EPA-600/D-84-022
1983
IMPLEMENTATION OF SEQUENCING BATCH REACTORS
FOR MUNICIPAL TREATMENT
by
E. F. Earth
Wastewater Research Division
Municipal Environmental Research Laboratory
Office of Research, and Bevelopment
U'.S. Environmental Protection Agency
Cincinnati,, Ohio 45268 .
for
6th Symposium on Wastewater Treatment
Sponsored byt
Environment Canada
Environmental Protection Programs'Directorate
Ottawa, Canada
Conducted at;
Meridien Hotel, Montreal, Canada
November 16-17, 1983
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
.OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, Ohio 45268
-------
NOTICE
'This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
11
-------
IMPLEMENTATION OF SEQUENCING BATCH REACTORS
FOR MUNICIPAL WASTEWATER TREATMENT
E, P. Barth
Wasewater Research Division
Municipal Environmental Research Laboratory
U.S. Environmental Protection Ageacy
Office of Research and Development
Cincinnati, Ohio 45268
ABSTRACT
Sequencing batch reactor (SBR) technology is beiag implemented at vari-
ous municipal sites in both the United States and abroad. Total life-cycle
cost savings, ease of operation and reliability favor this technology at
facilities sized up to 19,000 m^ per day" (5 mgd). Batch treatment has in-
herent advantages over continuous flow processes in many applications.
Current research investigations on SBR's concern controlling sludge
settleability, nitrification-denitrifIcatlon, and biological phosphorus re-
moval .
This paper has been reviewed in accordance with the U.S. Environmental
Protection Agency's peer and administrative review policies and approved for
presentation and publication.
INTRODUCTION
This paper discusses progress of full-scale implementation and labora-
tory research on batch reactors. In the United States and several other
countries sequencing batch reactors (SBR) are being designed and constructed
for municipal facilities to meet a variety of effluent requirements., Studies
are being conducted to seek operational patterns to achieve nitrification,
denitriflcation, and biological phosphorus removal.
Historical technology ia replete with examples of batch treatment of
municipal wastewater. Sidwiek and Murray (1) have outlined the evolution of
batch processes into continuous flow processes In England. Early, in land
treatment practices, it waa recognized that intermittent irrigation of
wastewater was necessary for reaeration, of the soil to occur. In the year
1897, ways to conserve land were investigated. Vessels were constructed
which contained various fine media (3mm) to try to duplicate soil surfaces.
The first of these were operated as simple FILL and DRAW in accordance with
intermittent land application experience. Later,, a cyclic process- was*
instituted which consisted of 2-hour FILL, 1-hour stand full and 5-hour
drain. The cycle waa repeated twice daily. Seventy-eight percent removal
of organic matter was reported. The need to supply oxygen in the proper
amount for active biological oxidation was the key finding of this early
work. From this point on, two divergent developments led co continuous flow
processes. One utilized larger media with greater void space so that waste-
water splashing from surface to surface entrained oxygen from the surrounding
air. This, of course, is the now familiar biological trickling filter. The
other retained the concept of a flooded vessel containing media, but institu-
ted continuous flow with suitable valving and supplied air by an external
blower. This technology was the progenitor of the Hays Process (2) and
-------
Imhoff's contact aerators (3). Conceptually, Imhoff discussed a variation
of contact aerators which described rotating biological contactors before
their actual invention.
The precursor to the various, now familiar, continuous flow activated
sludge processes was actually a FILL and DRAW system operated as a batch
process. Arden and Lockett (4) in 1914 were among the first to show the
benefit of retaining substrate adapted organisms for efficient treatment.
Working with 2.3 liter flasks containing Manchester, England, raw wastewater,
they showed that the batch aeration period to achieve nitrification could be
reduced from 5 weeks to 9 hours if the sludge that accumulated from each
batch were retained in the flask after decanting the nitrified liquid. They
coined the term "activated sludge" to describe the resultant biological mass.
In later full-scale studies these same investigations (5) showed efficient
treatment with an 8-hour batch cycle, as shown in Table 1.
TABLE 1. BATCH TREATMENT OF MANCHESTER WASTEWATER (modified from Ref. 5)
Year 1915
Analytical Results, mg/L
Saw Effluent
KMn04 (Oxygen
Absorption)
NH4-N
Orgaaie-N
N02-N
NQ3-N
124 18
37
12 1.9
1.4
Operational Cycle, hours
Filling,, 1
Aeration, 4
Settlement , 1
Discharge , 1
There were 3 cycles each day
and 20 percent activated
sludge, by volume, was
14 | retained in the reactor
after each cycle.
PMSSNT DAY BATCH PROCESSES
Many present day industrial processes are batch processes; such as the
dairy, steel, pharmaceutical and antibiotic, fine chemical, cosmetic, and
paint industries.
In some instances, state regulatory agencies require that toxic wastes,
such as cyanide, be detoxified by batch operation to insure the effluent can
be monitored before discharge.
