United States ' Municipal Environmental Research September 1980
Environmental Protection Laboratory
Agency Cincinnati OH 45268
Research and Development
International Seminar on
Control of Nutrients in
Municipal Wastewater
Effluents
OOOR80017
Proceedings
Volume II: Nitrogen
Hotel del Coronado
(San Diego)Coronado, Cahfornia92118
September 9, 10, and 11, 1980
-------
CONTROL OF NUTRIENTS
IN MUNICIPAL WASTEWATER EFFLUENTS
VOLUME II: NITROGEN
Proceedings of the International Seminar
San Diego, California
September 9-11, 1980
Seminar Convener:
E. F. Earth
Wastewater Research Division
Municipal Environmental Research Laboratory
Speakers:
Dr. W. Gujer, Dubendorf, Switzerland
Dr. C. Lue-Hing, Chicago, Illinois
Mr. E. R. Jones, Washington, D. C.
Mr. F. F. Sampayo, Toledo, Ohio
Mr. E. W. Knight, Chicago, Illinois
Dr. N. F. Matche, Vienna, Austria
Mr. D. E. Schwinn, Cazenovia, New York
Dr. H. J. Heimlich, Cincinnati, Ohio (Luncheon Address)
Municipal Environmental Research Laboratory
and
Center for Environmental Research Information
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
U.S. Environmental Protection Agency
Region V, Library
230 South Dearborn Street
Chicago, Illinois 60604
-------
AGENDA FOR
INTERNATIONAL SEMINAR
ON .
CONTROL OF NUTRIENTS IN WASTEWATER EFFLUENTS
SEPTEMBER 8, 1980
7:30 to 9:00 p.m. RECEPTION/EARLY REGISTRATION
VOLUME I Page
SEPTEMBER 9, 1980 PHOSPHORUS CONTROL TECHNOLOGY
7:30 to 9:00 a.m. REGISTRATION
9:00 to 9:15 WELCOME AND INTRODUCTION TO PROGRAM
Mr. Edwin Barth
Chief, Biological Treatment Section
Wastewater Treatment Division
U.S. EPA/MERL
9:15 to 10:05 NUTRIENT REMOVAL TECHNOLOGY - THE CANADIAN 1
CONNECTION
A presentation of the rationale for nutrient
control; the development of an R&D, legislative,
and technology transfer program; implementation
of low cost technology at existing municipal
plants; and impact and current status of control
technology.
Speaker: Dr. Norbert W. Schmidtke, Director
Wastewater Technology Centre
Environmental Protection Service
Environment Canada
Burlington, Ontario, Canada
10:05 to 10:20 COFFEE BREAK
10:20 to 11:10 PHOSPHORUS REMOVAL IN LOWER GREAT- LAKES MUNICIPAL 39
TREATMENT PLANTS
A survey of phosphorus removal processes of
various types, with statistical summary of lower
lakes facilities, histograms of performance and
loadings, and a discussion on phosphorus
availability in relation to treatment processes.
Speaker: Dr. Joseph DePinto
Department of Civil and Environmental
Engi neeri ng
Clarkson College, Potsdam, New York
iii
-------
VOLUME I (Continued) Page
11:10 to 12:00 EXPERIENCES AT GLADSTONE, MICHIGAN UTILIZING 91
ROTATING BIOLOGICAL CONTACTORS FOR BOD,
PHOSPHORUS AND AMMONIA CONTROL
A rotating biological contactor facility with
summary data on effluent residuals, key daily
operational points, actual cost data, and
recommendations on future facility design from an
operational standpoint.
Speaker: Mr. Willard Lee Morley, Superintendent
Water and Wastewater Treatment
City of Gladstone, Michigan
12:00 to 1:00 LUNCH
1:00 to 1:50 CONTROL TECHNOLOGY FOR NUTRIENTS IN MUNICIPAL 113
WASTEWATER TREATMENT IN SWEDEN
Necessity for nutrient control in Sweden,
techniques to translate basic nutrient research
into full-scale facilities, and extent of
implementation of nutrient control in municipal
facilities in Sweden.
Speaker: Dr. Bengt Gunnar Hultman
Swedish Water and Wastewater
Works Association
Stockholm, Sweden
1:50 to 2:40 RESEARCH ON PHOSPHORUS CONTROL IN JAPAN (separate
The type of research on phosphorus control being manuscript)
conducted in Japan, the reasons why phosphorus
control is necessary, and views of operating
facilities that utilize phosphorus removal
processes.
Speaker: Mr. T. Annaka
Department of Sewage and Serage
Purification
Ministry of Construction
Japan
2:40 to 3:30 ECONOMICAL AND EFFICIENT PHOSPHORUS REMOVAL AT A 139
DOMESTIC-INDUSTRIAL WASTEWATER PLANT
The combination of industrial and domestic waste
characteristics considered in the design of the
facility, a summary of several years of plant
efficiency, and the low-cost experience of
phosphorus control.
Speaker: Mrs. Doris Van Dam, Superintendent
Wastewater Treatment Plant
Grand Haven, Michigan
iv
-------
VOLUME I (Continued) Page
3:30 to 3:45 COFFEE BREAK
3:45 to 4:35 THE PHOSTRIP PROCESS FOR PHOSPHORUS REMOVAL 159
The PhoStrip process is discussed with emphasis
on efficiency, cost, and reliability in relation
to original design approaches.
Speaker: Mr. Carl J. Heim
Assistant Staff Engineer
Union Carbide Corporation
Linde Division
Tonawanda, New York
4:35 to 5:00 DISCUSSION ON PHOSPHORUS CONTROL TECHNOLOGY
-------
VOLUME II Page
SEPTEMBER 10, 1980 NITROGEN CONTROL TECHNOLOGY
8:00 to 8:50 EMERGING STRATEGY FOR NITROGEN CONTROL BASED ON 1
RECEIVING WATER QUALITY CONSIDERATIONS
Emerging nitrogen strategies and the need for
nitrification will be discussed, along with
research needs for nitrification to suit European
situations.
Speaker: Dr. Willi Gujer
Swiss Federal Institute for Water
Pollution Control
Dubendorf, Switzerland
8:50 to 9:40 FULL-SCALE CARBON OXIDATION/NITRIFICATION STUDIES 43
AT THE METROPOLITAN SANITARY DISTRICT OF GREATER
CHICAGO
Large-scale plant manipulations to accomplish
single-stage nitrification, with operational
control techniques related to nitrification
kinetics and to implications of control and costs
for a 1,300 MGD facility.
Speaker: Dr. Cecil Lue-Hing, Laboratory Director
Metropolitan Sanitary District of
Greater Chicago
Chicago, Illinois
9:40 to 10:30 PHOSPHORUS REMOVAL WITH IRON SALTS AT BLUE PLAINS 98
Data from the world's largest nutrient control
plant on mineral addition for phosphorus control.
Discussion of costs, alternate chemical
selection, and sludge production, plus what it
takes to put a plant of this size on-line.
Speaker: Mr. Ed Jones, Chief Process Engineer
Wastewater Treatment Plant
Washington, D.C.
10:30 to 10:45 COFFEE BREAK
10:45 to 11:45 NITRIFICATION AT LIMA, OHIO 129
Design of second-stage plastic media for
nitrification, summarizing several years of
efficiency data, operational control, and costs,
and relating these to design changes of second
generation designs.
Speaker: Mr. Felix Sampayo
Jones and Henry Engineers, Ltd.
Toledo, Ohio
VI
-------
VOLUME II (Continued) Page
11:45 to 1:30 LUNCH
Speaker: Dr. Henry Heimlich, Professor of
Advanced Clinical Studies
Xavier University
Cincinnati, Ohio
Author of the Heimlich Maneuver
1:30 to 2:20 OPERATING EXPERIENCE WITH A 30 MGD TWO-STAGE 153
BIOLOGICAL NITRIFICATION PLANT
A summary of efficiency data, control-loops,
operational modifications, and costs for the John
Eagan Plant.
Speaker: Mr. Earl W. Knight
Assistant Chief Engineer
Metropolitan Sanitary District
of Greater Chicago
Chicago, Illinois
2:20 to 3:10 NITRIFICATION-DENITRIFICATION IN FULL-SCALE 170
TREATMENT PLANTS IN AUSTRIA
Single stage nitrification/denitrification, plus
status of nitrification control in Austria and
the need for this technology.
Speaker: Dr. Norbert F. Matsche
Assistant Professor
Technical University, Vienna, Austria
3:10 to 3:25 COFFEE BREAK
3:25 to 4:15 SINGLE STAGE NITRIFICATION-DENITRIFICATION AT 194
OWEGO, NEW YORK
Second generation design for single-stage
nitrification/denitrifi cation systems, and a
real-world perspective on reliability, efficiency
demands, cost, and operation.
Speaker: Mr. Donald E. Schwinn, P.E.
Stearns and Wheler
Civil and Sanitary Engineers
Cazenovia, New York
4:15 to 5:00 DISCUSSION ON NITROGEN CONTROL TECHNOLOGY
vii
-------
VOLUME III Page
SEPTEMBER 11, 1980 COMBINED PHOSPHORUS AND NITROGEN CONTROL
TECHNOLOGY
8:15 to 9:05 DESIGN AND OPERATION OF NITROGEN CONTROL 1
FACILITIES AT TAMPA AND THE NSSD
Three-step nitrogen control at Tampa and two-step
nitrogen control at the North Shore Sanitary
District in Illinois. A summary of the design,
operation, and use of the unusual flexibility
buillt into these plants.
Speaker: Mr. Thomas E. Wilson
Principal Engineer
Greely and Hansen
Chicago, Illinois
9:05 to 9:55 PERFORMANCE OF FIRST U.S. FULL-SCALE BARDENPHO 34
FACILITY
A managed biological system for nitrogen and
phosphorus control.
Speaker: Dr. H. David Stensel, Manager
Sanitary Engineering Technology
Development, EIMCO PMD
Salt Lake City, Utah
9:55 to 10:10 COFFEE BREAK
10:10 to 11:00 DENITRIFICATION IN CONTINUOUS-FLOW SEQUENTIALLY 74
AERATED ACTIVATED SLUDGE SYSTEMS AND BATCH
PROCESSES
Present developments on batch systems controlled
by time-clocked valves and evolution into a
microprocessor-controlled municipal facility.
Speakers: Dr. Mervyn C. Goronszy
Senior Investigating Engineer
State Pollution Control Commission
.Sidney, Australia
and
Dr. Robert L. Irvine, P.E.
Deptartment of Civil Engineering
University of Notre Dame
Notre Dame, Indiana
viii
-------
VOLUME III (Continued) Page
11:00 to 12:00 NITROGEN AND PHOSPHORUS REDUCTION FROM LAND 118
APPLICATIONS AT THE DISNEY WORLD RESORT COMPLEX
Several approaches to attaining defined effluent
residuals and accumulating large amounts of
analytical data for this entertainment complex,
with data on phosphorus control in the activated
sludge system, overland flow, spray, and
perculation basins.
Speaker: Mr. Robert Kohl, Director
Reedy Creek Utilities Company, Inc.
Walt Disney World
Lake Buena Vista, Florida
12:00 to 1:00 LUNCH
1:00 to 1:50 EXPERIENCE WITH AMMONIA REMOVAL BY SELECTIVE ION 137
EXCHANGE AND CLOSED-CYCLE AIR STRIPPING
REGENERANT RENEWAL
A discussion of the Tahoe-Truckee Sanitation
Agency and the Upper Occoquan facility in
Virginia, covering a closed-cycle stripping
process in relation to efficiency and effluent
residuals, operational considerations, and cost
data.
Speaker: Mr. L. Gene Shur
Vice President and Director
CH2M-Hill Consultants
Corvallis, Oregon
1:50 to 2:40 NITRIFICATION AND PHOSPHORUS REMOVAL IN A 35 MGD 185
ADVANCED WASTE TREATMENT PLANT AT ROANOKE, VA
Design parameters related to operational results
for control of nitrification and phosphorus
residuals.
Speaker: Mr. Donald E. Eckmann
Alvord, Burdick, and Howson Engineers
Chicago, Illinois
and
Mr. Harold S. Zimmerman, Plant Manager
Waste Treatment Plant
Roanoke, Virginia
2:40 to 3:30 FULL-SCALE EXPERIENCE WITH TWO-STAGE 214
NITRIFICATION AND PHOSPHORUS REMOVAL
Accumulated efficiency and cost data from two
facilities, a summary of efficiency data, as
frequency distribution, overall costs, and
operational modifications necessary for enhanced
second generation design.
Speaker: Mr. WinfieldA. Peterson, Chief
Plant Operating Group N.E.
Metcalf and Eddy, Inc.
Boston, Massachusetts
-------
EMERGING STRATEGY FOR NITROGEN CONTROL BASED ON RECEIVING
WATER QUALITY CONSIDERATIONS
Dr. Willi Gujer
Head Engineering Science Department
Swiss Federal Institute for Water Resources
and Water Pollution Control, 8600 Diibendorf/
Switzerland
DEFINITION OF THE PROBLEM
INTRODUCTION
Until 1975 the goal of water pollution control in Switzerland
was to reduce the loads of biodegradable organic compounds dis-
charged into receiving waters. Based on this concept, secondary
treatment plants were designed to reduce BOD loads. Today 80%
of the population in Switzerland can potentially be connected
with existing treatment plants- in reality already 65% are
connected.
During the sixties, phosphorous removal in catchment areas of
lakes became required and today 283 of a total of 763 wastewa-
ter treatment plants make use of some form of phosphorous control.
In 1976 the new federal ordonnance for waste water discharge
was enacted. This ordonnance specifies numeric values for 52
chemical and physical parameters as required in treatment plant
effluents or as desireable (efforts shall be made towards meeting,..)
in receiving waters. In addition the ordonnance contains a
verbal description of the biological and aestetic state to be
-------
maintained in receiving waters. Switzerland now has a specific
goal for the state of receiving waters. This introduces the
necessity for regional considerations in the definition of
discharge standards.
In summary:
(a) Nutrient control becomes increasingly important, (b) treat-
ment plants exist to a large extent, they are designed however
for BOD removal only and (c) available funds are decreasing,
cost/benefit considerations therefore become increasingly re-
quired. In the future water pollution control must be based on
regional studies considering all sources of pollutant loads.
and indicating an optimal strategy to achieve desired goals.
This paper presents the results of a study for the definition
of a water pollution control strategy in an entire catchment
area. The study had pilot character in Switzerland, it consi-
dered a variety of pollutants (organic compounds, nutrients
such as P, NH , NO , NO and some heavy metals) and leads to-
wards the implementation of advanced waste water treatment
technology. Only the aspects of nitrogen control will be pre-
sented here. Proposed solutions may however only be fully ap-
preciated if other pollutants are considered.
NITROGEN STANDARDS IN SWITZERLAND
Considering fish toxicity, drinking water standards, dissolved
oxygen dynamics in rivers and groundwaters, available technolo-
gy etc., the federal ordonnance (1975) states tolerance levels
-------
for different forms of nitrogen in receiving waters and waste
water discharges as indicated in Table 1. Discharge standards
may be defined or lowered if substantiated by unsatisfactory
receiving water quality.
CHARACTERISTICS OF THE STUDY AREA (CATCHMENT AREA OF THE
RIVER GLATT)
The river Glatt originates at the effluent of the highly eutro-
phic Greifensee (lake) at 436 m above sealevel and discharges
after 36 km into the Rhine at 330 m above sealevel. Below the
2
Greifensee the river drains 260 km of densely populated areas
(>900 inhabitants/km ). 15 % of the area is developed, 52%
is agriculturally used and the remaining 33 % are mainly fo-
rests and swamps, as well as some undrained areas of the Zurich
airport and roads.
Table 2 indicates the statistic distribution of water flow in
the Glatt. Drinking water is imported into the area at a rate
of approx. 21-10 m /year. Rainfall adds up to 1000-1100 mm/
year, it is lowest during winter months. Groundwater infiltra-
tion from the Glatt river is observed in the lower part of the
area. Partially this groundwater is used for drinking water
supply.
Today 98 % of the population and industrial operations are con-
nected to 12 secondary treatment plants, most of them built
within the last 10 years. Fig. 1 gives an overview on the catch-
ment area and the location and size of treatment plants.
-------
Table 1: Tolerance levels for nitrogen based on Swiss Federal
Ordonnance for Waste Water Discharge (1975).
Form of Nitrogen
Ammonium NH . -N
4
Ammonia NH...-N
Nitrite NO -N
Nitrate NO -N
In running waters
not to be excee-
ded 347 days/year
0.5 mg/1 1)
0.08 mg/1
no toxicity 2)
5.8 mg/1
Not to be excee-
ded in waste
water discharges
variable
0.3 mg/1
1) may be exceeded if no drinking water is affected
(receiving water and ground water)
2) for rainbow trout (salmo gairdnerii) this value is
in the range of 0.05-0.1 mg/1 NO -N (Gujer 1978).
Table 2; Statistic distribution of flow rate of the river
Glatt (average year)
Fraction of time ,
the indicated flow-
rate is exceeded
5 %
50 %
95 %
97.5 %
Flowrates in River Glatt [m /sec]
Effluent of
Greifensee (lake)
2.4
3.3
7.8
9.5
Discharge into
Rhine
4.5
8.5
15.5
19.5
-------
LLJ
CL
SI °
°- o
O O LJLJ
^8 3 °-
LO ,
O • O
O
O
O
0 o o
o o o o
CN O o ^
v m CM ?
01
•H
E
_ac
o.
V
o-i
-------
STATE OF THE RECEIVING WATERS IN THE AREA
The state of the river Glatt and some of its tributaries is bad-
aesthetically, biologically and chemically - these rivers do
not fullfill legal requirements. The sediment is reducing as
indicated by ironsulfide precipitation on rocks and in sediment
beds. Protozoan colonies are abundant. Macrophyte growth is a
serious problem during summer months in many stretches of the
river. Fishing is not attractive and of low productivity in the
lower two thirds of the river, mainly because of adverse distri-
butions of fish populations.
Population growth and increased sewerage in developped areas
has increased the load of pollutants in the rivers inspite of
the construction of treatment plants. Fig. 2 indicates the ammo-
nium profile in the river Glatt as observed in the last 40
years.
The chemical characteristics of the river Glatt have been ob-
served recently (1973/74, Zobrist et a_l 1976) . The results of
this study, with regard to nitrogen compounds, are summarized
in Table 3. At the discharge into the Rhine nitrite and ammo-
nium exceed the limits as stated in table 1, ammonia (free NH_)
exceeds the limit in more than 10 % of the samples, nitrate
approaches the allowable concentration.
Groundwater, as produced by wells close to the river Glatt, have
nitrate concentrations close to those observed in the river
(Fig. 3). In bore holes far from the river this concentration
may however be greater.
6
-------
cn
ro
•P
-P
(0
rH
O
cu
>
•H
tf
CU
X!
-P
0
O.
U CU!
C -P
cu w
tP-H
O M
H XI
-P O
•H N
m ^
O r~
\
C ro
O r-
-H CTi
-P rH
4J
CU M
U CU
C cn
O XI
U O
00
cu
rH
X!
(0
QJ
ti
•H
K
0
-p
c
cu
&1
M
(d
X!
O
W
•rH
CU
CU
cn
G
cu
4H
•H
cu
£_l
o
m
o
-P
c
cu
rH
M_ i
1 1 [
c
o
1
~r-\
-P
X!
-H
}_l
•P
cn
-rl
Q
>i
U
fi
CU
J3
tr1
cu
>H
1
'C
tr
•r-
c
£*1
o
c
cu
d
CU
M
r_
MM
J^
0
T3
f^
d
o
p.
g
O .
U
cu
ft
_p
0\°
LD
(Ti
*o
O
ID
dP
i o
U )
c! *
o
•H
-P
tiC
m
c^
c)P
O
m
OV
i r^
LJ )
U
H
4-)
cn
H
i_l
CU
4->
U
(0
M
m
£
u
iJ 2 i-q 3 CQ S
r- oo ^r
oo r^ oo in • •
• • • • CTi CN
O ^ CN 00 rH rH
xf Vi) CN
rH •* n CN • •
• rH O
O n rH 00 rH H
LD P-
o o n r^ CN n
o CN o r°^ ^d1 ^o
" 1^ H f-4 t-M ^4
OO »* O CN
O VO 00 O • •
• • • • CN
•P -P -H r-l
•H (t) C O
SH M o ft cn
•P -P g g tn
•H -H g K CU -H
£2 »3 (^ ft EH ^
ITS
t)
i
-H
XI
g
M
o
rd
e
M
O
tr>
O
J
cu
o
•H
4-)
3
XI
•H
r4
-P
cn
•H
Q
-------
I
xf
CD
E
o
O
E
o
E
E
o>
CD
CO
k_
o>
2.0
1.0
0.8
0.6
0.4
0.2
0.1
0.08
0.06
0.04
0.02
1973/74
n = 95
1950/51
n=7
o>
•o
c=
1953
n = 35
•I'll • i •
0| 5 I I 15 I 25
km
O
:=3 Q_
MO
O>
•o
o>
o>
"P
o>
Js
CD
-------
60 i
50-
40
CO
O
CD
30-
20-
LU
O
O
O
10-
o:
k
X
« ,j \
_L\
•
Tolerance Level
> • »
\/v
V
/Drinking water
Tolerance Level Rivers
BH2
Observed Range in
River Glatt close to
Bore holes.
0
1959 61 63 65 67 69 71 73 75 77 79
YEAR
Figure 3
-------
In summary:
The state of the rivers in the study area does not fullfill
legal requirements. Improvements are necessary with regard to
many pollatants, the reduction of the load of several nitrogen
compounds is mandatory.
MODELLING THE DYNAMICS OF NITROGEN IN THE RIVER GLATT
In the river Glatt nitrification occurs throughout the year.
In the course of this porcess ammonium is converted to nitrite
and further to nitrate. Previous to this study a model for the
nitrification in rivers was developed (Gujer 1976, 1978) which
allows prediction of the amount of nitrification and the nit-
rite concentration without calibration. The model was verified
with data from the chemical survey of the river (Zobrist et al
1976). The results of this verification are indicated in Fig.4
for the observed ammonium load along the river and in Fig. 5
for the nitrite concentration as discharged into the river
Rhine. Conclusions from this modelling effort are:
1. The model is reliable and may be used to predict the
amount of ammonium converted to nitrate as well as
the nitrite concentration in the river.
2. The ammonium load discharged into the river Rhine is
significantly reduced by nitrification in the river
Glatt (Fig. 4).
3. The nitrite concentration in the river is controlled
by nitrification in the river itself. The main para-
meters that control nitrite in rivers are ammonium
concentration and temperature. Point sources for nitrite
(partially nitrifying activated sludge plants) may influ-
ence the nitrite concentrations in rivers for a few
kilometers along the river.
10
-------
o
CO
o
CM
3
CT>
•H
O
o
o
o
11V19
o
o
N -
Nl
11
-------
IT)
O
^/y H
#1
•o —\
?r
—
T- O
\Z°-
>-
/ -
CO
O
CM
O
CO CM
NOIlVdlN30NOO
0 CM 00
X
I—
o
O
3imilN
un
d)
cn
•H
12
-------
4. The successful reduction of nitrite in rivers below
0.1 mg NC>2-N/1 requires complete nitrification in
treatment plants to reduce nitrite discharge, and re-
sidual ammonium concentrations in the river below
0.2 mg NH4-N/1 in summer (20°C) and 0.4 mg NH4-N/1 in
winter (5oc).
5. Control of the nitrite concentration in the river im-
poses the most severe requirements for nitrogen con-
trol in waste water treatment plants.
NITROGEN TRANSFORMATIONS IN WASTE WATER TREATMENT PLANTS
Nitrogen Transformations in wastewater treatment plants are
complex. In domestic sewage, degradation of organic nitrogen
may approximately satisfy the metabolic needs for reduced nitro-
gen (mainly NH.) of the biomass. This discussion is therefore
limited to the processes nitrification (oxidation of NH. to
N0_ and N0_) and denitrification (reduction of NO., and N09 to
N2) .
In raw domestic sewage NH. loads are subject to extreme diur-
nal variations (Gujer and Erni 1978) which may result in ex-
treme variation of effluent NH. concentrations in nitrifying
wastewater treatment plants (Fig. 6). Today information on
activated sludge plants and to some extent also for tertiary
trickling filters allows the design of processes which do not
leak significant amounts of a NH. (Gujer 1977, Gujer and Erni
1978, USEPA 1975). Based on this information a design procedure
for activated sludge plants and tertiary trickling filters was
chosen which would limit peak and average NH. concentrations
to less than 2 and 1 mg N/l under winter conditions (10°C).
Nitrite discharge from activated sludge plants depends heavily
13
-------
lN3IYldd3
0)
sj
3
14
-------
on activated sludge growth rate (Fig. 7), hydraulic residence
time distribution and dissolved oxygen concentration (denitri-
fication) at the effluent of the aeration tank. Effluent nitrite
concentrations are significant in comparison to desired recei-
ving water concentrations, they are also subject to extreme
variations similar to NH. (Fig. 8). Today we lack reliable
methods to estimate effluent nitrite concentrations. For this
study the experimental results in Fig. 7 were used. This is
probably a conservative estimate since most activated sludge
plants in the area do not use single CSTR aeration tanks. The
utilized contacting patterns, theoretically, should yield
lower NO- concentrations.
Partial denitrification may be accomplished quite easily with
the aid of wastewater organics by contacting primary effluent,
activated sludge and nitrified effluent (recirculation) in an
anaerobic reactor. Dosing of external organic substrates (me-
thanol) was not considered in this study.
THE REGIONAL WATER POLLUTION CONTROL STUDY
SYSTEM ANALYSIS
This study was designed to discuss the different sources of
pollutants, the possibilities to reduce these pollutants and
the costs of these reductions and finally the manifestations
of the residual loads in the receiving waters. Dry weather as
well as wet weather situations had to be analysed.
15
-------
o
o
CM
I
00
o>
"eo
0>
I
o>
"cO
-------
CM
00
00
CM
00
<
Q
LLJ
l/6uj
!N3nidd3
17
-------
Considering the large number of existing boundary conditions
(12 existing wastewater treatment plants, 50 storage-sedimenta-
tions tanks for stormwater, regional interceptors etc) the
study relied heavily on engineering judgement with regard to
the choice of reasonable alternative strategies for water pollu-
tion control in the study area. However, an electronic data
processing system was designed which allowed design-cost-resi-
dual load calculations on an unified basis with high efficiency.
In the region two seperate water transport systems were defined:
a) an artificial sewer system for the transport of seperate
and combined sewage
b) an overland flow system (the river Glatt and its tribu-
tories).
The developed areas were separated into 120 subunits of homoge-
nous composition (housing developement, major industrial plant).
For each subunit, the load emission was estimated based either
on observed values (industry) or based on activity (population,
areas, industrial water consumption) and unit activity emission
rates. Calculated loads were (a) "transported" through a network,
designed according to physical interceptors, and (b) reduced
as predicted for treatment plants via a complex mass balance
model for these plants. Costs (investment and operation) were
estimated from detailled cost functions for all transport and
treatment links in the network.Changes of the existing transport
network and the treatment processes were defined by engineering
judgement ; reactor design and cost estimates for additional
construction was automatic - yielding a uniform basis for com-
parison of alternative strategies.
-------
Undeveleopped areas (forest, agricultural land) were separated
into 15 subunits. Pollutant loads from diffuse sources were
estimated via water flow (proportiaonal to area) and base concen-
trations. In overland flow only transport of pollutants was
considered - "selfpurification" where required, was predicted
"manually".
Concentrations of pollutants were calculated for several
"control points" in the main stream as well as in its tribu-
tories. Calculations for rain situations were more complex,
details are given later.
All model parameters were determined independant of data from
the study area and the model was used without calibration. Veri-
fication of the model within the study area was successful
where ever observed data was available (influent and effluent
of treatment plants, several situations in the river Glatt).
Based on this successful verification we concluded that extra-
polations were justified.
IDENTIFICATION OF NITROGEN SOURCES AND STUDY OF POTENTIAL
SOLUTIONS
For most nitrogen compounds the winter situation is critical.
Only if we consider the possible oxidation of ammonium in the
river, could nitrite accumulate to intolerable levels at ele-
vated temperatures during summer months (Fig. 5). Low water le-
vels are observed with equal frequency throughout the year. For
nitrogen control the study concentrated therefore mainly on
the winter situation.
19
-------
For a total of eleven different strategies costs and residual
loads were estimated. Only a few of these strategies are rele-
vant with regard to nitrogen control.
Pollutant emmission was based on todays activities increased
by 0 to 30 % as indicated in the regional development plan for
the next 20 years. After decades of rapid development, this
area has reached some form of saturation. The strategies rele-
vant to nitrogen control are:
Strategy 0: The existing treatment plants are used to treat
the prognosted loads. No additional investments are made. (Re-
ference Strategy).
Strategy 6; All but two small treatment plants are expanded to
allow full nitrification throughout the year (e.g. netgrowth
rate of activated sludge <0.13 d , aerobic reactor only).
Strategy 6*: indicates results for the nitrite inputs for stra-
tegy 6 during summer conditions.
Strategy 7 includes Strategy 6 plus partial denitrification
(-50 % NO_) in the larger 3 plants and in smaller plants where
existing excess aeration tank capacity could be converted into
denitrification reactors.
The results are as follows:
Ammonium; Fig. 9 indicates the cumulative load of NH.-N from
the effluent of the Greifensee to the discharge into the Rhine
as predicted for two different strategies for winter conditions,
20
-------
0}UI
cu
CD
0>
CO
c
o>
CO
o>
o
CO
o
Q_
CD
O
CD
CD
*»—
CD
tt:
CD
+-•
CO
0)
>J
3
cn
<0
o
co
>^
cn
CD
"to
(7)
illllllll
o:
UJ
I
LU
to
avoi
21
-------
Nitrification in the rivers is not considered; it would decrease
the residual NH loads as discharged into the Rhine by approx.
one third. During winter months the effluent of the Greifensee
carries a considerable load of NH. (circulation of a highly
eutrophic lake). During summer months NH in the effluent of
the Greifensee is negligible, nitrifying treatment plants dis-
charge less NH. and nitrification in the rivers is increased.
Diffuse sources of NH are negligible throughout the year.
Conclusion; The receiving water standards for NH. (0.4 mg
NH -N/l in winter, 0.2 mg NH.-N/l in summer) can be maintained
with strategy No 6. Major additional investments are necessary
in the region Diibendorf-Ziirich-Opfikon (Fig. 1) . An optimal
strategy for this region should be defined.
