WATER POLLUTION CONTROL RESEARCH SERIES
11010 ESQ 08/71
Design Guides for
Biological Wastewater
Treatment Processes
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and progress in the control and abatement of pollution
in our nation's waters. They provide a central source of
information on the research, development and demonstration
activities in the Environmental Protection Agency, through
inhouse research and grants and contracts with Federal, State,
and local agencies, research institutions, and industrial
organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications Branch
(Water), Research Information Division, R&M, Environmental
Protection Agency, Washington, B.C. 20^60.
-------
DESIGN GUIDES FOR BIOLOGICAL WASTEWATER
TREATMENT PROCESSES
by
THE CITY OF AUSTIN , TEXAS
and
CENTER FOR RESEARCH IN WATER RESOURCES
Environmental Health E.-.gineering Research Laboratory
Civil Engineering Department
The University of Texas
Austin, Texas
for the
ENVIRONMENTAL PROTECTION AGENCY
Grant No.: Project #11010 ESQ
August, 1971
For sale by the Superintendent of Documents, U.S. Government Printing Ollicc, Washington, D.C. 20402 - Price $1.75
-------
EPA Reviev Notice
This report has "been reviewed by the Environmental Protection
Agency and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the Environmental Protection Agency nor does
mention of trade names or commercial products constitute
endorsement or recommendation for use.
ii
-------
ABSTRACT
Tho objective of this report is to provide u sot oi 'ju I do lines lor the design
oi various biological treatment processes. Tho equations and factors which
must be considered in the design of the activated sludge system, the contact
stabilization system, trickling filter plants, aerated lagoons, and waste
stabilization ponds were based on operating data from full-scale plants at
the Govalle Wastewater Treatment Plant, the Williamson Treatment Plant,
and the Walnut Creek Plant operated by the City of Austin, Texas and other
operating data from the treatment plants where sufficient applicable data
were recorded.
The need for waste characterization including variations in quantities in flow
and composition of flow are emphasized. The applicability and limitations of
the design equations are presented. The significant design considerations
are discussed and design calculations included where these calculations
would be meaningful and in other cases, a design procedure is outlined in
detail.
111
-------
CONTENTS
Section PAGE
I Conclusions 1
II Recommendations 3
III Introduction 5
IV Municipal Wastewater Characteristics 5
V Basic Relationships for Biological Wastewater Treatment 35
VI Activated Sludge Process 63
VII Aerated Lagoon Process 119
VIII Trickling Filter Process ih'j
IX Waste Stabilization Ponds 179
X Acknowledgments 209
XI References 211
XII Symbols and Abbreviations 219
v
-------
FIOT'RES
PAGE
1-1 Effects of Population on Wastewater Flows 7
1-2 Probability Analysis of Wastewater Flow Data 8
1-3 Rainfall and Sewage Flow H
1-4 Probability of Sewage Flow 12
1-5 Sewage Flow Distribution, Govalle Plant, Austin 13
1-6 Daily Variations in Wastewater Flow & Composition 24
1-7 Monthly Variations in Wastewater Flow & Composition 16
1-8 Monthly Variations in Wastewater Flow Composition 17
1-10 Monthly Variations in Wastewater Flow & Composition 19
1-11 Monthly Variations in Wastewater Flow & Composition 20
1-12 Loading Characteristics 21
1-13 Probability of BOD Influent 23
1-14 Probability of BOD Load 24
1-15 BOD and Load Distribution 25
1-16 Probability of Susnended Solids 27
1-17 Total and Volatile Suspended Solids of Raw Sewage 28
1-18 Non-Biodegradable Volatile Suspended Solids in Raw Wastewater 29
1-19 Non-Biodegradable Suspended Solids in Raw Wastewater 31
2-1 Continuous Fermentation Results 37
2-2 Removal Rate and Remaining BOD as Functions of Aeration Time 42
2-3 Aerated Lagoon Interactions 44
VI
-------
FIGURE PAGE
2-4 Aeration Tank Types 53
2-5 Results of Tracer Studies from Aerated Lagoons 56
2-6 Tracer Response Curves 60
3-1 Aeration Tank Arrangements 64
3-2 Non-Biodegradable Suspended Solids in Mixed Liquor 68
3-3 Non-Biodegradable Solids of Mixed Liquor 69
3-4 Total Load and Soluble Effluent 71
3-5 Soluble Load and Effluent 72
3-6 Soluble COD Load and Effluent 74
3-7 BOD Load and BOD Effluent 75
3-8 BOD and Suspended Solids Effluent 77
3-9 Effluent Suspended Solids and Sludge Age 78
3-10 Oxygen Uptake Rates (Batch) 81
3-11 Oxygen Uptake Profiles 83
3-12 Influence of Temperature on Oxygenation Capacity 85
3-13 Oxygen Saturation and Depth of Air Release 87
3-14 Bubble Aeration - Oxygen Transfer (Schematic) 88
3-15 Typical Aeration Tanks 90
3-16 Mechanical Aerator Characteristics 91
3-17 Excess Sludge and (SS + BOD) - Load 95
3-18 Different Types of Final Clarifiers 100
3-19 Sludge Volume Surface Load and Effluent Suspended Solids 101
vii
-------
FIGURE PAUE
3-20 Improvements of Final Clarifiers 103
3-21 Temperature Effects on Growth Rate of Nitrifying Bacteria 106
4-1 Rate Constant - Laboratory Scale Aerated Lagoons 121
4-2 Lab Scale Aerated Lagoons - Evaluation of the Removal
Rate Constant, "K" 123
4-3 Aerated Lagoon Transient Conditions 124
4-4 Soluble Effluent - BOD vs Detention Time 126
4-5 Effluent Suspended Solids vs Effluent BOOT 128
4-6 Aerated Lagoons Temperature Prediction Nomograph 132
4-7 Power Level vs MLSS 135
4-8 MLSS and Power Level 136
4-9 Power Level and Degree of Mixing 137
4-10 Power Level for Oxygen Transfer 139
4-11 Aerated Lagoon Systems 142
5-1 Trickling Filter; Flow Diagrams 147
5-2 Trickling Filter Evaluation 151
5-3 Influence of n on k* 153
5-4 Determination of k* 154
5-5 Determination of k* 156
5-6 BOD Removal vs Specific Surface 158
5-7 BOD Removed and Hydraulic Detention Time over Specific
Surface 159
5-8 Relationship Between Fraction of Volume Required and Specific
Surface 161
5-9 Performance and Specific Surface 162
viii
-------
FIGURE PAGE
5-10 Range of Rate Constants 167
5-11 Removal and BOD Volume - Load 168
5-12 Removal and BOD - Volume - Load 170
5-13 Excess Sludge From Trickling Filters 172
5-14 Relationship of Specific Surface and Diameter for
Various Media 175
6-1 Mechanisms of Degradations in Facultative Ponds 184
6-2 BOD Removal in Laboratory Ponds 188
6-3 BOD Remaining in Lab Scale Ponds 189
6-4 Areal Loading Rate for 907 Removal 192
6-5 Detention Time for 90% BOD Removal 193
6-6 Effluent Ouality as a Function of Areal Loading 194
6-7 Range of Load and Effluent from Waste Stabilization Ponds 195
6-8 Experimental and Theoretical Effluent Ouality as a Function
of Areal Loading 196
6-9 Calculated Biodegradation Rate Constant, Assuming Com-
pletely mixed Conditions 198
6-10 Pilot Waste Stabilization Ponds 202
IX
-------
TABLES
No. PAGE
1-1 Wastewater Flows (Williamson 1970) 6
1-2 Typical Water Use 9
1-3 Ratios of the Coefficients of Variance 32
1-4 Composition of Untreated Municipal Wastewater 33
2-1 Design Equations for Biological Treatment Processes 51
3-1 Activated Sludge Process Design Chart 112
5-1 Incluence of n and O on Rate Constant k* 152
5-2 Trickling Filter Design 176
6-1 Loading Rate and Effluent BOD from Single Facultative
Waste Stabilization Ponds 199
6-2 Comparison of Design Requirements for a Conventional
Facultative Pond and an Anaerobic-Facultative Pond System 207
-------
CONCLUSIONS
The conclusions presented oclow are rxised on the results of laboratory and
field-scale experimentation with biological treatment processes at the
Govalle Wastewater Treatment Plant, the Williamson Creek Wastewater
Treatment Plant, and the Walnut Creek Wastewater Treatment Plant, all
operated by the City of Austin, Texas.
1. Wastewater in Austin, Texas is from domestic and residential origin
and the industrial contributions are insignificant in volume and concentration.
The rate of flow and the strength of the wastewater are markedly influenced
by the infiltration of rainwater into the collection system. The average
concentrations of the influent BOD and suspended solids is 155 mg/1. The
average flow of wastewater to the Govalle Plant on Sundays is 20 MGD and
on work days is 23.5 MGD. An average dry weather flow of 25 MGD will
occur 90 percent of the time whereas a flow of 30 MGD will occur 90 percent
of the time during periods of rainfall.
2. The conventional activated sludge process resulted in effluent suspencad
solids concentrations of 12 to 22 mg/1, a total BOD of 20 to 23 mg/1, and
a soluble BOD of three to seve,". mg/1. The activated sludge process was
evaluated at loading rates of C .23 to 0.28 pounas of BOD per pound of mixed
liquor suspended solids per day resulting in BOD removal rates of 0.198 to
0.234 Ib BOD/ib MLSS - day. The narrow range of organic loadings resulted
from dilution of the incoming wastewater during periods of heavy rainfall.
The system was operated at aeration times which ranged from 2.2 to 4.9
hours.
3. Tne results of laboratory scale and field-scale evaluation of the contact
stabilization process indicate that effluent concentrations of soluble BOD of
six to 11 mg/1 are possible at contact times as low as 15 minutes. In systems
in which the contact and stabilization basins are not physically separated,
back-mixing of the wastewater occurs at the point of introduction into the
aeration Dasin. Therefore, the contact time is somewhat longer than the
theoretical value. The data indicate that the average reduction in BOD for
the contact stabilizat ion system is 78 percent. BOD loading rates of 0.2
to 0.5 Ib BOD/lb MLSS - day based on the total quantity solids in the
aeration system were applied to the contact stabilization process.
4. Results of the studies of the pilot-scale trickling filter with corrugated
plastic medium indicate that the soluble effluent BOD of 9 .5 and 27.5 mg/1
are obtainable at hydraulic loadings of 63 and 189 MGAD, respectively.
These hydraulic loadings resulted in organic loading rates of 38 and 161 Ib
BOD/ 1000 cu ft-day. The performance of the trickling filter process is
controlled to a large extent oy the type of medium used. The hydraulic
characteristics of the medium affects the magnitude of the overall BOD
removal rate constant.