Most modern municipal treatment facilities utilize batch treatment in
one guise or another, such as anaerobic digestion, aerobic digestion or
intermittent discharge lagoons. Tchobanoglous (6) calls attention to the
fact that the classical biological oxygen demand test (BOD) is a batch assays
however, the resultant data are used to design continuous flow processes.
Renewed interest in main stream batch treatment resulted' from a mentor-
ial article by Pasveer (7). He reported on the conversion of a continuously
operated oxidation ditch; treating the waste from Sancta Maria Hospital,
Noordwijkerhout, Holland, Into a discontinuous discharge process with
intermittent aeration. The operational change resulted in the control of a
-------
-3-
filamentous activated sludge, improved clarification and yielded 90 percent
denitrlfleation.
Gronszy (8) has applied this concept of single tank treatment systems
in Australia. Extended aeration processes are operated with continuous
Inflow but discontinuous discharge. The liquid level in the tank varies
and various type decant devices are used. Aeration is provided in a cyclic
manner to encourage nitrification and denitrification.
These two approaches provided a high degree of treatment due to the fact
that they are both lightly loaded systems. Since the volume of the tank was
large in comparison to the influent flow, the possibility of short circuit-
ing influent to effluent was minor.
Irvine, the doyen of current investigations, of batch process, has
suggested a uniform- terminology and united the batch concept with modern
control strategies (9).
BATCH COMPARED WITH CONTINUOUS FLOW
Most simply stated, the conceptual difference between batch and contin-
uous flow is that continuous flow processes have spaclally related unit
operations, whereas unit operations are timed sequentially in batch proc-
esses .
The generic batch process can have several modifications, such as single
reactor,, multiple reactors in series or parallel, and sequencing reactors;
all with variable liquid levels. The most simple of these would be a vari-
able volume single reactor as depicted in Figure 1, which would be suitable
for a rural or small industrial situation where no flow occurs for part of
the day. The figure shows the. different time dependent modes that would
occur In a complete cycle of a 1-tank reactor. The modes are labeled accor-
ding to Irvine's (9) recommendation. The. volume and cycle time are only
Illustrative. For any batch system, FILL and DRAW must occur in each cycle,,
but REACT, SETTLE, and IDLE could be eliminated depending on the objective of
the treatment. Alternatively, other modes could be inserted by varying time
and operational controls within a cycle. Systems of this nature have a great
deal of flexibility. Table 2 compares batch and continuous flow processes.
In several instances drastic differences can be noted and their impact on
design, operation and effluent quality are apparent.
TABLE 2, COMPARISON OF BATCH AND CONTINUOUS PROCESSES
Parameter Batch Continuous
Concept Time Sequence Special Sequence
Inflow Periodic Continuous
Discharge Periodic Continuous
Organic Load Cyclic Even (by convention)
Hydraulic Load Cyclic Even (by convention)
-------
-4-
WASTEWATEB
EMOFIli}FHISBMI!t
M
START FILL
1SHR
QJHR
DISCHARGE
IFRUENT
WASTE
SLUDGE
SEQUENCE OF EVENTS FOR
6 HOUR SBR CYCLE IELEVATION VIEW}
LEGEND:
tx AUTOMATIC VALVE 1
n n no iirrcaiui wni-r
UJ tin _, _ ilfTcnmL CUJCX I
ť UQUlfl LiVIL SiMSOB J
BY
BT
FIGURE 1. SEQUENCE OF EVENTS FOR 6 HR SBR CYCLE
(ELEVATION VIEW)
-------
-5-
TABLE 2. COMPARISON OF BATCH AND CONTINUOUS PROCESSES (cont'd)
Parameter Batch Contiguous
Aeration Intermittent Continuous
Mixed Liquor Always in Reactor, Recycles through
No Recycle Reactor and Clarifler
Clarification Quiescent Hydraulics Hydraulic Motion
Flow Pattern Perfect Plug Complete Mix or
Approaching Plug
Equalization Inherent None
Flexibility Considerable Limited
Hydraulic Sizing Variable Uniform
Since the discharge is periodic, the possibility exists, within con-
straints, to hold effluent until some predetermined specific residual is
obtained.
The cyclic organic and hydraulic loading coupled with inherent equali-
zation allow control to be, exerted over substrate tension in the reactor by
suitably matching oxygen supply.
No recycle pumping is required, since the mixed liquor is always in the
reactor,,'thereby saving on energy. All the active microbial mass is avail-
able and not shunted through an inactive period in a clarifier.
Liquid solids separation occurs under near ideal quiescent conditions.
During the settling period there is no hydraulic motion due to inflow, out-
flow or recycle, as in continuous flow systems. The effluent exit piping
must be sized larger than Influent piping because flow accumulated over a
long time span is discharged in a much shorter time span.
