WATER POLLUTION CONTROL RESEARCH SERIES 12050 DRC 11/71
Experimental Evaluation of
Fibrous Bed Coalescers for
Separating Oil-Water Emulsions
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes
the results and progress in the control and abatement
of pollution in our Nation's waters. They provide a
central source of information on the research, develop-
ment , and demonstration activities in the Environmental
Protection Agency through inhouse research and grants and
contracts with Federal, State, and local agencies} research
institutions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Head, Publications Branch
(Water), Research Information Division, R&M, Environmental
Protection Agency, Washington, B.C. 20460
-------
EXPERIMENTAL EVALUATION OF FIBROUS BED OOALESCERS
FOR SEPARATING OIL-WATER EMULSIONS
Illinois Institute of Technology
Department of Chemical Engineering
Chicago, Illinois 60616
for the
ENVIRONMENTAL PROTECTION AGENCY
INDUSTRIAL POLLUTION CONTROL SECTION
Grant No. 12050 DRC
November 1971
For sale by the Superintendent of Documents, U.S. Government Printing Otlice, Washington, D.C. 20402 - Price $1.75
-------
EPA Review Notice
This report has been reviewed by the Environmental Protection
Agency and approved for publication. Approval does not signify that
the contents necessarily reflect the views and policies of the
Environmental Protection Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
11
-------
ABSTRACT
A 1 sq. ft. coalescer unit using filter press construction has
been designed for removing dispersed oil from water and tested on
both a synthetic stream and on an actual pollutant stream. The oil
removal efficiency was essentially 100$ at a superficial velocity of 1
fpm. The pressure drop increased from 3 to 25 psi over run times which
varied from 13 hrs. to 294 hrs., due to both accumulation of oil in
the bed and mechanical degradation of the fibers. Preliminary tests
indicated that the bed degradation phenomenon could be eliminated by
structurally stabilizing the compressed fibers with methacrylate resin.
Such fiber beds could be regenerated by various solvent treatments and
reused.
The performance of fiber glass coalescers was studied in depth using
a cell with an active area of 1.77 sq. in. The commercial fibers, with
phenol formaldehyde coatings and a fiber diameter of 3-2jj, gave
efficiencies of 90-99$ with bed densities of 12 lb./ft.3 when operating
at superficial velocities from 0.2 to 4 fpm on emulsions containing
50-500 ppm of oil.
The present design is suitable for large scale operation by the
use of both multiple cells and larger individual cells. It is estimated
that operating costs on the order of 0.13 $/!03 gal. are involved for
the worst case of single use of fiber. If the fibers can be regenerated
c - ' zing costs would be reduced to 0.01 $/!03 gal.
This report was submitted in fulfillment of Grant No. 12050 DRC
between the Environmental Protection Agency and the Illinois Institute
of Technology.
111
-------
CONTENTS
Section Page
I Conclusions 1
II Recommendations 3
III Description of Large Coalescer 5
IV Results of Large Coalescer Cell ?^
V Economics of Filter/Coalescer Operation 39
VI Description of Small Coalescer 43
VII Results of Small Coalescer Tests 59
VIII Microscopic Study of Fibrous Bed Coalescer 81
IX Stabilization Tests 85
X Acknowledgements 91
XI References 93
XII Glossary 95
XIII Symbols 97
v
-------
FIGURES
PAGE
1 GENERAL SCHEMATIC OF THE PLASTIC COALESCER CELL "
2 FINAL COALESCER UNIT 10
3 COMPLETE CELL IN HORIZONTAL POSITION WITH H
INFLUENT SIDE ON TOP
4 INFLUENT END PLATE PARTIALLY REMOVED, SHOWING 11
TOP SEAL FRAMES
5 ENTRANCE SIDE TO DIAPHRAGM ASSEMBLY I"1
6 SPECIAL INNER SEAL FRAME 11
7 OIL EFFLUENT FRAME OVER TFE SCREEN SECOND
STAGE 12
8 DISASSEMBLY VIEW "12
9 ASSEMBLY VIEW ^
10 TOP PERFORATED PLATE, BOTTOM ASSEMBLED DIAPHRAGM
WITH TOP GASKET IN PLACE
12
11 TOP GASKET REMOVED, SHOWING USED DIAPHRAGM.
SQUARE LINE AT EDGE IS METAL 0-RING SEAL WHICH
HAS CUT THROUGH FIBER BLANKETS 13
12 METAL 0-RING SEAL AFTER CUT EDGES HAVE BEEN
REMOVED 13
13 BOTTOM VIEW: SHOWS BOTTOM GASKET WITH GROOVE
FROM METAL 0-RING AND SPACER PLATE, ALL
RESTING ON BOTTOM PERFORATED PLATE 13
14 VIEW OF TREAD PLATE AND FORMED GASKET ON
PERFORATED PLATE 13
15 1/2 IN. OF 3.2y GLASS, 2 IN. OF 10.ly GLASS WITH
METAL 0-RING SEAL ON BOTTOM PERFORATED PLATE
AND GASKET. 1/8 IN. SPACER PLATE IN POSITION.
NOTE 1/2 IN. AND 5/16 IN. ALIGNMENT PINS 14
16 COMPLETE 5 INCHES OF FIBER GLASS 14
.VI
-------
FIGURES (CONTINUED)
PAGE
17 UPPER GASKET IN PLACE 14
18 UPPER PERFORATED PLATE IN PLACE 14
19 COMPLETE ASSEMBLY BUT NOT BOLTED TOGETHER 15
20 TFE SCREEN BEING INSERTED ABOVE WATER EFFLUENT
SPACER FRAME 15
21 OIL EFFLUENT SPACER FRAME 15
22 EFFLUENT SIDE OF DIAPHRAGM ALL ASSEMBLED 15
23 DIAPHRAGM THRU EFFLUENT SIDE ASSEMBLY 16
24 INFLUENT SPACER FRAME 16
25 INFLUENT END PLATE 16
26 COMPLETE ASSEMBLY 16
27A FIBROUS BED ASSEMBLY 18
27B SCREEN DETAILS 18
28 TEST ASSEMBLY 20
29 PHOTOGRAPH OF TEST ASSEMBLY 21
30 SUMMARY OF PERFORMANCE CHARACTERISTICS OF
1 SQ. FT. FILTER/COALESCER (SELECTED POINTS) 36
31 SCHEMATIC DIAGRAM OF RECIRCULATING SYSTEM 44
32 SCHEMATIC DIAGRAM OF ONCE-THROUGH SYSTEM 46
33 PHOTOGRAPH OF ONCE-THROUGH SYSTEM 47
34 PHOTOGRAPH OF INLET SECTION ASSEMBLY OF ONCE-
THROUGH SYSTEM 47
35 SMALL SCALE COALESCER CELL WITH FIBROUS BED
TEST SECTION 4?
36 PHOTOGRAPH OF SMALL SCALE COALESCER CELL 50
37 LIGHT TRANSMISSION APPARATUS b4
38 CALIBRATION OF LIGHT TRANSMISSION APPARATUS 56
39 PRESSURE DROP VS SUPERFICIAL VELOCITY FOR
SINGLE PHASE FLOW (VOID FRACTION 0.94) 62
vii
-------
FIGURES (CONTINUED) PAGE
40 PRESSURE DROP VS SUPERFICIAL VELOCITY FOR
SINGLE PHASE FLOW (VOID FRACTION 0.90) 6b
41 TIME-DEPENDENT BEHAVIOR OF PRESSURE DROP AT
DIFFERENT SUPERFICIAL VELOCITIES FOR BAKE-
LITE COATED FIBER GLASS FIBERS 69
42 COMPARISON OF TIME-DEPENDENT BEHAVIOR OF
PRESSURE DROP FOR COATED FIBER GLASS AND
UNCOATED GLASS FIBERS 70
43 TIME-DEPENDENT BEHAVIOR OF PRESSURE DROP AT
DIFFERENT VELOCITIES FOR 1/1 ILS OIL/KEROSENE
DISPERSED IN WATER 7'
44 COMPARISON OF TIME-DEPENDENT BEHAVIOR OF
PRESSURE DROP FOR ONCE-THROUGH AND
RECIRCULATING SYSTEMS /8
45 APPARATUS FOR HIGH SPEED CINEPHOTOMICROGRAPHY 8^
46 MICROSCOPIC PICTURES OF OIL DROPS COALESCING
IN A FIBROUS BED b-
47 ASSEMBLY FOR STABILIZATION TESTS 86
Vlli
-------
TABLES
No. Page
1 Large Scale Coalescer Tests with Double
Layer 3.2y Glass ... 25
2 Large Scale Coalescer Test. Standard Run
at 70°F 27
3 Large Scale Coalescer Tests with Samples from
Interlake Steel Corp. 28
4 Large Scale Coalescer Tests with Samples from
Interlake Steel Corp. 29
5 Large Scale Coalescer Tests - One Fourth
Standard Flow Rate 30
6 Large Scale Coalescer Tests - At Low Temperature 32
7 Large Scale Coalescer Tests - At Standard
Conditions 33
8 Summary of Test Results with Aluminum Cell 35
9 Physical Properties of the Fibers 51
10 Physical Properties of the Materials Used 53
11 Calibration of Light Transmission Apparatus 57
12 Single Phase Flow Through Coated Fiber
Glass Fibers 60
13 Single Phase Flow Through Coated Fiber
Glass Fibers 61
14 Single Phase Flow Through Uncoated Glass
Fibers 63
15 Single Phase Flow Through Uncoated Glass
Fibers 64
16 Preliminary Investigation of Coalescer
Performance 55
17 Coalescer Performance With and Without
Regeneration for Coated Fiber Glass Fibers 72
18 Coalescer Performance With Uncoated Glass
Fibers and Uncoated Glass Fibers Treated
With Various Hydrophobic Coatings 73
ix
-------
N°- Page
19 Time Versus Pressure Drop for Coated Fiber
Glass Fibers and Uncoated Glass Fibers 76.
20 Coalescer Performance Data for Recirculating
System 79
21 Coalescer Performance Data for Once-Through
System 80
22 Regeneration Tests on Stabilized Fibers 87
-------
SECTION I
CONCLUSIONS
1. A 1 sq. ft. coalescer cell using filter press construc-
tion has been designed and tested on a synthetic oil disper-
sion and on an actual pollutant stream. It was found to give
essentially 100% coalescence efficiency at a flow rate of
7.5 gpm (1 fpm velocity) over the entire run time. The con-
trolling parameter for a practical unit appears to be the life
of the bed. This is arbitrarily defined in the present work
as the through-put up to the point where the pressure drop
reaches 25 psi.
2. The present coalescer design may be used as is and readily
scaled up either by using multiple cells or by increasing the
area of individual cells. The capacity, corresponding to a
pressure drop of 25 psi, for a flow rate of 7.5 gpm was found
to be 710 gal. for a stream at 85°F and 100 gal. for a stream
at 65°F. These capacities are low but they do not preclude
practical operations since it is a simple matter to install
new fiber beds.
3. Stabilization of the fibers with regards to the mechani-
cal degradation was accomplished by coating the compressed
fibers with isobutyl methacrylate resin. This medium, which
showed the same coalescence efficiency and pressure drop
characteristics of the original phenol formaldehyde coated
fibers, could be regenerated by solvent treatment.
4. Operating costs on the order of 0.13 $/10 gal. are
involved for the single use of fibers. If the fiber can be
regenerated they would drop to 0.01 $/10 gal. or lower.
5. The factors controlling the coalescence performance of
fiber glass media on secondary emulsions of oil in water have
been evaluated in a small scale unit with an area of 1.77 sq.in.
6. Small scale tests have shown that commercially avail-
able fibers, 3.2y diameter and coated with phenol formalde-
hyde resin which are an effective coalescing medium for
secondary oil emulsions, have pressure drop characteristics
1
-------
which increase continually with time. This increase is
probably due both to oil accumulation in the fibers, a
reversible phenomenon, and mechanical degradation of the
fibers, an irreversible phenomenon.
-------
SECTION II
RECOMMENDATIONS
It is recommended that a demonstration unit be con-
structed using the final design developed here and evaluated
on a selected industrial pollutant stream. It is desirable
to use a unit of 25-100 gpm, 1 to 4 ft. active area, which
will operate on some percentage like 10% of the total stream,
to facilitate the necessary experimental changes.
It is recommended that the important parameters which
control the coalescence efficiency be further investigated.
Probably the most important parameter involves stabilization
and regeneration of the fiber medium. In addition the effect
of such parameters as the particle-size distribution, zeta
potential, media pore size distribution and wettability
should be quantitatively related to the coalescence per-
formance. A knowledge of these relationships is essential
to minimize upsets which will occur due to the variable
nature of any pollutant.
