beam, 36' length of hull, and 12' overall height. Weight
without fuel or crew is approximately 13,000 Ibs. The unit is
capable of being transported by trailer.
2. A catamaran type of hull is used to accommodate the 5* wide belt
which is suspended between the pontoons. The pontoons are 3' wide
each leaving a 6' width between pontoons, reduced to 5' at the belt.
3. Each pontoon contains an oval shape tank of 1,600 gallons capacity
for a total of 3,200 gallons cargo capacity.
4. Light draft is 14". Loaded draft with 3,200 gallons of oil-water
mixture at 8 Ibs per gallon is 3'-4". Depth of pontoons is 5'
leaving 20" freeboard in fully loaded condition.
5. The craft is to operate in 2* maximum height waves therefore,
sheer has been added at the forward end of the pontoons, with bow
rake and overhanging deck forward. Bulwark rake and slope have
been added to enhance the sea-keeping qualities of the craft.
6. A solid deck of raised tread plate spans between the pontoons,
except in way of the oil belt skimmer, to provide a dry, slip-
resistant deck.
7. A weathertight pilot house is installed at the bow, with doors port
and starboard, and containing propulsion and steering controls.
This location is chosen for good visibility and to obtain maximum
usable deck space.
8. A trash grid, extending from deck to keel between the pontoons,
should be installed to protect the skimmer belt from damage by
large floating debris such as logs.
9. Aluminum construction is contemplated to keep weight to a minimum.
Styrofoam, foamed in place between the pontoon plating and the cargo
tanks, up to the loaded draft, is considered a possibility to
protect the lightweight hull plating against damage.
Propulsion and Speed
1. Jet propulsion is recommended because of possible operation amid
debris which would tend to foul and damage propellers. Jet units
should be of a type which are capable of operating with control at
low speeds, and should be steerable and reversible with pilot house
control.
72
-------
2. Jet units should be capable of operating from low maneuvering speeds
for docking to full capacity of the propulsion power unit.
3. One propulsion unit is shown for each pontoon to enhance maneuver-
ability characteristics, and should enable the craft to turn in its
own length.
4. A speed of about 2^ knots is contemplated for the collection mode
of operation. A speed of 8 knots is contemplated for free running
without cargo and 5 miles per hour for free running when loaded
with cargo.
Power Unit
It is proposed that a single radiator-cooled gasoline engine rated at
85 hp be mounted on deck forward of the oil belt skimmer driving
hydraulic pumps to service all power requirements. With proper design,
this arrangement will minimize maintenance on engine equipment, shorten
controls from the pilot house for propulsion purposes, give good flexi-
bility for speed control on the propulsion jets and the oil belt skimmer,
and will make available the optimum power for propulsion in the free
travel mode and for transfer pumping to shore or barge unit. Approxi-
mate horsepowers contemplated are as follows:
1. Free travel mode - two propulsion units at 30 hp each
2. Collection mode - two propulsion units at 20 hp each, oil belt
skimmer at 7^ hp maximum, oil collection
transfer pump at 1/2 hp
3. Shore transfer mode - one self-priming centrifugal transfer
pump at 15 to 25 hp, for 500 gpm.
It is felt that with the use of a hydraulic system the oil belt skimmer
can obtain its required 5 to 1 speed variation without the need for a
mechanical variable speed unit. Propulsion units can be varied in speed
precisely to obtain optimum vessel speed for collection purposes. One
of the propulsion pumps can be utilized for driving the shore transfer
pump.
Outfit Equipment
The craft should be fitted with the following principal outfitting
equipment:
1. Mooring and anchoring - anchor, anchor line, deck cleats, fen-
ders, mooring lines
2. Personnel safety - railings and stanchions, life rings and
electrical water lights, life preservers
3. Fire protection - fire extinguishers
73
-------
4. Navigation - compass, fog bell, horn, ship-to-shore radio,
or walkie-talkie set
5. Miscellaneous - Pilot house seats, pilot house console, window
wiper, monomatic toilet (optional).