Nitrate; Fig. 10 indicates cunulative nitrate loads from the
effluent of the Greifensee to the discharge into the Rhine.
The effluent of the lake as well as diffuse sources are signi-
ficant. Drainage from agricultural areas is especially loaded
with nitrate during winter months, when vegetation does not re-
tain significant amounts of nitrogen. The most important nitrate
source is treated effluent even during winter months. Conside-
ring that today (Strategy 0) approx. 400 kg NH.-N are nitrified
per day in the river at winter conditions, nitrate loads come
close to the tolerable loads even with the reference strategy.
The nitrate loads will further increase with Strategy 6. Parti-
al denitrification in the large treatment plants (Strategy 7)
will reduce the nitrate loads substantially below tolerable limits
22
-------
iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiitiiiiiiiiiiiiiiiiimiiiiiiiii
I IIIMIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIItllllllllllllllll
iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiniiiiiiiiiiiiiiii
CWWXXXVWNXWXXXV
I Illlllllllllllllllllllllllllllllllllllllllllllllt
xvwvvxwxxxxwxv
iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiinniiiii
o|ui
(O
o>
o>
co
c
o>
<*—
o>
CO
o>
o
o
co
o>
cz
o
"-«-•
CO
o
•*-•
CO
<£>
a:
LU
I
LU
o
o
IT)
CN
o
o
p
CN
O
O
IT)
O
O
o
o
o
in
(P/NB>I) avoi 3ivdiiN
23
-------
Conclusion; With strategy No. 6 nitrate loads will come close
to tolerable loads in the Glatt. In the future partial denitri-
fication in treatment plants (Strategy No. 7) and non point
source programs (agriculture) are required to satisfy nitrate
standards. Since these predictions are based on prognosted
load data, strategy 7 may be delayed with regard to strategy 6.
Nitrite; The major sources for N0_ are the effluents of nitri-
fying activated sludge plants (Fig. 11). Today (Reference
strategy No. 0) nitrite standards are exceeded by far. Full
nitrification (Strategy No. 6) will allow to lower nitrite
loads on the river Glatt. The low resulting NH concentrations
will give the tendency to further nitrify N0_ to NO. in the ri-
ver throughout the year. Considering the conservative estimate
for nitrite production in treatment plants, it may be anticipa-
ted that nitrite is below the tolerable concentration in the
Glatt.
Conclusion; Nitrite proves to be the nitrogen compound most
difficult to control, and nitrite loads are most difficult to
predict. Research with regard to nitrite production and oxida-
tion in rivers and treatment plants is urgently needed.
«
PROPOSED SOLUTIONS AND COSTS
Regional considerations and the different control strategies
previously discussed allowed to identify the dominant point
sources for all nitrogen forms in the area Dubendorf-Zurich-
Opfikon. An optimal point source program for this area has to
24
-------
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIMIHIIII
II lllllllllllllllllllllllllllllllllllllllllllllll
XVVWV (.WXWXWXXV
Illlllllllllllllllllllllllllllllllllllllllllll
CD|
e
in
O
o
hiniiiiiiMiiiiiniiiM IIIIIIIIIIMIIII
VVSNVwVXXXVWVS
o>
J£
O
o>
CD
o
CO
o
OL
o>
? 1
o>
o
c=
o>
o>
v«_
"CO
O ~
CO
o
yz CD
I «
CO cO tO
r.'..'.! ;*l Illlllll 1XNXM
o:
LU
avcn
25
-------
be found. In addition a program for the reduction of non point
sources, mainly for nitrate, should be investigated.
Table 4 indicates the present state of the treatment plants in
the problem area. Excess hydraulic and solids handling capaci-
ty in Zurich combined with an overloaded treatment plant in
Dtibendorf, which lacks solids handling, and a plant in Opfikon,
loaded at design capacity,are a unique starting point for fur-
ther sanitation. In the course of the definition of an optimal
solution to this problem eight different stragies were evaluated
- three of them will be presented here.
Strategy 50; Each of the three plants will be upgraded for
full nitrification in activated sludge plants. Sufficient solids
handling capacity will be provided at each plant. Partial deni-
trification may be introduced later by addition of anaerobic reac-
tor volume to the aeration tanks (Fig. 12).
Strategy 52; The overloaded plant in Dtibendorf will be elimi-
nated. The sewage of Diibendorf may be treated in Zurich which
has excess hydraulic and solids handling capacity. Expansion of
the aeration tanks in Zurich is required to yield full nitrifi-
cation. The plant in Opfikon is upgraded as in strategy 50.
(Fig. 13)
Strategy 57; The plant in Zurich remains unchanged and yields
secondary treatment (no nitrification) for the sewage of Diiben-
dorf and Zurich. For nitrification and later denitrification
the secondary effluent of Zurich and Opfikon are combined and
26
-------
Table 4; Design capacity and actual load of the three waste-
water treatment plants in the problem area (major
ammonium point sources).
Existing
Plant (acti-
vated sludge
process)
Diibendorf
Zurich
Opfikon
In Operation
since
1964
1970
1962
Daily Peak Dry
Weather Flow
Design
m /s
0.30
1.50
0.25
Actual
m /s
0.36
0.90
0.24
Solids
Handling
Capacity
Utilizes
none
<80%
100%
Distance
3500 m
2200 m
27
-------
PRIMARY |
PRIMARY
SECONDARY (AS)
NITRIFICATION
SECONDARY (AS)
NITRIFICATION
PRIMARY |
SECONDARY (AS)
NITRIFICATION
FILTRATION | (FILTRATION | [FILTRATION |
1 i I .
RIVER GLATT
Figure 12
28
-------
SECONDARY (AS)
NITRIFICATION
SECONDARY (AS)
NITRIFICATION
FILTRATION
J L
FILTRATION
RIVER GLATT
Figure 13
29
-------
treated in a tertiary trickling filter (Fig. 14). In this stra-
tegy denitrification requires reconstruction of the secondary
treatment in Opfikon, since all organics in the Opfikon primary
effluent are required for partial denitrification of the sewage
of the entire area. Economically denitrification may therefore
only be introduced when the treatment plant in Opfikon re-
quires major reconstruction and expansion(approx. 1990) (Fig.15).
Total costs for the different strategies are predicted on a
yearly basis (capital plus operation) based on the following
assumptions:
Operating costs (Basis 1978) based on detailled analysis
of requirements for personel, maintenance, energy, chemi-
cals, solids handling etc. as experienced in Switzerland.
Capital costs (Basis 1978) based on present costs of new
construction, real interest rate (actual interest minus
inflation) = 2 %, depreciation of plants and sewers over
15 to 40 years.
Table 5 summarizes the predicted costs for the different stra-
tegies. Denitrification was not included in the cost predic-
tions, it contributes only marginally to total yearly costs.
Based on predicted costs, strategy 50 was eliminated from fur-
ther consideration. Political (legal combination of sewerage
districts) and technical (available land for expansion) consi-
derations require detailled analysis of strategies 52 and 57.
These further studies are now on their way.
Fig. 16 and 17 indicate concentration profiles for ammonium and
nitrate in the river Glatt (winterconditions, low flow, dry
weather)before (Strategy 0) and after (Strategy 52 and 57) in-
30
-------
CO
cu
-H
tr>
0)
-P
rd
n
-P
co
C
O
•H
4J
rd
-P
•H
C
rd
CO
-P
C
0)
M
0)
M-I rd
O Q)
M
CO rd
-P
CO >i
o -d
o d
4J
>i CO
rd -H
1 CO
0)
T3 U
fl M
rd 3
O
CO CO
-p
C -P
0) C
e -H
-P o
CO 04
cu
m
(d
EH
.p
d
CU CO
g -p
-P d
rd rd
0) rH
VH Pi
•P
M £H
Q) -H
-P -P
rd CO
£ -H
X
Q) 0)
CO &>
rd d
£ -H
T3
4-i rd
O M
Cn
co QJ
-P 3
CO
O S-l
U (1)
-p
rH M-l
rd rd
-p
o
H
d
d cu
35
rd rd
M 0)
tT1 M
DJ-P
j_i
m Q>
O -P
(d
CO S
-P
CO
O
U
Q)
£>
CO -H
-P -P
CO rd
O rH
U 0)
H
>i
rH
rd \
0) M
ro
o
H
CU
[>
CO -H
-P -P
co rd
0 rH
U CU
i_|
tn
d
-H h)
-P \
rd >-l
^ |X|
CO -H
•P -P
co rd
O rH
O QJ
•H
M h)
rd \
(U ^
K"I fo
ro
o
rH
(U
CO -H
-P 4J
d rd
CU rH
g Q)
_jj ^i
CO
/1\
;>
d M
H tM
VD
o
rH
d
0
"fH ^*l
4J tr>
rd CU
-P -P
•H rd
*^ Lj
M ^
rd -P
CO W
dp dp <*P
^& t**** ^^
•H 0 0
rH rH H
^-^
H
O O O
O 00 C^
00 rH in
CPi O^ 00
r^ H o
H rH 0
rH rH rH
O O O
[*— O^ CTi
rH CTi VO
ro CN i
o o cu
•a ^ tn
d -H -P
cu IM rd
Q Ql ^|
:3 O -P
Q 1 CO
1
. — .
CN
O
<
rM ^i
m 3 cr>
O JJ
-------
NITRIFICATION
TRICKLING FILTER
FILTRATION
RIVER 6LATT
Figure 14
32
-------
PREPRECIPITATIONJ
(P-CONTROL)
CO
O
DENITRIFICATION
NITRIFICATION
TRICKLING FILTER
o
z
o
ZJ
O
g:
o
LU
a:
FILTRATION
RIVER GLATT
Figure 15
33
-------
c
>
c
a
•4-
2
c/5
I
<
<
; /
-t/
•*-.
/ Q>
/ -1
/ <•*
/ «
/ k_
* G>
/ »2
/
7
1
D in o u~> o «^>
r> CM" c-j ^ *-' O
in
CN"
in
""a>
"cs
en
in
>s
CD
1
in
C^^
^^
CO
C
•^^^^^^
-co
•z
•o
ISI
A ^" ^
Q
-u.
5
o>
sv
to .
-g'
co
0)
5-i
3
O>
o>
co
-------
)
~cu
>
3
o>
o
c
(O
w.
CD
,2
I
O
<£>
\
\
\ °
. l =^
l\ F
a\ «
\-*->
'c\
L il
•-l
**r *
s\
h ^
1 f^
*i
o> 1
l\
fe 1\
^"* •
S \
£ \
c/> \
• t
• 1
•
•
CN
ir>
^^
CT>
0>
2
t/5
iii
o o o
in -^
-
\
A_
v
1
1
1
1
•
•\
\
1 *
O O r
CN — '
CD ci)
Olf=
CO Ic
•5°=
co
Q
-CD
- 2»
d^
^^^
• o
•j*
"N
-Q
o>
o>
C C
-u. ^^
H— »—
-j UJCD
0)
s-l
D
cn
•H
'ONOO 31VdilN
35
-------
troduction of nitrification in the entire area (Strategy 6).
For both NH. and NO legal tolerance limits may just be main-
tained in the future. The effect of selfpurification (nitrifi-
cation) does not significantly affect future N0_ concentrations.
NH concentrations will be lower at summer temperatures. Deni-
trification may have to be considered in the future.
The biological state of the Glatt will improve significantly
after the realisation of strategy 6. The reduction of the ammo-
nium and nitrite concentration is expected to improve fishing,
tertiary filtration will reduce sludge sedimentation in the
river and thereby reduce protozoan development, phosphorous
control by ferric precipitation is introduced mainly because
of the expected parallel improvement of organics removal which
will reduce heterotrophic production in the river. However this
upgrading of the existing treatment plants is merely a first
stage and will not yield a state of the river Glatt which ful-
fills all legal requirements of the Federal Ordonnance for
Waste Water Discharge (1975). Further treatment will be required
At this point it is however not feasible to design a complete
(final) sanitation program.
All point source programs should be accompanied by sanitation
progams for non point sources and the eutrophic Greifensee
(lake). These pollutant sources become increasingly important
as advanced wastewater treatment is introduced.
36
-------
STORMWATER SITUATION
Besides the low flow, dry weather situation, the receiving wa-
ter situation during rain events has been analysed. For this
purpose a complex simulation procedure has been designed
which allows estimation of pollutant loads during increased
water flow. The simulation procedure considers overland flow,
combined sewers overflow, separate sewers (25 % of sewered
areas) reduced treatment efficiency during increased combined
sewage flow, resuspension of sediments in the rivers etc. It
was successfully verfied (not calibrated) with data from three
storm events in the study area (influent and effluent of the
major treatment plant, water quality data in the river Glatt).
The simulation procedure was then applied to 13 different rain
events, representative for all rains during one year, with the
assumption, that all treatment plants are upgraded to yield
full nitrification during dry weather and winter conditions.
Table 6 summarizes pollutant concentrations during rain events
as discharged from the Glatt into the Rhine. Table 7 indicates
the average relative magnitude of water and ammonium sources
during rain events.
Several strategies were designed to reduce ammonium loads du-
ring rain events. Technically realistic and cost effective is,
to assign priority to upgrading wastewater treatment for 2-3
dry weather flows (nitrification) over other investments such
as treatment of combined sewer overflow or temporary storage.
37
-------
in
-p
rd
to x!
C -P
O
•H £
•P O
rd -H
•P O<
fi g
CD d
o to
fl to
O rd
U
CD
•P X!
£ -P
•P C
^ O
O Q)
04 W
ft)
ftS
C
O
•H
dUS
"H ^
- o *
PL. U O
>H
CD
X!
4J
CD
P (fl
CN O
4-1 O
O X!
O4
>1 tO
•P O
-H X!
O 04
rd
04 X!
rd -P
O -H
°i
ll
w ».
<4H
0) -H
rd
to
C
O
CD -rl
CD
4J
rd
CD ,
I O4 •
O •
rd
•H
4-1
CD
o
CD
0
to
CD
(d (tt Q)
04
i—I cd *d O.
>i rd X! ti -rl
rl U rd 3
TS tfl
O
CD
rl
rd
• 4->
4->
_ -P rd CD
rd rd rH t>i
rl rH O,
CD O CD
> -P x;
(d rl C -P
CD CD
•d > g -p
(U -rt 4-1 3
4-> >H rd O
rd CD X!
H (D SH Cn
3 X! -P d
U -P O
rH rH }H
rd C H X!
U -H rd 4J
CD
r-l
X)
rd
-
S
-p
•p
rd
i-H
^
Q)
^ Ql
•H fi
03 -H
CD K
5 n
-p
c c
•H -rl
c -d
o CD
•H cn
-P rl
rd rd
!H X!
-P O
C to
Q) -H
rj Q
G
O CO
U rd
CD
tn
S
CD
CN
rH
CD
rd H
CD \
S Cn
g
P
^
CD
4J
rd
CD rH
12 ^
CT*
>i g
SH
P
CN
00 O O V£J O rH
o un vo o o o
1 1 1 1 1 1
H rH
in o o •r o o
o ro •f o o o
ro ^
m o o
m o in ro o o
o I>D •<* o o o
1 I I 1 I 1
CN rn
in o o
•^ o o CN o o
o in ** o o o
•H in
in oo o o o
I
o un CN o o
PL,
^ h^ 1
C-t & ' * ' '
1 1 rH rd
•=!< ro u -P
a o o o s xi
3 Z P 4-> U D4
0)
-P
rd
r)
-P
(d
a
e
a
O
XI
M
rd
U
O
rd
rl
O
•d
CD
>
rH
O
CO
tfl
•H
P
0)
tr>
M
rd
X!
o
tfl
•rl
p
rl
CD
4-1
rd
•p
to
rd
O
CD UH
CD
U
c
rd
C
C
O
T3
M
O
rd
to
CD
Cn
to
to
•rl
CD
•P
rd
O
-H
to
CN
38
-------
Table 7; Relative significance of different yearly water and
ammonium inputs into the river Glatt during wet
weather. Calculations based on the assumption that
all treatment plants nitrify throughout the year.
Source
Relative Significance of Source
in % of Total Input During Wet
Weather
Flowrate
NH
Treatment Plants
Combined Sewer
Overflow
Seperate Sewers
Greifensee (Lake)
Infiltration and
Overload Flow
22
10
6
28
34
46
30
5
13
39
-------
CONCLUSIONS AND RECOMMENDATIONS
This water quality control study for an area with dense popu-
lation was very detailled - in general a less detailed
approach would suffice for the design of an optimal cost ef-
fective strategy. The study concentrated on the estimation
of pollutant loads as discharged into the rivers; the trans-
formation of pollutants in the rivers (selfpurification) have
not been considered to a significant extent. The short flow
time of the river Glatt (12-24 hrs depending on flow rate)
did not require consideration of selfpurification, at least
for a future state, when existing treatment plants are upgra-
ded.
The following conclusions were drawn from the study:
1. This study successfully allowed to identify the magni-
tude of point sources and non point sources for seve-
ral pollutants. For all nitrogen forms, point sources
dominate non point sources in this particulate area.
2. For the introduction of advanced waste water treat-
ment many boundary conditions have to be considered.
In this study area, secondary treatment to different
levels existed for virtually all sewered areas in
the region. This introduces the necessity of defining
specific treatment processes rather than uniform dis-
charge standards in order to define cost effective
strategies.
3. Even in high population density areas, control of
nitrogen to levels, as legally required in Switzer-
land, is possible with "conventional" technology. The
nitrogen form, most difficult to control in rivers,
seems to be nitrite. Our lack of knowledge of the
dynamics of nitrite in rivers and nitrifying treat-
ment plants is the limiting factor in the accuracy
of our predictions.
4. For the climatic conditions and the specific characte-
40
-------
ristics of this area, the study indicates, that sani-
tation of dry weather conditions should have priority
over the sanitation of wet weather conditions at least
for nitrogen compounds.
5. Pilot and full scale experience with tertiary trick-
ling filters and other fixed bed reactors is not suffi-
ciently documented to allow design without further pi-
loting.
6. Nitrite has not been sufficiently considered in water
quality control until now. The behaviour and signifi-
cance of nitrite in wastewater treatment, as well as
in receiving waters, merits further research.
ACKNOWLEDGEMENTS
This study was an interdisciplinary effort. Many collegues
from the Swiss Federal Institute for Water Resources and Water Pollution
Control (EAWAG)and the Office for Water Pollution Control of the
State of Zurich (AGW) as well as others have collaborated in
this study. I am especially grateful to the members of the
project team: V. Krejci, R. Schertenleib, W. Munz, H.R. Rhein,
E. Eichenberger and P. Perret. The study was financially sup-
ported in part by the State of Zurich.
REFERENCES
Federal Ordonnance for Waste Water Discharge (1975). Syst.
Collection of Federal Laws, No. 814.225.21 Bern, Switzerland.
Gujer, W. (1976) "Nitrifikation in Fliessgewassern - Fallstu-
die Glatt", Schweiz.Z.Hydrol. 38; 171-189.
Gujer, W. (1977) "Design of a nitrifying activated sludge
process with the aid of dynamic simulation". Prog.Wat.Tech.,
9, 323-336.
Gujer, W. and P. Erni (1978) "The effect of diurnal ammonium
load variation on the performance of nitrifying activated
41
-------
sludge processes". Prog.Wat.Tech. 10 No.516, 391-407.
U.S. Environmental Protection Agency (1975) "Process Design
Manual for Nitrogen Control", U.S. Government Printing Office
1975 - 630 - 902.
Zobrist, J, J.S. Davis and H.R. Hegi, (1976) "Charakterisie-
rung des chemischen Zustandes des Flusses Glatt", Gas Wasser
Abwass. 56, 97-114.
42
-------
FULL-SCALE COMBINED CARBON OXIDATION—
NITRIFICATION AT THE METROPOLITAN SANITARY
DISTRICT OF GREATER CHICAGO
By
Cecil Lue-Hing, Director of Research and Development
Booker Washington, Research Chemist
David R. Zenz, Coordinator of Research
Alan W. Obayashi, Formerly, Project Manager
T.B.S. Prakasam, Project Manager
SUMMARY AND CONCLUSIONS
During the period of October 1975 through June 1976, a
full-scale evaluation of combined carbon oxidation-nitrifica-
tion was conducted at the West-Southwest Sewage Treatment Plant.
The major objective of the study was to determine the validity
of previously formulated single-stage nitrification design
criteria with respect to the SRT required for the expansion and
improvement of the plant to meet applicable effluent NR.-N, BOD
and TSS standards. Battery D, the new 300 mgd secondary aera-
tion battery, was used to achieve this objective, with Batteries
A, B and C being similarly operated for purposes of comparison.
Based on the results obtained during the nine-month period
studied, the following conclusions can be made:
1. With environmental factors such as pH, alkalinity
and DO not being limiting, SRT was the major parameter
controlling nitrification.
2. The aeration tank suspended solids loading, rather
than the traditional parameter, BOD loading, deter-
mined the solids production, and consequently the SRT
43
-------
which could be maintained. This was due to most of the
total BOD being accounted for by the SS, with the BOD
to SS ratio in the influent to the aeration tanks being
highly variable, ranging from 1:1 during dry weather to
about 1:3 during storm flow conditions. On an overall
basis, 0.9 Ibs of SS were produced per Ib of influent
SS.
3. A 10-day design SRT for the W-SW Plant would provide
adequate flexibility to accommodate significant (.tran-
sient) increases in waste loadings in terms of NH.-N,
BOD and suspended solids, as well as decreases in the
nitrification rate at low wastewater temperatures of
10°C or less.
4. The design SRT for expansion and improvement of the
W-SW Plant should be 10 days. This would ensure suc-
cessful, year-round combined carbon oxidation-nitri-
fication, with the expected effluent quality being
less than the IPCB standards of 2.5 mg/1 NH.-N during
April through October, 4.0 mg/1 NH.-N at all other
times; and 10 mg/1 BOD and 12 mg/1 TSS on an average
monthly basis throughout the year.
5. Based on the assumption that MLSS concentrations will
be maintained at approximately 3000 mg/1, an increase
in total aeration tank volume to 411 million gallons
will be required in order to achieve a design SRT of
10 days. The resulting HRT at design flow of 1315 mgd
will be 7.5 hours.
44
-------
INTRODUCTION
The adverse impacts associated with ammonia discharges
into receiving waters have long been recognized. These concerns
are primarily in terms of the potential toxicity to fish and the
high oxygen demand exerted during bacterial stabilization of
ammonium to nitrate. Consequently, some regulatory agencies
have promulgated limits on effluent discharges of NH.-N,
The applicable NH.-N standards to be met by the Metropolitan
Sanitary District of Greater Chicago (MSDGC) were imposed by the
Illinois Pollution Control Board (IPCB) in 1972, and restrict
effluent NH.-N concentrations to 2.5 mg/1 during April through
October and 4.0 mg/1 at all other times(1). Compliance with the
numerical limits of the standards, as presently stated, will be
based on 24-hour composite samples. The standards will become
effective upon completion of the scheduled expansion and improve-
ment of the MSDGC's three major sewage treatment plants; namely,
Calumet, North Side and West-Southwest. In addition, compliance
with effluent BOD and TSS standards equalling 10 and 12 mg/1,
respectively, averaged over any consecutive 30-day period, will
also be required.
Process Considerations for NH4-N Removal
Due to the MSDGC's historical success with the activated
sludge process, only biological treatment methods were seriously
considered for ammonia removal during the early evaluations of
process alternatives. Physical-chemical treatment methods were
45
-------
rejected primarily because of the high cost of chemicals re-
quired to treat the large volume of plant flows which approxi-
mate 1.5 billion gals per day.
In the late sixties and early seventies, the MSDGC's re-
search on ammonia removal was heavily committed to the two-
stage biological process. This subsequently resulted in con-
struction of two comparatively small-scale, two-stage nitrifi-
cation activated sludge plants. The first of these, the 30
mgd Egan Plant, was completed in 1975; and the 70 mgd O'Hare
Plant was completed in 1980. However, even during this period
of new plant construction, the MSDGC was conducting single-
stage nitrification pilot studies at each of the three major
plants in order to investigate the potential economic and
operational advantages of single-stage nitrification (2, 3, 4).
The results obtained from these studies demonstrated the feasi-
bility of the single-stage nitrification process for both the
North Side and the West-Southwest (W-SW) Plants. Conversely,
the two-stage process was determined to be more practical for
the Ca.lumet Plant, which has a higher industrial input and
higher influent sewage BOD, suspended solids and ammonia nitro-
gen concentrations.
Theoretical Considerations for NH4-N Removal in Activated
Sludge Systems
An important fundamental parameter in designing and opera-
ting an activated sludge process to accomplish nitrification is
the concept of solids retention time (SRT), which was developed
46
-------
by Lawrence and McCartyCS), Mathematically, SET is the theo-
retical average retention time of suspended solids in an acti-
vated sludge system, and is the total mass (suspended solids)
in the system divided by the rate at which the mass is leaving
the system. Under steady-state conditions the reciprocal of
the SRT is the net growth rate. Thus, in order to maintain a
culture of nitrifying bacteria (and therefore oxidize ammonia)
in an activated sludge plant, the process must be operated
such that the net growth rate (1/SRT) is less than the maximum
growth rate (u ) of the nitrifying organisms. Otherwise, the
nitrifiers, being slower growing organisms, will progressively
diminish in proportion to the total microbial population and
eventually be washed out of the system.
Without any specific heavy metals or toxic organics being
present at concentrations which could inhibit nitrification,
the maximum growth rate of nitrifying bacteria (and hence the
minimurn SRT) is affected primarily by temperature, dissolved
oxygen concentration and pH. The EPA Nitrogen Control Manual(6),
which extensively discusses these parameters, lists the follow-
ing environmental conditions as being conducive to the growth
of nitrifying microorganisms: DO >2 mg/1, pH 7-8.4, and about
7 mg/1 of alkalinity per mg/1 of NH.-N oxidized. In situations
where these conditions can be maintained, the most important
design consideration is to determine or estimate the maximum
growth rate of the nitrifying bacteria (i.e., minimum SRT) at
a given wastewater temperature.
47
-------
The relationship between percent NH.-N remaining and SRT
is graphically illustrated in Figure 1. As shown, the SRT
maintained in the system should be greater than the minimum
solids retention time (6m) to sustain successful nitrification.
c
Therefore, the design SRT (8 ) will exceed the minimum SRT,
o
with the actual value of 8 depending on the magnitude of opera-
tional safety factor used in its computation. Generally, the
amount of safety factor employed is dependent on fluctuations
in influent sewage loadings, mainly with respect to ammonium,
BOD and suspended solids.
Objective and Experimental Plan
The objective of the study was to validate full-scale
single-stage nitrification design criteria for the expansion
and improvement of the West-Southwest Plant for the 1990
design year(7). These design criteria were initially developed
by Consoer, Townsend and Associates (C.T. & A.) from informa-
tion obtained through their nitrification studies conducted
at Flint, Michigan from October, 1970 through March, 1972(8).
Subsequent studies by the MSDGC during the 1974-75 winter(3)
supported the C.T. & A. design recommendations in terms of
hydraulic retention time (HRT) and several other criteria.
However, whereas an estimated SRT of 4.5 days would be main-
tained in the aeration tanks based upon solids loading figures
presented by C. T. & A., the MSDGC study recommended a design
SRT of 10 days. The available information indicated that a
48
-------
w
u
w
8 a
01
w
9
O
3
o
8
O
5
O
I
O
U. *
o
w
oc
o
o
o
in
9NINIVW3U N-frHN %
49
-------
10-day SRT would favor successful nitrification during winter
operation at W-SW, and provide an adequate safety factor
against future loading uncertainties. Thus, the present study
primarily attempted to firmly establish the design SRT for the
expanded W-SW Plant.
The experimental plan consisted of operating the new 300
mgd secondary aeration battery at W-SW, Battery D, in a single-
stage nitrification mode from October, 1975 through June, 1976,
thus encompassing both winter and fair weather periods. The
nine-month study was characterized by two phases of operation
which entailed controlling the primary influent source to
Battery D. Initially, during Phase 1, primary effluent was
principally from conventional rectangular settling tanks,
while during Phase 2 primary effluent from older Imhoff set-
tling tanks was used.
For purposes of comparison, Batteries A, B and C of the
4-battery plant were also operated to achieve single-stage
nitrification. The data obtained considerably facilitated
understanding the observations made in Battery D and in esti-
mating the minimum SRT required for nitrification.
50
-------
MATERIALS AND METHODS
Description of West-Southwest Sewage Treatment Plant
The West-Southwest (W-SW) Plant, which treats combined
sewer and storm flows, consists of two separate primary facil-
ities which discharge flow to an activated sludge system com-
prised of four aeration tank batteries designated A, B, C and
D. A schematic diagram indicating the general path of flow
through the W-SW Plant is shown in Figure 2.
Originally, the treatment facilities consisted of only
the West Side Plant, which was placed into operation in 1930
as a primary plant composed of two batteries of Imhoff tanks
for primary sedimentation, and sludge storage and stabiliza-
tion. Subsequently, in 1935, a third battery of Imhoff tanks
was added, thereby increasing the West Side design primary
treatment capacity to 472 mgd. With the addition of the South-
west Plant in 1939, the combined W-SW Plant was upgraded to a
secondary treatment facility. At that time two secondary
aeration tank batteries (A & B) were placed into operation
and received the primary effluent from both the Imhoff tanks
of the West Side Plant and the newly constructed Southwest
Plant primary settling tanks. Unlike the Imhoff tanks', the
Southwest primary tanks were of a modern design with settled
sludge being continuously drawn off. Thus, sludge storage
and stabilization were not features of this new primary system.
The remaining aeration tank capacity was added in 1949
and 1975, with the construction of Batteries C and D, respec-
51
-------
FIGURE 2
SCHEMATIC FLOW DIAGRAM OF THE
WEST-SOUTHWEST TREATMENT PLANT
•RAW SEWAGE
BAR
SCREENS
GRIT TANKS
(THE WEST SIDE)
(THE SOUTHWEST SIDE)
•RAW SEWAGE
CONVENTIONAL
PRIMARY
SETTLING
TANKS
AERATION CLARIFIERS
TANKS
FINAL
52
-------
tively, which increased the design secondary treatment capa-^
bility to the present level of 1200 mgd.