-------
SECTION III
DESCRIPTION OF LARGE COALESCER
Introduction
The object of the present program was to design and
breadboard on operational coalescer with 1 sq. ft. of active
area using filter press construction. Overall design speci-
fications were 1 fpm superficial velocity operating on an
influent containing 100 ppm of oil in the form of a secondary
emulsion. A practical definition of an emulsion as used in
these large scale tests is one formed by passage thru a pump
operating at 3450 RPM. The influent oil would be diluted
with at least 50% kerosene. The other 50% influent oil
being a representative pollutant material obtained from the
hot mill cooling water treatment system at Interlake Steel
Corporation. No. 30 lube oil was found to give comparable
results to the actual pollutant oil and was used in many
of the tests.
The coalescer bed was formed from standard, phenol
formaldehyde coated fiber glass and consisted of the three
layers: a) 2 inches of 10. ly glass, 0.6 Ib./ft. density,
b) 1/2, 1 or 1 1/2 in. of 3.2y glass, 0.6 Ib./ft.3 and
c) 2 inches of 10.ly glass 0.6 Ib./ft. . These fibers were
Owens Corning "Aerocor", 10.ly and "FM-004", 3.2y . The
bed, usually with 1/2 in. of 3.2y glass, was compressed
between screens to an overall thickness of 1/4 in. giving
an apparent overall density to the working membrane of
10.8 Ib./ft.3.
The above design specifications were based on the
correlations developed in the first phase work on this
project ( 1 ) and other work carried out at the Illinois
Institute of Technology (2-15 ). The design character-
istics were developed during this second phase by tests in
a small scale cell of 1.5 in. dia. which are given in
Section VI of this report.
-------
The structural design for large cells was developed in
two stages. In the first stage a plastic cell was used to
develop an operational model. This plastic construction
permitted ideas to be tried out at will with a minimum of
effort, expense and machining. While the filter press
principal is a very old art, it has not previously been
employed in filter/coalescer applications and considerable
developmental effort was required to develop a working design,
The second stage involved the construction of a metal unit
which, with multiple cells, could be used in small scale
demonstration unit at rates on the order of 25 to 50 gpm.
General Schematic Cell Description
The cells employed square frames, 16 in. outside square
(O.S.) by 12 in. inside square (I.S.). The overall cell
consisted of a front (l) and rear (n) end plate, an entrance
frame (2), diffuser plates @ and (T^fthe coalescer bed (§) ,
some type of support section (4) and \§), oil disengaging
frame @, an oil separating teflon screen (?), and the
effluent frame (Q) . The particular design employed in both
stages was selected to allow ease in modifications. Optimum
filter press design was not considered in this program. The
final design employs a unit cell length (without end plates)
of six inches which could be reduced to 2 inches by in-
creasing the structural complexity.
Preliminary Plastic Coalescer Unit
The plastic cell used in the preliminary tests is shown
in Figure 1 . The cell frames (except for that containing
the fiber bed), 16 in., outside square, by 12 inches, inside
square, were formed by laminating strips of methyl metha-
crylate plastic,14 in. by 2 in. by 1/4 in., to the desired
thickness. Laminating lacquer consists of methyl metha-
crylate monomer with 5% each of benzoyl peroxide catalyst
and paramethyl toluidine accelerators, heated to 80°C for
approximately 3 minutes so as to form a viscous liquid. The
-------
End Plate
Entrance Frame
Diffuser Plate
Screen Support
Coalescer Bed
Screen Support
Diffuser Plate
Oil Disengaging Frame
100 Mesh Monel Screen Coated with
0.005 In. Teflon
Effluent Frame
End Plate
FIGURE 1 GENERAL SCHEMATIC OF PLASTIC COALESCER CELL
-------
lacquer was applied to both surfaces of the lucite strips,
which were clamped together between wood strips and then
set by heating to 100°C for 1 hour. The edges of the frames
were given several seal coats of lacquer. 0-ring sealing
grooves were milled into the faces of the frames, nominal
dimensions for 3/16 in. cross section 0-ring being 13 in.
inside square, 5/64 in. deep and 1/4 in. wide. The thick-
ness of the individual 1/4 in. lucite strips varied up to
1/32 inches. Minor leaks, which continually developed on
bolting up the assembly due to cracking of material, did not
interfere with the preliminary test evaluations. These leaks
were eliminated in the final design.
The special frame for the fiber bed was constructed with
one of the 1/4 in. strips inset 1/2 in. to give a support
ledge for the fibrous bed assembly. The fiber bed was held
between two, 16 ga. sheets, drilled with 144-1/2 in. holes on
1 in. square centers which were located in the central
11 inch square. The plates with the fiber glass between
them (2 in. of 10.ly, 1/2, 1 or 1 1/2 in. of 3.2y, and 2
in. of 10.ly) was compressed to a thickness of 1/4
or 3/16 in. by 1/4 in. bolts spaced on 5-1/2- in.' square
centers. The assembly was then bolted to the projecting ledge
with 12 no* 8 bolts. This plastic assembly in operation had to
be mounted horizontally with the flow downward. The upper
circumferential 'edge was sealed by covering with 1 in. strip
of electrician's tape which in turn was covered with a
strip of silicone putty (John Mansville- Dux Seal).
The TFE screen M3J, Figure 1 , was mounted by taking
a piece, 13 in. square, bending the edges up 1/2 in. and
bolting the edges to the 'inside of a frame. A seal was
obtained by a strip of silicone putty around the edges.
The influent and water effluent streams flow thru
3/4 IPS nozzles in the center line of the end plates. The
oil effluent stream leaves thru a 1/2 in. IPS nozzle in the
wall of frame (s), Figure 1 .
-------
The end plates consisted of 1/4 in. steel plates stif-
fened with ribs of 1 1/2 in. x 1 1/2 in. x 1/4 in. angle
irons spaced 2 in. apart. The entire assembly was held
together by 8-3/8 in. bolts spaced on a 15 in. bolt square.
Final Coalescer Unit
The final large cell, constructed of aluminum, is shown
schematically in Figure 2. Photographs of this cell during
disassembly and assembly are shown in Figures 3-26.
The end plates(j_)and U^) consist of 16 in. x 16 in. x 2 in.
plates. The frames (3), (H) and us) are 16 in. outside square
(O.S.) by 12'in. inside square (I.S.) by 1 in. thick sawed
from plate stock. Each frame is provided with 4-1/2
in. FPT openings, located 3 1/2 inches from one edge. A
support plate QT) is 16 in. O.S. by 1 in. thick and the central
11 in. square area drilled with 144-1/4 in. holes on 1 in.
square centers. The various elements are sealed with 1/4 in.
0-rings. The 0-ring grooves are formed by seal frames,
(T), (IT), (To), ^2) , and (ij) . The inner frames (except(jT))
are 13 in. O.S. x 12 in. I.S. and the outer seal frame is
16 in. O.S. x 13 5/8 in. I.S. The inner seal frame (V) is
a solid plate with 144-1/4 in. holes which line up with those
in the diffuser plate (TJ so as to provide support via plate
QM. The seal frames are sawed from 1/4 in. plates. All of
these pieces are provided with 8-1/2 in. holes on a 15 in.
bolt square. The second stage screen barrier Q.3) consists
of a standard 100 mesh monel screen coated with 0.005 in. of
TFE.
-------
00©0©(!
\
\ \
n
u
n
\
u
n
u
n
u
D
D
D
D
D
D
D
(I) End Plate
(2) Seal Frame
(3) Entrance Frame
(T) Seal Frame
(5) Diffuser Tread Plate
(s) Coalescer Bed
(?) Diffuser Tread Plate
ff) Seal Frame
Support Plate
Seal Frame
Oil Effluent Frame
Seal Frame
100 Mesh Monel Screen Coated
with 0.005 in. Teflon
Seal Frame
Water Effluent Frame
End Plate
FIGURE 2 FINAL COALESCER UNIT
10
-------
FIGURE 3. COMPLETE CELL IN
HORIZONTAL POSITION WITH
INFLUENT SIDE ON TOP.
FIGURE 4. INFLUENT END PLATE
PARTIALLY REMOVED, SHOWING
TOP SEAL FRAMES.
FIGURE 5. ENTRANCE SIDE TO
DIAPHRAGM ASSEMBLY.
FIGURE 6. SPECIAL INNER
SEAL FRAME.
11
-------
FIGURE 7. OIL EFFLUENT FRAME
OVER TFE SCREEN SECOND
STAGE.
FIGURE 8. DISASSEMBLY VIEW.
FIGURE 9. ASSEMBLY VIEW.
FIGURE 10. TOP PERFORATED
PLATE, BOTTOM ASSEMBLED
DIAPHRAGM WITH TOP GASKET
IN PLACE.
12
-------
FIGURE 11. TOP GASKET REMOVED
SHOWING USED DIAPHRAGM.
SQUARE LINE AT EDGE IS METAL
O-RING SEAL WHICH HAS CUT
IHROUGH FIBER BLANKETS.
FIGURE 12. METAL'O-RING SEAL
AFTER CUT EDGES HAVE BEEN
REMOVED.
FIGURE 13. . BOTTOM VIEW: SHOWS
BOTTOM GASKET WITH GROOVE
FROM METAL O-KING AND SPACER
PLATE, ALL RESTING ON BOTTOM
PERFORATED PLATE.
FIGURE 14. VIEW OF TKEAD
PLATE AND FORMED GASKET ON
PERFORATED PLATE.
13
-------
FIGURE 15. 1/2 IN. OF 3.2u
GLASS, 2 IN. OF 10.lu GLASS
WITH METAL 0-RING SEAL ON
BOTTOM PERFORATED PLATE AND
GASKET. 1/8 IN. SPACER PLATE
IN POSITION. NOTE 1/2 IN. AND
5/16 IN. ALIGNMENT PINS.
FIGURE 16. COMPLETE 5 INCHES
OF FIBER GLASS.
FIGURE 17.
PLACt.
UPPER GASKET IN
FIGURE 18. UPPER PERFORATED
PLATE IN PLACE.
14
-------
FIGURE 19. COMPLETE ASSEMBLY
BUT NOT BOLTED TOGETHER.
FIGURE 20. TFE SCREEN BEING
INSERTED ABOVE WATER
EFFLUENT SPACER FRAME.
FIGURE 21. OIL EFFLUENT
SPACER FRAME.
FIGURE 22. EFFLUENT SIDE OF
DIAPHRAGM ALL ASSEMBLED.
15
-------
FIGURE 23. DIAPHRAGM THRU
EFFLUENT SIDE ASSEMBLY
FIGURE 24. INFLUENI SPACER
FRAME,
FIGURE 25. INFLUENT END
PLATE.
FIGURE 26. COMPLETE ASSEMBLY
16
,
-------
The fiber bed assembly is shown schematically
in Figure 27A. The fiber glass mats (D) , (ij) , and (l5) are
clamped between the two perforated plates (j$} , (T). The
screen used is detailed in Figure 278 and has 19% open area.
It is probable that the screen details are of little
importance since the coarse glass entrance Qj) and exit Q_5)
layers act as diffusers. The perforated plates are held in
place by the two diffuser plates (T) and (To) which consist
of 1/4 in. tread plates (overall thickness 5/16 in.) 18 in.
O.S. which are provided with 144-1/4 in. holes on the 11 in.
I.S. The space between the treads and the perforated
plates are sealed by a urethane rubber gasket \2\and QT) .
(Devcon Corp., Danvers, Mass., Flexane 85, Liquid Type.)
The gasket (approximately 16 in. O.S. by 12 in. I.S.) is
formed by casting a strip of urethane between 1/16 in. string
guides (thickness of viscous liquid over 1/16 in.), the
urethane thickens to a putty within 30 minutes at which time
the string guide is removed and the tread plate and perforated
plate are clamped together. (See note on indexing below.)
The 1/4 in. overall membrane thickness is maintained by
the thickness of the gaskets (^) and (T) and the spacer
frame (T). The gaskets are 18 in. O.S. x 12 in. I.S. x 1/16
in. thick and the spacer frame is 18 in. O.S. x 16 in. I.S.
x 1/8 in. thick. The membrane thickness can be varied by
changing the thickness of either the gaskets or the spacer
frame.
The membrane edge is sealed by the metal O-ring (If) which
is 1/8 in. cross sectional diameter x 13 1/2 in. I.S.
The fiber glass membrane is composed of 10.ly glass
fibers (nominal density 0.6 Ib./ft. , Owens Corning "Aerocor")
3
and 3.2]_i glass fibers (nominal density 0.6 Ib./ft. Owens
Corning FM-004).
The diaphragm assembly is drilled with 32-5/16 in. holes
on 17 in. bolt square and also 8-1/2 in. holes on 15 in. bolt
17
-------
00
V/////////////////////7/X
1/8"
FIGURE 27A FIBROUS BED ASSEMBLY
(T) Diffuser Tread Plate
(?) Urethane Rubber Gasket
(3) Screen Support
(4) Gasket
(5) Spacer Frame
(5) Metal 0-Ring
(T) Gasket
(?) Screen Support
(9) Urethane Rubber Gasket
@ Diffuser Tread Plate
^ Outside Bolt Square,
8-1/2 in. Diameter Holes
fi^ Inside Bolt Square,
32-5/16 in. Diameter Holes
6 rows/inch,
60 holes /in ,
.050"
t
Open area 19.4% ~~~ .