In situations where the influent flow is continuous, two or more tanks
operated" in sequence would be needed to treat the flow. A two-tank sequen-
cing batch reactor (SB!) schematic is shown In Figure 2. As one tank progres-
ses from the FILL to the REACT mode, flow is switched to the second tank,
which starts a new cycle in the Fill mode. When capital costs- and operation-
al controls are considered, Irvine and llchter (10) indicate a three-tank
system Is near optimum for flows above 3,785 m^d (1 mgd).
Because sequencing batch reactor technology is new to both designers and
treatment plant operators, the major applications are currently for facili-
ties of 19,000 m^ per day (5 mgd), or less. This lack of experience on large
size facilities is no barrier to implementation. Between the years 1983 and
2000, it is anticipated that 3,000 municipal facilities with design flows
less than 7,500 m^ per day (2 mgd) will need to be constructed in the United
States. About 2.5 billion dollars will be expended on these sized treatment
plants.
-------
-6-
OPSRATIONAL CONSIDERATIONS
Since the majority of the SBR's being constructed are of small size a
recent discussion of energy utilization at municipal facilities by Smith is
of interest (11). This discussion shows that in the 3,785 ť3 per day (1 mgd)
size range energy consumption is only 8.1 percent of the operating cost and
labor accounts for 68.2 percent of the cost. Therefore, the SBR process
with automated valving, and a process controller which eliminates major la-
bor times, is a logical selection at small facilities. In Australia, where
modifications of the SBR have been employed for several years, it is common
to have one operator manage four facilities in a several hundred square
kilometer area. Also, designers are beginning to recognize the attributes
of the SBR in many situations. Barth has noted flexibility, equalization,
quiescent settling, elimination of recycle pump, and process reliability,
as criteria that direct selection of batch systems for many installations
(12).
Bathija has described various SBR tank designs and operational schemes
(13). Racetrack, rectangular (length to width ~ 4:1), and circular SBR
facilities have been constructed. Recommended side-water depth is 3 to 4.5
m (10 to 15 ft). If both aeration and mixing are independently necessary,
a jet type aeration device is preferred. If separate mixing only is not a
process requirement any non-clogging or mechanical aeration system can be
specified.
Bathija also suggests SBR solutions to two common problem areas in
municipal treatmentj reduced hydraulic flow during the ear.ly design life of
a facility, and Increased hydraulic flow during storm events. During the
early design life period of reduced flow, liquid level sensors can be set at
a fraction of tank capacity. In this way, the length of treatment cycles is
the same as design, and power is not wasted by over-aeration during an exces-
sively long cycle time. There are several approaches to manage increased
hydraulic flow during storm flows. If level sensors are left in their usual
position, increased flows will cause the number of cycles per day to increase.
If this should cause a reduction in efficiency the level sensors could be
raised, within free-board limits, to keep the cycle time near design
requirements. Either of these two management approaches could be coupled
with modification of the time periods within a cycle. For instance, the
time devoted to IDLE could be eliminated and that time period shifted to
increase REACT time, yet not alter the overall design cycle time.
The SBR process has an outstanding reliability feature during Increased
hydraulic flows. The mixed liquor solids always remain in the reactor and
cannot be washed out by hydraulic surges. Also, there is no recycle flow
from a separate clarifier and, therefore, an SBR process does not rely on
return sludge capability to maintain an adequate level of mixed liquor sol-
ids during increased hydraulic flow.
EQUIPMENT SELECTION
Because the SBR process is rather new, several types of instrumenta-
tion are being pursued. Some designs are employing photoelectric level
sensors, others use float level switches for controlling reactor liquid
level. Automated valves for control of reactor influent and effluent range
from pneumatic devices to motorized gate valves.
-------
-7-
PLAN VIEW OF SBR
TANK31
EFFLUENT
VALVE I
CONTROLS |
LEVEL
PROCESS ,
CONTROLLER
VALVE I
CONTROLS |
1
AIR AND |
PUMP CONTROL
(EACH A I
JET} U
FLOATING
WEIRS ^
I DIRECTIONAL
JET
1 AERATORS
I
INFLUENT
PRIMARY
TANK
AERATION
~ TANKS
ALTERNATING CYCLES
FILL AND DRAW
AIR
COMPRESSOR
(TO EACH
JET)
FIGURE 2. SEQUENCING BATCH REACTORS
-------
-8-
Effluent discharge devices can be submerged standpipes, tipping weirs,
float supported collector pipes, or submerged pumps.
In all but the smallest systems, commercial process controllers are used
to time and control the process sequence. In systems under 20 oH/day (5000
gpd) simple time clock and electrical solenoid valves would be recommended.
INDUSTRIAL APPLICATION
Dairy Waste
The Springfield Creamery in Springfield, Oregon, produces a waste that
has a large concentration of soluble oxygen demand, intermittent flow, and
extreme pH fluctuation. Flow is 40 m^/day (10,000 gpd). An adjoining meat
packing plant has a large concentration of suspended oxygen demand, intermit-
tent flow, and stable buffered pH of about 7. Flow is 8 m^ per day (2000
gpd).