West Side Plant
Before the West Side influent raw sewage is treated in
the three batteries (A, B and C) of Imhoff tanks, the flow
passes through coarse and fine screens, an aerated grit cham-
ber, and scum skimming tanks. The sewage is then settled for
about two hours at design flow in Batteries A and B, and one
hour in Battery C. Jointly the three batteries of Imhoff tanks
provide primary sedimentation for approximately 50-55 percent
of all wastewater entering the combined treatment facilities.
Primary effluent from the West Side Plant combines with the
effluent from the Southwest primary settling tanks. The flow is
then further treated in the activated sludge system of the South-
west Plant.
Southwest Plant
The Southwest Plant is a conventional activated sludge
treatment facility consisting of unit processes of coarse bar
screens, aerated grit chambers, rectangular primary settling
tanks and four batteries of aeration tanks followed by final
settling tanks.
Total volume of each aeration battery is about 50 million
gallons, divided equally into eight, 4-pass tanks, which pro-
vide 4 hours detention time at a design flow of 300 mgd. All
four batteries are similar in arrangement to the schematic
diagram of Battery D shown in Figure 3. As indicated, influent
sewage is comprised of the two primary effluents and is mixed
53
-------
-J
*
.oS
*£o
zw
ss
S3
CO
tai
3
O
CO
IU
*
tt
I
IL,
o
o
<
o
I
IU
z
o
CO
ooo
ooo
ooo
ooo
ooo
ooo
UJ
_l
<
o
(/)
o
z
Gr
e
\
v
/ \
^_
-^>
-to-
•=)
10
«
-=)
SS
S3
M
>•
X
I-
9
o
•»
^ •- *
33*
iU
•!«i
0 i
10 >-•
I £ °
• °
• 2 *
g 2 i
5 i »
I- « O
• * z1
m
•e < •»
M J M
< o o
I I
/
0-KK-U.-WKW
>s/
I- < Z M •»
54
-------
with return sludge and distributed to the aeration tanks thru
a mixing channel. Porous plate diffusers located along one
side of each tank produce a spiral-type flow pattern and effect
mixing and aeration. However, whereas Batteries A, B and C
each have about 38,000 air diffuser plates, Battery D has
52,000—an increase of approximately 37 percent.
The aeration tanks are constructed in such a way that two
tanks discharge into a common conduit feeding six, 126 ft dia-
meter circular clarifiers. Therefore, with 4 pairs of aeration
tanks, each battery consists of 24 clarifiers; all of which are
center-feed, radial flow tanks. Normally, 30 percent sludge
recycle is maintained in Batteries A, B and C which presently
have limited return sludge capacity (130-140 mgd per battery). .
In comparison, Battery D has 100 percent design recycle capac-
ity at an influent battery flow of 300 mgd.
Following final clarification, the combined effluent from
the four aeration batteries is chlorinated prior to being dis-
charged into the Chicago Sanitary and Ship Canal. Currently,
about 800 mgd of wastewatex is given secondary treatment in the
existing facilities, with a typically very good quality effluent
being produced averaging less than 10 mg/1 for both BOD and TSS.
The planned expansion and improvements to the W-SW Plant will
accommodate an average (1990) design flow of 1315 mgd.
Operational Modes During the Nitrification Study
Phase 1 - Southwest Primary Effluent as the
Feed Source to Battery D
During Phase 1, extending from October 9, 1975 thru
55
-------
March 23, 1976, Battery D received primary effluent almost ex-
clusively from the Southwest preliminary tanks. This neces-
sitated diverting all West Side Imhoff tank effluent to Batter-
ies A, B and C during periods of dry weather flow. The only
time when this mode of operation was modified was during
periods of heavy rains or snow melts. Under these conditions
strict flow control could not always be maintained, and resulted
in small amounts of Imhoff effluent being pumped to Battery D.
Thus, except for periods of substantial runoff, the in-
fluent flow to Battery D (Southwest primary) was fairly con-
stant, and averaged 180 mgd during the first two months of the
study. With successful nitrification having been observed for
the first two months, the pumpage of Southwest primary effluent
to Battery D was increased and maintained at approximately 220
mgd thereafter in order to monitor performance at higher load-
ings.
In addition to exercising operational control of influent
source and quantity of flow to Battery D (i.e., HRT control),
SRT, MLSS levels, recycle rates, and air to sewage ratios were
also controlled. Typically, these parameters were maintained
within reasonably fixed ranges, with the exceptions mainly
being due directly or indirectly to storm runoff conditions.
Phase 2 - West Side Imhoff Primary Effluent
As the Feed Source to Battery D
The second phase of the study, during which the influent
source to Battery D was essentially restricted to the West Side
Imhoff tank effluent, covered the period from March 24 through
56
-------
June 30, 1976. As had previously been done with the South-
west primary effluent, the Imhoff effluent flow to Battery D
was maintained at approximately 220 mgd. The remaining West
Side effluent, in addition to the total Southwest primary ef-
fluent flow, was treated in Batteries A, B and C. However,
during periods of storm runoff Battery D was switched over to
Southwest primary effluent with West Side effluent being di-
verted to the three other batteries. This mode of operation
permitted more efficient utilization of the Southwest pumping
capacity during periods of high flow.
Due to a significantly higher solids production rate
being obtained in treating the Imhoff effluent, MLSS concen-
trations were increased to approximately 3500 mg/1 during
Phase 2 to facilitate maintaining SRTs equal to at least 5-6
days (compared with 2800 mg/1 and 9-10 days during Phase 1).
Operation at MLSS levels of 4000 mg/1 or greater (hence longer
SRTs) was precluded on the basis of the expected adverse impact
of the solids loading rates on final clarification during pe-
riods of storm water runoff.
Although the overall quantity of process air supplied to
the Battery D aeration tanks (0.7 cu ft/gal of sewage) approxi'-
mately equalled the ratio measured during Phase ly the amount
supplied to the influent passes was preferentially increased.
This ensured that dissolved oxygen concentrations in the first
passes, where the oxygen demand was highest, would be at least
2.0 mg/1 and thus adequate for carbonaceous and nitrogenous
oxidation.
57
-------
Modified Operation of Batteries A, B and C
When it bacame apparent in early November, 1975 that all
four batteries were nitrifying to various degrees, the study
was expanded to include more extensive monitoring of Batteries
A, B and C.
By coincidence, Battery A received approximately the same
quantity of flow as Battery D and thus served as a good com-
parison battery. However, unlike Battery D, Battery A received
a mixture of both influent streams throughout the study. Gen-
erally, West Side Imhoff effluent constituted the major in-
fluent source to Battery A during Phase 1, with the flow dis-
tribution being more nearly 50 percent during Phase 2.
The remaining primary effluent available for secondary
treatment after the adjustment of flows to Batteries A and D
was distributed about, equally to Batteries B and C. Therefore,
flows to Batteries B and C were normally lower, and resulted
in generally higher HRTs. Further, in most cases, flows to
both batteries were mixtures of the West Side and Southwest
primary effluents.
Commencing during the latter part of Phase 1, operations
in Batteries A, B and C were directed towards maintaining
minimum SRTs of 6 days. This required firmer control of in-
fluent flow, MLSS levels and sludge wastage rates. Also, at-
tempts were made to maintain DOs in the influent passes of each
aeration tank at or above 2.0 mg/1 by increasing the air supply,
58
-------
while simultaneously reducing the air input to the effluent
passes. However, this procedure was not quite as effective
in keeping DOs above the 2.0 mg/1 limit in Batteries A, B and
C as it was in Battery D.
Daily Monitoring and Sampling
Maintenance and Operations (M & 0) personnel at W-SW
retained operational control of Batteries A, B, C and D
throughout the nine-month study. Thus, changes in operating
parameters requested by the Research and Development (R & D)
staff were implemented by M & 0 to the extent consistent with
maintaining an efficiently operating sewage treatment plant.
Daily influent, return and waste sludge flows in each battery
were totalized and compiled by M & 0 and subsequently trans-
mitted to R & D.
Automatically-sampled and refrigerated composites 01
Imhoff effluent, Southwest primary effluent, and the clarified
effluent from each battery were collected each day. In addi-
tion, mixed liquor and return sludge samples from two selected
tanks in each battery were grab-sampled twice per 8-hour
shift by operating personnel, who simultaneously made Winkler
DO determinations on mixed liquor collected at the tank inlets,
mid-points and outlets.
Generally, two aeration tanks in each battery were pro-
filed once each week for NH.-N and combined NO-- and NO.,-N by
taking grab samples of mixed liquor at the inlet, 1/3, 2/3 and
59
-------
end of the first pass, and at the ends of the second, third
and fourth passes of the 4—pass aeration tanks. DOs were
determined at the various points using a field probe. The
information obtained provided an assessment of the NH.-N oxi-
dation rates occurring in each battery under the variable
operating conditions.
Chemical Analyses
Mixed liquor samples were analyzed daily for total sus-
pended solids (TSS) and volatile suspended solids (VSS), while
the return sludge samples were analyzed for TSS only. These
determinations utilized a TSS method(9) specificially adapted
for samples containing very high concentrations of suspended
solids. The analyses performed on the daily influent and ef-
fluent samples (pH, BOD, COD, TSS, VSS, TKN, NH4-N, NO.,- and
NO..-N) were all done according to procedures described in
Standard Methods(lO).
60
-------
RESULTS AND DISCUSSION
Performance of Battery D
Operational Parameters
Table 1 summarizes the operating conditions in Battery D
during Phases 1 and 2 and shows that influent flow rates were
about equal during both phases and resulted in equivalent HRTs
averaging approximately 5.5 hours. Based on overall weekly
averages, HRTs ranged from 4 to 7 hours, with the variability
being due to stormwater runoff and low dry weather flow,
respectively. Aeration tank wastewater temperatures are not
included, but fluctuated seasonally and ranged from 11 to 21°C
during Phase 1 and 13 to 21°C during Phase 2.
The difference in SRT maintained during the two phases
was the most dramatic difference in the operation of Battery D.
Although influent sewage flows and F/M ratios were approximately
equal during both periods of operation, the SRT established
during Phase 1 was about twice the SRT maintained during Phase 2
(9.8 vs. 5.4 days). In other words, suspended solids production
was nearly double utilizing the Imhoff tank effluent as the in-
fluent source versus using the Southwest primary effluent.
Mainly this was due to a higher suspended solids loading (93 vs.
124 tons/day) in the Imhoff primary, with a greater percentage
of the solids being more inert to biological oxidation. Thus
it was apparent that the BOD loading, as represented by the
F/M ratio, did not accurately reflect the potential solids pro-
61
-------
TABLE 1
Operational Parameters in Battery D During Phase 1
and Phase 2
Flow Rate, mgd
% Return Flow
HRT, hrs
MLSS, mg/1
SRT, days
F/M Ratio
( #BOD/#MLVSS/day )
NH4-N Loading
( #NH4 -N/ #MLVS S/day )
SS Loading (tons/day)
DO Concentration, mg/1
Phase 1
218
70
5.5
2900
9.8
0.19
0.029
93
2.9
Phase 2
214
70
5
3500
5
0
0
124
2
.6
.4
.17
.013
.3
(tank inlets)
Clarifier^Solids Loading
(Ibs/ftVday
SSR, gal/day/ft2
30
770
35
720
62
-------
duction. The implications of these results are especially
significant, and have been discussed in more detail by Obayashi
et al(ll) .
In general, DO concentrations at the aeration tank inlets
equalled a minimum of 2.0 mg/1, and averaged about 2.5 mg/1
during the nine-month period. The relatively few instances
when inlet DOs were depressed below 2,0 mg/1, to as low as 1.5
mg/1, were due to reductions in the process air supply to Bat-
tery D as a result of blower outages or reduced blower ef-
ficiency at warmer wastewater temperatures.
A fairly uniform rate of sludge return, averaging about
70 percent, was maintained during both phases as clarifier
2
solids loadings equalled 30 and 35 Ibs/ft /day, respectively.
During peak flows clarifier loadings were as high as 50 to 60
2
Ibs/ft /day. However, even at these loading rates there was
still no evidence of solids building up in the settling tanks/
and final clarification was good.
Analytical Parameters
Table 2 lists average influent and effluent wastewater
characteristics relative to both periods of operation. During
Phase 1 the NH,-N loading to Battery D was more than two times
higher than the NH.-N loading during Phase 2 because of the
higher concentration of NH.-N in the SW primary effluent. Of
paramount importance, however, were the consistently low efflu-
ent NH.-N values obtained from Battery D. For example, there
63
-------
TABLE 2
Influent and Effluent Characteristics of Battery D
During Phases 1 and 2
Influent
NH.-Njj mg/1
Total BOD, mg/1
Soluble BOD, mg/1
Suspended Solids, mg/1
%VSS
pH
Alkalinity (as CaC03) , mg/1
(N02-N+N03-N) , mg/1
Effluent
NH4-N, mg/1
Total BOD, mg/1
Suspended Solids, mg/1
PH
Alkalinity (as CaCO3) , mg/1
(N00+NO.j-N) , mg/1
Phase 1
13.3
93
36
102
73
6.8-7.8
190
1.8
0.6
4
4
7.1-8.2
100
14.5
Phase 2
7.1
91
30
139
64
6.9-7.8
164
0.7
0.2
2
3
7.0-8.0
102
6.4
64
-------
were only 20 scattered days during the entire study in which
the NH.-N residual was greater than 1.0 mg/1. Thus, the average
effluent NH.-N concentration did not exceed 1.0 mg/1 during any
30-day period, with the average values for Phases 1 and 2 being
0.6 and 0.2 mg/1, respectively (96% removal). The mg/1 of NH.-N
removed generally approximated the mg/1 of NO_- and NO--N formed,
therefore indicating that nitrification mainly accounted for
the NH.-N removal.
While it was demonstrated that a 10-day SRT was adequate
to sustain nitrification during Phase 1, the minimum SRT that
would sustain nitrification was not determined. However, the
significantly lower SRT of 5.4 days maintained during Phase 2
apparently had little effect on nitrification efficiency. The
increasing trend in wastewater temperature, which was coinci-
dental with the change in operation from Phase 1 to Phase 2,
perhaps enabled maintaining an adequate nitrifying mass, al-
though the SRT obtained was less.
As Table 2 indicates, there was sufficient alkalinity in
the wastewater to meet the requirements of NH.-N oxidation
during both phases. The alkalinity consumed was 6.9 and 8.8
Ibs/lb of NH.-N oxidized, respectively, which was reasonably
close to the theoretical value of 7.2 Ibs.
Battery D also performed efficiently with respect to BOD
and SS removal, achieving overall reductions greater than 96
percent for both parameters. The average monthly effluent BOD
65
-------
and SS concentrations did not exceed 5 mg/1 during either
phase of operation.
Performance of Batteries A, B and C
Battery A
Influent sewage flows to Battery A approximately equalled
the influent flow to Battery D. Thus, as Table 3 reveals, HRTs
in Battery A were approximately equal to HRTs obtained in
Battery D, averaging 5.6 hours. However, unlike Battery D,
Battery A treated a mixture of both primary effluents through-
out the study, with the flow distribution generally being no
more than 70 percent of either stream. Consequently, NH.-N
and BOD loadings were fairly consistent, averaging 0.22 Ibs
BOD and 0.024 Ibs NH4~N per Ib MLVSS per day during Phase 1,
compared with 0.20 and 0.018 Ibs BOD and NH4-N, respectively,
per Ib MLVSS per day during Phase 2.
As noted in Table 3, SRTs in Battery A averaged 4.6 days
during Phase 1 and 5.3 days during Phase 2. Higher average
SRTs (i.e., 10 days) could not be maintained due primarily to
the relatively high suspended solids loadings applied to Bat-
tery A averaging 107 and 113 tons/day, respectively. This
necessitated higher wastage rates in order to keep mixed liquor
levels within practical operating limits.
In spite of significant losses of process air through
long-standing leaks in some of the air headers, DO concentra-
tions at the inlets to the Battery air aeration tanks were
66
-------
TABLE 3
Operational and Analytical Parameters for Battery A
During Phase 1 and Phase 2
Flow Rate, mgd
HRT, hrs
SRT, days
F/M Ratio (#BOD/#MLVSS/day)
NH4-N Loading (#NH4-N/#MLVSS/day)
DO Concentration, mg/1
(tank inlets)
NH.-N, mg/1 in
out
Total BOD, mg/1 in
out
Suspended Solids, mg/1 in
out
Phase 1
219
5.6
4.6
0.22
0.024
2.1
10.0
4.9
94
8
117
8
Phase 2
219
5.
5.
0.
0.
1.
8.
3.
87
9
124
7
6
3
20
018
9
1
0
67
-------
typically in the range of 1.5 to 2.5 mg/1. As a result, inlet
DOs averaged approximately 2.0 mg/l during both phases of opera-
tion, and were appreciably less than 2.0 rag/1 only during June
(equalling 1.4 mg/1).
A high degree of nitrification was normally not observed
in Battery A, as evidenced by the effluent NH.-N data presented
in Table 3. On an average basis about 55 percent of the
NH.-N applied was removed, with the effluent NH.-N concentra-
tion being 3.9 mg/1. The applicable (future) NH.-N standards
of 4.0 mg/1 during winter and 2.5 mg/1 during summer were ex-
ceeded during all but 122 days of operation. Significantly,
about half of the 122 days of effluent NH.-N compliance occurred
during November 21 thru January 21. This two-month period was
characterized by inlet DOs being ^2,0 mg/1; HRTs "of 5-6 hours;
SRTs of 4-6 days; and aeration temperatures initially in the
range of 16-19°C. Subsequent temperature decreases to 11-12°C
during the next several weeks, accompanied by lower SRTs, re-
sulted in a deterioration in nitrification efficiency to less
than 50 percent removal (effluent NH.-N equalling 5 to 6 mg/1).
Removals did not increase above this level until wastewater
temperatures increased to 17-20°C.
Generally, effluent BODs and TSS averaged less than 10
mg/1, reflecting an excellent quality secondary effluent.
Even during those periods when the influent quality was the
poorest, or when clarifier solids loadings were high due to
68
-------
stormwater flows, the Battery A effluent averaged no more than
14 mg/1 BOD and 12 mg/1 TSS, respectively.
Batteries B & C
The average operating conditions and analytical quality
obtained with regard to Batteries B and C are summarized in
Tables 4 and 5>, respectively. In general, a similar daily
quantity and distribution of flow (Southwest primary effluent
and Imhoff effluent) were made to both batteries. Thus, HRTs
were roughly comparable throughout the study, averaging over-
all about 6-6.5 hours. BOD and NH.-N loading rates averaged
slightly lower in Battery C during both Phases 1 and 2. SRTs,
on the other hand, were approximately equal (averaging 4.5
days) during Phase 1, but during Phase 2 averaged about 25
percent higher in Battery C as a result of greater solids
wastage from Battery B.
With relatively few exceptions, aeration tank inlet DOs
equalled at least 2.0 gm/1 in both Batteries. Furthermore,
DO concentrations were invariably higher in the second, third,
and fourth passes of each tank, and were therefore adequate
for purposes of combined carbon oxidation-nitrification.
Since operating parameters in the two batteries generally
did not substantially differ, it was anticipated that NH.-N
removal efficiency would roughly correspond. However, parallel
performance was not observed, especially during Phase 2. As
noted, ammonia nitrogen removals during Phase 1 averaged 75
69
-------
TABLE 4
Operational and Analytical Parameters for Battery B
During Phase 1 and Phase 2
Flow Rate, mgd
HRT, firs
SRTf days
F/M Ratio (#BOD/|MLVSS/day)
NH4-N Loading (#NH4-N/#MLVSS/day)
DO Concentration, mg/1
(tank inlets)
NH.-N, mg/1 in
out
Total BOD, mg/1 in
out
Suspended Solids, mg/1 in
out
Phase 1
201
6.1
4.4
0.22
0.023
2.4
9.5
2.4
94
4
118
8
Phase 2
189
6.5
4.6
0.19
0.018
2.1
8.4
4.1
86
5
119
6
70
-------
TABLE 5
Operational and Analytical Parameters for Battery C
During Phase 1 and Phase 2
Flow Rate, mgd
HRT, hrs
SRT, days
F/M Ratio (|BOD/#MLVSS/day)
NH4-N Loading (#NH4~N/#MLVSS/day)
DO Concentration, mg/1
(tank inlets)
NH.-N, mg/1 in
out
Total BOD, mg/1 in
out
Suspended Solids r mg/1 in
out
Phase 1
195
6.3
4.5
0.20
0.019
2.4
9.0
3.2
94
5
122
7
Phase 2
189
6.
5.
0.
0.
2.
8.
1.
87
4
122
6
5
9
18
017
2
1
0
71
-------
percent in Battery B (2.4 rag/1 effluent NH.-N) and 64 percent
in Battery C (3.2 mg/1). During Phase 2, Battery C averaged
overall 88 percent removal (1.0 mg/1 effluent NH.-N) compared
with only 51 percent removal in Battery B (4.1 mg/1 NH.-N re-
mained) . In addition, although neither battery achieved a
high degree of nitrification during March when both SRTs and
aeration temperatures were lowest, Battery C recovered in
April with 85 percent removal as the average wastewater tem-
perature increased from the March low of 12°C to 15°C. Bat-
tery B, however, removed only 30 percent during this period.
The data suggest that with SRTs being about 20-25 percent
higher in Battery C, sufficient reserve nitrifying capacity
may have been provided to effect the higher removals of NH.-N.
Effluent BOD concentrations were consistently low in both
batteries, averaging 3-6 mg/1, throughout the nine months. Also,
suspended solids values typically equalled less than 10 mg/1,
and exceeded this level only during March in Battery B. Over-
all averages obtained during Phases 1 and 2 are shown in Tables
4_ and J5 for each battery.
Effect of Hydraulic Shock Loading on Nitrification Performance
During the period of February 23-March 11 shown in Figure 4,
an opportunity was presented to assess the effects of hydraulic
shock loading on nitrification in Battery D. The initial seven-
day period was characterized by influent sewage flows ranging
from 200-240 mgd, thus HRTs equalling 5-6 hours were obtained.
72
-------
FIGURE 4
EFFECT OF HYDRAULIC SHOCK LOADING ON
NITRIFICATION IN BATTERY D
400
300
200
1000
<90 800
20
O „ 600
-J O
mm
in oo
400
200
INFLUENT FLO
INFLUENT
SS LOADING
3.0
4.0
6.0
w
o:
o
15.0
01 10.0
£
5.0
0.0
INFLUENT NH4~N
EFFLUENT NH.-N
O
O
40
30
20
10
CO
NH.-N REMOVED
\
NH4-N APPLIED
24
26
28
II
FEBRUARY
MARCH
73
-------
However, during the next five days of March 1-5, daily rain-
fall averaged 0.7 inches per day and the subsequent stormwater
runoff produced flows averaging 345 mgd. This 1.5-fold increase
above the average dry weather flow reduced average HRTs to
about 3.5 hours. Following the cessation of storm flows by
March 6, HRTs returned to about 5 hours. Wastewater tempera-
tures ranged from 9-13°C over the eighteen-day period depicted.
The influx of suspended solids in the runoff impacted
Battery D in several ways. The higher SS loadings necessitated
higher solids wastage rates. Consequently, lower SRTs were
maintained during March 1-5 and March 6-11, equalling 4.4 and
5.7 days, respectively, compared with 7.2 days for the period
of February 23-29. Further, as a result of more inert solids
being carried into the system via the storm runoff, the vola-
tile fraction of MLSS decreased from 63.9 percent (2500 mg/1)
initially, to 61.2 percent (2100 mg/1) during the runoff period,
and finally averaged 56.7 percent (2200 mg/1) over the last
six days.
Even though there was considerable variation in influent
NH4~N values from February 23-March 5, effluent NH.-N concentra-
tions did not exceed 1.5 mg/1. Of particular interest is the
fact that effluent NH.-N values were consistently below 1.0 mg/1
during the period of storm runoff. However, as the storm flow
subsided, influent NH.-N concentrations progressively increased
over the next several days to about 2 to 3 times the March 1-5
74
-------
levels. The effluent NH.-N concentration also progressively
increased to 5.6 mg/1 on March 10. Thereafter, on March 11,
the influent NH.-N level declined, and a significantly lower
effluent NH.-N concentration was obtained (1.8 mg/1). Sub-
sequent data showed an effluent NH.-N concentration of less
than 1.0 mg/1.
Inspection of the daily NH.--N removals obtained in Bat-
tery D shows that 15,000-26,000 Ibs of NH.-N were removed per
day prior to the five days of high flows. Comparatively,
removals during the high flows of March 1-5 were in the
range of 12,000 to 20,000 Ibs per day. Since in both cases
the Ibs of NH.-N removed per day closely approximated the daily
loadings, the reduction in the Ibs of NH.-N removed during
March 1-5 illustrates the 20-25 percent lower average loading
rate observed during the latter five days. Thus, with NH.-N
loadings increasing substantially during March 6-10 to 15,000-
33,000 Ibs/day, the Ibs per day of NH.-N removed also increased
and roughly corresponded with the previously high levels
achieved during February 23-29. However, as mentioned, ef-
fluent NH.-N concentrations increased considerably.
In our opinion, these results indicate that relative to
the initial seven-day period, fewer nitrifiers were developed
in the Battery D aeration tanks during March 1-5 when NH.-N
loadings were substantially reduced. Moreover, independent of
the number of nitrifying organisms produced, fewer nitrifiers
were retained in the system during March 1 through March 11 due
75
-------
to the lower SRTs and MLVSS maintained. Therefore, with: (a)
lower concentrations of nitrifiers being present, (b) little
or no reserve capacity remaining under the existing operating
conditions, and (c) a slightly higher NH.-N loading rate during
March 6-10, it was not entirely unexpected that the effluent
breakthrough of NH.-N occurred following the hydraulic shock
loading.
Effect of SS Loading on the SS Produced and the Solids Reten-
tion Time
As noted in Figure 5, a fairly good correlation was shown
between the SS loading and the SS produced from both the South-
west primary effluent (Phase 1) and the West Side Imhoff ef-
fluent (Phase 2). Based on the SS loading and production data
obtained, a linear correlation coefficient equalling 0.97 and
0.81 was found during Phase 1 and Phase 2, respectively. Con-
versely, a poor correlation was observed between total BOD
loading and the suspended solids production during both phases.
In reporting on this, Obayashi et al(ll) found that the absence
of a good correlation with the BOD loading was due to most of
the total BOD being associated with the SS, with the BOD:SS
ratio in the influent to the aeration tanks being highly
variable. During dry weather the BOD:SS ratio equalled approxi-
mately 1:1, compared with 1:2 to 1:3 during storm flow periods
when SS inputs increased significantly.
On the average, the SS produced during Phase 1, per ton of
influent SS, was 33 percent lower than during Phase 2 (equal-
76
-------
FIGURE 5
THE EFFECT OP SUSPENDED SOLIDS LOADING ON THE SUSPENDED
SOLIDS PRODUCED IN BATTERY D DURING THE SINGLE
STA6E NITRIFICATION STUDY AT W-SW
5
O
z
o
O
bi
O
O
e
(L
CO
Q
J
O
(O
o
UJ
u
0.
«0
3
CO
150
• fu±
IOO
WESTS
PHASE II
!DE(IMHOFF)PR|MAF
y.l.i
' •
»
*+2.*
_^ _ i
O.5I
/
^
*\
IY EFFLUEN
x
\
/
^
V
/
/
PHASE 1
\ /
o
/
SOUTHWEST PRIMARY EFFLI
ys
r
0.80* -6. 5
2.0.97
f
\
/-
JENT
50 IOO
SUSPENDED SOLIDS LOADING, font/day
150
77
-------
ling 74 tons/ton versus 112 tons/ton). Thus, given the above
differences in solids production, and with the average influent
SS being 40 percent higher during Phase 2 (see Table 2), roughly
50 percent more SS were produced per mgd treated during Phase 2
than during Phase 1.
The differences in the solids produced during the two
phases were accounted for by the differences in the biodegrad-
ability of the solids as reflected in the lower unit BOD and
lower volatile content of the solids from the Imhoff tanks, as
opposed to the solids from the Southwest primary sedimentation
tanks. Suspended solids from the Imhoff tanks were more re-
fractory, and as a result relatively little auto-digestion oc-
curred in the aeration tanks when treating the Imhoff effluent
(11). Therefore, inasmuch as the amount of SS produced gen-
erally dictated the amount of SS wasted in order to maintain
a fairly constant MLSS concentration, this resulted in the
necessity of wasting more activated sludge from the system on
a daily basis during Phase 2 with Imhoff effluent as the in-
fluent source to Battery D. Consequently, as was pointed out
earlier, the SRT established during Phase 2 averaged 5.4 days,
compared with 9.8 days during Phase 1.
Effect of SRT on NH4-N Removal During the Winter Operation
The period from February 8 to March 20, 1976 comprised
the most critical of the study due to the fact that both low
aeration tank temperatures and low SRTs were obtained. Through-
78
-------
out this period temperatures averaged 12-13°C and were almost
as low (within 1°C) as those recorded in January, the coldest
month. Suspended solids loadings, however, were much higher
than those encountered in January, thereby producing signi-
ficantly lower SRTs ranging from 3.1 days in Battery B to 6.8
days in Battery D.
The BOD, NH.-N and hydraulic loadings to the four batteries
were similar and are summarized in Table 6. Note that the NH.-N
loadings ranged from 0.018 Ibs/lb MLVSS/day for Battery C to
0.023 Ibs/lb MLVSS/day for Battery D, reflecting the higher
NH4-N concentration of the Southwest primary effluent fed to
Battery D. On an average basis, the BOD to TKN ratios were
about 5, which could classify each battery as a combined car-
bon oxidation-nitrification facility(6). Approximately 5
hours detention time was achieved in all four batteries.
Table 7 summarizes the nitrification performance of Bat-
teries A, B, C and D during the critical period of 2/8 to
3/20/76, and shows that ammonia removal through Battery D
averaged 93 percent, with the effluent NH.-N concentration
averaging 0.7 mg/1. These data for Battery D remained consist-
ent with observations made under more favorable operating con-
ditions. In contrast, even though the average flows to Bat-
teries A, B and C were slightly lower, the NH.-N removals
ranged from only 41 percent in Battery A to 52 percent in Bat-
tery C. The average effluent NH4-N concentrations in Batteries
79
-------
TABLE 6
W-SW Operating Parameters During 2/8/76 to 3/20/76
(Phase 1)
Battery
A
B
C
D
NH4-N
0.020
0.020
0.018
0.023
Loading*
BOD
0.24
0.24
0.23
0.21
BOD/TKN
5.2
5,4
5.6
4.7
HRT
hrs
5.2
5.2
5.2
4.9
* Ibs/lb MLVSS/day
80
-------
TABLE 7
Nitrification Performance at W-SW During
2/8/76 to 3/20/76 (Phase 1)
Effluent NH.-N
Battery mg/1
A 5.2
B 4.5
C 3.7
D 0.7
%NH4-N SRT Inlet DO
Removal days mg/1
41 4.2 2.4
44 3.1 2.7
52 3.4 2.7
93 6.8 3.7
Aeration Tank Temperature - 12 to 13°C.
