WATER POLLUTION CONTROL RESEARCH SERIES • 17030 FWH 01/72
Filtration of Municipal Waste
with a Moving Bed Contactor
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, development, and demonstration
activities in the water research program of 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 Chief, Publications Branch
(Water), Research Information Division, R&M, Environmental
Protection Agency, Washington, D.C. 20460.
-------
FILTRATION OF MUNICIPAL
WASTE WITH A MOVING BED
CONTACTOR
by
F. 0. Mixon
Research Triangle Institute
Environmental Studies Center
Research Triangle Park
North Carolina 27709
for the
Office of Research and Monitoring
ENVIRONMENTAL PROTECTION AGENCY
Project #17030 FWH
Contract #14-12-895
January 1972
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C., 20402 - Price 60 cents
-------
EPA Review Notice
This report has been reviewed by the Environmental
Protection Agency and approved for publication.
Approval does not signify that the contents nec-
essarily 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 novel moving bed contactor has been utilized in filtration studies
of municipal waste at various stages within a trickling filter plant.
Granular, buoyant filter medium is slurried with process feed and in-
troduced to the bottom of a column equipped with lateral retaining
screens and filter medium harvesting machinery, both positioned toward
the top of the column. Within the column, filter medium rises by
buoyancy and forms a porous plug that traps suspended solids from the
feed stream. Filtered liquid is removed from the lateral screen, and
soiled filter medium is continuously removed from the column top, washed,
and recycled to the column bottom.
The process operates stably and dependably on all feeds tested—raw
wastewater, primary clarifier effluent, and trickling filter effluent.
Suspended solids removals of 60 to 80 percent can be achieved at column
loadings up to 7.5 gal./min/sq ft. Filtration of alum-coagulated feed
is less effective than that of untreated feed.
This report was submitted in fulfillment of Project Number 17030 FWH,
Contract Number 14-12-895, under the sponsorship of the Office of Re-
search and Monitoring, Environmental Protection Agency.
ill
-------
CONTENTS
Section
I Conclusions
II Recommendations
III Introduction
IV Design, Construction, and Fabrication
V Experimental Phase
VI Results
VII Related Studies
VIII Discussion and Cost Estimates
IX Acknowledgments
X References
XI Patent Status
XII Appendix
Page
1
3
5
9
15
23
29
35
39
41
43
45
-------
FIGURES
Page
1 Schematic Diagram—Moving Bed Filtration Process 10
2 Column—Detailed Drawing 11
3 Screen Section—Detailed Drawing 12
4 The Filtration Process 17
5 The Scraper and Bead Washer 18
6 Schematic Diagram—Chapel Hill, N. C., Waste
Treatment Plant 21
7 Electrophoretic Mobility of Alum-Treated Polystyrene
Particles 30
8 Proposed Combined Filtration-Incineration Process 34
VI
-------
TABLES
Page
1 Performance Summary, Moving Bed Filter—
Suspended Solids Removal 24
2 Performance Summary, Moving Bed Filter—
COD Removal 25
3 Performance Summary, Moving Bed Filter—
TOG Removal 26
4 Itemized Equipment Cost 35
5 MBC Process Operating Costs 37
6 Cost Comparisons 37
vii
-------
SECTION I
CONCLUSIONS
1. The moving bed process is mechanically stable, reliable, and de-
pendable.
2. With expanded polystyrene as the filter medium, pretreatment of
the filter medium with alum is necessary in order to get satis-
factory removal of suspended solids.
3. Suspended solids removals of from 50 to 80 percent were achieved
when processing trickling filter effluent, primary clarifier ef-
fluent, and raw wastewater at rates up to about 7 gal./min/sq ft.
4. Polystyrene has a negative surface charge and would be expected
to repel negatively-charged particulate matter.
5. Perlite, strengthened with some suitable coating, could be an
attractive filter medium.
-------
SECTION II
RECOMMENDATIONS
1. It is recommended that other filter media be evaluated prior to
commercial development.
2. Incineration of sludge directly from the media warrants investi-
gation.
3. Utilization of the moving bed unit for ion exchange or carbon
treatment should be studied.
4. Large-scale studies of the applicability of the continuous filter
as a replacement for more conventional solids removal techniques
should be undertaken.
-------
SECTION III
INTRODUCTION
Background
For a number of years the Environmental Studies Center of the Research
Triangle Institute has been actively participating in the development
and advancement of technology for the desalting of sea water. An im-
portant piece of equipment which has been developed in connention with
freezing processes for desalting has been the hydraulic piston wash
separator. Within the context of freeze desalting the wash separator
is utilized to separate ice crystals from the mother brine liquor and
to wash the brine from the surface of the ice crystals so separated.
A wash column as normally used for slurry separation is simply a column
whose upper section is equipped with lateral screens of such a mesh as
to be freely permeable to the passage of liquid but be impermeable to
the passage of ice crystals. A mechanical device for harvesting ice
crystals is located at the top of the column. In operation ice brine
slurry is pumped to the bottom of the column and brine flows up the
column and out through the laterally positioned filter screens. The
ice crystals are left behind to form a porous plug which ultimately
fills the entire column cross section. Once completely formed, the
plug is propelled upward by the pressure drop of the brine flow through
its interstices but is simultaneously replenished at its lower surface
by crystal deposition from the slurry. At the top of the column the
plug is harvested with some suitable device, melted, and a portion of
the product water is recycled to the top of the emerging plug as wash
water. This wash water drains downward through the plug by gravity
and joins the brine discharging at the filter screens.
Although wash columns as used in desalination have been primarily for
slurry separation, the concept of a moving bed solid-liquid contactor
constitutes an important bit of fallout technology from the desalination
program which can be of value in other areas, particularly in the areas
of waste treatment and water pollution control. To derive maximum bene-
fits from the technology transfer, it is, however, necessary to recog-
nize that the device should be considered in more general terms as a
moving bed solid-liquid contactor rather than simply as a slurry sepa-
rator or as a wash column.
About four years ago the Research Triangle Institute recognized, among
other things, the possibility of utilizing the moving bed contactor as
a device for removing suspended solids from industrial or municipal
waste streams. As utilized for the filtration function, the moving bed
contactor operates as follows: Granular filter medium is slurried with
the incoming process feed which is then fed to the bottom of the moving
bed contactor. Once inside the column, filter medium rises because of
both buoyancy and viscous drag to the bottom of the preformed porous
plug of filter medium where it is deposited and thus replenishes the
porous plug. The water is forced through the plug and is discharged by
-------
gravity at the lateral retaining screens, simultaneously undergoing fil-
tration and forcing the plug slowly upward by both drag and buoyancy ef-
fects. At the top of the column the emerging porous plug is harvested
with a mechanical scraper that discharges filter medium into an auxil-
iary tank from which it is washed and recycled to the process.
The moving bed contactor utilized as a filter for suspended solids in
waste treatment should compete with, and perhaps replace, the following
techniques currently in practice: (1) clarification in which one is
forced to depend solely on gravity for the removal of suspended solids;
(2) fixed bed filtration which offers the disadvantage of requiring
periodic backwashing as well as provisions for handling the process flow
during the backwashing operation; (3) microscreening which appears to be
limited by practical considerations to the polishing of streams contain-
ing relatively small amounts of suspended solids; and (4), moving bed
filters of countercurrent or crosscurrent design that depend primarily
on mechanical means for propelling the moving bed.
Compared to these techniques, the moving bed contactor used as a filter
offers a number of advantages; among them are the following:
(1) Simplicity. The unit and associated equipment are quite simple.
The column itself can be constructed of virtually any material
capable of containing wastewater at low pressures, and the only
mechanical appurtenance required is the overhead scraper assembly.