RE 27B SCREEN DETAILS
-------
square. Two of the 5/16 in. and two of the 1/2 in. holes are
used for alignment pining to maintain all pieces, particu-
larly gaskets (T) and (V) within 0.005 in. It was found
that only the central 4-5/16 in. holes are needed to pull
up the fiber glass bed near 1/4 in. spacing since the final
thickness is established when the cell is bolted together
through the 1/2 in. holes using 3/8 in. bolts. Only 6 bolts
are used since two center holes are used for alignment pins
(0.480 in. dia.).
Test Assembly
The test assembly is shown schematically in Figure 28
and pictured in Figure 29. it consists of four major com-
ponents, mounted on dollies; 1) the filter/coalescer,
2) the feed tank, 3) the pumping system, and 4) kerosene
pumping system. Only the final modification will be de-
scribed here which allows the unit to be operated con-
tinuously with a constant oil feed, automatic water makeup
and constant temperature via manual control. The flow rate
thru the cell is maintained by periodic adjustment.
The feed tank unit consists of a stirred 75 gal. tank
U.Q) . Excess water is added to the internal overflow tank
(T3) . The latter automatically adds any necessary volume of
makeup water. The system temperature is maintained at any
constant value by varying the amount of water to drain
(ml./min. to gpm) and thus the amount of cold water makeup.
Normally the pump suction is via line {9J to which is added
directly the oil feed via line (s). In the case of an
actual pollutant stream, the latter is added to the tank
put thru the system and effluent from cell returned to tank.
Thus the pollutant concentration varies as a first order
reaction and may be evaluated over an hour with decreasing
inlet concentration rather than in 7 minutes with once
through flow. The kerosene solvent which is added to the
suction line \9j does not enter the tank or interfere with
the evaluation. In preliminary tests oil was added batch-
19
-------
Liquid Ring Pump
By-Pass Line
Feed Line
Flow Indicator
5p Cotton Filter
Oil Feed Tank
Zenith Pump for Oil Feed
Oil Feed Line
Pump Suction Line
75 Gallons Polyethylene Tank
Stirrer
Makeup Water
Internal Overflow Tank
Overflow To Drain
Cell Inlet
Inlet Sample Valve
Coalescer Cell
Pressure Indicator
Oil and Air Vent Valve
Oil and Air Vent Valve
Pressure Indicator
Outlet Sample Valve
Cell Effluent
FIGURE 28 TEST. ASSEMBLY
-------
FIGURE 29. PHOTOGRAPH OF TEST ASSEMBLY
21
-------
wise to the tank every 1/2 hour so that influent concentra-
tions would vary between 100 to 25 ppm or otherwise as
desired.
The main system pump (T) is a liquid ring pump (Gould
Model 2520) which operates at 3450 KPM giving approximately
25 gpm at 25 psi. Its characteristics of pumping air or water
at up to 25 in. vacuum eliminates the problems of pump priming
which are present with centrifugal pumps. The oil feed is
added by a Zenith pump (7) driven by a variable speed Graham
transmission. The oil addition rates (2.2 ml./min. = 85 ppm
at 7.5 gpm to coalescer) are adjusted by nudging the setting
and by timing the RPM since the Graham settings are insen-
sitive in the low range.
The flow rate is determined by a rotameter (Fischer Porter
FP-1-35-G-10/83)float over 6.9 gpm and float for under 6.9
gpm. A filter (Fulflo Model 5y filter) was used in several
tests. Its use was required with the sample of overflow
from back wash which otherwise plugged the fiber bed within
minutes. Depending upon length of use, the 5y filter
cartridge would coalesce from 50 to 5% of the entering
emulsion.
22
-------
SECTION IV
RESULTS OF LARGE COALESCER TESTS
Preliminary Coalescer Tests
The first tests conducted on the plastic filter coalescer
were made at velocities of 0.1 to 0.8 fpm with an influent
containing 100-500 ppm of oil. The effluent contained 10-14
ppm when using 1/1 kerosene/130 oil and 22-33 ppm of 1/1
kerosene/ILS (pollutant oil recovered from the skimming tank
at Interlake Steel Corporation). The pressure drop across
the diaphragm increased from 3 to 9 psi in two hours for both
cases. An attempt to regenerate the bed by passing pure water
thru it at high velocity resulted in a diaphragm with essen-
tially zero coalescing ability.
The fiber bed employed in the first tests was maintained
by internal bolts between the diffuser plates. Since this
bolting technique presents a possible leakage path subse-
quent tests employed peripherial bolts to eliminate leakage
at these points.Tests were made at 0.6 fpm using a diaphragm
of 2 in. of 10. ly , 1 in. of 3.2y, and 2 in. of 10. ly glass
compressed to approximately 3/16 in. thickness. The in-
fluent contained 150 ppm of kerosene plus oil, and the efflu-
ent contained no detectable oil. The pressure drop across the
diaphragm increased from 2 to 6 psi in two hours.
Final Coalescer Tests
The data for the final large cell are summarized in
Table 8 and Figure 30. Run data is tabulated in Tables 1-7.
These show that in all runs the effluent stream contained
from 0 to 7 ppm of oil with an average of approximately 1
ppm (Table 8). Since readings are generally non-detectable,
the high values are probably due to zero point drift in the
light transmission equipment (calibration of light transmission
apparatus on page 53). These effluent concentrations are
independant of the test parameters: thickness and density
23
-------
of 3.2vi layer (Table 1), velocity through the bed (Table 5),
temperature (Table 5B and 6), length of run, and pressure
buildup.
The capacity of the coalescer bed, that is, the gallons
of influent with 85 ppm of oil that can be passed through to
a pressure drop buildup of 25 psi, was found to be 710 gal.
for a flow rate of 7.5 gpm (1 fpm velocity) and at temperature
of 85°F (Table 7 ). At 65°F the capacity drops to 100 gal.
(Table 6 ). At flow rate of 1.85 gpm (1/4 fpm velocity)
Table 5 , the capacity of the bed was 540 gal. for a tempera-
ture of 85°F and 280 gal. for 68°F. Table 3A shows that
the effluent from the sand filters can be coalesced effec-
tively. Table 4 shows that the back wash overflow can be
coalesced but requires a filter in the line to eliminate
plugging of the bed which occurred in the first test with
this material. The high turbidity readings of the coalescer
effluent are probably due to color bodies since visual
inspection did not indicate the presence of any emulsion.
Discussion of Results and Conclusions
The efficiency of the final large cell appears to be
essentially 100% under all conditions. The test data show
some points with effluent concentrations up to 7 ppm. These
are probably due to variations in the analytic method.
The controlling parameter for a practical unit appears
to be the life of the bed. This is arbitrarily defined in
the present work as the through-put up to the point where
the pressure drop reaches 25 psi. The rate of increase
in the pressure drop is a function of the flow rate, the
amount of oil in the influent, and the temperature of the
fluid. The pressure drop has never been observed in our
work to stabilize at any value even when subjected to the
single phase flow of pure water. Consequently, it appears
that the pressure drop is due both to gradual accumulation
of oil within the fibers and also mechanical degradation of
the fibers. During the coalescing phenomenon some +95% of
24
-------
ro
un
TABLE 1*
Large Scale Coalescer Tests With Double Layer 3.2y Glass
Time
hours
0
1/2
1
1 1/4
1 1/2
3 1/2
3 1/2
6
6 1/2
7
7 1/4
7 3/4
8
8 1/2
9 1/2
10
11
11 1/2
12 1/2
13
13
13 1/2
14
14 1/2
Oil
Added
ml.
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
SHUT
100*
100
100
100
Plow
Rate
GPM
ft7
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
DOWN OVER
6.4
6.4
6.4
6.4
Velocity
fpnv
1
1
1
1 '
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
NIGHT
0.85
0.85
0.85
0.85
Temp.
0F
_
68
64
83
80
81
66
68
68
72
73
75
77
78
80
82
83
84
65
68
74
78
Pressure
Drop
psi
1.0
_
_
1.8
4.0
6.0
7.5
7.5
7.5
7.5
7.8
7.8
7.8
8.0
8.2
8.2
9.0
9.0
12.0
12.0
12.5
12.5
Feed +
Tank
75
96
29
103
-
26
75
63
62
35
110
13
121
28
118
25
108
24
118
21
145
35
118
31
Oil ppm **
Influent +
to Bed
13
41
6
59
13
54
42
53
21
77
0
81
23
77
16
70
14
79
14
95
19
95
19
Effluent
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
SHUT DOWN OVER NIGHT
(Continued)
-------
TABLE 1 (Continued)
ro
01
14
15
15
16
16
18
20
21
1/2
1/2
3/4
3/4
1/2
1/2
100
100
100
100
NO
0
0
0
5.
5.
5.
5.
more
5.
5.
5.
Filter
32
33
34
35
51
0
0
0
0
0
5.
5.
4.
3.
1.
6
5
5
5
0
0
0
0
.75
.73
.73
.73
oil - water
3
0
4
Removed
8
0
0
2
04
0
0
0
0
0
0
0
0
.71
.67
.72
.77
.67
.53
.43
.14
56
-
-
-
16
-
-
0 63 62
- -
- -
- -
0
-
-
-
thru continuously
54
54
76
72
52
53
50
16
17
17
17
18
20
20
21
0 -
0 -
0 -
0 -
0 -
0 -
0 -
5 -
-
-
-
-
"
+ The 5u Filter in the line coalesced some of the oil
** By light transmission
* Test conditions tabulated in Table 8
-------
ro
TABLE 2*
Large Scale Coalescer Test. Standard Run at 70°F
Time
Hours
0
1/4
1/2
3/4
1 1/4
2 1/4
2 3/4
6 1/4
7 1/4
9 3/4
11 1/4
12 3/4
13 1/4
18 1/4
19
19 3/4
21 1/2
25 3/4
26 3/4
30 1/4
33 3/4
35 1/4
36 3/4
45 1/2
46 3/4
Flow
Rate
GPM
ft7
Oil added
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.4
7.3
6.4
6.1
6.1
6.0
4.9
4.9
Velocity
fpm
continuously
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.99
0.97
0.85
0.81
0.81
0.80
0.65
0.65
Temp.
oF
at 155 ml.
_
58
-
62
65
-
70
70
-
-
-
-
-
70
69
68
67
-
73
-
-
-
-
~
Pressure
Drop
psi
per hour
1.5
2.5
3.0
3.0
3.3
3.8
3.9
5.3
5.7
7.5
9.0
10.0
10.5
13.5
17.0
15.0
17.0
19.5
19.8
19.0
21.3
21.5
21.8
22.0
22.5
Oil ppm**
Influent Effluent
_ _
160 0
- -
60 0
- -
- -
70 0
75
- -
- -
- -
- -
- -
83 0
100
- -
87 0
-
85 0 End
** By light transmission
* See Table 8 for details
-------
ro
CO
TABLE 3*
Large Scale Coalescer Tests With Samples from Interlake Steel Corp.
Note: Samples consisted of 55 gal. The cell effluent was recycled back to the system
tank rather than running once through. This permitted the run to be extended over
a longer period with a continually decreasing (1st order) oil concentration in the
tank. The solvent kerosene was added to the pump suction at the rate of 2.2 ml.
per min.
Time
Min.
Run 3 A
0
15
17
20
32
38
Run 3B
38
38
40
41
47
49
Flow
Rate
GPM
ft2
Effluent
7.5
Velocity
fpm
Temp.
oF
Pressure
Drop, psi
Oil Concentration TU**
Tank Influent Effluent
from sand filters
1
79
Kerosene added continuously
7.5 1
7.5
7.5
7.5
Overflow
1
1
1
liquid from
-
82
the backwash
1.0
2.0
2.5
4.5
6.0
clarif ier
29 5
30
- 40
20 41 3.5
10 22 2.0
12 20 2.0
No kerosene added
7.5
Started
7.5
^2.0
O.n
1
through with
1
0.27
n
kerosene
78
-
"
8.5
25.0
25.0
_ _ _
25 47 3
_ _
_ _ _
** Read by Hach Turbidimeter
* See Table 8 for details
-------
Note;
TABLE 4*
Large Scale Coalescer Tests with Samples from Interlake Steel Corp.
Placed 5y filter before coalescer bed.
Time
min.
0
4
6
9
10
10
'13
17
19
22
23
29
37
Flow
Rate
GPM
ft*
Velocity
fpm
Pressure
Temp . Drop
°F psi
Oil
Tank
Concentration
Influent
to bed
TU***
Effluent
No kerosene added
7.5
Kerosene
7.5
7.5
Kerosene
7.5
7.5
7.5
7.5
7.5
7.5
7.5
1
added at
1
1
added at
1
1
1
1
1
1
1
1.5
2.2 ml. per min.
5.5 ml. per min.
