WATER POLLUTION CONTROL RESEARCH SERIES • ORD-17O5ODALO5/7O
GRANULAR CARBON
TREATMENT OF RAW SEWAGE
U.S. DEPARTMENT OF THE INTERIOR • FEDERAL WATER QUALITY ADMINISTRATION
-------
W TER POlLUTION. CJNIBJL RESER1 H SERIES
The Water Ik)llution Control Research Reports describe
the results and progress in the control and abateirent
of pollution in our Nation’ s waters. They provide a
central source of ir foimation on the research, develop-
nent, and c3enonstration activities in the Federal Water
Qn 1 ity Mministration, in the U. S. Departnent of the
Interior, through inhouse research and grants and con-
tracts with Federal, State, and local agencies, research
institutions, and industrial organizations.
A triplicate abstract card sheet is inchx3ed in the
report to facilitate ibformation retrieval. Space is
provided on the card for the user’s accession nurber and
for &1i tional uniterms.
Inquiries pertaining to Water Pollution Control Research
Reports s1 uld be directed to the Head, Project Reports
System, Planning and Resources Office, Office of Research
and Developient, Departnent of the Interior, Federal Water
Quality 1 änthistration, Rxin 1108, Washington, D. C. 20242.
-------
GRANULAR CARBON TREATMENT OF RAW SEWAGE
by
C. B. Hopkins
W. J. Weber, Jr.
R. Bloom, Jr.
FMC Corporation
Princeton, New Jersey 08540
for the
FEDERAL WATER QUALITY ADMINISTRATION
DEPARTMENT OF THE INTERIOR
Program #17050 DAL
Contract #14-12-459
FWQA Project Officer, Dr. C. A. Brunner
Advanced Waste Treatment Research Laboratory
Cincinnati, Ohio
May, 1970
-------
F k Review N tioe
!It is report has been reviewed by the Federal
Water Quality Administration and approved for
publication. Approval does not signify that
tbe cxzitents necessarily reflect the views
ai policies of the Federal Water Quality
Administration, nor does nention of trade
nates or xmtercial prodix ts cxnstitute
endorsaient or recxi iiendation for use.
‘or eale by the Superintendent of Docurnent , U.S. Government Printing Office
Washington, D.C. 20402- PrIce $1
-------
TABLE OF CONTENTS
ABSTRACT
INTRODUCTION . . 1
OBJECTIVES . . . . 3
SUMMARYANDRECOMMENDATIONS ....•.•••••• 4
EXPERIMENTAL
Test P rogr am . . . 6
Apparatus and Procedure . . . . . . . . . . . . . 6
Analytical Methods . . . . . . . . . . 18
RESULTS
Coagulation and Clarification Studies . . . . . . 21
Phase-One Operation: Preclarification . . . . . . 24
Phase-Two Operation: Extended Treatment . . . . . 43
Phase-Three Operation: Postclarification . . . . 52
Analysis of Spent Carbons . . . . . . . . . . . . 59
Carbon Particle Size Effects . . . . 59
Polishing Column . . . . . . . 59
DISCUSSION AND CONCLUSIONS
Clarification . . . . . . . . . . . 64
Activated Carbon Treatment. . . . . . . . . . . . 65
Polishing Treatment Concept . . . . 69
Carbon Particle Size Effects. . . . . . . . . . . 69
Proposed Treatment Scheme . . . . . . . . . . . . 70
Estimation of Treatment Cost. . • . . . . . . . . 76
REFERENCES . . . . . . . . . 82
-------
ABSTRACT
The primary purpose of this study has been the detailed
comparison of expanded-bed and packed—bed modes of operation
of activated carbon adsorption systems in the direct physico-
chemical treatment of raw sewage and/or primary effluent. A
major part of the project involved extended field testing of
a pilot scale treatment system of chemical clarification
followed by adsorption.
Pilot scale operations were carried out at the sewage
treatment plant of the Ewing-Lawrence Sewerage Authority (ELSA)
near Trenton, New Jersey. Selection of the coagulant and
determination of design parameters for the pilot plant
coagulation—clarification system were based on extensive
laboratory investigations of the coagulation of the Ewing-
Lawrence raw sewage and primary effluent.. Ferric chloride-
was chosen for use as the coagulant in the pilot operations.
Equivalent clarified feeds for the adsorbers were produced
from raw sewage and primary effluent with the same coagulant
dosage. The pilot study was carried out with primary effluent
as feed to the system.
The expanded—bed and packed—bed adsorbers were constructed
of plastic—coated 10-in, diameter steel pipe. These adsorbers,
each containing a 6—ft settled depth of activated carbon, were
arranged so that expanded—bed and packed-bed units of up to
24—ft carbon depth could be operated in parallel at specific
feed rates of 5—7 gpm/ft 2 . These studies extended over a
period of one year with the longest single run of the pilot
system being 125 days.
The expanded-bed and packed-bed modes of operation
demonstrated essentially equivalent removal of organic matter
from chemically clarified primary effluent. When a 24-ft
settled depth of activated carbon was used, both systems
consistently produced a clear, treated water with an average
organic content of only 3-5 mg/i, measured either as total
organic carbon (TOC) or biochemical oxygen demand (BOD).
The expanded—bed mode of operation offers several distinct
advantages. First, expanded—bed operation essentially eliminates
plugging or fouling with particulate matter and air binding,
both of which cause high pressure losses and frequent back-
washing requirements for packed—bed operation. In the expanded-
bed system, feed pressure remains constant and maintenance
requirements are minimal. Further, expanded—bed operation
provides the opportunity for a degree of natural aeration between
contacting stages which may be important to the production of
a stable, treated water.
Clarification with ferric chloride was found to remove
over 90% of the solids and at least 60% of the TOC from the
wastewater, including about 30% of the organic matter classified
by membrane filtration as soluble (SOC). In addition, phosphate,
-------
which is not normally removed to any appreciable extent in
conventional biological treatment of wastewater, was reduced
by over 90% during chemical clarification with the ferric
chloride.
The combined direct treatment system of chemical clarifi-
cation followed by adsorption in expanded beds of activated
carbon maintained, over a period of 125 days, an average of
93% removal of organic matter from the primary effluent
(equivalent to greater than 95% removal from the raw sewage)
measured as either TOC or BOD. This performance was maintained
in spite of variations in waste composition which would adversely
affect conventional biological processes. Even under the best
of operating conditions, conventional biological treatment can-
not produce nearly as high a quality effluent as was obtained
by this direct clarification-adsorption treatment. In addition,
it was demonstrated that further treatment of the effluent with
a small bed of fresh carbon could reduce the organic content
to 1-2 mg/l TOC.
Estimated cost, on a realistic 1969 economic basis, for
treating raw waste water in a 10 mgd plant by this process to
produce a high quality water is 20 cents per 1000 gal. with a
total estimated investment of $4,000,000. Expanded-bed
operation is slightly less costly than packed-bed operation.
If tertiary facilities were added to a conventional treatment
plant to provide an effluent of the same high quality, most
of the foregoing costs would be incurred in addition to the
costs for the conventional primary—secondary treatment.
This report was submitted in fulfill itent of Prograxt No. 17050 D L,
Contract No. 14-12-459, bet en the Fe ra1 Water Quality dministration
and F1’C Corporation.
-------
INTRODUCTION
A previous investigation 1 conducted by the FMC Corporation
for the Federal Water Pollution Control Administration demon-
strated that expanded-bed operation of granular activated
carbon contacting systems offers certain advantages over fixed
or packed-bed operation for the advanced treatment of secondary
effluents. Upon examining the results of other preliminary
studies on the direct physicochemical treatment of raw sewage 2 ,
and on consideration of the nature of the organic matter present
in raw sewage relative to that in secondary effluent, the
conclusion was drawn that activated carbon treatment of sewage
without prior subjection to biological treatment could be
highly effective and that the expanded-bed mode of contacting
might have considerable advantage for such treatment. As a
result, the treatment of settled sewage or primary effluent
by carbon adsorption was investigated on a pilot scale and,
as in the earlier investigations with secondary effluent,
expanded-bed adsorption systems were compared with fixed-bed
systems, for which some other experience had been reported in
the literature.
Research on an essentially physicochemical method of
treatment of sewage is highly desirable because it has become
increasingly apparent over the past several years that achieve-
ment of high levels of water quality demanded by progressive
water use and reuse requirements, and by requirements for more
effective water pollution control, necessitates the application
of improved techniques for waste water treatment. Conventional
secondary biological treatment processes do not provide the
degree nor the consistency of treatment requried for most
water reuse applications, nor do they provide a completely
satisfactory means for protecting natural waters from pollution
by waste discharges.
Well operated, modern biological sewage treatment plants
can provide approximately 95% removal of suspended solids (SS)
and reduction of BOD to 20 mg/i BOD but have difficulty main-
taining this level of treatment on a continuous basis. Although
the quality of the effluents from such plants has been considered
adequate to meet most discharge regulations and effluent standards
in the past, increased concern over the quality of surface waters,
and the problems of meeting increasing water demands from a
relatively fixed total water resource, have resulted in more
stringent demands for better water quality and more effective
pollution control. As a result, significant interest has focused
over the past decade on development of physicochemical processes
capable of consistently accomplishing the degree of treatment
required by more stringent effluent standards.”’ 5
Research and development on advanced physicochemical
processes for wastewater treatment has primarily been centered
on providing tertiary treatment for wastes which have already
undergone conventional secondary biological treatment. The
addition of tertiary-level physicochemical processes to
—l
-------
conventional secondary-level biological processes incurs sig-
nificant additional treatment expenses. Further, the effective
operation of a tertiary treatment system is dependent on the
consistent and efficient operation of the biological secondary
process, which remains subject to problems arising from changes
in waste composition, from large variations in flow which have
often had to be diverted, and from the presence of toxic materials
which disrupt biological oxidation processes.
The present work represents a major diversion from the
traditional concept of tertiary treatment and examines the use
of expanded-bed adsorption as a direct application of the physico-
chemical process for treatment of primary waste.
The concept of applying such a treatment directly to a
primary waste, rather than to a secondary effluent, derives
partially from observations regarding the apparent difficulty
of removing final traces of organic matter from secondary
effluents by treatment with activated carbon, as well as from
the relative economics of two—stage vs. three—stage treatment
systems. Several investigators have reported leakage of certain
organic fractions through activated carbon columns when the
latter were used to treat secondary effluent. 6 ’ 7 The nature of
this leakage is not exactly known, but, there is strong indication
that it is comprised partially of nonadsorbable bacterial cell
fragments and partially of small organic molecules which have
been extensively hydrolyzed in the biological treatment stage
and thus rendered more soluble and less subject to adsorption.
These observations suggest that a primary wastewater might
therefore be more suitable for direct treatment with activated
carbon than it would be after having undergone biological
treatment. Laboratory studies were performed several years
ago to test this hypothesis and the results indicated that high
levels of removal of organic material could be obtained. 2
—2—
-------
OBJECTIVES
The objectives of this project were to determine, on a
pilot scale, the feasibility of removing organic materials
from primary effluent or raw sewage using granular activated
carbon in expanded-bed adsorbers. Within this overall project
objective, specific experimental objectives were:
• To obtain sufficient data on a pilot scale for the
expanded—bed contacting system to demonstrate
effective performance, to prepare a preliminary
design of a treatment process, and to estimate the
cost of the process.
• To compare the performance of expanded—bed and
fixed-bed adsorption systems.
• To obtain data on the effects of pretreatments and
post-treatments for removal of suspended solids.
• To determine the need for aeration in the carbon-bed
systems to prevent septicity and to remove biological
materials from the carbon particles.
—3—
-------
SUMMARY AND RECOMMENDATIONS
The results of the experimental pilot-scale studies
presented in this report indicate that expanded-bed activated-
carbon adsorption can consistently, effectively, and relatively
economically treat a chemically coagulated and clarified raw
sewage or primary effluent to produce a clear high quality
effluent. This mode of adsorption demonstrated a high degree
of utilization of the capacity of the activated carbon, and a
potential for using biological activity in the adsorbers to
add to the organic removal capacity of the carbon bed, thus
reducing carbon dosage and frequency o- regeneration.
Pretreatment by chemical clarification, prior to adsorption,
demonstrated excellent capacity for solids removal, for removal
of a significant fraction of organic matter classified by
membrane filtration as soluble, and for effective removal of
phosphates.
Combining the two physicochemical processes of chemical
clarification and adsorption offers the potential for achieving
economical conversion of sewage to a reusable, non-polluting
water by removal of 95-97% of its organic matter, essentially
all of its suspended solids and turbidity, and in excess of
90% of its phosphate content. Cost estimates based on current
(1969) values indicate that this process can be applied in a
lO-mgd plant for a total cost of 20 cents per 1000 gal. of
product water, using a two-stage expanded—bed carbon contacting
system. A less detailed cost estimate presented in October
1967 for a primary-secondary-tertiary treatment system which
would produce an equivalent effluent, indicated 23 cents/l000
gal. for a 15-mgd plant.
The expanded—bed activated—carbon adsorber offers several
advantages, including:
little or no cleaning or backwashing required;
• low, and constant feed pressure requirements;
• the potential for simple aeration between stages,
if required; and,
the potential for promotion of biological growth
on the carbon without rapid plugging of the bed.
On the basis of these encouraging results, it is
recommended that:
• The expanded-bed activated-carbon adsorption treatment
be demonstrated on a variety of clarified sewage and
waste streams.
The preliminary treatment plant design and cost
estimate presented here be subjected to more rigorous
—4—
-------
design and analysis and be compared with other
potential processes for achieving an equivalent high
degree of organic, solids, and nutrient removal.
Process improvements be investigated to further increase
the benefit—cost ratio of such sewage treatment.
Means of capitalizing on the beneficial effects of
biological growths in expanded carbon beds be
thoroughly investigated.
• Further studies be carried out on the use of a rapidly
regenerated, fresh carbon polishing filter to reduce
the organic content of the effluent from a carbon
adsorption treatment to the ultimate low level on a
consistent and long-term basis.
—5—
-------
EXPERIMENTAL
TEST PROGRAM
The pilot plant test program for comparing the expanded
and packed-bed carbon contacting systems was carried out in
several phases as shown in Figure 1.