-------
5. The data observed for the field-scale aerated lagoons indicate that
soluble effluent BOD concentrations of less than five mg/1 are possible at
detention times of about 2.5 days. At detention times of less than one day,
the effluent soluble BOD was between seven and 19 mg/1. The performance
of the aerated lagoons are affected by the operating temperature, the power
level, and the concentration of suspended solids in the aeration basin.
The system is less efficient at low temperatures than at high temperatures.
A power level of 30 horsepower per million gallons or higher results in a
completely mixed system. At power levels of less than 30 horsepower per
million gallons, some of the suspended solids settle and a zone of anaerobic
activity will exist at the bottom of the lagoon. An aerated lagoon must be
followed Dy an additional pond or clarifier to reduce the effluent suspended
solids concentration to a level prescribed by most regulatory agencies.
Removal of suspended solids will also markedly reduce the total BOD of
the effluent.
6, The results of these investigations of stabilizationpondsindicate that
a three pond system consisting of a separate anaerobic pond with a short
detention time followed by a facultative pond and a smaller maturation ponj
produced the best effluent of the three systems evaluated. The effluent
quality expressed as total BOD increased as the areal loading increased
from 47 to 165 Ib BOD u/acre-day. However, the soluble effluent BOD was
independent of loading rate at the range of loading used in this study, and
the soluble BOD concentration was between two and ten mg/1. The soluble
effluent BOD concentration was also in the same range for the other two
systems.
7. Any of the biological systems described above can be used for the
effective treatment of municipal wastewaters provided the proper design
equation is applied and the proper design considerations are included in
developing the system.
-------
RECOMMENDATIONS
Design of municipal wastewater treatment plants should be based on the
best .available information and should take into consideration the variations
in flow and composition of the wastewater as well as the average flow and
quality characteristics. In those communities where a collection system is
already installed reliable quantity and composition data can be developed by
a survey of the existing system. In those communities where a collection
system does not exist, published information can be used as a basis for
design provided the information selected from the published sources is for
communities which have about the same population, commercial and industrial
activities, and climate.
The design equations presented in this report can be used for the design of
the unit processes selected for the biological treatment of municipal waste-
water. The coefficients and the rate constants which are included in the
discussions can be used for preliminary design, provided no other information
is available.
Design of the biological treatment processes should take into consideration the
type of preliminary or primary treatment employed. The type of sludge handling
and disposal system provided may have a bearing on the selection of the
biological treatment system. The available land and the degree of treatment
required by effluent standards may dictate the type of system required and
eliminate the use of other biological treatment processes.
The performance of a wastewater treatment facility will oe only as good as
the operation and maintenance of the system. Therefore, the qualifications
of the operator should be taken into consideration during the design of the
particular system.
-------
MUNICIPAL WASTEWATER CHARACTERISTICS
INTRODUCTION
The design of municipal wastewater treatment facilities must take into
consideration the quantity and composition of the wastewater. The average
and the range of flow must be included in the hydraulic design of the treat-
ment plant. The characteristics of the wastewater will dictate the type of
unit processes necessary for effective treatment.
The quantity and quality of the influent wastewater is affected by the land
use of the drainage area, the extent to which sanitary and storm water are
separated, the amount of infiltration, the rainfall pattern and the type of
industrial waste ordinance enforced by the municipality. The wastewater
generated in a particular area of a city is related to the water use pattern
which in turn is established by the price of water and the type of develop-
ment of the land. For example, the water use pattern is different for single
family residences, apartments, commercial and industrial developments.
Industries which operate only se sonally or which have batch processes
and institutions which have large transient populations such as universities
can markedly affect the quantity of wastewater which must be treated.
WASTEWATER FLOW
The necessary data can be collected for particular cities if the wastewater
collection system is already installed. The hydraulic characteristics of
the system and variations of the wastewater flows from different parts of
the city can be determined by field surveys. The quality characteristics
of the wastewater from the various areas can also be determined as part of
the field survey. The treatment plant can be designed based on the exist-
ing land use pattern and the respective qualitative and quantitative character-
istics of the return flows. The projection of future land use patterns and the
anticipated population densities will permit accurate estimates of the pro-
jected design flow to the proposed treatment facility. The quantity of
wastewater return flow generated in different portions of typical cities are
available in many of the classical textbooks or design manuals. (ASCE,
1960; Billings, 1971; Clark, 1971; Fair, 1966; Fair, 1971; Steel, 1960)
However, many of the reported values are for municipalities which do not
correspond completely to the climatological and land use pattern of the
municipality for which the treatment plant is designed. Therefore, in those
municipalities where the water distribution and wastewater collection
systems exist, it is a relatively easy task to correlate the water consump-
tion and return flows for various areas of the municipality. In this manner
a more accurate estimate of the return flow can be derived. The quality
characteristics of the composite wastewater can also be developed from
these data .
-------
In those cities where a wastewater collection system does not exist, it is
necessary to estimate the quantity and quality of wastewater. The land
use pattern and zoning regulations must be established so that an accurate
estimate of the population density can be developed for the present popu-
lation and for the projected future population. The total population to be
served by the municipal treatment plant must also be determined in order
to develop the flow for the design of the facility. The estimated water use
and wastewater generation patterns must be compared with and verified by
the results of other surveys conducted in communities of similar size and
the same geographical and climatological area . The industrial contribution
to the total flow and the characteristics of these return flows must also be
carefully defined.
The total flow generated by a community and the composition of the waste-
water are a function of the population served and the industrial development
in the particular municipality. The data in Figure 1-1 indicate the relation-
ship between population served and the average wastewater flow for com-
munities of populations between 2,000 and 50,000 in the State of Texas
(Williamson, 1970) . The range of average wastewater flows for commun-
ities of the same population is wide. A statistical evaluation of the avail-
able data can be used to determine the design flow. Graphical analyses
of return flow data are presentee in Figure 1-2 (Williamson, 1970) . The
per capita flow increases as the population served increases. These data
are for municipal flow with very little or no industrial flow included in
these data. The composition of the wastewater is also affected by the
population of the municipality and in turn by the quantity of water used.
The flow data exhibited a geometrically normal distribution and the statis-
tical parameters describing the wastewater flow data are summarized in
Table 1-1.
Table 1-1
Wastewater Flows (Williamson, 1970)
Geometric Geometric
Population Mean Flow Standard Deviation
(gallons/capita -da y)
2,500-4,999 67 1.47
5,000-9,999 74 1.61
10,000-14,999 75 1.62
15,000-24,999 89 1.64
25,000-49,999 101 1.36
6
-------
160
140-
o
o
CL
CP
120-
100
o
_i
Lu
o:
LU
80
60
40
20
1
1
0 10 20 30 40
POPULATION (1,000'S)
50
FIG.I-I
EFFECTS OF POPULATION ON WASTEWATER
FLOWS (Williamson 1970}
-------
400
200
<
o
O
or
ui
CL
en
2
O
_J
e>
100
50
20
J JL
J LJLJL
POPULATION
2,500-4,999
5,000-9,999
10,000-14,999
15,000-24,999
25,000-49,999
L
J I
I 5 20 40 60 80 95 99
PERCENT OF OBSERVATIONS EQUAL TO OR LESS THAN STATE VALUES
FIG.1-2
PROBABILITY ANALYSIS OF WASTE WATER FLOW
(Williamson 1970)
-------
Water use expressed on a per capita basis varies over a wide range for
different communities; therefore, the wastewater flow data can also be
expected to have considerable variation. Some typical ranges of water
use are illustrated in Table 1-2 .
Table 1-2
Typical Water Use (Fair, 1971)
Quantity, gallons/capita - day
Use Normal Range Average
Domestic 20-90 55
Commercial 10-130 20
Industrial 20-80 50
Public 5-20 10
Water Unaccounted for 5-30 15 _
Total 60-250 150
Water use in small communities which have very little industrial usage
can be estimated at 85 - 100 gallons per capita per day (gpcd) . The water
use for larger communities in which commercial and industrial water usage
is relatively high will reflect these water demands and the average water
use will be increased to approximately 150 gpcd or greater depending on the
type of industry.
The return wastewater flow into the collection system accounts for 60 to 70
percent of the water use. This percentage can be used to estimate the
return flow when other information is not available. For example, the re-
turn flow for a small non-industrial community would be 60 to 70 gpcd if
the water use was 100 gpcd. This flow range includes the average waste-
water flow reported in Table 1-1 for a community with a population of 2500
to 4999 people. The range of return flow for a larger community with a
water use of 150 gpcd would be 90 to 105 gpcd which includes the average
return flow of 101 gpcd reported in Table 1-1 for communities with a popula-
tion of 25 ,000 to 49 ,999 . Therefore, water use records can be used to
estimate the quantity and variation of wastewater flow necessary for the
design of a wastewater collection system and treatment facility.
The quantity of wastewater also is markedly affected by rainfall. Runoff
into combined systems is directly through the catch basins. However,
-------
rainfall may porcohito into sep
-------
0)
-C
o
c.
<
o:
8
6
4
2
0
MONTHLY
MAX PER DAY
40
U-X
0
MIN —o
GOVALLE,AUSTIN
1969
I I I l I I I
13 6 9 12
1969, MONTH
FIG.1-3
RAINFALL AND WASTEWATER FLOW
11
-------
99
95
G&f*^
>.
±80
— i
OD
GO
O
? 50
CL
— •»
Z
LJ
(_)
or 20
LU
Q_
10
5
i
i
1.0
0.2
i /
i /
RAINFALL AND ' EE f^*
IMFll TRATION •
1 IN r 1 L_ 1 r\M 1 1 \J IN t ^^
. /-A ' ••
^ \ , /
X / '
X «•
,\ no
A
9
— / ^P
'o •'
0° / /
. / •
fo /
20.0 mgd / /•
o ,1
\i /•
\ t ' w
//\
/ 23.5 mgd
0 •
SUNDAY/ /
/ • WORKDAYS
0 /^
" / /
0 I
I
1 GOVALLE, AUSTIN
^ 1969
/ i i i i 1
15
20 30 40 50
WASTEWATER FLOW (mgd)
FIG.1-4
PROBABILITY OF WASTEWATER FLOW
12
-------
50
E 40
o
_)
or
UJ
30
LU 20
(S)
<
10
0
MAY 20-21,1970
GOVALLE, AUSTIN
1969
I l i 1 i
6AM
6PM
6AM
TIME
FIG.1-5
WASTEWATER FLOW DISTRIBUTION
GOVALLE PLANT , AUSTIN
13
-------
400r
AVE«267.4
(mg/l)
AVE-215.0
(mg/l)
250r
173.8
(ma/I)
10
30
15 20
JULY, 1969
FIG.1-6
DAILY VARIATIONS IN WASTE WATER FLOW
COMPOSITION PHILADELPHIA NORTHEAST WATER POLLUTION
CONTROL PLANT-JULY, 1969
-------
reported for July, 1969, by the City of Philadelphia which has a combined
collection system which accepts industrial wastes. The daily variations
in the concentrations of BOD and SS are more pronounced than the daily
variations in the wastewater flow. It is difficult to attempt to correlate
the relationship of workday or Sunday to wastewater flow and composition
since the frequency and quantity of rainfall is not reported.