Upon learning of the inherent equalization of an SBR process, ease of
operation, wide flexibility of treatment options, and simple design
considerations, the two companies' plan to mix these waste streams for com-
bined treatment. The SBR. will be an earthen basin with dimensions of 7.5 m
x 4.5 m x 2.4 m (25 ft x 15 ft x 8 ft). Aeration will be provided by a
submerged pump with an air blower exhaust line tapped into the intake side
of the pump. Since the flows are intermittent, a single tank system will be
adequate.
Chemical Waste
Due to the facts that SBR's can be operated to produce a defined resid-
ual concentration of a pollutant with no possibility of short circuiting, and
that quiescent settling produces very clear effluent, the process has been
selected to treat a combination of landfill leachate and chemical manufactur-
ing wastes in Niagara, New York.
The plant wastes are generated from the industrial operations and the
leachate is collected from a remote landfill area via collection underdrains
and hauled by truck to the plant site. Both type wastes contain several
hazardous organic pollutants. Dally design flow is 600 m^.
The exsting treatment at the industrial site consits of a storage tank,
neutralization tank and carbon absorption columns; with discharge of the
column effluent to the city sewer. Operation has shown that with this
physical/chemical process, excessive soluble organic matter quickly satur-
ates the absorption capacity of the carbon columns requiring frequent thermal
regeneration of the carbon. Upgrading construction will add two biological
SBR's, to reduce the soluble organics, and provision for adding ammonia and
phosphate to serve as nutrients for the SBR biomass.
Many of the organic components of the mixed waste have slow rates of
biodegradation. To insure efficient degradation, the SBR contents will be
monitored for total organic carbon during REACT until a predetermined
concentration is achieved. Thus, the SBR will decrease the organic loading
to the carbon columns and efficient settling will not overly burden the
columns with suspended solids. Waste activated sludge from the SBR's will be
incinerated to destroy any sorbed hazardous pollutants on the sludge surface.
Pilot SBR operation at a scale of 50 m^/day (13,000 gpd) has shown a 90
percent reduction of soluble organic carbon from the influent mixed waste.
It is therefore anticipated that the useful life of the carbon, before
regeneration is necessary, will be increased by almost a factor of 10.
Figure 3 is a schematic of the full-scale process as it will exist in 1985.
-------
Chemical
Plant Waste
300 mVd
300 mVd
Haulftd in
.
Storage
Blend
/*
)
"WŤ 3 Biological
P04 SBR's
- Lime
Mixer Air
t
C
1 /
^\ Ł ^
i
)
V J \. S *~
Leachate x*~~~r_-"/ ^^^^
Sludge to Landfill
t-
E3
Air 1
^L_f^-
gfcg 1
ť **
-*
rS
o
O"
o
3
Y^
:x
o
c"
3
3
(A
V
To City
WAS to Sewer
Incineration
FIGURE 3. SBR-CARBOM ADSORPTION FOR INDUSTRIAL WASTE
-------
-10-
MUNICIPAL APPLICATION
Recognition of the virtues of SBR's and the emergence of reliable pro-
cess control hardware lead to the USSPA to fund a development project, conduc-
ted by The University of Notre Dane, to re-evaluate municipal batch treatment.
The project involved the conversion of the existing 1,500 m^/day (0.4 mgd)
Culver, Indiana, continuous flow activated sludge facility into a two-tank
SBR (14). The new flow scheme is the same as shown in Figure 2.
Primary effluent is directed to each tank on an alternate basis as
dictated by liquid level sensors in each tank. Pneumatic compression pinch
valves control tank switching. Aeration and mixing are provided, by direction-
al jet aerators. Effluent is decanted during DRAW by submersible pumpa
attached to flotation devices which are swivel mounted to allow changes in
elevation of the pump suction.. All of these functions are programmed into a
process controller which operates and monitors all aspects of treatment.
The controller programs for both Tank-1 and Tank-2 operate independently
in order to have greater mode flexibility within a cycle. However, a main
program monitors the Tank-1 and Tank-2 programs on a time-sharing basis to
insure correct sequencing.
The initial evaluation at the Culver facility used an anoxie period for
part of the FILL mode and this was programmed into the controller as FILL-1
and FILL-2. During FILL-1, only the mixing action of the jet pumps was
activated. During FILL-2, the jet pumps mixed and aearated simultaneously.
A typical cycle for the two tanks was as follows!
flours in Mode
Tank-1
Tank-2
Flll-1
1.1
1.2
Fill-2
1.8
1.9
React
0.8
0.6
Settle
0.8
0.8
Draw
0.7
0.6
Idle
0.8
0.9
The difference in times between the modes of the two tanks is due to
flow variation, since the system is controlled by liquid level sensors. At
the design flow of 1,500 m-^/d (0.4 mgd) each cycle is completed in 6 hours
and both tanks complete 4 cycles each day.
