Environmental Protection Technology Series
Pilot Plant Demonstration
of A Lime-Biological Treatment
Phosphorus Removal Method
Office of Research and Monitoring
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
-------
-------
EPA Review Notice
This report has been reviewed by the Office of Research
and Monitoring, Environmental Protection Agency, and
approved for publication. Approval does not signify
that the contents necessarily reflect the -views and
policies of the Environmental Protection Agency, nor
does mention of trade names or commercial products
constitute endorsement or recommendation for use.
-------
ABSTRACT
A 15,000-gallon per day pilot plant was constructed to demonstrate the
capabilities of a lime treatment process for phosphorus removal. The
process consists of lime treatment of the raw sewage with settling of the
resultant phosphorus-rich sludge in the primary clarifier. This is followed
by an activated sludge process where much of the remaining phosphorus is
incorporated into cell mass and subsequently removed.
Lime treatment of the raw sewage to a pH of 9.5 to 10.0 resulted in
phosphorus removals within the primary clarifier of about 80 percent. The
removal of BOD at this pH was about 60 percent as compared to the 35 percent
normally expected with settling alone. The biological process following lime
treatment neutralized this high pH due to microbial carbon dioxide production
down to pH 8.
Lime treatment prior to biological treatment resulted in a change of the
microbial life within the aeration tank over that normally expected. The
resultant higher sludge volume index must be taken into consideration in
design.
The process is capable of achieving removals of 90 percent of total phosphorus;
however, consistent removal is only assured at the 80% level.
This report was submitted in fulfillment of Project Number 17050 DCC, under
the (partial) sponsorship of the Environmental Protection Agency by the Kansas
State University, Department of Civil Engineering, Manhattan, Kansas 66506.
in
-------
CONTENTS
Page
CONCLUSIONS 1
INTRODUCTION 3
DESCRIPTION OF FACILITIES 7
RESULTS 15
MISCELLANEOUS STUDIES 37
DISCUSSION 39
ACKNOWLEDGMENTS 41
REFERENCES 43
APPENDICES 45
-------
FIGURES
No.
1. Phosphorus Removal versus Lime Addition 5
2. Schematic of Proposed Flow System 6
3. Schematic of Pilot Plant 8
4. Pilot Plant 9
5. Flow Control Device 10
6. Flow Measurement Apparatus 12
7. Flow Variations with Time, Manhattan Plant 16
8. COD Variations with Time, Manhattan Plant 17
9. SS Variations with Time, Manhattan Plant 18
10. Phosphorus Variations with Time, Manhattan Plant 19
11. Phosphorus Removal versus Primary Effluent pH 21
12. Primary Effluent pH versus Lime Dosage 23
13. Primary COD Removal versus pH 25
14. Primary SS Removal versus pH 26
15. Primary Sludge Production versus pH 31
16. Chemical Sludge Filtration Rate versus Polymer Dosage 33
-------
TABLES
No. Page
1. Average Wastewater Characteristics for Manhattan, Kansas .... 15
2. Alkalinity, pH, and Hardness of Manhattan Wastewater 20
3. Lime Dosage Variations for Different Periods of the Day for
a pH of 10.5 22
4. Typical Turbidity Analyses at pH 9.4 27
5. Effect of Lime Treatment on Chemical Characteristics
through Plant 28
6. 'The k Reaction-Rate Constants 35
7- ' Ratios of Lime Treated Sewage k values to Settled Sewage ... 36
yn.
-------
CONCLUSIONS
1. Lime treatment of raw sewage preceding biological waste treatment is
a workable method for phosphorus removal. The process is capable under
good conditions of 90 percent phosphorus removal. At a pH greater than
9.5 removals of 80% were almost always assured. The phosphorus concentration
of the wastewater under test was reduced from around 10 mg/1 down to about
1 mg/1. A minimum pH of 9.8 was required to obtain these removals.
2. The addition of lime to the raw sewage to a pH of 9.8 resulted in BOD
removals through the primary system of about 60 percent as compared to 30
to 35 percent removals without lime. Suspended solids removals did not
increase much although much of the solids in the primary effluent was
chemical precipitate which redissolved in the lower pH of the aeration
tank.
3. The microbial production of carbon dioxide within the aeration tank
is sufficient to neutralize the lime added in the primary system, even when
primary pH values up to pH 11 are used.
4. The lime treatment process has consistently resulted in a change of
microbial life within the aeration tank over that normally expected. The
rate of waste stabilization is not decreased, but a more voluminous sludge
is produced. With allowances for this, it should not prove to be a serious
hinderance to the process.
5. The lime treatment process produces two to three times more sludge than
would be produced in a conventional secondary treatment plant. Howeyer, with
proper polymer treatment the sludge can be easily and economically dewatered.
6. The recommended sludge treatment scheme would be polymer treatment,
vacuum filtration, or centrifugation, and then incineration and/or landfill.
7. Lime feed and control is not a highly critical operation so long as the
pH is greater than 9.5 to 9.8. Overfeeding up to a pH of 11 does not affect
the process and will only result in more sludge production and the wastage
of lime.
8. Primary clarifier surface overflow rates did not appear to be highly
critical for effective phosphorus removal from the primary unit. Most of
the floe is heavy and settles rapidly. However a residual fluffy floe still
will not settle well even down to overflow rates of 300 gpd/sf.
9. The addition of polymer to the lime treated raw sewage, either before,
during, or after lime treatment did not result in any significant increase
in phosphorus removal through the primary unit.
-------
INTRODUCTION
Removing phosphates from sewage at the waste treatment plant rather
than from the products making up the sewage, is a philosophy which is
receiving increased attention. The difficulty in finding a phosphate
substitute for detergents and the realization that removal from detergents
will only result in a partial solution to the problem has again focused
attention on phosphorus removal at the treatment plant.
Phosphorus quantity in sewage has increased almost four fold in the last
twenty years, due mainly to the use of synthetic detergents. These quan-
tities are substantiated both in the sanitary engineering literature and
in industrial phosphate production statistics.
Phosphorus removal was first attempted by the removal from final secondary
effluent (1)(2). This resulted in high chemical costs and,where lime was
employed, recarbonation to lower the pH after chemical precipitation had to
be practiced. The emphasis switched to removal by biological means, either
by algae (3), or by the activated sludge organisms as "luxury up-take" (4)(5).
These methods have not provided consistently good removals. Emphasis has
recently been placed on chemical precipitation of the phosphates in the
activated sludge aeration chamber (6)(7). This offers some advantages of
operation but the chemical costs are still quite high.
Rather than adding an additional process to a conventional plant, a systems
approach, integrating the entire treatment system, seemed to be the logical
way to proceed. Since the phosphate is precipitated readily by a variety
of chemicals,, it seemed essential that a chemical process should be a part
of the overall system.
Based on a review of the literature and jar tests in the laboratory, lime
was selected as the best coagulant from the standpoint of both cost and
efficiency. An examination of a typical curve, Figure 1, where phosphate
remaining is plotted against lime addition,yields the answer to the
reasoning behind the proposed process. The general shape of this curve
is applicable whether the wastes used are treated or untreated.
If 95% phosphorus removal from the wastewater given in Figure 1 is desired,
it will be necessary to reduce the phosphate concentration from 11.5 mg/1,
point A on the curve, to 0.6 mg/1, point C. From a practical viewpoint it
can be seen that the initial lime dosage is quite effective in removing
relatively large quantities of phosphorus. As the phosphate concentration
is reduced to a low level, below 3 mg/1, the lime dosage required to remove
a unit of phosphorous begins to increase quite rapidly. For this reason lime
treatment can best be used to remove the initial fraction of phosphate from
the wastewater.
The micro-organisms in activated sludge can readily utilize phosphorus at
very low concentrations. Uptake of phosphorus by the micro-organisms in
biological waste treatment systems is related to the amount of cells
synthesized which in turn is related to the organic load stabilized.
-------
[95% REMOVAL
I
r*
K
100 200
^ mg/l Co (OH) 2
Fig. 1 Phosphorus Removal versus Lime Addition
4
-------
Combining these two concepts it is immediately apparent that lime treatment
should precede rather than follow biological treatment for the most
economical design. By adding 100 mg/1 lime to the wastewaters in Figure 1,
it would be possible to drop the phosphate concentration to 2.3 mg/l» point
B on the curve. Thus, the lime treatment effects a 79% phosphate reduction.