This innate simplicity means, of course, low initial cost and low
maintenance.
(2) Applicability. The filter can handle suspended solids concentra-
tions as high as those of raw municipal wastewater and as low as
those of secondary effluents. Plugging is alleviated simply by
increasing the flow rate of filter medium through the filter col-
umn.
(3) Continuity. The fact that the unit operates continuously with-
out interruption for backwashing simplifies both materials han-
dling and process automation.
(A) Stability. The filtration unit has a type of inherent head-loss
stability in response to fluctuations in both feed rate and feed
solids concentration. For example, should the amount of liquid
to be handled suddenly increase, there will be an accompanying
tendency for increase in the pressure drop across the porous plug,
with the consequent result that the plug will rise more rapidly
and thus compensate for the increased flow. Similar behavior
should result from an increase in solids concentration in the
feed. In this circumstance the tendency toward higher pressure
drop would be caused by the permeability drop due to solids depo-
sition within the filter. Again the plug will rise and be re-
plenished more rapidly.
-------
(5) Analogy to graded filter media. The fact that the moving bed
of filter medium moves co-currently with the water stream under-
going filtration causes the filtration process to simulate that
achieved in fixed-bed batch filtrations with graded media fil-
ters. In the moving bed unit, clean filter medium is deposited
at the bottom of the moving plug and accumulates increasing solids
deposits as it moves upward through the bed. Thus, there exists,
in steady operation, a spatial variation in solids deposition
ranging from a minimum at the bottom of the plug to a maximum
at the top of the plug. The permeability varies in inverse fash-
ion, being lower at the top of the plug and increasing as one
proceeds downward. Achieving this behavior is the rationale
behind graded-media filters, since to increase production between
backwashes, it is desirable to filter first through coarse, high-
permeability material. Filtration in this sequence offers the
advantage of more efficient removal of suspended solids than does
filtration through a medium of uniform porosity (1).
Purpose
The purpose of the work reported herein has been to conduct an experi-
mental evaluation of the moving bed contactor (MBC) adapted as a filter
for removing suspended solids from municipal waste streams.
Scope and Program
The project as planned and conducted has involved the construction and
debugging of a moving bed contactor of sufficient size to be representa-
tive of municipal-scale waste treatment equipment. The solids removal
capabilities of the unit were evaluated over the range of flow rates
from 2 to 10 gpm per square foot, utilizing as feed to the unit process
streams taken from various points within a typical trickling filter mu-
nicipal waste treatment plant. In particular, the performance of the
unit has been studied when filtering trickling filter effluent, primary
clarifier effluent, and raw wastewater.
It was originally intended to study solids removal using a single filter
medium, i.e., expanded polystyrene. However, for reasons discussed in
subsequent sections of the report, it was found necessary to use a mix-
ture of about equal parts by volume of expanded polystyrene along with
low density polyethylene blow molding resin.
Within the limitations of project timing and funding, it was also planned
to study the filtration of alum-coagulated trickling filter effluent,
primary clarifier effluent, and raw sewage.
Provision was made to vary the depth of the filter medium from about 1
to 6 feet.
-------
SECTION IV
DESIGN, CONSTRUCTION, AND FABRICATION
A schematic diagram of the moving bed contactor (MBC) as adapted for
the filtration operation is shown in Figure 1. The integrated process
can be broken down into the following sections which will be illustrated
and described individually: the column, the screens, the scraper sec-
tion, the bead washing section, the sludge settling section, and the
bead recycle section.
The Column
A detailed drawing of the column is shown in Figure 2. The column is
2 feet in diameter with an 8 foot straight section equipped with a 3-inch
flanged section at the top and a conical bottom with a 2-inch standard
flanged pipe fitting. Translucent fiberglass, manufactured by Justin
Enterprises, Fairfield, Ohio, was chosen as the material of construction
for the column with the object of providing for visual observation of
the bottom of the bed of porous material. The column had liquid-level
strips painted at 90-degree intervals around its circumference for easy
measurement of the bed depth.
As it turned out in practice, the lower level of the porous plug could
be observed through the column only on certain occasions, i.e., after
the addition to the bottom of the plug of clean, white filter medium or
during rapid motion of the filter medium associated with draining the
column. Thus, the fiberglass was only marginally adequate for viewing
events within the column. It proved, however, to be quite satisfactory
from the standpoint of its corrosion resistance and mechanical strength.
The Screens
A detailed drawing of the screen section is presented in Figure 3. This
piece of equipment was custom fabricated for the moving bed contactor by
the Wedgewire Corporation of Wellington, Ohio, from their Wedgewire
"Kleenslot" screens. The screen section was constructed from Type 403
stainless steel which proved to be entirely adequate for corrosion re-
sistance. Slot spacing was specified at 0.020 inches. For this design
the open area of standard Wedgewire screen is such that the hydraulic
resistance of the screen is negligible compared to that of the particles
retained by the screen. This piece of equipment proved entirely ade-
quate and caused no difficulty at all throughout the duration of the
project.
The Scraper Section
The scraper section was fabricated utilizing the screw flights of two
9-inch diameter by 5-feet long screw conveyors. These screw flights
were mounted in parallel within a steel shroud designed to fit on top of
the column. The two screw flights were connected with No. 40 chain and
6-inch sprockets so as to run simultaneously and at the same speed.
-------
Effluent*
Influent
Feed
Drum
Overflow
Bead Washing Section
4CU
Sludge Settling Tank
Bead Recycle Tank
Col
Sludge
X—I X
Feed
Pump
Drain
Wa«h
Pump
Figure 1. Schematic Diagram—Moving Bed Filtration Process
-------
H— 2'6"
Flange
o
oo
2'0"
315
225
180'
Liquid Level
Strip
45°
90*
Liquid Level
Strip
135°
Liquid Level
Strip
Figure 2. Colu»n~Detailed Drawing
11
-------
3/8" x 1/2" Ear 11-1/4" Long
Spaced at 60*
3/8" Holes, Top Plate i, Bottom
Plate, Spaced at 60°
3/8"
1-1/4" Nipple-
Standard Pipe — v
Threads \
pi
«
"7T"~"~7
• •
• • j
_^_»«
—
U- 6" — *
1 , 1 0" A
1
1
f
r
I
3/8"
*-2"
Figure 3. Screen Section—Detailed Drawing
-------
They were driven with a one-quarter horsepower, 6 rpm gear motor ar-
ranged to produce a rotational speed of the screw flights of about 2 rpm.
This apparatus proved to be satisfactory for removing the solids laden
filter medium from the top of the column prior to washing.
The Bead Washing Section
From the scraper section, the filter medium was discharged by gravity
into a 2-foot diameter funnel which served as the bead washing section.
The underflow of the funnel was directed into a tank used for sludge
settling and bead separation. From this tank, the sludge settling
tank, liquid was recycled with a small centrifugal pump (about 500 gal-
lons per hour) to the bead washing funnel to serve as wash liquid.
This liquid was discharged into the bead washing funnel in a tangential
fashion with the washing action being much like that in a dentist's
spittoon. Filtered solids were washed from the beads by the liquid
turbulence. Throughout the program there was no indication that uti-
lizing dirty water as the bead washing liquid was disadvantageous.
The Sludge Settling Section
From the bead washing funnel the beads, together with the solids-laden
wash water, were discharged into the sludge settling tank, a tank 2 feet
by 1.5 feet by 3 feet deep having one side cut to serve as an overflow
weir maintaining a liquid depth of 2 feet in the tank. The tank was
equipped with internal baffles to prevent the turbulence generated by
the small centrifugal pump from reaching to the bottom of the tank and
to the vicinity of the sludge discharge. During operation sludge would
settle to the bottom of the tank and be discharged periodically to waste
through a timed solenoid valve, and laundered floating beads would be
discharged at the overflow weir for recycling to the column.