76 2.9
3.8
4.2
- -
- -
- -
80 7.0
27
_
12
13
-
6
-
4
12
10
9
10
-
-
-
14
15
18
6.8**
_
6.0
_
4.5
-
4.0
-
4.0
3.0
*** By Hach Turbidimeter
** Color bodies in the material are probably responsible for these readings
* See Table 8 for details
-------
TABLE 5*
Large Scale Coalescer Tests - One Fourth Standard Flow Rate
CO
o
Time
Hrs.
Run 5 A
0.0
0.25
0.5
1.0
1.5
2.0
3.0
3.5
5.5
6.5
7.0
8.5
9.5
10.5
12.0
13.0
14.0
24.0
25.0
27.0
30.0
32.0
34.0
36.0
38.0
45.0
Velocity
fpm
Flow
Rate
GPM
Ft7
Temp.
OF
Pressure
Drop
psi
TU**
Influent
Effluent
ppm*
Influent
**
Effluent
High temperature
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
-
-
66
-
72
72
80
78
-
84
84
88
88
84
87
78
81
91
84
88
89
77
83
86
86
87
(17"H20)
(30"H20)
0.5
1.30
1.30
1.40
1.50
1.50
1.60
1.60
1.60
-
1.70
1.75
1.87
2.20
2.25
3.20
3.35
3.50
3.90
4.60
4.45
4.70
4.75
5.20
-
-
44.0
31.0
85.0
20.0
-
11.0
17.0
18.0
16.0
15.0
15.0
17.0
15.0
16.0
16.0
16.0
15.0
16.0
16.0
17.0
16.0
16.0
-
-
_
0.45
0.35
0.34
0.35
-
0.30
0.32
0.30
0.28
0.28
0.28
0.31
0.31
0.39
0.45
0.70
0.71
0.95
0.96
0.96
0.86
0.91
-
-
102.0
125.0
107.0
187.0
97.0
-
67.0
90.0
95.0
85.0
83.0
83.0
90.0
83.0
95.0
90.0
83.0
85.0
87.0
90.0
90.0
87.0
90.0
-
0.0
-
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.0
2.0
4.0
4.0
4.0
5.0
4.0
-------
TABLE 5* (continued)
47.0
49.0
51.0
54.0
56.0
58.5
60.0
68.0
71.0
73.0
75.0
78.0
80.0
83.0
93.5
95.0
97.0
Run
99.0
103.0
104.0
106.0
107.0
108.0
109.0
117.0
119.0
120.0
121.0
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
5B Low
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1/4
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
88
90
81
85
86
88
83
88
94
84
90
89
91
88
97
86
95
5.40
5.45
5.80
5.70
5.80
6.00
6.20
6.60
6.85
7.00
6.90
7.10
7.25
7.50
8.60
8.40
8.30
16.0
17.0
17.0
17.0
17.0
17.0
15.0
17.0
17.0
17,0
16.0
16.0
17.0
17.0
18.0
16.0
16.0
0.93
0.95
1.10
1.10
1.00
1.00
0.99
0.95
0.60
1.40
1.20
1.10
0.90
1.10
0.56
0.96
0.94
87.0
90.0
90.0
90.0
90.0
97.0
90.0
81.0
80.0
83.0
83.0
87.0
87.0
87.0
83.0
83.0
87.0
3.0
5.0
5.0
5.0
4.0
4.0
4.0
3.0
2.0
7.0
6.0
5.0
4.0
3.0
4.0
7.0
4.0
temperature
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
68
69
67
68
68
69
68
68
69
68
68
10.00
11.50
12.00
13.60
14.00
14.50
15.00
16.20
18.80
19.30
19.80
16.0
19.0
17.0
16.0
17.0
16.0
16.0
17.0
16.0
17.0
16.0
0.36
0.39
0.26
0.25
0.28
0.22
0.25
0.21
0.23
0.25
0.23
83.0
92.0
87.0
87.0
90.0
87.0
85.0
83.0
85.0
87.0
87.0
0.0
1.0
0.0
0.0
0.0
0.0
0.0
1.0
0.0
0.0
0.0
*** By light transmission
** By Hach turbidimeter
* See Table 8 for details
-------
CO
ro
TABLE 6*
Large Scale Coalescer Tests - At Low Temperature
Time
Hrs.
0.00
0.25
0.50
1.00
1.30
2.00
2.30
3.00
3.30
4.00
4.30
5.00
6.00
7.00
8.00
9.00
10.00
Flow
Rate Velocity Temp
GPM fpm °F
ft2
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
-
69
69
64
66
65
-
64
64
64
63
64
64
64
63
65
65
Pressure
Drop
psi
Pl
1.75
2.40
3.10
4.00
4.90
5.60
6.25
6.90
7.60
8.40
9.20
10.10
12.10
14.15
16.00
18.20
19.50
P2
0.75
0.75
0.75
0,75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.80
0.80
0.80
0.90
0.75
TU **
Influent
AP
1.00
1.65
2.35
3.25
4.15
4.85
5.50
6.15
6.85
7.65
8.45
9.35
11.30
13.35
15.20
17.30
18.75
18.0
18.0
18.0
17.0
18.0
20.0
18.0
20.0
20.0
18.0
18.0
20.0
19.0
23.0
20.0
Effluent
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-
.20
.27
.16
.19
.16
.20
.22
.30
.31
.32
.26
.30
.27
.32
.30
ppm
Ii. fluent
-
81.0
80.0
85.0
85.0
85.0
-
85.0
83.0
85.0
85.0
83.0
83.0
83.0
85.0
85.0
85.0
***
Effluent
-
0.0
1.0
1.0
0.0
0.0
-
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.0
3.0
2.0
*** By light transmission
** By Hach turbidimeter
* See Table 8 for details
-------
TABLE 7*
Large Scale Coalescer Tests - At Standard Conditions
CO
CO
Time
Hrs.
0.00
0.25
0.50
0.75
1.00
1.30
2.00
3.00
4.00
5.00
6.00
13.00
14.00
15.00
17.00
19.00
21.00
23.00
24.00
26.00
28.00
30.00
37.50
42.50
44.00
45.00
46.00
47.00
Flow
Rate
GPM
ft"2"
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
Velocity
fpm
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Temp.
°F
91
90
85
85
88
88
88
87
87
84
84
84
84
84
86
86
88
87
84
85
84
87
86
84
86
90
Pressure
Drop
psi
0.75
1.25
1.60
1.70
1.90
2.15
2.35
2.60
2.80
3.00
3.20
4.40
4.50
4.60
4.80
5.00
5.30
5.55
5.70
5.90
6.20
6.55
8.00
8.90
9.20
9.35
9.60
9.80
TU**
Influent
*«
19.0
18.0
18.0
17.0
18.0
19.0
18.0
19,0
20.0
19.0
18.0
19.0
20.0
-
18.0
19.0
19.0
20.0
19.0
20.0
18.0
20.0
20.0
19.0
20.0
20.0
20.0
Effluent
_,
0.25
0.22
0.20
0.18
0.20
0.25
0.25
0.26
0.28
0.23
0.25
0.26
0.24
-
0.25
0.27
0.25
0.22
0.20
0.19
0.18
0.20
0.18
0.20
0.22
0.18
0.18
ppm
Influent
_
83
85
85
81
85
85
83
85
83
85
83
85
85
-
83
85
85
83
85
85
83
85
85
83
85
87
85
***
Effluent
_
0.0
0.0
0.0
0.0
0.0
1.0
0.0
0.0
1.0
0.0
0.0
0.0
2.0
-
1.0
1.0
1.0
1.0
1.0
2.0
1.0
2.0
1.0
1.0
1.0
2.0
1.0
-------
TABLE 7* (continued)
48.00
49.00
50.00
51.00
52.00
53.00
59.00
68.00
68.50
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
1
1
1
1
1
1
1
1
1
86
80
82
84
84
84
88
88
88
10.25
11.10
11.00
11.30
11.60
11.90
14.40
17.70
17.90
19.0
20.0
18.0
20.0
19.0
19.0
22.0
20.0
21.0
0.25
0.21
0.20
0.25
0.25
0.24
0.25
0.28
0.30
85
85
83
85
85
85
87
85
85
2.0
1.0
1.0
2.0
2.0
1.0
2.0
2.0
2.0
*** By light transmission
** By Hach turbidimeter
* See Table 8 for details
-------
CO
en
TABLE 8
Summary of Test Results with Aluminum Cell
Run
No.
1
2
3A
3B
4
5A
5B
6
7
1
2
3
4
*
**
Inches
Uncom.
.-., Fibers
Ol1 10. lu/
3. 2|/10. In
K/#30
K/#30
K/P ***
K/P
K/P
K
K
K
K
Nominal
2/1/2
24/2
2/^/2
L
2/^/2
24/2
1
2/^/2
2/^/2
2/^/2
2/^/2
thickness.
Compressed density
Compressed
Density
Ib/ft3
12.
10.
10.
10.
10.
10.
10.
10.
10.
in .
in
Extrapolation by linear
Sample of pollutant
ppm
T 1 1
0
8
8
8
8
8
8
8
8
of 10.
Ib/ft3
Flow
Rate
GPM
7
7
7
7
.7
1
1
7
7
.5
.5
.5
.5
.5
.85
.85
.5
.50
W3
ratioing
stream from
last
Temp
°F.
-80
~ 70
~ 80
-
-80
-85
68
64
85
.2|J/ 10.
point
Pressure
Drop vs. Time At AP=25 psi
psi hrs. gal. hrs.
9.
19.
6.
-
7.
8.
19.
18.
17.
0
5
0
0
3
8
8
9
22.5 415 56
26.0 210 28
1/2
1/6
2/3
97 540 294
121 (280) (152)
10 100 13
68.5 710 96
Efflu-
Tank Influent ent Remarks
13-80* 0* with 5M filter
85* 0*
29-12** 40-20** 0* ILS4 sand filter
30-20** 47-30** 0* ILS back wash overflow
27-13** 10* 0* ILS backwash overflow
with 5 |a filter
85* 85* 3*
85* 85* 3* Excess cooling water
85* 85* 3*
85* - 2*
1[J. fibers
Interlake Steel Corp.
*** P stands for pollutant
-------
25.0
CO
CTl
10
15 20
Time, Hours
25
30
35
40
FIGURE 30 SUMMARY OF PERFORMANCE CHARACTERISTICS OF 1 SQ. FT. FILTER/COALESCER
(SELECTED POINTS)
-------
coalesced oil passes through the bed. The rate of degrada-
tion does not appear to be a function of the actual pressure
drop. Abrupt increases occur if the runs are not continuous.
Previous work on removal of the oil held up in the used beds
by regeneration treatments, such as solvent extraction,
usually resulted in a bed with poor coalescence efficiency.
Treatments with fiber lubricants, which made the surface more
hydrophobia, were unsuccessful in stabilizing the beds.
Recently (see Stabilization Test Section) a small bed 1/2
in. effective diameter, was treated with 10% isobutyl metha-
crylate resin in methylene chloride. This bed gave ^100%
coalescence effectiveness and similar pressure buildup as
the regular large cell. Treatment of this used bed with
hexane or methanol solvent to remove the oil restored the
bed to the original characteristics of low pressure drop
and 100% coalescer effectiveness found in unused fibers.
37
-------
SECTION V
ECONOMICS OF FILTER/COALESCER OPERATION
We have made rough estimates of treatment coats for
worst case conditions of 910 gal/ft? at 85°C which correspond
to results from actual tests on fiber as received from manu-
facturer. Preliminary tests indicate the fibers can be sta-
bilized and the medium regenerated. If so, medium life could
be extended many fold, such as 10 or 100 times. Under these
conditions the cost of fibers and labor would decrease pro-
portionally. Likewise the heating cost could be eliminated.
1. Fiber glass filter bed
a) Material cost: At the moment the replacement cost of
the fine glass membrane is the controlling cost since at 7.5
gpm and 85°F we get 710 gal. thru when AP reaches 25 psi.
Preliminary tests indicate that this membrane can be regenerated
and thus the life (before compression) will be some multiple
of this. The actual life will depend on the efficiency of the
filter and the nature of the stream being processed, i.e. the
dirt which leaks thru and plugs the fine glass.
The bed consists of 2 in. of coarse glass, 1/2 in. of fine
glass and 2 in. of coarse glass.
Only the fine glass layer would have to be replaced at
present due to fiber migration.
Cost of fine glass in $/10 gal. of effluent
1. 1/2 in. layer $57.00/103ft2.
2. Cells 4 ft. x 4 ft. with 3.5 ft. x 3.5 ft. active area
3. $/103 gal. = 1§2) ( _!i_ ) Ifll = 0.10
(10J) 3.52 710
b) Labor Costs: The labor cost with a modern automated
filter press, for standard operating and change over is minimal.
Again change over will constitute over 90% of the cost.
Assuming a large unit - 100 cells, each 42 in. x 42 in. active
39
-------
area, the actual time to change the pink layer would vary from
1 minute down to 10 seconds. The shut down and start up times
of 10 to 20 minutes per cell can be neglected. Using labor
costs of 5.75 $/hr. and supervision of 50% of labor (treatment
of waste water - waste oil mixtures, US Dlf FWPCA 12010 EZV 02/70
May 1970 p. 81) and 1 min. per overall labor burden per all.