In the first phase, the effect of primary effluent pre-
treatment on the performance of expanded and packed beds of
activated carbon for removal of organics was evaluated by
using untreated and chemically clarified primary effluent feeds.
In the second phase, the expanded and packed beds were
compared in an extended operation with chemically clarified
primary effluent as feed.
In the third phase, post-clarification of the effluent
from expanded-bed and packed-bed contacting of primary effluent
was studied for comparison with results obtained in the second
phase.
Supporting laboratory experiments were conducted to select
coagulants, develop parameters for the design of the clarification
system, and compare clarification prior to and following carbon
treatment.
Two special small—scale experiments were conducted: 1) to
study the effect of particle size on adsorption rate; and, 2) to
evaluate the use of a small bed of fresh activated carbon as a
final polishing treatment to produce a water with a minimum
organic content.
APPARATUS AND PROCEDURE
Preliminary Studies of Coagulation and Clarification
Standard laboratory jar tests were conducted on a Phipps-
Bird gang stirrer with six metal blade stirrers, 1 x 3 in. All
tests were run in 1—liter beakers using 1-liter samples of
primary effluent. Visual observations of the process at specific
time intervals were used to check the various steps. Turbidity,
pH and other required analyses were run by procedures described
later in this report.
A variety of inorganic coagulants and organic polyelectrolyte
coagulants were evaluated over a range of dosages, alone and
in various combinations for effectiveness on ELSA primary effluent.
Test conditions involved rapid mixing at 180 rpm for 15 minutes
followed by slow mixing at 20 rpm for 15 minutes. The samples
were then allowed to settle for 30 minutes before measurement
of turbidity for evaluation of the degree of clarification.
To develop parameters for designing the pilot scale
clarification system, the effects of mixing and flocculation
were examined at constant dosage of the selected coagulant. A
—6—
-------
PILOT PLANT OPERATION
PHASE I PRIMARY EFFLUENT & CLARIFIED FEED
PHASE II CLARIFIED PRIMARY EFFLUENT FEED
PHASE III PRIMARY EFFLUENT FEED
POST CLARIFICATION LABORATORY STUDIES
CARBON PARTICLE SIZE EXPERIMENT
SHORT BED FINAL CARBON TREATING
LABORATORY TESTING AND OTHER STUDIES
12-FT BEDS
24-FT BEDS
24-FT BEDS
FIGURE 1
TEST PROGRAM FOR STUDY OF TREATING PRIMARY EFFLUENT
AND CLARIFIED PRIMARY EFFLuENT IN EXPANDED AND
PACKED BEDS O’ ACTIVATED CARBON
YEAR
MONT H
1968 1969
NOV. DEC. JAN. FEB. MAR. APR. MAY JUN. JUL. AUG. SEP.
I I I
F
I
I I
I I
I I
-------
series of tests was conducted in which stirring speeds of 80
and 160 rpm and times of 0 to 8 minutes were used for the rapid
mix period, and 20 to 80 rpm and 5 to 15 minutes for the slow
mix period. Thirty minutes settling time was allowed before
decanting supernatant for measurement of turbidity.
Pilot Plant Installation
The pilot plant used for the experimental work was the
same as that described in the report of the preceeding study
on the treatment of secondary effluent in expanded-bed adsorbers.’
It was located at the sewage treatment plant of the Ewing-
Lawrence Sewerage Authority (ELSA) near Trenton, New Jersey
which serves residential, commercial, and industrial areas
within the two townships of Ewing and Lawrence. The sewage
consists of about 25% industrial waste and 75% domestic
waste. This is a trickling filter plant which includes four
circular, primary sedimentation basins. For this study,
primary effluent was taken from one of the sedimentation basins
and siphoned through a 300-ft long, 1-1/2 in. polyethylene
pipe to a pump at the pilot plant site. Primary effluent was
used because of the presence in the raw sewage of varying
amounts of solids which would have been difficult to handle
in the small lines and valves of the pilot system. The feed
to the pilot plant, therefore, consisted of settled raw
sewage and, in addition, the liquid from return sludge and
supernatant from the anaerobic sludge digestor.
The experimental apparatus at the pilot plant was set
up on a poured concrete slab installed for this purpose next
to the ELSA return pump building. The coagulation-clarification
system, pumps, and controls were located in a 10-ft x 16-ft
building constructed on the slab. The carbon adsorbers were
internally coated steel-pipe columns resting on the slab and
supported by an angle iron frame. The filter and other tanks
were located on the slab next to the building. Connections
between the columns and valves were rubber hoses which were
passed through the building walls. Photographs of the apparatus
at the pilot plant site are shown in Figure 2.
Clarification System
The clarification system consisted of two 55-gal. drums
for rapid mix, coagulation and floccuLation followed by an
up-flow clarifier and dual media filter, as shown in Figure 3.
The primary effluent was pumped through a float valve which
controlled the level in the rapid mix compartment in the upper
part of the first 55-gal. drum. The coagulant, a 30-weight
percent aqueous solution of ferric chloride, was fed by a
metering pump into the stream of primary effluent, discharging
from a 5/8—in, nozzle into an elbow to impart a circular motion
in the rapid-mix compartment. A motor—driven propeller was
used to provide additional rapid mixing. After an average
—8—
-------
FIGURE 2a PACKED-BED AND EXPANDED—BED ADSORPTION COLUMNS
AND APPURTENANCE. BUILDING IN BACKGROUND HOUSES
CLARIFICATION UNITS, PUMPSI CONTROLS, AND AUTOMATIC
SAMPLING EQUIPMENT
UPFLOW CLARIFIER AND PUMPS
FIGURE 2b
FIGURE 2c FLOW METERING AND CONTROL PANEL
FIGURE 2 PHOTOGRAPHS OF PILOT PLANT INSTILLATIUrI
-------
RAPID NIX
CHAMBER
0
PRI MARY
EFFLUE
PUMP
DUAL MEDIA CLARIFIED FEED
5.5gpm FILTER
38id.
UP-FLOW CLARIFIER
45id.4 STRAIGHT SIDE
60’ CONE BOTTOM
RESERVOIR
5.SQpm
©
L LOCCULATORS ’
BACKWASH
COAGU LANT
PUMP
Fe C 13
9 COAL’
9 SAND
61 COARSE SAND
BACKWASH
FIGURE 3 FLOW DIAGRAM OF CLARIFICATION SYSTEM
-------
detention time of two minutes at the design flow of 5.5 gpm,
the mixture flowed down into the lower part of the first drum
for flocculation with slow mixing, into the bottom of the second
drum for further flocculation with slow mixing, then out at the
top and into the clarifier. The slow stirring in both tanks was
provided by 24-in. x 2-in. x 1/2-in, redwood paddles mounted, with
the 24-in, dimension in the vertical position, at a 7-in, radius
to a vertical shaft driven by a constant speed motor. The first
flocculation tank was also fit with three vertical stators made
of redwood, 24-in. x 1-1/2-in. x 1/2-in., attached to the side
of the tank. The detention times at 5.5 gpm were 7 minutes in
the first flocculation chamber and 9 minutes in the second. The
motor—driven paddles in these tanks could be operated at various
speeds to provide different degrees of mixing.
The up-flow clarifier designed for this project was a
400-gal. capacity, shop fabricated, steel cylindrical tank,
3-ft 9-in, in diameter with a 4-ft high. straight section and
a 600 cone bottom. Flocculated water entered a central 8—in.
diameter chimney, discharged at a depth of about 4-ft below the
surface, then flowed upward to the overflow trough at the
surface. The detention time at 5.5 gpm was approximately 1 hour.
The product water from the clarifier flowed to the dual-media
filter, which consisted of 9-in, of anthracite coal (effective
size 0.59 mm) over 9-in, of filter sand (effective size 0.62 mm)
supported on gravel with a pipe underdrain. Filtered water was
pumped to a 250-gal. reservoir to provide feed to the pumps
for the carbon column systems.
Sludge was pumped from the clarifier by a positive pressure
pump attached to the bottom of the cone and operated by a cycle
timer to remove and discard sludge at predetermined intervals.
Usually, this pump was operated for 1 minute at a time, three
times per hour to discharge about 15-gal. per hour of sludge.
Carbon Adsorption Systems
The carbon adsorbers were vertical columns constructed of
internally coated 10-in, diameter steel pipe as shown
schematically in Figure 4. Each column was charged with
85 pounds of 12 x 40 granular activated carbon* which provided
a 6-ft deep bed of settled carbon. The carbon was supported
on a 6-in, layer of gravel and coarse sand over an inverted
5-in, diameter cone-shaped distributor. The columns designed
for packed bed operation were 9-ft tall to allow a backwashing
and disengaging zone. Those designed for expanded bed operation
were 12-ft tall including at the top a 1-ft section of li-in.
i.d. transparent pipe to permit observation of the effluent from
*pittsburgh CAL activated carbon, Calgon Corp., Pittsburgh,
Pennsylvania.
— 11 —
-------
___________ HOSE CONNECTION FOR
_________ ( il COLUMN FEED OR PRODUCT
II in. O.D. ACRYLIC PIPE
3/8 in. WALL
EXPANDED BED ONLY
1/8 in. RUBBER GASKETS
7U’)
3/4 in. BOLTS
lOin. STEEL PIPE l/4in.WALL
1501b. FLANGES
BED DRAIN
3/4 in. I.P. COUPLING
HOSE CONNECTION FOR
COLUMN FEED OR PRODUCT
PLASTIC CONE 5in. DIA.
COVERED WITH GRAVEL.
FIGURE 4
PILOT ADSORBER COLUMN DETAIL
— 12 —
-------
the expanded carbon beds and to provide a visual check to insure
that carbon was not being washed out of the column. Hose
connections were provided at the top and the bottom of the columns.
The behavior of the 12 x 40 carbon in an expanded bed
in the column at 5 gpm/ft 2 is shown in Figure 5.
Four constant displacement pumps driven by electric motors
through variable speed drives were installed so that feed could
be supplied to as many as four carbon adsorber systems. Reinforced
rubber hose, 5/8-in. i.d., was used for the connecting lines to
and between the columns to provide for ease of installation and
rearrangement. All of the flow controls, including in-line
valves, pressure gauges, flow meters (water meters), and solenoid
sampling valves were mounted on one central operating panel
within the building. Stream flows were maintained at constant
rate by adjustment of the pump drives. The treated water from
the expanded-bed columns was discharged into a drum so that any
carbon particles carried out could be collected and returned to
the column if necessary. All product water was returned to
the sewage treatment plant.
The entire system was designed for essentially automatic
operation. A technician visited the plant daily to take samples,
adjust flows and perform any routine maintenance required.
Twenty-four hour composite samples of the primary effluent
feed to each carbon column and the product water from each were
collected automatically. Timer—controlled solenoid valves
opened at 20-minute intervals to draw 100-ml samples. When
composite samples were collected for periods longer than one day,
appropriate adjustments were made either in the time the valve
was open or in the interval between samples. These samples
were composited in 5-gal. polyethylene bottles in an acid medium
to maintain stability and prevent deterioration or biological
activity over the sampling periods. Spot samples were collected
by hand.
Chemical and biochemical analyses were performed at the
FMC Chemical Research Center. Analytical determinations on
composited samples included TOC, SOC, and SS. BaD, turbidity,
phosphates and nitrates were run on spot samples brought unacidified
to the laboratory for immediate analysis. Analytical procedures
are discussed later in this Section.
The packed beds required frequent backwashing to dislodge
the collected solids that caused buildup of pressure drop. The
most effective procedure found for cleaning the carbon beds
consisted of lowering the water level, injecting air into the
bottom of the column for 5 to 10 mm. and then backflushing
with clean water to sweep away the dislodged solids. The
sludge resulting from the carbon cleaning operation was collected
in a drum so that any activated carbon lost from a column could
be returned to that column.
— 13 —
-------
too -
Particle Size
12/40
Static
— Region
I
ion ‘
l0
Tern p era t u re
o to°c
• 20°C
A 25°C
Water Flow Rate, gpm/ft 2
FIGURE 5
BEHAVIOR OF EXPANDED BEDS OF ACTIVATED CARBON
‘75
0,
/
Total Bed
Height
0
0
8
0
A
125 - 0
A
0
-C
C)
150
I
CD
0
(0
. 4 -
U)
9-
0
C
CD
0
L
CD
a
.4-
- C
C)
CD
I
V
Mix ng
Reg ion
A
75
0
-7
6
.4-.
Li..
CD
5U 1
C t,
C
E
40
0
E
0
L
‘4-
0
C
I t ,
. 4-
LI
20
x
0
L
0 .
0.
1<
A
50 -
0
0
25
0
0
5
— 14 —
-------
Phase-One Operation: Pre-Clarification
Parallel studies under the first phase of the
experimental program examined the relative effectiveness of
treating primary effluent by activated carbon with and without
pre—clarification. These studies were conducted at the pilot
plant facility over an operating period of approximately 75
days. Four adsorption systems were used; two systems operating
on unclarified primary effluent and two on chemically clarified
effluent, as shown in Figure 6. Each carbon system contained
a 12-ft depth of activated carbon, 6-ft in each of two columns.
Two systems were operated in expanded—bed mode, while
the other two systems were operated in packed-bed mode. Each
system treated approximately 3600 gal. of primary effluent per
day, corresponding to a carbon column loading of approximately
5 gpm/ft 2 of column cross section area.
The rate of flow to the expanded-bed and packed-bed
systems operating on primary effluent was increased to 7 gpm/ft 2
for the final part of this experiment.
Phase—Two Operation: Extended Treatment
In the second phase of the program, chemically clarified
primary effluent was treated in expanded-bed and packed-bed
adsorbers for an extended period of continuous operation of
125 days. The carbon columns were arranged in two sets, each
set with four columns in series to provide 24-ft of activated
carbon in the systems, as shown in Figure 7. For this test
fresh 12 x 40 activated carbon was charged to each column,
and clarified effluent was pumped to the carbon beds at about
5 gpm/ft 2 , or about 3600 gallons per day to each system. The
expanded-bed and packed-bed systems had each treated about
450,000 gallons of clarified primary effluent when the run was
terminated to allow time for Phase Three experiments.