It would have required 150 mg/1 more lime to obtain an additional phosphate
removal of only 16%, 0.6 mg/1. Biological treatment can be just as effective
at a phosphate concentration of 2.3 mg/1 as at 11.5 mg/1. In essence, the
microbes are used to remove that fraction of phosphates which require large
chemical dosages. A schematic diagram of the proposed flow system is presented
in Figure 2.
Initial laboratory research demonstrated that this phosphorus removal method
appeared to offer an economical solution to the problem. The purpose of this
research was to expand the laboratory study to include larger scale pilot
plant testing. The primary objective of this investigation was to develop
design and operating criteria for this recently developed phosphate removal
method of lime precipitation prior to biological treatment. Exploration of
expected operating problems, process sensitivity, and reliability were also
objectives of this investigation. Besides functioning as a phosphate removal
process, the process was to be examined as a treatment scheme which may extend
the useful life of overloaded plants, to handle slug loads, and for other
situations where a coagulant aid for sewage would be required.
-------
LINE
RAW
SEWAGE
RAPID MIX
ACTIVATED
SLUDGE
FIMAL\ EFFLUEMT
SED.
SLUDGE DISPOSAL
Fig. 2 Schematic of Proposed Flow System
-------
DESCRIPTION OF FACILITIES
The research was conducted with a 15,000 gallon per day activated sludge
pilot plant which was erected at the Manhattan, Kansas Waste Treatment Plant.
Figure 3 is a schematic of the set-up and Figure 4 is a photograph of the
pilot plant. Many equipment modifications were made during the early part
of the project. The final modifications will be reported upon.
The raw sewage is pumped from the inlet well of the Manhattan Sewage
Treatment Plant clarifier by a variable speed positive displacement pump.
This pump operates under approximately five feet of suction. The pump is
a Jabsco Model 6400, rubber impellered, positive displacement type. The
pump is modified by removing the guard teeth within the pump on the in-let
and out-let sides. The pump was belt driven by a 1-HP variable speed motor.
The suction line from the clarifier to the pump was 11/2 inch in diameter.
To capture unusally large solids which might plug the pump and to prevent
loss of prime on the pump a 30 gallon air tight container was installed
immediately prior to and in line with the pump. A vacuum pump was attached
to this tank to withdraw captured air withdrawn with the sewage from the
primary clarifier and to maintain a water level above that of the sewage
pump. The vacuum pump operated in response to a signal from a mercury float
switch within the tank to maintain a constant level. This tank was equipped
with a snap-off lid so that the accumlated solids and floatables could be
removed at a weekly interval.
The pumping rate was controlled by a program controller to simulate the
variable flow of sewage that would be expected to a treatment plant under
gravity flow conditions. A 24 hour clock with an attached cam operated
a lever connected by flexible wire to a counter weighted pulley attached
to the motor speed control knob. This system is more fully described in
an article published by the project director (8). A photograph of the
system is shown in Figure 5. Lime slurry was injected into the line
following the pump and immediately before the waste entered a 20 gallon
reaction tank (20 gallon water system pressure tank). Following this tank
were the immersion type pH electrodes. The tank allowed reaction time for
the lime and resulted in some flocculation. A Speedomax-H pH Meter, recorder
and controller by Leads and Northrup was the heart of the pH control. pH
was sensed by the immersed pH electrodes equipped with automatic temperature
compensator. The pH was recorded. A controller operated a variable speed
tubing pump which injected a lime slurry into the line depending on whether
the actual pH was above or below the controller set point. The controller
never did function to give good variable control, but operated only as an
on-off device. This, however, appeared satisfactory even though some over
shoot of pH always occurred. Lime was prepared in a 50 gallon barrel equipped
with an 8 inch motor driven propeller to maintain lime in its slurry form.
This can also be seen in Figure 5. A 50 pound bag of hydrated lime was mixed
with 50 gallons of water as required. Lime usage was determined by measuring
the drop in slurry level.
-------
Fl NAL
SED.
SLUDGE HOLDING
AND MIXING
FLOW
MEASURING
2=2
Fig. 3 Schematic of Pilot Plant
-------
vO
Fig. 4 Pilot Plant
-------
Fig. 5 Flow Control Device
10
-------
The lime treated sewage entered the primary settling tank which was an
8' x 8' hopper bottomed clarifier. At a 20,000 gallon per day design flow
the over flow rate would be 300 gallons per day per square foot. Settled
sludge was pumped from this tank automatically on an hourly basis by an
air lift pump.
Supernatant flowed over a V-notch weir to the aeration tank. The aeration
tank is a 3,000 gallon cylindrical tank, 8T in diameter and 8'4" long. Air
is supplied to the tank through rubber boot diffusers supplied by Smith and
Loveless Corporation. Two 2-HP blowers, each delivering 55 cubic feet per
minute at 3.5 psi are provided. One blower is adequate to furnish sufficient
oxygen so that the blowers can be alternated in usage.
The final clarifier is an 8' x 8' hopper bottom clarifier, identical in size
to that of the primary clarifier. Activated sludge return is accomplished
by a continuously operating air lift pump which can be controlled by a ball
valve. An air lift scum return device is used to transfer any scum back to
the aeration tank. Activated sludge is wasted by a 1 1/2 inch tap into the
activated sludge return pipe. Wasting is done manually by means of a gate
valve. The flow is measured in a weir box just prior to discharge to the
plant sewer. The weir is a 22 1/2 degree V-notch. The flow recording device
is a Stevens Model 61-R Flow-Meter with continuous strip chart recording in
gallons per day and equipped with a continuous volume totalizing display.
The recorder is run by an 8 day spring wound clock and is equipped with a
30 day chart. Figure 6 is a photograph of the weir box.
Raw sludge from primary settling can be pumped into a 200 gallon stock tank
where it can be held for measurement. This was the method for determining
sludge quality and quantity. An entire days production could be captured
and sampled. Waste activated sludge could also be drained into a similar
tank for flow measurement and sampling. The two type of sludges could be
blended into a third tank for further studies. These tanks can be seen in
the photograph, Figure 4.
Ins trumentation.
The turbidity of both the primary effluent and the final effluent was
continuously monitored. Due to the coating tendency of lime a surface
scatter turbidimeter was thought to be the only feasible method. A Hach
continuous reading surface scatter turbidimeter was selected for this.
The turbidity was continuously recorded with a 2 inch Rustrak strip chart
recorder. The primary and final effluents were siphoned by a 1/2 inch
plastic tubing back to the control building at a rate of 1/4 to 1/2 gpm.
A device was constructed to direct the flow of either the primary effluent
or the final effluent to the turbidimeter for a period of one hour. At the
end of one hour the turbidimeter chamber was automatically emptied and the
flow from the other effluent source returned to it. Thus, both effluents
could be monitored with one instrument. A Hach Model 31, continuous
recording phosphate analyzer was also installed to take the same flow
that was going to the turbidimeter. This system operated on a colormetric
11
-------
Fig. 6 Flow Measurement Apparatus
12
-------
method with the Aminonaphtholsulfonic Acid method for ortho phosphate.
Results were printed out on a 2 inch Rustrak recorder. Since it was a
colormetric procedure interferences could be expected due to solids present.
Also, it was recognized that only ortho phosphate would be measured. The
installation of this machine was not designed to eliminate laboratory
analysis but only to monitor the effluents for any sudden changes in ortho
phosphate concentrations. Daily maintenance was required on both instruments
to keep them functioning. The major problem was build-up of solids and slime
because of biological solids in the secondary effluent.
Operational Problems.
As with any other pilot research facility, this plant was not without its
operational problems. Some would only occur with a plant of this size while
others could be expected in a full scale plant. As long as a large solids
trap was ahead of the pump, plugging by large sewage solids and rags was no
problem. A Jabsco pump was very reliable and very seldom plugged. The
20 gallon reaction tank for lime additions was a source of problems. Solids
and matted hair eventually resulted in plugged lines. This problem
necessitated an occasional cleaning. The line following lime addition was a
1 1/2 inch ABS plastic line. This would coat with precipitate and would
eventually plug. Every three to four weeks the line had to be flushed with
acid to clean it out. Similar problems may be expected in a large plant.
Coating of the primary effluents weirs was also encountered. This could be
cleaned with a hose which is normal maintenance anyway at existing plants.