(1) 101 = .Q16 $/1Q3
z
(60) (3.5) 710
2. Pumping Costs
p=,rQ-K.=~« equals 15 psi (.5 to 25). Using 0.01 $/KWH,
ctVGjTclCj G
75% overall efficiency of large system, the ocst calculation
is:
( li!_ ) (15) (144) ( ) - 0.746 = 0.067 KW/ft? area.
7.48 60 (550) (.75)
2
gpm psi , in. v min. KW/HP
7 2 3
gal/ft;5 ft. sec. .ft. /sea. ,ff .
HP
(0.01) (0.067) ( i^- ) ( ) = .0016 $/103 gal.
7.5 60
ft. gpm min.
3. Heating Cost
Assume that the effluent is at 60°F and must be main-
tained at 85°F using heat exchanger with 90% heat recovery
and heat at $1/106 Btu.
(103craL) (8.33) (85-60) (-(1-0) ^10) ) = .0125 $/103 gal.
40
-------
4. Capitol Costs
Direct quotation on 42 in. iron frame 100 units, 2100
2
ft and completely automated is $50,000.
Cost/ft.2 = 50,000/2100 = 24 $/ft2
Installation at 30% give (1.3) (24) = 31 ,$/ft.2
Using straight line amortization over 10 years and a
90% stream factor this capitol cost can be calculated
(neglecting land, buildings, other special accessories such as
heat exchangers).
gal./ft?/10 years = (7.5) (60) (24) (356) (.9) (10) = 35.4 x 106 gal.
dep/103 gaL = 31 X 10 , = .0009 $/103 gal.
35.4 x 10b
5. Summary
The cost in $/10 gal. are summarized in the table below.
Fiber 0.10
Pumps 0.002
Labor 0.016
Heating 0.013
Filter coalescer .0009
0.13
41
-------
SECTION VI
DESCRIPTION OF SMALL COALESCER
Experiments were carried out to determine the degree of
separation in the small coalescer cell together with the
settler.
Recirculating System
Figure 31 is the schematic diagram of the apparatus used
in this study- The emulsion was made in a 6 liter baffled
stainless steel mixing tank MJ using a model F Lightnin
Laboratory mixer (2j which had a 2-inch impeller and speeds
could be varied from 800 to 1600 rpm but experiments were run
at 1600 rpm. The agitator speed was measured with a strobo-
scope. A cooling coil (T) made from 3/8 O.D. copper tubing
was put into the tank to maintain the constant temperature
of the emulsion by controlling the circulation of tap water.
A Jabsco 1725 rpm rubber paddle pump (V) was used to circulate
the emulsion through the system. Before the emulsion went
into the system, it passed through a two liter settler (ji)
where entrained air was removed. A standoff pipe (6J at the
top of the settler allowed separation and removal of the air.
It is essential to remove all entrained air before proceeding
to the optical cell of light transmission device. After
passing through the settler, the emulsion was metered through
one of the two rotameters,(V) . After metering, the emulsion
passed through the inlet optical cell (jy of the light
transmission apparatus, then through the fibrous bed Mn where
it coalesced. The separated oil stream before the teflon
coated screen, and resulting emulsion after passing through
the teflon coated screen and outlet optical cell (lO) was
returned to the baffled mixing tank. For some runs instead
of teflon coated screen a 2 liter settler U-U in series with
coalescer cell was used to separate coalesced oil. The
pressure drop across the fibrous bed was measured at the
43
-------
water out-*i'
v/ater in*-i
(2)
(T)
(7)
(D
(?)
?
Standard Optical
Cell
Photocomparator
Recorder
6 Liter Baffled Stainless
Steel Tank
Mixer
Cooling Coil
Rubber Paddle Pump
Air Settler
Air Removal Port
Flow Indicators
Inlet Optical .Cell
Fibrous Bed Test Section
Outlet Optical Cell
Oil Settler
I
Pressure Indicators
Manometer
FIGURE 31 SCHEMATIC DIAGRAM OF RECIRCULATING SYSTEM
-------
inlet and outlet of the cell by pressure gages Q.J) or by a
mercury manometer (0) .
The recirculating system has several disadvantages:
there is a change in inlet drop size distribution with time
because of the mixing of streams with different drop size
distributions and the inlet oil concentration cannot be kept
constant, due to the oil holdup without adding fresh oil in
the mixing tank. A once-through system was constructed later
to overcome these disadvantages.
Once-through System
Figure 32 is the schematic diagram and Figure 33 is the
photograph of the apparatus used. The emulsion was made, in
a 30 gallon stainless steel tank fl) by recycling the oil/water
mixture through the by-pass line. A glass fiber filter \2J
could be valved into the by-pass line to remove particulate
matter present in the water. The filter was employed prior
to making up the emulsion. To control the temperature a
heat exchanger \3J was installed between the centrifugal pump
(T) and by-pass filter assembly. This assembly is shown
in Figure 34. The remaining part of the system was essen-
tially the same, shown on the right of the centerline in
Figure 31,as that used in recirculating system except that
the effluent from the cell, settler and the optical cell was
discharged into the drain.
Coalescer Cell
The coalescer cell is shown in Figure 35 and its photo-
graph in Figure 36. The cell consists of an external 2 1/2
in. O.D. x 6 in. Plexiglas tube (V). Inside the outer tubes
are placed three retainer Plexiglas tubes (T) which are
sealed by 1/8 in. O-rings. Four 1/8 in. mesh wire screens(T)
were used to hold the bed between the center and the inlet
retainer tubes. A 100 mesh teflon coated screen \5J was
held between center and outlet retainer tubes. The Plexiglas
45
-------
Standard Optical
Cell
Photocomparator
Recorder
30 Gallons Stainless
Steel Tank
Glass Fiber Filter
Heat Exchanger
Centrifugal Pump
Air Settler
Air Removal Port
Flow Indicators
Inlet Optical Cell
Fibrous Bed Test Section
Outlet Optical Cell
Oil Settler
Pressure Indicators
Manometer
FIGURE 32 SCHEMATIC DIAGRAM OF ONCE-THROUGH SYSTEM
-------
FIGURE 33. PHOTOGRAPH OF ONCE-THROUGH SYSTEM,
FIGURE 34. PHOTOGRAPH OF INLET SECTION ASSEMBLY
OF ONCE-THROUGH SYSTEM.
-------
cell allowed visual observation of the bed operation. The
coalesced drops of the dispersed phase were removed through
an outlet (7) at the top of the cell. The desired compres-
sion of the bed was achieved by the adjustment of the tie
bolts (IT). The flow cross-section of the cell was 1.77 sq.
inches.
The once-through system used a cell of similar construc-
tion but with a smaller flow cross section of 1 sq. in. This
allowed the long period of operation of the bed, than one with
1.77 sq. in. cross-section for a given volume of emulsion in
the feed tank.
The coalescer beds employed a 3.2y fiber glass coated
with bakelite (Owens Corning FM-004). In many cases front and
rear layers of a 10. ly bakelite coated fiber glass (Owens
Corning - Aerocor) was also used.
Some runs were also made with uncoated 5.6y fibers which
were used as received and also with different hydrophobic
coatings. The size distributions and physical properties of
these fibers are given in Table 9 .
The fiber beds of 2 in. dia. were cut from fiberglas
sheets by a 2 in. O.D. cutting die. The pads were weighed
and placed in Plexiglas cell as shown in Figure 35. The
first layer (V) is coarse coated fiberglas fibers of 10. ly
mean diameter, the second coalescer layer (jT) is of fine
coated fiberglas fibers of 3.2y mean diameter, and third
separator layer (T) is same as first layer. In some runs
uncoated fibers of 5.6y mean diameter were used as the
coalescer layer. Some runs were made only with the single
coalescer layer of various fibers used. The desired bed
density and length of the bed was determined by draw up of
the tie bolts.
48
-------
©
©
(T) Retainer Plexiglas Tubes
(3) Wire Screens
(T) 10. ly Coarse Fiber Glass Layers
(T) 3.2p Fine Fiber Glass Layer
(5J 100 mesh Monel Screen Coated with.
0.005 in. of Teflon
(?) External Plexiglas Tube
(?) Coalesced Oil Outlet
(8) Tie Bolts
FIGURE 35 SMALL SCALE COALESCER CELL WITH FIBROUS BED TEST SECTION
-------
en
o
FIGURE 36. PHOTOGRAPH OF SMALL SCALE COALESCER CELL
-------
TABLE 9
Physical Properties of the Fibers
Fine fiberglass (pink) Owens Corning FM-004
Size distribution
Interval
Microns
1-2
2-3
3-4
4-6
6-8
8-10
No. of
fibers
11
11
18
8
2
0
2
Weight/ft. Density
0.025 lb./ft.2 2.57^0.02 g./cm.
3
50
Mean diameter = 3.2y
Coarse fiberglass (yellow) Owens Corning "Aerocor"
Size distribution
Interval
Microns
1-3
3-5
5-7
7-10
10-15
15-25
>25
No. of
fibers
2
5
4
17
18
3
1
Density
2.57±0.02 g./cm.3
50
Mean diameter = 10.ly
Uncoated glass fibers
Size distribution
Internal
Microns
1-3
3-5
5-7
7-9
9-11
11-13
No. of
fibers
10
24
18
8
7
3
Density
2.50 g./cm.
70
Mean diameter = 5.6y
51
-------
Oil Dispersions
Kerosene, 1/1 #30 single grade mobil oil/Kerosene and
1/1 ILS extracted oil/Kerosene were used as the dispersed
phase. Some runs were also made with 0.1/1, 0.5/1 #30 oil/
Kerosene. In all cases water was used as the continuous
phase. Each run was made with fresh material. Before and
after the apparatus was used, it was cleaned by flushing
with water. Physical properties of the material used are
given in Table 10.
Concentration Measurement by Light Transmission
The light transmission device used to determine
the oil concentration is described by Sherony (16).
Measurement of Light Transmission
The diagram of the apparatus is shown in Figure 37.
Light from a 100 watt bulb passes through an optical cell,
which is filled with emulsion, and strikes the photocell.
Light passing through another optical cell, filled with
pure continuous phase, is also monitored by a photocell.
The signals from both of these photocells are fed to a
photocomparator. This photocomparator has been connected
to a recorder. Details of the comparator circuit and
instructions are given by Sherony (16).
Calibration
Both the outlet and inlet optical cells were calibrated
for the Kerosene/water system. The same calibration was
used for 1/1 #30 oil/Kerosene dispersed in water and 1/1 ILS
oil/Kerosene dispersed in water.
An emulsion was prepared by adding known volume of
dispersed phase to the continuous phase. The solubility of
Kerosene in water at 75°F is approximately 20 x 10~6 volume
per volume of continuous phase. This volume was subtracted
from the total volume of dispersed phase added to give the
52
-------
TABLE 10
Physical Properties of the Materials Used
Temp.
Material
sem
#30
ILS
a
oil/Kerosene
oil/Kerosene
JL A-
dynes/cm.
37.
23.
17.
8
7
5
gm
0
0
0
3
./cm.
.816
.840
.860
»a
gm./cm.sec.
0.
0.
0.
0191
2320
3600
gm.
0
0
0
<_,
/cm. sec.
.0092
.0092
.0092
oF
75^2
75±2
15-2
CO
-------
Emulsion
in
Optical Cell
1.12 in. I.D.
and 6 in.long
Settler
Coalescer
Cell
Settler
Emulsion
Out
Photocell
Light Source
Aperture
1 in. dia.
To Photo-
comparator
FIGURE 37 LIGHT TRANSMISSION APPARATUS
-------
concentration of the dispersion. The emulsion was agitated
and recycled for about an hour before samples were with-
drawn from the tank and put into the two optical cells.
The corresponding voltage readings were taken on the recorder
connected with the photocomparator. Figure 38 and Table 11
show the calibration of light transmission apparatus for
different concentration of dispersed phase. The photo-
comparator was balanced before and after calibration and
during actual runs by filling the optical cells with pure
continuous phase and setting the potentiometer.
Operating Procedure
The tank was filled with 6 liters of the continuous
phase, and dispersed phase was added to bring the emulsion
to the required concentration. The emulsion was agitated
and recycled through the pump for about an hour. The
temperature of the emulsion was maintained at 75° - 2°F by
circulating water through the cooling coil. The run was
started at the maximum flow rate to drive out all the air
from the bed and the system. The flow rate was then
gradually decreased. In preliminary runs, the desired flow
rate was maintained for about an hour, readings of pressure
drop across the bed and inlet and outlet oil concentration
were taken. Since the calibration given in Figure 38 is
only good upto 230 ppm oil, higher concentrations were
measured by diluting the sample with pure water.
In most runs, because of the time-dependent behavior
of the pressure-drop, both pressure drop across the bed
and inlet and outlet oil concentration data were taken as
a function of time at various flow rates.
At the end of the run, the system was flushed with water
and the next run was started with fresh materials.