6. Electric power - engine attached alternator, shore power
connection, battery charger, 12 volt batteries, electric
power panel.
7. Lighting - general lighting including wiring, navigation lights,
flood light, spot light, yellow flashing light, fused lighting
panel.
8. Oil Recovery Equipment - shore transfer hose, 3" or 4" size.
9. Self-priming Bilge Pump - with suctions to propulsion compart-
ments and to holds
Bow Booms
Booms are shown port and starboard, hinged at the decks. Distance between
the leading edges of the booms is 15', giving a concentration factor
of 3 to the surface oil being fed to the collector belt. Buoyancy
will be installed near the leading edges of the booms to retain free--
board in waves. The lower edges of the booms will be flanged inboard
to prevent escape of oil under the booms. The booms will be raised
above the water during the free travel mode, by hand winches mounted
on the top outboard corners of the forward bulwark. The boom inner
surfaces will be of light gage aluminum, plywood, or durable fabric,
supported by a lightweight truss of aluminum tubing, and will overlap
the bow ends of the pontoons. Booms will be removable for shipping.
Surface area covered by the craft with booms extended, at a speed of
2.5 knots (15,000 feet per hour), is about 5 acres per hour.
Options
1. Craft without pontoon storage tanks - This option would provide
pontoon depth of about 3'-6". Collection of the oil would be into
rubber storage bags stowed on deck and launched off of the stern
as they are filled to be picked up by accompanying barge. A small
surge tank would be installed on the collector transfer pump to
permit uninterrupted collection during transfer of collecting bags.
With this arrangement, the oil belt skimmer, engine and transfer
pump should be moved forward to increase clear deck spaces aft.
2. Larger craft to accommodate two 5' wide belt skimmers - This concept
would result in a semi-portable unit; in other words, one that would
have to be dismantled to be trucked from one site to another.
74
-------
Two pontoon hulls, each 6* wide and from 35' to 50' in length,
would be shipped side by side on a trailer. The deck and oil belt
skimmer module would be 11' to 12" wide, would also be trailable,
and would be assembled to the pontoons with structural bolts.
Connection of the hydraulic supplies to the propulsion units would
be by quick-disconnect hydraulic hose.
An erection crane would be required to assemble the unit. It is
contemplated that assembly could be done afloat in protected waters
to avoid the necessity of elaborate launching devices.
A craft designed to this concept would have considerably greater
storage capacity and would be capable of operation in more severe
sea conditions than the smaller craft described in this report.
Trash Collector - A basket type collection unit may be suspended
between the pontoons, extending from the bows to the oil collector
unit. The collector should contain sufficient flotation to support
its own weight, should extend from the bottom on the pontoons to
6" above the loaded waterline at the sides and up to the underside
of the deck at its after end. Bow doors should be fitted, or the
trash grid previously described should be hinged at the deck to
permit basket removal.
75
-------
SECTION X
ACKNOWLEDGEMENTS
Rex Chainbelt Inc. extends its sincere appreciation and gratitude to the
following people, without whose cooperation project completion would
have been difficult if not unachievable.
Mr. Ralph Rhodes and Mr. Kurt Jakobson of the EPA Agricultural and
Marine Pollution Control Branch, each of whom served as Project
Officer during respective contract time periods. Mr. J. Stephen
Dorrler provided valuable technical input to the project.
Mr. Clifford Mortimer, Director, Center for Great Lakes Studies,
The University of Wisconsin-Milwaukee, and Mr. R. A. Stearn,
R. A. Stern Inc., Marine Architects, who served as project
consultants.