Phase-Three Operation: Post-Clarification
At the conclusion of the 125 day run with clarified
primary effluent, the two four—column adsorption systems still
containing the same carbon, were switched directly to primary
effluent and operated for an additional 30 days. During this
period, post-clarification was studied in the laboratory and
by feeding the adsorber effluent to the chemical clarification
unit.
Particle Size Studies
Two parallel glass columns, 6—in. diameter x 9—ft tall,
were charged with 4-ft settled depths of an experimental activated
carbon of widely different particle size ranges. These columns
were operated in expanded mode with clarified primary effluent
as feed.
— 15 —
-------
PACKED
ADSORBERS
ADSORBERS
CHEMICAL
CLARIFICATION
0 i
EXPANDED PACKED EXPANDED
FIGURE 6 EXPERIMENTAL SET-UP FOR 12-FT CARBON BEDS
-------
PRIMARY EFFLUENT T HEMICAL FILTER
CLARIFICATION
I I
I -I
I
I
EXPANDED-BED ADSORBERS PACKED BED ADSORBERS
FIGURE 7 EXPERIMENTAL SET-UP FOR 24-FT CARBON BEDS
-------
To prepare the carbon in the two particle size ranges
studied, a batch of experimental activated carbon was passed
through a screening apparatus and separated into several fractions.
The two fractions selected for this experiment nominally 8 x 16
and 50 x 100, are described in Table I. Particle size 8 x 16
includes mostly particles passing a U. S. Standard Sieve No. 8
and retained by a U. S. Standard Sieve No. 16; 50 x 100 particles
include mostly those passing a No. 50 sieve, and retained by a
No. 100 sieve.
Separate rotameters and valves were used to regulate
the feed rate to each test column from a single pump. The
flow rate used for both columns was dictated by the maximum
rate that would retain the expanded bed of 50 x 100 carbon
in the adsorption column. With the 9-ft column and a settled
depth of 4-ft, the bed expansion was limited to 100%, which was
obtained at a specific flow rate of 1 gpm/ft 2 . At this flow
rate no expansion was observed in the bed of 8 x 16 particles.
Spot samples of the feed and product from the columns were
taken daily for TOC analyses.
Polishing Column Studies
A glass column with 1-in 2 cross section was filled
to a depth of 1-ft with fresh 12 x 40 activated carbon and
operated in a packed—bed mode with product water from the
24—ft expanded bed adsorber treating clarified primary effluent.
Feed rate was controlled by a valve from an elevated supply
tank. Composite samples of the feed and product were collected
at frequent intervals during a day of testing, and preserved
with acid for TOC analysis the next day. This polishing
column concept was examined in four separate runs on four
different days with fresh carbon in the test column for each
run.
ANALYTICAL METHODS
Suspended solids concentration of the primary effluent
and treated water samples was measured by a procedure 8 involving
the use of 0.45—micron membrane filters. Before use, the
membrane filters, which in manufacture are treated with an
organic conditioning agent, were washed in distilled water to
remove this agent. They were then dried in individual
desiccators to constant weight. After filtration of a sample,
each filter was dried again in the same desiccator and weighed
to determine weight gain by retention of suspended solids. The
filtrates were collected to provide samples for SOC analyses.
The Beckman Carbonaceous Analyzer was used for organic
carbon analysis on well-mixed composite samples, directly for
TOC, and, after filtration through the membrane filters, for SOC.
— 18 —
-------
TABLE 1
Properties of Activated Carbons Used in The Studies
of The Effect of Particle Size on
Adsorption of Organics From Clarified Primary Effluents
Carbon Designation 8 x 16 50 x 100
Sieve Analysis , U.S. No. % U.S. No. %
% Retained 8 0.0 40 0.4
10 0.8 50 0.6
12 7.4 80 57.8
16 67.2 100 22.0
20 16.0 140 16.8
pan 8.6 pan 2.4
Average Particle Size, mm 1.4 0.2
Bed Expansion , Rate Rate %
% of Settled Depth 0.45 0 0.45 45
at Rates in gpm/ft 2 0.95 0 0.95 100
2.7 1
5.0 2
Adsorptive Properties ,
Iodine No. mg/g 507 605
Methylene Blue mg/g 110 160
— 19 —
-------
Determination of BOD of unacidified spot samples was
performed by the dilution procedure described in Standard
Methods. 9
All turbidity determinations were made with a Hach Model
2100 Photoelectric Turbidimeter.
Total phosphate was determined by ASTM Procedure D515-
60T— on samples after digestion to convert all phosphate to
the ortho form.l°
Nitrate was determined by ASTM Procedure D992-52.-
Munonia and organic nitrogen were determined by Kjeldahl
Procedure as outlined in Standard Methods.9
— 20 —
-------
RESULTS
COAGULATION AND CLARIFICATION STUDIES
Coagulant Selection
The most effective reduction in the turbidity of the
primary effluent obtained in the laboratory jar tests was
generally from a value in excess of 30 Jackson Turbidity Unit
(JTU) to a level of about 1.0 JTU. The dosage ranges for
inorganic coagulants providing better than 90% turbidity
removal for the several samples of primary effluent examined
were:
Alum — 200 to 300 mg/i as Al 2 (SO ) 3 •18H 2 O
Ferric Chloride - 200 to 350 mg/i as FeC1 3 ’6H 2 O
Lime 300 or more mg/i as Ca (OH) 2
These coagulant dosages were consistently effective for
clarification of primary effluent samples collected on several
different days, even though the samples of primary effluent
analyzed during this period varied in p11 from 7.2 to 8.8, and
in alkalinity from 200 to 300 mg/i.
Ferric chloride and lime both produced rapidly settling,
strong floc particles, These flocs were resistant to breakage
during transfer and reformed rapidly after dispersion to settle
and provide a clear supernatant. Ferric chloride produced a
neutral water; lime.a strongly alkaline water with a pH of
9.5 or more. Coagulation with alum was slower, required closer
pH control, produced a more fragile floc, and did not produce
as clear a supernatant. Combinations of lime with ferric
chloride were found to be effective.
Several anionic, nonionic, and cationic organic
polyelectrolytes were tested in doses ranging from 0.1 to
5 mg/i, but none was found to’be particularly effective as
a primary coagulant. As secondary floc strengtheners, used
in combination with iron or alum, some of the polyelectrolytes
produced larger floc particles which settled more rapidly than
those formed with alum or iron alone. Turbidity removal was
generally no better for the combination, however, and the
dosage of primary coagulant could not be reduced much below
that required when only the metal salt was used.
Ferric chloride used alone gave consistently good
clarification and rapid settling of floc, and produced an
essentially neutral clarified water with a PH from 6 to 7.
For these reasons and also because of the relative ease of
handling and feeding the small quantities required, ferric
chloride was selected for use as the coagulant for the pilot-
plant operations, although lime may well be preferred for
larger scale operation.
— 21.. —
-------
Coagulation Conditions
The results of the tests for selection of optimum
conditions for use of the ferric chloride coagulant may be
summarized by the following observations:
• Initial rapid mix for a short period of time was
found desirable for thorough dispersion of the
reagent without adversely affecting the later steps.
• Extended high speed mixing restricted the growth of
floc particles during subsequent flocculation under
conditions of slow stirring.
• In the flocculation step, mixing at excessively high
speeds (80 rpm) resulted in small, poorly settling
floe particles; excessively low speeds (20 to 30 rpm)
produced growth of larger particles which were less
effective for turbidity removal.
• Moderate agitation (40 rpm) after a short rapid mix
seemed most effective for good coagulation and
turbidity removal by producing a large number of
relatively small, dense floc particles.
• A final slower mixing (20 rpm) to permit flocculation
of the particles formed during coagulation at moderate
speed after an initial rapid mix, appeared to offer
an advantage.
Design of Pilot Plant System
The pilot plant clarification system was designed to
provide a series of mixing conditions similar to those found
to give best results in the jar tests. To translate the
laboratory scale stirring conditions to the pilot plant design,
the degree of agitation was expressed in terms of the velocity
gradient, G, which is a measure of energy dissipation in a
mixing or stirring system. 11 ’’ 2 Values for G were calculated
for the conditions used in the jar test assuming a drag
coefficient of 1.2 for metal blades, and water velocity equal
to 3/4 of the paddle-tip radial velocity. Values for the
velocity gradient calculated for paddle speeds up to 180 rpm
ranged to 90 fps/ft, as illustrated in Figure 8, with values
reported by Hudson’ 2 for comparison. While there were
inconsistent results, the experiments which gave the best
clarification were operated at values of G in the range of
10 to 20 fps/ft.
The velocity gradient in the rapid mix. section of the
pilot plant system cannot be estimated accurately, but to
approach the high velocity required, a combination of jet
mixing action and the propeller recirculation was used. Values
— 22 —
-------
120
100
80
60
40
20
‘4-
‘I )
Q.
‘4-
a
I—
z
Li
I—
0
0
-J
Li
>
FIGURE 8
RPM OF PADDLE
VELOCITY GR/\DIENT vs. PADDLE RPM IN J/ R TEST APPARATUS
40 80 120 160
200
— 23 —
-------
for the velocity gradient were calculated for the pilot plant
coagulation system assuming a drag coefficient of 1.8 for the
wooden paddles and a water velocity equal to 3/4 of the radial
velocity at the center of the paddles. Figure 9 shows the
relationship between paddle rate and the calculated values for
G. The system was operated continuously at 18 rpm for the
entire experimental period.
The up—flow clarifier for the pilot plant system was
designed to provide a settling period of at least 30 minutes
with minimum agitation and effective retention of the floc
in a floating sludge blanket.
Operation of the Pilot Plant System
The pilot scale coagulation—clarification system provided
good removal of suspended solids over a large percentage of the
operating periods, as may be observed from the clarified effluent
data presented in Figures 10 and 11. There were, however, during
the Phase—One experiment, certain periods when coagulation with
the ferric chloride was poor. During these periods the pH of
the primary effluent was observed to be as low as 6.6 compared
to the usual pH of 8 or more. Alkalinity measurements on
several samples during these periods indicated values as low
as 100 ppm as CaCO 3 , compared to the usual levels of 200 to
300 observed during other times. Addition of sodium carbonate
in solution to the mix tanks to provide an additional alkalinity
of about 100 mg/i as CaCO 3 during periods of low pH restored
good flocculation and clarification.
Further results on the effectiveness of the coagulation-
clarification for removal of organic matter (TOC, SOC and BOD)
and phosphate within the context of the overall treatment
sequence are presented in the following sections of this report.
These results may be summarized by stating that the chemical
clarification system was highly effective •not only for removal
of suspended matter and production of a clear feed to the
carbon columns, but also for removal of dissolved organic matter
and phosphates.
PHASE-ONE OPERATION: PRECLARIFICATION
Clarification
Results of analyses for TOC and SOC for the primary effluent
and chemically clarified effluent are given in Figures 12 and
13. A consistent reduction of TOC resulted, as might be expedted
because of the removal of suspended matter. However, in addition,
a consistent removal of SOC was observed.
The weights of TOC and SOC removed in the clarification
system, presented in Table 2, were calculated from the volumes
treated and the analyses of the feed streams to the carbon
— 24
-------
30
RPM OF PADDLES
FIGURE 9
VELOCITY GRADIENT vs. PADDLE RPM FOR DRUM FLOCCULATORS
25
20
I-
z
w
15
I0
>.
I—
0
-J
U i
>
5
5 10 15 20
25
— 25 —
-------
80
E
C,)
-J
0
C l)
I ii
z
LU
Cl)
C l)
28 13
CLARIFIED PRIMARY
EFFLUENT
DATE, FEBRUARY
FIGURE 10
REMOVAL OF SUSPENDED
SOLIDS FROM
PRIMARY EFFLUENT
U
LEGEND:
PACKED
0 EXPANDED
BED PRODUCT
U
BED PRODUCT
OPEN-CLARIFIED FEED
U
U
£
U
£
BLACK-UNCLARIFIED FEED
UU
U
£
60
40
20
£
£
£
A
a
aa
£
IMARY EFFLUENT
U
a
A
a
a
a
0
a a
£
a
a
0
U
a
U
U
a
a
U
a
a
a
0
a
0
0 17 24 3 10 17 24 31 2 7
MARCH
APRIL
BY CLARIFICATION AND ACTIVATED CARBON BEDS
-------
FIGURE 11 REMOVAL OF
SUSPENDED SOLIDS FROM PRIMARY
EFFLUENT
-J
0 )
E
C,)
-J
0
C l)
w
z
w
0.
C,)
C,)
DATE, MAY JUNE JULY AUG
BY CHEMICAL CLARIFICATION
AND ACTIVATED CARBON BEDS
-------
I I I I
80-
60-
40-
20
O0_-
FIGURE 12
50
I I
100 150
VOLUME TREATED,
200 250
1000 gal.
REMOVAL OF TOTAL ORGANIC CARBON FROM PRIMARY
-j
E
z
0
4
C)
C)
z
4
CD
0
-J
4
I-
0
I-
PRIMARY EFFLUENT
CLARI Fl ED
PRIMARY EFFLUENT
EFFLUENT BY FERRIC CHLORIDE COAGULATION-CLARIFICATION
-------
I I
1
40
30
20
10-
0
I I I I I
0 50
100
150
2
00
VOLUME
TREATED,
1000
gal.
250
CLARIFIED
PRIMARY
EFFLUENr
FIGURE 13
REMOVAL OF SOLUBLE ORGANIC CARBON FROM PRIMARY
EFFLUENT BY FERRIC CHLORIDE COAGULATION-CLARIFICATION
t.J
-J
E
z
0
()
C-)
2
C D
0
w
-J
-J
0
C l )
PRIMARY
EFFLUENT
-------
TABLE 2
TOC and Soc content of Primary Effluent and
clarified Primary Effluent Feeds to Columns and
Effect of Coagulation With Ferric Chloride:
Phase One Pilot Operation
Period - January 27 to April 11, 1969
Organic Content of Untreated Primary Effluent Feed
Volume, Gal. lb TOC lb SOC
Feed to PCU* 296,275 150.0 76.6
ECU 321,612 162.4 82.8
Total 617,887 317.4 159.4
In Primary Effluent
lb/bOO gal. 0.514 0.258
Organic Content of Clarified Feed
Net Feed to Clarifier 510,068 262.0 131.6
To PCC 252,484 52.8 44.9
ECC 257,584 53.8 45.6
Total Product 510,068 106.6 90.5
Organics Removed ky Clarifier , lb 155.4 41.1
% 59.3 31.2
Note - “10% Wasted as Sludge
Ferric Chloride used = 666 lb = 1.3 lb/1000 gal. = 156 mg/i
*Activated Carbon Column Designations:
PC = Packed EC = Expanded U = Untreated Feed C = Clarified Feed
— 30 —
-------
column systems. The average concentration of TOC was reduced
from 60 to 25 mg/i, and Soc from 31 to 21.5 mg/i by clarification,
which thus removed an average of 59.3% of the TOC and 31.2% of
the SOC from the primary effluent.