55
-------
en
cn
-P
rH
O
Kerosene dispersed
in water
Outlet Optical Cell
A Inlet Optical Cell
150 200
Oil Concentration, ppm
250
300
FIGURE 38 CALIBRATION OF LIGHT TRANSMISSION APPARATUS
-------
TABLE 11
Calibration of Light Transmission Apparatus
System: Kerosene dispersed in water
Solubility: 20 ppm at 75° - 2°F
concentration Inlet cell Outlet cell
ppm millivolts millivolts
5 0.040 0.060
10 0.070 0.110
25 0.175 0.200
55 0.405 0.420
75 0.462 0.542
100 0.560 0.640
130 0.662 0.740
180 0.730 0.820
23C 0.778 0.860
57
-------
SECTION VII
RESULTS OF SMALL COALESCER TESTS
Single Phase Flow Through Fibrous Beds
The pressure drop vs flow rate data were obtained for
single phase flow using pure water flowing through single
layer beds. The data for coated fibers (3.2y mean diameter)
and uncoated glass fibers (5.6y mean diameter) are plotted
in Figures 39 and 40 and are presented in Tables 12,13,14,15.
For the same bed conditions, the pressure drop across the
bed for the case of coated glass fibers is much higher than
for uncoated glass fibers for same flow conditions. A slight
time-dependent behavior of pressure drop was observed for
coated fiberglass fibers.
The data were found to follow Darcy's Law. The pressure
drop was proportional to flow rates except for 0.25 in,
coated fiberglass bed at higher flow rates as shown in Fig. 40,
ycvL
The permeabilities K ( r ) were computed from slopes of
the curves given in Figures 39 and 40, and are given in the
Tables 12,13,14,15. The permeability values for uncoated
glass fibers are the same as reported by Spielman (17).
However, the permeability values are much lower in the case
of coated fiberglass fibers for the same conditions.
Two Phase Flow Through Fibrous Beds
The coalescence and pressure drop data of the prelimi-
nary investigation are shown in Table 16. The coalescence
of Kerosene; 0.1/1, 0.5/1, and 1/1 #30 oil/Kerosene; and 1/1
ILS oil/Kerosene dispersed in water at superficial velocities
from 0.15 to 1.96 ft./min. with inlet oil concentration of
about 50 and 500 ppm. With coalescer layer of 5/16 or 5/64
in.length and about 11.6 Ib./ft. density, oil in the
effluent stream varied from 7 ppm to generally nondetectable
oil for all systems at all velocities investigated. The
coalescence of 1/1 ILS oil/Kerosene in water with 5/64 in.
59
-------
TABLE 12
Single Phase Flow Through Coated Fiberglass Fibers
Fiber Type: Coated fiberglass fibers
Mean Diameter: 3.2y
Void fraction: 0.94
Bed density: 8.7 lb./ft.3
Bed length: 0.13 in.
Permeability
Velocity
f t ,/min.
0.4
0.73
1.09
1.44
1.62
1.80
2.26
AP
psi
0.38
0.68
1.03
1.38
1.53
1.70
2.20
Reynolds ,_ y^vL .
number o Ap
(10 x cm )
0.018 0.024
0.012
0.010
0.009
0.008
0.006
0.004
60
-------
TABLE 13
Single Phase Flow Through Coated Fiberglass Fibers
Fiber Type: Coated fiberglass fibers
Mean diameter: 3.2 y
Void fraction: 0.90
Bed density: 15.8 lb./ft.3
Bed length: 0.25 in.
Permeability
Velocity
ft. /min.
0.40
0.73
1.09
1.44
1.62
1.80
2.26
Ap
pai
2.40
4.23
6.11
7.87
8.24
9.10
10.59
M V -I-J
Reynolds y ,_ c .
number o Ap
(10~6 x cm2)
0.008 0.007
0.014
0.020
0.027
0.031
0.037
0.043
61
-------
Ch
ro
in
O
Q
QJ
S-l
tn
0)
5.0
4.0
3.0
2.0
1.0
0.5
1.0
T
T
Bakelite coated fiber glass fibers (3.2y
Bed Density = 8.7 lb./ft.3
Bed Length = 0.13 in.
Uncoated glass fibers (5.6y)
Bed Density =8.7 lb./ft.3
Bed Length = 0.13 in.
Spielman's Data
1.5 2.0 2.5 3.0
Superficial Velocity, ft./rain.
4.0
FIGURE 39 PRESSURE DROP VS SUPERVICIAL VELOCITY FOR SINGLE PHASE FLOW (VOID
FRACTION 0.94)
-------
TABLE 14
Single Phase Flow Through Uncoated Glass Fibers
Fiber Type: Uncoated glass fibers
Mean fiber diameter: 5.6y
Void fraction: 0.94
Bed density: 8.8 lb./ft.3
Bed length: 0.13 in.
Permeability
Velocity
ft . /min .
0.40
0.73
1.09
1.44
1.70
1.80
1.95
2.26
3.29
Ap
psi
0.05
0.08
0.12
0.15
0.16
0.17
0.18
0.20
0.32
Reynolds
number
0.012
0.023
0.034
0.045
0.053
0.056
0.061
0.070
0.102
K (= -? )
o Ap
(10~6 x cm2)
0.186
63
-------
TABLE 15
Single Phase Flow Through Uncoated Glass Fibers
Fiber Type: Uncoated glass fibers
Mean fiber diameter: 5.6y
Void fraction: 0.90
Bed density: 15.8 lb./ft.3
Bed length: 0.25 in.
Permeability
velocity
ft. /min.
0.40
0,73
1.09
1.62
2.26
3.29
Ap
psi
0.12
0.23
0.33
0.50
0.71
1.04
ycvL
Reynolds K (= T )
number ,- ~
(10 D x cni )
0.012 0.148
0.023
0.034
0.050
0.070
0.102
-------
20.0
C_n
16.0
en
N,
Qi
O
Q
0)
CO
M
04
12.0
8.0
4.0
I 1
Bakelite coated fiber glass fibers (3.2y)
Bed Density =15.8 lb./ft.3
Bed Length = 0.25 in.
Uncoated glass fibers (5.6y)
Bed Density =15.8 lb./ft.3
Bed Length =0.25 in.
x
s
X
1.0
2.0 3.0
Superficial Velocity, ft./min.
4.0
FIGURE 40
PRESSURE DROP VS SUPERFICIAL VELOCITY FOR SINGLE PHASE FLOW (VOID
FRACTION 0.90)
-------
TABLE 16
Preliminary Investigation of Coalescer Perfoi
inan ~e
Dispersed
Phase
Kerosene
Kerosene
Bed
density
Ib/ft?
*
C.L.
**F.L.
*** g L
C.L.
F L.
S.L.
= 11 .7
= M.7
- 11.8
****
= 11.6
= 14.7
= 11.6
Pre-soaked with
Bed
Velocity Pressur
length
inches
C.L.
F.L.
S.L.
T.L.
C. L
F.L.
S.L.
T.L.
-5/32
= 1/8
= 5/32
- 7/16
. =5/64
= 1/16
- 5/64
= 7/32
ft- m in
1 96
1.09
0,48
0.15
1.96
1.09
0.48
0.15
High
Cone .
23.3
20.4
11 .0
9.2
21.5
15.8
8.1
3.9
e dicp psi iniet
Low
Cone .
30.0
28.2
15.8
6,0
30.0
20.8
11 .2
5.2
Cone . -p pm
rl'qii
512
512
512
512
450
450
450
450
Low
67
67
67
67
53
53
53
53
O ullet
High Low
U
0
0
0
30
17
6
17
0
0
0
7
0
1
0
3
dispersed phase
0.1/1 #30 oil/kero.
0.5/1 #30 oil/kero.
1/1 #30 oil/kero.
1/1 ILS oil/kero.
C.L.
F.L.
S.L.
C.L.
F.L.
S.L.
C.L.
F.L.
S.L.
C.L.
F.L.
S.L.
= 11.6
= 14.4
= 11.8
= 11.6
= 14.5
= 11 .5
- 11.7
= 14.4
- 11.8
= 11.8
= 14.5
= 11 .7
C.L.
F.L.
S.L.
T.L.
C.L.
F.L.
S.L.
T.L.
C.L.
F.L.
S.L.
T.L.
C.L.
F.L.
S.L.
T.L.
= 5/64
= 1/16
-5/64
= 7/32
= 5/64
= 1/16
= 5/64
= 7/32
= 5/64
= 1/16
= 5/64
= 7/32
= 5/64
= 1/16
= 5/64
= 7/32
1.96
1.09
0.48
0.15
1.96
1.09
0.48
0.15
1.96
1.09
0.48
0.15
1.96
1.09
0.48
0.15
18.0
12.2
7,0
4.8
15.3
10.0
5.2
2.5
16.0
10.8
5.8
3.0
17.8
14.6
9.8
5.0
30.0
1 1.2
5.0
24.8
15.2
8.0
3.2
23.3
14.0
6.9
2.9
24.8
16.0
8.2
3.8
510
51-0
51-0
510
495
495
495
495
460
460
460
460
570
570
570
570
48
48
48
48
60
60
60
60
68
68
68
68
65
65
65
65
0
0
0
0
3
1
0
1
1
2
1
3
5
6
3
4
L
2
1
0
1
0
0
0
1
0
0
0
0
0
0
1
-------
Table 16 continued
1/1 ILSoil/kero.
1/1 ILSoil/kero.
C.
F.
S.
C.
F.
S.
L.
L.
L.
L.
L.
L.
= 5.9
= 7.3
= 5.9
= 8.7
= 11.8
= 8.2
C.L.
F.L.
S.L.
T.L.
C.L.
F.L.
S.L.
T.L.
= 5/64
= 1/16
= 5/64
= 7/32
= 5/64
= 1/16
= 5/64
= 7/32
1.96
1.09
0.48
0.15
1.96
1.09
0.48
0.15
7.
5.
3.
2.
6.
5.
3.
"
0
2
5
0
9
5
8
"
480
480
480
480
470
470
470
470
70
91
130
62
35
22
_
-
-
-
_
01
* Coalescer layer
** Filter layer
*** Separator layer
**** Total length
-------
thick and 5.9 lb./ft.3 and 8.7 lb./ft.3 densities of
coalescer layer was found to be poor.
The pressure drop across the bed was found to be time
dependent. The data shown in Table 16 are taken at the end
of about 30 minutes.
The time-dependent behavior of pressure drop for Kerosene
dispersed in water with inlet oil concentration of about
100 ppm for four different velocities is shown in Figure 41.
The pressure drop continuously increases with time and does
not appear to reach steady state value. To check this, a
test was made under the similar conditions for a long period
of time. The data for the test are plotted in Figure 42.
In 810 minutes the pressure drop across the bed reached a
value of 44 psi at 1.95 ft./min. which was approximately
the maximum discharge pressure of the pump. The time-
dependent behavior for 1/1 ILS oil/Kerosene dispersed in
water is shown in Figure 43. The coalescence efficiency
was essentially 100%.
The rise in pressure drop across the bed with time
is due to increase in holdup of the dispersed phase. Re-
generation of the bed using water and methanol to remove the
accumulated oil was attempted. The pressure drop and coales-
cence performance data for Kerosene dispersed in water are
given in Table 17. The methanol regeneration gave almost
the similar pressure time characteristics as the fresh bed
but the coalescence performance was poorer than fresh bed.
Beds could not be regenerated by water.
Most of the data reported in the literature have been
taken using uncoated glass fibers. Tests were made to
evaluate both uncoated and coated glass fibers, the latter
coatings being commercially available silicone materials
that produce hydrophobic surfaces. The pressure drop and
coalescence performance data for Kerosene dispersed in water
are given in Table 18.
68
-------
25.0
01
04
o
Q
CD
cn
0)
S-l
20.0
15.0
10.0
5.0
50
System: Kerosene dispersed
in water ""
Fiber: Bakelite coated
fiber glass fibers(3.2y)
Bed Density: C.L. = 11.7
lb./ft.3
F.L. - 14.5
lb./ft.3
S.L. = 11.7
lb./ft.3
Bed Length: C.L. = 5/64in
F.L. = l/16ia
S.L. = 5/64in
Inlet oil cone.: - 100 ppm
100
150 200
Time, Minutes
250
300
FIGURE 41
TIME-DEPENDENT BEHAVIOR OF PRESSURE DROP AT DIFFERENT SUPERFICIAL
VELOCITIES FOR BAKELITE COATED FIBER GLASS FIBERS
-------
50.0
rH
tn
0,
O
M
Q
0)
M
3
U)
U)
0)
40.0
30.0
20.0
10.0
System: Kerosene Dispersed in water
Bakelite coated fiber glass fibers
3.2y)
Bed Density
11.7 Ib./ft.
11.8 lb./ft.
Bed Length
Uncoated glass fibers (5.6y)
Bed Density
11.7
14.4
11.7 lb./ft.
Bed Length
5/64
1/16 in.
5/64 in.