The pH electrodes were also subject to this lime fouling. Several days
accumulation resulted in loss of sensitivity. The electrode assembly was
removed about every two days, dipped into acid which quickly removed the
deposits, and returned to service. Calibration was checked weekly but it
usually did not require adjustment.
Another problem encountered in this plant was that the sludge did not settle
well in either the primary or the final tank. The solids tended to cling to
the 60° sides or the solids would bridge over near the sludge pump inlet.
If the interior sides of the settling tanks were not cleaned at least every
other day by high pressure water, septic sludge resulted in the primary, and
denitrification occurred in the secondary. This resulted in a large amount
of floating solids occurring in both tanks. This was not a fault of the
process but instead was the fault of the equipment. In summary the major
operational problem which can be attributed to the system itself is the lime
fouling of both the monitoring and control equipment.
13
-------
RESULTS
Pilot Plant Influent Wastewater Quality
The influent wastewater for the pilot plant was raw sewage from the city
of Manhattan, Kansas. The city of Manhattan serves a sewered population
of approximately 40,000. Kansas State University with its student popula-
tion of about 15,000 is included within this figure. There is very little
industry within this town so industrial wastes are a very minor constituent
of the wastewater. This sewage can be classified as primarily domestic.
Samples of Manhattan wastewater were collected from June 1970 through
August of 1971. Samples were generally taken twice weekly and were collected
at the hours of 3:00 P.M., 9:00 P.M., 3:00 A.M. and 9:00 A.M. Analyses were
normally run on each sample and a composite value calculated according to
flow. At times the samples were composited first and a single analyses run.
The following results in Table 1 are the averages of the raw sewage samples
showing the variations throughout the day.
Table 1. Average wastewater characteristics for Manhattan, Kansas.
Time COD SS Total P
mg/1 mg/1 mg/1
3 AM 270 100 7.71
9 AM 395 185 8.63
3 PM 575 215 12.17
9 PM 510 225 10.34
Calculated Composite
Based on Typical Flow 465 195 10.01
These were then composited according to a typical plant flow, Figure 7, to
give the calculated composite values. Figures 8, 9, and 10 depict some
variations from two sampling runs with samples collected hourly. These
values are in very close agreement with the values of Table 1. Analyses
were made periodically throughout the study to determine the hardness,
alkalinity, and pH of the Manhattan wastewater. The average composite
values are reported in Table 2. These values did not change greatly over
the period of the study.
15
-------
CD 3
LJ
<
o: 2
^
O
_i
U.
3/2- 3/3,197!
, 6
-_0 o-- 3/31-4/1. 197]
I I I I I I I I I I I I I
"1 i .^ ' "
I (ME OF DAY
Fig. 7 Flow Variations with Time, Manhattan Plant
-------
iZOOr
1000
\ 800
Q
-z.
< 600
U)
Q
LJ
X
o
400
o
i: 20°
LJ
i:
o
2/E- 3/3 1971
C--"-0- 3/31-4/1 1971
I__LLJL_L_L_L
JLJ.
PM
8 10 12 2 4 6 8 10 12
AM N OON
TIME OF DAY
Fig. 8 COD Variations with Time, Manhattan Plant
-------
00
10001
800
_J
\
600
CO
Q
O
CO
400
Q
Id
Q
2
UJ
0.
CO 200
ID
CO
3/2-3/3 1971
_o0__ 3/31-4/1 J97I
I I I I
6
PM
8
10 12 2
M1B
TIME OF DAY
6
AM
8
10
12 2
NOON
Fig. 9 SS Variations with Time, Manhattan Plant
-------
~- 3/2- 3/3 1971
_A\/E3/2 9J, MG/
Fig. 10 Phosphorous Variations with Time, Manhattan Plant
-------
Table 2. Alkalinity,, pH, and hardness of Manhattan wastewater,
composite values. Expressed as mg/1 CaCO .
PH
7.2
Alkalinity
235
Calcium
110
Magnesium
40
Total Hardness
150
In general it appears that the Manhattan wastewater can be considered to be
a very typical domestic sewage.
Effect of pH on Phosphorus Removal
Many variables of operation changed throughout the period of this study,
flow rate, method of lime feed, method of pH control, temperature, waste
characteristics, etc. The single factor that most affected phosphorus
removal however was pH. This was expected since the mechanism of removal
is by precipitation with the solubility of the calcium phosphate hydroxy
apatite precipitate being a function of hydroxide ion concentration.
Over the period of the study the pH of the lime treated sewage was varied
from slightly above neutral up to pH 11.4. Figure 11 is a plot of the
phosphate removal through the primary system versus primary effluent pH for
select runs. "Selected runs" refer to 24 hour samples which were collected
at a time when the plant was known to be operating consistently for a period
of several days prior to a sampling day. Problems were often encountered
with flow, pH control, or lime feed and all of these systems had to be
working for a period of time to obtain valid data.
Since most of the phosphorus is removed in the primary precipitation phase,
it will be analyzed separately. Figure 11 shows some increase in removal
as pH increases to 9, but it is quite slight. Between pH 9 and pH 10
phosphorus removal increases rapidly. Above pH 10 removal levels off
rapidly and we get very little additional removal by increasing the lime
dosage any further. The desired operating pH depends upon what removal is
desired through the system. Operating pH's of around 10 are capable of
80 percent removals through the primary alone while operating at pH's of
11 or 11.5 could result in removals of 90 percent or better. At these
higher pH's other factors occur to reduce the desirability of operating
at high pH.
Even though soluble phosphorus was very low at a pH 10, it was difficult to
get removal consistently above 80 percent. The reason for this was incom-
plete settling and floe carry over. Changing the flow rate and consequently
the surface overflow rate of the primary clarifier, which varied by a factor
of three during the study: did not seem to appreciably change the amount of
this floe carry over. There was always some light fluffy floe particles of
about pin head size that could be seen passing over the primary clarifier
wiers.
20
-------
lOOr
o
60
co
o 40
Q_
CO
CD
20
1
8 9
PRIMARY EFFLUENT pH
10
II
Fig. 11 Phosphorus Removal versus Primary Effluent pH
-------
Phosphorus Removal through the Biological Process.
Removal of phosphorus through the secondary portion of the plant was muc
poorer than anticipated. For the data between pH 9.5 and pH 10.5 only an
additional 2.5 percent removal was obtained through the biological portxon
of the plant. This can be explained partially by the nature of the^plant.
The final settling tank was a single hopper bottomed tank with an air lift
sludge return. The tank had sides sloped 60 degrees and much trouble was ^
encountered in getting solids returned from the secondary clarifier. Solids
tended to hang up on the sides of the tank where they underwent anaerobic
action and denitrification occurred. This resulted in much of the sludge
floating to the surface of the clarifier. In general the final clarifier
was usually anaerobic due to the inability to return solids properly.
Anaerobic conditions tend to cause the release of some of the phosphorus
that is either within or absorbed to the cells. Studies on the biological
process for removing phosphorus have emphasized the need to return the sludge
promptly and avoid anaerobic conditions unless you wish to strip the phos-
phorus from the sludge. The poor removal through the biological portion of
the process is not considered to be a fault of the process, but instead is
a fault in the plant design.
pH versus Lime Requirements.
Lime was purchased in 50 pound bags of high calcium, hydrated lime (Ca(OH)2).
Its strength was checked several times by titration against standard acid and
it was found to average about 94 percent CaCOH^. Lime was made up in a
slurry form by mixing with tap water and was kept stirred by mechanical means.
Lime dosage was determined by measuring the drop in lime slurry within the
feed barrel. The decrease over a measured period of time was converted to
weight of lime and was then divided by the total flow for the measured time
to get mg/1 of lime added. Figure 12 is a plot of some typical values
showing lime dosage versus pH of primary effluent. The values given in
Figure 12 are average values over 24 hours. Any instantaneous lime concen-
tration demand to maintain a desired pH will vary throughout the day as the
strength of the sewage changes and as the alkalinity may change. Some values
giving an indication of this magnitude are presented in Table 3. These are
for a time period of 2 days during which the pH was constant at pH 10.5.
Table 3. Lime Dosage Variations for Different Periods of the Day
for a pH of 10.5.