The process conditions at Culver were:
F/M 0.2 kg BQDs/kg MLSS/d
SRT 20 days
MLSS 2,200 mg/L (70 percent volatile)
SVI 110 ml/g
One year of data and operational experience has been documented for BODj
and SS control. Table 3 provides monthly average wastewater and SBR effluent
constituents for 8 months of 1980. The City of Culver has; an effluent
phosphorus limitation of 1 mg/L. To achieve removal of phosphorus, ferric
chloride is dosed directly into the primary effluent channel leading to Tank-
1 and Tank-2.
Very high efficiency for BOD5, SS and TP is achieved at Culver with the
SBR process. Nitrate nitrogen of about 2 to 5 rag/1 is usually present in the
-------
-11-
Culver raw wastewater due to run-off from fertilized farm land Infiltrating
sections of the collection system. Table 3 shows that this influent nitrate
is denitrified during the anoxic FILL-1 mode of the SEE..
TABLE 3. CONSTITUENTS OF CULVER, INDIANA PROCESS STREAMS
MAY - DECEMBER, 1980
Milligrams per Liter
Location
Raw Wastewater
Primary Effluent
SBR Tank-1
SBR Tank-2
BOD 5
173
132
8
9
SS
138
81
7
9
TP
6.3
5.2
0.4
0.4
NH4-N
20
19
19
18
Oxidized
Nitrogen
2.8
2.6
0.4.
0.4
Overall Removal,
Percent
95
94
94
86
Process conditions to achieve nitrification and denltrlfication were
implemented during a second study period. The anoxic period for FILL-1 was,
reduced to provide a longer period of "dissolved oxygen supply for nitrifica-
tion. The efficiency of the SBR process during the last five months of 1981
is given ia Table 4.
Nitrification and denltrification occurred almost simultaneously, as
evidenced by the fact that neither ammonium or nitrate nitrogen reach high
concentrations in the reactors. Apparently as ammonium nitrogen is trans-
formed to nitrate, during the low load period at the start of FILL, there, is
subsequent denitrification due to the high mixed liquor concentration and
differential D.O.. content of the reactor. When mixed liquor is pumped
through the aerated jet nozzle a high 0,0., suitable for nitrification is
present; then as the mixed liquor circulates as a bulk liquid localized low
D.O. conditions can occur.
Once nitrification is established in the Culver SBR, ammonium.nitrogen
concentration in the reactor can never reach influent levels due to the
dilution of influent by the residual reactor contents left after IDLE, and
the biological transformation to nitrate. Accordingly, the nitrate concen-
tration cannot reach high levels, and even a low kinetic rate for denitrifi-
cation would yield fairly efficient nitrogen removal.
-------
-12-
TABLE 4. EFFICIENCY OF CULVER SBR FOR NITRIFICATION
AND DEVITRIFICATION
AUGUST - DECEMBER, 1981
Milligrams per Liter
Location
Raw Wastewater
Primary Effluent
SBR Tank-1
SB! Tank-2
BOD 5
158-
110
10
9
SS T?
150 7
82 5
5 0.7
6 0..8
NI4-N
20
18
1
1
Oxidized
Nitrogen
2
2
1
1
Overall Removal, 94 96 89 95
Percent
LABORATORY RESEARCH ON SBR's
Controlof Sludge Bulking
Working with 4 L SBR's, fed a soluble synthetic wastewater, Chlesa has
correlated sludge volume index (SVI) with aeration time during the FILL peri-
od (.15). He used the following strategies to manage the substrate tension
in the SBR's:
Control
Strategy
A
B
C
D
E
F
Unaerated /Unmixed
FILL, hr
2.00
1.75
1,50
1.25-
1,00
0.00
Aerated
FILL, hr '
0.00
0.25
0.50
0.75
1,00
2.00
Aerated
REACT, hr
4
4
4
4
4
4
SETTLE, DRAW
and IDLE, hr
2
2
2
2
2
2
Very spectacular results were obtained as shown in Figure 4. To con-
firm this observation, an SBR operated under control strategy F, with an SVI
of 600 mL/g, was switched to strategy A; and in 15 days the SVI decreased to
LOO mL/g.
These researchers explain these observations on the basis of competitive
growth of one group of organisms in opposition to other groups. They postu-
late that strategy A, where substrate tension surrounding the organisms
reaches a high level during FILL, and subsequent oxidation of organics and
endogenous metabolism during REACT, which places the organisms in periodic
starvation, selectively encourages the growth of floe forming organisms.
-------
-13-
20 30 40
Time, days
SO
FIGURE 4.. SVI AS A. FUNCTION OF CONT1DL STRATEGY
(FEOM SEP. 7)
-------
-14-
Fllamentous organisms survive best when the substrate tension is constantly
maintained at a low level such as in strategy F.
The sludge developed under strategy A had a greater specific oxygen
uptake rate than did the F strategy sludge; and the A sludge was much more
efficient in substrate removal rate than the F sludge.