120
240
360
480
Time, Minutes
600
720
840
960
FIGURE 42
COMPARISON OF TIME-DEPENDENT BEHAVIOR OF PRESSURE DROP FOR COATED FIBER
GLASS AISiD UNCOATED GLASS FIBERS
-------
25. O
System: 1/1 ILS oil/Kerosene
dispersed in water
Fiber: Bakelite coated fiber
glass fibers (3.2y)
Bed Density
Bed Length;
C.L. = 11.8 Ib./ft.
F.L. - 11.0 Ib./ft.:
S.L. = 8.9 Ib./ft.
C.L. = 5/64 in.
F.L. = 1/16 in.
S.L. = 5/64 in.
Inlet oil cone.: - 100 ppm
100
150
Time, Minutes
200
250
3QQ
FIGURE 43
TIME-DEPENDENT BEHAVIOR OF PRESSURE DROP AT DIFFERENT VELOCITIES FOR
1/1 ILS OIL/KEROSENE DISPERSED IN WATER
-------
TABLE 17
Coalescer Performance With and Without Regeneration
for Coated Fiberglass Fibers
System: Kerosene dispersed in water
Bed density: *C.L
**F.L
= 11.6 lb./ft.
= 11.5 lb./ft.
***S.L.
Bed length:
Fresh Bed
C.L.
F.L.
S.L.
Time Velocity
min. ft
0
20
21
40
41
70
71
100
Regeneration
0
30
31
60
Regeneration
0
10
20
21
35
. /min.
1.95
1.95
0.48
0.48
1.09
1.09
0.15
0.15
with
1.95
1.95
1.09
1.09
with
1.95
1.95
1.95
1.09
1.09
- 9.2 lb./ft.
= 5/64 in.
- 1/16 in.
= 5/64 in.
Ap
psi
3.8
15.5
5.0
10.5
15.0
17.0
2.0
3.0
water at 1.09
20.5
25.0
12.8
16.5
methanol
3.0
12.0
15.4
9.0
12.0
Inlet oil
cone. , ppm
_
120
-
58
-
50
-
53
ft. /min. for
250
-
135
_
168
100
_
70
Outlet oil
cone. , ppm
_
0
-
1
-
1
-
1
about an hour
40
-
25
_
8
13
_
9
* C.L. = Coalescer Layer
** F.L. = Filter Layer
*** S.L. - Separator Layer
72
-------
CO
TABLE 18
Coalescer Performance with Uncoated Glass Fibers
and Uncoated Glass Fibers Treated with Various Hydrophobic Coatings
Time for Velocity
Fiber each run ft./min.
min.
(a) Bakelite coated, Single
coalescer layer cell 7/32 in.
thick & 12.2 lb./ft.3 bed
density
(b) Uncoated glass, single
coalescer layer cell, 7/32 in.
thick and 12.4 lb./ft.3 bed
density
(c) Uncoated glass, three layer
cell with inside coalescer layer
5/64 in. thick and 11.8 lb./ft.
bed density
(d) Uncoated glass coated with
L-45 "Silicone" (Union Carbide)
in benzene, three layer cell as in
(c) with 11.7 lb./ft.3 bed density
(e) Uncoated glass coated with 1%
Z-6040 "Silane" (Dow Corning) in
water three layer.,cell as in (c)
with 11.7 lb./ft. bed density
5
5
10
10
10
30
30
30
30
30
30
30
30
30
30
30
75
30
30
30
45
30
30
4.41
3.29
2.26
1.62
1.09
4.41
3.29
2.26
1.62
1.09
4.41
3.29
2.26
1.95
1.09
4.41
2.26
1.62
1.09
4.41
2.26
1.62
1.09
Pressure
drop
psi
42.0
43.0
40.0
35.0
22.5
6.2
4.9
4.0
3.4
3.2
5.8
4.5
3.4
3.4
2.2
7.0
9.5
10.0
9.5
6.5
8.8
10.2
10.8
Inlet
oil cone.
ppm
215
150
117
69
40
160
290
175
260
160
25
140
65
81
49
72
55
96
137
82
44
62
102
Outlet
oil cone.
ppm
2
1
1
1
0
8
8
10
7
8
12
17
12
15
9
17
12
14
15
15
11
13
12
(Continued)
-------
TABLE 18 (Continued;
Time for
each run
ruin .
(f) Uncoated glass coated with
0.3% Z-6020 "Silane" (Union
Carbide) in water, three layer .,
coll as in (c) with 11.7 Ib./ft.
bed density
(g) Uncoated glass, coated with 1%
A-172 "Vinyl Silane" (Union Carbide)
in water, three layer cell as in (c)
with 11.8 Ib./ft. bed density
(h) Uncoated glass, coated with
DC-200 in toluol, three layer _
cell as in (c) with 11.7 Ib./ft.
bed density.
30
40
50
30
30
30
40
30
30
30
Velocity
f t ./min.
4.41
2.26
1.62
1.09
4.41
2.26
1.62
1.09
1.62
1.09
Pressure
drop
psi
11.2
14.0
14.2
12.5
4.5
11.0
12.5
11.8
9.2
14.5
Inlet
oil cone.
ppm
93
65
59
117
44
160
87
220
59
140
Outlet
oil cone.
ppm
40
21
40
60
13
13
12
12
18
12
-------
The data show that practical pressure drops are obtained
with uncoated glass and similarly with uncoated glass treated
with different hydrophobic coatings. In all cases, the
coalescence performance is poorer than the commercially
available bakelite coated fiberglass even when using much
thicker coalescer layer.
The time-dependent behavior of the pressure drop was
also observed in case of uncoated glass fibers. The Table 19
gives time vs pressure drop data for bakelite coated and un-
coated glass fibers. These are plotted in Figure 42.
The pressure drop data on the uncoated glass fibers
indicates that there was initial slow rate of increase
followed by rapid rate of increase. The final hour in-
dicated no increase in pressure drop. The coalescer per-
formance appears to be increasing with increase in pressure
drop. However, the performance is still much poorer than
bakelite coated fiberglass which gives the same optimum
performance of 1 to 2 ppm at both low and high pressure
drops. The rate of increase of pressure drop in case of
bakelite coated fiberglass is higher than uncoated glass
fibers.
Good coalescence data requires a system free of solid
particles. The accumulation of these solid particles in
the bed changes the surface characteristics and hence the
pressure drop behavior and coalescence performance. Micro-
scopic examination of bed at the end of a run revealed the
presence of black particles probably from the rubber impeller
of the pump. Consequently, a new 3/4 HP centrifugal pump was
used to eliminate the possibility of any particles from
rubbing surfaces.
The data reported here and those reported in the litera-
ture indicate that the lower pressure drops are obtained with
uncoated glass fibers. The oil removal efficiency [1 - outlet
oil cone./inlet oil cone.] is however much lower. Tests
75
-------
TABLE 19
Time Versus Pressure Drop for Coated Fiberglass Fibers and Uncoated Glass Fibers
CTi
Fiber
(a) Bakelite coated fiber
glass, three layer cell with
inside coalescer layer 5/64 in.
thick and 11.7 lb./ft.3
bed density
(b) Uncoated glass, three layer
cell with inside coalescer layer
5/64 in. thick and 11.7 lb./ft.3
bed density
Time
min .
6
15
30
45
180
240
300
360
480
600
750
810
15
45
60
120
180
270
300
360
velocity Pressure
ft. /min. drop
psi
1.95 10.0
13.8
17.0
18.8
25.8
29.2
30.7
32.0
35.4
39.5
43.5
44.0
1.95 1.2
1.8
2.4
6.8
12.8
17.8
19.0
19.0
Inlet
oil cone.
ppm
96
-
63
84
-
-
62
-
-
-
117
50
89
65
50
77
50
-
Outlet
oil cone.
ppm
1
-
2
2
-
-
2
-
-
-
48
27
25
20
12
12
12
-
-------
made with uncoated 8y glass fibers of uniform size (which
was used by Sherony Cl6) ) gave steady state pressure drop
shown in Figure 44, however oil removal efficiency was only
10 to 15%. The bakelite coated fiber glass, which exhibit
higher and time dependent pressure drops, give oil removal
efficiencies between 90 and 99%. The data taken in this
study and given in Table 20 permits the following ob-
servation. The oil removal efficiency for bakelite coated
fibers is more than 90%. The lower efficiency values given
in Table 20 are due to the rapid change in inlet oil con-
centration (outlet oil concentration is almost constant)
with time. The change in inlet oil concentration is due
to the oil holdup which occurs in a recirculating system.
Consequently a once-through system was set up. The data
for Kerosene dispersed in water with once through system is
given in Table 21. The data reveal that the inlet oil
concentration is almost constant and oil removal efficiency
is above 90%. The slight increase in outlet oil con-
centration with time is due to air bubbles passing through
the light transmission apparatus. The pressure drop vs.
time plot for both recirculating and once-through system
shown in Figure 44 indicates similar time-dependent behavior
of pressure across coated fiber glass beds.
77
-------
25.0
--4
CO
T~ 1
System: Kerosene dispersed in water
Recirculating System
Fiber: Bakelite coated fiber glass
fibers (3.2y) _
Bed Density: 70.3 Ib./ft.
(single coalescer layer)
Bed Length: 0.125 in.
Once-Through System
Fiber: Uncoated glass fiber
Bed Density: 10.30 Ib./ft.
(single coalescer
layer)
Bed Length: 0.125 in.
Recirculating System
Fiber: Uncoated Glass (8y
Bed Density: 15.8 Ib./ft.
Bed Length: 0.25 in.
50
100
150
Time, Minutes
200
250
300
FIGURE 44 COMPARISON OF TIME-DEPENDENT BEHAVIOR OF PRESSURE DROP FOR ONCE-THROUGH
AND RECIRCULATING SYSTEMS
-------
TABLE 20
Coalescer Performance Data for Recirculating System
System: Kerosene dispersed in water
Fiber: coated fiber glass
Bed density: 10.3 lb./ft.3
Bed length: 0.125 in.
Time Velocity
min. ft./min.
0 1.95
23
30
35
40
*
50
55
60
62 1.09
65
70
85
90
Ap
psi
3.0
6.6
7.6
8.2
8.6
-
9.5
10.2
10.4
6.4
-
8.0
8.9
9.3
Inlet
oil cone.
ppm
-
57
31
20
11
-
93
76
52
-
48
40
17
10
Outlet
oil cone.
ppm
-
6
4
4
4
-
6
7
7
-
5
5
3
3
oil removal
efficiency
-
0.89
0.87
0.76
0.65
-
0.93
0.90
0.86
-
0.89
0.88
0.82
0.70
* Flow of emulsion was stopped through the cell, 2 ml. of
Kerosene was added and emulsified for about 1 hr. It
was then passed through the coalescer cell.
79
-------
TABLE 21
Coalescer Performance Data for Once-Through System
System: Kerosene dispersed in water
Fiber: coated fiber glass
Bed density: 10.3 lb./ft.3
Bed length: 0.125 in.
Time
min.
5
10
16
20
30
50
60
75
90
120
150
165
Velocity Ap
ft. /min. psi
1.09 3.48
-
4.93
5.35
6.34
7.91
8.82
9.99
11.15
13.43
15.39
16.06
Inlet
oil cone.
ppm
_
175
-
175
175
175
172
170
168
165
162
160
Outlet
oil cone.
ppm
_
1
-
0
8
8
9
9
10
11
12
14
oil removal
efficiency
_
0.99
-
1.00
0.96
0.96
0.95
0.95
0.94
0.93
0.93
0.91
80
-------
SECTION VIII
MICROSCOPIC STUDY OF FIBROUS BED COALESCER
The qualitative analysis of coalescence phenomena in a
fibrous bed was carried out by observing the bed in operation,
The observations were made under a microscope and recorded
by high speed motion picture photography.
The optical system for taking pictures through micro-
scope is shown in Figure 45, and has been described by
Kintner et al. (18). The approach of the oil drops (average
size 10y) to the fiber and the coalescence of the drops
are shown in Figure 46.
It was observed that coalescence occured in two ways.
The most common type of coalescence was that between a drop
in the main stream and a drop attached to a fiber when
collision occured. The second type of coalescence resulted
when two drops were in contact on a fiber. In all systems
investigated it was possible to find drops attached to
fibers. Systems which showed a high population density of
drops attached to fibers, coalescence frequency with
attached drops was high.
From the above observations it is believed that the
primary mechanism of coalescence in fibrous beds involves
attached drops, and that the three phase systems in which
there are relatively few drops of secondary emulsion
attached to the fibers will be systems with poor coalescing
effectiveness.
81
-------
H
UsU
B
CD
A. Bausch & Lomb illuminator,
6-volt, 18-ampere lamp
B. Rack for heat absorber and
light filters
C. Condensing lens system
D. Substage condenser and iris
diaphragm
E. Flow cell containing fibers
F. Microscope optical system
G. Light trap
H. Fastax WF3 camera
FIGURE 45 APPARATUS FOR HIGH SPEED CINEPHOTOMICROGRAPHY
-------
FIGURE 46. MICROSCOPIC PICTURES OF OIL DROPS COALESCING IN A
FIBROUS BED.