The monthly variations in the wastewater flow and composition for the
Northside, Calumet, and Southwest Plants of the Chicago Metropolitan
Sanitary District are presented in Figures 1-7, 1-8, and 1-9, respectively.
Domestic and industrial wastewaters are collected in a combined system;
therefore, the data represent the influence of stormwater runoff. These
data indicate that the average concentrations of BOD and SS in the untreated
wastewater entering the Southwest Plant which receives most of the indus-
trial wastes is about ten times higher than that entering the other two
plants.
The effects of industrial wastes and stormwater on the composition of the
wastewater are indicated by the wide range of concentrations included
between the monthly maximum and minimum concentrations reported. The
monthly average flow, BOD and SS are relatively constant for the Northside
and Calumet Plants compared to those for the Southwest Plant.
The average monthly data for the two treatment plants in Fort Worth, Texas,
are presented in Figure 1-10. The collection systems are separate. The
flow rate, BOD and SS are relatively constant throughout the year. The
variations in the concentrations of SS and BOD are attributable to industrial
wastes discharged into the collection system.
The data for the City of San Antonio, Texas, are presented in Figure 1-11.
The variations in wastewater flow can be related to the infiltration of rain-.
fall as indicated by the data. The decrease in BOD as the flow rate increased
indicated the diluting effect of infiltration. The suspended solids concen-
tration remained relatively constant indicating that the infiltration did not
dilute the solids. However, the increased flow resulting from infiltration
may have resuspended some of the heavier solids which may have settled
in the collection system and offset the diluting effect.
BOD Loading
The interrelationship of flow, concentration of BOD in the wastewater and
the total organic load to the treatment plant can be used to evaluate the
dilution which generally results from infiltration of rainwater. The data
presented in Figure 1-12 indicate the variations in wastewater flow, concen-
tration of BOD and BOD load expressed in thousand pounds/day for the Govalle
-------
o
500
400
£ 100
TMAX
40O
300
o 200
o
CD
100
0
500r
_ 400
•o
1 300
O 200
u.
100
M
A M J J A S 0
1967
FIG.I-7
MONTHLY VARIATIONS IN WASTEWATER FLOW$
COMPOSITION (Northside plant)
CHICAGO METROPOLITAN SANITARY DISTRICT
N
MAX
MAX
16
-------
600
5 500-
I
400
to
o
300
LU 200
o
100
v>
T MAX
CJ
O
00
500
400
300
200
100
0
400r
^ 300
O»
E
£ 200
O
u! 100
JFMAMJJASO
1967
FIG.1-8
MONTHLY VARIATIONS IN WASTEWATER FLOW
COMPOSITION
CHICAGO CALUMET PLANT - 1967
N
rMAX
17
-------
6000
g 5000
0 4000
8^ 3000
a 9
£~ 2000
en
8 1000
n
•
-
-
-
-
\
^
-. —
• d
•
•
\
• i
, !
\
rMAX
4000
9 3000
g 2000
CD
1000
0
1000
800
£ 60O
9
1 400
o
i 200
0
*
-
H
H
-
•
. «
. «
i
•
•
"*--•.
• «
i i
>
/
•
i ,
> m
m
\
1 1
i
•
. •
[X
1
^M—W
. -
J
•MAX
MEAN
MIN
1
JFMAMJJASONU
1967
FIG.1-9
MONTHLY VARIATIONS IN W&STEWATER FLOW
COMPOSITION CHICAGO SOUTHWEST PLANT-1967
18
-------
• VILLAGE CREEK PLANT
+ RIVERSIDE PLANT
LJ
400
uiB^ 300
Q.-IOI
400
200
30r
il
20
10
_ +. j
• i
h I
i
0 N I
1966
w^
K — • — "(~-~
1 I
D J
•f ^
1 i
F M
1967
-+_
i
A
— f . i
r — ^. H
^^^-—*r
i i i
M J J A S
FIG.I-IO
MONTHLY VARIATIONS IN WASTEWATER FLOW
COMPOSITION-FORT WORTH,TEXAS
19
-------
u
4
6
. U
-
M
U
•w
u
• LEON CREEK PLANT
Mi
U
DRILLING ROAD PLANT
400
300
o 200
o
CD
100
lOOr
70
FIG.I-II
N
MONTHLY VARIATIONS IN WASTE WATER FLOW 4
COMPOSITION - SAN ANTONIO,TEXAS (1966)
20
-------
40
O 20
0
400
\
r\
# »
\
Q
O
DO
200
Q o
< ^
o
QQ
0
80
\ A /-, MAX
\
i i
"•' MIN ' \X~"
i j i i i ' ' ' ' ' ' ' ' ' ' ' ' ' '
0
V_/_
MONTH 8 12
—1968-
4 8
- 1969 —
12 4
— 1970
FIG.1-12
LOADING CHARACTERISTICS
MONTHLY AVG,MAX,MIN
GOVALLE
21
-------
Wastewater Treatment Plant. These data are the monthly average, the
maximum, and the minimum values which occurred during the time period,
July, 1968,through June, 1970. The variation in wastewater flow ranged
from slightly less than 20 MGD to over 45 MGD and represents about a
two-fold variation. However, the concentration of BOD varied from less
than 50 mg/1 to over 400 mg/1 while the BOD load ranged from about
10 ,000 to 80 ,000 pounds of BOD per day. The ratio of maximum to mini-
mum BOD concentration and load are about 8 to 1. Therefore, total BOD
load is more dependent on the concentration of BOD than on the rate of
wastewater flow.
Statistical analysis of the BOD concentration and load are presented in
Figures 1-13 and 1-14, respectively. These data are described by a
logarithmic or geometric distribution. There is some deviation from the
linearity at a BOD concentration of less than 100 mg/1 and at a BOD load-
ing of less than 15,000 Ib/day. The slopes of the two lines are almost
identical and this similarity indicates that the BOD load is affected much
more by the variation in BOD concentration than in the wastewater flow.
The data indicate that during the two year period of record the average BOD
into the Govalle Wastewater Treatment Plant was 155 mg/1 and the average
BOD load was 30,000 pounds per day.
The relationship of wastewater flow, BOD concentration, and BOD load
during a 24-hour period is presented in Figure 1-15. The data in Figure
1-15 are based on the analyses of four-hour composite samples which were
prepared by mixing hourly samples. These data were collected during July
16, 1970, and July 17 , 1970, which represents a period of dry weather flow.
The concentration of BOD is highest for the sample collected during 12:00
noon to 4:30 p.m. period and decreases for each of the following four-hour
composite samples. The sample collected between 8:30 p.m. and 12:30 a.m.
is the lowest. The flow rate also is highest at noon and decreases steadily
until 4:30 a.m. The BOD load is maximum during the four-hour period of
12:30 and 4:30 p.m. , decreases steadily and reaches a minimum between
4:30 and 8:30 a.m. The peak BOD load is 2500 pounds of BOD per hour
while the average load was 1400 pounds per hour. The ratio of the peak to
the average was 1.7 . The average load during the day-time period from
12:30 p.m. to 8:30 p.m. was approximately 2,200 pounds of BOD per hour.
The ratio of this average load to the 24-hour average load is about 1.5.
During this eight-hour period, about half of the total load arrives at the
plant.
The data indicate that transient loading conditions should be considered in
the design of biological treatment facilities. The BOD load to the plant
expressed in pounds per hour provides a better estimate of the load on the
facility than the variations in flow and BOD concentration.
22
-------
99
95
t 90
m
3 80
GO
0
Q_
• ^_
Ld
O
o:
LU
Q_
20
10
5
i
10
I.U
/
*
0
7
~ ?
_ • \.
/ I55mg/l
•
%
GOVALLE, AUSTIN
> 1969
1 1 II
50 100 200 400 1000
BOOT (mg/l)
FIG.1-13
PROBABILITY OF BOD INFLUENT
23
-------
CD
99
95
90
80
CD
050
OL
UJ
O
cr
UJ
10
1.0-
02
30,000 LB/DAY
GOVALLE.AUSTIN
1969
. I .1.1.1.1
10
20 40 60 100
LOAD (lOOOIbBOD/day)
FIG.1-14
PROBABILITY OF BOD LOAD
-------
TO
H
^
O
40
20
en
JE
Q
O
GO
Q
<
O
Q
O
CD
0
200
100
0
2000
1000
0
830 (230
430
7-16-70
8
12
30
j.
430 Q30
7-17-70 —
FIG.H5
BOD AND LOAD DISTRIBUTION
GOVALLE, AUSTIN.TEX
25
-------
Suspended Solids
The quantity of sludge requiring handling and disposal is directly related
to the suspended solids concentration in wastewater. Statistical evalua-
tion of the suspended solids data for the Govalle Treatment Plant during
1969 is shown in Figure 1-16. These data are described by a normal dis-
tribution. The mean suspended solids concentration is 155 mg/1 .
A portion of the volatile suspended solids are biodegradable. Inefficient
removal of the volatile solids adds a BOD load to and increases the oxygen
requirements of biological treatment. The relationship between the volatile
suspended solids and total suspended solids for the Govalle and Hyperion
treatment plants are presented in Figure 1-17 . This linear relationship
between the volatile fraction and the total suspended solids indicates that
the volatile fraction accounts for 86.5 per cent of the suspended solids.
A fraction of the volatile suspended solids are not biodegradable, and
therefore accumulate in the biological treatment system. Smith and Eilers
(1969) suggested that the biodegradable fraction of the volatile suspended
solids can be calculated by Equation 1-1:
_ .ultimate BOD of Suspended Solids ,
ovb ov ( COD of Suspended Solids ' 1-1
The ratio of the five-day BOD to the ultimate BOD of the suspended solids
is assumed to be 0.9. Therefore, the nonbiodegradable suspended solids
can be calculated from Equation 1-2:
(Xo " Xov'
in which:
X = total suspended solids (mg/1)
X = volatile suspended solids (mg/1)
ov
X = non biodegradable volatile suspended solids (mg/1)
ovn
X = non biodegradable solids (mg/1)
on
BOD = five-day BOD of the suspended solids (mg/1)
COD = the COD of suspended solids (mg/1)
The relationship between the volatile suspended solids and nonbiodegradable
volatile suspended solids in the untreated wastewaters is shown in Figure 1-18.