The team effort required to design, build, and test the experimental
prototype skimmer was due to attendant individual contribution of the
following Ecology Division personnel:
Mr. Arlyn Albrecht served as Project Director during the initial phases
of the contract. The materials selection studies were conducted by
Mr. Egmont Helmer while Mr. Robert Scholz was responsible for the
mechanical design and much of the concept development. Messrs. John
Pernusch, Mahendra Gupta, Charles Hansen, Donald Murray, Ronald Holasek
and Kenneth Witter provided valuable assistance during the design and
data evaluation phases, with Mark Scholz and Timothy Riesing operating
the prototype skimmer during the test phase.
Dr. Robert W. Agnew served as the Project Director during the final
twelve months of the project and was the principal author of this report.
The support of the project by the Environmental Protection Agency and
the willing assistance and helpful advice of EPA personnel is gratefully
acknowledged.
77
-------
SECTION XI
REFERENCES
1. Swift, W. H., £t al, "Oil Spillage Prevention, Control and
Restoration - State-of-the-Art and Research Needs", Journal,
Water Pollution Control Federation. 41, p. 392, 1964.
2. "Combating Pollution Created by Oil Spills - Volume I: Methods",
Arthur D. Little, Inc., June 30, 1969.
3. O'Sullivan, A. J. and Richardson, A. J., "The Torrey Canyon Disaster
and Inter-Tidal Marine Life," Nature. 214, p. 448, 1967.
4. "Oil Spill Technology Makes Strides", Environmental Science and
Technology. 5, 8, p. 674, 1971.
5. "Study of Equipment and Methods for Removing Oil From Harbor Water",
Battelle Memorial Institute, August 25, 1969, Clearinghouse Report
AD 696980.
6. Barridge, S.A.S et al, "The Properties of Persistent Oils at Sea",
J. Inst. Petrol.. 549 539, November, 1968.
7.. Blocker, P. C., "Spreading and Evaporation of Petroleum Products on
Water", paper presented to the fourth International Harbor Conference,
Antwarp, June 22-27, 1964.
8. Brockis, G. J., Comments in discussion of paper, reference (6).
9. Torrey Canyon Pollution and Marine Life0 edited by J. E. Smith,
University Printing House, Cambridge, England, 1968.
10. Hughes, P., "A Determination of the Relation Between Wind and Sea
Surface Drift", Quart. J. Royal Meteorol. Soc., 82, p 494-502, 1956.
11. "Notes on Industry's Oil Spill Control Activities", Ocean Industry.
p. 46-60, June, 1970.
12. "Using Chemicals for Cleaning Up Oil Spills", Ocean Industry.
p. 35-42, August, 1970.
13. Proceedings, First Joint Conference on Prevention and Control of
Oil Spills, sponsored by API and EPA, New York, December 15-17, 1969.
14. Proceedings, Second Joint Conference on Prevention and Control of
Oil Spills, sponsored by API and EPA, Washington, D.C.S June 15-17,
1971.
79
-------
15. "Study of Equipment and Methods for Removing or Dispersing Oil
From Open Waters", Battelle Memorial Institute, August, 1970,
Clearinghouse Report AD 716-792.
16. "Corralling Oil on the High Seas", Chemical Engineering, 78, 18,
August, 1971.
17. Technical Bulletin, "An Introduction to the JBF Family of
Skimmers", JBF Scientific Corporation, Burlington,, Massachusetts,
1971.
18. Schatzberg, Paul and Nagy, K.V., "Sorbents for Oil Spill Removal",
Proceedings of Joint Conference on Prevention and Control of Oil
Spills, Washington, D.C., June, 1971.
80
-------
SECTION XI
APPENDICES
Page No.
A. RESULTS OF THE BENCH SCALE OIL SKIMMER 82
Selection of Belt Materials 82
Preparation of Test Belts 84
Belt Scale Apparatus 85
Test Procedures 85
Tests at Constant Oil Level with Various Foams 87
Tests at Varying Belt Speeds and Diminishing Oil
Layer Thickness 87
High Viscosity Oils 87
Conclusions 87
B. COMPUTER PROGRAM 91
Table II-l - Program Listing Petski 91
Table II-2 - Definition of Variables 93
81
-------
APPENDIX A
BENCH SCALE SKIMMER INVESTIGATIONS
Preliminary studies were conducted to select and evaluate various sor-
bent materials for use on the oil skimmer. In addition, a bench scale
model was constructed and operated to determine the basic feasibility
of the proposed concept and assist in developing mechanical design
criteria. The basic methodology and results obtained in these studies
are presented in this section.