-------
STABILIZATION TESTS
Preliminary tests on stabilizing the 3.2y coalescing
fibers were made by treating them with 10% isobutyl metha-
crylate resin dissolved in methylene chloride. This material
was used because of availability. Information and samples
of phenol formaldehyde resins which could be used have not
been obtained as yet after two months of effort.
The following technique of applying the resin should be
directly applicable to forming the diaphragm in situ in
the large cell. The 3.2y glass, after being saturated with
the solution of 10% resin in methylene chloride, is com-
pressed between 2 layers of 1 in. 10. ly glass on each side to
1/4 in. The two end layers of 10.ly glass were then replaced
with fresh 10.ly glass, compressed to 1/4 in., clamped, and
dried at 100°C for 1 hour. The final assembly of 2/1/2 in.
of uncompressed 10.ly/3.2y/10.ly glass corresponds to the
diaphragm assembly used in Runs 2 thru 7.
The diaphragms were evaluated in the test unit shown in
Fig. 47. The fiber bed(7_)is maintained at 1/4 in. spacing by
means of spacer cylinders(\p, between perforated disk(jy. The
disk is provided with 1/2 in. opening in the holder(j). The
spacer bars(T)are bolted(2)between the saddles Qjwhich are
threaded to(^ The periphereal seals are provided by the 0-ring
(?) placed in very shallow grooves. This construction insures
that any leakage thru the entrant layers of 10.ly glass can
not bypass to the exit side and contaminate the effluent.
The flow area is only nominally 1/2 in.
Stabilization Tests^-Results
The results of the tests on the stabilized fibers are
shown in Table 22. The same performance and pressure drop
as in the large cell are obtained. Treatment of the
diaphragm with hexane after 20 hours resulted in a regenerated
85
-------
©
Spacer Cylinder
Tie Bolt
Saddle Clamp
Inlet Port
Qj) Perforated Disks
© 0-Rings
(7) Fiberous Bed Test Section
FIGURE 47 ASSEMBLY FOR STABILIZATION TESTS
86
-------
TABLE 22
Regeneration Tests on Stabilized Fibers
Flow
Rate
Time Velocity QPM
hours fpm f^
0
.45
.53
.53
.62
.78
.98
1.02
1.06
1.20
1.28
1.50
1.75
2.00
2.50
3.00
3.50
4.00
4.50
5.00
3.1
6.2
.3
.8
1.2
.7
.7
.9
1.1
1.0
1.0
.9
.9
.6
.9
.8
.9
.9
.7
.8
22.50
46.40
2.24
5.98
8.98
5.24
5.24
6.74
8.23
7.48
7.48
6.74
6.74
4.49
6.74
5.98
6.74
6.74
5.24
5.98
Pressure
Temp . Drop
°F psi
_
-
-
-
96
92
90
90
86
86
85
84
84
85
86
84
84
85
87
85
1.93
1.93
0.97
1.93
2.90
-
3.62
4.39
5.93
6.03
6.40
6.90
8.10
9.09
10.50
12.00
13.50
14.50
16.00
18.00
Oil T
Influent
24.0
_
_
10.0
17.0
16.0
14.0
12.0
20.0
18.0
20.0
19.0
21.0
20.0
18.0
U*
Effluent
3.00
1.50
.85
.75
.75
.95
_
1.10
1.20
.95
.87
.73
.75
.82
.75
.73
.80
.83
.92
SHUT DOWN OVER NITE
5.00
5.25
0
.02
.09
.17
.25
.33
.50
.75
1.00
1.25
1.75
2.25
2.75
3.25
3.75
3.75
5.00
5.75
6.75
7.75
.6
.5
-
2.1
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
4.49
3.74
Remove
15.70
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
7.48
-
-
Diaphragm,
-
-
-
78
-
78
78
80
80
80
79
79
79
78
78
79
79
80
78
78
18.00
18.00
washed
-
-
2.50
1.70
2.20
2.70
3.30
3.50
3.90
4.10
4.20
4.20
4.20
4.20
4.20
4.50
4.80
4.90
5.00
5.00
12.0
-
with Hexane
33.0
22.0
-
-
-
25.0
18.0
17.0
21.0
19.0
19.0
21.0
20.0
18.0
18.0
18.0
18.0
20.0
20.0
21.0
1.20
-
-
-
.75
1.50
1.80
2.00
2.40
2.75
3.40
3.60
3.60
3.50
3.60
3.10
3.10
2.80
1.50
1.20
0.86
0.75
87
-------
TABLE 22 (Continued)
8.75
9.75
10.75
11.75
20.25
20.50
24.00
28.00
1.0
1.0
1.0
1.0
0.5
1.0
1.0
1.0
7.48
7.48
7.48
7.48
3.74
7.48
7.48
7.48
80
78
79
80
-
79
81
82
5.50
5.80
6.00
6.30
6.00
11.40
12.50
18.50
22.0
22.0
22.0
21.0
-
27.0
29.0
30.0
0.63
0.56
0.50
0.52
-
0.38
0.48
0.50
Regeneration with Water
0 0.03 0.23 58 17.50
1.75 0.03 0.23 - 17.50
Treated with Hexane without drying
0 0.0 0.00 61 17.50
Regenerated with Methanol
2
3
3
4
0
.00
.50
.75
.30
0
4
3
-
-
.70
.50
.00
5
33
22
-
-
.24
.60
.4
61
60
60
60
60
3
4
14
14
- -
.50
.00
.00
.50
-
-
"
By Hach turbidimeter
-------
diaphragm. After the second AP buildup the diaphragm was
regenerated with methanol. These preliminary tests on
regeneration involved removing the diaphragm from this cell
rather than in situ treatment.
Stabilization Tests-Discussion
The present results are very preliminary but indicate
that the 3.2y fiber bed can be stabilized by cementing
together with a resin. Furthermore, using a methacrylate
resin which contains numerous polar groups, the coalescence
efficiency remains the same 100% as the original commerically
available fibers with some type of bakelite coating.
There are numerous possibilities open in this type of
stabilization work. The most desirable would be to use such
a resin that is more oleophilic or non polar in nature. This
might result in the pressure drop building up to some constant
value so that regeneration is not necessary. Regeneration
techniques should involve in situ methods. This should be
no problem. In addition, regeneration using steam or air
rather than solvents would have considerable advantages.
89
-------
SECTION X
ACKNOWLEDGEMENT
This work was carried out at the Illinois Institute of
Technology under the direction of Dr. W.M. Langdon and
Dr. D.T. Wasan. The bulk of the experimental work was
carried out by Mr. P. Naik. Messrs. K. Nelson, G. Ramsey,
S. Browning, L. Ksycki, L. Gupta and M.B. Ranade also
participated in the experimental program. Photographs of
equipment were taken by Mr. Brian Swanson.
Industrial effluent material for testing was supplied
by Interlake Steel Corporation, Riverdale, Illinois. The
large scale cell was supplied by Ickes Braun Glass Company,
Aplakisic, Illinois. Teflon screen material for the second
stage was supplied by General Steel Tank Company of Reedsville,
North Carolina.
We wish to acknowledge the many helpful suggestions
provided by Dr. R.C. Kintner and Dr. B.S. Swanson, Illinois
Institute of Technology.
This project was supported by the Environmental Protection
Agency, Grant 12050 DRC and the advice and assistance of
Mr. E.L. Dulaney, Washington Office, and Mr. Clifford Risley,
Jr., Chicago Office, are hereby acknowledged.
91
-------
SECTION XI
REFERENCES
1. Langdon, W.M. and Wasan, D.T., Report, March, 1970,
FWPCA, Grant 12050 DRC.
2. Bhutan!, S.K., M.S. Thesis, 111. Inst. of Tech.,
Chicago, 111., 1967.
3. Chandak, S.S., M.S. Thesis, 111. Inst. of Tech.,
Chicago, 111., 1967.
4. Davis, B.C., U.S.P., 1, 535, 768, April, 1925.
5. Gudesen, R.C., M.S. Thesis, 111. Inst. of Tech.,
Chicago, 111., 1964.
6. Kyan, C.P., M.S. Thesis, 111. Inst. of Tech.,
Chicago, 111., 1964.
7. Malhotra, R.K., M.S. Thesis, 111. Inst. of Tech.,
Chicago, 111., 1969.
8. Patel, J.G., M.S. Thesis, 111. Inst. of Tech.,
Chicago, 111., 1970.
9. Rose, P.M., M.S. Thesis, 111. Inst. of Tech., Chicago,
111., 1963.
10. Sareen, S.S., Rose, P.M., Gudesen, R.C., Kintner, R.C.,
A.I.Ch.E. J., 12, 1045 (1966).
11. Sherony, D.F., M.S. Thesis, 111. Inst. of Tech.,
Chicago, 111., 1967.
12. Sherony, D.F., Ph.D. Thesis, 111. Inst. of Tech.,
Chicago, 111., 1969.
13. Spielman, L.A., Ph.D. Thesis, Univ. of Calif., Berkeley
Calif., 1968.
14. Tan, C.B., M.S. Thesis, 111. Inst. of Tech., Chicago,
111., 1968.
15. Tyebjee, T.T., M.S. Thesis, 111. Inst. of Tech.,
Chicago, 111., 1969.
93
-------
16. Sherony, D.F., Ph.D. Thesis, 111. Inst. of Tech.,
Chicago, 111., 1969.
17. Spielman, L.A., Ph.D. Thesis, Univ. of Calif.,
Berkeley, Calif., 1968.
18. Sherony, D.F., Tan, C.B., and Kintner, R.C., I&EC
Fundamentals, 6, 592 (1967).
94
-------
SECTION XII
GLOSSARY
Coalescence Efficiency - Percentage of inlet oil removed
by coalescence.
Diaphragm - Active section of the coalescer bed.
FPT - Female pipe thread.
Holdup - The volume of dispersed phase retained by the bed
per volume of fibers.
ILS - Interlake Steel Corporation.
IPS - International pipe size
Oil Removal Efficiency - Percentage of inlet oil removed by
the coalescer bed.
TFE - Teflon.
TU - Turbidity unit.
95
-------
K Permeability, cm.
SECTION XIII
SYMBOLS
2
V Superficial velocity, cm./sec.
L Bed length, cm.
-4
y Unit of length, micron (1 micron = 1 x 10 cm.)
y Viscosity of the continuous phase, gin./cm. sec.
\^f
y. Viscosity of the dispersed phase, gm./cm.sec.
p Density of the dispersed phase, gm./cm.
a Interfacial tension, dynes/cm.
97
-------
1
Accession Number
w
5
2
Organization
Department of Chemical
Subject Field & Group
05D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Engineering
Illinois Institute of Technology
Chicago, Illinois 60616
Title
Experimental Evaluation of Fibrous Bed Coalescers for Separating Oil-Water
Emulsions
1 Q Authors)
W.M. Langdon and
D.T. Wasan
16
21
Project Designation
Project No. 12050
me
Note
22
Citation
23
Descriptors (Starred First)
^-Dispersed Oil Removal from Water
25
Identifiers (Starred First)
-*Fibrous Bed Coalescers, -x-phenol Formaldehyde Coated Fiber Glass Fibers, Teflon
Coated Screen, Kerosene, #30 Lubricating Oil
27 Abstract & i sq. ft. coalescer unit using filter press construction has been designed
for removing dispersed oil from water and tested on both a synthetic stream and on an
actual pollutant stream. The oil removal efficiency was essentially 100$ at a superficial
velocity of 1 fpm. The pressure drop increased from 3 to 25 psi over run times which
varied from 13 hrs. to 294 hrs., due to both accumulation of oil in the bed and mechanical
degradation of the fibers. Preliminary tests indicated that the bed degradation phenomenon
could be eliminated by structurally stabilizing the compressed fibers with methacrylate
resin. Such fiber beds could be regenerated by various solvent treatments and reused.
The performance of fiber glass coalescers was studied in depth using a cell with an
active area of 1.77 sq. in. The commercial fibers, with phenol formaldehyde coatings and
a fiber diameter of 3.2*, gave efficiencies of 90-99$ with bed densities of 12 lb./ft.3
when operating at superficial velocities from 0.2 to 4 fpn on emulsions containing 50-
500 ppm of oil. The present design is suitable for large scale operation by the use of
both multiple cells and larger individual cells. It is estimated that operating costs on
the order of 0.13 $/!03 gal. are involved for the worst case of single use of fiber. If
the fibers can be regenerated operating costs would be reduced to 0.01 $/10~i gal.
This report was submitted in fulfillment of Grant No. 12050 DEC between the
Environmental Protection Agency and the Illinois Institute of Technology.
Abstractor
W. M. Langdon
Institution
Illinois Institute of Technology
WR:102 (REV JULY 1969)
WRSI C
SEND. WITH COPY OF DOCUMENT, TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 20240
UU.S. GOVERNMENT PRINTING OFFICE: 1972-484-484/86 1-
------- |