26
-------
99 -
<
•
7
95
90
80
CD
<
CD
O
(T
Q-
Ld
o 20
LJ
"- 10
J 155
•
GOVALLE,AUSTIN
1969
0 100 200 300
SUSPENDED SOLIDS (mg/l)
FIG.1-16
PROBABILITY OF SUSPENDED SOLIDS
27
-------
250
g 150
o
CO
£ 100
Q
2
LU
QL
O
X
50
0
= 0.85
GOVALLE (RAW)
HYPERION (SETTLED)
0 50 100 150 200
XQV VOLATILE SUSPENDED SOLIDS (mg/l)
FIG.1-17
TOTAL AND VOLATILE SUSPENDED
SOLIDS OF RAW SEWAGE
-------
240
200
60
c/)
O)
LU
20
o
> 80
O
X
40
0
GOVALLE
HYPERION
0 25 50 75 100 125
X0vn NON-BIODEGRADABLE VOLATILE SS (mq/l)
FIG.I-18
NON-BIODEGRADABLE VOLATILE SUSPENDED
SOLIDS IN RAW WASTEWATER
29
-------
The data indicate that the concentration of nonbiodegradable volatile solids
increases as the volatile suspended solids concentration increases. The
relationship between the nonbiodegradable suspended solids and the total
suspended solids concentration in the untreated wastewater is more linear
as shown in Figure 1-19. The slope of this line is 0.6 which indicates
that about 60 per cent of the total suspended solids in the sample are non-
biodegradable .
Parameters for Estimating Organic Contents of Wastewaters
The organic fraction of municipal wastewaters can be characterized by
organic carbon in conjunction with the classical BOD and COD information.
These three parameters are used interchangeably at times; however, each
measures a different characteristic of the wastewater. The BOD indicates
the fraction of the wastewater which is degradable by the seed microorgan-
isms during a specific incubation time. The more slowly or less readily
degradable material is generally not reflected in the standard five-day BOD
test. On the other hand, the COD includes organic and inorganic materials
which exert a chemical oxygen demand. Much of this material may be non-
biodegradable. The TOG represents an estimate of the biodegradable and
nonbiodegradable carbon in the wastewater material. A relationship or
ratio between any two of the parameters must be developed for the particular
wastewater in order to effectively use these three parameters in the design
of the treatment plant. The relationship of the BOD and TOG is linear and
may be expressed by Equation 1-3:
BOD = a (TOG) - b 1-3
The constants a and b can be evaluated graphically. However, the number
of uncertainties associated with the BOD analysis make a true evaluation
difficult. The value of these constants is different for the influent and the
effluent samples since the effluent of a wastewater treatment plant contains
the less readily biodegradable material. The coefficient of variance for
the data observed at the Govalle Wastewater Treatment Plant and those pre-
sented by Wuhrmann (1964) are shown in Table 1-3.
The data indicate that at low concentrations of BOD the ratio of BOD to TOG
varies much more widely than the similar ratios for COD to TOG. Some of
this variation may be attributable to the limitations of the BOD analysis at
low substrate concentrations. Wuhrmann (1964) indicated that the BOD
analysis represents oxygen uptake in a limited period of time by an unknown
number of organisms and an unknown substrate. In spite of the shortcomings
of the BOD determination, the design information presented in these guide-
lines are based on the BOD since sufficient TOG and COD data and exper-
ience on which to base the design of a municipal treatment plant are not
available.
30
-------
240
200
C/)
Q
_J
O
CD
O
LxJ
Q
160
20
UJ
8> 80
Z>
C/)
o
X
40
0
0
7
i
GOVALLE
HYPERION
30
60
90
120 150
Xon NON BIODEGRADABLE SS (Ing/I)
FIG.I-19
NON BIODEGRADABLE SUSPENDED SOLIDS
IN RAW WASTEWATER
31
-------
Table 1-3
Ratios of the Coefficients of Variance
Ratio Influent Effluent
Govalle Experiments
BODT/TOCT 0.91 1.38
BODF/TOCF 1.44 1.48
CODT/TOCT 0.95 1.12
CODF/TOCF 1.20 1.16
Wuhrmann's Experiments
BODT/TOCT 0.99 1.48
Nitrogen and Phosphorus
Municipal wastewater contains other constituents in addition to BOD and
suspended solids which may affect the design and performance of biolog-
ical treatment plants. Specifically nitrogen and phosphorus are considered
to be nutrients and indeed are required by the bacteria and other micro-
organisms responsible for the removal and degradation of the organic con-
stituents of wastewater. Nitrogen and phosphorus are generally present
in municipal wastewater at concentrations much higher than that required
to sustain microbial growth. The excess nutrients are discharged in the
effluent of a conventional biological treatment plant. The nitrogen and
phosphorus along with carbon have been responsible for accelerating
eutrophlcation or the aging process of streams and lakes.
The compositions of municipal wastewaters are presented in Table 1-4.
The nitrogen and phosphorus concentrations illustrate the relative concen-
trations in comparison to the other components. Owens (1953) reported
that the per capita phosphorus contribution to municipal wastewater varied
from 1.5 to 3.7 grams per person per day with an average of about 2.0 gm/
cap-day.
32
-------
Table 1-4
Composition of Untreated Municipal Wastewater
Gulp (1967) Merrell (1965) Oswald (1961)
BOD mg/1 200-400 164-630 168
Phosphorus mg/1 as P 8.3-10 12-32* 10.7
Nitrogen
Organic mg/1 as N 10-15 17-42 26.6
Ammonia mg/1 as N 25-35 29-36 33.3
Nitrate and Nitrite mg/1 as N C 0.10-0.27 1.4
PH 7.2-7.4 6.9-7.8 9.3
*Orthophosphate
33
-------
BASIC RELATIONSHIPS FOR BIOLOGICAL WASTEWATER TREATMENT
BIO DEGRADATION OF ORGANIC COMPOUNDS
Municipal wastewater is composed of a mixture of dissolved, colloidal
and particulate organic and inorganic materials. The concentration of
any individual component is continuously changing as a result of sediment-
ation, hydrolysis, and microbial transformation and degradation of organic
compounds .
The environment in a reactor permits the bacteria to use the organic material
as a substrate for growth and a source of energy. When the substrate is
depleted, the bacteria use cellular material for energy and complete auto-
oxidation results in cell death. The organic material associated with these
dead cells includes soluble biodegradable and particulate nonbiodegradable
components .
The heterotrophic microorganisms which use organic compounds as a source
of energy and synthesis play a most important role in the biological treat-
ment of wastewaters. Autotrophic bacteria which oxidize inorganic com-
pounds are also present. Typical autotrophic bacteria are the Nitrosomonas
and Nitrobacter which oxidize ammonia to nitrite and nitrate nitrogen re-
spectively. Algae, which are microscopic plants, are capable of auto-
trophic and heterotrophic synthesis. The role of algae will be discussed
in connection with stabilization ponds.
Biological processes may be operated aerobically or anaerobically. In the
first case free dissolved oxygen is maintained in the environment and micro-
organisms use this source of oxygen during the degradation of organic material
A simplification of this reaction is illustrated in Equation 2-1:
Organic material + O + NH + P bacteria > New Cells + CO + HO 2-1
The degradation of cell material is illustrated in Equation 2-2:
2-2
(C H NO ) _P + O,, - > CO9 + H9O + NH + Polysaccharide-like material
0 / Z 1U &• t* £• 3
Free dissolved oxygen is not present in the anaerobic environment. In
fact, oxygen is toxic to the truly anaerobic organisms. Under anaerobic
conditions facultative bacteria convert organic material to intermediate
organic acids and new cell material as illustrated in Equation 2-3.
-------
Organic Matter ——:—• > Organic Acids + Cells 2-3
Bacteria
Methane-forming bacteria which are true anaerobes convert the organic
acids to methane and carbon dioxide gases as shown in Equation 2-4.
Methane
Organic Acids ——7—7 7 CH. + CO. + Cells 2-4
Bacteria 4 2
The energy available to bacteria for synthesis and growth in an aerobic
system is greater than under anaerobic conditions. The methane gener-
ated anaerobically contains most of the energy which otherwise would have
been available for cell synthesis. Therefore, the quantity of sludge pro-
duced in anaerobic systems is much less than that generated during the
aerobic biodegradation of organic material.
Biological utilization of organic material involves a series of enzyme catalyzed
reactions. Simple dissolved organic material is easily incorporated in the
cells and utilized for energy. However, the more complex organic materials
must be dissolved or hydrolyzed by exoenzymes released outside the cell.
This phenomenon may provide an explanation for the rapid removal of soluble
organic material in the contact tank while the particulate and colloidal
material undergo biodegradation in the stabilization tank of the contact
stabilization process.
The aerobic biological treatment processes currently used in practice include
various modifications of activated sludge process, trickling filters and aerated
lagoons. Waste stabilization ponds are facultative systems in which aerobic
and anaerobic zones of activity occur. The anaerobic process is generally
applied for the digestion of sludges in many municipal systems.
The design and effective control of biological treatment processes require
a basic understanding of the interdependence of the various biodegradation
reactions. The basic correlations were developed from continuous flow
reactors and reported by Herbert (1956), Monod (1949), and Schulze (1964).
Glucose was the single substrate used in all these studies. Under these
conditions only one type of microorganism will develop rather than the mixed
culture which results from the use of the heterogeneous substrate such as
municipal wastewater.
The general results of these studies are presented in Figure 2-1. These
relationships indicate:
(a) The growth rate increases sharply at low concentrations
of substrate; however, as the substrate concentration reaches
a maximum a maximum rate of growth is reached.
36
-------
S (SUBSTRATE)
S (SUBSTRATE)
FIG.2-1
CONTINUOUS FERMENTATION RESULTS
37
-------
(b) the oxygon uptako rate increases linearly as the substrate
removal rate increases.
(c) there is a linear relationship between the growth rate and
the substrate removal rate.
(d) substrate removal rate can be used to substitute for the
growth rate in design procedures .
The substrate removal rate may be expressed in terms of pounds of BOD
per pound of MLSS per day. High removal rates are possible only when
the soluble substrate is high. However, at low substrate concentrations,
the removal rate increases markedly with a slight increase in the sub-
strate concentration.
These relationships are not directly applicable to multiple component
substrates and mixed cultures of bacteria. Therefore, the mathematical
formulations developed for glucose must be modified for municipal waste-
waters. The equations presented below provide a basis for understanding
the interaction between bacteria and a single substrate and are used to
develop the applicable design equations for the various processes employed
in the biological treatment of municipal wastewaters.
The relationship between the growth rate and the substrate concentration
can be written in Equation 2-5.
in which:
k +S
m
growth rate and maximum growth rate (day )
2"5
S = substrate concentration (mg/1)
k = substrate concentration at |_t = 0.5|a (mg/1)
m Inax
b = endogenous respiration rate (day )
This equation was developed by Michaelis-Menten for enzyme kinetics and
was first applied to describe growth k netics in biological waste treatment
by Monod (1959) . The endogenous respiration rate, b, is frequently omitted
from this equation. This endogenous rate is only detectable at extremely
low substrate concentrations and compared to the maximum growth rate is
almost negligible. However, the endogenous rate (b) must be considered
in developing any growth kinetics or any substrate removal kinetics .