The authors point out the great flexibility of the SBR process to
control substrate tension at any desired level. It can be noted in the above
strategies that A would be equivalent to contact stabilization and F would be
similar to a completely mixed activated sludge process.
Biological PhosphorusRemoval in an SBR
Discussions at the June 1982 "Workshop on Biological Phosphorus Removal"
showed a similar process feature for three proprietary phosphorus removal
systems (16). This feature is shown conceptually in Figure 5. When activar
ted sludge mixed liquor is subjected to an anaerobic or anoxic environment*
with influent organic substrate, there is coincident leaching of phosphorus
from the cellular material and sorbtion of organic matter. When environmen-
tal conditions are changed to an aerobic state,, phosphorus is removed from
solution and organics are oxidized.
The following several theories have been proposed to explain this find-
ing:
a) Anaerobic-aerobic staging can result in selection of a biological
population that is capable of storing polyphosphates. The actual mechanism
for storage and release and the basis of population selection is not known.
b) The regulation of the adenosiae dlphosphate ratio to adenosine
trlphosphate (ADP/ATP) controlled by the reversible enzyme polyphosphate
kinase has been suggested as a mechanistic factor. When cells are placed in
a "stress condition" such as lack of oxygen during an anoxic period, large
amounts of polyphosphates are then stored in the cells, when subsequent aero-
bic conditions are provided and cellular growth is rapid.
c) Another theory suggests that polyphosphate storage ability of certain
bacteria allow them to compete and survive In an alternating anoxic-aerobic
environment. The energy derived by hydrolysis of polyphosphate (ATP) may be
used directly for transport of substrate across the cell membrane in the
anoxic condition.
d) Polyhydroxy butric acid (PHB) has also been suggested as an important
consideration. It has been noted that PUB is formed during anoxic conditions,
and organisms such as Acineto_bacter_ can accumulate both PHB and phosphates.
These various mechanistic theories seem to show that some type condition-
ing occurs during the anoxic period; whether enzymatic hydrolysis, uptake of
a metabolite or environmental stress that confers a competitive advantage to
certain phosphate storing organisms when conditions are switched to an aero-
bic, state.
*The term anaerobic refers to environments which have no measurable concen-
trations of either oxygen or oxidized nitrogen. Often times appreciable
concentrations of carbonaceous materials must be present to maintain this
condition. Anoxic refers to environments which have no dissolved oxygen but
have oxidized nitrogen present. However, these'two terms were used almost
Interchangeably in the workshop discussions (16).
-------
, -15-
Anaerobic Aerobic
Environmental Conditions
5v BIOLOGICAL TOOSflOEDS AND BODg DDE
TO AHAIEOBIC-MIOBIC COH1ACT11SG
-------
The PhoStrip process utilizes this anaerobic or anoxic leaching to
remove phosphorus from a portion of mixed liquor in a side stream stripper
tank. Both the A/0** and Bardenpho*** processes remove phosphorus by baf-
fling the biological reactor to maintain the mainstream flow environmental
conditions suitable for each process' operation.
Also, all three processes were reported to be subject to inhibition of
phosphorus leaching during the anaerobic or anoxic phase if nitrate ion was
present. If nitrate was present, provision for denitrification before
phosphorus leaching had to be provided.
All of these applications for phosphorus removal are for continuous flow
systems. The Municipal Environmental Research Laboratory is employing an SBR
for studying biological phosphorus removal because the batch approach should
allow close control of environmental conditions, A schematic of the 4 L
bench-scale reactor used for these studies is shown in Figure 6. A vast
number of operational schemes can be employed uging a microprocessor control-
led SBR. In order to achieve the release and uptake of phosphorus noted in
Figure 5, the following cycle scheme-was employed:
Cycle1 Event
FILL*
Purpose
Add substrate
STRESS
4.5 hr
AERATE
WASTE SLUDGE
1 hr
4 sec
Cause phosphorus
leaching and
denitrification
Cause phosphorus
uptake, and
oxidize organics
Remove phosphorus
.from reactor,
and control SET
Mixer on
No aeration
(Time overlaps
with FILL
time) -
Aeration
Aeration
SETTLE
DRAW
IDLE
15 min
5 min
10 min
Form clear super-
natant
Discharge effluent
Adjustment of cycle
length
No aeration
Ho aeration
No aeration
Biospherics, Inc., 4928 Wyaconda Rd. , Rockville, MD 20852
**Air Products, Inc., Box 538, Allentowti, PA 18105
***EIMCQ-PMD, 3839 S. West Temple, Salt Lake City, UT 84115
^Primary effluent from Cincinnati's Muddy Creek treatment plant used.
-------
-17-
The total cycle time was 6 hours and therefore the SBR emptied and filled
4 times each day.
The results achieved are shown in Figure 7. During FILL the phosphorus
concentration in the reactor is affected by the phosphorus content at the end
of the last cycle event (either DRAW or IDLE), the concentration in the feed
entering the reactor and the leaching of phosphorus from mixed liquor solids.