Time Interval Lime Dosage
8:30 a.m. - 4:30 p.m. 340 mg/1
4:30 p.m. - 8:00 a.m.- 302 mg/1
8:00 a.m. - 2:30 p.m. 332 mg/1
2:30 p.m. - 4:30 p.m. 360 mg/1
4:30 a.m. - 8:30 a.m. 320 mg/1
22
-------
121
Q.
h-
UJ
ID
_1
U_
U_
LL!
en
10
cr
Q.
8
7
0
100 200 300
LIME DOSAGE
MG/L
400
Fig. 12 Primary Effluent pH versus Lime Dosage
23
-------
Lime Feed and pH Control
There are several alternatives to controlling the amount of lime to be fed
to the process. This research explored these and found that several methods
would work. The best method of course would be to have a pH sensor, recorder
and controller to automatically increase or decrease the lime dosage to
maintain a constant pH at all times. This was the method of choice in this
project. However, as mentioned earlier, the design of the system did not
allow the controller to function as it should and the pH at the sensor would
overshoot the set point. The lime feed pump did not operate as a variable
speed pump but instead was strictly an on-off pump. By the time the small
portions of overlimed sewage and underlimed sewage reached the settling tank
they were generally well enough mixed to give a resulting pH equal to the
desired pH. This probably had some adverse effect upon the process but it
was not felt to be great. Another method of lime feeding would be to have
the lime be fed proportional to the flow. Thus, rather than be tied into
a pH sensor the variable lime feed device could be connected into the plant
flow measurement system for control. Table 3 showed that although lime
demand did vary over the day it didn't vary greatly, less than +10 percent
of the average. This system was used for many runs, especially when the pH
meter or electrode system was under repair. This system was able to hold
a constant pH nearly as well as the regular pH control system.
Another method would be to feed the same amount of lime over the entire
day or possibly change it every four to eight hours. This would create the
greatest variability due to both the varying flow and the varying strength
of sewage, but it may have applications for small plants or under temporary
conditions. The dosage could be set to give the desired pH at the maximum
flow. This would result in overdosing at lower flows but it has been
found that this process can handle overdoses very well. The major problem
involved here is wastage of lime and increased production in sludge.
Changes in Other Waste Constituents with Lime Treatment
An additional benefit to be realized by lime treatment of raw sewage is the
removal of waste constituents other than phosphorus. The removal of addi-
tional organic material from the primary system reduces the organic load
in the secondary portion of the plant and may result in some cost savings
for that portion of the plant. Lime has always been recognized as a
coagulant aid. It was first used to treat sewage back in the mid 19th
century (9). It was recognized at that time that lime combined with
phosphates provided better clarification of sewage. Phosphates were added
at some plants as a coagulant aid and a patent was issued for that process.
Our present sewages do not require a phosphate addition since this has been
added to sewage by way of the detergents. Lime treatment for clarification
enjoyed a resurgence in this country in the mid 30's but again fell out of
favor as biological processes began to be added which eliminated the need
for additional treatment within the primary treatment.
Figures 13 and 14 plot COD and SS removals from the primary system versus
pH, which correlates with lime dosage. COD removal shows a marked increase
24
-------
lOOr-
80
to
Cn
I
#
60
Q 40
o
u
20
I
I
8 9
PRIMARY EFFLUENT pH
10
II
Fig. 13 Primary COD Removal versus pH
-------
100
I
8 80
u
en
8 so
.J
O
CO
O
U40
Q
u
CL
i
8 9
PRIMARY EFFLUENT pH
10
tl
Fig. 14 Primary SS Removal versus pH
-------
from about 20 percent to 25 percent COD removal without lime treatment up
to a 60 to 70 percent removal at a pH of about 10.0. BOD values were not
run continually like the COD's but the removal of BOD is roughly about the
same as the COD removals. Suspended solids removal did not show a great
increase in removal by the addition of lime. However, the solids in the
primary effluent are not the same solids that entered the tank in the raw
sewage. Primary effluent suspended solids appeared to be primarily precipi-
tate carry over, the combination of the calcium phosphate and calcium
carbonate precipitate. These solids would redissolve in the lower pH of the
aeration tank. Even though the solids removal did not show an increase on a
weight basis, the primary effluent with lime treatment was of considerably
more clarity than the primary effluent without lime treatment. This is
supported by observation of turbidity values. Turbidity values varied from
around a low of 10 standard turbidity units (STU) in the early morning hours
to a peak of about 40 STU in the late afternoon hours. Primary effluent
without lime treatment would usually range over 100 STU during the day time
hours. Table 4 gives some results of turbidity analyses for an operating
pH of 9.4.
Table 4. Typical Turbidity Analyses at pH 9.4
Time Settled Sewage Primary Effluent Final Effluent
Lime Treated
9:00 a.m.
3:00 p.m.
9:00 p.m.
3:00 a.m.
130
150
89
45
2
54
41
28
1
4
2
1
There is little change in the weight of suspended solids in the primary
effluent for lime treated versus non-treated sewage but yet the clarity with
lime treatment is much better. This indicates that the smaller colloidal
particles of the sewage are being removed and replaced by the larger chemical
precipitate.
The effect of lime dosage on alkalinity, pH, and hardness can best be shown
by a typical analyses run on a 24 hour composite sample. This is illustrated
in Table 5.
27
-------
Table 5. Effect of Lime Treatment on Chemical Characteristics
Through Plant
Sample pH Calcium Magnesium Pheno. Total
mg/1 CaCO mg/1 CaCC>3 Alk Alkalinity
Raw Sewage 7.1 96 36 0 202
Primary Effluent 9.6 136 36 80 232
(Lime Treated)
Final Effluent 8.0 160 28 0 107
It should be pointed out that the composite is made from only four samples
over a 24 hour period so that due to retention time through the units a
completely accurate picture cannot be shown. This shows for example in the
discrepancy between the calcium content of the primary effluent and the final
effluent and yet no calcium is added within this stage.
The general conclusion from this and the rest of the data is that the hard-
ness will be increased but not in direct proportion to the lime added. Some
calcium carbonate will precipitate removing some of the added calcium. This
is also supported by the reduction in total alkalinity from the raw sewage
to the final effluent indicating that some of the bicarbonate has been
precipitated as calcium carbonate.
Effects of Lime Treatment on the Biological Treatment Process
Laboratory research by the investigator in an earlier study (10) had shown
that the buffering capacity of the aeration tank in a complete mixed system
could handle fairly high primary effluent pH values. The results of this
pilot study has supported this original claim completely.
The pH in the aeration tank was essentially the same as the final effluent
at all times. The normal aeration tank pH with no lime treatment was about
7 to 7.2. Primary effluent pH values of about 9.5 resulted in aeration tank
pH values of 7.5 to 8.0. Lime dosage was increased slowly over a period of
several weeks until the lime treated sewage had a pH of only 8.5. It can be
concluded then that the buffering capacity and the carbon dioxide produced
by the metabolic activity is sufficient to neutralize the high pH values
following lime treatment of sewage. The addition of lime treated sewage to
an activated sludge system will have an affect upon the microbial life within
the tank. Protozoa life, however, did not change from what would be con-
sidered a normal activated sludge distribution. The predominant species was
ciliates and the population changed from free swimming ciliates to rotifers
depending upon the organic load imposed upon the tank. A laboratory study
consisting of batch feed units, one unit receiving settled sewage, the other
28
-------
unit receiving lime treated sewage, was conducted during the course of the
field study. This also helped provide a comparison between the sludges.
The activated sludge produced with lime treatment did not have the
characteristic dark brown color usually associated with activated sludge.
This sludge was instead a light brown-gold in color. This was especially
noticeable in the laboratory units where a direct comparison could be made.
The sludges in both the laboratory and the field units were more filamentous
than the controlled laboratory unit receiving only settled sewage. This was
a very fine filamentous growth which appeared to be actinomycetes. It was
mixed in with normal appearing activated sludge floes. The filamentous
growth wasn't predominant but it was sufficient to result in bulking sludge.
At times the pilot plant had sludge volume index values up to 600. A more
normal value was a sludge volume of 300 which is about what was experienced
in the laboratory unit also.
There was no^ real explanation found for this trend to sludge bulkiness.
Iron shortage was suspected and iron was added to the aeration tank on a
continuous basis for several weeks. This was at a concentration of 3 to 5
miligrams per liter of iron. The sludge volume index decreased drastically
but it is felt that this was due to precipitation and weighting of the sludge
rather than a change in the microbial nature of the mixed liquor suspended
solids itself. Microscopic observation still showed the presence of about
the same number of filamentous organisms. This increased density with
chemical addition may provide a clue, however, to the reason for settling.