Selection of Belt Materials
Trade directories as well as firms with whom Rex Chainbelt Inc. has
previous contacts were contacted to obtain information regarding mater-
ials that were potentially suitable for oil belt skimmers. Thirty-three
suppliers of various types of sorbent materials were contacted. Of
those responding less than half had materials which were judged to be
suitable for oil recovery. Certain sorbent materials which appeared
to be quite promising were unavailable for evaluation because of the
proprietary interests of the suppliers.
Materials which were available and which were evaluated include polyure-
thane foams and various types of felts.
Samples of polyurethane foam were obtained from two sources for
evaluation:
1. Scott Paper Company, Foam Division, Eddystone, Pennsylvania
2. Foam Rubber Products, Milwaukee, Wisconsin
Samples of felts were obtained from:
1. Synthetic Fiber Felts, GAF Corporation Industrial Products
Division, Greenwich, Connecticut
2. Natural Fiber Felts, Western Felt Works, Chicago, Illinois
The samples of foam and felts obtained were subjected to screening tests
to determine which of the various densities and porosities available
appeared to be most promising for a skimmer belt.
The following test procedure was used to evaluate the various foams and
felts:
1. The foam or felt material was cut to predetermined dimensions
and volume
82
-------
2. The test material was weighed.
3. The test material was then dipped into a mixture of oil-
Milwaukee tap water for three seconds. This mixture consisted
of tap water at room temperature topped by a 1/2" layer of oil.
4. The test material was then drained until the frequency of
dripping was greater than once every 10 seconds and weighed.
5. The liquid (oil and water) was then squeezed out of the
material and collected.
6. The test material was then weighed
7. The oil and water content of the liquid was determined by
gravity separation.
8. The oil recovery was then calculated from the difference in
weight before and after squeezing the liquid from the test
material. This recovery was then related to the volume of
the test material.
Two oils were used in these tests:
1. Light oil, viscosity - 60 SUS/100°F
2. Heavy Oil, viscosity - 3100 SUS/100°F
The test results ruled out the felts as potential materials for skimmer
belts. The felt materials absorbed some oil but were not sufficiently
resilient to release of oil under pressure. As a result the felt mater-
ials were considered unsatisfactory for use as a skimmer belt.
The results of the laboratory screening tests on the polyurethane foams
indicated that the recoveries of 60 SUS/100°F oil + water ranged from
1.32 gal/cu ft up to 4.38 gal/cu ft and that of 3100 SUS/1000°F oil
+ water from 1.72 to 5.06 gal/cu ft. The heavy oil water mixtures
contained from 11 to 19% water. In addition, water was contained in the
oil in the form of an emulsion and/or a solution. It was estimated that
the concentration of water in emulsion and/or solution was less than 1%.
It was evident that foam with large pores corresponding to a pore rate
of 30 pores or less is incapable of holding oil to a satisfactory degree
independent of oil viscosity. There was a tendency for increased capac-
ity to hold oil with increased pore rates (smaller pores) which was
particularly pronounced with the high viscosity oil.
Up to 69% of the pores were filled with liquid in the tests with heavy
oil and up to 60% on those with light viscosity oil. The amount of
oil/cu ft in the foam was practically independent of the thickness of
foam in the thickness range of 1/2 to 2 inches investigated.
83
-------
The conclusions derived from these screening tests were:
1. Polyurethane foams are feasible for use as belt materials.
2. Small pore foams are expected to yield higher recovery rates
than large pore foams.