Adsorption
The results of analyses for TOC and soc on unclarified
primary effluent and the treated water leaving both the
expanded-bed and the packed-bed systems, are presented in
Figures 14 and 15. The cumulative amounts of TOC and SOC
charged to and removed by the systems are shown in Figure 16.
In this plot, the slope of each line represents the fractional
removal of organic carbon. During this operation, the expanded
bed handled slightly more water than the packed bed due to the
down time of the packed bees for column cleaning, which was
necessary to remove accumulated suspended solids.
As noted previously, after approximately 2/3 of the 75
day operating period of this experiment had been completed
the rate of primary effluent feed to the columns was increased
from 5 gpm/ft 2 to about 7 gpm/ft 2 . At the higher flow rate,
the carbon beds continued to remove about the same proportion
of TOC ard SOC from the primary effluent.
At the termination of the run, the expanded bed had
removed 33% ar d the packed bed 43% of the weight of TOC applied,
and 3 % and 40%, respectively, of the weight of SOC.
The results of TOC and SOC analyses from the parallel
column systems operating on clarified primary effluent are
shown in Figures 17 and 18. The cumulative amounts of
organic matter charged to and removed by the carbon systems
are given in Figure 19 for TOC and Figure 20 for SOC. The
expanded and packed beds each removed over 50% of the weight
of TOC and SOC charged during the 75 days of operation. A
comparison of the values of TOC and SOC shows that most of
the organics present in the clarified feed and column products
consisted of SOC. At the time the test was terminated, both
systems were still removing 45% of the SOC applied, as indicated
in Figure 20.
Data for each of the four systems, prpsented in Table 3,
show the total volume treated ard the quantities of TOC and
SOC removed during the period of this test. Average concentrations
of TOC and SOC were calculated for the feed and the product
for each of the systems, as was the organic loading on the
activated carbon at the end of the experiment. The fraction
of TOC and SOC removed from the clarified primary effluent by
the carbon beds was greater than that removed from the
unclarified primary effluent.
— 31 —
-------
I
PACKED
BED
12’ DEPTH
100
VOLUME
TREATED, 1000
FIGURE 14 REMOVAL OF TOTAL ORGANIC CARBON
FROM PRIMARY
PRIMARY
EFFLUENT
-J
E
z
0
C-)
C-)
z
4
(9
0
-J
4
I—
0
I-
80
60
40
20
EXPANDED
BED
IN CREASE
50
RATE
5to7 gpm /ft 2
150
200
250
300
gal.
EFFLUENT IN BEDS OF ACTIVATED
CARBON
-------
I I I
I I
PACKED
50 100 150
VOLUME
TREATED,
1000 gal.
FI-GURE 15 REMOVAL OF
CARBON FROM PRIMARY
-J
E
z
0
4
C-)
0
z
4
0
U i
-J
cc
-J
0
( )
PRIMARY EFFLUENT
40
30
20
l0
:XPANDED
/
/
BE
BED
\ /
‘I
V
12’
DEPTH
INCREASE
RATE
5T07 9
pm/ft 2
200
250
300
SOLUBLE ORGANIC
EFFLUENT
IN BEDS OF ACTIVATED CARBON
-------
J: I I I I I I I I I I I I I
70
60
50-
-o
0
w
>
0
Li
z
0
3O
20
I0
I I I I
120 130 140 150 160
lb
PERFORMANCE OF ACTIVATED CARBON IN 12-FT BEDS
7 gpm ft 2
5 gpm ft
0
z
4
0
0
,
:XPANDED
BED
PACKED BED
SOLUBLE ORGANIC CARBON
TOTAL ORGANIC
CARBON
O 10 20 30 40 50 60 70 80 90
ORGANIC CARBON
FIGURE 16
I I I I I I __ I I I I I I
100 110
APPLIED,
FOR REMOVAL OF ORGANIC CARBON FROM PRIMARY EFFLUENT
-------
50 100 150
200 250
VOLUME TREATED,I000 gal.
FIGURE 17 REMOVAL OF TOTAL
ORGANIC CARBON FROM
I
3O
4
C)
CLAR FlED
(.aJ
U i
PRIMARY EFFLUENT
C)
z
4
(9
0
-J
0
I—
20
I0
S
S
S
ED BED
12’
DEPTH
CLARIFIED PRIMARY EFFLUENT
BY ACTIVATED CARBON
-------
100
150
200
VOLUME T
REATED, 1000 gal.
30
-J
[ 0
CLARIFIED
20
PRIMARY EFFLUENT
I
I
l0
0 50
PACKED BED
EXPANDED BED
12’ DEPTH
FIGURE 18
250
REMOVAL OF SOLUBLE ORGANIC FROM
CLARIFIED
PRIMARY EFFLUENT BY ACTIVATED CARBON
-------
I I I I I I I I
30
0
— 0 00
0 0
25- co
0 0
> D OD o0
o 0
00
U i
20 oo°
PACKED BEDS
a3
DD0 ,0 BEDS
C)
C)
410-
0
o
4
0 I I I I I I I
0 5 10 15 20 25 30 35 40 45 50
TOTAL ORGANIC CARBON APPLIED, lb
FIGURE 19 PERFORMANCE OF ACTIVATED CARBON IN 12-FT DEEP
BEDS FOR REMOVAL OF TOTAL ORGANIC CARBON FROM
CLARIFIED PRIMARY EFFLUENT
-------
I
1 -I I I
I I I
Do
PACKED BEDS P O
00
EXPANDED
—o
I I I I I I
BEDS
0 5 10
15
20
25
30
35
SOLUBLE
ORGANIC
CARBON APPLIED,
FIGURE 20
0
0 d o0
0 c0
lb
I I
40 45 50
PERFORMANCE OF ACTIVATED CARBON IN 12-FT DEEP
BEDS FOR REMOVAL OF SOLUBLE ORGANIC CARBON
FROM CLARIFIED PRIMARY EFFLUENT
w
LU
z
0
C.)
C.)
z
CD
0
LU
-J
-J
0
0)
25
20
15
10-
5
0
DOD
000
F
F
-------
TABLE 3
Removal of TOC and Soc From Primary Effluent
and Clarified Primary Effluent by Adsorption
in 12-Ft Carbon Beds: Phase One Pilot Operation
Period - January 30 to April 11, 1969
COLUMN PCU ECU PCC ECC
FEED Untreated Clarified
TOTAL VOLUME TREATED, gal. 296,275 321,612 252,484 257,584
Total Organic Carbon
In Feed, lb 150.0 162.3 52.8 53.7
Prod. 84.7 109.8 21.6 26.3
Removed 65.3 52.5 31.2 27.4
Avg. Conc. Feed, mg/i 60.7 60.5 25.1 25.0
Prod. 34.3 40.9 10.2 12.2
Percent Removed 43.5 32.3 59.1 51.0
Soluble Organic Carbon
In Feed, lb 76.6 82.8 44.9 45.6
Prod. 45.9 51.5 19.3 21.6
Removed 30.7 31.3 25.6 24.0
Avg. Conc. Feed 31.0 30.9 21.3 21.2
Prod. 18.6 19.2 9.2 10.0
Percent Removed 40.1 37.8 57.0 52.7
lb TOC per lb AC 0.38 0.31 0.18 0.16
lb SOC per lb AC 0.18 0.18 0.15 0.14
— 39 —
-------
The results of five—day BOD determinations are given
in Figure 21. Separate analyses indicated that the primary
clarifier removed about 50% of the BOD from the raw sewage.
The BUD of the primary effluent at the time the samples were
taken varied from about 90 to 40 mg/i. The chemical
clarification step consistently removed about 70% of the BOD
from the primary effluent to produce a clear water with an
average BUD of about 15 mg/i, varying from 5 to 29 mg/i, over
the period of the run. The BOD of the clarified and carbon
treated water ranged between 3 and 16 mg/i.
Samples were periodically analyzed for total phosphate.
The results of these analyses are presented in Figure 22 with
phosphate reported as P0 4 3 . As expected, the carbon bed
alone removed little or no phosphate from the untreated primary
effluent. With clarified primary effluent, the phosphate
was lower after carbon adsorption in most cases, which suggests
that some of the precipitated phosphate was carried out of the
clarifier arid trapped in the carbon bed. The clarification
of primary effluent with ferric chloride usually reduced the
phosphate to less than 5 mg/i from its initial concentration
of from 15 to 40 mg/i.
Analyses for various forms of nitrogen were also conducted
on the samples. While nitrate would not normally be expected
to appear in significant concentration in raw sewage or primary
effluent, several analyses indicated nitrate concentration of
5 to 15 mg/i, as NO 3 in the ELSA primary effluent. The city
water during this period was found by infrequent analyses
to contain as much as 5 mg/i. Nitrate was only minimally
affected by clarification but was removed essentially completely
on passing through the carbon bed, probably by biological
reduction. This removal of nitrate in the carbon beds was
not observed on the first samples analyzed, when there was
only fresh activated carbon in the system. In a few infrequent
araiyses for ammonia and organic nitrogen, it was found that
neither coagulation nor carbon adsorption had any effect on
the ammonia concentration of the primary effluent, which ranged
from 20 to 50 mg/i. Organic nitrogen was reduced by clarification
from about 4 to 7 mg/i in the primary effluent to 2 to 3 mg/i
in the clarified effluent, and after carbon treatment to about
1 mg/i. These removals were in about the same proportion as
the organic matter removals.
The composite samples collected from the operating systems
were filtered to provide a measure of the suspended solids
retained on membrane filters with effective openings of 0.45
microns. The results of these analyses were shown earlier
in Figure 10. Clarification consistently reduced the suspended
solid content to about 10 mg/i. The packed beds of activated
carbon removed additional suspended solids, as expected. The
expanded beds were also observed to remove some suspended
solids from the primary effluent, possibly by collecting these
— 40 —
-------
80-
LEGEND
EXPANDED BED PRODUCT-
PACKED BED PRODUCT
BLACK-UNTREATED FEED
OPEN—CLARIFIED FEED
30-
20-
10-
00
0 000
oBo°
CLARIFIED PRIMARY
EFFLUENT
FIGURE 21
FEB
I MAR I APR
REMOVAL OF BOD FROM PRIMARY EFFLUENT BY
CHEMICAL CLARIFICATION AND ACTIVATED CARBON BEDS
93
PRIMARY EF LULNI
70-
60-
U
040
.
U
U
U
.
.
.
U
.
U.
.
U
0
0
0
8
08
— 41 —
-------
35
Lv
• EXPANDED BED PRODUCT
£ PACKED BED PRODUCT —
BLACK-UNTREATED FEED
OPEN-CLARIFIED FEED
LEGEND:
FEB MAR APRIL
FIGURE 22
REMOVAL OF PHOSPHATE BY CHEMICAL CLARIFICATION
AND 12-FT ACTIVATED CARBON
•
30-
PRI MARY
EFFLUENT
_i25
E
20
C ,,
4
£
w
I-
4
=
( )
0
I
a-
a
15
I0
5
I
ARIFIED PRIMARY EFFLUENT
0
- LA
— 42 —
-------
solids in the lower portions of the bed where there was
relatively little bed expansion (see Figure 5).
The packed-bed systems required frequent cleaning to remove
the accumulated solids. The first and second columns operating
on primary effluent were cleaned 38 and 11 times; on clarified
feed, the first and second columns were cleaned 31 and 4 times,
respectively. During this same period of operation with primary
effluent, the first expanded bed required cleaning twice, the
second bed once. In these cases there appeared to be plugging
in the bottom of the column only, which did not affect the
operation of the expanded bed. With clarified feed, the
expanded beds did not require any cleaning.
PHASE-TWO OPERATION: EXTENDED TREATMENT
Clarification
The coagulation-clarification system worked well during
this operational phase, with only a few minor upsets caused by
blockage of the primary effluent line to the pilot plant site.
At no time during this phase of study was the alkalinity
insufficient for proper coagulation with ferric chloride, which
was used at an average rate of about 170 mg/i. The warmer
wastewater (about 15°C) during the spring —summer period
of operation was easier to clarify than the cold wastewater
(5-10°C) of the Phase-One experiments.
Results of TOC analyses of the primary effluent and
clarified primary effluent are given in Figures 23 and 24,
along with the cotresponding data for the effluents from the
carbon columns. The weights of TOC and SOC involved in the
clarification step are given in Table 4. During this period,
about 64% of the TOC and 33% of the SOC was removed by chemical
clarification. The concentrations of TOC and SOC were essentially
equal for the clarified effluent, demonstrating the effective
operation of the clarification system.
Adsorption
Because the TOC and SOC values for the clarified primary
effluent feed were nearly equal, only the TOC results are plotted
in Figures 23 and 24. The TOC data for the feed and the effluents
from each column in the four-column, 24—ft deep carbon bed systems
are also shown in Figure 23 for the expanded-bed adsorbers and
in Figure 24 for the packed-bed adsorbers. For about the first
72 hours of operation (185 gallons throughput), the fresh carbon
reduced the TOC of the effluent to 1-2 mg/i. For most of the
remainder of the 125 days (about 450,000 gallons) of operation
at constant flow, both carbon systems reduced the TOC to about
4 mg/i in the final product water. The weak primary effluent
near the end of the run resulted from periods of heavy rain in
the area. In Figure 25, the weights of TOC removed are plotted
against weights applied for the 6-ft, 12-ft and 24-ft bed depths
— 43 —
-------
50 100 150 200 250 300 350 400
VOLUME TREATED, 1000 gal.