The increased primary clarification due to lime treatment will remove more
of the nonvolatile and heavier suspended solids than will conventional
settling. This means that there is less build up of inert solids in the
aeration tank. It is believed that these heavier inert solids may help
the settling in the normal activated sludge process. Their absence from
the lime treatment process means that even if the sludge was of the same
microbial content it may settle slower. The mixed liquor volatile suspended
solids content of the activated sludge in the lime treated process also
supports this conclusion. Volatile solids values were generally in the
range of 80% to 85%. This compares to 70% to 75% normally expected in
conventional activated sludge and with about 75% found in the laboratory
unit receiving settled sewage.
Sludge Production.
One of the major costs and problems generally associated with chemical
treatment of sewage is the handling of the larger quantity of sludge that
is produced. As mentioned earlier in the report primary sludge was pumped
automatically on an intermittent basis by an air lift pump. This sludge
was pumped into two 200 gallon tanks where an entire day's sludge could be
accumulated. After 24 hours the volume was recorded. The tanks were then
stirred and sampled. The values so determined were calculated as pounds
of dry solids per million gallons of flow.
Sludge tended to hold up in the hopper bottom tank with the result that it
was not always determined for certain that the sample collected included
29
-------
the entire days production. The inside tank walls were hosed down at the
start and finish of a sludge collection run in an attempt to assure better
results.
A plot of primary sludge production versus pH of primary effluent is shown
in Figure 15. This illustrates the rapid increase in sludge production
with increasing pH. The calculated values shown for no lime addition is
based upon the average raw sewage suspended solids concentration and an
estimated suspended solids removal of 60%. At an operating pH of 9.5, over
three times as much sludge will be produced with lime treatment as without.
The volatiles solids content of this lime sludge averages between 45% and
50%. In comparison, the volatile solids content normally expected for
sludge produced without lime treatment is in the range of 60% to 65%.
Sludge density is another important characteristic. The density of the
sludge produced by lime treatment was generally about 3% after 30 minutes
of settling. As will be discussed later, the addition of polymers to the
sludge greatly increased this density.
Secondary Sludge Production.
Determination of biological solids production was a much more difficult
thing. Sludge wasting was a very intermittent thing as for many periods of
time solids lost in the effluent were sufficient to hold the mixed liquor
suspended solids at a constant level. An estimate of this production was
made by following the increase in mixed liquor suspended solids over a
period of one to two weeks and using the average effluent suspended solids
values. The biological solids production from four determinations made in
this fashion for primary pH values between 9.5 and 10 were 42, 50, 46, and
55 or an average of 48 mg/1. These values were the production per pass
through the aeration tank. Average COD of the primary effluent was
230 mg/1 for these periods for a solids production to COD ratio of 0.208.
Assuming a BODr to COD ratio of 0.42 for primary effluent after lime
treatment, a figure substantiated from 10 samples for which BOD was run,
the biological solids production could be estimated at about 0.5 pounds of
biological solids produced per pound of BOD5 added to the aeration tank.
This figure is high but within the realm of reason. Sludge age values
averaged about 12 to 15 days for a mixed liquor suspended solids level of
around 2000 mg/1. This is a longer sludge age than normally encountered for
aeration times of 8 to 10 hours, but it can be explained by the fact that a
much lower strength waste is being used due to the pre-treatment procedures,
and also that more of the inert non-volatile solids has been removed in
the primary treatment system.
Sludge Treatment.
The lime treatment process produces considerably more sludge than what would
be produced in a conventional treatment plant. The total sludge produced in
a typical lime treatment process may total 3750 Ibs per million gallons of
which only about 7% by weight may be secondary sludge. In a conventional
30
-------
6000
CD
^5000
Q
2
O
4000
o
cj
Q
O
QC
Q.
U
CO
1000
I
10
PRIMARY EFFLUENT PH
Fig. 15 Primary Sludge Production versus pH
II
-------
plant total sludge production may be around 1500 Ibs. per million gallons
of flow of which over 35% by weight may be secondary sludge.
Lime sludge was found to dewater very well with proper polymer treatment.
In the original laboratory research Dow A-21 was found to be about the only
polymer that really worked well (10) . Dow Chemical has replaced this with
A-23 which works even better although the price per pound increased over that
of the A-21. Figure 16 is a plot of filtration rate versus polymer dosage
for lime sludge alone and for mixtures of lime sludge and waste activated
sludge. This mixture is approximately the proportion that would be expected
from a lime treatment plant. The mixing of activated sludge with the lime
sludge results in a decrease in filtration rates. Whether this is entirely
due to the solids themselves or mainly due to the diluting effect of the less
concentrated activated sludge cannot be ascertained. However, even at the
lower filtration rates for the combined sludge, filtration rates of 5 lbs/sf/
hr can still be obtained for polymer dosages of 2.5 Ibs/ton of dry solids.
This is equivalent to a chemical cost for vacuum filtration of about $9 per
million gallons of flow. This cost is within the range normally expected for
dewatering a mixture of raw and activated sludge from a conventional plant.
Biological solids are difficult to dewater by either vacuum filtration or
centrifugation. The reduction of biological solids requiring treatment
and the reduction in percentage of total sludge treated is a definite
advantage of this process.
Ultimate sludge disposal may be by incineration or land fill directly. The
use of anaerobic or aerobic digestion is not recommended for the treatment
of this sludge. Either process could result in the release of some of the
phosphate into the supernatant which would have to be treated again. The
primary reason for not digesting this sludge, however, is that the chemical
sludge would occupy too much digester volume. Digester requirements may
have to be more than doubled in order to obtain the same degree of digestion.
The rapid concentrating effect and the conditioning for the dewatering of
this lime sludge by Dow A-23 makes the large volume of chemical sludge
produced in this process less of a liability than what might have been
expected.
Effect of Lime Treatment on the Biodegradability of the Waste.
Observation of pilot plant data from this study revealed in the early runs
that the treatment plant was achieving very good overall BOD and COD
removals. Total BOD removals up to 98% were obtained, which is well above
that normally expected in a biological treatment process. A laboratory
study was set up to determine and compare the biodegradability of lime
treated sewage versus settled sewage.
Some recent research has indicated that the lime treatment results in
hydrolysis of the high molecular weight compounds with the result that
resulting lower molecular weight compounds are more amenable to carbon
absorption (11). Lower molecular weight compounds should also be more
32
-------
CO
\
CO
CD
O
UJ
a:
§3
a:
^
u_
UJ
Q
ID
100 °/o L»ME SLUDGE
AvE. SOLIDS 3.7 %
0 ACTIVATED SLUDGE
75 c/o LIME SLUDGE
AVE. SOLIDS 2.9 °/o
I
I
0
I
2 ' 3
LBS / TON DRY SOLIDS
POLYMER DOSAGE
Fig. 16 Chemical Sludge Filtration Rate versus Polymer Dosage
-------
amenable to biological treatment. Another theory is that more participate
and colloidal BOD is removed by lime coagulation leaving primarily a soluble
readily degradable material. These theories were explored by laboratory
determinations of k rates (BOD reaction rate constant) by means of a Warburg
apparatus.
Two 3-liter laboratory activated sludge units were maintained on a batch
feed basis. One unit received settled sewage from the Manhattan wastewater
treatment plant while the other unit received lime treated sewage from the
pilot plant. Each day 1.5 liters were decanted off following settling and
replaced with either the settled sewage or the lime treated sewage.
It was theorized that the biological solids which were developed on lime
treated sewage could also have an effect on the biodegradability, so the
experiment was designed to explore this effect also.
Three different operating pH values were used; 9.8, 10.5, and 11.5. Raw
sewage was treated with lime to these pH values flocculated and the super-
natant decanted. The different combinations used were settled sewage with
microbial solids grown on lime treated sewage, settled sewage with microbial
solids grown on settled sewage, lime treated sewage with microbial solids
grown on lime treated sewage, and lime treated sewage with microbial solids
grown on settled sewage. All microbial solids were thoroughly washed of
any remaining substrate before addition to the Warburg flasks. A detailed
description of the research procedure is presented by Flournoy (12).
Table 6 summarizes the results of this study. The kj values reported are
to the base 10. Corrected values refer to the fact that the uptake rates
were corrected for endogenous respiration as measured by an unfed blank.