3. It is advantageous to use as thick a foam as is compatible with
the mechanical functioning and desired service life of the belt.
Based on these conclusions it was decided to test 1 inch and 2 inch foam
thicknesses and foams with pore rates of 45 to 80 ppi. It was also
decided to test an ester base foam (Scott Industrial Foam brand) against
an ether base foam (Foam Rubber Products) because the former is supposed
to provide higher resistance against hydrolyzation by sea water.
Preparation of Test Belts
Since the polyurethane foam is mechanically too weak to be used as such
in a belt, it was necesaary to equip the foam with a supporting backing
of a higher strength material. The backing materials selected were:
1. Neoprene
2. Polypropylene Filter Cloth
The neoprene backing was adhesive bonded by Stephenson and Lawyer in
Grand Rapids, Michigan with an oil resistent adhesive of their selection.
The adhesive used by Stephenson and Lawyer was not disclosed. An experi-
mental adhesive (*) made by the Hughson Chemical Company in Erie,
Pennsylvania produced a bond between polyurethane foam and neoprene back-
ing which was of satisfactory strength and resistant to petroleum oils
and water.
The polypropylene filter cloth was Mero Fabric No. ICH-27, construction
60 x 32, weight 8.2 oz., supplied by Mero and Company, Inc., Chicago,
Illinois. The foam was sewn to the filter cloth. Sewing reduced the
effective width and cross section of the foam to about 3.9 in2 from
6 in2 for 1" thick foam of 6" width.
(*) Adhesive TS 1966-40A, Catalyst CS9988
84
-------
One belt of each of the following design and materials were tested:
Designation
Upholstery
Upholstery
Scott Indus.
Scott Indus.
Scott Indus.
Kind
Polyether
Polyether
Polyester
Polyester
Polyester
Thickn.
in.
1
1
1
2
2
Poro-
sity
PPJ
80-100
80-100
80
80
45
Backing
Material
Effective Cross
Section of
6" Wide Belt
polypropylene
Neoprene
Neoprene
Neoprene
Neoprene
3.88 in2
6 in2
6 in2,
12 in,
12 in
Bench Scale Apparatus
The bench scale oil belt skimmer used in the tests is shown in Figure 1-1.
Its main features are a pair of wringer rolls located above each other
and between them the skimmer belt. The lower roll was connected to the
driving mechanism. The upper roll was loaded by a dead weight. The
liquid wrung out of the belt collected in a container located below the
wringer roll. From this container the liquid is discharged through a pipe
either into the tank of the machine or outside the tank. The belt hung
freely in the tank in a wide loop with the foam covered side toward the
water. The wide loop results from a 3rd roll located parallel to the
wringer in a horizontal plane some distance away from the roller. Varia-
ble speed control, speed indicator, a water level control (overflow weir)
and an oil supply pump with hand adjustable oil supply valve completed
the arrangement.
Test Procedures
All but one test in this phase of investigation were run with Mobil
Naprex oil (#920) having a viscosity of 60 SUS @ 100°F and specific
gravity of 0.862 @ 60°F. The oil layer thickness (1/4") was controlled
only at a belt speed of 26 fpm because of the operational difficulties
in controlling the oil layer at higher belt speeds. However, for
higher belt speed and thicker oil layer tests, a modified procedure was
utilized. A known amount of oil was placed in the tank (corresponding
to a given known oil layer thickness) prior to an experiment. No
additional oil was added to replenish that removed. The oil was removed
from the water by the skimmer belt resulting in a gradually and progres-
sively decreasing oil layer thickness. The quantity of oil removed was
measured at predetermined time intervals and plotted against time of
operation resulting in curves from which the recovery rate change with
diminishing oil layer thickness could be obtained by differentiation.
Samples of water-oil mixture were collected in a weighed 5-gallon
container at predetermined time intervals and after reweighing, the
contents were transferred into a transparent plastic separatory funnel.