81
-------
A and B exceeded the future NH.-N standard of 4.0 mg/1, and
were 5.2 mg/1 and 4.5 mg/1, respectively.
With DO concentrations averaging greater than 2.0 mg/1
at the inlets to the aeration tanks, the most important factor
affecting the nitrification performance of the four batteries
appeared to be the SRT. As may be seen, the SRT in Battery D
was 6.8 days, which was approximately double the SRT in Bat-
teries B and C, and also considerably higher than the 4.2 day
SRT maintained in Battery A. Apparently, the SRTs of 3 to 4
days established in Batteries A, B and C were not adequate in
maintaining a substantial population of nitrifying bacteria,
and may have been close to the "wash out" or critical SRT. On
the other hand, the SRT maintained in Battery D compared with-
in 10 percent of the 7.3 day SRT established during the North
Side single-stage nitrification study (.2), in which successful
NH.-N removal occurred through the winter with aeration temper-
atures being as low as 11°C.
Figure 6 shows that relationship between NH.-N removal and
SRT. Weekly averages for SRT and effluent NH4~N were plotted
for Batteries A, B, C and D for the entire winter period of
December 1975 thru March 1976, during which time wastewater
temperatures ranged from 11 to 13°C. Due to fluctuations in
loading and flow, the data depicted in Figure 6 reflect non-
steady state conditions.
It may be noted that SRTs in Battery D were generally above
8 days, whereas SRTs in Batteries A, B and C were almost exclu-
82
-------
s
-------
sively less than 8 days, and typically below six days. As
shown, excellent nitrification (greater than 95 percent NH.-N
removal) was achieved at SRTs of 8 days and above. A relatively
high level of nitrification was also obtained in the range of 6
to 8 days SRT, with NH.-N removals being about 75 percent or
more. However, below a six-day SRT nitrification was quite vari-
able, and at 3-4 day SRTs the nitrification efficiency was con-
sistently low.
Based on these data, the critical or limiting SRT for nitri-
fying bacteria at 11-13°C was estimated to be approximately 2.5
days. This finding compares favorably with the estimates of the
limiting SRT at various temperatures by other investigators
(Table 8). Of particular note is the study by Sawyer et al(12)
in which estimates of the maximum growth rates of nitrifying
bacteria were made at four controlled temperatures, feeding
Imhoff effluent to once-thru pilot scale reactors. A good agree-
ment was observed for the results obtained at 10°C (3.0 days)
when compared to the estimated limiting SRT of 2.5 days at 12°C
for this study. In addition, the values obtained at 12°C by
Gujer and Jenkins(13), 2.5 days, and Knowles et al(14), 2.9 days
(from their predictive model) also compared equally well. The
predictive model developed by Knowles et al is of the following
y = 0.47e°-098(T-15)
m
Where: y = maximum growth rate of ,
nitrifying bacteria, day"
T = Temperature, °C
84
-------
00
W
9
EH
01
O
•P
id
tn
-H
-P
CO
>
fl
H
W
j3
o
••H
Vl
id
^1
Q
id
0)
4J
o
id
CQ
tn
C
•rl
«H
•H
lj
-p
•rl
JJ
4-1
O
EH
CO
C?
•H
-P
•H
e
•H
•o
0)
-P
id
c!
-3
-P
01
W
01
CD
o
CJ
Q)
rl
0)
is
rl
0)
-P
id
CD
4J
01
id
u
o
•
(X
EH
13
<0
-P
id
CO
•p
a
*>,
01
^i
id
•d
»
EH
5
CO
tn
C
•H
-P
i
•H
e
•H
•d
CD
-P
e
-•H
-P
01
w
n
01
OI
rH
in
J__l
"fl1
rH
(M
,_)
O
rH
00
vo
n — .
H m
^•^ *O ^^
01 -*
— C CP C
oi -H C £
rH ^ -H O
rH 0) £^ «
Id ra O rH
4J TJ -P
(DC *-P ' 0)
id CQ 0) O
rl 0) r4 C
0) Vl rH rl (U
>i 0) £ id M
> -r-i O ffl ^
flj 2 C fO
CO O « i-3
rl
0)
^
•H
rt
CD CD W CD
tji cn fl) CT>
id id g id
^ £ id £
0) d> 43 0)
CO CO EH CO
o
•
i-H
rH
rH
.
rH
00 O
• •
0 01
V
'^
•
rH
O) O\
r-H rH
n
•
01
in
•
OJ
o
•
n
00
m
OJ o
• •
T
rH
•
r~
vo
rH
01
^
CD
rl
•d
C
rtj
c3
id
j^
0}
•d
o
o
•H
-P
0)
4J
a
CO
<*
•
0
. .._l
rH
rl
_f*
0)
Q
,J
bd
g
id
01
id
id
rl
PL.
£>1
rl
rH
rj
O
CM
o
•
rH
K*1
'd
3
4J
01
CO
•H
EH
0)
tn
(d
]j
O
CO
in
•
01
85
-------
This equation appears to adequately represent most of the
values in Table 8 which are plotted in Figure 7 in terms of
ym (-1/SRT ^ ) and temperature.
86
-------
FIGURE 7
MAXIMUM GROWTH RATE OF NITRIFYING BACTERIA
AT VARIOUS TEMPERATURES
2.0
-8
O .4
or
.3
6
FROM KNOWLES ET AL(I4)
* .2
X
2
j ,
^^^
X
i i I I
1 I 1 1 |
8 10 12 14
TEMP ERATUR E
16
18
20
22
SAWYER ETAL (12) x
GUYER AND JENKINS (13) A
KNOWLES ET AL (14) o
PRAKASAM a LOEHR (17) •
THIS STUDY O
87
-------
DESIGN CONSIDERATIONS
Determination of the Design SRT
In our opinion, the importance of incorporating a reason-
able degree of flexibility (i.e., margin of safety) into the
design of a combined carbon oxidation-nitrification system was
clearly demonstrated by the results obtained during this study.
This requirement for an adequate safety factor cannot be over-
emphasized as the key to successfully maintaining nitrification
during periods of high strength waste loadings or sustained low
temperature conditions. Thus, in general, the major considera-
tions to be made in the determination of the safety factor, and
therefore in the selection of the design SRT, are with respect
to aeration temperature and waste strength, with DO and other
environmental factors assumed not to be limiting. Consequently,
since the maximum growth rate of the nitrifying bacteria is
primarily affected by temperature, determination of the design
SRT required in order to achieve the desired NH4~N removals
should be predicated on the lowest temperatures expected to be
encountered in the aeration tanks.
Based on historical data, the lowest sustained aeration
temperature likely to be experienced at W-SW is 10°C. As may
be recalled from Figure 6, Battery D successfully maintained
nitrification throughout the winter at temperatures ranging from
11-13°C (averaging 12°C), with SRTs being about 8-10 days. On
the other hand, Batteries A, B and C achieved only marginal
88
-------
nitrification during this period as average SRTs equalled 5.0,
4.4 and 4.5 days, respectively. Therefore, on the basis of
these results and the published literature previously cited
with regard to SRT and temperature considerations, it would
appear that a design SRT of 10 days at 10°C should be adequate
for treating the W-SW influent wastewater.
A 10-day design SRT for the expanded and improved W-SW
Plant also provides sufficient flexibility to accommodate
significant increases in waste loadings in terms of BOD, sus-
pended solids and NH.-N, which could otherwise result in de-
terioration of the effluent quality. In fact, NH.-N break-
throughs in response to transient increases in NH.-N loading
were recorded on several occasions in Batteries A, B and C,
being indicative of the generally insufficient nitrifying
populations maintained in these batteries. However, this
occurred on only one occasion in Battery D (documented in
Figure 4), and resulted in the future effluent NH ,-N daily
standard of 4.0 mg/1 being exceeded on two consecutive days.
As was pointed out, with the notable exception of the estab-
lished SRT, Battery D was maintained at roughly equivalent
operating conditions as the other three batteries.
With the anticipated implementation of the MSDGC's Tunnel
and Reservoir Plan (TARP) for pollution and flood control, it
is expected that hydraulic shock loadings to the W-SW Plant
will be substantially reduced but not completely eliminated.
89
-------
Estimation of Suspended Solids Production
As mentioned, it was revealed in a previous publication
by Obayashi et al(ll) that the suspended solids loading to the
(Battery D) aeration tanks, rather than the total 5-day BOD
applied, was more indicative of the suspended solids produc-
tion rate in the system. This observation was attributed to
the relatively low soluble BOD concentration (approximately 30
mg/1) which was found in both the West Side Imhoff primary ef-
fluent and the Southwest preliminary effluent, with most of
the total BOD being associated with the suspended solids. Fur-
ther, owing to the combined stormwater — domestic sewer system
in use in the MSDGC service area—the BOD to SS ratio was
highly variable, ranging from 1;1 during dry weather to about
1:2 or 3 during storm flow periods.
The report also indicated that the percent volatile con-
tent of the suspended solids significantly affected the sus-
pended solids production, with approximately 30-40 percent
less solids being produced per Ib of influent suspended solids
from the Southwest primary effluent (73 percent volatile) than
from the West Side Imhoff effluent (64 percent volatile). The
lower solids production obtained was accounted for by an over-
all greater destruction of the more easily biodegradable sus-
pended solids in the SW primary effluent, as evidenced by the
90
-------
volatile content being reduced from 70-75 percent to 60*-65
percent in the activated sludge process. In comparison, the
volatile content of the influent suspended solids and the
mixed liquor was about the same (60-65%), thus reflecting lit-
tle or no biological oxidation, when the primary effluent was
from the West Side,
Based on the average results of the study, a suspended
solids yield coefficient of 0.9 Ibs of SS produced per Ib of
influent SS was obtained and was subsequently used to estimate
future SS production at W-SW. This value (0.9 Ibs/lb) repre-
sents the average of Phases 1 and 2, and is in agreement with
earlier MSDGC studies conducted at both the North Side and
West-Southwest Plants.
Aeration Tank Volume Requirements for Single Stage Nitrification
At West-Southwest
Table 9 lists several design alternatives, along with the
basic assumptions on which the design aeration tank volume re-
quirements for single-stage nitrification at West-Southwest are
predicated. These assumptions reflect projected conditions at
the W-SW Plant for the 1990 design year, and were obtained from
the results of a solids study(7) conducted by the Engineering
Department projecting future solids loadings.
Following the completion of TARP, the winter design sewage
flow of 1315 mgd (1358 ragd-summer). should essentially constitute
the maximum flow to the aeration tanks. Therefore, with the
(assumed) average influent suspended solids concentration to the
91
-------
TABLE 9
An Evaluation of Various Design Parameters for
Nitrification at West-Southwest
Assumptions:
1. design flow (winter) = 1315 MGD
2. average influent TSS = 105 mg/1
3. solids yield coefficient = 0.9 Ibs/lb inf. TSS
4. solids recycle rate = 50 percent
5. design flow (summer) = 1358 MGD
6. design final tank surface settling rate = 800 gpd/sq ft
SRT, MLSS, HRT, Solids Loading Rate to
days mg/1 Required Volume, MG hrs Clarifiers, Ibs/ftVday
10 2500 493 9.0 25.0
3000 411 7.5 30.0
3500 352 6.4 35.0
9 2500 444 8.1 25.0
3000 370 6.8 30.0
3500 317 5.8 35.0
8 2500 394 7.2 25.0
3000 329 6.0 30.0
3500 282 5.1 35.0
7 2500 345 6.3 25.0
3000 288 5.2 30.0
3500 246 4.5 35.0
6 2500 296 5.4 25.0
3000 247 4.5 30.0
3500 211 3.8 35.0
92
-------
aeration tanks being 105 mg/1, the anticipated design solids
loadings should equal about 575 tons/day. Accordingly, given
a solids yield coefficient of 0.9 Ibs of solids produced per
Ib of influent suspended solids, the daily quantity of sus-
pended solids to be wasted would amount to 518 tons. Thus, in
order to achieve the recommended design SRT of 10 days, the
solids inventory (mixed liquor suspended solids under aeration)
would have to be equivalent to approximately 5180 tons. Prac-
tically speaking, this requirement can only be met by sub-
stantially increasing the aeration volume above the present
204 million gallon capacity. Consequently, by maintaining MLSS
at approximately the 3,000 mg/1 level as recommended, although
operation at higher MLSS concentrations would not be precluded,
411 million gallons of total aeration volume will be needed.
The resulting design HRT at 411 MG of aeration volume is
7.5 hours (with the design sewage flow being 1315 mgd). Al-
though this design value exceeds the average HRT of 5.5 hours
that was observed in Battery D during Phase 1 at a comparable
SRT of 10 days, this can be attributed to the relatively low
solids yield coefficient of 0.72 Ibs/lb obtained during this
period while treating Southwest preliminary effluent. In
other words, if a higher solids yield coefficient had been
obtained during Phase 1, for example 0.9 Ibs/lb, then Battery D
could not have operated at 10 days SRT due to higher solids
wastage requirements. Given this particular situation, provid-
-------
ing a larger aeration volume would consequently permit main-
taining the desired 10 day SRT. However, the corresponding
HRT would then be increased in proportion to the increase in
the aeration volume.
Summarizing the above, the required aeration volume needed
to achieve desired levels of nitrification and carbon oxidation
in the future West-Southwest Plant is 411 million gallons, an
expansion of approximately 207 million gallons over the present
aeration capacity of 204 MG. As discussed and shown in Table 9,
the required volume is based on the following design criteria:
1. SRT = 10 days
2. MLSS = 3>000 ing/1
3. Suspended solids produced = 0.9 Ibs SS produced
Ib influent SS
4. Sewage flow = 1315 ragd
5. 1990 influent suspended solids = 105 mg/1.
It should be clear that the success of the nitrification
design depends on the existing older facilities (Batteries A,
B and C) being sufficiently rehabilitated, as planned, to the
extent necessary to affect carbon oxidation and NH.-N removal
consistent with the demonstrated performance of Battery D.
Further, the scheduled rehabilitation of the West Side Imhoff
tanks must also be completed.
Estimated Construction Costs
The estimated construction costs for implementing the
indicated expansion and improvements to the West-Southwest
94
-------
Plant are about $475 million, based on an Engineering News
Record (ENR) Chicago area January, 1980 construction cost
index of 3300(18). Included in this cost estimate are approx-
imately $61.3 million for a new blower facility.
95
-------
References
1. "Rules and Regulations of the Illinois Pollution Control
Board," Chapter 3, Water Pollution, Rule 406, 1972.
2. Sawyer, B., A.W. Obayashi, and C. Lue-Hing, "Pull-Scale
Single Stage Nitrification Study at the North Side Sewage
Treatment Plant," MSDGC Research- and Development Report,
75-26, October, 1975.
3. Washington, B., A.W. Obayashi, C. Lue-Hing, and D.R. Zenz,
"Single Stage Nitrification Study at the West-Southwest
Treatment Plant," MSDGC Research and Development Report,
76-2, November, 1975.
4. Prakasam, T.B.S., C. Lue-Hing, E. Bogusch, and D.R. Zenz,
"Pilot-Scale Studies of Single-stage Nitrification," Jour.
Wat. Poll. Con. Fed., Vol. 51, p. 1904 (1979).
5. Lawrence, A.W., and P.L. McCarty, "Unified Basis for
Biological Treatment Design and Operation," J. Sanitary
Engr. Div. Amer. Soc. of Civil Engr., Vol. 96, p. 757
(1970).
6. "Process Design Manual for Nitrogen Control," U.S. EPA, Of-
fice of Technology Transfer, Washington, D.C., October,
1975.
7. "Design Criteria, Expansion and Improvement, West-Southwest
Sewage Treatment Works, Rev. No. 4," Department of
Engineering, Metropolitan Sanitary District of Greater
Chicago, June, 1975.
8. Beckman, W.J., et al, "Combined Carbon Oxidation Nitrifica-
tion," Jour. Wat. Poll. Con. Fed., Vol. 44 p. 1916 (1972).
9. Smith, J.I., "Investigation of a Rapid Method for Sludge
Solids Estimation," Sewage Works J., Vol. 6, p. 908 (1934).
10. "Standard Methods for the Examination of Water and Wastewater,"
American Public Health Assn., Inc., 13th Ed., New York, N.Y.(1971)
11. Obayashi, A.W., B. Washington, and C. Lue-Hing, "Net Sludge
Yields Obtained During Single Stage Nitrification Studies
at Chicago's West-Southwest Treatment Plant," Proc. of the
32nd Annual Purdue Industrial Waste Conference, p. 759
(1978).
12. Sawyer, B., A.W. Obayashi, C, Lue-Hing and D,R, Zenz
"Estimation of the Maximum Growth Rate of Ammonia Oxidizing
Nitrifying Bacteria Growing in Municipal Wastewater," Paper
presented at the 52nd Annual WPCF Conference, Houston, Tex.,
October, 1979.
96
-------
References (Cont'd)
13. Gujer, W., and D. Jenkins, "The Contact Stabilization Pro-
cess-Oxygen and Nitrogen Mass Balances," San-Engr. Res.
Lab. Rept No. 74-2, Univ. of Calif. Berkeley (1974).
14. Knowles G., A.L. Downing, and M.J. Barrett, "Determination
of Kinetic Constants for Nitrifying Bacteria in Mixed
Culture, with the Aid of an Electronic Computer," J. Gen
Microbiology, Vol. 38, p. 263 (1965).
15. Lawrence, A.W., and C.G. Brown, "Design and Control of
Nitrifying Activated Sludge Systems," J. Water Poll. Con.
Fed., Vol. 48, p. 1779 (1976).
16. Poduska, R.A., and J.F. Andrews, "Dynamics of Nitrification
in the Activated Sludge Process," Jour. Water Poll. Con.
Fed., Vol. 47, p. 2599 (1975).
17. Prakasam, T.B.S., and R.C. Loehr, "Microbial Nitrificaiton
and Denitrification in Concentrated Wastes," Water Res.,
Vol. 6, p. 859 (1972).
18. "Master Design Program for Treatment Facilities," Department
of Engineering, Metropolitan Sanitary District of Greater
Chicago, January, 1980.
97
-------
PHOSPHORUS REMOVAL WITH IRON SALTS AT BLUE PLAINS
Edgar R. Jones, P.E.
Chief Process Engineer
Bureau of Wastewater Treatment
District of Columbia Government
Washington, B.C.
BACKGROUND AND INTRODUCTION
The District of Columbia's Wastewater Treatment Plant at Blue Plains
is a regional treatment plant located along the Potomac Estuary in the
nation's capital. (See Figure 1.) The plant's service area is approximately
725 square miles with a population equivalent of 2,200,000. An average
flow of 330 mgd of medium strength sewage from primarily residential and
office complexes is treated daily. Flows from the District of Columbia
and Maryland account for 95% of the flows received with the balance coming
from Virginia. (See Figure 2.) The wastewater treatment scheme
incorporates primary sedimentation followed by a modified aeration activated
sludge process (See Figure 3.). On-going expansion will add nitrification
and multi-media filtration to the wastewater train in the early 1980's.
Sludge processing operations include gravity and flotation thickening,
anaerabic digestion and elutriation, vacuum filtration and sludge disposal
by an amalgram of composting, trenching, incineration and land spreading
methods. (See Figure 4.)
Discharges to the Potomac Estuary are regulated by an NPDES permit and an
Order of Compliance. Effluent quality criteria is designed to enhance the
water quality in the Estuary. Maximum pound loadings for various pollutants
under various flow conditions in the Estuary were developed by EPA in 1971
(1). In their study, EPA noted a 12-fold and 9 -fold increase in phosphorus
and nitrogen loadings to the estuary from 1913 to 1970. The algae population
98
-------
\
-ANACOSTU fflVtlt
UXATKM tut
6UN1TON COVC
OCCOOUAM »AV'
LEGEND
• MAJOR WASTE TREATMENT PLANTS
A GAGING STATION - WASHINGTON. O.C.
A DISTRICT Or COLUMBIA
B ARLINGTON COUNTY
C ALEXANDRIA SANITATION AUTHORITY
D FAIRFAX COUNTY - WESTGATE PLANT
E FAIRFAX COUNTY - LITTLE HUNTING CREEK PLANT
F FAIRFAX COUNTY - OOGUC CREEK PLANT
G WASHINGTON SUBURBAN SANITARY COMMISSION - PISCATAWAY
H ANDREWS AIR FORCE BASE - PLANTS ONE. FOUR
I FORT BELVOIR - PLANTS ONE. TWO
J PENTAGON
K FAIRFAX COUNTY - LOWER POTOMAC PLANT
Source - EPA TR 35, Ref. 1
CHCSAPfAff
ICMJL M ULO
Figure 1. Potomac Estuary
99
-------
g
CD
-P
(0 >t
0) CO
-------
9 <
Z to
< P
>i
cr ;=
SECONDA
LARIFICAT
o
; i
cr z
< 0
P
R *
S HJ
LLJ ^*
CO <
i \
z
o
!_ .__
PRIMARY
EDIMENTA1
(O
CC
LU
CD
0 2
LJ <
H X
< 0
tr
<£
cr
o
VACUUM
FILTRATIO
i \
z
o
h-
<
o;
1-
o
_J
Ld
O
pi
O
Z
to i
ocro
Q.OZ
2
o
o
'• t
-^
DIGESTIOf
2
ID
Z3
O
5
" j
o
tu -
0 Z
Q LiJ
z> *
Ij o
to x
1
r—
111
cr
H
f
Z
FILTRATIO
i
en
o
rfl
tn
c
•H
-P
03
•H
X
H
C
•H
fO
(U
D
rH
CQ
oo
0)
.J
3
CT>
-i-t
t,
f
101
-------
C
TRATION
A
_J
U.
j
|
<
C
U
^M
K
2
J
>
Q
-?
C
2
C
«.
U
4
n
4
{
4
i
(
/
5
CC
NFECTION
U)
Q
ft
»
i
•
C
r
»
»
r
i
»
«
f
5
>
)
j
•*
r
>•
r
=x
r
L
>
IWASTEWATER
Krf
-°tfi
ze>£ .-
w Z (/)
*I5
O o
0-ZO
U H_J
i k
— V7 -r
' 2 g ^ o
§ 3 _, = 5
fe g n 3 E
u. x ^ —
1 i
in
» *~" Z „ . ' L > Q
• > g * i
2 o 3
tD X ®
h-
01 OL
< M
5
i
g
L^
o:
tu
z
Q"
K8
^S
ii n
•^ u.
o
ns
a;
M
D
4J
3
C
•1-1
IX)
0)
a
iH
CQ
^
OJ
3
cn
•H
fa
102
-------
became unbalanced with dcminance by blue-greeen algae coincident with
the phosphorus and nitrogen increases. Since wasta^ater loadings increased
from 42 mgd to 325 mgd during that same period, EPA concluded that
unfavorable ecological changes were due to the phosphorus and nitrogen
loadings fron wastewater treatment plant discharges.
To maintain algal standing crops below nuisance levels under severe
summer conditions, EPA concluded that phosphorus concentrations should
be limited to 0.03 to 0.1 mg/L as P in the estuary. Similar lijnitations
were imposed on nitrogen concentrations. Consequently, discharge limitations
for Blue Plains were formulated that would effect the necessary nutrient
restrictions in the Estuary. Table 1 summarizes Blue Plains NPDES
limitations along with Order of Compliance iterim requirements staged to
coincide with construction events.
To meet the stringent effluent quality criteria established for the Blue
Plains wastewater treatment facility, numerous unit processes were reviewed
in an extensive pilot plant program. The research effort was carried out
jointly by the District of Columbia Government and the EPA. Various process
combinations were compared in terms of process reliability, relative costs,
land requirements, chemical availability and treatment capabilities of
existing facilities during and after construction. The process scheme
selected utilized the existing modified aeration activated sludge process
for secondary treatment to conserve limited site aera and to produce an
effluent compatible with the subsequent biological nitrification-denitrification
process. Denitrification has been held in abeyance pending additional
estuary studies. Alum or ferric chloride is designed for phosphorus removal
in the modified aeration system.
103
-------
in
0)
tf
H
vo
o
ro
o
n
I 1
(0
E-i
0)
T3
(0
-U
•H
g
W
Q
d.
(0
Q
I
O
ro
H
E-i
M
2
H
o
ro
ro
I I
I I
ro ro
\ \
o ro
rH
I I
O
•H
4J
(0
O
•H
IM
•H
i-i
4J
•H
S
(0
C
o
CO
m
m
Q
O
CO
W
CO
(0
4J
o
E-i
*» rH
104
-------
Iron salts, either FeCl., or FeS04 are used at Blue Plains because of
costs and local availability. Both chemicals are industrial waste by-
products from major sources within 150 miles of the plant. FeCl., is added
prior to secondary clarification. When used in lieu of FeCL,, FeSO. is
added in the aeration basin for ferrous iron to ferric iron conversion by
oxidation. Anionic polymers are added with the iron salts to improve
the settling characteristics of the mixed liquor suspended solids.
The purpose of this paper is to present on analysis of Blue Plains operating
data showing phosphorus removal efficiencies, chemical sludge quantities
and costs related to phosphorus removal. Full-scale process performance
data will be compared to pilot plant data, plant design criteria and NPDES
and compliance order requirements.
CHEMICAL ADDITION IN SECONDARY
During the initial compliance period, effluent limitations were met by
adding 35 mg/1 of FeCl^ to half of the secondary process. The 35 mg/1
was a criteria established in the pilot plant studies. The 35 mg/1 dose
was quickly recognized as too sludge intensive for full plant application.
A polyelectrolyte was substituted for part of the metal salt dosage. Full
plant addition of 25 mg/1 FeCL, and 0.3 mg/1 of an anionic polymer minimized
the amount of chemical sludge generated and achieved the desired phosphorus,
BOD and suspended solids removals during the Interim I period.
Once Region III EPA recognized phosphorus discharges were as low as 1.6
mg/1 routinely, the Order of Compliance was revised to restrict phosphorus
discharges to that lower level. Table 2 summarizes plant performances under
three degrees of compliance.
105
-------
CN
0)
(0
D
4-1
U
(U
o
C
ft}
a,
o
(0
g
g
D
CO
0
o
C
(0
o
U-l
S-i
(U
ft
4-J
C
03
rH
ft
o
•H
^1
QJ
ft
0
O
C5
fd
•rH
rH
a>
g
0
u
CO
•H
> o
0) oo
Oi r-l V£>
--' (0 00
H S oo
g ON
•H r- i
Q) 3
-P rfj
C
H
H C*
r-
g H 0
•H 3 ON
r-l l"3 CM
CD 1 X.
.p VO ON
C r~ o
H 4J 00
D
o
VD
r-i r-
tO CX O
•rH
g
«.
U
O
r-l
fn
rH
Cn
g
».
in
Q
O
CQ
rH
\^
Cn
g
^
CO
CO
rH
"s^
tn
g
«.
ft
r-H
X
g
ro
r-l
u
0)
106
-------
A decline in phosphorus influxes to Blue Plains has reduced chemical
demands below original projections. Phosphorus reduction trends are
shown in Figure 5. Better than expected phosphorus discharges are
attributable to the lower phosphorus influxes.
Rate of phosphorus insolubilization by FeCl., was separated from the
biological phosphorus uptake and quantified. Figure 6 is a plot of
phosphorus removal in the secondary process as a function of FeCl,
addition. Each point plotted represents a monthly average as summarized
in Appendix A. The slope of the line of best fit is 0.0926' pounds of
phosphorus removed per pound of FeCl, added. Inversely, 10.8 pounds of
FeCl, is required to remove one pound of phosphorus. The Fe/P molar
ratio was 1.8. The Y - intercept, 2.1 mg/1, represents the biological
phosphorus uptake occurring in the activated sludge process. The phosphorus
to volatile suspended solids ratio (P/VSS) in the secondary waste sludge
was 0.029 and is consistant with the P/VSS ratio of 0.03 calculated for
the waste sludge in 1970 and 1971 when FeCl, was not added.
By far, the most alarming problem associated with phosphorus removal with
iron salts is the inability of feed systems to resist the process chemical.
Blue Plains has experienced over 30 failures in sections of rubber lined
pipes and fittings. Those failures have occurred in suction, discharge,
and transmission lines. On one occasion, an 8-inch PVC transmission
line burst sending fragments of PVC flying 50 feet in all directions and
spilling approximately 3000 gallons of FeCl.,, in addition, no less than
20 FeCl, measuring cylinder heads in the metering pumps have failed
spilling FeCl, in the pump vicinity. As a result of the many failures- in
various system components, spilled FeCl, has seriously injured three
107
-------
tr
<
LU
a:
<
Q
LU
<
O
A
?
ID
k
fi
}
IJJ
^
-------
n
H
0
OJ
JC
4J
w
w
0)
o
o
Vj
0.
c
o
(0 tJ
N C
•H 0
rH O
12
3
iH
O
en
c
03
-P
co
-u r-
W CTi
03 H
W
W 0)
C XI
o
o
o
0) I
10 3
3 M l-~
^-i OQ r^
O o>
£ 0) iH
to -u
-------
employees and caused extensive damage to the equipment.
Faulty rubber lined pipe is normally replaced by PVC or fiberglass pipe
as ruptures occur. The permanent fix on the measuring cylinders was to
replace the kynar coated cylinders with solid polypropylene ones. Slight
microscopic imperfections in the egg-shell thin kynar coating were suspected,
allowing FeC3-3 to attack the aluminum core.
Since effluent phosphorus concentrations are well within permit limitations,
the phosphorus analyzers installed to trim the metering pumps are not
utilized. As tolerances decrease, the analyzers will be put on-line to
support the flow pacing mechanising built into the metering pumps.
SOLIDS GENERATION WITH FeCL,
One adverse consequence of chemical addition for phosphorus removal is the
increased sludge mass. Besides the chemical precipitates of FePO. and
Fe(OH)3 complexes, improved suspended solids captures add to the amount
of waste sludge requiring disposal. Data and procedures predicting
quantities of chemical sludge as a function of chemical added or phosphorus
removed are scarce. In their design manual for phosphorus removal (2),
EPA estimates solids generation rates using simple stoichiometric relationships
and then allowing for extra sludge by multipling the stoichiometric result
times a 35% safety factor, i.e. 1.35 multiplier. The following analyses
of Blue Plains data will support at least a 35% safety factor.