In the case depicted, the residual phosphorus in the SBR was diluted as
primary effluent, with a smaller concentration entered, and a gradual increase
to 33 g/m^ occurred as mixed liquor cellular leaching occurred. During the
STRESS period a plateau concentration existed, until aeration was cottmenced,
and cellular uptake reduced the concentration to 2.5 g/m3. Effluent filtered
orthophosphorus concentrations approaching 0.0 g/m^ have occasionally been
noted..
Nitrate nitrogen remained at about 1 g/m^ during STRESS due to the
occurrence of denitrification. A dramatic increase of nitrate formation
occurs once aeration is commenced and dissolved oxygen increases.
The SBR was operated under this regime for 30 days (120 cycles) using
the Muddy Greek primary effluent. SBR effluent residual orthophosphorus was
routinely between 2.5 and 0.5 g/m^. Influent total phosphorus varied between
20 g/m^ and 15 g/m^, with orthophosphorus between 20 g/nH and 6 g/m3. Muddy
Creek wastewater is primarily domestic waste.
The process management to control biological phosphorus removal can, be
illustrated by referring to Figure 7. During FILL and STRESS, conditions
have to be provided to cause denitrification of any nitrate and insure that
leaching from the cellular mass occurs. STRESS periods ranging, from 2 to 6
hours have been studied. Three hours appears to be the minimum time that can
be employed. An aeration period of at least 1 hour is necessary to reincorpo-
rate the leached phosphorus and oxidize organlcs. Aeration times greater
than this lead to less efficiency due to lysis of cells during endogenous
respiration. Wasting- of sludge from the SBR must be done when the biological
solids have the highest content of phosphorus. This would be at the end of
AERATE, which yields a dilute waste activated sludge of the same solids
content as the mixed liquor. Or, wasting can be done near the end of SETTLE,
which provides a thickened waste activated sludge. Caution is required here
in case low dissolved oxygen in the sludge blanket causes leaching of phospho-
rus. Thus, overall efficiency of biological phosphorus removal dictates that
SETTLE time be as short as possible, yet provide good solids separation. A
minimum of 15 minutes and a maximum of 30 minutes has been satisfactory.
Time devoted to DRAW is not critical to biological phosphorus uptake
directly. Mainly, on full-scale SBR's this time period would be related to
effluent weir design to Insure proper hydraulics within the reactor, and
hydraulics .of discharge appurtenances. Provision for an IDLE period is
optional in SBR operation for biological phosphorus removal. A short IDLE
period, either quiescent or aerated would have no impact on overall efficien-
cy. IDLE is simply a convenience of adjusting cycles between reactors if two
SBR's are operated alternately. The same SBR process cycle described above
was employed with wastewater from Cincinnati's Little Miami facility. This
waste contains a major fraction of wastes from a variety of industrial
processes. Consistent biological phosphorus removal could not be achieved.
Over a three month period variations to FILL, STRESS, and AERATE were attempted
-------
-18-
<0
If
Jl'
n-
f6T~\_!
~F k
!*
i j
L
k-g==3
r-$^-n
/ V\ A
/ l\ \ -- \
n i /ť -\
i i ' ffi'n
j . -...-wy
7 * '" '" T/
\A/ J
,
I
s
^ T -. _u " r-gfn
gl
N
0
_ 1 _, Ť
II S a
.1 Ť TJ
It a 5
1 ť2
\ 5(e
1 |
[1
1 1
1
!
11.
7TTU
' U- S
to t
Łl 1
-Tl
03
CO
td
U-
Cfl'
1
CO
a
M'
&U
a
S-
-------
SEQUENCING BATCH REACTOR BIOLOGICAL PHOSPHORUS
REMOVAL
5.0-
/Ť.i
/m3
2.5-
10-
0/m3
5-
30-
g/m3
15-
- NOTES
pH RANGE 6.8
TEMPERATUR
DISSOLVED OXYGEN IN SBR MIXED LIQUOR MLSS 3000
MLSS 5.9%
i T y T T J T 7 f f f | T
N03 N IN FILTRATE FROM SBR
.___
ť i ] i- i j i
QRTHO PHOSPHOROUS IN FILTRATE FROM SBR
,'*"^ _ + ,.--Ť-_-^ -Ť^
,'' ^*~~
\ s
^ ^'
^
|_ ' ll ' ' 12 ' ' 13 ' ' 14 '
m
-------
-20-
without success. Efficient removal with residual orthophosphorus values of
0.5 to 1.0 g/w?, would be noted for several days; then deterioration of
efficiency would occur.
These SBE studies, on two wastewaters, have not as yet revealed the
controlling parameters to insure consistent, highly reliable biological
phosphorus removal. This same conclusion was noted for the three continuous
flow biological phosphorus removal processes dicussed at the Annapolis work-
shop (16).