As can be seen from the MBC process schematic, Figure 1, two streams
flow into the sludge settling tank: beads plus wash water from the
bead washing section and a flow coming from the by-pass pump. The pur-
pose of the latter flow is twofold: first, to refill the tank after
periodic sludge withdrawal and second, to impinge upon the floating
beads and wash them over the weir into the bead recycle tank. This
stream was operated continuously at a rate no greater than necessary
to perform its function, normally about 3 gpm.
The Bead Recycle Section
The beads being washed over the overflow weir from the sludge settling
section were discharged into the bead recycle tank, a tank of dimensions
1 foot by 1.5 feet by 3 feet deep. From this tank the beads were dis-
charged continuously to the suction of the main feed pump supplying in-
fluent to the column. The bead-laden discharge of this pump was piped
directly to the bottom of the column proper.
13
-------
Sequence Timing in Normal Operation
The operation of the bead washing and recycling systems was controlled
by a set of timing relays designed to produce the following sequence of
events:
1. At the beginning of the cycle, open the solenoid valve in the sludge
settling tank.
2. At 15 seconds, close the solenoid valve and activate the bead
scraper and the centrifugal bead washing pump. The time provided
for sludge drainage (15 sec) is sufficient for withdrawal of about
2 gallons of sludge.
3. At 30 seconds, deactivate the bead scraper. At this point in the
cycle approximately 1/2 cubic foot of beads have been dropped into
the bead washing section and are being gradually cleansed and dis-
charged by the flow from the centrifugal wash pump, which continues
to operate for a time.
4. At 2 minutes, deactivate the centrifugal wash pump. At this time
in the cycle the recycle flow is washing beads over the outlet weir
and the sludge is being allowed to settle in the bottom of the
sludge settling tank.
5. At 4 minutes, repeat the cycle, starting with Step 1 above.
14
-------
SECTION V
EXPERIMENTAL PHASE
Development and Debugging of the Equipment
During the design and construction of the equipment, the usual assort-
ment of difficulties that one normally runs into in an experimental
project were encountered. There were numerous leaks to be plugged, and
several failures of purchased equipment were experienced. For example,
a centrifugal pump proved to be faulty, and a circuit breaker in the
main power panel failed. On one occasion an unexpected overnight drop
in ambient temperature resulted in freezing and rupture of several
plastic pipes.
On several occasions throughout the winter of 1970-1971, moist, solids-
laden beads emerging from the water at the screen level in the column
froze together during their transit from the screen section to the
scraper section. This problem was particularly serious with some of
the initial scraper designs that were limited in their ability to re-
cycle beads at a sufficiently rapid rate. In fact, several design modi-
fications of the bead scraping system were attempted before a system
was found which proved adequate to transport even partially frozen beads
at the required rates.
The bead scraper as originally designed was a circular piece of perfo-
rated steel plate with half-inch perforations on 1 inch centers. A
spindle was welded to- the center of the plate and mounted so that its
axis coincided with the axis of the column. In operation the plate was
driven with a gear motor at a rate of about 2 rpm. Presumably the beads
rising from below would pop through the perforations in the plate and
then be directed with a nonrotating baffle into a lateral discharge
chute leading from the column to the wash tank. A vapor lift pump was
installed to pump wash water from the wash tank into the top of the
discharge chute.
The airlift method of pumping was selected at the outset to avoid the
plugging of pumps and orifices with sewage-laden beads. This method of
transporting wash fluid was soon discovered to be inadequate, and the
airlift pump was quickly replaced with a submersible centrifugal pump.
Our anticipation of difficulties in pumping bead slurries with the small
centrifugal pump proved realistic. Intermittent difficulties were en-
countered with this particular pump throughout the course of the project.
The rotating-plate bead scraper turned out to be an inadequate design
for the handling of wet, solids-laden beads. It simply would not convey
enough wet beads to the discharge chute.
To remedy this problem a modified scraper assembly was designed, fabri-
cated, and installed. The new conveyor was a 5 foot section of 9-inch
diameter screw conveyor which was mounted on top of the column -utilizing
an especially constructed adaptor section. The adaptor section was
15
-------
designed and constructed to give a smooth transition from the 2-foot
diameter circular cross-section of the top of the column to the 1 foot
by 2 foot rectangular cross-section entrance to the conveyor shroud.
This particular arrangement did not function properly because the beads
had difficulty moving through the circular to rectangular transition in
the adaptor section. Though the screw flights would turn on schedule,
they would not pick up and move beads.
The next modification of the scraper section was to remove the adaptor
section completely and to mount the conveyor shroud directly atop the
column with segmentally-shaped baffle plates covering the open area.
This arrangement still seemed to provide too much constriction in the
bead path for the beads to be able to move freely from the column into
the conveyor.
The next and final arrangement was a double-barreled conveyor. An addi-
tional 5-foot section of 9-inch diameter screw flights was procured and
was mounted with its axis parallel to the existing conveyor. A new and
larger shroud was fabricated which housed both conveyor screw flights
and which was large enough to freely admit beads coming from the top of
the column. The final configuration for the scraper section is shown in
Figures 4 and 5, which are photographs of the equipment taken near the
termination of the experimental phase and thus show the arrangement of
the equipment as it finally evolved.
In addition to the bead scraping section, it was also necessary to make
design modifications to the bead washing section. As originally designed,
the beads were discharged into a small chute from the column and were
then washed down into the main bead washing tank with the air lift pump.
This method of washing was changed when the first scraper arrangement
was changed. A subsequent configuration was to let the beads discharge
directly from the screw conveyor into the bead washing tank. This ar-
rangement did not provide for efficient washing of the beads. They
would drop in bulk from the conveyor to the tank without being cleaned
in the process. It appeared that somehow the beads needed to be re-
strained for more efficient washing between the discharge of the con-
veyor section and the main bead tank.
To test this idea, a 1 foot by 2 foot by 4-inch deep tank was placed so
as to catch the beads being discharged from the conveyor section. Wash
water from a submerged centrifugal pump in the main bead washing tank
was directed onto the beads in this intermediate tank. It was found
that the short holdup of the beads in the intermediate tank, together
with the swirling action of the wash water introduced into this tank
by the centrifugal pump, provided a washing action better than that
heretofore obtained.
The final modification to the bead washing system was to replace the
rectangular tank with a circular, funnel-shaped tank functioning in a
manner similar to a dental spittoon and providing still more efficient
washing action.
16
-------
Figure 4. The Filtration Process
17
-------
oo
fJIK!
B ^
Figure 5. The Scraper and Bead Washer
-------
Period of Successful Operation
During the time of evolution and debugging of the mechanical problems,
the filtration performance of the unit was, for reasons to be discussed
shortly, somewhat below what one would expect. As a result, the initial
phases of the project were concerned not only with ironing out mechani-
cal details but also with attempts to improve the filtration performance.
As is often the case, after a preliminary induction period in which
everything seemed to go wrong, everything seemed to fall into place, and
the unit began to deliver acceptable filtration with good mechanical
dependability.
From this time on, the usual procedure was to leave the unit running
around the clock and often right through the weekend. Daily samples of
influent and effluent were routinely secured and submitted for analysis
of total organic carbon, chemical oxygen demand, and suspended solids.
At the time of sampling, feed rate, feed temperature, and pressure drop
across the bed were noted. On many occasions additional samples were
taken for special studies which were undertaken as time and funding
permitted.