The observed chemical solids generation rate was 1.12 pounds of solids
per pound of FeCl-. added. The major device used to measure the chemical
sludge quantities generated was a mass balance around the point of chemical
addition. As a check, mass balances with and without chemical addition
110
-------
Mass balances were performed for four, two-year periods as summarized in
Table 3. Background data without chemical addition was codified for the
period between January 1969 through December 1970. Mass balances during
a two year period of Alum and FeCl., trials were reduced in a mass balance
for January 1972 through December 1973. Another mass balance for the period
between January 1975 through December 1976 was produced for the period where
reduced FeCl- dosages were tolerated while a polymer was used to aid MLSS's
settleability. The mass balance for January 1977 through December 1978
reveals the effect of heavy recycled solids fron gravity thickening. The
mass balance for each period is enclosed in Appendix B.
Figure 7 is a bar graph of total waste activated solids for the four
periods identified in Table 3. Chemical solids (ChemS) are shown
in perspective with biological waste activated solids (BWAS). BWAS equals
inert waste activated solids (IWAS) plus volatile activated solids (VWAS).
Bar C shows the pronounced increase of inert solids due to chemical solids
while Bar D reflects an increase in solids loading/casting effects due
to the plant recycled solids situation.
FeCl., addition appears to have little effect on the biology in the high
rate secondary process at Blue Plains. The biological solids production
factors were relatively consistent before and after FeCl., addition. Table
4 summarizes the observed solids yields. Also, as shown in Table 5, the
%VS (BWAS) data provides additional credence to the biological growth factors
and the mass balances. Table 6 presents Blue Plains chemical solids production
factors for both FeCl and Alum.
Ill
-------
a
UJ
1
O
CO
<
>
CO
I
CO
o
o
01
< CO
u g
5 —i
LU O
I CO
CO
e
_J
O
O O O O
(U 0)
CO
(0
S
0)
Cn
r^
0)
D
Ava/sai eoi 'sanos
3isvM
112
-------
n
(0
EH
C
O
• H
-P
•r-t
TJ
ti
03
O
x;
u
LU
O
CO Z
CO
C0>
< LL LL Q- LL CL
CO
O
LU
G
IT)
00
O
LU
G
HI
^
O
O
<
a:
o
o
cc
LL
C/D G
C/5
O —
> Q
O LU
LU O
LU
> E
> LU
< 0-
LU X
X LU
113
-------
4H
C
£_|
P*
CO
tn -H
>i r- cn o r~ co
03 "3- ^ m r- co
O • • • *
0 O 0 0 Xl
r<
CO
CO
H
cn
XI
tn
rH
cu
•H
>H
x:
5
0
o
"3*
rH
(U (0
rH 0
JQ --H u}
(0 CP pi
EH 0 o
O tj
•H ^j*
CQ r
^^ 1 T i
T3
CD
>
/i i
rH
%
T3
in 1
tn
T3
•rl
O
in
-P
OJ
in ro m oo C
CN r~ co co co -H
v^ • • • •
O O O O 'IS
i
tn
T3
•H
rH
0
tn
T3
0)
i_i
j
u
u
P4 co
CO <
tn
J2
O
CN
-------
in
(0
O
CO
0)
LU LU
f- O
CO Q
o <
o >
CQ <
CO
CM O
• •
O r-
00 CO
LO
CM
CO
CM
CO
c
0)
CU
D
cn
0)
— i
•H
(0
•H
O
Jj
o
D
O1
•H
J
T3
0)
X
•H
s
UJ CO
^^ ^^
^Q^
5 LU 0s
jHjui
-i 5 o
< 2: Q
H- H- D
OOlj
h- < CO
CM CO
O LO
CO fN
P
^211j
rt) — J Cj
LL < G
CM
0 S
o
CO
CM
CO
CO
CM
CM
CO
LO
CO
LO
CM
G
O
CC
LU
a.
CO O
CO
2 _
£ 2
£ s
F 5
< CO -J
CC LU 13
UJ O «
^§
LU
CC
cc
CO
CO
O LU
-J LU
LL LL
LU LL C
3^3
CO I LL <
• •
r- CM
115
-------
C/5
VO
S-i
O
4->
U
(0
fej
c
o
•H
-p
u
3
Cp
cn
O
O
<
u.
UJ
O
Q
D
_l
CO
CO
CD
O
<
o
o
03
in
•
o
CO
O
uu
U5
CO
CO
03
m
(0
u
•H
E
0)
JC
u
ID ——
CD }- uj ^
< Z « O
O < O ^
O -i Q ^
in
in
II
0 ^
0 ^
ALU
UJ
u.
_cp
o
0)
116
-------
were compared for similarity of non-chemical related parameters. Following
is a recap of the generalized solids generation formulas used in the
Blue Plains secondary process:
(1) TWAS = BWAS + ChemS = WASTE + SSeff
(2) BWAS = IWAS + VWAS
(3) IWAS = kx x SSinf
(4) 1/SRT = k2 x F/M
(5) VWAS = k3 x BCD
(6) ChemS = k4 x FeCl3
(7) F = BCD . _ - BCD ff + 0.68 x 1.42 x VSS .,
inf eff eff
(8) % VS (TWAS) = VWAS x 100 = %MLVSS
TWAS
(9) %VS (BWAS) = VWAS x 100
BWAS
Where:
TWAS = Total waste activated solids
BWAS = Biological waste activated solids
ChemS = Chemical waste solids
IWAS = Inert waste activated solids
VWAS = Volatile waste activated solids
WASTE = Waste secondary sludge
SS. _ = Suspended solids in secondary influent
BC^T"- = BOD5 in secondary influent
SS ^suspended solids in secondary effluent
VSS __ = Volatile suspended solids in secondary effluent
BCD6 = BOD5 in secondary effluent
FeCl- = FeCl., added in secondary
SET = Solids residence time
F/M = Food to mass ratio
F = BODj. insolubilized in secondary
BODr= BOD5 removed in secondary
M = Mixed Liquor volatile suspended solids mass
k,, k«, k, k. = Solids production factors derived
from actual operating data
%MLVSS = Per cent mixed liquor volatile suspended solids
117
-------
The stoichiometric factor can range fron 0.94 to 0.66 Ibs solids/Ib
FeCl- as the Fe/P molar ratio varies from zero to infinity. An adjusted
estimated range of values for the solids/FeCl, ratio using the EPA
multiplier of 1.35 is 1.27 to 0.89. The 1.12 Ibs chemS/lb FeCl3 factor
correlates well with EPA estimates.
COSTS
Recent O&M costs for the past three fiscal years (FY 77-79) are presented
in Appendix C. FeCl3 requirements in the period were 58,000 Ibs/day for
phosphrous removal. To properly identify the costs associated with
phosphorus removal, the cost for chemicals must be added to the extra cost
for handling and disposal of the chemical sludge. For the three year period,
FeCL, costs were 6.8C/0±>- Raw sludge disposal costs by trenching were $35/ton.
Filter cake production of chemS's as a function of phosphorus assuming
chemical condition requirements of 8% FeCl_ and 27% lime on a vacuum filter
producing a 22% cake are as follows:
1.12 x 1.35 x 58,000 _ 21? wet tons
0.22 x 2000 Day
Ignoring the chemical conditioning costs and combining the two major cost
items, the average annual cost for phosphorus removal during the past three
years was as follows:
ITEM Daily Quantity Annual Cost
$ ID/tear
FeCl3 58,000 Ibs/Day $1.44
Raw Sludge 217 wet ton/day 2.77
$4.21
Since the phosphorus removed chemically in the past three years was
0.0926 x 58,000 = 5370 lbs/t»ay, the O&M cost to remove phosphorus chemically
was greater than two dollars per pound, i.e. 4,210,000/365 = $2.15/1£> P .
5370
Figure 8 shows the O&M costs for
FeCl and sludge disposal as a function of phosphorus removed.
118
-------
'S1SOD W 3 O
o
o
o
in
(N
o
o
o
o
CN
o
o
o
in
o
o
o
o
o
o
o
in
o
o
o
o
o
o
o
o
fS
o
o
o
O
in
Q
LJJ
LU
o
§
CO
O
X
O Q.
o
O u_
10 O
C/)
D
Z
D
§2
o
o
u>
O
LU
>
O
2
LU
£
(/)
D
a:
O
i
a.
(S>
O
LU
U
l±
LU
Q.
lo
' ro
(0
>
0)
w
d
M
o
to
o
si
CO
EH
CO
O
u
O
•
oo
o;
o
o
o
(N
tN
in
o
Q3ACMI3H d qi/$ 'S±SOD
119
-------
Utilizing FeS04 will reduce the costs by a third since FeSO. will be
supplied at no costs.
RECOMMENDATIONS
Following are suggested recommendations for future investigations:
1) To reduce sludge requirements, biological phosphorus
removal mechanisms should be further evaluated to reduce
chemical requirements and sludge generation in the phosphorus
removal process.
2) Where commercial grade chemicals are being used for phosphorus
removal, cleaper industrial waste by-products should be examined
for possible use.
CONCLUSION
In the modified aeration activated sludge process at Blue Plains, FeClo
is added to remove phosphorus and enchance SS's and BODj. removal efficiencies.
Concern over increasing solids generation rates due to chemical sludge
resulted in a substitution of polymer for a portion of the FeCl.^ dose. A
significant quantity of costly sludge production has been avoided at a
substantial saving. One pound of FeClo creates 1.12 pounds of chemical
sludge but only removes 0.0926 pounds of phosphorus in the process. Cost
to remove phosphorus with FeCl,. is over two dollars per pound. Eventual
O&M costs to remove phosphorus will drop once FeSO. is used since FeS04
will be provided free of charge.
BIBLIOGRAPHY
1. Jaworski, N.A., Leo J. Clark, and Kenneth D. Feigner, "A Water
Resource - Water Supply Study of the Potomac Estuary,"
Technical Report 35, April 1971.
120
-------
2. "Process Design Manual for Phosphorus Removal," USEPA,
Technology Transfer, April 19-16.
121
-------
APPENDIX A
INSOLUBILIZATION OF PHOSPHORUS IN THE EAST
SECONDARY PROCESS AT BLUE PLAINS
PHOSPHORUS CONCENTRATIONS, MG/L
MONTH
JUN 77
JUL 77
AUG 77
SEPT 77
OCT77
NOV 77
DEC 77
JAN 78
FEE 78
MAR 78
APR 78
MAY 78
JUN 78
JUL 78
AUG 78
SEPT 78
OCT78
MEAN
STAN-
DARD DEV
0-IN
MGD
173
173
193
215
168
172
199
202
174
193
201
213
214
228
243
231
207
200
B
FeCI3
KIP/DAY
40.7
38.5
44.2
47.8
47.4
44.3
49.8
40.0
39.5
46.0
38.2
31.2
35.1
40.0
40.8
42.1
36.2
41.9
MG/L
28.2
26.7
27.5
26.7
33.8
30.9
30.0
23.7
27.2
28.6
22.8
17.6
19.7
21.0
20.1
21.9
21.0
25.1
4.56
TOTAL
PIN
(A)
5.7
6.0
5.3
5.5
5.9
5.4
4.7
4.6
5.3
5.2
5.1
4.8
4.7
5.0
4.9
4.6
5.2
5.2
0.45
TOTAL
POUT
(B)
1.7
2.2
1.9
1.9
1.6
2.0
1.3
1.4
1.9
1.2
1.5
1.8
1.9
2.1
1.7
1.7
2.0
1.8
0.28
DIS-
SOLVED
POUT
(0
0.9
1.1
0.9
0.8
0.5
0.7
0.3
0.3
0.5
0.3
0.8
1.0
1.0
1.1
1.0
0.7
0.9
0.76
0.273
INSOLU-
BILIZED
P
(A-C)
4.8
4.9
4.4
4.7
5.4
4.7
4.3
4.3
4.8
4.9
4.3
3.8
3.7
3.9
3.9
3.9
4.3
4.42
0.492
WASTE VSS
KIPS/DAY
103
90
138
125
109
99
112
123
101
134
145
136
109
115
110
128
176
121
NOTES: BIOLOGICAL P UPTAKE: P/VSS RATIO
1) 2.1 * 8.35 • 200/121 - 0.029
2) W/O FeCI3 (CY70& 71) « 0.029 TO 003
122
-------
APPENDIX B
SECONDARY SOLIDS PRODUCTION MASS BALANCE FOR CY
nASS BnLAuCE "A "
7O
225
BOD
,520
N
263 MGD Secondary Influent
mg/1
kips/day
Kips/day | - %VS~
Reactor Loading
/2/7_Kips/day_| _
3 A O""'MGD 'Re turn
Secondary Reactors
MLSS = 60? mg/1
Reactor Vol.=25 MG
5"O Kips/day Net Growth
/ V 92 KipsTday ] gq2 %VS
Reactor 'Effluent
SRT=
P/H=
Days
/Day
Secondary Clarifiers
Surf. Area =
Clar. Vol. =2O MG
/W Kips/day |
MGD Pischarge
/63 Kips/day
/. 6 2 MGD
5VS
Waste
SS
BOD
//Z
Chemical
_FeClq
Polyner
kips/day
N
262 MGD Secondary Effluent
ng/1
kips/day
1) TV/AS = Total Waste Act. Solids = Waste + SSeff = 275 Kips/Day
TV/AS = Biological Waste Act. Solids (BWAS) + Chemical Solids (ChemS)
ChemS = 1.12 * FeCl = Q Kips/Day
BWAS = TWAS - ChemS =_^2j7£" Kips/Day
BWAS = Volatile Solids (VWAS) + Inert Solids (IWAS)
VWAS = %MLVSS * TWAS/100 = 22 / Kips/Day
IWAS = BWAS - VWAS
IWAS/SSlnf =
ki •
Kips/Day
VWAS/BWAS =
O,8'02
2) 1/SRT = k
2
VWAS/F
F/M
0.75
SRT = Solids Residence Time = M/WAS
k0 = Observed Yield F/M = Food to Mass Ratio
VWAS/BODR
/. 05
F =
M =
BOD
inf
- BOD
VA *MLVSS
pff
8.35
0.68 * 1.JI2 * VSS
pff
M =
123
Kips/Day
Kips
-------
SECONDARY SOLIDS PRODUCTION MASS BALANCE FOR CY 72 i ~13
MASS
Kips/day
MGD
Return
Kips/day
MGD
5VS
was
SS
22Z
BOD
387
7.7
N
23.7
277 MOD Secondary Influent
mg/1
kips/day
11/7 Kips/day I —
3/1 MOD Reactor Loadin
.Secondary Reactors
MLSS = 75*3 mg/1
Reactor Vol.=2O.5"MG
36 Kips/day Net Growth
if 53 Kips/day I 75", 3 *VS
MGD Reactor Effluent
Secondary Clarifiers i~2-C)
Surf. Area = ^,?0x/0*SF
Clar. Vol. = 2O MG
7973 Kips/day
MGD Discharge
SS
BOD
N
276 MGD Secondary Effluent
SRT=
Days
Chemical
kips/day
'Alum
ng/1
kips/day
1)
TWAS
TWAS
Total Waste Act. Solids = Waste + ssoff = 2.77 Kips/Day
Biological Waste Act. Solids (BWAS) + Chemical Solids (ChemS)
ChemS = 1.12 * FeCl = 2O Kips/Day
BWAS = TWAS - ChemS^ ^57 Kips/Day
BWAS = Volatile Solids (VWAS) + Inert Solids (IWAS)
VWAS = %MLVSS * TWAS/100 = 209 Kips/Day
IWAS = BWAS - VWAS -
k1 = IWAS/SS1 f = Q.12
Kips /Day
WAS/BWAS = O.SI3
2) 1/SRT = k2 * F/M
k0 = VWAS/F = 0,33
SRT
Solids Residence Time = M/VWAS
Observed Yield F/M = Food to Mass Ratio
VWAS/BODR
F = BODlnf - BODpff
M = VA *MLVSS * 8.35
0.68 * 1.42 * VSS
pff
F
M
124
252
103
Kips/Day
Kips
-------
SECONDARY SOLIDS PRODUCTION MASS BALANCE FOR
MASS BALANCE "C'v
75 '
Xips/day [ 663
- MOD " "Re'furn
Kips/day
0,
-------
SECONDARY SOLIDS PRODUCTION MASS BALANCE FOR CY 77
MASS BALANCE "DIX
Kips/day | - %VS
? MOD Return '
Kips/day 6f.75VS
MGD
Waste
ss
135
33?
BOD
N
352
3O/ MOD Secondary Influent
mg/1
kips/day
Kips /day | -
MOD Reactor Loading
Secondary Reactors
MLSS = /6^ mg/1
Reactor
Kips/day Net Growth
Klps/dayj
3f/ MOD Reactor Effluent
Secondary Clarifiersy 7/
Surf. Area =^^0x/o3 SF
Clar. Vol. =3^ MG
3S/
MGD Discharge
SS BOD
N
7O
MGD Secondary Effluent
SRT=0.77 Days
Chemical
Polymer
jcips/day
ng/1
kips/day
1) TWAS
TV; AS
= Total Waste Act. Solids = Waste + SSeffm
Kips/Day
Biological Waste Act. Solids (BWAS) + Chemical Solids (ChemS)
ChemS = 1.12 * FeCl = 7/ Kips/Day
BWAS = TWAS - ChemS^= ^7/ Kips/Day
BWAS = Volatile Solids (VWAS) + Inert Solids (IWAS)
VWAS = %MLVSS * TWAS/100 = 2?O Kips/Day
IWAS = BWAS - WAS = &/
k = IWAS/SSi f = _C
Kips/Day
VWAS/BWAS = 0,
2) 1/SRT = k2 * F/M
k = VWAS/F = 0,88
Solids Residence Time = M/V\VAS
F/M = Food to Mass Ratio
1.42 * vSSpff
VWAS/BODR
SRT
k? = Observed Yield
F = BODlnf - BODpff + 0.68
M = VA *MLVSS * 8.35
F = 322 Kips/Day
M =
126
Kips
-------
4J
C
cu
O <*>
0)
in
rH
CM
CN
CN
a\
CM
10
vo
rH
•
CM
CM
00
CD
CM
00
•
in
m
rH
O
o
o
EH
O
U
in
•
m
CN
•
in
CM
in
o
•
CM
CM
oo
CM
•
O
01
•,
CM
CM
cr>
r^
1
r*
r^
>H
fe
CO
z
H
5
O4
W
D
J
CQ
C5
S
CO
EH
CO
O
O
S
vB
O
CJ
X
1— 1
Q
01
•p
01
0
U
w
-P
c
""•
o
0
n
O
O
O
m
in
0
o
m
CTl
O
o
0
o
0
o
CT>
ID
in
•
CM
o
o
00
o
o
00
in
c
0
o
CM
0
o
CM
VO
•
00
0
0
0
o
0
o
m
O
O
VD
VO
m
o
ro
•
o
CM
r-
•
0
vo
•
m
CM
•
o
o
•
CM
^^t
^^y
o
o
o
CM
v>
,»W
O
0)
'a
r-i
CU
&
3 >i
•H rd
to T3
T3 M
0) (U
-p a
w
£l
H
rd
•P
O
4J
1
•§
01
(U
cn
T3
3
rH
CO
J>^
rd
TJ
CO
C
o
.p
^g
CN 0
^ ^3
/i\ ^0
E W
•H
ij ®*
"id
>»
rd
5.
01
JQ
r-(
0
O
CO
00
rH
-JQ
s<
•A -0
S °
O •
rH vo
.
rd
•X3
01
J3
r-<
0
o
S'5
00 ^
—o
Or-'
Q)
*0
0)
total
CO
01
rH
rd
U
•H
0)
£
Q
CQ
a
w
4J
rd
U
.p
n
0
U
iH
Q)
C
0
to
M
0)
a<
EH
(U
o
c
c
0)
JJ
c
•H
Ul
o
•H
4J
•H
rH
•H
rd
01
8,
CO
-H
a
0)
TJ
3
rd
-------
I
< en
O Q
5 3
ui O
x
o
i-
a:
LU
z
O Q
< 3
5
o
1
1
1 1
1 1
1 1
1 1
1 1
\ 1
1
1
f
\
1
1
1
1
1
Cfl
Q
O
H
tf
W
o
u
0)
Q
1
CTl
^O
C
ft)
CO
o
Q)
Q
1
(N
r-~
C
(0
vo
U
d)
Q
1
in
r-
C
(0
00
u
cu
Q
1
[^
1 —
C
ns
U
UJ
I
O
O)
<
8
in
8
en
I
u1
en
en
en
<
K
en
en
o
o
en
en
<
CD
en
111
11
II
II1
11
II
en
,<.
>
03
O
O
O
O
Q
g
cr
UJ
Q.
xva/sai soi 'sanos Q3ivAii3v
128
-------
NITRIFICATION AT LIMA, OHIO
Felix F. Sampayo
Member
Jones & Henry Engineers, Limited
INTRODUCTION
The City of Lima, located in northwestern Ohio (40°45' lati-
tude), has a population of approximately 53,000. The City is
an important industrial center with an excellent rail and high-
way network. Average monthly temperatures range from 2.4 C
in January to 23.2°C in July.
A substantial part of the City is served by combined sewers.
Combined sewer overflows and wastewater treatment plant efflu-
ent are discharged to the Auglaize River. During dry weather,
treatment plant effluent constitutes the majority of the stream
flow.
In the late 1960s, the City began planning a pollution control
program to reduce combined sewer overflows and to improve the
existing secondary treatment plant. The program called for the
first flush of the combined sewer overflows to be collected and
transported to the treatment plant where the wastewaters would
receive at least primary treatment and chlorination. In late
January 1970, the City of Lima authorized Jones & Henry
Engineers, Limited to prepare a report recommending improve-
ments to the treatment plant. In May 1971, a report covering
phosphorus removal and miscellaneous improvements was submitted
to the City. Shortly after the report was completed, the City
129
-------
was requested by the State regulatory agency to investigate the
possibility of producing a nitrified effluent in order to
reduce the ammonia concentration in the Auglaize River.
This paper summarizes the results of the pilot studies used in
the design of the nitrification facilities, discusses the
selected treatment process, and presents operating results and
costs for three years of full-scale operation. Overall nitro-
gen control efficiency, process reliability, operational con-
trols and problems, and capital costs are also discussed. The
paper concludes with recommendations directed to more cost-
effective second generation facilities and suggests areas for
additional research.
PILOT STUDIES
Several processes were considered for possible use in producing
a nitrified effluent. The use of a one-stage or two-stage
activated sludge for nitrification was investigated and aban-
doned as impractical. During wet weather, nitrifying organisms
were washed out of the system due to high flows from the com-
bined sewers. Laboratory studies verified that one-stage and
two-stage activated sludge were ineffective, and that break-
point chlorination of the effluent would have been very cost-
ly. A decision was made to investigate nitrification towers
following the existing activated sludge. This concept appeared
to offer advantages such as low area requirements, stable per-
formance, and easy operation. The possible disadvantages
130
-------
included high capital and operation costs required to pump the
secondary effluent to the nitrification tower.
The studies were conducted in 1972-73 using a pilot unit con-
sisting of a steel shell 3 feet in diameter and 30 feet high
with 21.5 feet of plastic filter media. The media used in the
study was Surfpac, a product marketed at that time by Dow
Chemical.
The experiments showed nitrification towers could produce very
low ammonia levels during the summer months. Winter ammonia
levels would be higher because the ammonia concentration in the
tower effluent increased as the waste and air temperatures
decreased.
No net reduction in BOD_ was obtained which pointed to the
development of very specific nitrifying cultures in the tower.
Changes related to suspended solids concentration and pH were
negligible. No substantial sloughing of solids occurred at any
time as the high application rates kept solids from accumu-
lating on the surface of the media.
Study results showed nitrification units should be designed on
the basis of Total K-N (Kjeldahl) Nitrogen loading. Within the
flow ranges investigated, the hydraulic application rate did
not appear to be a significant parameter. The report of the
pilot studies recommended that nitrification towers be designed
for a TK-N loading of 0.18 pounds per square foot per day.
This loading was expected to produce an effluent containing
131
-------
2 mg/1 NH -N during the sununer months and 7 mg/1 NH3 during
the winter. Another study recommendation was not to include
either post-nitrification settling tanks or effluent polishing
filters since the tower effluent contained an average of only
15 mg/1 suspended solids.
THE PLANT
Construction of the improvements to the wet stream processes
began in January 1974 and was essentially completed in the fall
of 1976. The expansion of the anaerobic digesters and dewater-
ing facilities began in the fall of 1977 and was completed in
mid-1979.
The plant is designed for an average dry weather flow of 18.5
mgd and for a peak flow of 53 mgd. Under normal conditions,
the secondary and advanced treatment portions of the plant
operate at a peak rate of 33 mgd.
The improved activated sludge plant includes screening, grit
removal, primary settling, aeration, final settling, nitrifi-
cation towers, chlorination, and phosphorus removal. The
chemicals used for phosphorus removal are ferric chloride and
anionic polymer. Sludge treatment and disposal consists of
gravity thickening, anaerobic digestion, vacuum filtration,
sludge cake storage, and land spreading. Normal sludge
treatment/disposal uses thickening, digestion, and land spread-
ing of liquid sludge. Vacuum filtration and sludge storage is
used to provide backup to the landspreading program. Figure 1
132
-------
Ferric
Chloride
Bypass
Polymer
Raw
Sewage
Screenings
>
To Disposal To Disposal
Waste Activated Sludge j I*
~ ff
To Sludge Treatment
and Disposal
To
Auglaize
River
Recycle
To Auglaize River
Figure 1. City of Lima, Ohio
Wastewater Treatment Plant
Wet Stream Process Diagram
133
-------
illustrates the wet stream treatment process. Figure 2
diagrams sludge handling and disposal.
The activated sludge process was designed on organic loadings
ranging from 22 to 53 pounds BOD per 1,000 cubic feet of aera-
tion tank capacity. The precise loading is dependent on flow.
Over the range of organic loadings, BOD removals ranging from
92 to 66 percent were anticipated. The suspended solids remov-
als in secondary were projected to be between 75 and 60 per-
cent, depending on the flow to the system.
Two nitrification towers, each with a diameter of 106 feet,
were designed in accordance with the experimental results. The
media used in the full-scale installation was supplied by
Goodrich. The basis of design for the treatment plant, the
description of the individual treatment units, and the pro-
jected plant effluent are shown in Table 1.
THE NPDES PERMIT
The plant operates under a National Pollutant Discharge Elim-
ination System (NPDES) Permit issued on September 19, 1977 that
expires on June 30, 1980. The pertinent conditions of the Per-
mit may be summarized as follows:
Concentration (mg/1)
Parameter 30-Day 7-Day
Suspended Solids 14 20
BOD5 9 13
Ammonia (N) 2 4
Total Phosphorus 1 1.5
134
-------
Primary
Sludge
Gravity
Thickeners
Waste
Activated
Sludge
Anaerobic
Digesters
Overflow
Supernatant
To Primary Settling Tanks
Land Application
Vacuum
Filters
Filtrate
Filtrate
Holding
Tank
Filtrate
To Primary Settling Tanks
Landfill
Figure 2. City of Lima, Ohio
Wastewater Treatment Plant
Sludge Process Diagram
135
-------
Table 1
City of Lima, Ohio
Wastewater Treatment Plant
DESIGN CRITERIA AND DESCRIPTION OF PLANT
Average Daily Plow: 18.5 mgd
Peak Flow Through Secondary and Tertiary Facilities: 33 mgd
Peak Flow Through Primary Treatment: 53 mgd
Unit
Bar Screens (2)
Grit Removal Basins (2)
Primary Settling
Tanks (7)
Size
1 @ 5' wide
1 @ 6' wide
1 @ 20' x 20'
1 @ 24' x 24'
2 @ 2,964 sf
2 @ 3,600 sf
2 e 4,803 sf
1 @ 4,900 sf
730,250 cf total
Aeration Tanks (5)
Aeration Blowers (5)
Final Settling Tanks (4) 115' dia. x 14' swd
3 § 9,300 SCFM
2 @ 10,100 SCFM
Capacity and/or
Operating Conditions
53.0 mgd
16.0 mgd
23.1 mgd
53.0 mgd @ 1,900 gpd/sf
33.0 mgd @ 1,200 gpd/sf
18.5 mgd @ 650 gpd/sf
7.1 hrs @ 18.5 mgd
4.0 hrs § 33.0 mgd
1,760 cf air/lb BOD applied
33.0 mgd @ 794 gpd/sf
18.5 mgd @ 445 gpd/sf
Nitrification Towers (2) 106' dia. x 21.5' deep 18.5 mgd @0.73 gpm/sf
18.5 mgd @ 0.18 Ibs TK-N/sf
33.0 mgd @ 1.30 gpm/sf
33.0 mgd @ 0.32 Ibs TK-N/sf
Chlorine Contact
Tanks (2)
37,970 cf total
Phosphorus Removal Chemical Pumps
FeCl3 (2)
Polymer (2)
Sludge Thickeners (2)
Anaerobic Digesters
Primary (2)
Secondary (1)
210 gph each
210 gph each
75' dia. x 11' swd
85' dia. x 22' swd
85' dia. x 22' swd
18.5 mgd @ 24 minutes contact
33.0 mgd @ 15 minutes contact
25 mg/1 Fe
0.2 mg/1
21.4 Ibs/sf primary sludge
2.4 Ibs/sf chemical sludge
3.9 Ibs/sf secondary sludge
21 days detention
136
-------
Table 1
City of Lima, Ohio
Wastewater Treatment Plant
(Continued)
Capacity and/or
Unit Size Operating Conditions
Sludge Holding Tanks (2) 70' dia. x 32' swd 103,000 cf total
Vacuum Filters (3) 12' dia. x 10' 1,130 sf total
Supernatant and Filtrate
Holding Tank (1) 25' dia. x 81 swd 4,000 cf
Design Effluent Quality
BOD: 9 mg/1 @ 18.5 mgd
SS: 14 mg/1 @ 18.5 mgd
NH3-N: 2 mg/1 (summer) - 30-day average
7 mg/1 (winter) - 30-day average
P: 1 mg/1 - 30-day average
137
-------
These conditions apply to the secondary and tertiary treatment
effluent. The winter NH^-N limitation in the Permit is more
restrictive than the plant was designed for.