Presently it would be recommended that any facility plan for biological
phopsphorus removal be guided by pilot plant studies prior to final selection.
Additionally, in those cases where a. high reliability of phosphorus removal
is necessary, it would be recommended that a chemical addition system be
installed as backup equipment.
REFERENCES
1. Sidwick, J. M., and Murray, J. E., "A Brief History of, Sewage Treatment
(Part 1)." Effluent and Water Treatment Jour., Feb., 65 (1976).
2. Hurley, J., "Contact Aeration for Sewage Treatment." The Surveyor, 10,
183 (1943).
3. Imhoff, K.., and Fair, G. M., Sewage treatment.. John Wiley and Sons-,
Inc., New York, N.Y. (1940).
4. Ardern, E., and Locketf, W. T., "Experiments on the Oxidation of Sewage
without the Aid of Filters." Jour. Soc. Chemical Ind., 33, 523 (1914).
5. Ardern, E., and Lockett, W. T., "The. Oxidation of Sewage without the- Aid
of Filters, Part III." Jour. Soc. Chemical Ind., 34. 937 (1915).
6. Tehobanoglous, G.., WaatewaterEngineering; Treatment Disposal Reuse.
McGraw-Hill, Inc., New ^ork, N.Y. (1979).
7. Pasveer, A., "A Case of Filamentous Activated Sludge." Jour. Water Poll.
Control Fed., 41_, 1340 (1969).
8. Goronszy, M. C., "Intermittent Operation of the Extended Aeration Process
for Small Systems." Jour. Water Poll. Control Fed., 51, 274 (1979).
9. Irvine, R. L., Sequencing Batch Reactors - An Overview," Jour. Water
Poll. Control Fed., 51^, 235 (1979).
10. Irvine, R. L., and Richter, R. 0., "Computer Simulation and Design of
Sequencing Batch Biological Reactors." 31st Industrial Haste Conference,
Purdue* University (1976-).
11. Smith, J. M., "Energy Recovery and Conservation for Low-Cost Systems."
Workshop on Low-Cost Treatment. Clemson University, April 1983.
12. Barth, E. F. (III)., "Sequencing Batch Reactors: Doing More with Less,"
Federal Water Quality Association Newslette, Vol. 10, No. 1, October
1982.
-------
-21-
13. Bathija, P, R,, "The Jet Age in Sequencing Batch Reactors," Seminar on
Emerging Technologies, Boston, Massachusetts, December 15, 1981.
14. Irvine, R. L., Ketchum, L. H., Breyfogle, R., and Barth, I. P., "Municipal
Application of Sequencing Batch Treatment." Jour. Water Poll. Control
Fed., 55, 484 (1983).
15. CMesa, S. C., and Irvine, R. L., "Growth and Control of Filamentous
Microbes in Activated Sludge." 55th Annual Conference Water Pollution:
Control federation, St. Louis, Missouri, October 1982.
16. Workshop on Biological Phosphorus Removal in Municipal Wastewater
Treatment. Summary report, Annapolis, Maryland, June 1982.
17. Hong, S., et al., "A Biological Wastewater Treatment System for Nutrient
Removal." Air Products and Chemicals, Inc., June 22, 1982.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing/
1. REPORT NO.
EPA-600/D-84-022
4. TITLE AND SUBTITLE
5. REPORT DATE
Implementation of Sequencing Batch Reactors
for Municipal Treatment
1983
6. PERFORMING ORGANIZATION CODE
7, AUTHOFMS)
Edwin F. Barth
8, PERFORMING ORGANIZATION REPORT NO.
9, PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Municipal Environmental Research Lab., WRD
Cincinnati, Ohio 45268
10. PROGRAM ELEMENT NO.
CA2B1B
11. CONTRACT/GRANT NO.
N/A
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Municipal Environmental Research Lab., WRD
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Conference Prpceedings
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
Paper presented at Sixth Conference on Wastewater Treatment, Montreal, Canada
-November 16-17, 1983, sponsored by Environment Canada .
16.' ABSTRACT
Sequencing batch reactor technology is being implemented at various municipal sites
in both the United States and abroad. Total life cycle cost savings, ea-se of operation,
and reliability favor this technology at facilities sized up to 19,000/m3} per day (5 mgd)
Batch treatment has inherent advantages over continuous processes in many
applications.
Current research investigations on sequencing batch reactors concern controlling
sludge settleability, nitrification, denitrification, and biological phosphorus removal,
17.
KEY WOflOS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN 6NDSD TERMS C. COSATi Field/Group
Batch Reactors
Municipal Wastewater
Secondary Treatment
Advanced Waste Treatment
Phosphorus Removal
Nitrogen Removal
Process Controller
Sludge Volume Index
13B
IS. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Reportl
Unclassified
20. SECURITY CLASS (Thispage}
Unclassified
21. NO. OF PAGES
.. - 24
22. PRICE
EPA Form 2220-1 13-73)
------- |