38
-------
Theissier (1936) presented a different expression for the growth rate which
is illustrated as Equation 2-6.
(1 _ e"cS) - b 2-6
These equations are quite similar; however, the equation presented by
Michaelis-Menten will be used to develop other design formulations.
The relationships between oxygen uptake rate and growth rate with substrate
removal rate are represented in Equations 2-7 and 2-8.
k = a1 —— + b1 (oxygen uptake rate) 2-7
r XAt
= a -^-— - b (growth rate) 2-8
in which:
k = oxygen uptake rate (mg (X/mg cells-day)
r ^
—— = substrate removal rate (mg substrate/mg cells-day)
XAt
a, a1 = substrate utilization constants
b, b1 = endogenous respiration constants (day )
X = concentration of active cells (mg/1)
Combining Equations 2-5 and 2-8, the substrate removal rate can be defined
by Equation 2-9.
_AS_ , , Sv 2_9
XAt R Vkm + S'
in which:
k = u /a = maximum biodegradation rate (mg of substrate
R max
per mg of cells per day)
This equation is applicable only to completely mixed reactors. The relation-
ship can be modified for plug flow or batch reactors as shown in Equation
2-10.
39
-------
2-10
The negative sign indicates that the substrate Ecmovaj^-pate decreases with
increasing time. The integration of Equation 2-10 results in Equation 2-11
oO
c k
m m
The exponents in Equation 2-11 indicate the substrate removal is a function
of time-dependent and substrate-dependent variables.
REMOVAL KINETICS
The expression presented in Equation 2-11 can be modified to describe the
specific physical system used for waste treatment. The removal rate is
constant if a single organic compound in a high concentration is to be
treated. Under these conditions k is very small and the amount of sub-
strate removed (S - S) is high; therefore, the first exponent is negligible
and the substrate°removal controls. However, linear removal is almost
never observed for municipal wastewater. In general, it may be assumed
that k is much larger than the effluent substrate concentration, S, and
that (Ic1 + S) is proportional to k . Equation 2-9 can then be rewritten as
Equation 2-12. m
^ - *s
in which:
k = kR
k
m
Integration of this expression results in Equation 2-13.
-Bft
S
o
= e
2-13
The change in substrate concentration in a completely-mixed reactor can
be described by a mass balance assuming that no biodegradation takes place,
by Equation 2-14.
-------
Q S At - Q S At - VAS = 0 2-14
o e
This equation may be rearranged and presented as Equation 2-15.
Q (S - S )
AS _ o e
At V 2-15
Under steady-state conditions the change in substrate concentration with
time is equal to the amount of substrate that must be removed; therefore,
combining Equations 2-12 and 2-15 results in Equation 2-16.
2_16
The hydraulic detention time is equal to the volume divided by the rate
of flow (t = — ) , and in a completely-mixed reactor the effluent concentra-
tion S is equal to the concentration of substrate in the reactor (S) ; there
fore, ihe effluent concentration may be defined by Equation 2-17.
S
i + kxt
These equations deal primarily with the soluble material and must be
modified for municipal wastewaters since only 40 to 60 percent of the total
BOD entering the system is soluble. A relatively short period of time of
mixing the incoming wastewater with the activated sludge is necessary for
the removal of the soluble components of the waste. The rate of removal
is dependent on the mixed liquor suspended solids concentration. The
kinetics of removal of soluble organic compounds was described in Equation
2-12. The effect of aeration time on the substrate removal rate and the
quantity of substrate remaining are illustrated in Figure 2-2. A higher
removal rate is observed at short aeration times; however, as the aeration
time increases , the amount of BOD remaining decreases and the overall
substrate removal rate decreases until an equilibrium level is reached.
Beyond this equilibrium point the effluent BOD concentration remains con-
stant and the removal rate continues to decrease for a period.
The concentration of microorganisms in the activated sludge process can be
controlled to some extent by the rate of sludge wasting. However, in the
aerated lagoon process the level of microorganisms is self-controlled and
is affected by the overall detention time. The rate of growth of microorgan-
isms in the activated sludge process was presented in equation 2-8 as a
ratio of the weight of sludge produced during the process per weight of
sludge in the reactor, (•M . The inverse of the growth rate is in fact the
XAt
-------
UJ
or
Q
O
GO
C/)
I
O
C/)
LU
CL
I
LU
o:
TOTAL BOD INFLUENT
EQUILIBRIUM
POINT
AERATION TIME
mn.
AERATION TIME
FIG.2-2
REMOVAL RATE AND REMAINING
BOD AS FUNCTIONS OF AERATION TIME
-------
sludge age (G) of the solids in the activated sludge process which represents
the length of time the sludge is in the system. The sludge age is equal to
the hydraulic detention time in flow through processes such as the aerated
lagoon process. Equation 2-8 can be rewritten in terms of sludge age by using
Equation 2-12 to define the substrate removal rate as Equation 2-18. This
equation can be further reduced to Equation 2-19 which can be rearranged and
the following expression for effluent substrate concentration results.
2-18
-j- = akS - b 2-19
S =-^~ +^- 2-20
akt ak
This equation indicates that the effluent soluble substrate concentration in
a flow-through system such as the aerated lagoon and certain waste stabil-
ization ponds is a function of only the detention time and is independent of
influent concentration. The suspended mixed liquor solids or algae which
develop in the process are not included in the substrate. The effects of the
influent and effluent substrate concentrations and the required detention
time can be calculated by a trial and error approach or by a graphical sol-
ution using Equations 2-21 and 2-22.
a (S - S)
X = 2-21
a 1 + bt
,S - S.
x = J_2 1 2-22
a Skt
in which X indicates the active microorganism population in the process.
The curves3presented in Figure 2-3 are based on an experimentally deter-
mined constant, k=0.20, a=0.65, and b = 0.15 and represent the interrela-
tionship among the active mixed liquor suspended solids, the remaining BOD,
the detention time and the incoming BOD for an aerated lagoon system. The
data indicate that:
(a) The mass of active mixed liquor suspended solids is dependent
on the concentration of the BOD in the incoming wastewater.
(b) There is a maximum concentration of active mixed liquor sus-
pended solids which occurs at a one-day detention time for
influent BOD concentrations from 50 - 100 mg/1.
U3
-------
cr 120
0=0.65
b=O.I5
150 mg/l BOD
^,
if. • \
\/. N
100 mg/l BOD
50 mg/l BOD
••^
l.o days
0
O.i
0.2 0.3 0.4 0.5
BOD REMAINING (S/S0)
FIG.2-3
AERATED LAGOON INTERACTIONS
-------
(c) At the one-day detention time, the concentration of active
mixed liquor suspended solids increases with the increasing
concentration of incoming BOD.
(d) The percent BOD remaining increases from about ten percent
to about 20 percent as the BOD coming into the system decreases
from 150 to 50 mg/1 at a one-day detention time.
(e) The concentration of active mixed liquor suspended solids
decreases as the detention time increases above one day. This
decrease in suspended solids is a result of endogenous respira-
tion or autooxidation of the synthesized cellular material.
(f) The active mixed liquor suspended solids concentration is also
lower at detention times less than one day. This phenomenon
may be attributed to the fact that sufficient time is not avail-
able for synthesis of the incoming BOD to cellular material.
At the shorter detention times the removal of BOD is only in
the range of from 50 to 80 percent. However, the concentration
of active mixed liquor suspended solids is not markedly differ-
ent at the half-day detention time than at the one-day detention
time.
(g) The maximum removal of BOD occurs at the maximum active
mixed liquor suspended solids concentration and increases
as the detention time increases beyond that time at which
the maximum active mixed liquor suspended solids concen-
tration develops. At the maximum mixed liquor suspended
solids concentration, the removal of BOD is in excess of
90 percent.
The maximum active mixed liquor suspended solids in aerated lagoons, is
important when considering the kinetics of substrate removal. The data
indicate that at detention times longer than that required for development
of maximum mixed liquor suspended solids concentration, and in this case
more than one day,the rate of substrate removal decreases. The results of
full-scale experimentation with aerated lagoons at detention times between
three and eight days indicate that the soluble BOD in the effluent was about
the same in all cases. The detention time at which the maximum active
mixed liquor suspended solids concentration occurs is similar to the equil-
ibrium point illustrated in Figure 2-2. At higher concentrations of mixed
liquor suspended solids in the activated sludge process, the equilibrium
point is reached at shorter detention times than in the aerated lagoon pro-
cess. However, it could be expected that the equilibrium point in the
activated sludge process and in the aerated lagoon process occur at about
the same sludge age.
-------
Smith and filers (1969) reported that the dotd observed
-------
QX
AX.
n
in which:
W
X
Q
X
(
X.
X
V
t
RS
AXn
therefore:
X V
a
Xt
*«
Xt
X V
a
Xt
Xt
2-24
waste sludge flow (MGD)
concentration of suspended solids in return sludge (mg/1)
wastewater flow (MGD)
concentration of effluent suspended solids (mg/1)
MLSS concentration (mg/1)
Mass of solids in aeration tank (Ib)
Volume of aeration tank (million gallons)
aeration time (days)
active biological solids formed (Ib)
nonbiodegradable suspended solids accumulated (Ib)
sludge wasted daily (Ib/day)
sludge lost daily in effluent (Ib/day)
sludge mass in aeration tank (Ib)
rate of biological solids formation (Ib/lb - day)
rate of accumulation of nonbiodegradable solids
(Ib/lb - day)
The sludge age (G) in days may therefore be expressed as:
f-*
X V
a
WXRS * QXe
2-25
-------
and Equation 2-24 may be transformed to:
, AX AX
i_ = _ a _ n
G Xt Xt 2"26
These mathematical models do not include the nondegraded biodegradable
suspended solids since at the range of sludge ages commonly used in
practice, the quantity of these solids is approximately constant and are
included in the nonbiodegradable fraction. The rate of formation of bio-
logical solids can be determined from Equation 2-27 which includes the
total suspended solids concentration in the mixed liquor.
AX S - S*
_ §_=a* LO - e)
Z Z/
Xt Xt
in which:
a* = synthesis of soluble BOD removed to suspended
solids
b* = rate of endogenous respiration of active solids
(Ib/lb - day)
fc - S*)
o e
- — - = rate of BOD removal (,, .1D . - r
Xt lb MLSS - day )
S* = filtered effluent BOD
"
The excess sludge includes the influent nonbiodegradable suspended solids
and the rate of accumulation may be expressed as:
AX QX X
_n = _2n =_on_
Xt XV Xt z°
in which:
X = nonbiodegradable suspended solids in the influent
(mg/1)
-------
Equation 2-26 may be rewritten as:
= a*
G
S_ c; *
o .