FIGURE 23 TREATMENT OF PRIMARY EFFLUENT BY CLARIFICATION
-J
D I
E
z
0
4
0
0
z
4
0
-J
0
I-
60
50
40
30
20
I0
0
450
AND ACTIVATED CARBON IN EXPANDED BEDS
-------
I
60
-J
a’
E SO
z
0
0
0
I Z
30
I Q
-J
0
I—
I0
PRIMARY
CLARIFIED
150
VOLUME
200
250
TREATED, 1000
FIGURE 24 TREATMENT OF PRIMARY EFFLUENT BY CLARIFICATION
EFFLUENT
PRIMARY EFFLUENT
50
S
%
—
%
‘ I2 FT CARBON
\
100
gal.
350
400
450
AND ACTIVATED CARBON IN PACKED BEDS
-------
TABLE 4
Removal of TOC and Soc From Primary Effluent
by Chemical Clarification and Adsorption in
24-Ft Carbon Beds: Phase-Two Pilot Operation
Period - April 29 to September 2, 1969
Packed Beds Expanded Beds
TOTAL VOLUME TREATED, gal. 447,464 449,680
Avg. Avg.
Weight Conc. Weight Conc.
Total Organic Carbon lb mg/i lb mg/i
In Primary Effluent 172.5 46.3 173.0 46.3
In Column Feed 62.0 16.6 62.1 16.6
Removed 110.5 110.9
Percent Removed by Clarifier 64.1% 64.1%
Amount Removed in
First Carbon Bed 25.4 6.8 25.1 6.7
Four Carbon Beds 46.7 12.6 45.9 12.3
Percent Removed by Carbon 75.3% 74.0%
Average TOC of Product 4.0 4.3
Soluble Organic Carbon
In Primary Effluent 79.0 21.1 80.6 21.4
In Column Feed 53.8 14.4 53.9 14.4
Removed 25.2 26.7
Percent Removed by Clarifier 31.9% 33.1%
Amount Removed in
Four Carbon Beds 38.0 10.3 38.2 10.2
Percent Removed by Carbon 70.6% 70.8%
Average SOC of Product 4.0 4.1
Loading on Activated Carbon
TOC on First Bed lb/lb 0.30 0.30
TOC on Four Beds 0.14 0.14
SOC on Four Beds 0.11 0.11
— 46 —
-------
20 30 40 50
TOTAL ORGANIC CARBON APPLIED, lb
EFFECTIVENESS OF 24 FT OF
ACTIVATED CARBON FOR
.. . .
PACKED BED ADSORBERS
EXPANDED
BED
FT
LU
>
0
LU
z
0
cc
0
0
z
0
-J
F—
0
I—
ADSORBERS
40
30
20
10
0
I2FT
.
6FT
BED DEPTH -
0
l0
FIGURE 25
60
REMOVAL OF ORGANIC CARBON
FROM CLARIFIED EFFLUENT
-------
in the two adsorption systems. A summary of TOC and Soc values
for the entire run is presented in Table 4.
By the end of the Phase-Two experiment, the 85 pounds of
activated carbon in the first 6-ft bed in both the expanded-
bed and packed-bed systems had removed 25 pounds of TOC, thus
the TOC loading on the activated carbon was about 30% by weight.
The removals of BOD in the clarification and adsorption
systems during this experiment are shown in Figure 26. Chemical
clarification removed about 2/3 of the BOD from the primary
effluent, and the adsorption systems further reduced the BOD to
an average of less than 5 mg/i in the final product water. In
addition, there was an immediate oxygen demand of 1 or 2 mg/i
in most of the samples for which the data were obtained.
Phosphate removal for this experiment is given in Figure 27.
As expected, the ferric chloride reduced the phosphate content
by about 90 percent from an average of about 30 mg/i to an
average of about 3 mg/i. The activated carbon appeared to give
a slight additional removal of phosphate, probably due to
collection of floc particles and inorganic precipitates.
The removal of SS by clarification and adsorption was
presented in Figure 11. The chemical-clarification generally
reduced the SS content of the primary effluent to about 10 mg/i
which includes a significant amount of inorganic solids. The
same pattern is seen in the turbidity removal presented in
Figure 28. The spike showing high turbidity and suspended
solids in the clarified effluent in early June was caused by
a failure of the coagulant feed pump before this sample was
taken.
Both activated carbon systems accomplished further reduction
in turbidity as shown in Figure 28, producing clear effluent
with 1-2 JTU turbidity most of the time. The first packed bed
was cleaned 48 times during Phase Two, and the other three
beds 5, 2 and 1 times to remove the collected solids while
the expanded beds required no cleaning at all.
The Phase-Two experiment was conducted during the spring
and summer months, when the wastewater was warmer than during
the Phase—One experiment. This warmer water permitted
development of anaerobic conditions in both the expanded-bed
and packed-bed carbon systems. The reducing conditions in the
carbon beds were indicated by the presence of traces of H 2 S
in the carbon-treated water and by the formation of a slight
haze in this water upon standing for some time. In some
cases, the treated water contained sufficient ferrous iron to
give a floc after exposure to air. On a few occasions, the
iron content+ f the treated water was found to be as high as
5 mg/i as Fe, which on exposure to air would oxidize and
precipitate.
— 48 —
-------
60
50
1
E40
0
30
20
I0
FIGURE 26
REMOVAL OF BOO FROM PRIMARY EFFLUENT BY
CHEMICAL CLARIFICATION AND 24 FT ACTIVATED
CARBON
— 49 —
-------
30
25
20
15
I0
5-
0
a-
U)
a
-j
E
U i
I
a-
Cl)
0
=
a-
-J
0
I-
FIGURE 27
CLARIFIED PRIMARY EFFLUENT
0._EXPANDED BED
o< M Ao 0
PRODUCT
°PACKED BED
no
PRODUCT
MAY
JUNE
JULY
REMOVAL OF PHOSPHATE FROM PRIMARY EFFLUENT BY
CHEMICAL CLARIFICATION AND 24 FT ACTIVATED CARBON
PRIMARY EFFLUENT
— 50 —
-------
FIGURE 28
REMOVAL OF TURBIDITY FROM PRIMARY EFFLUENT BY
I—
-)
I—
I-
U i
40
30
20
I0
CHEMICAL CLARIFICATION AND ACTIVATED CARBON BEDS
-------
Several attempts were made to control the problems
associated with the anaerobic conditions in the columns.
Chlorine, as sodium hypochiorite, was added to the water from
the clarifier and appeared to be moderately effective in
reducing the H 2 S odor in the clarified primary effluent.
Aeration of the clarified effluent by bubbling air through
the feed tank to the carbon columns reduced the H S odor,
but also produced a biological floc in this tank, which tended
to plug the packed column more rapidly. Neither of these
solutions provided completely for control of the evolution
of H 2 S or the instability of the final treated water.
Injection of oxygen from a cylinder of the compressed gas at
a rate that would provide about 10 mg/i of 02 in the influent
to the columns, did reduce the evolution of H 2 S but did not
eliminate the instability of the treated water. Although the
injected oxygen usually provided a dissolved oxygen (DO)
concentration of 10 mg/i in the feed streams to the first
carbon beds, the concentration of DO after the first bed in
each system was 1 mg/l or less.
PHASE-TH.REE OPERATION: POSTCLARIFICATION
Adsorption
After termination of the Phase-Two experiment, the four-
column systems containing partially spent carbon were again
fed unclarified primary sewage, and the effluents from the beds
were evaluated for post-clarification. During this 28 day
period of operation no bed cleaning was required for the first
5 days. After that the first packed bed was cleaned 10 times
and the other three required cleaning 4, 4 and 2 times,
respectively. In spite of this almost daily cleaning, the
total pressure in the packed bed system frequently went to 75
psig as compared to a maximum of approximately 40 psig during
the Phase-One or Phase-Two experiments with clarified primary
effluent. Results of analyses for TOC and SOC during this
period are shown in Figures 29 and 30. Because of the down
time required for cleaning the packed beds, and the slower rate
of flow through the packed beds resulting from the higher
pressure drop, the expanded—bed system which did not require
cleaning treated about 15% more water over the same period.
The removal of TOC by the expanded-bed system progressively
declined until the total 24-ft was no more effective than the
first 6-ft of the packed bed. The removal of SOC indicated a
similar decline in the effectiveness of the expanded-bed adsorber
under these conditions. The extensive solids removal by the
filtering action of the packed bed, as indicated in Figure 31,
probably contributed to most of this difference.
Post Clarification: Laboratory Studies
In jar tests conducted during Phase One and reported in
Table 5, it was found that the coagulant dosages required
for effective coagulation of primary effluent and primary
— 52 —
-------
I I
100
VOLUME
FIGURE 29 TREATMENT OF
BEDS THAT
- PACKED BEDS
/
0
TREATED, 1000 gal.
PRIMARY EFFLUENT IN ACTIVATED
HAD TREATED 45O,OOO GALLONS
50 100
CARBON
OF CLARIFIED
PRIMARY
ELUENT—
7—6ft
12ff
24ft
POST-
CLARIFI ED
60
50
40
30
20
l0
-J
E
z
0
4
0
0
z
4
0
-J
4
I—
0
F-
U i
I
0
50
PRIMARY EFFLUENT - REMOVAL OF TOTAL ORGANIC
CARBON
-------
I I
-J
a’
E
z
0
0
0
U,
C..,
I ii
-J
-J
0
C l)
30-
20-
10
EXPANDED BEDS
I I I
0
PACKED BEDS
50 100 0
VOLUME
TREATED, 1000 gal.
FIGURE 30 TREATMENT OF PRIMARY
50
100
EFFLUENT IN ACTIVATED CARBON
BEDS THAT HAD TREATED 45O,OOO GALLONS OF CLARIFIED
PRIMARY
EFFLUENT
ft
PRIMARY
EFFLUENT
N
ft
24ft
PRIMARY EFFLUENT - REMOVAL
OF SOLUBLE ORGANIC CARBON
-------
x
5 810121511192224262913
SEPTEM BER
PRIMARY EFFLUENT
EXPANDED BEQ
12’ DEPTH
24’ DEPTH
PACKED BEDS
DEPTH
DEPTH
FIGURE 31
SUSPENDED SOLIDS IN TREATMENT OF PRIMARY
EFFLUENT IN ACTIVATED CARBON BEDS THAT HAD
TREATED 450,000 GALLONS OF CLARIFIED
PRIMARY EFFLUENT
x
120-
100-
x
80-
.
.
-J
E
U)
-J
0
U)
w
z
U i
U)
U)
60-
.
40-
20
— 55 —
-------
TABLE 5
Laboratory Comparison of Coagulant Requirements
For Primary Effluent and for the Effluent From
12-Ft Expanded Beds of Activated Carbon
(Jar Tests: 2 Mm Rapid Mix, 15 Mm Slow Stir, 30 Mm Settling)
Sample FeC1 3 Turbidity Sample FeCl Turbidity
Date and pH mg/i JTU Date and pH mg/i JTU
3/14 Primary 0 31 3/20 Primary 0 27
Effluent 77 9.3 Effluent 76 8.2
pH 7.7 96 3.3 pH 7.6 115 2.9
115 1.5 153 6.5
115 1.8
Adsorber 0 23
134 3.0
Product 76 6.2
153 4.0
pH 7.8 115 2.3
Adsorber 0 28 153 2.5
Product 57 5.9
3/21 Primary 0 16
pH 8.1 77 Effluent 38 5.8
77 6.0
96 3.3 pH 8.5 76 2.8
115 1.7
115 1.0
115 3.2 Adsorber 0 32
153 2.3 Product 38 8.8
pH 8.5 76 4.1
3/18 Primary 0 34 115 2.6
Effluent 95 5.5
115 3.3 4/3 Primary 0 32
134 2.1 Effluent 76 19
pH 8.6 115 2.3
Adsorber 0 33 - 153 2.7
Product 95 4.2
170 3.5
pH 7.4 115 1.8
134 1.9 Adsorber 0 30
Product 76 13
3/19 Primary 0 29 pH 8.6 115 4.3
Effluent 76 6.7 153 2.7
pH 7.5 115 2.1 170 3.5
153 2.2
4/10 Primary 0 42
Adsorber 0 28 Effluent 96 16
Product 76 6.2
pH 8.9 115 9.5
pH 7.7 115 4.5 134 6.9
153 2.5
Adsorber 0 38
Product 96 9.4
pH 8.6 115 8.4
134 3.2
— 56 —
-------
effluent which had been passed through the activated carbon
systems were essentially the same. Variations in dosages
required for samples of either stream taken at different times
were greater than variations in the dosages required for
samples of the two different streams taken at the same time.
For example, the amount of ferric chloride required to reduce
turbidity to 2 to 3 JTU in the jar tests during this period
varied from 75 to 150 mg/i as FeC1 3 for both the primary
effluent and for the carbon adsorber product, whereas the
dosages required for samples of the two different waters on
any given day were nearly equal.
Results of jar tests of post—clarification following
carbon treatment in 24-ft beds run during Phase Two are
presented in Table 6. During this entire test the columns
operated in an anaerobic condition, which clearly affected
the coagulation results. It can be observed from Table 6 that
there were times when coagulation of the adsorber product water
with the ferric chloride was poor, the addition of the coagulant
producing a black, finely divided solid. After aeration, either
by allowing the sample to stand in air or by bubbling air
into the sample, coagulation did take place.
Lime appeared to be a somewhat superior coagulant for
post—clarification of the water produced from carbon treatment
of primary effluent under anaerobic conditions. With lime
there was no color in or dark appearance of the supernatant
even when coagulating samples for which ferric chloride was
ineffective.
In Phase One, as noted above, ferric chloride was found
effective for coagulation of carbon-treated primary effluent.
There were, however, no anaerobic conditions apparent in the
columns during the Phase One tests. Lime was not evaluated at
that time although it may be presumed that it also would
have been effective.
Post-Clarification: Continuous Operation
The treated water from the expanded-bed and packed-bed
systems was fed during separate periods to the pilot plant
clarification system, which was therefore operating at one-half
of the rate used for the previous operations. The TOC levels
in the post—clarified carbon column effluents were shown in
Figure 29. Suspended solids data are not presented. Clarification
was inconsistent for the effluent from either column, and in
order to permit coagulation it was necessary to operate the
rapid mix propeller at a rate that would aerate the mixture.
The clarified product never appeared to be as clear as clarified
primary effluent, at times appearing yellow or dark. On occasions
a scum of iron oxide formed on the surface of the clarifier.