A separation time of 1/2 - 1 hour was allowed and the separated oil was
measured after draining out the water.
85
-------
\
FIGURE I-l
OVERALL VIEW OF BENCH SCALE OIL SKIMMER
86
-------
Tests at Constant Oil Level with Various Foams
A summary comparison of all the tests at the identical test conditions
of 26 fpm belt speed and 1/4 inch oil layer thickness is shown in
Table 1-1. The results indicated that the recovery rate of light
(60 SUS 100°F) oil was highest with the 2 inch thick Scott Industrial
foam at 80 ppi porosity and lowest with the one inch thick upholstery
foam sewn to a polypropylene filter cloth backing. Oil recovery rate
was proportional to foam thickness at the identical porosity of 80 ppi
but water content in the recovered mixture increased almost 5 times
with double the foam thickness. The recovery rate for light oil was
reduced by larger pore size (45 ppi) and the rate of water absorption
in the recovered liquid was 3 times that of the smaller pore size
(80 ppi) foam.
Tests at Varying Belt Speeds and Diminishing Oil Layer Thickness
The effect of belt speed was tested in a series of tests conducted with
a one inch thick upholstery foam at an initial oil layer thickness of
1.25 inches. The oil layer diminished gradually as the test progressed
and oil was removed from the surface of the water. The results of
these tests are summarized in Figure 1-2.
Basically these tests indicate that the oil pickup rate decreases
logarithmically with diminishing slick thicknesses for a given belt
speed. In addition it was found that the water content of the recover-
ed oil-water mixture generally increased with decreasing oil slick
thickness.
High Viscosity Oils
Attempts were made to evaluate the pickup rate of //6 Fuel Oil - Bunker C
by means of the bench scale skimmer. Many problems were encountered
but the primary one was the difficulty in getting the highly viscous
oil to flow to the skimmer so that it could be picked up. A very
minimal amount of data indicated two basic conclusions:
1. That wringer rollers cannot adequately remove Bunker C from
low porosity (80 ppi) polyurethane belts.
2. That the rate of oil pickup is substantially less for Bunker
C than for low viscosity oils.
Conclusions
The following conclusions were drawn based on the bench scale testing
phase of this project and were used as the basis for material selection
and design criteria for the experimental prototype evaluation.
87
-------
CX3
00
Test //
45-48
53-56
58-61
63-66
68-72
Kind of Belt
Upholstery Foam
Polypropylene Filter Cloth
Backing 4.5" Effective
Width; 3.88 in2 cross sec.
Upholstery Foam, Neoprene
Backing
Scott Indust. Foam
Neopreme Backing
Scott Indust. Foam
Neoprene Backing
Scott Indust. Foam
Neoprene Backing
Foam
Thickness,
Inches
1
1
1
2
2
Foam
Properties
(ppi)
80-100
80-100
80
80
45
Average
Water
Volume
f/\
\'»)
40.8
18.1
5.2
Water
Gal/Hr.
34.3*
48.6
50.8
24.9 100.7
75.6
77.7
* Corrected to 6 in cross section: 6.00
3.88
1.55
Remarks: Polypropylene backed belt 115" long, 2.71 cycles/min at 26 fpm;
Neoprene backed belt 100" long, 3.12 cycles/min. at 26 fpm;
All belts in contact with liquid over 20 inches of belt length.
C/3
c
w o
«^g
n> o *~o
M o>
rt ^ 5&
WOW
•a H- o
n> M 23
n> -
a- o
O *rJ
ro H-
t-h r1 r1
1 ^ H
(D ^
^N H K
OJ ?d
^•3-0
r-o H- g
o ^
*^ 3 SS
O (B O
M CO M
(D CO
CO I
3*^1 w
H- -P- S
3 = §
n
o
3
CO
&
Pi
-------
o
5100
o
-------
1. A 2 inch thick Scott Industrial foam of 80 ppi porosity
produced the highest oil recovery rates at all belt speeds
and all oil layer thicknesses.