STARTUP
The nitrification facilities began operating in late summer,
1976. Startup progressed on schedule after damage to the plas-
tic media in one of the towers was repaired. Damage resulted
from mechanical failure of one of the distributor arms. The
facility began nitrifying in about eight weeks, and by early
November was producing the expected effluent values. The time
required for the start of nitrification was essentially the
same as that found in the pilot studies. The startup for the
rest of the plant, including phosphorus removal, presented no
particular problems.
OPERATING CHARACTERISTICS, OPERATING
PROBLEMS, AND CORRECTIVE MEASURES
During 1978 the plant removed approximately 96 percent of the
BOD and 94 percent of the suspended solids in the wastes. The
aeration system operated at a loading of 12.61 pounds BOD per
1,000 cubic feet and used 2,786 cubic feet of air per pound of
BOD removed. During 1979 the BOD removal was about 97 percent
while that for suspended solids was 93 percent. The aeration
system operated at a loading of 21.72 pounds BOD per 1,000
cubic feet and used 1,334 cubic feet of air per pound of BOD
removed.
138
-------
Operation of the nitrification facilities has been remarkably
free of problems. The towers are operated at 100 percent
recirculation throughout the year with no attempts made to
optimize recirculation rates. The operational simplicity of
the system is greatly appreciated by the plant personnel.
The towers have sloughed off solids once since they began oper-
ating. This occurred late in the summer of 1979 and lasted for
a period of approximately two hours. No decrease in process
efficiency was reported following slough-off.
The nitrification facilities have not experienced significant
operating problems during about 3.5 years of operation. The
plant superintendent reported icing problems two or three times
during the winters of 1977 and 1978, two of the coldest winters
on record for the Lima area. During these occurrences, ice
along the filter walls built up and stopped the distributor
arms. The operators broke the ice and the towers were put back
into operation.
At the beginning of the winter of 1979, operating personnel
capped the end nozzle in each of the distributor arms, elimina-
ting ice formation from splashes on the walls. No icing prob-
lems were experienced this past winter.
RESULTS
The results for BOD, suspended solids, ammonia, and dissolved
oxygen during the first three full years of treatment facili-
ties operation are shown in Tables 2, 3, and 4. Of special
139
-------
Table 2
City of Lima, Ohio
Wastewater Treatment Plant
RAW WASTEWATER AND EFFLUENT CHARACTERISTICS (1977)
Flow (mgd) Average Raw Wastewater Average Final Effluent
Month
Jan.
Feb.
March
April
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Avg.
Day
13.64
13.67
14.41
15.50
15.21
15.20
16.75
16.30
16.34
14.16
15.25
22.85
High
Day
16.31
23.06
26.86
25.28
28.26
21.96
20.11
23.75
31.28
26.08
24.77
48.60
Low BOD
Day (mg/1)
10.39 102
8.37
9.26
7.75
8.60
9.90
12.92
11.50
10.97
8.20
9.48
10.30
82
53
93
116
119
102
80
90
103
109
96
SS
(mg/1)
93
97
85
110
160
124
103
87
98
122
130
124
P
(mg/1)
11.4
6.0
2.4
6.0
5.8
6.5
4.2
12.0
16.3
14.6
15.4
12.4
BOD
(mg/1)
3.1
2.9
3.1
4.0
2.0
2.1
2.1
1.5
1.7
1.2
1.3
2.4
SS
(mg/1)
3.6
7.0
10.0
6.0
8.4
8.3
4.7
2.8
2.9
1.9
2.5
5.6
P
(mg/1)
1.2
0.4
0.0
1.0
0.7
0.8
0.7
4.0
4.3
3.9
4.0
1.7
NH3-N
(mg/1)
4.0
4.6
1.2
1.5
1.0
1.7
1.2
1.4
0.1
0.1
0.1
0.0
D.O.
(mg/1)
9.5
9.0
8.9
10.2
10.0
9.3
9.2
9.2
9.3
9.7
9.9
11.6
140
-------
Table 3
City of Lima, Ohio
Wastewater Treatment Plant
RAW WASTEWATER AND EFFLUENT CHARACTERISTICS (1978)
Flow (mgd) Average Raw Wastewater Average Final Effluent
Month
Jan.
Feb.
March
April
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Avg.
Day
14.93
12.99
27.55
19.80
10.76
9.29
8.85
9.17
8.06
8.53
8.93
11.94
High
Day
35.57
17.69
60.83
36.67
19.98
16.37
21.41
16.28
15.15
23.64
24.52
36.50
Low BOD
Day (mg/1)
9.50 71
9.17
9.30
7.80
6.95
5.35
5.76
9.17
5.65
5.74
5.47
6.60
131
82
64
89
86
93
102
105
127
135
145
SS
(mg/1)
117
146
96
75
116
143
128
164
150
145
147
139
P
(mg/1)
5.4
5.6
3.6
2.9
4.8
5.2
5.3
6.6
6.5
7.3
7.3
6.1
BOD
(mg/1)
1.3
1.3
5.2
3.2
1.6
4.1
4.1
4.2
5.9
3.5
4.9
6.2
SS
(mg/1)
4.4
4.2
6.9
6.7
2.8
6.8
5.5
6.9
8.2
6.0
5.6
19.9
P
(mg/1)
0.17
0.18
0.33
0.24
0.42
0.61
0.96
1.50
1.20
0.63
0.70
0.98
NH3-N
(mg/1)
2.14
2.80
3.05
0.37
0.45
0.54
2.00
1.23
1.97
1.14
1.93
1.13
D.O.
(mg/1)
11.3
10.6
10.6
10.6
9.9
9.1
9.7
8.6
8.7
9.9
9.2
10.0
141
-------
Table 4
City of Lima, Ohio
Wastewater Treatment Plant
RAW WASTEWATER AND EFFLUENT CHARACTERISTICS (1979)
Flow (mgd) Average Raw Wastewater Average Final Effluert
Month
Jan.
Feb.
March
April
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Avg.
pjy_
12.31
14.58
21.60
23.31
16.47
10.89
13.20
17.31
14.16
10.65
19.11
18.15
High
Day
41.72
39.53
47.73
45.11
48.76
20.32
30.50
42.22
38.74
20.36
47.02
48.00
Low BOD
Day (mg/1)
7.47 145
6.35
11.12
9.11
8.64
8.23
7.88
10.08
8.22
6.90
8.38
9.41
116
115
170
202
246
208
139
158
129
147
116
SS
(mg/1)
159
157
147
110
141
108
122
128
105
129
102
112
P
(mg/1)
5.8
6.2
4.4
3.4
5.4
6.0
5.0
4.5
5.7
6.2
4.4
4.4
BOD
(mg/1)
4.7
10.1
6.2
7.1
6.5
5.3
5.4
3.0
2.3
2.1
4.8
5.5
SS
(mg/1)
7.8
8.5
12.8
10.5
16.6
12.8
5.6
6.6
5.4
6.3
9.6
9.2
P
(mg/1)
0.42
1.23
0.57
0.72
0.89
0.75
0.67
0.60
0.90
0.71
0.74
1.23
NH3-N
(mg/1)
1.24
3.92
1.40
0.85
1.15
1.09
1.90
0.45
0.80
0.68
1.38
0.59
D.O.
(mg/1)
10.7
11.9
11.1
10.0
10.6
10.7
10.9
10.7
10.7
11.8
13.4
13.2
142
-------
significance were the results obtained during the exceptionally
cold winter of 1977-78 and 1978-79. The Tables show the quali-
ty of the effluent is generally better than required by the
NPDES Permit. The only parameter the plant has had difficulty
meeting consistently is phosphorus.
Effluent COD, pH, and nitrate nitrogen are shown in Table 5.
Table 6 shows the variations in effluent NH--N concentrations
and effluent temperatures for 1978 and 1979.
During 1979, plant personnel began taking approximately four
measurements per month of TK-N in the nitrification towers
influent and effluert. The average for these values is shown
in Table 7. Table 8 shows loading to the tower (Ibs TK-N/sf/
day) and the resulting effluent NH--N concentration for the
year 1979.
In February 1980, samples were collected upstream and down-
stream of the nitrification towers, and analyzed for BOD and
suspended solids. The average of the seven samples analyzed
are as follows:
BOD - 11 mg/1 upstream; 2 mg/1 downstream
SS - 22 mg/1 upstream; 10 mg/1 downstream
SLUDGE PRODUCTION
The nitrification towers are not followed by settling tanks,
therefore no sludge is collected. The pilot studies leading to
the design showed a sludge collection system would not be
required. The findings of the pilot studies have been largely
confirmed during operation. The towers have sloughed off
143
-------
Table 5
City of Lima, Ohio
Wastewater Treatment Plant
COD, pH, AND NH3-N IN PLANT EFFLUENT
(1978-1979)
COD (mg/1)
PH
1978
1979
Avg.
35
38
High
65
66
LOW
21
22
High
8.1
7.8
Low
7.0
7.2
N03-N
Avg. High Low
11.7 14.1 8.1
16.8 26.3 8.1
Table 6
City of Lima, Ohio
Wastewater Treatment Plant
VARIATIONS IN EFFLUENT AMMONIA
CONCENTRATION AND TEMPERATURE (1978-1979)
Month
1978
January
February
March
April
May
June
July
August
September
October
November
December
1979
January
February
March
April
May
June
July
August
September
October
November
December
NH^-N (mg/1)
Avg.
2.14
2.80
3.05
0.37
0.45
0.54
2.00
1.23
1.97
1.14
1.93
1.13
1.24
3.92
1.40
0.85
1.15
1.09
1.90
0.45
0.80
0.68
1.38
0.59
High
3.40
5.00
10.00
1.70
2.20
1.40
7.90
6.06
5.46
4.50
6.89
4.17
2.82
9.52
2.80
3.40
8.50
4.18
8.10
1.12
2.60
2.60
4.45
2.20
Low
1.00
0.40
0.00
0.00
0.00
0.00
0.10
0.15
0.27
0.22
0.18
0.13
0.25
0.89
0.10
0.15
0.32
0.14
0.12
0.16
0.19
0.17
0.11
0.12
Temperature (°C)
Avg.
9.3
6.9
9.7
11.5
15.5
21.2
23.6
23.4
22.8
22.3
18.5
18.1
10.0
9.0
12.3
14.9
17.3
21.4
23.6
23.9
22.3
18.6
14.2
11.1
High
12.1
9.4
15.0
15.6
20.0
24.4
26.1
25.6
25.0
24.4
23.3
23.3
12.0
13.9
18.9
18.9
20.6
23.3
26.1
25.6
25.6
21.7
17.8
13.9
Low
6.1
5.0
4.4
7.8
12.8
19.4
20.6
20.6
20.6
19.4
14.4
6.1
1.0
3.9
6.7
11.1
13.3
19.4
20.6
22.2
18.9
16.1
10.0
8.9
144
-------
Table 7. City of Lima, Ohio
Wastewater Treatment Plant
AVERAGE TK-N CONCENTRATION IN
NITRIFICATION TOWERS INFLUENT AND EFFLUENT (1979)
Month
January
February
March
April
May
June
July
August
September
October
November
December
TK-N Concentration (mg/1)
Tower
Influent
13.70
17.15
9.34
7.58
23.90
14.58
9.33
4.82
2.41
2.71
2.82
2.14
Tower
Effluent
3.48
8.43
4.18
2.30
5.59
7.92
7.78
3.81
1.68
1.62
2.10
1.97
Table 8. City of Lima, Ohio
Wastewater Treatment Plant
TK-N LOADING VERSUS NI^-N IN EFFLUENT (1979)
Month
January
February
March
April
May
June
July
August
September
October
November
December
Flow
(mgd)
12.31
14.58
21.60
23.31
16.47
10.89
13.20
17.31
14.16
10.65
19.11
18.15
Tower
TK-N
(mg/1)
13.70
17.15
9.34
7.58
23.90
14.58
9.33
4.82
2.41
2.71
2.82
2.14
Influent
TK-N
(Ibs/day)
1,406.5
2,085.4
1,682.5
1,473.6
3,282.9
1,324.2
1,027.1
695.8
284.6
240.7
449.4
323.7
Tower Loading*
Ibs TK-N/
sf/day
0.08
0.12
0.10
0.08
0.19
0.08
0.06
0.04
0.02
0.01
0.03
0.02
Tower
Effluent
NH3~N (mg/1)
1.24
3.92
1.40
0.85
1.15
1.09
1.90
0.45
0.80
0.68
1.38
0.59
*Total area of towers = 17,650 sf.
145
-------
solids only once for a period of about two hours during the
three and one-half years of operation.
CONSTRUCTION COST
The improvements to the wet stream portion of the plant were
built for a construction cost of $11,295,000. The major
improvements included: a new administration building and
laboratory, reconditioning of existing primary and final set-
tling tanks, additional aeration tanks and blowers, new final
clarifiers, new secondary effluent pumping station, nitrifica-
tion towers, equipment to store and feed phosphorus removal
chemicals, improvements to chlorination facilities, new sludge
thickeners, and extensive piping changes. The project was bid
in 1973 and completed in the fall of 1976.
The major improvements to sludge treatment and disposal includ-
ed additional secondary digester capacity, vacuum filters and
vacuum filter building, a sludge cake storage area and build-
ing, supernatant and filtrate storage tank, and a garage for
sludge trucks. The project was bid in 1976 and completed in
1979. The total construction cost was approximately $3,582,000.
Both projects received Federal EPA grants for 75 percent of the
eligible portions.
OPERATION AND MAINTENANCE COST
Operation and maintenance costs averaged $142.55 per million
gallons in 1978 and $138.93 per million gallons in 1979. Total
146
-------
operation and maintenance costs for both years are shown in
Table 9.
TABLE 9
CITY OF LIMA, OHIO
WASTEWATER TREATMENT PLANT
OPERATION AND MAINTENANCE COST FOR
WASTEWATER TREATMENT PLANT (1978 AND 1979)
Item 1978 1979
Payroll $356,666.58 $415,957.31
Power 172,208.54 217,028.92*
Chlorine 3,833.08 5,545.90
Chemicals** 55,367.50 74,604.53
Miscellaneous 66,346.64 97,228.62
$654,422.34 $810,365.28
* The cost for power averaged 0.0213/KWH.
** Ferric chloride and polymer.
DISCUSSION
Figures 3 and 4 show the data derived from the pilot studies
and used for design, and the operating results for 1979. The
results predicted by the pilot studies have been confirmed
under actual operation. This strongly supports the concept of
designing nitrification towers on the basis of TK-N loads.
The nitrification efficiency of the total system meets design
expectations. For part of the year (1979), the secondary plant
was nitrifying well as evidenced by the low TK-N in the secon-
dary effluent. During that time, the towers functioned as
polishing facilities. When the secondary was not nitrifying or
not nitrifying well, the towers provided the necessary ammonia
oxidation. Stable performance has been achieved with a minimum
147
-------
4-1
c
03
4J
C
0) 4J
e c
4-J 0)
03 3
QJ ^H
0) i-i
4J 0)
ro 3
3 O
0) E-i
4->
cfi C
03 -H
s
2
^ I
O <^
•H re
.c 2
O
en
^ >
03
S 0^ 03
•H C 4-1
J -rH tT3
iw 03
o o
CJ
•
r*l
0)
0)
e
p
5
(l/Biu)
| N'EHN
148
-------
o =
— 3
£L U-
0<3
<3
U) ~"
*~ o>
0 .£
o
to
-U
c
03
c
(1) -p
s c
-P 0)
03 3
OJ !-i
4-> 0)
03 S
5 O
0) E-<
4J
U) C
03 -H
s
2
- I
O oo
•H SC
X3 S
O
C/3
E cn
J -H rg
t3 Q
M-l 03
O O VJ
J OJ
>1 ^J
-U Z C
•H I -H
U « S
•^r
OJ
^
a
en
149
-------
of operational adjustment to the nitrification facility. The
operators simply set the recycle rate to 100 percent.
The plant has produced the desired results while treating the
highly variable wastewaters generated by a partially combined
sewerage system. In any single month the flow can range from
less than half to more than twice the design average. The
monthly average for BOD in the raw sewage has ranged from
53 mg/1 to 246 mg/1. The monthly average for suspended solids
in the raw sewage has ranged from 87 mg/1 to 164 mg/1.
SUGGESTED AREAS OF ADDITIONAL RESEARCH
The following areas for additional research are suggested:
1. Potential for reducing the height of the media. No
information has been developed on the minimum
height that will give the desired results.
2. Performance of nitrification towers when operated
as a combination of carbonaceous BOD removal and
nitrification.
3. The influence of the surface area per unit volume
of media.
4. The effect of forced air ventilation on the per-
formance.
POTENTIAL AREAS FOR COST SAVINGS
The only area identified where the design could be made more
cost-effective is in the material of construction for shells
housing the trickling filter media. Metal or fiberglass panels
could probably be substituted for the concrete used at Lima.
150
-------
CONCLUSIONS
The stability of the nitrification process under highly vari-
able flow conditions is evidenced in Tables 2, 3, and 4. The
flow to the plant in any one day can range between about twice
and half the average for the month. Hourly variations are con-
siderably greater. The process produces the high degree of
nitrification projected from the pilot studies.
The use of nitrification towers following activated sludge con-
sistently produces a high quality effluent. The BOD, suspended
solids, and NH..-N values have been low for the first three
years of operation. Final settling following the towers has
not been necessary as the effluent contains very low suspended
solids.
A secondary benefit derived from the use of nitrification
towers is the high dissolved oxygen concentration in the efflu-
ent. The plant effluent is normally saturated with oxygen.
The performance of the full scale plant has confirmed the
design criteria derived from pilot studies. Very low ammonia
concentrations can be obtained even during the cold winters
experienced in midwestern United States. The process used at
Lima, Ohio, single-stage activated sludge followed by nitrifi-
cation towers, is relatively easy to operate and reliably pro-
duces a high quality effluent.
151
-------
ACKNOWLEDGMENTS
The author gratefully acknowledges the operating and data col-
lecting efforts of the Superintendents for the Wastewater
Treatment Plant. Mr. Roland Nevergall, now retired, assisted
greatly during the pilot and startup phases. Mr. Jerry Coffey
has been of great assistance in the recent past.
152
-------
OPERATING EXPERIENCE WITH A 30 MGD TWO-STAGE
BIOLOGICAL NITRIFICATION PLANT
Earl W. Knight
Assistant Chief Engineer
Metropolitan Sanitary District of Greater Chicago
INTRODUCTION
The Metropolitan Sanitary District of Greater Chicago is located
within the boundaries of Cook County in Illinois. The District
encompasses an area of 866 square miles, has a present population
of approximately 5,400,000; and serves 124 member municipalities.
The District owns and operates seven treatment plants with a total
treatment capacity of 1869 MGD: 1755 MGD secondary and 114 MGD
tertiary.
The John E. Egan Water Reclamation Plant (WRP) one of the seven
plants, is located in Northwest Cook County in an unincorporated
area of Schaumburg and serves an area of approximately forty-four
square miles. This area encompasses most of the upper Salt Creek
drainage basin and includes all or parts of Palatine, Schaumburg,
Hoffman Estates, Arlington Heights, Roselle, Schaumburg, Elk Grove
Village, Rolling Meadows and Inverness. Construction of the plant
began in 1971 and the plant started treating sewage on December 16,
1975. The plant was constructed at a cost of $43 million.
The plant is designed as a 30-million gallon per day (MGD), two-stage
activated sludge system with dual media filtration. The plant con-
sists of control, maintenance, pretreatment, filter, digester, labo-
ratory, and thickener buildings; three pump houses; four aeration
tanks, four digesters, and twelve settling tanks. The plant is
capable of providing complete treatment for flows as high as 50 MGD.
153
-------
Primary treatment can be provided for an additional 75 MGD. All
effluent flows are chlorinated.
The facilities provided (Figure 1) comprise coarse screening; pump-
ing; fine screening; grit removal; two stages of aeration each fol-
lowed by settling; gravity filtration through dual-media filters;
and chlorination. Provision has been made for the addition of
aluminum salts in the first aeration stage for reduction of phos-
phorous and the addition of methanol in the filter influent for
the reduction of nitrogen if these reductions become necessary.
Facilities for handling waste activated sludge from the two aera-
tion stages include flotation thickeners and anaerobic digesters.
A centrifuge building to provide dewatering of the digested sludge
before disposal is currently under construction. Until the centri-
fuge facilities are completed the digested sludge is being pumped
to a sewer to the District's Northside Sewage Treatment Works.
TREATMENT REQUIREMENTS AND DESIGN CRITERIA
The bases for the degree of treatment provided in the design of the
John E. Egan Water Reclamation Plant were the Illinois Sanitary
Water Board Rules and Regulations which were in force at the time
of design. Since the flow of Salt Creek downstream of the plant
outfall is dominated by the Egan Plant effluent flow, the most
stringent effluent requirements were applied in the design of the
plant. The most pertinent effluent requirements were as follows:
BOD 4 mg/1
Suspended Solids 5 mg/1
154
-------
CO
10
0)
o
0
i-l
cu
c
0)
-p
(0
dJ
S-i
E-i
S-i
(0
OJ
EH
5
C
a
c
JC
o
0)
iJ
3
155
-------
Ammonia Nitrogen 2.5 mg/1
Nitrate Nitrogen 45 mg/1
Fecal Coliforms 2000 per 100 ml.
The design criteria provided for an average flow of 30 MGD with a
range of 15 MGD to 50 MGD for complete treatment. An additional
requirement was for a peak wet weather flow of 125 MGD to receive
a minimum of primary treatment. Sludge treatment was to be pro-
vided by flotation thickening and high rate anaerobic digestion.
The aeration tanks (first and second stage) were to be designed
to provide three hours detention at 50 MGD with diffused air aera-
tion and were to be capable of conventional, contact stabilization,
or step aeration processes. The settling tanks were to be designed
for an overflow rate of 1430 GPD/S.F. and a detention time of two
hours at 65 MGD. The storm water settling tanks were to be designed
for an overflow rate of 1660 GPD/S.F. and a detention time of 1.7
hours at 75 MGD. The sand filter loading for design was 5 GPM/S.F.
at 50 MGD. The digesters were to provide a fourteen-day detention
time. Figure 1-A documents the success of the design.
PLANT START UP
The John E. Egan WRP began receiving sewage for treatment on
December 16, 1975. The plant had been seeded with 100,000 gallons
of waste activated sludge from the District's Hanover Park WRP.
The limited supply of solids available made it necessary to mini-
mize the flow entering the plant until a population of organisms
adequate to provide treatment had grown in the aeration tank. This
procedure allowed better control of plant processes during the
156
-------
QUALITY DESIGN ACTUAL AVERAGE ANNUAL PERFORMANCE
PARAMETER AVERAGE
1976 1977 1978 1979
BOD
(MGA)
SS
(M3A)
NH4-N
(M3/L)
FLOW
(MGD)
4
5
1.5
30
4
4
2.4
12.2
3
2
.8
15.4
4
3
1.7
18
4
2
1.1
18
Figure 1-A. Comparison of Design Treatment Quality
With Actual Effluent Quality
157
-------
shakedown period and accomplished the following:
(1) Minimized the discharge of pollutants during biological
conditioning.
(2) Minimized the discharge of ammonia until the slow growing
nitrifiers could be established to provide treatment.
(3) Minimized waste sludge production. The digestion facility
was not operational until the end of February, 1976.
Flow into the plant was controlled by removal of the bulkheads from
only one of the two intercepting sewers entering the plant. This
limited the flow to approximately one third of that available.
Mixed liquor suspended solids (MLSS) was less than 300 mg/1 from
the start-up on December 16, 1975, until January 23, 1976. In spite
of the low MLSS, effluent BOD5 ranged from 1 to 39 mg^/1 and effluent
suspended solids ranged from 3 to 31 mg/1. Both parameters averaged
approximately 20 mg/1. First stage MLSS increased steadily to more
than 2000 mg/1 by February 5, 1976. As the MLSS increased the
effluent quality steadily improved until it was consistently able
to meet the 4 mg/1 BOD5 and ^ m9/! suspended solids criteria 55 days
after start-up.
The difficulty in obtaining a sufficient MLSS concentration was
unexpected but the problem was attributed to low plant flows allow-
ing the settling of some of the solids in sections of channels and
aeration tanks. The presumption was that the settling of solids
would not allow an increase in MLSS until the sections trapping
solids reached an equilibrium rate of gain and loss. This rate was
attained in January, 1976, when MLSS started to increase.
158
-------
Significant nitrification was first observed in the first stage
aeration effluent on February 2, 1976. Nitrification continued
to improve as the solids in the first stage increased. Ammonia
nitrogen in the first stage effluent was reduced to less than
1 mg/1 by February 29, 1976, just seventy-four days after plant
start-up. Solids from the first stage were used to seed the second
stage on March 2, 1976, and, on April 16, 1976, the remaining
bulkheads blocking flow to the plant were removed. This addi-
tional flow did not affect the effluent quality.
PLANT OPERATIONS AFTER START-UP 1976 - 1980
The John E. Egan Water Reclamation Plant serves a rapidly develop-
ing suburban area of approximately forty-four square miles. The
average daily flow, increasing yearly since the plant opened in
1975, was approximately eighteen million gallons during 1979.
This flow is sixty percent of the design flow for the plant.
Figure (2) shows actual average daily flow compared to average
daily flow estimates for design purposes.
ins U7t u;j I«B an
Figure 2. A comparison of Estimated Sewage Flows
with the Actual Flows
159
-------
The aeration systems are operated as conventional activated sludge
systems and solids are wasted from the first stage as necessary.
Normally, solids are not wasted from the second stage. Table 1
presents data showing 1979 plant performance, treatment efficiency,
and permit requirements.
Table 1
Plant Performance, 1979
Sample
Raw
Final Efficiency
Effluent
Permit Limits
BOD5
SS
NH3-N
129 mg/1
180 mg/1
14.6 mg/1
4 mg/1
2 mg/1
1.1 mg/1
96.9
98.9
92.5
(1975)
4 mg/1
5 mg/1
1.5 mg/1
(1979)
10 mg/1
12 mg/1
*
* 1.5 mg/1 April 1 - November 1
4.0 mg/1 Nov. 1 - April 1
OPERATIONAL PROBLEMS
Beginning in December, 1977, one-half of each aeration system was
taken out of service to test the treatment systems at design flow.
This test continued through September, 1978, and flow through the
plant averaged eighteen million gallons per day during this period.
Excluding September, the effluent ammonia averaged 0.9 mg/1 for the
one-tank operations with an influent concentration of 14.5 mg/1
representing a removal efficiency of 93.8%.
MLSS levels of 1000 - 2000 mg/1 were maintained during the period
from initial start-up until September, 1978. Nitrification was
maintained with little difficulty except for periods when power
160
-------
outages or mechanical problems caused lowered D.O. concentrations
and inhibited growth of the nitrifying organisms. However, these
problems were transient and caused no prolonged upset of the treat-
ment process. Major repair work was performed on the aeration tank
weirs in September, 1978. This work required a shutdown of the
entire plant for several hours, and also required that one first
stage aeration tank be drained. This operation apparently caused
the growth of nitrifying organisms to be inhibited as there was a
noticeable increase in the concentration of ammonia nitrogen in
the plant effluent for several days after the plant was restored
to service. Figure 3 shows percent ammonia nitrogen removal com-
pared to MLSS for the period September, 1978 to December, 1978. The
decrease in ammonia nitrogen removal efficiency shown on Figure 3
in December, 1978, was attributed to a four-hour power failure.
The effect of this power failure lasted well into February, 1979,
when ammonia nitrogen removal efficiency was restored.
An analysis of plant operating data was undertaken to determine
what factors caused the upset of nitrification. This analysis showed
that past power outages and low D.O. periods had caused little or
no disruption of treatment processes, but that the MLSS in the
second stage aeration tanks had been less than 500 ppm when the
upset of the nitrification process occurred. Further analysis of
the data led to the development of Table 2 which shows the relation-
ship between MLSS and nitrification relative to efficiency and
reliability. This table shows 1978 data and uses 1.5 mg/1 as the
maximum allowable ammonia nitrogen in the plant effluent. Based on
the relationship shown in Table 2, 500-600 mg/1 MLSS range was
161
-------
CI/DW) SSTW
o
o
m
CN
o
o
o
CN
I
O
O
o
o
o
o
o
in
I
w
PQ
S!
w
u
w
Q
W
CQ
O
E-l
u
o
w
EH
cu-
W
cn
i
o
g
cu
c
0)
o
4-1
•r-l
s
o
c
o
o
ro
en
T:VAOW3H % - *-HN
162
-------
selected for operational purposes as the minimum level for reliable
maintenance of the nitrification process at the Egan Plant.
Table 2
2nd Stage
MLSS (mg/1)
<100
100 - 200
200 - 300
300 - 400
400 - 500
500 - 600
600 - 700
700 - 800
800 - 900
900 - 1000
>1000
Nitrification Process at Egan Plant
(Reliability)
% within each
range below 1.5 mg/1
(Efficiency)
Average %
Nitrified
32.4
53.0
67.0
70.2
90.1
83.6
91.2
81.5
91.8
94.0
94.7
100
82.6
58.0
56.8
17.1
29.7
20.0
36.4
16.1
11.8
7.5
Maintenance of second stage MLSS is a continuing problem at the
Egan Plant. Three possible reasons for this problem, all related
to current underloading of the plant, are given below:
(1) The long retention times in first stage cause much of the
nitrification to occur in that stage leaving a negligible
amount of ammonia nitrogen for growth of the nitrifiers in
second stage. This problem is most apparent in the warm
summer months.
(2) The first stage produces an effluent which has a suspended
solids concentration of 5 - 10 mg/1 on a regular basis
163
-------
causing little solids to enter second stage.
(3) The long solids retention times in the second stage cause the
development of pin flock which carries over the weirs into
the filters thus removing solids from the second stage.
The first problem has been overcome by running only half of the
first stage system during the warm months. This reduces the deten-
tion time, and the amount of nitrification occurring in first stage,
thus providing nutrients for the nitrifiers in second stage.
The second and third problems are interrelated. The suspended
solids concentrations for second stage influent and effluent are
nearly equal, or, what solids, enter the system leave the system.
In addition, some portion of the solids are utilized within the
tanks by the nitrifying bacteria and are "lost". The solution to
the first problem provided little or no relief for problems (2) or
(3). Two possible solutions to these problems would be to increase
the second stage influent solids or to reduce the effluent solids.