— 57 —
-------
TABLE 6
Laboratory Evaluation of the Post-Coagulation
of Effluent From 24-Ft Beds of Activated Carbon
(Jar Tests: 2 Mm Rapid Mix, 15 Mm Slow Stir, 30 Mm Settling)
Dose Appearance
Date Sample Coagulant mg/l Floc Supernatant Comments
9/11 ECU4 FeC 1 3 40-95 Poor Black
ECU4 FeCl3 50 Fair Haze Tested after standing
100 Fair Slight Haze for 2 hours in air.
Primary FeCl3 50 Good Slight Haze
(and Raw) 100 Good Clear
9/16 ECU4 FeC13 150 Good Clear Aeration Increases
180 Good Very Clear pH from 7.9 to 8.6
Primary FeCl 3 60 Fair Haze
(and Raw) 72 Good Clear
9/17 ECU4 FeC1 3 50 Fair Haze
75 Fair Slight Haze
100 Good Clear
120 Good Very Clear
Lime 75 Fair Haze
125 Good Slight Haze
PCU4 FeCl 3 50 Poor Black
75 Poor Black-Grey
Lime 75 Fair Haze pH 9.4
125 Good Slight Haze pH 9.8
9/19 ECU4 FeC1 3 72 Fair Haze
108 Good Slight Haze
144 Good Very Clear
Lime 100 Good Slight Haze pH 9.4
150 Good Very Clear pH 9.8
9/23 Primary FeC1 3 108 Good Slight Haze
144 Good Very Clear
Lime 125 Good Slight Haze pH 9.6
175 Good Very Clear pH 10.0
ECU4 FeC1 3 108 Fair Haze Dark
144 Good Haze Dark
Lime 125 Good Slight Haze pH 9.4
175 Good Clear pH 9.8
— 58 —
-------
ANALYSIS OF SPENT CARBONS
At the conclusion of the continuous treatment experiments,
the partially spent carbons were removed from each column,
weighed, and subjected to several analyses. The results of
these analyses are given in Table 7. In addition, the weight
of SOC removed by each column is reproduced in this table to
facilitate comparison with the apparent increase in weight
of the activated carbon.
The total weight of the dry spent carbon from each column
was calculated from the total weight of the drained carbon and
the moisture content as determined by drying samples of the
carbon at 140°C to constant weight. Likewise, amounts of
volatile matter (at 900°C) were determined on samples of the
carbons, and the total devolatilized weights calculated from
these values. In general, the carbon in the first bed in each
series contained more volatile matter, and showed a greater
increase in weight.
Iodine and methylene blue adsorption tests were run on
the fresh activated carbon and on samples of the dried and
devolatilized spent carbons. These measurements indicated
that a considerable portion of the original adsorptive capacity
of the carbon was recovered by treatment at 900°C for two
hours in the absence of air. All of the carbons showed an
increase in ash content, some of which may have been due to
iron from the coagulation operation.
CARBON PARTICLE SIZE EFFECTS
At the low flow rate of 1 gpm/ft 2 used for these particle
size studies, no expansion or particle motion was observed in
the bed of 8 x 16 adsorbent. Therefore, this column was
subject to the usual problems of a fixed bed. There were
instances in which solids accumulation resulted in the entire
bed being lifted as one mass of carbon by the upf lowing
wastewater. It was necessary to clean this bed frequently to
break up the plugs of carbon.
Data on the removal of TOC for the two columns are presented
in Figure 32. initially, removal of TOC for the two columns
was similar but the TOC removal by the bed of 8 x 16 particles
declined to about 2/3 of the TOC removal by the 50 x 100
particles by the end of the run.
POLISHING COLUMN
During the course of the Phase—Two operation, use of a
fresh activated carbon post—treatment was evaluated for
periods up to 48 hours. The results of these experiments,
presented in Table 8, show effective reduction of the TOC which
remained after the treatment provided by the 24-ft of activated
— 59 —
-------
TABLE 7
Analyses of Spent Carbons From The Pilot Operations
Volatile Ash,
SOC Matter, Dry Methylene Blue
Drained Dry Devolatilized Removed, Moisture, Dry Basis, Basis, Iodine Number, mg/g Adsorption mg/g
Column Wt., lbs Wt., lbs Wt., lbs lbs wt. % wt. % Wt. % Dried Devolatilized Dried Devolatilized
Phase I
Tests
PCU I 207.2 120 102.8 42 14.4 7.15 610 865 230 375
PCU2 182.1 108 92.9 30.7 40.6 14.0 7.3 625 880 230 360
ECU1 195.9 112 95.5 42.7 15.0 7.4 630 900 230 395
ECU2 188.1 108 92.4 31.3 42.8 14.2 7.5 610 880 230 360
PCC1 186.9 108 92.6 42.2 14.2 5.9 600 900 200 355
PCC2 183.7 102 90.0 25.6 447 11.5 5.9 700 910 220 380
ECC1 204.6 13.5 99.0 43.6 14.1 5.8 590 870 200 360
ECC2 180.0 104 91.5 24.0 42.4 11.8 5.8 700 940 240 345
Original 85 1.1 5.3 940 465
Phase II
Tests
PCC1 182.0 107 90.9 25.4 38.3 19.0 6.1 390 760
PCC2 174.9 106.4 91.0 12.0 41.3 14.4 7.5 510 790
PCC3 179.1 101.9 89.7 6.5 43.8 11.7 7.8 640 860
PCC4 186.1 93.8 86.7 2.9 50.2 7.6 7.4 800 940
ECC1 188.8 99.8 83.5 25.1 47.1 16.3 7.0 415 775
ECC2 185.4 105.0 90.4 15.3 43.3 13.9 7.3 500 790
ECC3 185.4 100.9 89.5 5.0 45.6 11.2 7.4 620 870
ECC4 183.2 89.2 83.6 2.4 51.3 7.3 7.0 730 910
Original 85 0.5 1.0 5.7 1120 520
All columns originally charged with 85 lbs of activated carbon.
-------
PRODUCT FROM
50 X 100
VOLUME
TREATED, 1000 gal.
FIGURE 32
TREATMENT
UPWARD THROUGH 4-FT DEEP BEDS OF ACTIVATED CARBON
-‘20
E
CLARIFIED PRIMARY
EFFLUENT
0
‘-a
z
0
4
C)
C)
z
4
0
4
I—
0
I-
‘5
I0
5
PRODUCT FROM 8X 16
-
/
0
0
I 2 3 4 5 6
OF CLARIFIED PRIMARY EFFLUENT FLOWING
WITH DIFFERENT PARTICLE SIZES AT 1 gpm/ft 2
-------
TABLE 8
Evaluation of Fresh—Carbon Polishing of
Product Water From 24-Ft Expanded-Bed Adsorbers
Column: 1-in. 2 l’ Deep
92 Grams Activated Carbon
Carbon Treated Clarified Primary Effluent Feed ‘ 5 gpm/ft
Time TOC
Date Hours In Out
7/22
7/25
7/29
8/27
1
2
8
2
5.5
1.5
4
4.8
2.0
6
3.8
1.5
8
4.5
3.5
23
5.0
4.0
3.5
2.8
2.0
<0.5
0.5
<0.5
1
1.5
<0.5
2—1/2
1.0
<0.5
5
1.0
<0.5
7
1.5
<0.5
1
4.0
<1.0
2
3.5
1.5
3
4.0
<1.0
7
3.2
1.5
12
3.5
1.5
24
5.0
2.5
48
5.0
4.0
— 62 —
-------
carbon to a level of 1 to 2.5 mg/i for a period of 24 hours.
The 24—ft carbon beds were anaerobic during this operation,
and at times produced unstable product water which became
cloudy. However, the product from the short polishing bed
was stable even after the carbon bed had been on stream for
48 hours.
— 63 —
-------
DISCUSSION AND CONCLUSIONS
CLARIFICATION
Chemical clarification with ferric chloride as used in
this study proved very effective for consistent removal of
suspended and colloidal matter from a primary effluent.
During the extended Phase-Two test program (125 days) the
clarification system consistently produced a reasonably clear
effluent and, at the same time, removed TOC. It was expected
that removal of suspended solids would result in TOC reduction,
but the removal of TOC was somewhat higher than anticipated.
The extent of removal of TOC by coagulation may be related to
the nature of the solid matter in the particular wastewater
studied.
A somewhat surprising result was the accompanying average
removal of 32-33% of the SOC by the chemical clarification
step. This SOC removal may be a matter of definition, in that
SOC was defined as the organic carbon measured in the filtrate
passing a 0.45-micron membrane filter. Thus, the removal of
SOC by clarification indicated that a significant portion of
this organic matter may have been present in the form of
colloidal or particulate matter smaller than 0.45 microns or
was reduced by biological activity. Removal of this fraction
may have been especially beneficial to the subsequent treatment
by activated carbon, since colloidal particulate matter is not
adsorbed nor otherwise readily removed by activated carbon in
either an expanded or packed bed.
The TOC and SOC removal was accompanied by a significant
reduction in the BOD. The average BOD of the primary effluent
averaged about 50 mg/i since the ELSA plant primary settler
removed about 50 percent of the raw sewage BOD. The chemical
coagulation and clarification removed another 70 percent resulting
in a clarified primary effluent with an average BOD less than
that expected in secondary effluent from the trickling filter
treatment.
As expected, the coagulation system accomplished very
effective removal of phosphorous in addition to providing a
high degree of clarification and removal of organic matter.
It is presumed that the phosphate was removed by precipitation
of a ferric phosphate. The phosphate therefore consumed a
part of the ferric chloride. Thus, ferric chloride dosage
requirements can be expected to vary with variations in the
phosphate content of the sewage.
Since phosphate removal from waste effluents is a matter
of considerable interest, this factor adds another advantage
to the chemical pretreatment, which, while accomplishing
clarification for more effective removal of organic matter by
activated carbon, also provides a high percentage of phosphate
reduction.
— 64 —
-------
ACTIVATED CARBON TREATMENT
Removal of Organic Matter
The principal study directed toward meeting the objectives
of this project was the extended test run of Phase Two which
was a direct comparison of the effectiveness of the expanded-
bed and packed-bed modes of carbon adsorption for treating
primary effluent. Sufficient stages of carbon beds were used
to accomplish essentially complete organic removal. In
preparation for the extended run, the effects of pretreatment
of the primary effluent were evaluated in Phase One using shorter
activated carbon beds which were not expected to accomplish
as much organic removal. On the basis of analyzing the possible
pretreatments and running initial laboratory studies, the Phase
One pilot plant runs were limited to a comparison of the effects
of a high degree of clarification using chemical coagulation
against no pretreatment.
The results indicated that the use of chemical coagulation
and clarification was quite desirable. When the suspended
solids in the primary effluent were removed, nearly all of the
organic matter remaining was that defined as SOC , which could be
effectively treated by the granular activated carbon. The result
was a relatively low organic content in the effluent from the
two column (12-ft carbon bed) systems. Therefore, in the
extended pilot plant test run of Phase Two using two four-column
(24-ft carbon bed) systems, the primary effluent was highly
clarified prior to feeding to the adsorbers.
In the Phase One studies the expanded-bed mode of contacting
w s somewhat inferior to the packed—bed mode, particularly, as
would be expected, for removal of TOC from untreated primary
effluent. With the clarified effluent feed, the difference
between the two modes of carbon contacting was less but still
existed, probably due to suspended solids that remained in the
primary effluent when the coagulation—clarification system
operated poorly as noted in the earlier discussiOn. In the
Phase Two runs, the performance of the expanded-bed and packed-
bed adsorbers was essentially equal for organic carbon removal.
In Phase Two, the clarification w s more consistently effective
and TOC essentially was all SOC. Thus, the effect of organic
matter present in the effluent as solids was eliminated. During
this test, after the clarification had removed substantial
quantities of TOC and SOC from the primary effluent, both
adsorption systems reduced the organic content to an average of
about 4 mg/i, which appeared to be a lower limit. This was an
average removal of 75% of the TOC and 70% of the SOC. Both
adsorber systems continued producing this low organic content
effluent essentially throughout the 125 days of test and it
appeared that they could have continued this effectiveness for
an additional time. However, the test was stopped to allow time
to investigate post—clarification during the Phase Three period.
— 65 —
-------
The conclusion that the carbon adsorption effectiveness
could have continued beyond the 125 day test period is supported
not only by noting that the organic content of the final
effluent was not yet increasing, but also by considering the
behavior of the first column in each series compared to that
of the other three columns. Both of the first columns removed
on the average more than half (55%) of the organic matter
removed by the total system. This resulted in the first beds
being loaded to about 30% by weight with TOC while the average
TOC loading for the entire system of four beds was 13.5%.
Despite the 30% loading, the first beds were still removing
about 25% of the TOC from the clarified feed at the termination
of the tests. If TOC represents 50% by weight of the organic
matter, then the loading on the carbon in the first bed was
actually 60% by weight which is substantially greater than the
usual adsorption capacity of carbon. This high loading may
have been due to solids collection on the carbon. However, it
is also believed that biological activity in the bed actually
accounted for some of the apparent organic removal ascribed
to the carbon. This affect of biological activity is discussed
further later in this Section.
BOD removal in the clarification-adsorption system roughly
paralleled the TOC removal. The activated carbon adsorption
systems further reduced the BOD to 4-5 mg/l, as may be noted
in Figure 24. These results indicate that the BOD was largely
present in the particulate and colloidal matter and not as
soluble organic (SOC). The BOD of the product, representing
only 4—5% of the raw sewage BOD — or 95 to 96% removal — was
considered to be acceptable and no efforts were made to find ways
to achieve further reduction.
Thus, the treatment of sewage by clarification and adsorption
achieved high removal of both organic carbon and BOD, as contrasted
with primary-secondary treatment plants which are reported to
be capable of removing 90% of the BOD but less (80%) of the
total organics.k Adsorption can remove the organics that are
considered refractory to biological processing, and at the same
time remove most of the biodegradable organics in the waste-
water.
Based on the organic removal results during the test period
and ignoring any additional removal capability of the system,
it is possible to estimate a carbon dosage for this method of
treatment using either the expanded-bid or packed-bed mode of
adsorption. It is assumed that an adsorption system would
consist of at least two stages and that operation would be
countercurrent. When the carbon in the first phase became
saturated, it would be removed from the system for regeneration
and the second stage would then become the first stage.