The overall timing of the experimental work was approximately the fol-
lowing: From the inception of the project, approximately 9 months were
required for procurement, construction, and debugging of the equipment,
including the modifications previously discussed. Thereafter, about
2-1/2 months were spent investigating the performance of the filter
when handling trickling filter effluent. Then about 1/2 month was
spent treating primary clarifier effluent. Following this, about 1/2
month was spent processing raw sewage through the moving bed unit.
There followed a month in which the unit was not operated at all. The
shutdown was caused by a flash flood that completely submerged our pump
drivers as well as those associated with the main waste treatment plant.
When the plant was again on stream, raw sewage was not available to the
moving bed filter because of a pump failure. About 3 weeks were spent
collecting additional data on the processing of primary clarifier efflu-
ent; thereafter, the balance of the experimental time available, about
a month, was spent doing additional studies with raw sewage.
The Chapel Hill Waste Treatment Plant is normally operated with substan-
tial recycling from the final clarifiers to the primary clarifiers.
During the period of study of the moving bed filter, the recycle rate,
defined as recycle rate divided by feed rate, ranged from about 1.5 to
about 4.5 with most of the values exceeding 2. Such recycle ratios are
typical of high-rate trickling filter plants but not of low-rate plants
(2). With these recycle rates, the primary clarifier effluent is more
nearly like trickling filter effluent than would be the case with no
recycle. It should be remembered that the data here reported are for a
high-rate plant.
Initially in the investigation attempts were made to filter untreated
trickling filter effluent utilizing the polystyrene filter media.
These efforts proved fruitless. The data indicated that substantially
19
-------
no solids were being removed during the filtration. The next step was
to attempt to filter alum-coagulated trickling filter effluent at column
loadings of around 4 to 5 gallons per minute per square foot. Filtra-
tion performance when processing alum-treated feed was quite poor. How-
ever, we discovered more or less fortuitously that if the alum flow to
the column were stopped, filtration performance improved significantly.
The conclusion is inescapable that alum treatment modifies the poly-
styrene beads in some fashion so as to greatly improve their performance
as a filter medium. Following discovery and confirmation of this be-
havior, all subsequent filtration studies were made utilizing alum-
treated beads.
The mechanical changes that were necessary to cause the unit to function
properly have been discussed in a preceding section. In addition to
these changes, it was necessary to make certain modifications to obtain
satisfactory filtration performance. It was observed that the polysty-
rene particles, after having been in service for several months, would,
as a result of hydrostatic and hydrodynamic pressures, be deformed from
spheres to disc-shaped particles. This squashing of the individual par-
ticles was accompanied by reductions both in the total bed volume and
in the bed porosity. These changes in the structure of the bed could
and did on several occasions lead to malfunctioning of the unit in the
sense that the lateral discharge screens became overloaded and the ex-
cess liquid overflowed to the top of the column, through the bead
scraper, and into the bead washing section. To alleviate this required
some means of maintaining the porosity of the bed by inhibiting the
bulk deformation of the polystyrene particles. This was done by mixing
in with the 0.040-inch diameter polystyrene particles approximately an
equal part by volume of a particulate polyethylene blow molding resin
with particle size approximating 0.125 inch. The effect of the poly-
ethylene was to provide bed rigidity while, at the same time, permit-
ting the smaller polystyrene particles to accomplish effective filtra-
tions. All but one week's worth of significant data were taken with
the mixture of particles. This arrangement seemed to work quite satis-
factorily. It was necessary, however, to take steps to prevent the
polystyrene particles from entering the suction of the centrifugal bead
washing pump. It proved to be quite easy for the rigid polyethylene
particles to jam between the impeller and the housing of a centrifugal
pump and to thus cause a pump stopage.
In Figure 6 is shown a schematic of the Chapel Hill, North Carolina,
Waste Treatment Plant. The locations from which trickling filter efflu-
ent, primary clarifier effluent, and degritted sewage were diverted to
feed the moving bed contactor are indicated on the schematic.
MBC feed was sampled at the overflow of the feed drum and MBC effluent
was taken from the catch trough around the screen section. Routine
analyses included COD (chemical oxygen demand) and suspended solids,
performed according to "Standard Methods" (3), and TOC (total organic
carbon), performed according to "FWPCA Methods" (4). Usual procedure
was to perform COD and TOC measurements on samples from which solids
had not been removed but which had been homogenized. Thus COD and TOC
data include contributions from suspended matter.
20
-------
Raw Wastewater
Removal Point
Primary
Sedimentation
Clarifier
Effluent
Removal Point
Biological
Filtration
Final
Sedimentation
Figure 6. Schematic Diagram—Chapel Hill, N.C., Waste Treatment Plant
21
-------
SECTION VI
RESULTS
The raw data obtained on the moving bed contactor when utilized as a
filter are reproduced in the Appendix. These results are summarized
and discussed in this section.
The performance of the unit when removing suspended solids is summarized
in Table 1. The data indicate that when processing trickling filter ef-
fluent, the removal efficiency decreases as column loading or liquid
handling capacity increases. At the lowest flow rate, 3.1 gallons per
minute per square foot, the percent of suspended solids removed by the
filter seems unusually low, 55.4 percent. These numbers, however, are
the result of averaging over 2 days of operation and consequently should
be interpreted more as an indication rather than as a statistically
significant conclusion.
On one occasion the trickling filter effluent was sampled during a per-
iod of unusually high sloughing. The feed concentration of suspended
solids to the moving bed unit was 451 mg/1 with the moving bed removing
all but 56 mg/1. The capability of the filter to remove a higher per-
centage of a more concentrated feed stream appears to be a generally
valid characteristic.
When removing suspended solids from primary clarifier effluent at 5.9
gallons per minute per square foot, the filter removed 53.9 percent of
the feed solids. This figure represents a somewhat lower removal level
than found for unsettled trickling filter effluent at similar filtration
rates.
The unit appears well suited to the removal of suspended solids from raw
sewage. As indicated by the data in Table 1, from 62 to 82 percent of
the suspended solids in raw sewage was removed by the moving bed unit
at flow rates ranging from around 3.5 to 7.5 gallons per minute per
square foot. No difficulty was experienced during this phase of the
work with filter blinding or with operational instabilities.
The performance of the moving bed unit for COD and TOC removal is sum-
marized in Tables 2 and 3, respectively. Since some of the COD and TOC
is dissolved and the balance suspended, the removal of a certain per-
centage of the suspended impurity results in the removal of a smaller
percentage of the total impurity. Since organic removal efficiencies
are so strongly dependent on the nature of the suspended solids, the
removal of the latter is the primary criterion of process efficiency.
The trend toward removal of higher percentages of the impurities in
stronger feed streams is evident.
The behavior of the unit was studied with 200 mg/1 alum addition to the
feed. Under no circumstances did alum addition appear to be helpful.
However, as previously noted, in order to achieve satisfactory perform-
ance, it was necessary to subject the filter medium to some sort of alum
23
-------
Table 1
Performance Summary
Moving Bed Filter
Suspended Solids Removal
Feed
Trickling Filter
2 day average
9 day average
4 day average
8 day average
7 day average
Primary Clarifier
1 day sample
7 day average
1 day sample
Raw Wastewater
7 day average
2 day average
5 day average
1 day sample
Feed Rate
gpm/sq ft
Effluent
3.1
4.1
5.3
6.9
7.9
Effluent
4.3
5.9
7.2
3.4
4.3
5.6
7.5
In
37.0
57.8
72.0
33.8
28.1
80
62.9
62
145.1
189.5
156.0
83
Suspended
Solids
(mg/1) Out
16.5
10.2
18.3
12.8
12.4
48
29.0
57
34
34.5
59.4
20
Percent
Removed
55.4
82.4
74.7
62.2
54.3
40.0
53.9
8.0
76.6
81.8
61.9
75.9
This set of data excludes one anomalous point for which the influent
suspended solids level was 451 mg/1. Effluent concentration for this
sample was 56 mg/1.