2. Oil recovery rates are proportional to the cross sectional
area of the foam.
3. Foams with small pore, such as 80 ppi, are not suitable for
the recovery of very high viscosity oil. It is likely that
an optimum foam porosity exists for each oil viscosity.
4. Oil recovery rates increase with increasing belt speed.
The information obtained was insufficient to indicate conclu-
sively whether an optimum speed exists.
5. Oil recovery rates decrease with decreasing oil layer
thickness and water content of the recovered liquid increases
simultaneously.
6. The upholstery foam with a permeable polypropylene backing
recovered a higher percent of water than upholstery foam of
the same thickness with a neoprene backing. No advantage of
using the permeable polypropylene backing can be seen in
recovering light oil. In recovering heavy oil the relatively
light weight of the backing may be an advantage in helping
to make the belt float.
7. The polyether based upholstery foams pick up a higher percen-
tage of water than the polyester based Scott Industrial foam.
The polyester based foam should be used in future tests.
8. The bench scale tests have, in spite of the experimental
difficulties, provided much insight into the potential
problems of oil belt skimmers.
90
-------
APPENDIX B - COMPUTER PROGRAM PETSKI
TABLE II-l
PROGRAM LISTING PETSKI
10:RUN NUMBER ////////
20: AMOUNT OF OIL SPREAD //////.//// GAL INITIAL OIL DEPTH
//.//////// INCHES
30 READ R,T8,T9,D1,D2,D3,D4
40 V5=(D1-D2)*13.1*12
50 D6=12*V5/(364.4*7.48)
60 PRINT USING 10, R
70 PRINT USING 20,V5,D6
80 PRINT
90 PRINT
100 PRINT"TIME VOLUME VOLUME PERCENT COLLECTION SAMPLE
CUMULATIVE
110 PRINT" INTO OF OF OIL RATE MIDPOINT
OIL PCT
120 PRINT"RUN SAMPLE OIL MIX OIL
VOL REC
130 PRINT" (MIN.) (GAL) (GAL) (PCT) (GPM) (GPM) (MIN)
(GAL) (PCT)
140 PRINT
ISO://////.// ////.//// ////.//// //////.//// //////.///' ////.//// ////////.// //////.//
160 T7=0
170 V4=0
175 F3=0
180 V8=0
190 READ Tl
200 IF Tl<0 THEN 390
210 READ T2,T3,V1,V2
220 T4=T1+T2/60
230 T5=T4+T3/60
240 T6=(T4+T5)/2
250 F1=V1/(T3/60)
260 IF V2>5 THEN 280
270 GO TO 290
280 V2=V2/3785
290 F2=V2/(T3/60)
91
-------
TABLE II-1 CONT.
300 P1=(V2/V1)*100
310 V4=V4+(((F2+F3)/2)*(T6-T7))
320 D5=12*(V5-V4)7(7.48*364.4)
330 F3=F2
340 T7=T6
350 P2=(V4/V5)*100
360 V8=V1+V8
370 PRINT USING 150,T4,V1,V2,P1,F1,F2,T6,V4,P2,D5
380 GO TO 190
390 V9=((D4-D3)*16.1*12)+V8
400 PRINT
410 PRINT USING 420,V9
420:TOTAL MIXTURE COLLECTED #////#.// GAL'
421 PRINT
422 P3=100*V4/V9
423 A1=V9/(T8+T9/60)
424 A2=V4/(T8+T9/60)
425 PRINT USING 426,P3
426:PERCENT OIL IN THE MIXTURE //##.## PERCENT
427 PRINT
428 PRINT USING 429,Al
429:AVERAGE COLLECTION RATES: MIXTURE ##//#.« GPM
430: OIL «#//.« GPM
431 PRINT USING 430,A2
432 PRINT
433 PRINT
434 PRINT
435 PRINT
436 GO TO 30
999 END
92
-------
TABLE II-2
DEFINITION OF VARIABLES
Variable
Al
A2
Dl
D2
D3
D4
D5
D6
Fl
F2
F3
PI
P2
P3
R
Tl
T2
T3
T4
T5
T6
T7
T8
T9
VI
V2
V4
V5
V8
V9
Average oil-water mixture collection rate
Average oil collection rate
Initial oil depth in the supply tank .