— 66 —
-------
In this test program, if it is considered that the first
beds of 85 pounds of carbon each were spent after treating
450,000 gallons each (4500 bed volumes for each four-column
system) this would represent a dosage of 189 pounds of carbon
per million gallons of effluent treated. To estimate system
dosages from this loading for the first beds, several factors
must be taken into account: (1) the carbon might have been
exhausted earlier if the sewage had maintained its strength
rather than being reduced in organic content by heavy rain
during part of the 125 day run; (2) the second beds in the
system had removed half as much TOC as the first bed, Figure 25
and Table 4, and if these beds were moved to the first position
with half of their capacities exhausted, they would only be
able to treat half the effluent, thus doubling the carbon
requirements; (3) the carbon after regeneration would be expected
to have a lower capacity than fresh carbon. On these bases,
a conservative estimate of the activated carbon requirements
for such continuous processing would be about 500 pounds per
million gallons of water.
The Phase Three experiments which demonstrated that
post—clarification was not a desirable approach for either
expanded—bed or packed-bed modes of adsorption, also
demonstrated the adverse effects that large amounts of suspended
solids may have on carbon adsorption systems after the carbon
has been in use for some time. In this test period, the
filtering action of the packed beds was very apparent in that
they definitely showed greater capacity than the expanded beds
to remove TOC and BOD, and to some extent, SOC. Using the
carbon that had previously been used in the Phase Two
experiments, the TOC removal data initially indicated that the
expanded beds were slightly inferior to the packed beds. As
the run progressed, however, the TOC removal effectiveness of
the expanded beds drastically declined. During this time,
the first packed bed (6-ft) removed as much TOC as the entire
24-ft expanded-bed system and the first two packed beds (l2-ft)
removed as much SOC as the 24-ft expanded bed.
However, during the course of this operation, it was
necessary, because of excessive pressure drop, to clean the
packed bed daily to remove the accumulated suspended solids.
This frequent cleaning of the first packed bed was apparently
beneficial to the performance of the subsequent packed beds in
the series. Because the pressure drop remained low, the expanded
beds were not cleaned during this period and the suspended
solids content of the product water frequently exceeded that of
the feed.
This test series definitely confirmed the conclusion
from the Phase One studies that removal of suspended solids
before adsorption is beneficial to the efficient operation of
— 67 —
-------
activated carbon systems. In the case of the packed bed, a
thoroughly clarified feed permitted operation at lower pressure
and less frequent backwashing. For expanded-bed systems, clarified
feed is also desirable for efficient operation by preventing
excess solids accumulation with a resulting decrease in the
adsorptive efficiency of the activated carbon.
Biological Activity in the Carbon Beds
The high organic loadings observed in the first beds in
the activated carbon adsorber systems and the continued organic
removal under these conditions indicated that another mechanism
was involved in addition to that of adsorption. Activated
carbon does provide an excellent surface for concentration of
organic biological substrate materials which appeared to provide
a beneficial biological growth. This biological activity did
not appear to hinder the adsorption process in any observable
fashion.
During the course of the Phase Two extended run, the
biological activity on the activated carbon beds generally
produced anaerobic conditions as evidenced by evolution of H 2 S.
Without aeration or other approaches to control the anaerobic
environment, this biological activity had the disadvantage of
creating an immediate dissolved oxygen demand (IDOD). This
IDOD was probably due to the H 2 S and the presence of iron carried
over as Fe(III) from the coagulation stage and then reduced to
Fe(II) as a result of the anaerobic conditions in the carbon
adsorption systems.
Control of the anaerobic conditions and the resulting
objectionable H 2 S formation was not completely solved by the
application of either hypochiorite, air or oxygen prior to the
first carbon beds only. Even with a concentration of 10 mg/i
DO in the feed to the first bed there was no DO remaining in
the feed to the next three beds in series. This apparently
controlled the conditions in the first bed, but not in the
later beds.
It appears that control of the anaerobic conditions will
require having a positive DO in the feed to each column. With
expanded—bed adsorbers, this can be accomplished by using open
vessels and overflow troughs designed to provide aeration.
Operational Advantage of Expanded-Bed Contactors
Although there were no significant differences in the
organic removal characteristics of the expanded—bed and packed-
bed adsorbers with highly clarified feed, there were, as
observed previously in pilot studies on the treatment of
secondary sewage effluents 1 , some significant operational
differences. None of the four expanded—bed adsorbers required
any cleaning or maintenance over the four—month period of
operation with the clarified primary effluent feed; flow rates
— 68 —
-------
remained constant without appreciable increases in head loss
through the four columns in series. Conversely, even with the
highly clarified primary effluent, the head loss in the packed-
bed adsorbers increased steadily, requiring increased pumping
pressures and frequent cleaning and backwashing.
POLISHING TREATMENT CONCEPT
At the start of the Phase Two experiment the first 6 ft
of activated carbon initially removed essentially all of the
TOC. After about 72 hours, the effectiveness of the first bed
had changed somewhat to give an effluent containing about 6 mg/l.
This level then very slowly increased to about 12 mg/i during
the next 1800 hours while the three additional columns in the
system reduced the TOC to a level of 2 to 6 rng/l. These results
indicated that the carbon may have rapidly become saturated with
some fraction of the organic matter and that this effect prevented
complete organic removal from then on. Therefore, it was proposed
that the use of a final polishing bed of fresh activated carbon
might permit a closer approach to achieving continuously organic-
free, renovated water. In the preliminary experiments reported,
small columns containing a 1-ft bed depth of fresh activated
carbon were found to be effective for removal of most of the
remaining TOC over a period of about 24 hours. If further tests
confirm that this polishing action can be depended upon with
regenerated carbon, development of a very low cost regeneration
or a means of achieving high loading to make the concept
economically feasible would appear justified.
The organics adsorbed in the short bed following carbon
treatment are evidently only weakly adsorbed and at a slow
rate, and thus could possibly be removed readily in a rapid,
simple reactivation system. This approach should be investigated
further in an effort to devise a treatment scheme for complete
organic removal in applications where the specifications for
reused water are especially demanding. The main carbon beds
would be used to adsorb the major part of the contaminants with
the final polishing bed used to complete the organic removal
and to provide the higher quality product water.
CARBON PARTICLE SIZE EFFECTS
It was proposed by Morris and Weber 13 that the expanded-bed
mode of operation wou1d allow for the use of smaller particles
of granular activated carbon, with a greater capacity and a
higher rate of adsorption of organics from solution. The present
experiment on the effects of particle size was conducted to
determine on a reasonable scale if it would be practical to
use smaller particle carbon in an expanded—bed adsorber and if
indeed advantages would accrue. Two widely different-sized
fractions of an experimental carbon were used. Based on diameters,
the 8 x 16 fraction had an average particle size seven times
larger than the average particle size of the 50 x 100 fraction
— 69 —
-------
(1.4 vs. 0.2 mm). Thus, the bed of smaller particles presented
about seven times as much external surface to the water. Although
the two size fractions were taken from the same batch of activated
carbon, it was found that there was a significant difference in
the adsorptive capacity of the two as measured by both iodine and
methylene blue adsorption.
The system containing the smaller particles was more
effective for removal of TOC from clarified primary effluent
for the entire test, with a greater difference developing as
the run progressed, suggesting a larger adsorptive capacity as
well as a higher rate of adsorption for these smaller particles.
While there may have been other factors operating, the
50 x 100 particles provided greater organic removal. However,
while there were no operating problems with the smaller particles
at this flow rate and complete expansion of the bed was attained,
a practical application would probably require higher specific
flow rates to minimize the size of the adsorption vessels. Some
particle size range intermediate between the two chosen for these
tests may be optimum to provide a balance between capacity of the
adsorbent and vessel size.
PROPOSED TREATMENT SCHEME
On the basis of the results obtained in the extended test
in Phase Two of this study, a complete physicochemical wastewater
treatment system is proposed. The concept proposed consists of
clarification of raw sewage followed by adsorption of the soluble
organics remaining in the clarified effluent on activated carbon.
Pilot—scale experiments were conducted with primary effluent to
avoid problems due to large or fibrous solids, and by using
ferric chloride for convenient metering of the coagulant solution.
The proposed scheme is based on coagulating Thw sewage with lime.
To obtain data for design and analysis purposes, laboratory
jar tests were conducted on raw sewage, primary effluent and
secondary effluent taken at the same time. Coagulation with
ferric chloride at 120 mg/l or lime at 250 mg/l produced very
clear supernatant in all cases. Results of clarification and
subsequent contacting of the supernatant are given in Table 9.
While the initial TOC values are low, there appears to be little
difference in the effectiveness of the coagulants for removing
organic carbon, or on the subsequent adsorption. Gross solids
present in the raw sewage settled rapidly under the conditions
in the jar test, and did not appear to affect the coagulation.
Similar experiments were conducted to determine the effect
of pH on organic removal by adsorption on granular activated
carbon after clarification with lime or ferric chloride. Results
of these tests in Table 10 show insignificant effect due to
coagulation or adsorption at the higher pH.
The use of lime as a coagulant provides good clarification,
a rapid settling sludge, and in addition, permits the use of a
— 70 —
-------
TABLE 9
Laboratory Studies of Coagulation With Ferric Chloride
and Lime and The Effects on Subsequent Carbon Adsorption
TOC After Carbon
coagulation Contacting 2 of
The Supernatant
TOC TOC 700 2000
Date Waste Stream mg/i Reac ent’ pH mg/i mg/i mg/i
8/19 Raw Sewaae’ 7.7 30.5 Fed 3 6.3 13.0 7.5 6.5
Lime 10.8 16.0 10.0 5.0
Primary 7.3 29 FeC 1 3 6.0 13.0 5.0 5.5
Effluent Lime 10.7 12.5 4.5 4.5
Secondary 7.3 18 FeC 1 , 6.0 5 3.5 2.5
Effluent Lime 10.7 6 2.5 4.0
8/26 Raw Sewage 7.9 50 FeC1 , 6.3 19.0 5.0 3.0
Lime 10.6 21.5 7.5 6.0
Primary 8.5 31 FeC1 , 6.5 12.0 5.0 3.0
Effluent Lime 10.8 11.0 6.0 4.0
Secondary 7.0 9.5 FeC1 3 6.0 6.5 <1 <1
Effluent Lime 10.6 6.5 1.5 <1
9/9 Raw Sewage 7.6 35 FeC1 , 6.2 11.5 4.0 4.0
Lime 10.8 14.0 8.0 4.0
Primary 6.9 19.0 FeC 1 , 5.9 5.5 3.0 3.5
Effluent Lime 10.7 7.5 1.8 2.5
Secondary 6.8 12.5 FeC1 3 5.8 4.8 2.5 1.0
Effluent Lime 10.8 7.0 2.5 2.0
‘Coagulant dosage: FeC13 120 mg/l, lime 250 mg/i
2 pulverized Activated Carbon @ 700 and 2000 mg/i for 1 hour then filtered
3 From ELSA Plant
— 71 —
-------
TABLE 10
Evaluation of The Relative Effects of pH and
Pre—Coagulation on The Effectiveness of
Treatment of Primary Effluent by Activated Carbon
*pH adjusted with BC].
No change in pH observed during adsorption
4
FeC13
140
6.7
5
Ca (OH) 2
420
11.5
7.9
6
Ca (OH) 2
390
11.5
PRIMARY EFFLUENT (TOC 38 mg/i, 1000
ml samples)
Jar Test No. 1
2
3
Reagent Added NaOH
Amount mg/i -
NaOH
-
None
pH Obtained 11.5
11.5
Appearance after
1 hr settling
Decanted After 1:45 hrs
pH for Adsorption
TOC of Supernatant mg/i
Treated Supernatant with
2000 mg/i Granular Carbon
TOC After 1/2 hours. 20
1—1/2 hours 22
3 hours 21
22
18
17
Cloudy Cloudy Cloudy Clear Clear Clear
7.0*
37
11. 5
35
7.0*
33
6.7
16
7.0*
18
11.5
15
20
10
10
11
23
10
10
10
— 72 —
-------
simple method for recovery that also insures destruction of the
usual sewage solids. The sludge from lime coagulation of raw
sewage will be thickened and incinerated to destroy the raw
sewage organic solids and provide regenerated coagulant. Lime
is the preferred coagulant for raw sewage.
A recommended flow sheet for the wastewater treatment
concept is given in Figure 33 and a possible plant lay-out in
Figure 34. In this scheme, coagulant is added to the raw sewage,
and flocculation takes place in a chamber which provides moderate
agitation for an average detention time of 15 minutes. Clarification
takes place in a sedimentation basin with an average detention
time of two hours.
The clarified effluent is then passed through activated
carbon adsorption units for removal of dissolved organics.
The preferred mode of operation is an expanded bed, which permits
the use of simple open—top concrete contacting basins and
relatively trouble—free operation. The use of open tanks with
trough-type overflows at the surface of the contacting basin
provides a means of additional aeration of the wastewater during
treatment, thus controlling anaerobic conditions such as those
observed in the closed systems in the pilot plant. Two—stage
contacting of the activated carbon is proposed to provide efficient
utilization of adsorptive capacity by countercurrent movement
of the activated carbon within the system. The plant lay-out is
based on installing five adsorption units of two stages each.
When the granular carbon in the first stage of one unit is spent,
that unit is to be taken off stream while the spent carbon is
removed and regenerated in the multiple-hearth furnace system
provided. During the. time this unit is off-stream for the
carbon regeneration, the other four units will run at 25% higher
feed rate each. Upon completion of the carbon regeneration, the
regenerated carbon will be returned to the adsorber which will
become the second stage of that unit; the former second stage
with partially spent carbon becoming the first stage. Feed would
then be evenly divided to the five units until another first
stage carbon bed is spent.
The water resulting from the clarification and activated
carbon treatment will enhance the quality of surface waters and
with chlorination would be suitable for many uses. A final
filtration treatment may be desirable to insure a crystal clear
effluent for some reuse applications. This post-filtration
would remove any suspended matter generated in the carbon
columns which might otherwise be released into the final treated
water.