Table 7 shows the ratio of k^ values of lime treated sewage to that of
settled sewage. The values are the average of three separate runs at each
pH. The term SS solids refers to microbial solids developed on settled
sewage while lime solids refers to the use of microbial solids developed
on lime treated sewage.
This data shows that the kj value, the BOD reaction rate constant, is
increased significantly by lime treatment. This would mean that the effluent
from lime treatment is more biodegradable than the effluent from conventional
settling. However, the thesis that the microbial solids developed on lime
treated sewage may aid in this increased removal proved to be false. In fact,
there seemed to be a definite relationship the other way, indicating that
microbial solids developed in this substrate were not as efficient.
Because these two conclusions tend to nullify each other, the only real
conclusion that can be made is that the biological process is not signifi-
cantly altered by prior lime treatment of the waste. However, this in
itself is a significant conclusion.
34
-------
TABLE 6. The k Reaction-Rate Constants
Runs
la
Ib
Ic
Lime
Treated
Sewage
PH
9.8
9.7
9.8
MLSS
Cone.
mg/1
1000
2000
3000
Final
Suspended
Solids
Cone, in
Mixture
mg/1
200
400
600
Combinations
Uncorrected
SS
SS
.36
.64
.98
Solids
Lime
.75
.69
1.37
Lime
SS
.49
.86
.88
Solids
Lime
.67
1.01
1.04
SS
SS
.43
.72
1.29
Corrected
Solids
Lime
.86
.96
1.86
Lime
SS
.63
1.93
.89
Solids
Lime
.85
1.15
1.15
2a
2b
2c
10.5
10.5
10.5
1000
2000
3000
200
400
600
.41
1.06
1.07
.87
1.29
1.15
.23
.42
.93
.47
.28
1.00
.49
.97
1.12
1.38
.22
.
.97
.42
1.04
3a
3b
3c
11.5
11.5
11.5
1000
2000
3000
200
400
600
.54
1.00
1.28
.90
1.31
1.47
.48
.73
.61
.22
1.08
.69
.63
1.00
1.23
1.89
.48
.77
.29
1.25
..___
Before neutralization.
"Cone, of 1 ml activated sludge solution added to flask.
-------
TABLE 7. Ratios of Lime Treated Sewage k Values to Settled Sewage
Co
Lime
Run Treated,^
Sewage pET ;
la
Ib
Ic
Average
2a
2b
2c
Average
3a
3b
3c
Average
9.8
9.7
9.8
10.5
10.5
10.5
11.5
11.5
11.5
MLSS,9,
ConcU;
mg/1
1000
2000
3000
1000
2000
3000
1000
2000
3000
SS Solids
2.1
1.1
1.4
1.5
2.1
1.2
1.4
1.6
1.7
1.3
1.1
1.4
Lime
Solids
1.4
1.2
1.2
1.3
2.0
0.7
1.1
1.3
0.5
1.5
1.1
1.1
Before Neutralization.
"Cone, of one activated sludge solution added to flask.
-------
MISCELLANEOUS STUDIES
Addition of Polyelectrolyte to Increase Primary Clarification.
Difficulty was encountered in obtaining removals of phosphorus of greater
than 80% from the primary portion of the plant. As was discussed earlier
the major carry over of phosphorus was not due to soluble phosphorus but
unsettled precipitate which appeared as pin point flocks. It had already
been demonstrated that the Polymer Dow A-23 worked very well in treating
this lime sludge for dewatering. If was theorized that the addition of
polymer immediately after lime addition would aid in coagulation. Polymer
addition was tried at several points of the process from before lime addition
to just prior to where the flow entered the primary tank. The best point,
as determined by floe observation, was immediately after lime addition.
A two week series of polymer addition was initiated with lime dosages set
to give a resultant pH of about 10. The polymer dosage was increased until
a visual difference could be discerned in the floe. This required a dosage
of about 1 mg/1 of Dow A-23. A dosage of 1.5 mg/1 was selected for the run.
Over a two week period there appeared to be a slight improvement in the
phosphorus removals and the suspended solids removals but there was not
enough improvement to establish a statistically reliable difference. A
polymer dosage of 1.5 mg/1 would require 12.5 pounds of polymer per million
gallons of flow for a cost of $25 per million gallons. This is more than
the lime cost alone. It was hoped that some of this polymer cost could be
recovered by the reduced requirement for polymer for sludge conditioning.
This was not the case, however. Apparently, the long retention time with
this biologically active raw sludge reduced the effectiveness of this polymer
for sludge dewatering. Approximately the same amount of polymer had to be
added to dewater as was required to dewater the lime sludge alone. The
polymer addition was increased to 2.5 mg/1 with no apparent increase in
performance.
The conclusion can definitely be drawn that addition of this polymer with
the lime was of no economic benefit.
Primary Sludge Recirculation.
Recirculation of chemical sludge to improve clarification has long been
practiced in the water treatment industry. A short study was undertaken in
this research to determine if primary sludge recirculation would aid in
chemical reduction and phosphorous removal. The sludge was pumped from the
primary clarifier hourly into the 200 gallon sludge storage tank. Sludge was
pumped from the bottom of this tank by a rubber impellered Jabsco pump, the
same model as used for pumping raw sewage to the plant. A variable speed
motor driving this pump allowed the sludge to be pumped back into the influent
line just prior to where it entered the primary clarifier. Trouble was
encountered with the mechanics of the system but it was operated off and on
for three days. The sludge return rate was high, about 50% of the raw flow
so retention time within the settling tank was effectively reduced. Only
several grab samples were obtained for this time. Continuous turbidity
37
-------
monitoring and soluble phosphorous monitoring showed no improvement in the
primary effluent. Lime dosage remained the same to allow a pH of about 10.
After about the second day of recirculation, with some waste to take care
of sludge produced, the sludge became more septic and the primary effluent
worsened. By visual observation it was noted that the primary effluent was
more turbid and flocculation was poorer with the sludge recirculation.
This portion of the study had to be dropped due to failure of the mechanical
equipment. It is realized that because of the shortness of this study that
definite conclusions cannot be drawn but it is worth mentioning because of
its implications. This sludge is not an inert chemical sludge as is normally
encountered in water treatment. It is instead a mixture of a highly
putrescible organic sludge and a chemical sludge. Recirculation of this
type of sludge would greatly lengthen the time that this putrescible portion
of the sludge would remain in contact with the carriage water. Decomposition
of this organic matter results in a solubilization which would tend to release
organic matter, or BOD, back into the system. With recirculation of this
high solids sludge at the low strength night time flows it would appear that
it would be possible to have an effluent worse than the influent.
Ideally, it would appear that the chemical sludge and raw sewage solids
mixture should be removed from the primary tank as rapidly as possible and
not be returned as a recirculated sludge.
38
-------
DISCUSSION
Lime treatment of raw sewage prior to biological treatment appears to offer
one solution to the removal of phosphorus from sewages. The process is
capable of consistent removals in the area of 80% but it cannot be relied
upon if removal greater than this is consistently desired.
Lime control is best done with a control system which will allow for a
constant pH. It was found that maintaining a constant pH is not highly
critical in this process. As long as the pH is above 9.5 good removals will
occur. About the only detriment caused by a lime overdose is an increase in
sludge and a wastage of lime. The biological process following lime treat-
ment is not highly sensitive to these changes in pH as long as the primary
effluent stays below pH 11. This process involved a complete mix activated
Sludge system. A conventional system or a contact stablization system would
be expected to be more susceptible to this higher pH because the waste would
not be distributed over the entire tank contents. Even though more care
would need to be exercised with these other biological treatment systems, it
is felt that this process could still be used. This could also include the
trickling filter treatment method.
The production of carbon dioxide by microbial life within the aeration tank
is sufficient to neutralize the excess lime from the chemical treatment
phase. Effluent pH values were below pH 8.5.
Lime treatment prior to biological treatment will result in a change of the
micro'bial life within the aeration tank as compared to what it would be
without lime treatment. This was evidenced by changes in the sludge
appearance and also by an increase in the sludge volume index. Although
the sludge volume index was about three times that normally expected it did
not result in any major problems in the operation of the plant.