-------
Table 2
Performance Summary
Moving Bed Filter
COD Removal
Feed
Trickling Filter
2 day average
9 day average
4 day average
9 day average
7 day average
Feed Rate
gpm/sq ft
Effluent
3.1
4.1
5.3
6.9
7.9
In
118.5
115.3
181.8
157.1
133.2
COD
(ma/1) Out
106
73.9
108.3
98.6
118.7
Percent
Removed
10.5
35.9
40.5
37.3
10.9
Primary Clarifier Effluent
1 day sample
7 day average
1 day sample
Raw Wastewater
7 day average
2 day average
5 day average
1 day sample
4.3
5.9
7.2
3.4
4.3
5.6
7.5
227
162.3
163
385.1
160.0
476.0
294
173
127.2
154
171.4
132.0
266.4
154
23.8
21.6
5.5
55.5
17.5
44.0
47.6
25
U.S. Er
-------
Table 3
Performance Summary
Moving Bed Filter
TOC Removal
Feed
Trickling Filter
2 day average
9 day average
4 day average
9 day average
7 day average
Primary Clarifier
1 day sample
7 day average
1 day sample
Raw Wastewater
7 day average
2 day average
5 day average
1 day sample
Feed Rate
gpm/sq ft
Effluent
3.1
4.1
5.3
6.9
7.9
Effluent
4.3
5.9
7.2
3.4
4.3
5.6
7.5
In
35.0
39.9
52.0
48.1
34.9
74
82.7
79
126.0
133.0
189.6
99
TOC
(mg/1) Out
31.0
24.2
36.8
41.6
30.7
60
71.1
74
53.0
64.5
127.0
58
Percent
Removed
11.4
39.2
29.3
13.6
12.0
18.9
14.0
6.3
57.9
51.5
33.0
41.4
26
-------
pretreatment. A continuation of alum addition seemed to have no desir-
able effect. Also, once accomplished, the alum treatment of the beads
seemed permanent. In our experience, a second treatment was not needed
for the duration of the project, some 6 months.
A sequence of a half-dozen samples was taken while processing trickling
filter effluent and submitted for total phosphate analysis. This was
done during the two weeks following alum pretreatment of the beads but
without continuing alum addition. The results show little, if any,
phosphate removal.
'The suspended solids content of sludge as discharged from the sludge
settling tank was most often in the range of 300 to 600 mg/1. However,
on occasion, values as low as 183 mg/1 were observed. A sequence of
four runs producing sludge in excess of 1200 mg/1 solids showed that
the unit was capable of producing a thicker sludge provided more atten-
tion was given to the operation of the sludge settling section. On
two occasions, values in excess of 2600 mg/1 were observed. The solids
content in discharge sludge can be raised by adjusting the timing se-
quence to provide for longer sedimentation periods during the bead har-
vesting cycle, by more effective baffling in the sludge settling tank,
and by more careful placement of the intake of the centrifugal bead
washing pump so as to avoid undue sludge dilution.
27
-------
SECTION VII
RELATED STUDIES
During the course of the investigation several independent studies were
undertaken. These studies were aimed in part at explaining the reasons
for the observed behavior of the filtration unit and in part to estab-
lish the feasibility of ideas for follow-on developmental work.
Electrophoretic Mobility Studies
During the course of the investigation it was observed that polystyrene
beads treated with alum produced more effective removal of pollutant ma-
terials during filtration of secondary effluent compared to similar
beads which had received no alum treatment. It was surmised that this
difference in behavior could be attributed to a modification of the sur-
face charge, or zeta potential, of the polystyrene beads resulting from
alum treatment. To investigate this hypothesis, experiments were under-
taken by Professor Charles R. O'Melia of the Department of Environmental
Sciences and Engineering of the University of North Carolina at Chapel
Hill to determine the effects of alum treatment on the surface proper-
ties of polystyrene under chemical conditions analogous to those in the
effluent from the waste treatment plant.
Unexpanded beads were crushed with a hammer, ground with a mortar and
pestle, and wet-sieved. Crushed polystyrene which passed a 105y sieve
and was retained on 1.2y filter paper was thus obtained. This frac-
tionated material was suspended in 5 x 10 M NaHC03 which had been
neutralized to pH 7.5 with sulfuric acid. The resulting suspension
(pH 7.5, ionic strength about 5 x 10~3) was similar in these respects
to secondary effluent from the Chapel Hill Waste Treatment Plant.
Varying quantities of alum were added to aliquots of this stock suspen-
sion accompanied by slow stirring and pH adjustment with NaHCO^.
Electrophoretic mobility measurements of the particles in these sus-
pensions were made using a microscope and Briggs cell (5) .
The results of these experiments are presented in Figure 7. Particle
surface characteristics (electrophoretic mobility, (y/sec)/(volt/cm),
or zeta potential, millivolts) are plotted as functions of applied
alum dosage (log scale, moles/1 of aluminum sulfate or mg/1 of aluminum).
The polystyrene particles had an average electrophoretic mobility of
-2.9 (y/sec)/(volt/cm) with addition of alum. This corresponds to a
zeta potential in the order of -40 millivolts. While these values are
functions of pH and ionic strength, it is probable that this material
is negatively charged in the absence of alum or other coagulants under
all conditions of interest in water and wastewater.
Increasing addition of alum caused the electrophoretic mobility to be-
come less negative, indicating a reduction in the surface charge of the
polystyrene. In no case did the alum neutralize this negative charge,
i.e., bring the negative mobility or zeta potential to zero. At high
29
-------
-10
o
-20
4-1
C
-------
alum concentrations (>10 M), it is probable that the particles observed
were primarily precipitates of aluminum. The surface properties of
aluminum hydroxide should maintain their negative characteristics at all
ionic strengths in suspensions at pH 7.5. At lower pH levels, however,
these materials could assume a positive charge.
A maximum electrophoretic mobility is observed at an alum dose of about
10~^M. This is difficult to interpret and explain. Possibilities in-
clude (a) experimental error, (b) ionic strength effects (the ionic
strength of the solution increases with the alum dose due to the alum
itself and to the NaHC03 added), (c) increasing hydrophilicity of the
polystyrene particles with increased alum dosage, and (d) the changing
character of the particles being observed (polystyrene particles are
observed at low alum dosages; some aluminum hydroxide precipitates were
probably observed at high alum dosages). Regardless of the cause, how-
ever, it is apparent that polystyrene particles remain negatively charged
at pH 7.5 at all alum dosages.
The following conclusions can be stated:
1. Polystyrene is negatively charged in water.
2. Polystyrene treated with alum is negatively charged in water at
pH 7.5 regardless of the alum dosage.
3. Increased filtration efficiencies of alum-treated polystyrene which
are observed in wastewater at pH 7.5 does not appear to be ascribed
to charge effects. It is plausible that alum treatment improved
filtration in these systems by rendering the surfaces of the poly-
styrene (a) more hydrophilic, so that materials in suspension were
better able to contact the surface of the beads, or (b) more
"sticky" or chemically reactive as is the nature of the adsorbed
hydroxo-aluminum species.
Perlite Studies
At the beginning of the project polystyrene was selected as the filter
medium because of its desirable density and abrasion resistance. Based
upon the conclusions of the electrophoretic mobility study, it appears
that the hydrophobicity and negative surface charge of polystyrene make
it possible that much better filtration performance could be obtained
with a filter medium not exhibiting these undesirable characteristics.
An attractive alternative filter medium appears to be perlite, which is
a naturally occurring siliceous volcanic rock. It is processed to form
a foamed, or popcorn-type, particle structure which contains countless
tiny air bubbles separated by thin, glassy membranes. Perlite is nor-
mally used as a refractory insulation, a concrete or plaster aggregate,
or a fertilizer carrier.