L2 - e
Xt
X
- b*
on
Xt
2-29
In this equation, the effluent concentration is expressed as the BOD of the
filtered sample.
Typical values of the constants in Equation 2-29 are a * = 0.6 to 0.65 pounds
of active cells formed per pound of BOD removed, b* = 0.075 pound per
pound per day and the nonbiodegradable suspended solids in the influent
are equal to 60 percent of the total solids in the influent. These data may
be used for preliminary calculations or for checking existing calculations
if no other information is available. This model for excess sludge production
should be applicable to all aerobic processes; however, there are some
limitations. In low rate trickling filters and in stabilization ponds, algae
and other microorganisms are produced and tend to control the quantity of
solids which are generated. Therefore, Equation 2-29 is not applicable
in these cases in the present form.
OXYGEN UPTAKE
The general equation for oxygen uptake rate was developed and presented
as equation 2-7; however, this equation can be rewritten in a more practical
form as Equation 2-30.
R = a1
,S - S*
( o e)
+ b'X
2-30
in which:
R
S - S*
o e
X
b1
Oxygen uptake rate (Ib/day)
BOD removal rate (Ib/day)
mass of mixed liquor suspended solids (Ib)
oxygen required for synthesis of BOD removed
,lb Oxygen.
Ib BOD
.
oxygen required for endogenous respiration
, Ib Oxygen ,
llb MLSS-day
-------
Typical values of a' and b1 for municipal wastewater are a'=0.53 (0.40 to
0.65) and b'=0.15 per day, respectively. The value for b' is affected mark
edly by the temperature; and the value presented is typical of summer condi
tions and is widely used. Equation 2-35 represents the oxygen required
for the biodcgradation of organic material; however, at low organic loading
rates , nitrification can be expected. The oxygen requirements for nitri-
fication may be expressed as a stoichiometric relationship developed by
Downing (1964):
2 NH. + 4 0_ - > 2 NO + HO + 4 H+ 2-31
ft ^ O £
This reaction indicates that approximately 4.6 pounds of oxygen were
required to convert one mole of ammonia -nitrogen to nitrate nitrogen. The
oxygen per pound of nitrogen converted can be calculated as — = 4 . 60 .
The rate of oxygen uptake for nitrification can be expressed as Equation
2-32.
ANH
4.60--3 2-32
in which:
R = oxygen uptake rate for nitrification (Ib/day)
ANH = ammonia nitrogen removed (Ib)
O
t = aeration time (days)
In those treatment plants which have a low organic load to the aeration
system, partial nitrification has been reported. Consequently, if nitrifi-
cation is not included in calculations of oxygen required, higher values
of the coefficient, a', must be used in Equation 2-30. The total oxygen
including nitrification can be expressed as:
S - S* ANH
R=a'(°t 6) +b'X + 4.60— j-^- 2-33
DESIGN EQUATIONS AND APPLICATIONS
The various design equations for biological treatment processes are summarized
in Table 2.1. The applicability of the various equations and limitations are
also included in this table. Typical values of the constants used in the equa-
tions are presented .
50
-------
TABLE 2-1
S -kXt
So
S -kAvDQ~n
0
S 1
S 1 + kXt
o
S 1
so i + kt
„ 1 , b_
a akt ^ ak
S - S*
- a ( • • ' ) + b X
S -S* ANH
= a (• ) + b X + 4. CO
. S - S*
1 , o e s
G * ( X t } -b
. S - S* X
1 ^r ° ei i -* -i- on
G ~ a ( x t } - b + xt
S + X
1 a* ( ° ° ) -b*
G v Xt '
t - t ft20 ' T
'T ~ 120 &
Q20 " T
kT ~ 20W
Activated
Sludge
Batch tests plug
flow system
uniy nign ioau
rates G (1 day
_ 1 f\ C O
a — u . o o
b1 = 0.15
a — u . jj
b' = 0.15
a = 0.75 (settled)
a — \ 10 (raw\
b = 0.075
a* = 0.63; b* = 0.075
B3 U . D A
on o
a* = 0.6
b* - 0.075
9 = 1.0 - 1.04
Aerated
Lagoons
Detention time < 2
days k = 8-15
only t (2 days
a-0 . 63 , b-0 .15
k=0.10-0.20
-.1 n ";i K1 n i "»
Completely mixed
lagoons only
ai nQQ v.1 ni^
Completely mixed
lagoons only
9 = 1.03 - 1.10
Trickling
Filter
Modified form
S*/SQ* = 0.50
k* = 0.010-
0.020 (0=MGAD1
design
design
in a modified
in a modified
form
9 = 1.035
Waste
Stabilization
Pond
Ponds in series
ajd. baffled
lab units
Overall performance
~ 0 .1 - 0.2
9 = 1.085
-------
MIXING MODELS FOR REACTOR
Reactors used in wastewater treatment are aeration tanks, aeration lagoons,
trickling filters , waste stabilization ponds and anaerobic digestion tanks .
Mixing is essential to the maintenance of a high rate and a high degree of
biodegradation. In the activated sludge process and the aerated lagoon
process , mixing is accomplished by aeration which also provides the oxygen
essential to the function of the process. Waste stabilization ponds are
mixed by the action of wind along the surface which tends to turn over the
entire body of water. The microorganisms in the trickling filter are fixed as
slime layers on the surface of the packing medium. Therefore, the mixing
is accomplished by distributing the incoming wastewater uniformly over the
entire surface of the filter medium.
The kinetic models presented earlier in this chapter describe either of two
idealized flow systems, namely the completely mixed or the plug flow
system. In the completely mixed system, the incoming material is instan-
taneously distributed throughout the entire tank volume; therefore, the
concentration in the tank is the same at all locations. Typical flow patterns
for the two types of systems are illustrated in Figure 2-4. Theoretically,
no longitudinal mixing takes place in the plug flow system, and any material
introduced with the influent will be associated with this liquid parcel as
it passes through the tank. An aeration tank in which the waste is intro-
duced at one end and travels longitudinally to the effluent at the other end
and trickling filters may be considered plug flow reactors.
Longitudinal plug flow-type tanks were originally preferred for the activated
sludge process because early investigators indicated that the performance
of the system was primarily based on the aeration or contact time of the
biological solids in the mixed liquor and the substrate in the wastewater.
In the plug flow system, the aeration time was predictable; however, the
possibility of shortcircuiting was expressed by Kurd (1929) , Calvert and
Bloodgood (1934) , Haseltine (1932) , Kessener (1935) , and Kehr (1936) . The
more recent literature indicates that extensive studies in which tracer re-
sponse techniques were applied in order to determine the "residence time
distribution" of fluid elements in the reactor have been published.
Several models have been proposed to characterize the non-ideal flow patterns
in aeration tanks , namely:
(a) subdivided tank model
(b) tanks-in-series model
(c) dispersion model
52
-------
PLUG FLOW
RETURN SLUDGE
WASTEWATER
RETURN SLUDGE
WASTEWATER
COMPLETELY MIXED
4
4
4
4
4
4
4
4
4
4
COMPLETELY MIXED
RETURN SLUDGE
WASTEWATER
FIG.2-4
AERATION TANK TYPES
53
-------
The subdivided tank rnodol do.sc:ribos Lhf: conditions which exist, if the
original tank was subdivided into smaller individual flow regions connected
in series and/or in parallel. Each of the individual regions is usually re-
stricted to the simple case of plug flow, completely mixed, single parameter
dispersion, and dead space. The type of flow connecting the various regions
can be categorized as cross-flow, bypassing flow, and recycle flow from
end to beginning of the regions. These generalized models have a large
number of parameters which must be evaluated and in practice are difficult
to determine. Levenspiel (1962) reported that "an unrealistic many para-
meter model may closely fit all present data after the fact, but may be
quite unrealistic for prediction in new untried situations."
The tank-in-series model is somewhat simpler to understand. An aeration
tank is visualized as a number of equal-volume completely mixed compart-
ments. The effluent concentration of a tracer for this particular model can
be calculated by the following equation.
C nn (JL, ""' e -nt/1 2_34
Co (a - 1)! >-
in which:
t = time measured from the initial addition of the tracer
(hours)
t" = the theoretical detention time = the volume of the tank
divided by the flow (hours)
C = tracer concentration at the exit of the tank measured at
time t (mg/1)
C = average tracer concentration = mass of tracer added
° divided by the tank volume (mg/1)
n = number of tanks in series
This model describes the mixing in an aeration tank in terms of one single
parameter, n, the number of completely mixed tanks in series. The value
of n may be determined from experimental response curves by comparing
dimensionless response curves with a set of curves with various values of n.
2
Levenspiel (1962) related the number of tanks to the variance a of the
experimental response curve as shown in Equation 2-35.
-------
2-35
The number of tanks calculated from liquation 2-35 will not in general be;
an integer. This paradox cannot be theoretically justified because in
equation 2-34, n is assumed to be an integer.
Thomas and McKee (1944) reported good correlation between experimental
and theoretical response curves which were predicted by Equation 2-34
for a compartmentalized laboratory aeration tank. This model was also
used for the evaluation of data observed during mixing studies in aerated
lagoons at the Williamson Creek Wastewater Treatment Plant in Austin,
Texas. These data are presented in Figure 2-5. The recovery of the tracer
in the effluent (C/C ) is plotted versus the ratio detention time t/ F. The
data indicate that at energy inputs of 0.03 HP/1000 gallons of tank volume
the tracer recovery is very close to the theoretical recovery curve. The
initial peak seems to be the result of some short circuiting. The results
observed at a power level of 0.015 HP/1000 gallons indicate a deviation
of the actual tracer recovery curve from that expected for an ideal com-
pletely mixed basin. Therefore, at the lower power level a number of
completely mixed tanks-in-series would better represent the situation.
The theoretical number of tanks-in-series was 1.42, 1.2, and 4.1, respec-
tively at power levels of 0.054, 0.032, and 0.015 HP/1000 gallons . The
actual value of the numbers of tanks-in-series should not be over-
emphasized since a repetition of the test could result in different distri-
bution and the actual value may be higher or lower than a previous test.
These results do indicate that at the higher power levels , this particular
tank will function more like a completely mixed system than at lower
power levels .
The dispersion model was considered to describe longitudinal mixing in
reactors by Dankwerts (1953) , Levenspiel (1957, 1959, 1962) and Van der
Laan (1958) . An analogy between mixing in actual flow and molecular
diffusion was presented; therefore, back mixing of fluid flowing in the
axial direction may be represented by an equation similar to Pick's second
law of molecular diffusion. Equation 2-36 represents the mathematical
model considering a uniform intensity of backmixing .