The major disadvantage of this process is the larger quantities of sludge
which is produced. Any other chemical precipitation process for phosphorus
removal also suffers this problem. Sludge production, including the chemical
primary sludge and the waste activated sludge is about two and one-half
times greater than for a comparable plant operating without phosphorus
removal.
This sludge is, however, easily dewatered by chemical treatment. The
recommended sludge treatment scheme would be polymer treatment, vacuum
filtration or centrifugation, and then land fill or incineration.
Many plants use anaerobic digestion followed by some form of ultimate sludge
disposal. The effect of this chemical sludge upon the anaerobic digestion
process was not a part of the study. The major detriment would probably be
the decrease in digester volume due to the inert chemical sludge. The
chemical sludge should result in a well buffered system within the digester.
Some solubilization and return of phosphorus to the system by digesters
39
-------
supernatant return is quite likely, but the high ion concentration of
calcium and phosphorous in this sludge should prevent most of it from
solubilizing. This is an area which requires further study.
This process is very applicable where phosphorus removal is required at
existing plants. Relatively little modification is required to existing
facilities. Lime storage, lime feeding and pH control equipment need to be
added. The existing tanks can be utilized. What additional needs to be done
to handle the increased sludge would vary with each individual plant. This
research has shown that this process is one that should be considered when
phosphorus removal is required. The final process to be used should be
selected only after bench scale or pilot scale tests have been completed
with the waste in question.
40
-------
ACKNOWLEDGMENTS
The support of the city of Manhattan, Kansas, for providing the site for
the project, the utilities and the wastes is gratefully acknowledged.
Special thanks must go to Virgil Spain, Superintendent of Waste Treatment
and his crew for helping in the operation of the plant and in collection
of the many samples.
Mr. Ralph Flournoy, graduate student, performed many of the analyses and
assisted in the research.
The support of the project by the Office of Research and Monitoring,
Environmental Protection Agency and the Grant Director, Mr. Ed Earth,
is acknowledged with sincere thanks.
41
-------
REFERENCES
(1) Lea, W. L., Rohlich, G. A., and Katz, W. S. "Removal of Phosphates
from Treated Sewage," Sewage and Industrial Wastes, 26, 261 (1954).
(2) Malhotru, S. K., Lee, G. Fred, Rohlich, G. A. "Nutrient Removal from
Secondary Effluent by Alum Flocculation and Lime Precipitation," Int.
Journal of Air and Water Pollution, J3, 487 (1964).
(3) Bogan, R. H., Albertson, 0. E., and Pluntze, J. C., "Use of Algae in
Removing Phosphorus from Sewage," Journal of the Sanitary Engineering
Division, ASCE 86, No. SA5 (Sept. 1960).
(4) Vacker, D., Connell, C. H. and Wells, W. N., "Phosphate Removal Through
Municipal Waste Water Treatment at San Antonio, Texas," Journal WPCF,
39:5:750 (1967).
(5) Levin, G. V. and Shapiro, J., "Metabolic Uptake of Phosphorus by
Waste Water Organisms," Journal WPCF, 37:6:800 (June 1965).
(6) Ekerhardt, W. A., and Nesbitt, J. B., "Chemical Precipitation of
Phosphate within a High Rate Bio-Oxidation System," Proceedings 22nd
Annual Purdue Industrial Waste Conference, Lafayette, Indiana (1967).
(7) Nesbitt, J. B., "Phosphorus RemovalThe State of the Art," Journal
WPCF. 41:5:701 (1969).
(8) Schmid, L. A., "Look What $25 Can Buy," Water and Wastes Engineering,
Vol. 8, No. 9 (Sept. 1971).
(9) Pearse, et al., "Chemical Treatment of Sewage," Sewage Works Journal,
7_, No. 6, p. 997 (1935).
(10) Schmid, L. A., McKinney, Ross E., "Phosphate Removal by Lime Biological
Treatment Scheme." Journal WPCF, 41:7:1259 (1969).
(11) Zuckerman, M. M., and Molof, A. H., "High Quality Reuse Water by
Chemical-Physical Wastewater Treatment," Journal WPCF, 42:3:437 (1970).
(12) Flournoy, Ralph, "The Effect of Lime Treatment of Sewage on Oxygen
Uptake by Activated Sludge," M.S. Thesis, Kansas State University, 1970.
43
-------
APPENDIX
45
-------
TABLE A-l. Pilot
RAW SEWAGE
DATE
6/24
7/ 6
7/ 8
7/22
7/27
8/ 4
8/17
10/19
10/20
10/27
12/21
1/25
2/18
3/31
4/ 5
TIME
COMP**
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
1500
2100
0300
0900
COMP
1500
2100
0300
0900
COMP
FLOW
6500
10000
8000
8500
10750
11500
13000
14000
14000
14000
11500
10750
10000
8000
7000
4000
5000
6300
9000
6000
3000
5000
5500
COD
316
485
300
437
399
385
360
410
472
469
442
445
359
600
662
280
345
512
530
582
332
448
500
SS
254
260
300
151
178
179
215
200
195
205
150
192
90
190
240
70
140
174
250
270
130
270
244
P
9.6
14.2
11.5
12.4
14.1
11.0
14.4
14.8
13.4
12.7
12.6
13.2
10.0
10.0
8.6
7.5
7.0
8.5
10.8
6.9
6.6
12.2
9.6
Plant Operating Data for Selected*
PRIMARY EFFLUENT
COD
159
151
170
177
179
303
146
175
181
189
251
275
202
204
249
196
147
204
209
205
213
145
195
%R
50
69
43
60
55
21
59
57
62
60
43
38
44
66
62
31
59
60
61
65
36
68
61
SS
92
83
74
54
81
132
69
68
62
78
204
157
84
60
60
40
16
48
112
112
296
68
126
%R
64
68
75
64
55
26
68
66
68
51
20
76
68
75
43
88
73
55
59
--,
75
48
P
2.6
2.3
1.9
2.9
4.2
6.8
2.8
3.2
2.5
3.1
5.7
7.6
1.1
1.3
1.4
1.2
1.3
1.3
2.1
1.6
1.2
1.4
1.7
%R
72
84
82
77
70
38
81
78
81
76
55
42
89
86
83
83
81
84
81
76
82
89
82
PH
9.3
9.8
9.8
11.5
11.2
7.3
11.0
9.5
9.5
9.3
9.4
8.7
10.7
9.5
9.2
9.4
9.5
9.4
10.1
10.1
10.2
10.4
10.2
MLSS
2150
2700
2050
4025
1800
2750
2460
1590
1590
2590
1770
1800
1500
4000
4000
1660
1660
Runs
FINAL EFFLUENT
COD
21
15
10
40
30
63
35
32
32
22
28
20
40
82
61
53
61
66
82
45
37
49
59
%R
93
97
97
91
93
84
90
92
93
96
94
96
89
86
91
81
82
87
85
92
88
89
88
SS
20
11
8
4
16
44
34
4
4
4
18
15
--
36
2
10
6
16
24
16
18
18
20
P
2.6
1.8
2.0
2.9
3.0
12.2
1.7
2.5
2.1
2.7
3.5
6.0
1.3
2.8
2.0
2.3
2.1
2.3
1.6
1.3
1.5
1.4
1.4
%R pH
74 7.9
87 8.5
82 8.1
77 8.2
78 8.1
7.2
89 8.2
84 8.1
85
79 8.0
73 7.8
54 7.1
87 8.0
73 7.3
77 7.5
66 7.5
69 7.7
73 7.5
85 7.5
81 7.4
77 7.4
88 7.5
84 7.5
Selected* Runs when all systems were functioning properly
COMP** Composite value
-------
Table A-l Continued
RAW SEWAGE
PRIMARY EFFLUENT