Many of the commercially available grades of perlite are fragile and
would tend to excessive attrition in the moving bed filter. It is thus
necessary to obtain a more robust particle for use as a filter medium.
31
-------
According to one manufacturer, it may be possible to induce less expan-
sion upon popping the the particles, and thus as an integral part of
the manufacturing process, wind up with stronger particles. An alter-
native method is to coat the particles with silicate.
A jury-rigged fluidized bed assembly was set up to demonstrate the
feasibility of a fluidized-bed technique for application of a refractory
coating (sodium silicate) to the perlite particles. The technique used
was to dilute the sodium silicate with water to reduce its stickiness
before gradual addition of the diluted solution to a bed of perlite
particles fluidized with heated air. The silicate addition can be done
either intermittently or slowly enough to lag the drying so that the
particles never become wet enough, to hinder the fluidization. By add-
ing a dye to the silicate solution, we were able to confirm that the
fluidized bed technique uniformly coated the particles.
To test filtration with perlite, a small, fixed-bed filter was constructed
and used in a batch manner to filter effluent alum sludge from a
fluidized-bed sludge blanket clarifier. A relatively coarse perlite was
used, with size distribution as follows: 50 percent +8 mesh, 45 percent
-8 to +16 mesh, and 5 percent -16 mesh. The sludge from the clarifier
was filtered at a rate of 12 gallons per minute per square foot. The
suspended solids concentration in clarifier sludge was 1,727 mg/1. The
filter effluent suspended solids concentration was 121 mg/1 after the
filtration of 34 gallons of sludge per gallon of beads and 890 mg/1 after
the filtration of 136 gallons per gallon of beads. This study lends
support to the possibility of utilizing.the moving bed filter for the
dewatering of dilute sludges as well as to the feasibility of perlite
as an alternate filter medium.
An advantage of utilizing refractory-coated perlite as the filter medium
is the capability of this material to withstand sludge incineration
temperatures. This makes possible the direct incineration of sludge
from the filter medium without the necessity for washing the filter
medium-and subsequent dewatering of the wash liquor. An initial dem-
onstration of the feasibility of this idea involved taking a sample of
sludge-laden perlite and heating it to incineration temperatures with a
propane torch. The sludge burned off evenly and smoothly, and no diffi-
culties were apparent that would interfere with the operation of a con-
tinuous process.
In order to evaluate the heat requirements that would be needed for a
direct incineration process, several analyses were made of the moisture
content and incinerable sludge content of beads as they emerged from
the top of the column with the following results:
32
-------
Percent
Water
34
31
21
31
23
25
29
Average
2
Ib solids
26.2
23.8
17.5
44.3
16.4
16.7
20.7
23.7
Percent
Solids
1.3
1.3
1.2
0.7
1.4
1.5
1.4
These results indicate that sludge-laden beads, because of their high
moisture content, will not sustain combustion without auxiliary heat.
In some applications it may be advantageous to supply this heat in
order to achieve direct incineration.
An additional application possibility for the moving bed filter is as
a contactor for carbon treatment. A method for coating perlite parti-
cles with carbon was demonstrated in a small laboratory study by pyroly-
sis of particles that had been wet with a concentrated sugar solution.
The activity of such particles was not studied. Continuing work is
needed to establish the conditions under which the char becomes most
adsorbent. This would then permit investigation of the performance of
the moving bed unit as a carbon contactor.
A projected integrated process utilizing filtration, adsorption, and
direct incineration is shown schematically in Figure 8.
33
-------
BEADS
FEED
V
PRODUCT
HEAT
EXCHANGER
F
U
R
N
A
C
E
Figure 8. Proposed Combined Filtration-Incineration Process
-------
SECTION VIII
DISCUSSION AND COST ESTIMATES
In this section is presented an estimate of the costs of filtration
with the moving bed process and a comparison of these costs with those
of other solids removal techniques. The estimate is for a 1 mgd process,
assuming 200 mg/1 suspended solids removed by Dylite F-40-C polystyrene
resin costing $0.25/lb and expanded to a bulk density of 5 Ib/ft .
Equipment costs were taken from standard reference sources and the fil-
ter vessel cost was predicted from the data of Bauman (6) on a price
per pound basis for a 13-foot diameter vessel 20 feet tall of 3/8 inch
carbon steel (unfired pressure vessel). Cost items were corrected to a
1969 basis by using the PHS Sewage Treatment Cost Index (7) where ap-
propriate.
The following table summarizes the cost of the principal items of equip-
ment (PIE).
TABLE 4
Itemized Equipment Cost
1 MGD Filter Plant
Filter vessel $ 5,400
Pumps (w/drive)
(1) Primary feed $ 900
(2) Filtrate 150
(3) Slurry 2,700
(4) Sludge 180
$ 1,380
Tanks
Filtrate surge $1,800
Sludge settling 640
$ 2,440
Media scraper 500
Media expander (infrared heated) 2,500
Initial media inventory 3,500
Site preparation and plant connections 2,500
Miscellaneous 1,000
TOTAL PIE $19,220
35
-------
The total capital cost for the principal items of equipment (PIE) was
multiplied by a factor which takes into account the following items.
1) As percent of PIE
a) Plant erection: 30 percent
b) Instruments: 4 percent
c) Contingencies: 10 percent
d) Engineering: 10 percent
e) Interest on investment during construction: 4 percent
2) As a rate on total investment
a) Supplies and maintenance materials: 0.5 percent per year
b) Maintenance labor: 0.5 percent per year
c) G + A: 30 percent of (a + b)
d) Payroll extras: 15 percent of (a + b)
e) Taxes and insurance: 2.0 percent per year
f) Amortization at 6 percent at 25-year life (7.8 percent per year)
Combination of all the above factors gives a factor of 0.0518 percent on
(PIE) in dollars per day as total debt service. (One operating year =
330 days.)
The power requirements for the process have been estimated at 145 kwh/day
for the main feed pump and 18 kwh/day each for the bypass pump and the
bead scraper for a total requirement of 181 kwh/day. At $0.01/kwh, the
total power cost is 0.18C/1000 gal.
An estimate of 0.5C/1000 gal. was taken for loss of filter medium.
Experience has shown that medium losses can easily be kept within this
figure.
Labor costs were found by assuming 3 man-hours per day and 6 man-hours
per day for 1 and 5 million gallon plants, respectively, at $3.50 per
hour. Thus, labor costs for each plant are !.!<; and 0.42C/1000 gal.,
respectively. It might be reasonable to suppose, however, that the
operating labor for the smaller plant could be absorbed by existing
personnel. Labor costs for small plants are usually disproportionately
large in comparison to other operating costs on a per gallon basis.
A summary of these cost estimates follows:
36
-------
TABLE 5
MBC Process Operating Costs
Capacity 1 MGD 5 MGD
Debt service 1.00 0.53
Power cost 0.18 0.18
Medium loss 0.50 0.50
Labor 1.10 Q.42
2.780/1000 gal. 1.630/1000 gal.
These costs compare quite favorably with estimates for other solids re-
moval processes. The following table compares MBC costs with those for
filtration utilizing a mechanically propelled bed (9), with those for
microscreening (8), and with those for fixed bed filtration (8).
TABLE 6
Cost Comparisons
1 MGD 5 MGD
Moving Bed Contactor 2.80/1000 gal. 1.60/1000 gal.