2-36
55
-------
OO54? HP / IOOO t,Al
NUMBER OF TANKS I V,;
10 20 3040
O0322 HP / IOOO GAL
NUMBER OF TANKS = 1201
0.6 08 1.0 20 30 40
£ 20
co
OOI52 HP/ IOOO GAL
NUMBER OF TANKS =4 147
I O j
IDEAL COMP MIXED
MEASURED
0 02 04 06 08 10 20 3040
TIME t/t
FIG. 2-5 RESULTS OF TRACER STUDIES
FROM AERATED LAGOONS
-------
in which:
C = the concentration
t = time
x = longitudinal distance
D = longitudinal dispersion coefficient which characterizes
the degree of backmixing
The application of this model has been previously limited to flow through
tubular and packed bed reactors. The dispersion term includes molecular
and turbulent or eddy diffusion. The term longitudinal is used in order to
distinguish between radial mixing or dispersion. The application of the
dispersion model assumes that lateral mixing is sufficient to insure a
uniform concentration of tracer at any given cross-section. Solutions to
the second order differential equation for a tracer pulse input to a closed
vessel were developed by Thomas and McKee (1944) and Miyauchi (1953) .
A dispersion number was developed as a constant in the solutions of these
equations. This dispersion number is presented in Equation 2-37.
_D_
pL 2-37
in which:
D = dispersion coefficient (sq ft/hr)
^i = bulk velocity along the tank length = flow rate of water
divided by tank cross section (ft/hr)
L = tank length (ft)
Van der Laan (1958) and Timpany (1966) developed techniques for estimating
dispersion number from experimental response curves. The dispersion
coefficients can be derived from this dispersion number.
Boyko (1968) , Boyko and Murphy (1968) and Timpany and Murphy (1967)
reported the results of measurements on laboratory and full-scale spiral
flow aeration tanks and the conclusions are summarized as follows:
(a) The dispersion coefficient is dependent on the geometry of
the tank cross section and the specific air flow rate. The
57
-------
following expression may be used to calculate the dispersion
coefficient.
„ 110T,r2 0.346 2-38
D = 3.118 W q
in which:
W = tank width (ft)
q -- specific air flow rate (scfm/1000 cu ft of tank volume)
f\
The dispersion coefficients measured for full-scale tanks
ranged from 5000 to 7500 sq ft/hr.
(b) The dispersion coefficient was not affected by water tempera-
ture in the range of 12 to 30 C .
(c) There was no significant difference in the value of the dis-
persion coefficient observed in experiments using coarse
bubble aeration (sparger) and fine bubble aeration (saran
wrapped tubes) at the same air flow rates .
The dispersion coefficient also represents the degree of backmixing in a
particular cross-secion of the tank. The dispersion number, however, is
dependent on the dispersion coefficient, detention time, and tank length.
The dispersion number can be expressed in terms of the average time:
P_=Dt 2-39
uL 2
4-J
Substituting the general expression for the dispersion coefficient, the
dispersion number can be redefined as:
a-- 3.118 ft' ?n°-346 2-40
uL LA
The constants in Equation 2-40 must be verified experimentally; however,
the terms 3.118 and 0.346 are presented at this point to indicate the rela-
tive degree of influence that each constant has on the dispersion number.
Therefore, air flow rate has a relatively small influence on the mixing in
a spiral flow tank compared to other variables . The tank geometry on the
other hand has the most influence on the mixing. Shallow wide tanks with
long detention times more closely approximate the completely mixed system
58
-------
in which the dispersion number (——) approaches infinity. On the other
hand, long narrow tanks with relatively short detention times more closely
approximate plug flow conditions in which the dispersion number approaches
zero. The results of tracer studies in one aeration tank at the Govalle
Wastewater Treatment Plant in Austin, Texas indicate that dispersion coef-
ficient was equal to about 3500 sq ft/hr (Halbert and Malina , 1970) . This
value of dispersion coefficient is much lower than those reported by
Murphy, Boyko, and Timpany (1967, 1968) .
The method of evaluating the data from tracer studies has considerable
influence on the results obtained. The aeration tank at the Govalle Plant
is not a single straight longitudinal tank but is divided in the middle so
that the flow pattern is around the end. The transition in flow pattern oc-
curring at the mid-point of the tank may have an effect of the dispersion
coefficient. The degree of backmixing at the opening between the two tanks
may have been less than expected in the normal tank cross-section. The
cross-sectional areas of the tank and the opening were 375 square feet
and 150 square feet, respectively. The measured dispersion number was
0.06 sq ft/hr and indicates that the tank represents a plug flow system.
The measured response curves for around the end tanks i s shown in Figure
2-6. Each curve represents the response curves at different stations located
along the length of the tank. Station 6 was located in the effluent of the
tank and this response curve indicates that actual detention time was 0.86
of the theoretical detention time. Station 1 was located at a distance
equal to a volume which represents about 40 percent of the theoretical
detention time.
Batch biological reactors were also operated during the tracer studies in
order to determine a biodegradation rate constant. Theoretically the
quantity of substrate remaining along the length of the tank is predictable,
if the biodegradation rate constant and the dispersion number are known.
Dankwerts (1953) developed an equation to incorporate these two parameters
and a solution for this particular equation was developed, Wehener and
Wilhelm (1957) . This equation was used to evaluate the rate constant data
and the dispersion number; however, the measured soluble substrate con-
centration along the length of the tank decreases at a much faster rate
than the calculated value (Halbert and Malina, 1970).
Additional mixing studies should be carried out in tanks of different geometry
and with different diffuser arrangements in order to determine the condition
of flow. In aerated lagoons and tanks operated at low power levels, short-
circuiting can easily be detected by tracer studies. These tracer studies
should be conducted at conditions as close to steady state operation as
59
-------
6.0
'o
AERATION TANK "GOVALLE
0
0
0.4
FIG.2-6
TRACER RESPONSE CURVES
60
-------
possible. Development of dissolved oxygen profiles, substrate profiles
and determination of rate constants should accompany the tracer studies
In aerated lagoons the tracer study data should be supplemented by pro-
files of suspended solids at various locations and at different depths in
the lagoons .
6l
-------
ACTIVATED SLUDGE PROCESS
PROCESS DESCRIPTION
The activated sludge process is probably the most commonly used system
for the biological treatment of municipal and industrial wastewaters. The
essential components of the activated sludge plant are an aeration tank with
the necessary aeration equipment, a final clarifier in which the biological
solids can be separated from the liquid effluent, and the necessary pumps
to recycle the concentrated solids to the aeration tank. The incoming waste-
water is mixed with the returned sludge solids in the aeration tank and this
mixture is aerated for about two to six hours. The intensity of aeration
must be sufficient to mix the tank contents. The mixture of incoming waste-
water and the activated sludge solids is commonly called the mixed liquor.
Most of the settled sludge is returned from the clarifier to the aeration tank;
however, a portion of the sludge is wasted. This excess sludge is a result
of bacterial growth as well as the accumulation of nonbiodegradable sus-
pended solids which enter the system.
The activated sludge contains a mixture of a wide variety of microorganisms
which are capable of degrading different organic compounds. Inactive sus-
pended materials also accumulate in the biological floe which forms during
the aeration. The bacteria and other microorganisms in the activated
sludge are generally found in most municipal wastewaters. The mass of
activated sludge increases as the microorganisms degrade the organic
material in the wastewater and grow in the aerobic environment of the aera-
tion tank. The bacterial growth rate during the initial stages is very high
because the substrate concentration is relatively high compared to the
bacterial population. After a week or two of operation of the activated
sludge treatment plant, the desired mixed liquor suspended solids concen-
tration and the desired degree of treatment are attained.
The activated sludge process has been modified in numerous ways since
1914 when the process was first applied to wastewater treatment. In the
original process which is called the conventional activated sludge process,
the wastewater and return sludge entered at one end of a long narrow tank.
The mixed liquor flowed in the longitudinal direction and with the help of
the diffused air aeration system, a spiral flow resulted. The mixed liquor
left the other end of the long narrow tank..
A number of the modifications in the activated sludge process are illustrated
in Figure 3-1. The basic differences occur in the point at which the incoming
63
-------
CONVENTIONAL PROCESS
WASTEWATER
SLUDGE RtTURN
STEP AERATION
WASTEWATER
I 4 4.
SLUDGE RETURN
CONTACT STABILIZATION
WASTEWATER
SLUDGE RETURN
CONTACT STABILIZATION
WASTEWATER
SLUDGE RETURN
COMPLETE MIXED
WASTEWATER
SLUDGE RETURN
FIG. 3-1 AERATION TANK ARRANGEMENTS
-------
wastcwater is introduced into the aeration tank and mixed with the return
sludc, 3, the reaeration of the return sludge prior to coming into contact with
the incoming wastewater and the type of flow pattern that might be typical
of the system and operation.
Aeration of the return sludge has been practiced for many years. The original
purpose for reaeration was to maintain the return sludge aerobic because the
sludge would turn anaerobic in the final clarifier since the quantity of return
sludge was relatively small. Kraus (1945) and Hatfield (1931) pumped anaer-
obic digester supernatant into the reaeration tank and effectively treated the
supernatant with the return sludge. In both cases the concentration of
mixed liquor suspended solids was high and a long aeration time was main-
tained. This type of system also minimized the shock loads to the aeration
tank since the flow of digester supernatant was distributed more uniformly
throughout the day and a higher concentration of solids was available to
utilize the organic material in the supernatant.
The contact stabilization process was introduced independently by Ullrich
and Smith (1951) as well as by Eckenfelder and Grich (1956) . This process
was based on observations made in the conventional activated sludge
system which indicated that the soluble fraction of the organic material in
wastewater was almost entirely removed after very short aeration times.
Therefore in the contact stabilization process, the incoming wastewater
is mixed with the return reaerated sludge for very short times of about 30
minutes to two hours. The sludge solids are then separated from the liquid and
the concentrated sludge is reaerated prior to being mixed with the incoming
wastewater. The aeration times in the contact zone must be long enough
to permit a reduction of the concentration of soluble orgamcs to the pre-
determined effluent concentration. However, in the contact zone, the
nonsoluble colloidal and particulate materials are also incorporated in the
activated sludge floe particles and are removed in the final clarifier. The
sludge is aerated for an additional two to four hours in the stabilization or
reaeration tank. In many plants, the stabilization zone and the contact
zone are not physically separated; therefore, it is difficult to calculate the
exact time which may be attributed to contact and to stabilization.