FINAL EFFLUENT
DATE
V 7
4/19
5/ 4
5/ 6
5/10
5/13
TIME
1500
2100
0300
0900
COMP
1500
2100
0300
0900
COMP
1500
2100
0300
0900
COMP
1500
2100
0300
0900
COMP
1500
2100
0900
COMP
1500
2100
0900
COMP
FLOW
11000
10000
5000
5500
7940
10000
8500
5500
4000
7130
12000
12000
12000
12000
12000
10000
10000
10000
10000
10000
9000
10000
9000
9500
6000
7000
5000
6000
COD
508
533
196
234
422
708
745
284
348
585
545
510
247
454
439
705
655
357
451
542
450
506
354
437
628
408
398
478
SS
280
290
90
260
250
230
290
50
180
206
160
170
80
150
140
210
230
140
230
203
210
120
170
167
220
120
220
187
P
8.9
10.8
8.0
9.2
9.4
13.0
11.9
8.9
8.9
11.3
14.0
11.9
6.8
9.5
10.5
11.1
13.7
7.1
7.4
9.9
17.2
14.2
7.9
13.3
15.3
12.9
8.3
12.1
COD
189
196
171
133
179
228
252
208
124
216
235
255
163
171
206
192
212
98
114
154
309
225
96
209
189
198
115
167
%R
63
63
13
43
58
68
66
27
64
63
57
50
34
62
53
73
68
73
74
72
31
56
73
52
70
51
71
65
SS
124
128
96
68
111
60
124
76
20
77
64
32
8
28
33
64
40
12
28
36
136
88
32
85
44
68
20
44
%R
56
56
74
56
74
57
90
63
60
81
90
81
76
70
83
92
88
82
35
27
81
49
80
43
91
77
P
4.0
5.4
3.6
3.5
4.3
2.2
1.6
2.2
2.7
2.1
2.9
3.1
1.9
1.7
2.4
1.1
0.8
0.3
0.9
0.8
6.9
1.9
0.4
3.1
1.6
1.3
0.5
1.1
%R
55
50
49
61
54
83
86
75
70
82
89
74
72
82
77
91
94
95
87
92
40
87
96
77
89
90
95
91
PH
9.8
9.7
9.8
9.9
9.8
9.9
9.8
10.7
11.1
10.1
9.0
9.1
9.1
9.0
9.1
9.4
9.4
9.5
9.9
9.5
9.4
10.2
10.8
10.1
10.0
10.6
10.8
10.5
MLSS
1790
1790
2425
2425
1030
1030
1075
1075
1220
1220
1210
1210
COD
15
41
15
48
29
80
40
32
40
53
72
56
64
60
63
43
24
51
67
46
73
45
45
54
52
42
63
52
%R
97
92
92
80
93
89
46
89
89
91
87
89
74
87
86
96
94
86
85
92
84
99
87
88
92
90
84
89
SS
18
50
20
62
36
14
4
4
16
9
34
20
4
14
18
4
4
8
66
20
4
4
16
8
12
8
2
7
P
3.1
2.9
3.0
3.4
3.1
1.3
2.2
2.6
1.2
1.8
3.5
3.3
2.1
1.7
2.7
0.7
0.8
0.5
1.3
0.9
1.7
2.1
1.0
1.6
0.9
1.1
0.9
1,0
%R pH
65 7.5
73 7.4
63 7.4
63 7.4
67 7.4
90 8.3
82 8.1
71 7.9
87 7.9
84 8.1
75 7.2
72 7.1
68 7.2
82 7.4
75 7.2
96 7.2
94 7.2
98 7.4
82 7.9
91 7.4
90 7.8
86 7.6
88 7.8
88 7.7
94 7.9
92 7.8
89 8.2
92 7.9
-------
Table A-l Continued
00
RAW SEWAGE
DATE
5/18
5/24
5/27
6/ 7
TIME
1500
2100
0300
0900
COMP
1500
2100
0900
COMP
1500
2100
0300
0900
COMP
1500
2100
0300
0900
COMP
FLOW
7000
6000
6000
5000
6000
5000
5000
5000
5000
8000
8000
8000
8000
8000
12000
10000
6000
6000
8500
COD
437
346
193
447
348
831
508
438
592
512
351
351
399
403
758
534
273
146
428
SS
310
110
70
160
162
360
130
60
183
260
140
130
190
180
490
310
310
200
328
P
12.5
11.1
5.8
7.7
9.2
10.6
9.0
6.6
8.8
8.5
7.7
6.7
7.1
7.5
12.0
8.4
6.8
3.7
7.6
COD
156
183
140
146
156
269
223
169
220
179
179
116
160
153
138
111
100
125
PRIMARY EFFLUENT
%R
64
47
28
77
55
67
56
62
63
65
49
71
60
80
74
59
32
71
SS
56
68
60
48
58
68
40
14
41
104
176
48
44
93
60
44
20
41
%R
82
38
24
70
64
81
69
77
78
49
63
77
63
87
86
36
88
P
1.5
1.3
1.3
2.3
1.6
1.3
0.9
0.3
0.8
1.4
1.5
1.0
1.0
1.2
1.1
1.2
0.8
0.2
0.8
%R
88
88
78
70
83
88
90
96
91
83
81
86
86
84
91
85
72
97
90
PH
9. 8
10.0
9.6
9.5
9.7
9.7
10.5
10.8
10.3
9.7
9.7
9.7
9.6
9.7
10.6
10.3
10.8
11.0
10.6
MLSS
1590
1590
2010
2010
1870
1870
2775
2775
COD
30
33
30
37
32
62
154
131
116
37
37
45
45
41
15
15
8
35
18
FINAL EFFLUENT
%R
94
90
85
92
91
93
70
70
80
93
89
87
88
90
98
71
97
76
96
SS
12
4
14
10
10
20
10
10
13
8
6
4
8
7
5
5
4
5
5
P
1.0
0.9
1.9
1.0
1.2
0.9
0.9
0.8
0.9
1.7
1.6
1.5
1.5
1.6
0.9
0.9
1.1
1.0
1.0
%R
92
92
67
87
87
91
89
87
90
80
80
77
78
79
93
89
84
72
87
pH
7.7
7.9
8.1
8.0
7.9
7.7
7.9
7.9
7.8
7.2
7.4
7.4
7.5
7.4
7.1
7.3
7.4
7.4
7.3
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
2. Repeat No.
2.
3, A ccession No.
w
4. Title PILOT PLANT DEMONSTRATION OF A LIME-BIOLOGICAL
TREATMENT PHOSPHORUS REMOVAL METHOD
7. Authors)
Schmid, Lawrence
i. Report Date
6,
8. Performing Organization
Report No,
9, Organization
Kansas State University
Department of Civil Engineering
Manhattan, Kansas 66506
10. Project No.
17050 DCC
11. Contract/Grant No.
17050 DCC
u.
Type ef Reptat and
Period Coveted
12. Sponsoring Orgaommtioa
15. Supplementary Notes
Environmental Protection Agency report number,
EPA-R2-73-159, June 1973. ,
16. Abstract A 15,000 gpd pilot plant was constructed to demonstrate the capabilities of a
lime treatment process for phosphorus removal. The lime treatment of raw wastewater re-
moves the bulk of the phosphorus, and a subsequent biological process removes an addition*
increment of phosphorus via cell synthesis.
The pilot plant used for the study was a package-type, prefabricated unit. Addition*
small tanks were provided for sludge storage and measurement. A variable speed motor was
mated to a program controller to duplicate a diurnal flow pattern through the pilot plant
Lime was introduced into the raw wastewater in response to an automated pH control system
Instrumented systems were also developed for turbidity and ortho-phosphate concentrations
in the plant effluent. The raw wastewater entering the pilot plant was obtained from
the wet well of the Manhattan, Kansas Municipal Plant.
Lime treatment of raw wastewater to a pH of 9.5 to 10.0 resulted in phosphorus re-
moval of about 80 percent in the primary clarifier. The removal of BOD at this pH level
was about 60 percent as compared to 35 percent normally expected in primary settling.
The biological activated sludge process following the lime treatment neutralized the
primary effluent due to the microbial carbon dioxide production. Aerator pH's were in
the neighborhood of 8.0.
Lime treatment prior to biological treatment resulted in a change of the
settleability of the mixed liquor. Rather high sludge volume indexes were routinely
encountered. The process is capable of 90 percent phosphorus removal; however, con-
17a. Descriptors
siscentremoval is only assured at cue ou percent level.
*Nutrient Removal, *Chemical Coagulation, #Biologiical Treatment, Sludge, Municipal
Wastewater, Biological Oxygen Demand
17b. Identifiers
^Instrumentation, Diurnal Flow, Turbidity, Program Controller, Respiration Rate*;
Phosphate Analyses
17c. COWRRField & Group
IS. A vailability
19.
Security Class
(Report)
Security Class
(Page)
21.
Ho. of
Pages
Send To:
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 2O24O
Abstractor E. F. Earth
Institution NERC-Cincinnati, Ohio
WRSIC 1O2 (REV. JUNE 197!)
------- |