The sizes of the various processing units included in the
overall treatment scheme are based on the results of this work
with one sewage source. It is anticipated that results similar
to those reported here can be achieved in larger plants operating
on sewage with different characteristics. However, there could
— 73 —
-------
RAW SEWAGE
AIR
CLARIFIER 2-STAGE
CARBON CONTACTORS
EXPANDED BEDS
CARBON REGENERATiON
I 4
L
c j fT11 r 1
-j
DRAIN
TANK
MULTI-
HEARTH
FURNACE
STORAGE
TANK
FIGURE 33
PROPOSED SCHEME OF TREATMENT OF RAW SEWAGE BY
CHEMICAL CLARIFICATION AND ADSORPTION ON ACTIVATED CARBON
AERATED GRIT CHAMBER
AND FLOCCULATOR
I———
I
TREATED
WATER
— — — — —
UME
MULTI-
HEARTH
FURNACE
SLUDGE
THICKENER
DRUM
FILTER
SLURRY
TANK
-------
RAW SEWAGE
FLOCCULATION CHAMBER
SETTLING CHAMBERS
Ii II II II II i
I
cM
00000
II II II II II ii
I EXPANDED
>3ERS
CARBON
REGENERATION
BED AD-
. . ADSORBER FEED
\TANKS PUMPS
00000
ALTERNATE
PACKED BED ABSORBERS
TREATED WATER
FIGURE 34
PROPOSED ARRANGE.MENT OF PROCESSING UNITS
FILTER
SLUDGE
THICKENER
GRIT REMOVAL
AIR
Q
Q
— 75 —
-------
be considerable differences in the response of other wastewaters
to the various processes included in this overall treatment
scheme. Tests of the process with additional sewages are required
to show the universality of the concept and effectiveness.
ESTIMATION OF TREATMENT COST
Capital and operating costs were estimated for a 10 mgd
plant to produce a clear, low TOC and BOD water by clarification
and carbon adsorption using the proposed process scheme. The
capital cost estimates, Tablesil a d 12, inclu 1aUofthe.necessaxy
equipment, foundations, buildings, piping and carbon and provisions
for utilities. Allowances for engineering and construction costs
and profits are included.
To arrive at the capital cost estimate, a fairly detailed
breakdown of the individual units in the process was prepared.
Then costs for these items were found in several referencesik,J 5 ,l$, 7
and a most reasonable price was selected. The individual cost
estimates and the individual unit sizes are presented in Table 13.
Usually these prices tended to be on the high side in order to
place the estimate on a conservative basis. The prices so
obtained were adjusted to be representative of price levels
expected by the end of 1969 by using an estimated ENR index of
1300.18 In a preliminary estimate such as this one, the prices
estimated for the individual units are not as accurate as those
that would be obtained for an engineering study with actual
quotations. It is believed, however, that the total capital cost
estimate is reliable.
The operating costs were developed from a procedure
recommended by the FWPCA 19 for maintenance and overhead charges
and include reasonable labor requirements, utilities and supplies
including replacement carbon and lime. This estimate is presented
in Table 14. All of the factors used in arriving at this estimate
are also shown on the table.
These estimates indicate that the total costs for producing
a clear, low-carbon, low-phosphate effluent are l9.7 /l000 gal.
for an expanded-bed system and 20.5’ /1000 gal. for the packed-
bed system including amortization of the capital over 24 years
at 6% interest. 20 These treatment costs appear to be reasonable
for the quality of water such a plant would produce from sewage.
— 76 —
-------
TABLE 11
Estimated Capital Costs for Treatment
of Raw Sewage by Clarification
Basis: 10 mgd
Equipment Piping Total
Pretreatment and Flocculation $ 99,800 $ 3,200 $ 103,000
Clarification and Sludge Handling 731,000 13,500 744,500
Plant Cost 830,800 16,700 847,500
Instrumentation 5% equip. 41,500
Painting and Insulation 2% equip. 16,600
Buildings and Structures 25% Plant 212,000 270,100
Physical Costs $1,117,600
Engineering
Home office 18% 201,000
Field 17% 190,000
Contractor 5% 55.000 446,000
Base Costs $1,563,600
Contingency 15% of Base Costs 235,000
Auxiliary Facilities $1,798,600
Power @ lOO/KW 26,000
Roads, walks, fence 110,000 136,000
Total Plant Cost $1,934,600
-------
TABLE 12
Estimated Capital Costs for Treatment of
Clarified Raw Sewage by Adsorption in Activated Carbon Beds
Basis: 10 mgd
Expanded
Packed
Equipment
Adsorption System
Regeneration System
Piping
Adsorption System
Regeneration System
Total
$ 172,900
129 ,900
$ 302,800
$ 195,000
13,500
$ 208,500
$ 511,300
$ 179,200
129 ,900
$ 309,100
$ 233,000
13,500
$ 246,500
$ 555,600
Instrumentation
Painting and Insulation
Buildings and Structures
$ 25,000
10,000
127,000
$ 162,000
$ 28,000
16,500
140 ,000
$ 184,500
Engineering
Home Office
Field
Contractors
Base Cost
Contingency 15% of Base Cost
Auxiliary Facilities
Power
Fuel Oil
Roads, Walks and Fence
$ 6737300
$ 120,000
115,000
34,000
$ 269,000
$ 942,300
142,000
$1,084,300
$ 25,000
20,000
60,000
$ 105,000
$ 740,100
$ 134,000
126 ,000
37,000
$ 297,000
$1,037,100
155,000
$1,192,100
$ 54,000
20,000
60,000
$• 134,000
Total Plant Cost
$1,189 ,300
Sl,326,100
Activated Carbon
288,000
$1,477,300
288,000
$1,614,100
Physical Costs
— 78 —
-------
TABLE 13
Estimated Equipment Costs for Direct Treatment of
Raw Sewage by Chemical Clarification and Adsorption
EQUIPMENT COSTS INSTALLED
Pretreatment and Flocculation
NO .
1 Bar Screen and Rake
1 Bypass Gate
3 Raw Sewage Pumps - 5 mgd each
1 Grit Removal & Flocculation Chamber
16’ x 15’ x 60’ (15 mm.)
1 Grit Removal Equipment
2 Aeration Diffusers - 30’ long each
1 Blower - 1,000 cfm 7 psi
TOTAL
Air Piping
Clarification and Sludge Handling
TOTAL COST
$ 14,000
5,000
20 ,000
28 ,800
22,000
3,500
6,500
$ 99,800
$ 3,200
1 Clarification Chamber
100’ x 100’ x 11’ deep (2 hrs.)
4 Tank Equipment - 25’ x 100’
2 Sludge Pumps - 200 gpm each
1 Sludge Thickener Tank - 25’ dia. x 10’
1 Thickener Mechanism 2
1 Sludge Filter - Vacuum Drum 800 ft
1 Sludge Cake Conveyor
1 Multiple Hearth Furnace - 13’ dia. x 8
hearth
1 Slurry Tank - 16’ dia. x 20’ deep
2 Slurry Feed Pump - 100 gpm each
1 Storage Tank - 2 @ 17’ x 17’ x 20’ deep
TOTAL
Piping - Sludge Lines
$ 85,000
155,000
9,600
23,000
7,000
115 1000
2 , 500
300,000
4,800
9,600
19,500
$731,000
$ 13,500
— 79 —
-------
TABLE 13 (Cont’d )
Activated Carbon Adsorption Systems
Packed Bed
NO .
10 Contactors — 19’ dia. x 20’ high
Internal Coating
Underdrain Blocks
Surface Wash System
Pumps, Main - 5 mgd, 60’ head each
Surface Wash
Backwash - 4,000 gpm
TOTAL
Piping Adsorber
Backwash
$102,000
15,000
14,000
5,000
36,000
1,200
7,200
$179 ,200
128,000
69,000
36,000
33,000
TOTAL - Packed Bed
Expanded Bed :
$412,200
10 Contactors - 17’ x 17’ x 20’ deep
Internal Distributor
Overflow Troughs
6 Pumps - 5 nigd, 30’ each
TOTAL
Piping - Adsorber
Backwash
Transfer Lines
TOTAL - Expanded Bed
$ 94,400
36,000
10,000
32,500
$172 ,900
128,000
31,000
36,000
$195,000
$367,900
CARBON REGENERATION
1 Transfer Pump - 400 gpm 100 psi
1 Transfer Pump - 50 gpm 50 psi
1 Drain Bin - 16’ dia. x 12’
1 Storage Bin - 16’ dia. x 12’
1 Dewatering Feed Screw
1 Multiple Hearth Furnace — 54” 10 x 8
hearth
TOTAL
Piping
TOTAL - Carbon Regeneration
$ 2,200
1,200
9,800
9,800
1,900
105,000
$129 ,900
$ 13,500
$143,400
6
1
1
TOTAL COST
Transfer Lines
— 80 —
-------
TABLE 14
Estimated Annual Operating Costs for Treatment
of Municipal Wastewater by Clarification and
Adsorption in Expanded-Bed and Packed-Bed Adsorbers
Basis: 10 mgd
Pretreat
Clarify, etc .
Annual Cost
Carbon Treatment
Combined
1.
Operating Labor*
Expanded
$ 43,300
$ 43,300
$ 86,600
$ 86,600
$ 43,300
2.
Maintenance Labor - 3% Plant
Phys. Costs
33,500
20,100
22,100
53,600
55,600
3.
Maintenance Materials — 2%
22,300
13,500
14,800
35,800
37,100
Plant Phys. Costs
4.
Maintenance Supplies — 15% of
2 + 3
8,400
5,000
5,500
13,400
13,900
5.
Supervision — 15% of 1
6,500
6,500
6,500
13,000
13,000
6.
Payroll Overhead — 15% of 1 +
2
11,500
9,500
9,800
21,000
21,300
7.
General Overhead — 30% of 1 +
2 + 6
26,500
21,900
22,600
48,400
49,100
8.
Insurance — 1% of Plant Phys.
Costs
11,200
6,700
7,400
17,900
18,600
9.
Carbon Makeup — 5% @ $.28/lb
27,500
27,500
27,500
27,500
10.
Lime Makeup — 25% @ $20/T
23,000
23,000
23,000
11.
Fuel — Q $0.50/MM Btu
62,000
13,000
13,000
75,000
75,000
12.
Power — $0.01/kwh
19,500
13,500
25,500
33,000
45,000
13.
Amortization — 24 years @ 6%
154,100
117,500
127,700
271,600
281,800
Total Annual Cost
$421,800
$298,000
$325,700
$719,800
$747,500
Treatment Cost — /1000 gal.
11.56
8.164
8.924
19.72
20.48
* 2 shift men + 2 day men @ $4.00 per hour
-------
REFERENCES
1. Hopkins, C. B., Weber, W. J., Jr., and Bloom, R., Jr.,
“A Comparison of Expanded-Bed and Packed-Bed Adsorption
Systems”. Report No. TWRC-2, 1968, 74 pp, R. A. Taft
Water Research Center, U. S. Department of the Interior,
Cincinnati, Ohio.
2. Weber, W. J., Jr., and Kim, J. G., “Preliminary Evaluation
of the Treatment of Raw Sewage by Coagulation and Adsorption”,
Technical Memorandum, TM-2-65, Sanitary and Water Resources
Engineering Division, The University of Michigan, Ann Arbor,
Michigan, May 1965.
3. “Nutrient Removal and Advanced Waste Treatment”, Proceedings
of the Technical Symposium, FWPCA, U. S. Department of the
Interior, Cincinnati, Ohio, April 29-30, 1969.
4. Stephan, D. G., and Weinberger, L. W., “Water Reuse-Has it
Arrived”, Journal Water Pollution Control Federation , 40,
4, 529 (April 1968). —
5. “Advanced Waste Treatment Research”. Federal Water Pollution
Control Administration Summary Report, Advanced Waste
Treatment, July 1964-July 1967, FWPCA Publication No. WP-20-
AWTR—19, 1968, 96 pp, R. A. Taft Water Research Center,
U. S. Department of the Interior, Cincinnati, Ohio.
6. Joyce, R. S., Allen, J. B., and Sukenik, V. A., “Treatment
of Municipal Wastewater by Packed Activated Carbon Beds”,
Journal Water Pollution Control Federation , 38, 5, 813
(May 1966) . —
7. Parkhurst, J. D., Dryden, F. D., McDermott, G. N., and
English, J., “Pomona Activated Carbon Pilot Plant”, Journal
Water Pollution Control Federation , 39, 10, part 2, R70
(Oct. 1967). —
8. Winneberger, J. H., Austin, J. H., and Klett, C. A., “Membrane
Filter Weight Determination”, Journal Water Pollution Control
Federation , 35, 6, 807 (July 1963).
9. “Standard Methods for the Examination of Water and Wastewater”.
12th Ed., Amer. Pubi. Health Assn., New York (1965).
10. “Manual on Industrial Water and Industrial Wastewater”,
American Society for Testing Materials, Philadelphia, Pa., 1963.
11. Trainirjg Course Manual - “Technical Seminar on Advanced
Waste Treatment”, FWPCA Publication, U. S. Department of
the Interior, 1967.
12. Hudson, H. E., Jr., “Physical Aspects of Flocculation”,
Journal American Water Works Association , 57, 885 (1965)
— 82 —
-------
13. Morris, J. C., and Weber, W. J., Jr., “Adsorption of
Biochemically Resistant Materials From Solution 1.”, FWPCA
Publication No. AWTR-9, 1964, R.A. Taft Water Research Center
U. S. Department of the Interior, Cincinnati, Ohio.
14. Bauman, H. Carl, “Fundamentals of Cost Engineering in the
Chemical Industry”, Reinhold, New York 1964.
15. Chilton, Cecil H., “Cost Engineering in the Process Industries”,
McGraw-Hill, New York, 1960.
16. Anon. Water and Wastes Engineering, 6 #9, 68—82 (1969)
17. Smith, Robert, “Costs of Conventional and Advanced Treatment
of Wastewater”, Jourhal Water Pollution Control Federation ,
40, 1546 (1968)
18. Extrapolated From Indices Published in Engineering News
Record Magazine.
19. Final Report for FWPCA, Contract No. 14-12-105, Swindell-
Dressier.
20. “This Week in Tax Exempts”, Investment Dealers Digest ,
October 28, 1969.
— 83 — 15. S. GOVERNMENT PRINTING OFFICE 1970 0- 408-392
------- |