Mechanically Driven Bed 7.2
Microscreening 1.2 1.1
Fixed Bed 6.5 3.6
Surface loadings forming the bases for these costs estimates are the fol-
lowing: for the moving bed contactor, 7 gpm/ft^; for the mechanically
driven bed, 2 gpm/ft^; and for the fixed bed, 4 gpm/ft^. The cost basis
for microscreening is not reported, but typical loadings for such equip-
ment are about 10 to 30 gpm/ft^.
The moving bed unit should be easily scalable to a diameter of about 10
feet. A unit of this size should handle almost one million gallons per
day. For larger requirements, modular construction would probably be
desirable.
37
-------
SECTION IX
ACKNOWLEDGMENTS
We wish to acknowledge the assistance and contributions of the following:
The Environmental Protection Agency and Mr. James F. Kreissl, Project
Officer, for financial support and many valuable suggestions.
The University of North Carolina Wastewater Research Center for pro-
viding facilities, services, and analytical assistance.
Mr. Bill Sun for day-to-day attendance of the equipment.
Professor Charles O'Melia, UNC School of Public Health, for electro-
phoretic mobility studies.
Eliza Rucker for secretarial and computational assistance.
Dr. J. C. Orcutt, co-inventor and co-applicant along with F. 0. Mixon,
for patent coverage of the MBC.
39
-------
SECTION X
REFERENCES
1. Ives, K. J., "Research on Deep Filters," Trans. Instn. Chem. Engrs.,
43., pp. T238-T247 (1965).
2. ASCE Manual of Engineering Practice No. 36 (WPCF Manual of Practice
No. 8), Sewage Treatment Plant Design, New York (1959).
3. American Public Health Association, Inc., Standard Methods for the
Examination of Water and Wastewater, New York (1965).
4. Federal Water Pollution Control Administration, FWPCA Methods for
Chemical Analysis of Water and Wastes, (November, 1969).
5. Black, A. P., and Smith, A. C., "Determination of the Mobility of
Colloidal Particles by Microelectrophoresis," Journal, Amer. Wat.
Works Assn. , 54_, pp. 926-934 (1962).
6. Bauman, H. C., paper presented to A.A.C.E., June (1958).
7. Smith, R., PB 182 158, "Cost of Conventional and Advanced Treatment
of Wastewaters," July (1968).
8. Smith, R., and McMichael, W. F., "Cost and Performance Estimates for
Tertiary Wastewater Treating Processes," Robert A. Taft Water Research
Center, Cincinnati, Ohio (June 1969).
9. Bell, G. R. , ^t _a!L, "Phosphorus Removal Using Chemical Coagulation
and a Continuous Countercurrent Filtration Process," FWQA Water
Pollution Research Series, 19010 EDO 06/70 (June 1970).
41
-------
SECTION XI
PATENT STATUS
The initial patent application on the Moving Bed Contactor was filed
1 September 1970, Serial No. 68,611. On November 24, 1971, the initial
application was disallowed. An amendment was prepared and filed on the
patent on February 16, 1972. On March 21 all claims were allowed.
43
-------
SECTION XII
APPENDIX
Performance Data—Moving Bed Filter
45
-------
APPENDIX
Performance Data
Moving Bed Filter
Date
1/5
1/6
1/7
1/8
1/11
1/12
1/13
1/18
1/19
1/20
1/22
1/25
1/26
1/27
1/28
1/29
2/1
2/2
2/3
2/4
2/5
2/8
2/9
2/10
2/11
2/12
Rate
(gpm)
6.8
7.5
7.6
8.3
8.6
8.5
8.2
8.4
8.8
9.1
8.9
8.8
8.9
8.6
16.9
13.2
14.1
14.3
14.2
Temp
°F
60
53
59
59
61
61
58
58
58
54
58
50
51
51
51
57
49
56
51
COD
(mg/1)
In Out
184
154
139
127
95
147
155
137
149
109
194
169
256
140
85
269
115
165
145
57
64
66
91
276
148
123
244
132
148
139
133
102
120
149
131
94
139
137
237
110
107
209
105
262
220
TOG
(mg/1)
In Out
21
51
80
59
47
69
58
55
52
56
60
41
54
25
38
63
54
55
41
34
70
39
53
73
22
20
24
21
82
41
25
80
49
49
38
58
35
36
23
33
56
43
48
37
31
47
34
45
90
Suspended Sludge
Solids SS
(mg/1) (mg/1) Remarks
In Out
19
73
116
120
30
78
41
31
31
30
27
43
29
29
22
50
44
28
28
35
92
31
52
61
11 Bed depth at 1.5 ft. Fresh, untreated
4 filter medium. Processing unsettled
2 trickling filter effluent.
1
34
16
3
106
23
22
25
24
16
14
22
18
25
47
21
18
37
48
44
61
66
-------
Date
2/15
2/16
2/17
2/18
2/22
2/23
2/24
3/22
3/23
3/24
3/25
3/26
3/29
3/31
4/1
4/2
4/5
4/6
4/7
4/8
4/9
4/12
4/13
4/14
4/15
4/16
4/19
4/20
Rate
(gpm)
14.5
16.3
11.7
12.1
7.3
7.8
7.4
12.1
16.3
9.8
4.3
1.8
.7
6.4
9.6
10.3
8.1
11.5
5.3
10.3
9.8
8.7
6.3
10.1
11.3
9
13.1
Temp
°F
53
60
58
59
61
63
59
61
64
60
61
51
59
59
63
61
61
59
58
64
61
62
63
62
62
63
73
COD
(mg/1)
In Out
168
162
147
160
232
146
177
273
210
255
265
233
179
213
147
200
139
175
186
111
132
133
104
123
132
88
109
163
149
90
130
188
117
128
232
180
210
263
208
142
238
188
202
153
130
121
68
187
123
150
113
127
142
160
TOC
(mg/1)
In Out
52
57
50
53
75
55
73
73
62
78
75
74
66
62
67
72
46
53
72
37
51
56
46
56
63
44
48
60
56
36
45
69
43
58
87
46
57
74
66
61
69
89
73
44
39
45
33
62
53
57
64
63
64
57
Suspended Sludge
Solids SS
(mg/1) (mg/1) Remarks
In Out
68
52
54
34
60
38
84
69
87
104
136
125
66
111
107
90
89
70
146
43
46
62
34
18
34
20
25
58
40
20
25
27
12
79
88
51
88
156
101
22
182
82
90
95
22
49
38
133
63
100
30
30
71
64
At this point the bed depth for filtra-
tion was increased
by adding another box
of beads. After this addition the fil-
tration depth was 4
ever, showed signs
the beads compacted
ft. This depth, how-
of getting smaller as
during filtration.
Filtration depth increased to about 6 ft
at this point.
Scraper assembly modified. Screw section
mounted directly in
converging section.