These alternatives were rejected because it would be undesirable
to try to produce a poorer quality effluent in first stage to
increase the solids in the second stage influent and to reduce the
SRT in second stage by reducing the MLSS thus reducing effluent
solids carryover would cause the nitrification to suffer as was
indicated by the previous analysis.
An acceptable solution to this problem has been found to be transfer
of solids directly from first stage MLSS, to second stage MLSS. The
first stage solids contain nitrifiers to assist in the development
of nitrifiers in the second stage and provide a net influx of
164
-------
solids to second stage. The transferred first stage solids also
exhibit better settling characteristics and provide more capture
of the pin flock in the second stage settling tanks.
OPERATIONAL CONTROL
Several automatic control loops are designed into the second stage
system. These include such parameters as Dissolved Oxygen (DO) and
Return Sludge.
The DO is a set point controller allowing for more air to enter the
tank if the DO falls below the set level. Each pass of the aera-
tion system has a DO probe indicating the DO level in that pass
which is the basis for determining if more air is required.
The return sludge can be set at a fixed return rate or as a percent-
age of the current sewage flow rate. In either case the automatic
valves open or close to maintain the requested sludge flow rate to
the air lifts.
Measurements of other parameters used for operational control are
made manually. These include MLSS, settleability, and ammonia
nitrogen.
In the second stage system, the nitrifying organisms are sensitive
to the dissolved oxygen level and, thus, the DO is maintained above
2 mg/1 at all times to prevent an upset of these organisms. The
required DO level is tapered upward from first through third pass
to insure adequate oxygen for nitrification.
An adequate solids retention time (SRT) must be maintained in the
165
-------
second stage to allow for growth of nitrifying organisms. The second
stage SRTs1 are normally greater than twenty days which should be
adequate for nitrification to occur.
CAPITAL AND OPERATING AND MAINTENANCE COSTS
The contract for the construction of the John E. Egan Water Recla-
mation Plant was awarded on November 4, 1971, for $43,259,000, or
$1,442,000 per MGD (based on 30 MGD design flow). Although the
average daily design flow is 30 MGD, portions of the plant, speci-
fically the pretreatment phase, the stormwater settling tanks, and
the thickener building were constructed to meet some or all of the
future year 2020 average design flow of 90 MGD. In addition, por-
tions of the plant were designed to handle service areas greater
than the sewage treatment service area. These include the sludge
thickening and digestion facilities which will handle the sludge
produced by the nearby 72 MGD O'Hare WRP, which began operating
May 12, 1980; the maintenance shop and storeroom areas which service
the entire Northwest area; and the laboratory which performs analysis
and research for the whole north area comprising four treatment
plants. Extracting these portions from the contractor's item by
item bid listing yields a construction cost of $37,764,700 for a
30 MGD, 2-stage activated sludge sewage treatment plant, or
$1,258,900 per MGD.
The cost for nitrification, specifically the second stage aeration
and settling tanks and associated piping and equipment is estimated
to be $5,818,900 for nitrification, or $195,000 per design MGD.
The annual maintenance and operations (M & 0) cost for the plant
166
-------
is obtained from data compiled from all personnel timesheets,
purchase invoices and material distribution records. The total
M & 0 cost for 1979 for the entire plant was $2,051,599 or $112,601
per MGD (based on the 1979 average treated flow of 18.22 MGD).
The breakdown by phase of treatment shows maintenance and operation
cost for 1979 for nitrification to be $362,862, or $19,916 per MGD
for the 1979 yearly average flow of 18.22 MGD.
Table 3 summarizes the extracted capital cost and the 1979 main-
tenance and operations cost for the J.E. Egan W.R.P.
167
-------
Table 3
John E. Egan Water Reclamation Plant
Capital and Maintenance and Operations Cost
CONSTRUCTION COST
A. TOTAL $43,259,000 OR $1,442,000/MGD*
B. 30 MGD PLANT 37,764,700 OR 1,258,900/MGD*
C. NITRIFICATION 5,848,900 OR 195,000/MGD*
MAINTENANCE AND OPERATIONS COST - 1979
A. TOTAL $2,051,599 OR $112,601/MGD**
B. NITRIFICATION 362,862 OR 19,916/MGD**
* BASED ON DESIGN FLOW OF 30 MGD
** BASED ON 1979 AVERAGE TREATED FLOW OF 18.22 MGD.
168
-------
CONCLUSIONS AND RECOMMENDATIONS
(1) Means of readily transferring MLSS from first stage to second
stage should be incorporated into the design of a two-stage
system.
(2) A minimum solids level must be maintained in the second stage
aeration tanks to provide reliable nitrification. The MLSS
required, however, can be much lower than that required in the
first stage carbonaceous system.
(3) Once nitrifying organisms are established in the second stage,
the nitrification occurs very rapidly in the aeration tank,
usually in the first half. This indicates that the second
stage aeration tanks may be designed smaller than those in
the first stage. This area requires further research.
eld
169
-------
NITRIITCATION-DENITRIFTCATION IN FULL-SCALE
TREATMENT PLANTS IN AUSTRIA
N. F. MATSCHE
Assistant Professor
Technical University,Vienna,Austria
INTRODUCTION
Austria is situated in the center of Europe with the main part
of the country north-east of the Alps. The country has many na-
tural lakes which are of prime importance for recreation in a
country that is economically dependent to a large extent on
tourism. The decreasing water quality in some of these lakes in
the past years could be stopped by means of big investments in
sewers and treatment plants. When it was possible all effluents
were diverted from the catchment of the lakes or the plants dis-
charging to lakes were equipped with phosphorus removal which
in most cases means simultaneous precipitation. The removal of
nitrogen from waste water will be more important in the future
with an increasing amount of surface water being used for water
supply and an ever increasing number of hydroelectric plants on
the rivers which increases the tendency of algal nuisances under
certain climatic conditions. At the moment nitrogen removal is
mainly an operational advantage, saving a significant amount of
170
-------
energy in treatment plants with nitrification. On example of 3
different treatment plants with simultaneous nitrification-deni-
trification the process will be discussed. In conclusion the
conditions for simultaneous nitrogen removal and the advantages
of the process will be dealt with.
TREATMENT PLANTS WITH NITRIFICATION-DEtTITRIEIGATIQN
VESHBA BLUMENTAL
Nitrogen removal on a large scale single stage activated sludge
plant was reported for the treatment plant Vienna Blumental for
the first time in 1971• In this plant waste water of approxima-
tely 200 000 PE is treated without primary sedimentation in 2
aeration tanks with mammoth rotor aeration (Figure 1). Besides a
high BOD removal a significant removal of nitrogen of up to 90%
could be obtained. The mechanism of this single stage activated
sludge process for BOD-removal, nitrification and denitrifica-
tion could be explained by the simultaneous presence of aerobic
and anoxic zones in the aeration tanks (MATSCHE 1977, MATSCHE
and SPATZIERER 1977). In order to keep these conditions which
are essential for the simultaneous nitrification and denitrifi-
cation the oxygenation capacity had to be adjusted according to
the specific oxygen demand of the mixed liquor. This could be
performed with an automatic control system (USRAEL 1977) where
ML from the aeration tank is pumped to an aerated control tank
with a fixed rate. It could be shown that the DO in this control
171
-------
EXCESS ACTIVATED
SLUDGE ,
^ —fe I
r\
t
1
w
r\
I
t
w
t
UESINGTAL
SEWER
AERATION
TANKS i
RETURN SLUDGE
PUMPI STATION
d 1
GRIT CHAMBER
SCREEN
PUMP STATION
FINAL SETTLING
TANKS
OPERATORS
BUILDING
Figure 1. Treatment Plant Vienna-Blumental
Aeration Tanks:
Total Volume 12000 m3; 2 tanks,
each 150 x 17 x 2,5 m; each tank
equipped with 6 pairs of mammoth
rotors, each 75 kW, 15 m long.
Final Tanks:
Total volume 9400 m3
2 tanks of 45 m 0
172
-------
tank was inversly proportional to the oxygen uptake rate of the
HL and could be used for the continuous control of the aeration
of the plant.This control scheme has been successfully in opera-
tion for 4 years turning the aerators on and off according to
the actual demand in the aeration tanks. In order to improve
the use of this system a series of investigations was performed in
which the performance of the plant under different operating
conditions was studied. The main operating conditions that
differed in the distinct investigations were
Food Microorganism-Ratio F/M
MLSS M
Oxygen Supply OC/OU
Temperature
The main results of the experiments are given in Table 1. The
F/M-ratio varied within a range of 0,12-0,29 kg BODr/kg MLSS.d
and was mainly influenced by the varying MLSS.
During Period 1 MLSS were kept as high as 6,8 g/1 which redu-
ced the F/M ratio to 0,12 kg/kg.d. The plant was nitrifying
nearly completely under these conditions and denitrification
was working satisfactorily as well. The removal of TKN amounted
to 96 % whereas total nitrogen (TN) was removed to 86 %. The
total nitrogen content of the effluent was as low as 3,5 mg/1
of which 2,5 mg/1 were present as nitrates.
A reduction of the MLSS to 3 g/1 during Period 2 caused an in-
crease of the F/M-ratio to 0,29 kg/kg.d. In spite of the low
173
-------
Table 1. Results of Operation, Vienna-Blumental
Period
Q,-Flow
a
F-Vol.Load.
M-MLSS
F/M-SL.Load.
BODc-Inf.
BODc-Eff .
COD-Inf.
COD-Eff.
OKN-Inf .
NH4-N-Inf.
NH^-N-Eff.
org.N-Eff .
ri-TKN
NO^-N-Eff .
TI-TN
EN./P
OC/F
OC/OUR
T
1
5^,5
0,84
6,8
0,12
194 '
9
355
31 |
25 '
13
0,2
0,8
96
2,5
86
0,98
1,76
1,15
20
2
58,2
0,86
3,0
0,29
186
9
342
41
24
14
2
1,5
85
6
60
0,79
1,42
1,19
20
3
61,2
1,08
5,7
0,19
223
11
369
39 |
28 ~1
15
7
0,3
74
1
70
0,52
0,94
0,76
20
4
75,5
0,98
6,8
0,14
155
5
394
28
27 '
17
1,7
0,4
92
3
81
0,80
1,44
1,12
12,5
.
Dim.
103 m3/d
kg BOD/m^.d
kg/m')
kg/kg. d
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
%
mg/1
%
kWh/kg BOD^
kg OVkg BOD
kg/kg
°C
174
-------
sludge age nitrification was still nearly complete as the NH^-N
in the effluent was only 2 mg/1. However the TN-removal decreased
to 60 % since the NO^-N in the effluent increased to 6 mg/1.
Due to the low HLSS the anoxic zones were not sufficient to
achieve a satisfactory denitrification.
During the following Period 3 "the control system was adapted to
a lower energy level and the specific energy supply was reduced
to 0,52 kWh/kg BOD,-. Nitrification was significantly reduced
bringing the NH^-N in the effluent to 7 mg/1. With full deni-
trification the TN removal could be kept at 70 % however.
An increase in the energy supply and in the MLSS in Period 4-
could again improve the plant performance and nearly complete
nitrification was obtained even under winter conditions (12,5°C).
A comparison of the different investigations shows that at a
volumetric loading of roughly 1 kg BODc/m .d the BOD,- removal
amounted to 95 % with a BOD^ below 10 mg/1 in the effluent in
most cases. The varying energy for aeration between 0,5 and
1,0 kWh/kg BOD,--load did not show a significant influence on
the BODt--removal efficiency and so did the change in tempera-
ture. There was no significant difference between results at
12°C and 20°C.
With an increase of the F/M ratio the COD removal efficiency is
only slightly reduced from 91 % (F/M = 0,12) to 88 % (F/M =
= 0,29). As a consequence variations in the load hardly have
175
-------
an influence on the COD removal efficiency. A continuous regis-
tration of TOG in the influent and effluent of the plant could
confirm these results. Variations between 50 and 200 mg TOC/1 in
the influent of the plant resulted only in values from 8 to 12
mg TOC/1 in the effluent.
A significant influence of the operation parameters was effec-
tive for the TKN removal. A reduction in efficiency was obser-
ved with both decrease in applied energy for aeration and in-
crease in 3?/M ratio. The use of only 0,52 kWh/kg BODn reduced
the TEN removal to 74- % leaving 7 mg of NH^-N in the effluent.
As compared to Period 1 the low temperature in Period 4 only
caused a minor reduction from 96 to 91 % of TEN with NH^-N in
the effluent below 2 mg/1
The most significant differences between the different modes of
operations can be demonstrated with the TH-removal. The TN-re-
moval of 86 % under optimal conditions during Period 1 compares
well with results of earlier investigations and seems to be the
upper limit of the removal efficiency under present loading
conditions. With an operation at a low F/M-ratio of 0,12-0,13
kg/kg.d the low temperature of 12,5 °C did not significantly
affect the TN-removal which amounted to 81 %. A reduction in the
ML3S from 6,8 g/1 (Period 1) to 3,0 g/1 (Period 2), however, re-
duced the TN-removal from 86 % to as low as 60 %. In this case
the oxygen uptake in the anoxic zones was not sufficient for
the complete reduction of the produced nitrate and the NO^-N in
176
-------
the effluent rose to 6 mg/1.
The reduction of energy for aeration in Period 3 had a positive
influence on the denitrification. Nearly the whole nitrate pro-
duced is eliminated, however, the nitrification was incomplete
leaving 7 mg NH^-N in the effluent and reducing the TN-removal
to 70 %.
FELDKIRCH-MEININGEN
The treatment plant at Keiningen is a regional plant that re-
ceives the waste water of the town of Feldkirch and some surroun-
ding communities. The whole area is situated in the catchment of
Lake Constance, which means that "beside a BOD,- below 20 mg/1 in
the 24 hour composite sample the total P in the same sample has
to be below 1 mg/1.
The design is based on the load that is expected for the year
2000 with a BOD,- of 15000 kg/d. The conception of the plant is
very similar to Vienna-Blumental exept the presence of primary
sedimentation. The plant consists of
Pumping Station: 3 screw pumps, 640 1/s each
Screens , width 25 mm
Aerated grit chamber
Primary sedimentation: 2 rectangular tanks (60x15x2 m)
1800 m^ each
Aeration: 2 tanks (160x17x2,8 m) 7500 m* each, 8 mammoth
rotors, length 7i5 m per tank
177
-------
Final sedimentation: 2 circular tanks (0 45 m) 5100 nr
each
Return sludge pumping: 2 variable speed screw pumps,
100-400 1/s each
Simultaneous precipitation: addition of
At present one aeration tank is used for stabilization of the
excess sludge which is afterwards stored in one of the primary
and final clarifiers for agricultural application.
The plant is in operation since one year and a recent investiga-
tion (April 1980) could show the excellent performance of the
process, With a total flow of 12 000 mvd the BOD^ load amoun-
ted to approximately 3800 kg/d which is only 25 % of the design
load. The results of this investigation are summarized in Table 2.
The oxygenation capacity of the rotors is controlled by the
immersion depth using DO probes in the tank. The mean DO was
kept at 0,8 mg/1. Based on the chemical results the total oxygen
uptake was calculated including nitrification-denitrification
and amounted to 4400 kg O^d. With approximately 4600 kg/d of
oxygen transferred into the aeration tank the ratio of OC/OU
was 1,05. The experience from Vienna Blumental to keep this
ratio slightly above 1 resulted in an even improved nitrogen
removal in this treatment plant with primary sedimentation.
178
-------
Table 2. Results of Operation, Feldkirch-Meiningen
(Flow 12120 m3/d)
(MLSS 3,8 kg/m3)
T "1
COTD
IN
NH4-N
N07-N
TP
po4-p
Influent
310
773
24,4
6,9
5,8
4,6
Primary
Effluent
261
519
14,8
8,7
-
4,2
3,9
Final
Effluent
5
27
2,6
<0,1
2,5
0,1
<0,1
mg/1
mg/1
mg N/l
mg N/l
mg N/l
mg P/l
mg P/l
179
-------
ZELLERBECKER
The treatment plant Zellerbecken has been designed for the
treatment of sewage from Zell am See and Kaprun. Both places
are well known tourist resorts for winter and summer. As a con-
sequence the load of the plant can vary in the range of 6:1.
The design of the plant was based on a variation in the BOD^
load between 560 and 3150 kg/d. Biological treatment with sepe-
rate aerobic sludge stabilization has been chosen as treatment
process.
The plant consists of:
2 screens (width 25 and 10 mm)
Aerated grit chamber
7,
Aeration: 4 tanks, 600 nr each
7.
Final sedimentation: 2 tanks, 2050 nr each
Return sludge pumping: 2 variable speed screw pumps,
80-190 1/s each
•z
Stabilization: 2 tanks, 1000 nr each
7,
Sludge thickener, 300 nr
The aerobically stabilized excess sludge is either used in
agriculture or can be dewatered and composted together with
garbage. The aeration of the 4 aeration tanks and the 2 stabi-
lization tanks is performed by cone aerators which can be ope-
rated at two different speeds. By means of an adjustable weir
at the effluent of the aeration tanks the immersion depth of
the aerators can be controlled so that the oxygenation capacity
180
-------
can be changed in the range of 6:1. The plant was put into opey
ration 1976 and after a short starting period a number of in-
vestogations on nitrogen and phosphorus removal started (Table 3)
(v.d.EHDE, SPATZIERER). The plant is very flexible and can be
operated with one, two, three or four aeration tanks (Fig.2).
It is also possible to operate in two seperate lines, each con-
sisting of two aeration tanks and one final sedimentation tank.
For the investigations on nitrogen removal the system with pre-
denitrification was used. Instead of an internal recirculation
the return sludge v;as used the flow of which could be varied
between 80 and 580 1/s. The waste water and the return sludge
were fed to the first aeration tank which should serve as the
denitrification tank. Therefore the aerator was operated at the
lower speed in this tank to keep the oxygen input as low as
possible. Under these conditions approximately 50 % nitrogen
removal was achieved (Period 1). In order to obtain increased
denitrification the oxygen input in aeration tank 4- was de-
creased and as a result the nitrogen removal reached 75 % with
(Period 2) the same BOP and COD removal as in the previous
period.
During the following investigation (Period 3) it was tried to
adjust the oxygen input according to the oxygen consumption. This
could be met by keeping the DO between 1 and 2 mg/1 in aeration
tank 3. The application of this control decreased the power
consumption under otherwise unchanged conditions from 1800 kWh/d
to 1300 kWh/d. The TN-removal increased over 80 %. During this
181
-------
Table 3. Results of Operation, Zellerbecken
.... ,_—...-..— ii - -r- —
Period | 1
Qd-Flow 4750
P-Vol.Load 0,47
M-MLSS 9,4
P/M 0,051
! BOD5-Inf. 229
BODc-Eff. 5
COD-Inf. 431
COD-Eff. 28
TN-Inf. 35,8
HH^-N-Eff. 0,2
NOj-N-Eff. 15,4
TN-Eff. 15,9
TP-Inf. 11,4
TP-Eff. 8,6
EN/F 1,69
OC/P 'I 2,67
T | 17
2
4390
0,43
9,1
0,047
236
6
502
28
44,6
0,2
10,3
11,0
13,1
9,9
1,82
. 2,83
18
3
4900
0,40
9,6
0,043
201
5
428
36
39,2
4,0
3,3
7,6
11,6
= 4,0
1,38
2,16
17
4
4610
0,47
9,1
0,054
241
4
463
26
41,3
0,9
2,3
3,4
11,7
3,0
1,27
2,00
16
5
4300
0,35
8,8
0,030
196
4
349
24
35,9
0,2
3,9
L 5,4
8,9
4,3
1,56
2,45
15
m5/d
kg BOD/m^.d
kg/nr
kg/kg . d
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
kWh/kg BOD5
kg Op/kg BOD
°C
182
-------
LU
3
U.
U_
UJ
to
<
o
I
UJ
1
en
<
g
5
y
x
o
u
u_
cr
UJ
Q
c
0)
,x
u
(U
0)
CS3
-U
c
(0
c
0)
jj
(0
a;
CN
•H
183
-------
period the phosphorus removal without chemical precipitation
amounted to 65 % as compared to 20-30 % during Periods 1 and 2.
A comparatively high concentration of ammonia (4 mg NK^-N/1)
in the effluent indicated, that nitrification was limited by a
lack of oxygen. In order to get full nitrification DO in tank 3
was increased to 1,5-2,5 mg/1 (Period 4). As a result the energy
demand increased to 1400 kWh/d. Nitrogen removal exceeded SO %
and 73 % of total phosphorus was removed. The return sludge
flow was kept constant at 100 l-'s during Periods 1-4. An increase
in flow to 160 1/s (Period 5) left the nitrogen removal at the
90 % level, however, the phosphorus removal dropped to 50 %.
On the basis of a nitrogen mass balance the processes in the
denitrification tank could be studied. It could be shown that
besides denitrification simultaneously also nitrification
occured in this tank. Oxygen supplied to this tank is utilized
not only for the oxidation of carbonaceous compounds but also
for nitrification (Table 4). In addition to the determination of
nitrogen compounds measurements of the oxygen uptake rate of
the ML with and without inhibition of nitrification (addition
of allylthiourea) were performed. From the results one could
expect that the high oxygen demand for carbon oxidation and ni-
trification (°uc+^) compared to the relatively small oxygen in-
put (01) would result in a high amount of denitrification. In
order to decide whether nitrification or denitrification is pre-
dominant, the difference between oxygen uptake for carbon oxi-
dation (OUp) and oxygen input (01) has to be calculated.
184
-------
In case the difference of OI-OUC is positive nitrification will
be predominant; if this difference is negative denitrification
will be predominant. The observed results (Table 5) agree well
with the nitrogen mass balance.
During the whole year 1977 the performance of the plant was in-
vestigated. The BOD,--loading and the F/M-ratio during this year
varied between 0,3-0,6 kg BOD^/nAd and 0,04-0,12 kg/kg.d re-
spectively. With COD concentrations of 300-55° nig/1 in the in-
fluent a mean value of 25 mg COD/1 in the effluent was obtained
indicating a COD removal of more than 90 %. Th^ total nitrogen
concentration in the effluent was most of the time significant
below 10 mg/1 (mean value 6,6 mg/1). With a TN concentration
of 30-55 fflg N/l in the influent the removal amounted to more
than 80 % (Figures 3,4).
CONDITIONS FOR SIMULTANEOUS NITROGEN REMOVAL
Nitrification can only be performed by the spezialized nitri-
fying bacteria. On the other side a great number of aerobic
bacteria in the ML is able to use nitrate instead of DO as an
oxygen source. As a consequence the process of denitrification
is closely related to the activity of the bacteria which in case
of ML can best be expressed by the oxygen uptake rate (OU). The
OU of the ML depends on a number of influencing parameters like
F/M-ratio, temperature and addition of substrate. The denitrifi-
cation rate is influenced by the same parameters as could be
demonstrated with several measurements (MATSCHE 1979). For
185
-------
Table 4. Nitrogen Mass Balance of Aeration Tank 1
ua^e
; 12.8.
.
13.8.
Time
14.00
17.00
19.00
21.45
9.30
13.00
iNrU-lN JhlU^-IV
(mg N/l)
4,8
': 2,5
2,6
5,4
1,6
1,5
(mg N/l)
8,2
4,4
6,6
8,7
0,3
3,3
Table 5. Oxygen Uptake Rate with (OUc+N) and without
Nitrification (OUC) Compared to Oxygen Input (01)
Date
Time
OIL
OI
12.8. ; 14.00
:, 16.30,
13.8.
18.45 '
21.45 ;
9.30!
13.30
37
38
41
34
22
33
(mg02/l.h)! (mg02/l.h)j(mg02,/l.li)
___
36,0
35,0
33,5
27,5
100
94
92
78
75
82
30,0
i_
-4,5
-2,0
-6,0
-0,5
+5,5
-3,0
186
-------
50
COD (Effluent)
TOC mg/L
0
COD-Influent
300r550mg/L
COD
TOC
-1 -L. J_»» Montn
8 10 12 (1977)
ABR(kg.BOD5/m-d)
0,2 -
8
10
F/M
kgBOD5
kgMLSS-d
•0,12
-0,08
-0,04
> Month
12 (1977)
Figure 3. Treatment Plant Zellerbecken,
Results of Operation in 1977
COD, TOC in Effluent and Organic
Loading
187
-------
15 -
Total N Influent
30 T 55 mg/ L
N03-Nj
Month
8 10 12 (1977)
kWh/kgBOD5
Temperature = 10,4
Figure 4. Treatment Plant Zellerbecken,
Results of Operation in 1977,
Nitrogen Concentrations in
Effluent, Specific Energy
Consumption and Temperature
in Aeration Tank
188
-------
practical purposes it is recommended to use 70-75 % of the oxygen
uptake for carbonaceous compounds (OU^) for the estimation of
the denitrification.
This estimation method takes care of all situations in the plant
excess organic substrate at peak loads or carbon limitations
under endogenous conditions. Depending on the carbon-nitrogen
ratio in the influent the ratio of the oxic and of the anoxic
part of plants with simultaneous nitrification and denitrifica-
tion can be estimated and varied accordingly in order to obtain
optimal process performance. The volume of the aeration tank
(V-ni) can be devided in an anoxic fraction a . V.^ and in an
oxic fraction (1-a) VAT. In order to achieve full denitrifica-
tion the oxygen uptake in the anoxic fraction must exceed the
amount of oxygen supplied by nitrate. The required oxygen up-
take can be expressed by
OUre ^ 2,9 . ND
ND = 1T0 - Njg - NEp
The available oxygen uptake is
OUav = a ' OUC * °>75
ouc = 0,7 . -r] . COD
For extensive denitrification OU has to be equal to OU
av ^ re
a . 0,7 . 0,75 . T! CODn = 2,9 . NT,
1ST
n; c D
a = p, 2
TI COI>0
189
-------
where NQ CODg nitrogen, COD in influent
N-p, nitrogen for denitrification
NES nitrogen in excess sludge
N-g-p nitrogen in effluent
a anoxic fraction of aeration tank
f] COD removal efficiency
The anoxic fraction of the aeration tank depends on the ratio
of nitrogen and COD in the waste water. Since this ratio changes
between different days and within the course of a day it would
tie necessary to have a flexible partition well for a denitrifi-
cation stage for optimal performance. Plants with simultaneous
nitrification-denitrification can approach this demand by tur-
ning on or off the aerators which influences the oxic and an-
oxic fractions in the tank.
SUMMARY AND CONCLUSIONS
The demand of energy for an activated sludge plant is mainly
influenced by the energy for aeration (v.d.EMDE). Nitrification
and denitrification have a significant influence on the oxygen
demand of the process (Table 6). With full nitrification (30 mg/1
NO^-N in effluent) a process operated at a F/Ii-ratio of 0,2 kg/
/kg.d uses 65 % more oxygen as compared to the same process
without nitrification. In case the 30 mg NO^-N/1 are denitri-
fied, however, the additional oxygen demand is reduced to 24 %.
The DO that is kept in the aeration tank has an even higher in-
fluence on the energy for aeration (Table 7)« In case a DO of
190
-------
Table 6. Relative Oxygen Demand as a Function
of Nitrification and Denitrification
(F/M = 0,2 kg/kg.d, NQ = 40 mg/1, N£3 =10 mg/1)
With Nitrification
With Denitrification
mg/1
relative oxygen
demand
denitrified rl
mg/1
relative oxygen
d emand
0
10
20
30
1
1,22
1,43
1,65
0
10
: 20
j
50
1
1,08
1,16
i 1>24
i
Table 7. Relative Demand for Oxygenation as a
Function of DO in Aeration Tank
DO in
Aeration Tank
0
1
relative
OI/OU
1
1,13
1,29
1,80
191
-------
4 mg/1 is kept in the aeration tank the necessary oxygenation
exceeds the oxygen demand "by 80 %. One should therefore operate
a plant with a DO as low as possible because this can save most
of the energy. Depending on the geometry of the system with a
low DO denitrification can in many cases be obtained as well
which results in a further saving of energy for aeration. Due
to the variations in oxygen uptake the aeration system should
be as flexible as possible. This is true for aeration tanks
with circulating ML and mammoth rotor aeration that are used in
Vienna Blumental and in Feldkirch-Meiningen. However, it could
be shown that under proper operating conditions excellent ni-
trogen removal was possible in an aeration tank cascade (treat-
ment plant Zellerbecken) as well. In Vienna Blumental and in
the treatment plant Zellerbecken the process works without
primary sedimentation which instead of the presence of filamentous
organisms in the sludge results in a well settling sludge and
the possibility to work with high MLSS. To keep a high MLSS is
advantageous for denitrification. The application of the same
concept is, however, also possible with primary sedimentation
as could be shown in the plant Feldkirch-Meiningen.
For the optimal performance of a denitrification plant the
adaptation of the oxygen supply according to the oxygen uptake
is essential. With low loaded plants (volumetric loading below
0,5 kg BODc/m^.d) a control by means of DO measurements in the
aeration tank seems to be sufficient. The application of a
continuous device for the measurement of oxygen uptake rates
192
-------
separated from the aeration tank is a suitable method for
higher loaded plants.
LITERATURE
v.d. EMDE W. (1980): Untersuchungen liber Energieeinsparungen
beim Belebungsverfahren. Abwassertechnisches Seminar,
TU Hiinchen, April 1980
v.d. EMDE W., SPATZIERER G. (1978): Das Klarwerk Zellerbecken.
Osterr. Wasserwirtschaft JK), 85-94-
HATSCHE N. (1972): The Elimination of Nitrogen in the Treatment
Plant of Vienna Blumental. Water Research 6, 4-85-4-86
MTSCHE N. (1977): Removal of Nitrogen by Simultaneous Nitrifi-
cation-Denitrification, Progr.Wat.Techn. Vol.8, Nos. V5,
625-637
HATSCHE N., SPATZIERER G. (1977): Investigations towards a Con-
trol of Simultaneous Nitrogen-Elimination in the Treat-
ment Plant Vienna-Blumental. Progr.Wat.Techn. Vol.8, No.6,
501-508
HATSCHE N. (1979); Influencing Parameters on the Nitrification-
Benitrification Performance of a Single Stage Activated
T>ri
Sludge Plant. 3 IAWPR Workshop, Vienna Sept. 1979
U3RAEL G. (1977): Control of Aeration at the Treatment Plant
Vienna-Blumental, Progr.Wat .Techn., Vol.8, No.6, 24-5-249
193
-------
SINGLE-STAGE NITRIFICATION-DENITRIFICATION
AT OWE GO, NEW YORK
BY D. E. SCHWINN AND D. F. STORRIER
INTRODUCTION
In some areas of the United States, water quality criteria require
«
the removal of nitrogen from wastewater treatment plant effluents. Facili-
ties for nitrogen removal can cost considerably more to construct and
operate than conventional secondary treatment facilities. This increased
cost results from the fact that the most reliable biological system
available necessitates three separate stages of activated sludge treatment
with different types of process control in each of the three stages. Addi-
tionally, methanol must be added to multi-stage systems for nitrogen
removal. Within the past few years the energy situation has resulted in
methanol shortages and much higher costs.