Final oil depth in the supply tank
Initial mixture depth in the collection tank
Final mixture depth in the collection tank
Equivalent oil depth on the water surface
Initial oil depth on the water surface
Mixture collection rate during sampling
Oil collection rate during sampling
Internal variable (F3=F2)
Percent oil in sample
Percentage of total oil collected
Percent oil in the total mixture collected
Run number
Time (integer minutes) at the start of sample
collection
Time (remaining seconds) at the start of sample
collection
Sample collection time
Time at start of sample collection
Time at end of sample collection
Time at midpoint of sample collection
Internal variable (T7=T6) . >
Time (integer minutes) at the end of the run
Time (remaining seconds) at the end of the run
Total volume of the sample
Volume of oil in the sample
Cumulative estimate of the volume of oil collected
Volume of oil spread on the surface
Cumulative volume of samples
Total volume of oil-water mixture
Units
gal/min
gal/min
ft
ft
ft
ft
in
in
gal/min
gal/min
gal/min
percent
percent
percent
min
sec
sec
min
min
min
min
min
sec
gal
gal or ml
gal
gal
gal
gal
TO SIGNIFY THE END OF A DATA SET ENTER A -1 AS THE LAST PIECE OF DATA
FOR THAT RUN
ftU.S. GOVERNMENT PRINTING OFFICE:1972 514-147/49 1-3
93
-------
1
5
Xcresxion Number
2
Subject Field &. Group
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
PRY rHATNRFT.T IMP. MTT UATTIfFF WTCrOWQTM
ECOLOGY DIVISION
Title
A FREE FLOATING ENDLESS BELT OIL SKIMMER
1Q Authors)
AGNEW, ROBERT W.
i JL Project Designation
EPA CONTRACT 14-12-908 PROJECT 15080 GBJ
21 "
22
Citation
Environmental Protection Agency report
number EPA-R2-72-006, August 1972
oo Descriptors (Starred First)
25
Identifiers (Starred First)
27 Abstract
A free floating endless belt oil skimmer was developed as a means of recovering
spilled oil from surface waters. The skimmer utilizes a unique high efficiency conveyor
wringer to power and wring the belt. The belt is designed to float on the water surface
and responds rapidly to the shape of the waves, thereby maximizing oil-sorbent contact
time. Evaluation of the skimmer was conducted in a 60 foot diameter annular test tank
under the conditions of slightly progressive waves having an amplitude of two feet.
One foot wide neoprene backed polyurethane foams were utilized as the sorbent material.
The experimental results indicate that the oil pickup rates will vary with the
belt speed, oil slick thickness and belt porosities. Oil pickup rates of 8.3 and
3.7 gpm per foot of belt width were attained for i?2 Fuel Oil and Bunker C oil respec-
tively at a slick thickness of 0.10 inches. The recovered liquid contained approxi-
mately 50-70% oil at 0.10 inch slick thickness.
A conceptual design of a five foot wide boat mounted skimmer capable of
harvesting approximately 5 acres per hour of spilled oil is presented.
This report was submitted in fulfillment of Project Number 15080 GBJ,
Contract 14-12-908, under the sponsorship of the Office of Research and Monitoring,
Environmental Protection Agency.
(AGNEW, REX CHAINBELT INC.)
Abstractor
Robert W, Agnew
Institution
Rex Chainbelt Inc., Ecology Division, Milwaukee, Wisconsin
WR;102 (REV. JULY 1869)
WR5I C
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 20240
------- |