-------
00
Date
4/21
4/22
4/23
4/26
4/27
4/28
4/29
5/3
5/4
5/17
5/18
5/19
5/20
5/21
5/24
5/25
5/26
5/27
6/14
6/15
6/23
6/24
6/25
6/28
6/29
6/30
7/1
7/2
Rate
(gpm)
11.3
11.0
9.5
10.5
10.4
5.3
7.9
10.3
10
9.8
8.2
10
8.5
5.6
6.3
7.2
10.4
10.8
9.7
12
16.3
16.5
17.2
17
16.5
Temp
°F
70
70
63
69
70
69
69
62
71
74
80
71
71
72
73
74
82
76
79
79
79
81
80
78
80
COD
(mg/D
In Out
105
115
124
149
123
95
149
115
115
150
131
95
142
127
85
95
99
84
97
160
222
210
135
127
125
156
179
196
129
182
73
131
154
356
71
164
221
40
139
98
77
92
83
159
98
93
TOC
(mg/1)
In Out
47
42
41
63
64
52
43
49
56
44
89
47
72
83
67
50
51
48
37
39
36
71
57
44
48
48
50
67
77
57
41
58
22
56
92
132
26
67
98
28
67
41
27
35
29
52
32
34
Suspended Sludge
Solids S;> Remarks
(mg/1) (mg/1)
In Out
22
25
21
69
106
55
54
50
29
22
32
212
59
11
147
29
68
44
21
13
43
105
96
44
53
51
49
84
123
67
24
68
28 Alum addition to feed started at 200 i
45
125
238
42
20
24
157
112
23 Alum addition stopped.
9
5
8
38
15
12
-------
Date
7/5
7/6
7/9
7/12
7/13
7/14
7/15
7/16
7/19
7/20
7/21
7/22
7/23
7/26
7/27
7/28
7/29
7/30
8/2
8/3
8/5
8/6
8/9
8/10
8/11
8/12
8/13
8/16
8/17
8/18
Rate
(gpm)
13.4
19
22.3
22.5
21.4
21.5
19.8
22.5
22.5
9.9
11.8
16.1
21.7
14.1
16.1
13.4
13.4
16.1
14.1
13.1
13.1
13.1
22.5
23.4
25.6
23.4
21.6
23.4
25.6
Temp
79
81
73
79
80
79
80
80
80
80
80
78
78
78
80
80
80
80
80
80
80
80
80
80
78
74
78
74
75
COD
(mg/1)
In Out
88
265
125
130
74
102
156
510
88
127
153
77
85
198
298
271
184
149
95
90
132
80
98
102
125
144
182
99
80
69
132
118
81
71
78
117
185
36
94
135
50
82
70
235
216
203
89
73
66
93
27
73
107
134
120
170
73
51
TOC
(mg/1)
In Out
27
55
42
34
31
3G
47
138
31
48
33
28
26
111
80
62
47
25
28
22
32
92
26
39
42
32
26
37
24
21
40
35
28
29
28
36
137
25
28
35
19
18
113
72
59
28
22
20
12
24
20
28
32
35
38
22
24
21
Suspended Sludge
Solids SS
(mg/1) (mg/1) Remarks
In Out
30
299
29
16
19
28
56
451
22
81
53
16
17
99
202
123
109
27
25
22
27
25
19
18
25
39
29
35
21
5
26
20
12
9
17
18
56
8
16
24
6
7
33
165
88
133
15
66
7
6
5
13
11
10
30
13
12
2
Polyethylene added
From 7/16 until 8/24
introduced after the
rather than before.
ference in suspended
suits .
Started alum addition
Stopped alum addition
Total P04 (mg/1)
In Out
7.5 8.4
6.8 6.5
8.0 8.5
6.6 6.3
4.2 8.3
plant recycle was
primary clarifier
No significant dif-
solids content re-
at 200 mg/1.
.
-------
i-n
O
Date
8/19
8/20
8/24
8/25
8/26
8/27
8/30
9/3
9/8
9/9
9/15
9/16
9/17
9/20
9/21
11/3
11/4
11/5
11/8
11/11
11/16
11/17
11/18
11/19
Rate
(gpm)
25.6
25.6
12.0
12.8
13.7
13.1
13.4
18.7
22.5
18.7
18.7
18.7
7.5
7.3
10.0
11.8
12.5
18.8
18.2
18.2
16.6
Temp
°F
82
78
78
78
78
78
79
80
80
81
79
78
76
63
68
69
70
71
68
69
71
COD
(mg/1)
In Out
153
115
58
101
130
140
252
227
185
163
455
453
439
513
520
212
179
235
182
128
218
110
176
157
168
127
110
63
53
82
93
224
173
146
154
226
345
227
248
286
170
133
175
142
70
170
115
143
119
80
TOC
(mg/1)
In Out
46
24
33
27
31
118
61
74
79
79
167
152
87
240
302
62
55
63
95
88
95
62
92
77
86
48
28
37
25
21
22
111
72
60
64
74
174
120
84
126
131
47
41
51
77
76
85
59
80
62
72
44
Suspended Sludge
Solids SS
(mg/1) (mg/1) Remarks
In Out
33
26
25
44
49
76
80
40
62
202
145
118
193
122
59
51
56
68
50
82
71
42
89
47
69
101
11 Total P04 (mg/1)
11 In Out
7 7.1 6.4
19
11
34 Transferred to primary clarifier effluent.
48
3
57
50 Transferred to raw wastewater.
75
37
69
66
13"
15
21
33
14 Increased bed depth to 6 ft and started alum
treatment for bed conditioning.
38 406 Clarifier effluent.
35 198
27 512
39 427
23 638
38 471
52 508 Raw wastewater.
-------
Date
11/22
11/23
11/29
11/30
12/1
12/2
12/3
12/6
12/7
12/9
12/16
12/17
12/20
Rate
(gpm)
13
23
11
9
10
10
10
11
9
9
.4
.5
.0
.9
.8
.6
.4
.0
.4
.4
Temp
°F
66
65
61
64
62
61
62
69
67
71
68
67
COD
(mg/1)
In Out
160
294
404
450
419
244
520
317
342
351
149
187
194
345
377
393
132
154
133
278
227
115
120
205
122
137
130
158
213
162
64
63
TOG
(mg/1)
In Out
142
124
99
123
150
142
82
110
172
103
115
49
57
66
144
130
134
54
75
58
54
91
71
32
28
64
31
39
39
44
57
67
20
17
Suspended
Solids
(mg/1)
In Out
260
119
83
174
178
115
131
153
140
125
144
50
49
63
196
133
169
38
31
20
23
42
20
21
32
73
27
73
80
108
113
56
13
8
Sludge
SS
(mg/1) Remarks
334
679
2,601
1,177
824
2,956
350
1,496
1,349 Alum added.
1,236
1,237
293
411
355
183
-------
SELECTED WATER i. Report No.
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
4. Title
2, 3. Accession No.
w
5. Report Date
FILTRATION OF MUNICIPAL WASTE WITH A MOVING
BED CONTACTOR,
7. Author(s)
MIXON, F. 0.
9. Organization
Research Triangle Institute
Environmental Studies Center
Research Triangle Park, North Carolina 27709
6.
8. Performing Organization
Report No.
10. Project No.
17030 FWH
12. Sponsoring Organization
15. Supplementary Notes
11. Contract/Grant No.
14-12-895
13. Type of Report and
Period Covered
16. Abstract
A novel moving bed contactor has been utilized in filtration studies of municipal
waste at various stages within a trickling filter plant.
Granular, buoyant filter medium is slurried with process feed and introduced to
the bottom of a column equipped with lateral retaining screens and filter medium
harvesting machinery, both positioned toward the top of the column. Within the
column, filter medium rises by buoyancy and forms a porous plug that traps suspended
solids from the feed stream. Filtered liquid is removed from the lateral screen,
and soiled filter medium is continuously removed from the column top, washed, and
recycled to the column bottom.
The process operates stably and dependably on all feeds tested—raw wastewater,
primary clarifier effluent, and trickling filter effluent. Suspended solids
removals of 60 to 80 percent can be achieved at column loadings up to 7.5 gal/
min/sq ft. Filtration of alum-coagulated feed is less effective than that of
untreated feed.
17a. Descriptors
*Filtration, *Separation techniques, *Sewage treatment,
17b. Identifiers
Continuous filtration, *Floating media, *Suspended solids, Trickling filter
17c.COWRR Field & Group 05 D
18. Availability
19. Security Class.
(Report)
20. Security Class.
(Page)
21. No. of
Pages
22. Price
Send To:
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 20240
Abstractor James F. Kreissl
Institution EPA-NERC, Cincinnati, Ohio
WRSIC 102 (REV. JUNE 1971)
<3P 0 9 13.261
------- |