Recent research has demonstrated that the functions of the three
different stages can be combined in a single treatment stage under con-
trolled operating conditions to achieve 75 percent or more nitrogen removal
without the addition of methanol. However, there is a lack of detailed
design and operating data for engineers and operators to implement single-
stage nitrification-denitrification. The major objective of this study was
to operate a full-scale plant to determine the feasibility and reliability of
the process, to identify design features needed to be incorporated by engi-
neers, and to develop operating techniques to ensure optimum performance.
Schwinn is partner and Storrier is project engineer with Stearns & Wheler,
Civil and Sanitary Engineers, Cazenovia, New York.
194
-------
Because nitrification-denitrification can become more difficult to control
under extremely cold wastewater temperatures, a full-scale plant at Owego,
New York, was selected which performs under wastewater temperatures
ranging from 8° to 22°C.
195
-------
BACKGROUND
Considerable research has been reported on biological processes
for nitrification and denitrification to provide nitrified effluents or nitro-
gen removal from wastewater. Three basic classes of organisms are
required to achieve the removal of carbonaceous and nitrogenous materials
in a suspended growth biological reactor. Aerobic heterotrophs provide
carbonaceous removal by microbial assimilation and oxidation. Aerobic
autotrophs, collectively known as nitrifiers, provide oxidation of nitrogen
in a two-step microbial reaction: first, ammonia is converted to nitrite
by Nitrosomonas; and then, nitrite is converted to nitrate by Nitrobacter.
Facultative heterotrophs provide denitrification by nitrate respiration,
i. e. , the microbial reduction of nitrate to nitrogen gas. The nitrate
radical acts as the electron acceptor and organic carbon sources serve
as electron donors under anaerobic or anoxic conditions. Facultative
organisms are also capable of oxidizing carbonaceous material under
aerobic conditions.
Optimum environmental and operating conditions for each of these
classes of bacteria differ from one another. Nitrifying bacteria have more
specific environmental requirements than the heterotrophic bacteria
responsible for carbon removal. The nitrification rate in the activated
sludge process reportedly reaches a maximum at dissolved oxygen (DO)
concentrations of approximately 2 mg/1 or above, and decreases to zero
as the DO concentration decreases to zero. Although some denitrification
can occur in aerobic systems, maximum denitrification rates occur at DO
concentrations near zero, when ample organic carbon is available. Because
of the varying conditions best suited for each type of bacteria, two-stage and
196
-------
three-stage systems for nitrification-denitrification with supplemental
carbon addition received the most attention initially. However, because
multi-stage systems have high capital and operating costs, the search
for more economical alternatives has intensified.
197
-------
DESCRIPTION OF TREATMENT FACILITIES
Owego Water Pollution Control Plant No. 2 (Figure 1) is located
in the Hamlet of Apalachin, Town of Owego, in the southern portion of
Central New York State near Binghamton. The Town of Owego and the
treatment plant itself are located on the banks of the Susquehanna River.
The plant was designed for a year 1990 flow of 7, 600 cu m/day (2. 0 mgd)
and was placed in operation in 1971.
Flexibility was provided in the design to allow operation during
the initial low flow years as an extended aeration plant and operation in
the conventional activated sludge or contact stabilization mode as waste -
water flows increase over the design life of the plant. During the year
previous to start-up of the single-stage nitrification-denitrification study,
the plant had been operated as a low-loaded conventional activated sludge
treatment plant. Currently, plant flow is approximately 1, 900 cu m/day
(0. 5 mgd). A schematic flow diagram of the treatment facilities is shown
in Figure 2. A summary of key plant components is presented in Table 1.
The wastewater at Owego is typically domestic in character. While
operating in the conventional activated sludge mode prior to initiation of
this study, the plant had been consistently achieving BODj. and suspended
0
solids (SS) removals above 90 percent.
198
-------
FIGURE 1. OWEGO WATER POLLUTION CONTROL PLANT NO. 2
199
-------
si
2 I
to o
X CC
z <
Ul
CO
II
cr ui
CL CO
Z
o
O DC
u
o
t/>
UJ
h—
«/)
<
3*
j
Ul
o
3
(n
z
cc
3
^_
Ul
cc
A
L
* + 1 -*
o
X
o
Ul
z
o
M
0
cc
Ul
|
*y
r z
i
i
,
CO
o
:D
z
K
Z
o
o
Z
—
JT
^
lii
<
' . 1
_l
<
z
C
L
f
'V y
1
< uj
« 0 §
Ul
z
p
m N
g
Ul
* 0 Z
o o
^2) , — Ul
w Z ^ 2
€/) P CC O
Z Ul IS!
0 < '2
o i o
4 .»
r
0
z
p
j
Ul
{/
5
1
j
)
h
/
» ^
_
1-
,z
JLU
— °i
tr. <
3^
Xx™\
v^/
00
^*-"^
t-o
W_J
z=>
Q
2= + ^~~
O cr
z uj
II
II
Is
CM
X Z <
O uj _j
en o a.
c\i
UJ
cr
^
g
u.
200
-------
TABLE 1. MAJOR PLANT COMPONENTS
Component
Bar Rack, Hand Cleaned
No. of Units
Unit Capacity
Comminutor
No. of Units
Unit Capacity
Grit Separators
Cyclone Degrltter, Primary Sludge
Cyclone Size
Flow Meters
Parshall Flume
Flow Range
Flow Tube (Return Sludge)
Flow (Waste Sludge)
Flow Tube (Thickened Sludge)
prlrn="-y Settling Tanks
No. of Units
Total Surface Area
Diameter
Average Depth
Aeration Tanks
No. of'Tanks
No. of Compartments Per Tank
Volume of Compartments
First Compartment (N-l and S-l)
Second Compartment (N-2 and S-2)
Depth
No. of Mechanical Aerators Per
Compartment
Aerator Horsepower
Return Sludge Rate
Final Settling Tanks
No. of Rectangular Units
Length
Width
Average Depth
Surface Area
Total Volume
Chlorine Contact Tanks
No. of Rectangular Units
Length
Width
Depth .
Total Volume
Sludge Digestion Tanks
No. of Units
Diameter
Sldewall Operating Depth
Sludge Thickening Tank
No. of Units
Diameter
Sldewater Depth
Sludge Disposal
Tank Truck
Tank Capacity
Emergency Open Sludge Drying Areas
Total Area
ISU
1 I
19,000 cu m/day 5 mgd
1 1
19,000 cu m/day 5 mgd
1
0.3 m
0.23 m
0-19,000 cu m/day
1
1
1
2
230 sq m
12 m
2.7 m
2
2
1,190 cu m
760 cu m
3.7 m
1
11 kw
1,900-5,700 cu m/day
2
26 m
4.4 m
2.6 m
240 sq m
630 cu m
1
27 m
4 m
1.8 m
200 cu m
2
12 m
9 m
1
7.3 in
2.4 m
1
7.6 cu m
2
4,050 sq m
ft
1
12 1n.
9 1n.
0-5 mgd
1
1
1
2
2,500 sq
40 ft
9 ft
2
2
42,000 cu ft
27,000 cu ft
12 ft
1
15 hp
0.5-1.5 mgd
2
85 ft
14.5 ft
8.5 ft
2,600 sq ft
167,000 gal
1
88 ft
13 ft
6 ft
52,000 gal
2
40 ft
30 ft
1
24 ft
8 ft
1
2,000 gal
2
1 acre
201
-------
PROCESS MODIFICATIONS
To achieve single-stage nitrification-denitrification in an activated
sludge system utilizing wastewater organics as the carbon source for deni-
trification, the operational plan was to provide sufficient solids retention
time (SRT) to obtain a stable nitrifying population and to provide alternating
aerobic-anoxic conditions within the reactor. To increase the concentra-
tion of solids, and thereby increase the SRT, it was anticipated that the
sludge wasting rate would have to be significantly reduced. Automatic
timers were installed to cycle the mechanical aerators in Compartments
N-l and S-l (see Figure 2) to provide the alternating aerobic-anoxic condi-
tions. No additional equipment was added to provide mixing during the
anoxic stage when the aerator was off. It was anticipated that the residual
rolling action caused by mechanical aeration would keep the mixed liquor
solids suspended to a reasonable extent if the anoxic cycle was kept rela-
tively short. Mechanical aerators in Compartments N-2 and S-2 were
allowed to provide continuous aeration to oxidize residual ammonia (NH -N)
and carbonaceous material and to prevent sludge bulking in the secondary
clarifiers caused by denitrification of nitrified mixed liquor.
In preparation for the change to the single-stage nitrification-
denitrification process, the mixed liquor suspended solids (MLSS)
concentration was increased. Starting in November 1974 sludge wasting
was discontinued. The MLSS concentration increased from 1, 000 mg/1 in
mid-November to 3, 600 mg/1 in early January 1975. Both aeration tanks
were put into service to fully establish the extended aeration process. The
hydraulic retention time (HRT) in the aeration tanks increased from approxi-
mately 12 hours to 24 hours. The process of building up the MLSS inventory
was continued.
202
-------
A major operating modification necessary to initiate the study was
the bypassing of raw wastewater directly into the aeration tanks, rather
than through the primary settling tanks as normally done. The influent
was piped directly to the aeration tanks to enhance the carbon/nitrogen
ratio entering the aeration tanks. This was done to maintain a relatively
high BOD/TKN ratio for effective denitrification.
One of the most important features of process development and
operation was the creation of alternating periods of aerobic and anoxic
conditions. Computations indicated that the aerators installed at Owego
could supply the oxygen required for carbonaceous oxidation and for
nitrification, even when operating on a 50 percent on-off cycle. This
would then allow adequate time for denitrification to occur during the off
cycle. As the aerators are of the variable submergence type, the oxygen
required can be provided in a short aerobic cycle at high submergence, or
during a long aerobic cycle at low submergence. Therefore, the duration
of the aerobic and anoxic cycles could be balanced, if desired, in accordance
with nitrification and denitrification kinetic rates.
The following equipment and process modifications were also made:
a. Installed automatic sampler housings.
b. Repaired aeration tank weir drives.
c. Installed manual bar screen at aerator influent channel
to minimize rag accumulations on aerator blades.
d. Installed flow splitter at aerator influent.
e. Balanced flows to final tanks.
f. Repaired solenoids on return sludge suction valves.
g. Installed laboratory equipment and minor electrical
and physical equipment.
h. Wasted activated sludge to control MLSS at 3, 000 mg/1.
203
-------
i. Modified recycle rate to optimize clarifier performance
and reduce rate if possible when excess supernatant
appeared in 30-minute return sludge settling sample.
The study was divided into two phases: warm weather operation and
cold weather operation. Shorter HRT's were investigated during each phase
of the study once the process had been established at longer HRT's, assuming
that sufficient SRT was provided throughout. Operation in the winter was of
particular interest because cold weather was considered to be the critical
controlling environmental factor of the study.
The timers on the aerators in Compartments S-l and N-l were
pulsed on a 30-minute staggered on-off cycle to begin Phase I (warm
•«
weather operation) of the study. The aerators in Compartments S-2 and
N-2 were set to remain running continuously. Figure 3 shows a typical
aerobic-anoxic cycle. This initiated the denitrification phase of the single-
stage nitrification-denitrification process utilizing wastewater organics as
the carbon source for denitrification. The process was operated in this
manner for the remainder of the study period, varying only the aeration
tankage in service and the submergence of aerator blades. No attempt
was made to fine tune the process by optimizing the aerator on-off cycles.
Sludge disposal was as normally practiced, employing two-stage digestion
of thickened sludge followed by ultimate disposal to the land.
The critical phase of the project, Phase n, took place during the
late winter and spring of 1976 when cold temperature performance was
studied.
Table 2 lists the sequence of events for the entire study period.
204
-------
LU
I-
=)
2
i
UJ
^.
h-
LU
_J
o
>
o
o
X
o
o
CD
O
DC
UJ
O
Q.
CO
UJ
cc
Z)
— O O> CD t^ tP IT) ^- lO CVJ — O
(|/£»UJ) N39AXO Q3A10SSIQ
UJ
205
-------
TABLE 2. SEQUENCE OF EVENTS
Pre-Study
November 1974
January 7, 1975
January 27, 1975
March 1, 1975
Phase I Start-Up
March 18, 1975
March 24, 1975
April 18, 1975
Phase IA
May 5, 1975
May 28, 1975
June 13, 1975
Phase IB
July 17, 1975
August 1975
September 8, 1975
October 1975 thru
January 1976
January 28, 1976
Phase II Start-Up
February 3, 1976
February 12, 1976
Phase IIA
March 10, 1976
March 1976
Phase IIB
May 1, 1976
May 28, 1976
Sludge wasting discontinued
Second aeration tank put in service
Primary claritiers taken off line
Grant funded
Initiation of operational and equipment modifications
Nitrogen series analysis started
Changed labs (Technicon Auto-Analyzer to wet chemistry)
Air pulsed on 30-minute cycle
Aerator blade submergence increased to 5 inches
Aerator blade submergence increased to 7 inches
North aeration tank (N-l and N-2) taken out of service
Primary digester malfunction—taken out of service
Discontinued study until cold weather period
Monitored on random basis
Hydraulic washout of solids
Aerator blade submergence decreased to 5 inches
Measures taken to reestablish process; north aeration
tank put in service; all aerators placed in continuous
operation
Pulsing started; resumed monitoring program
Began cleaning out primary digester
North aeration tank (N-l and N-2) taken out of service
Study terminated
206
-------
DISCUSSION OF RESULTS
Single-stage biological nitrification-denitrification was shown to be
a viable nitrogen removal process under full-scale plant operation for both
summer and winter conditions. Suitable environmental conditions for
nitrogen removal were created by simple operational changes and without
the supplemental addition of methanol. Wastewater organics proved to be
adequate as a carbon source for denitrification in lieu of methanol. By
operating at sufficiently low F/M ratios and correspondingly high SRT's
under warm and cold weather conditions at HRT's of about 13 to 16 hours
and 20 to 24 hours, respectively, stable carbonaceous-oxidizing, nitrifying,
and denitrifying bacterial populations were maintained. As shown in Table 3,
the largest constituent of Total-N in the effluent was generally NO^-N.
Influent and effluent NO2~N concentrations were negligible. With the
exception of the latter part of Phase II, effluent NBL-N levels varied from
0. 3 to 2. 2 mg/1, indicating a high degree of nitrification. Good removals
of Org-N were also consistently achieved. Except for a two-week period
in Phase I when too little available carbon impaired denitrification, nitrogen
removals during Phase I averaged 76 to 81 percent with both aeration tanks
in service and 77 to 86 percent with one aeration tank in service. In
Phase II, with two tanks in operation, nitrogen removals averaged about
78 percent. Nitrogen removal was initially good in Phase II with one
tank in operation but rapidly dropped to about 50 percent because of strong
supernatant returns from the anaerobic digester. One digester had to be
removed from service because of a structural failure, leaving one digester
in operation and that became increasingly overloaded.
Careful operational control of solids inventory was found to be a
major factor in obtaining efficient year-round nitrogen removal. Other
207
-------
•1
"j
<
0
£
a
c:
2
H
O
O
ct
H
2
CO
K
CQ
H
*o
t£ 0)
tf
V«
5 w
bo
_g
2
1
lo
Q)
Q(
f^
OJ
CM
>3
Q
w"
•* co r-
CD O CD
r- co CD
in r- m
co co* t—
*-)*—(»-<
1 1 1
-H CM CO
co' eg" •*'
»-(
O r- "-1
• .
eo co "i
..rj CD co c-
Mj ...
> CD co m
<|
•*->
O »H LO CD
H
t~*
a
0)
bo
C
0)
r*»
M>.
• • •
ca o co
CO CO CD
1 1 1
co in CD
r-' CM' t-'
~ '-I CM CO
N
Q
co'
co in co
^ co" co
••H *-(
MJ ^ f_ -;
>\ o eg t-
<| •'J* -"J" •'S*
~
-^
cm
2l
i
eg2
O '
X to
2 0
2.
>«•
S
CD CO CO
Tjl' ^.' T-!
"^ r~ 1
•
C
CO CO CO
o o" o
2 -
i j~
oo t"
u P
0-5
2 S
TfbC
ft S
2 ^
l_-
H
, •
£
tf
W
Iin CM r—
. . .
eg eg CM
Im o -o1
CM CM eg
*-H r~l i-(
ICM CO I—
eg" —< --i
.
c
CD •«»> *j<
r-' o •*"
>-il CM CO CO
so
c|°~
1 * .
r= 0) a
C -4J _i
C CD C
>-< a v
£H
0) 1 CO CO CD
^0| TH »-» T-l
Cl III
ffll '-> ID t-
Pj| T-l rt rH
S~| •»> CO CO
>| ^ „ ^
•o
o
H
V
P.
o
E
H
1 rt CD
.*. co "<
P 3 i J.
2 S •? £ *
< S 3 5
PH
eg
CO
CO
CO
r-
i
eg
^
eg
.
»-*
CO
CO
•<}<
03
in
i
CO
•&
CO
M<
r-
m
in"
•^
CD
^
CD
O
CO
eg
eg
CO
«-i
T-i
eg
r-
•-I
to
o
eg
i
CO
o>
i
•^
r-
i-t
.
^-<
f-
co
CO
CM
in
CO
eg
CO
in
r-
o
CO
CO
o
•<»"
CO
o"
eg
CO
CO
¥-1
in
i-("
•-,
co
CM
T-,
eg
t
CO
v->4
CD
T-,
CO
1
s
CD
a ~
CU
w
CD
co"
r-
•*
o
»^
t
in
eg
•cf
.
eg
CD
r-
CM
in"
<<
I
0
o"
CO
CD
•*'
CD
•*'
CO
CO
in"
•«"
TH
r-
»-(
CD
co'
T-,
CO
o
CD
CD
•-«
0
*-l
r
CO
a>
»H
1 CO
< 0
a -
w J
co «
< |
a •*
PH
<»•
r-
,r-
0
CM
CM
•*
CM
rH
.
nn
to
co'
'•i*1
o
in
i
CO
eg
CM
i-i
ID
O
CO
CO
CD
to
CO
o"
•a1
T-.
»-.
CD
»-*
CO
o"
«»4
»-H
CM
CM
»-,
1
CD
C3
TH
t,"
a
<
CM
«-t
r-
c«
r-i
»-«
f
•*
CD
TH
.
vH
v-t
o
*-f
CD
r-'
••S"
1
CO
CD"
eg
i—
o
iH
CO
CD
CO
CO
r-
•H
CM
V,
1
CQ co
•— ' i
•-1 J<
S fr
< |
B •*
P,
co
r-
m
•*
t-
*-«
i
•H
a>
t-
.
eg
co
CO
*-(
CO
vH
•*
m
CO
eg
CO
CO
CO
CO
CO
»-H
m
vH
T-l
'J1
T-i
m
m"
^-i
»-i
t-"
r-
a>
«-^
CO
*-4
1
CM
•-*
CM
^1
co
CM
1
O
v-4
>>
a)
S
_X
T3
C
a>
a
D.
<
o
0)
CO
at
^»
a!
"O
fl> -
lly nltrog
Devlatior
* -o
•o C
JB5
V C
0.5
*H r*s
g co
o «
o
c5Q"
b 03
TH OJ
208
-------
important factors were influent characteristics and loadings, including
wastewater temperature, COD/TKN ratio, excessive wet weather waste-
water flow rates and excessive solids in digester supernatant returns.
In early summer in Phase I, the combination of a higher bacterial
metabolism rate resulting from increasing temperature, and a lower
COD/TKN ratio, resulted in a deficiency of available carbon for denitri-
fication. To reduce the rate at which carbon was consumed during the
aerobic cycle, and thus to increase the amount of carbon available during
the anoxic cycle, the north aeration tank was removed from service. This
resulted in an immediate return of denitrification and previously observed
effluent values.
Some problems were encountered in attempting to restore nitrifica-
tion following a solids washout due to high settling tank loading rates that
resulted from infiltration/inflow problems in the collection system prior to
the start of Phase IL Although solids were rapidly built back up and the
biomass responsible for carbonaceous removal was quickly reestablished,
the nitrifiers were not reestablished until approximately four weeks after
the washout. Conversely, denitrifying bacteria were found to be highly
responsive and readily established. It appeared that when conditions were
favorable for good nitrification efficiency, a high degree of denitrification
was readily achievable by providing an anoxic period and an adequate
carbon source.
The failure of the process to efficiently remove nitrogen in the
latter part of Phase II is believed to be primarily attributable to digester
problems causing a return of excessive inert solids in digester supernatant
and not to temperature effects at lower retention times. Data from the
period between Phase I and Phase II suggest that single-stage nitrification-
denitrification at SRT's of less than 20 days is feasible during winter
conditions with adequate operational control. Spot checks indicated that
nitrification-denitrification at such SRT's was maintained at wastewater
temperatures as low as 11°C with one aeration tank in service until the
209
-------
hydraulic washout of solids occurred. Insufficient data, however, were
collected during this period prior to the washout to fully evaluate the
process under such conditions. Additional research is needed to deter-
mine minimum reactor volume under winter conditions.
Process modifications and operational changes performed to
create suitable environmental conditions for the nitrification-denitrification
process did not adversely affect the plant's ability to remove BOD, COD,
and SS. Removal efficiencies were equal or superior to previous plant
performance. As shown in Table 4, excellent removals of these three
pollutants were consistently achieved throughout the study period until
near the end of Phase n when strong digester supernatant returns caused
a slight decrease in treatment efficiency. BODc removal efficiency
normally ranged from 94 to 97 percent, except late in Phase n when the
efficiency decreased to 86 percent. Effluent COD values over various
time periods averaged between 35 and 78 mg/1. As effluent BOD^ values
were consistently less than 10 mg/1, the data suggest that residual COD
consisted of mostly refractory material. SS removals generally exceeded
95 percent. On many occasions, equipment and piping objects located
6 to 8 feet below the water surface of the final settling tanks could clearly
be seen.
There are many existing extended aeration plants in the nation
which have the capability for nitrification and denitrification, as well as
many such plants under design or construction. This study demonstrated
that a significant improvement in national water quality could be achieved
at very little increased cost by the use of single-stage nitrification-
denitrification, especially in water quality limited areas where nitrogen
removal is mandatory.
210
-------
u
c
0
•H
U
w
C
a>
g
-l-l
to
0)
QJ
rH
.Q
fO
EH
CD
t-
0>
rH
1
£
H
rp
•
^i
£
n
cc
M
1
O
rH
J>
Q
§
CD
1
•— <
t>
*^
5
CO
t* r-
CO t*- CM CD CD CO O
CO CO I OJ 1
»-H in co co
^ ^ »~* CO
r~*
CD
CO CD CO
r- 10 >rt o CD *~*
in eg i i *^
»-H c*3 mo
^*
r-l
CO
CD
t—
CTJ
*~~*
1
r— (
W
^
£
in
CD
r-»
i
ffi
W
CO
rS
£,
"
. 1 C^ *"^ ^
j^l ^ o co r- CM *-*
Oj CD in i P CD
^1 *-H ^ ^ O>
^' CN
I— (
CO
o
*~*
1
CO
1
.
— -*
w
rH
O CD CO
CD CD CO rH CO -H
C- ^< 1 »-* 1 CO
*"* CD in O)
o
CO
T-I CD CM
O3 O> CM i-t -*fi r-t
CD ^ l rH 1 fO
1—1 co r— o
CO
CM
CM O CD
£jj ^j* r— rp co •**" »~<
5j CO CO 1 I iC
^-i] •"• O> CO CD
^1 co
CO
1
f
rH
•*":
z
O
CO CM rH
co in co CD CM »"< .
O co i i r-
CM CO ^ CD
»-l
^
CD
in
••H
t
•tf
r— J
r -i
co
•£
Q<
1
>;
3
rH
CD O5
CO CD CM in rH CO
CD ^ 1 1 CO
rH JO CO CO
CD
•-H
« > rt — —
£ < co os < w « "S 5?
"" 3 -§.
Q3 '. I D
o -^ t: K o
ra r5 w i£ o
o
CO ^3*
^* O> ^* ro O oo
r- ^ i CD CM i
CO O t-
O CM
CO
t~
rH rH
rH r^ co CD in in
in »-H i co *-i i
CO —< CM
CO CM
CO
•*
CO O
CO CM CD OO CO CO
CO O 1 ^J1 CM I
CO —1 CO O
CO rH
rH
CD
o •-'
CO K3 t- in CM C-
^ rH 1 CO CM 1
CO rH CO O
CO rH
rH
r-
rH CO
r-* co in oo o CD
CO rH 1 r- rH 1
CO rH CM C~
CO CD
CM
en CM
~ 0
r- CO j^ CO -^ rH
co ^ , in CM i
0 ^
CM
CD
rH CO
co CD co in in in
CM CO 1 CO rH t
TT rH [O O>
**« rH
CM
CO
CM CM
co rH -^ in CD co
ro CD i in >-H i
CM CO T)<
ro CM
0
in
, CD CM
^ in . t— ro rH
C- rH CM in CM rH
^< in i i
O CO
r- -1
in
CM
CM •"< . CO O O
CM O CM CO m CM
in i* i i
CO CM
CO — <
rH
^~ S-*
rH •—*
^* QJ — Q)
. . bo . . bo
^ fll r> , OJ
< W P3 < OT CS
i i
V_J *«H
£ 5
CM
CO
OO
t-
CO
CO
CD
r~
CO
CO
CO
CO
rH
,
O
CO
o
r-
co
r-
.
rH
0
CM
rH
CO
O
,
CO
00
CO
r-
CO
c
0
3 "BO
*O f*
a> £
rf
C—
ro ro r-
o in i
CM -I CD
rH
rH
CO
r-
CD CM '-<
in CM i
rH in
CO
rH
CM
t-
CO CM CM
co in r
rH O
ro
•*
0
ro in CM
rH CM
CO
rH
CO
CD
CO ^ CM
ro ^ i
rH O
2
co
o
« CM CD
TJl CO 1
CM rH ^J-
CO
CD
LO
^ CM ^J*
CO CO 1
CM — ' CO
O
rH
CD
CO
in co co
in r- i
-< o
uo
CO
r- co o
co CM r-
rH rH |
CD
CO
o
CO ^ CM
CM O CO
CM CM 1
CO
CD
^~
— ' a)
. . bo
be o e
> rt
< co" PJ
i
c
rH
•tf
in o co
rH rH |
CO
CO £M CO
1
in
t-
CD r- CM
t
rH
CO
t- CO CO
rH rH t
CO
O
F~ CO *""*
1
CM
CM
CD ^ rH
1
rH
CM
in **• rH
|
rH
•* CM r-
i
rH
•^« CO CM
rH
1
rH
CO ^ T*<
rH rH J,
rH
^-^
— ~ a>
. . bo
< CO iS
i
W
00
CM
co
CM
CD
ro
CO
CD
CD
CD
CO
CO
1/5
.
CD
CO
in
.
r-
CO
CO
.
r-
01
•f
.
c-
CK
Ol
*
t-
co
CO
•ct1'
CO
c
o
"o
T)
CD
Pi
c
o
*^
•-
QJ
Q
•a
L.
as
•a
c
at
CO
n
Q
CO
211
-------
ACKNOWLEDGMENTS
The interest and support of Mr. William E. Engelhard, Supervisor
of the Town of Owego, and the members of the Town Board are acknowledged
with sincere thanks. Employees of the Water and Sewer Department of the
Town of Owego who carried out the major share of plant operation and
control included Mr. Daniel G. Thorne, Chief Plant Operator, and
Mr. Burton E. Schoonover, Laboratory Technician. This study was
partially sponsored by a grant from the United States Environmental
Protection Agency, Grant No. 803618-01.
212
-------
REFERENCES
1. Earth, E. F. , R. C. Brenner, and R. F. Lewis. Chemical Control of
Nitrogen and Phosphorus in Wastewater Effluent. J. Water Pollution
Control Federation, Vol. 46, No. 12, 2040, 1968.
2. Bishop, D. F. , J. A. Heidman, and J. B. Stamberg. Single-Stage
Nitrification-Denitrification. J. Water Pollution Control Federation,
Vol. 48, No. 3, 520, 1976.
3. Heidman, J. A. , I. J. Kugelman, and E. F. Barth. Plug Flow Single-
stage Nitrification-Denitrification Activated Sludge. Presented at:
49th Annual Conference of the Water Pollution Control Federation,
Minneapolis, Minnesota, October 3-8, 1976.
4. Kugelman, I. J. Status of Advanced Waste Treatment. Presented at:
Long Island Marine Resources Council, Hauppauge, Long Island,
New York, June 10, 1971.
5. Lawrence, A. L. , and C. G. Brown. Biokinetic Approach to Optional
Design and Control of Nitrifying Activated Sludge Systems. Presented
at: Annual Meeting of the New York Water Pollution Control Association,
New York, New York, January 23, 1973.
6. Ryan, R W. , and E. F. Barth. Nutrient Control by Plant Modification
at El Lago, Texas. EPA-600/2-76-104, U. S. Environmental Protection
Agency, Cincinnati, Ohio, 1976.
7. U. S. Environmental Protection Agency, Office of Technology Transfer.
Nitrification and Denitrification Facilities. Prepared by: Metcalf &
Eddy, Consulting Engineers, Boston, Massachusetts, August 1973.
8. U. S. Environmental Protection Agency, Office of Technology Transfer.
Process Design Manual for Nitrogen Control. Prepared by: Brown and
Caldwell, Consulting Engineers, Walnut Creek, California, October
1975.
9. U. S. Environmental Protection Agency, Office of Technology Transfer.
Process Design Manual for Upgrading Existing Wastewater Treatment
Plants. Prepared by: Metcalf & Eddy, Consulting Engineers, Boston,
Massachusetts, October 1974,
2, _L 3 , U S GOVERNMENT PRINTING OFFICE 1980-657-165/0078
-------
------- |