The completely mixed system has been used extensively in laboratory-scale
units in evaluating the various parameters which affect the activated sludge
process. However, this process has only been recently installed for the
treatment of municipal and industrial wastewaters. The development of
surface and mechanical aerators which may be used separately or in con-
junction with diffused air systems has made the use of completely mixed
tanks more popular. In the completely mixed tanks, the concentration of
oxygen and of the substrate is constant at each point throughout the entire
tank volume. The system permits automatic control of oxygen transfer and
65
-------
is relatively resistant to shock loading since the influent wastewater is
distributed uniformly throughout the entire tarn; volume .
e
extended a';rn process ir; a modification (>\ the activated sludge
process and is cruimcteri/.ed by relatively low organic loadings; therefore,
the sludge which is produced during biodegradation is also stabilized to
some extent in the aeration tank. However, in the extended aeration pro-
cess the nonbiodegradable solids tend to accumulate and must be removed
periodically.
The activated sludge process can be operated as the conventional activated
sludge process and the contact stabilization process. The flow pattern in
the aeration tanks can be categorized as the plug flow or the completely
mixed. The performance of the various process modifications and flow
schemes can only be compared if each system is operated at an identical
organic loading and with identical wastewater. The results of laboratory-
scale experiments comparing contact stabilization and the conventional
activated sludge processes indicated that the effluent BOD concentration
for each process was almost identical at organic loading rates of 0.40 to
1.90 pounds of BOD/pound of mixed liquor suspended solids -day (Water
Pollution Research Laboratory, 1967 , 1968) .
PERFORMANCE OF THE ACTIVATED SLUDGE PROCESS
The first order substrate removal model does not normally apply for the
organic loadings at which most activated sludge plants are operated.
Therefore, data are compared on the basis of detention time, BOD volume
load (pounds of BOD per 1000 gallons of tank volume per day) and organic
loading (pounds of BOD per pound mixed liquor suspended solids or per
pound of mixed liquor volatile suspended solids per day). However, neither
the MLSS nor MLVSS concentrations define the inert fraction of the solids
present in the mixed liquor. The nonbiodegradable solids in the mixed
liquor include the inert suspended solids in the influent and in the return
sludge and may be represented as Equation 3-1 which can be modified and
used to calculate the nonbiodegradable suspended solids.
QX = WX -:- QX 3-1
on n en
X =X .-p- 3-2
n on t
in which:
X ,X = influent and effluent nonbiodegradable suspended
on en solids (mg/1)
66
-------
X - nonbiodegradable suspended solids in the aeration
tank (mg/1)
G ;- sludge age (days) (total suspended solids in the. system
divided by the total suspended solids leaving the system
per day)
t = theoretical detention time (days) (aeration tank volume
divided by wastewater flow rate)
The interrelationship between the concentration of nonbiodegradable sus-
pended solids in the mixed liquor, the sludge age, and the theoretical
aeration time are shown in Figure 3-2. These curves are based on an
influent nonbiodegradable suspended solids concentration of 50 mg/1.
Therefore, at a sludge age of four days and a theoretical detention time
of six hours the accumulation of nonbiodegradable suspended solids in
the mixed liquor is about 750 mg/1.
The relationship of the nonbiodegradable fraction of the mixed liquor
suspended solids and the ratio of the influent BOD to influent suspended
solids is presented in Figure 3-3. These data are from pilot-scale studies
conducted by Wuhrmann (1964) a.id from the experimental work conducted
at the Govalle Treatment Plant in Austin, Texas,and the Hyperion Treatment
Plant in Los Angeles, California. The accumulation of nonbiodegradable
suspended solids in the mixed liquor range from 40 to 70 percent of the
total mixed liquor suspended solids and was affected by the ratio of in-
fluent BOD to influent suspended solids (so/xo) which ranged from 0 .5 to
1.75.
The curves in Figure 3-3 reflect the efficiency of primary clarification at
the particular treatment plant. The relatively high percentage of nonbiode-
gradable solids in the mixed liquor at the Govalle Plant is indicative of no
primary clarification. The relatively low ratio reported for the Hyperion
Plant is typical of efficient primary treatment. The data reported by Wuhrmann
(1964) for the operation of a pilot plant reflect the different analytical pro-
cedures used for determining suspended solids. The Wuhrmann data are
based on a membrane filtration technique which captures about 10 to 15
percent more of the suspended solids than the glass fiber or paper filter
techniques used at the Govalle and Hyperion Plants. If the results reported
by Wuhrmann are adjusted, the data would compare quite favorably.
The operating conditions under which these data were collected were influent
BOD of 90 to 200 mg/1, and influent suspended solids concentrations of 70
to 250 mg/1. The mixed liquor suspended solids concentrations was 600 to
6000 mg/1 and the BOD loading rate varied from 0.17 to 2.0 Ib BOD/lb MLSS-
day. The curves indicate that the influence of the loading is relatively
67
-------
1*4000
LU
mo 3000
CT
C/)
2000 -
1000 -
0
6 8 10
SLUDGE AGE
FIG.3-2
NON-BIODEGRADABLE SUSPENDED
SOLIDS IN MIXED LIQUOR
68
-------
CO
CO
_J
o
70
60
o
cr
u_
UJ
CD
50
Q
<40
O
LJ
Q
O
CD 30
I
o
X
2°
0
0
0.5
\
\
\
A \
\
\
AAA
WUHRMANN
GOVALLE
HYPERION
.0
.5
2.0
S0/XQ RATIO BOD AND SS INFLUENT
FIG.3-3
NON-BIODEGRADABLE SOLIDS OF
MIXED LIQUOR
69
-------
negligible At times, it may be useful to compare data based on the mixed
liquor active suspended solids concentration (MLASS) which can be obtained
by subtracting the nonbiodegradable fraction from the total suspended
solids concentration. For example, at a loading rate of 0 .3 Ib BOD/lb
MLSS - day, the rates of BOD removal based on the active mixed liquor
suspended solids concentration would be 0.55 and 0.85 Ib of BOD/lb
MLASS - day, respectively. The ratio of MLASS to MLSS for Hyperion and
Govalle were 0.55 and 0 .35, respectively. These data indicate that the
loading rate at the Govalle Plant is about 60 percent higher at the Hyperion
Plant. Loading based on volatile suspended solids would exhibit similar
differences. However, the mixed liquor suspended solids (MLSS) concen-
tration is generally the only measure of the solids in the aeration tank at
most plants; therefore, a comparison of performance data for various plants
must be based on the total mixed liquor suspended solids .
The relationship between the effluent filtered BOD and the BOD loading
(Ib BOD/lb MLSS - day) is illustrated in Figure 3-4. The BOD loading
ranged from 0.2 to 1.0 Ib BOD/lb MLSS - day. The effluent filtered BOD of
the Govalle samples scattered widely from about three mg/1 to over 11 mg/l
for loadings of less than 0.2 to about 0.5 Ib BOD/lb MLSS - day. The
scatter in the data from these experiments was considerably greater than
that for the data observed at Hyperion which indicated a slight increase in
the effluent BOD as the loading increased from 0.2 to almost 1.4 Ib BOD/lb
MLSS - day. The Hyperion data seemed to indicate a linear relationship
between the effluent filtered BOD and the BOD loading range.
It should be pointed out that data presented in Figure 3-4 are not comparable
theoretically since the Hyperion Plant and one of the three plants at the
Govalle Plant were operated as the conventional activated sludge process
whereby the other two plants at the Govalle Plant were operated as the con-
tact stabilization process. If one considers that the removal rate of soluble
substrate is not affected by the nonsoluble organic material, the relation-
ship which results comparing the soluble BOD loading rates and the effluent
soluble BOD concentration can be calculated and plotted as shown in
Figure 3-5. The curves indicate that there is a linear relationship between
the effluent soluble BOD and the soluble BOD loading. The loading rates
for the contact stabilization plant were based only on the volume of the con-
tact zone, the MLSS in the contact zone, and the incoming wastewater flow.
These data indicate that the concentration of soluble material in the effluent
increases as the loading expressed as soluble BOD also increases. In
general, at the same soluble BOD loading the residual BOD at the Govalle
Plant was higher than that reported at the Hyperion Plant. The results of
laboratory-scale experiments at the Govalle Plant indicate that the soluble
BOD in the effluent of contact stabilization process was between three and
six mg/1 at soluble organic loading rates of 0.3 to 2.3 Ib BOD/lb MLSS - day
70
-------
12
o
o 8
CD
Q
LU
o:
I4
u.
u_
UJ
0
0
GOVALLE
HYPERION
0.2 0.4 0.6 0.8 1.0
BOD LOAD TOTAL (Ib/lb/day)
FIG.3-4
TOTAL LOAD AND SOLUBLE
EFFLUENT
71
-------
12
1*10
o
o
00
S 8
o:
LU
n- 6
h-
z
UJ
3 4
UL
LU
0
0
GOVALLE
HYPERION
0.2 0.4 0.6 0.8 1.0
SOLUBLE BOD LOAD (Ib/lb/day)
FIG.3-5
SOLUBLE LOAD AND EFFLUENT
72
-------
(Berryhill, 1970) . These residual BOD concentrations are in the same range
as that reported for the Hyperion Plant effluent. The relationship between
the effluent soluble COD and the soluble COD loading rates are presented
in Figure 3-6. These data indicate that the effluent soluble COD increases
at n much more rapid rate with increasing soluble COD loading at the Hyperion
Plant than at the (kwallc Plant.
The discrepancy between the laboratory-scale data and the full-scale data
observed at the Govalle Plant may be explained by the fact that the laboratory-
scale runs were performed about six months after the full-scale experiments,
and the composition of the wastewater was different. The relatively high
effluent BOD at the Govalle Plant indicate that these residual organic materials
are degradable but at an extremely show rate.
The more useful presentation of data for practical purposes is the relationship
between the total effluent BOD and the loading rate expressed in terms of
total BOD entering the plant. This relationship is presented in Figure 3-7,
for data observed at the Hyperion Plant, the Govalle Plant, and the pilot-
plant studies by Wuhrmann (1964) . These data scatter considerably over
a loading rate of about 0.1 to almost 2.0 Ib BOD/lb MLSS - day. The re-
sults of the three different experiments should be discussed individually.
The results for the Hyperion Plant were reported by Smith and Eilers (1969) .
The wastewater flow for all the experiments was 50 MGD. The theoretical
aeration times were four, five, and six hours, respectively, and were main-
tained by using different numbers of aeration tanks. The loading rate was
changed by varying the mixed liquor suspended solids concentration in the
aeration tank from 600 to 3500 mg/1. The surface area of the settling tanks
were constant for all experiments and provided an overflow rate of about
530 gallons/sq ft-day. The increase in total BOD in the effluent was the
result of the increased effluent suspended solids concentration caused by
the increased loading rates.
During the Govalle experiments, the wastewater flow varied slightly and
the organic loading was changed by controlling the mixed liquor suspended
solids concentration in the aeration tanks. The overflow rates for the
clarifiers fluctuated from 750 to 1000 gallons/sq ft - day. Therefore, part
of the scatter of the effluent BOD is caused by the suspended solids in
the effluent. However, during most of the runs, partial nitrification took
place. It is therefore safe to assume that the nitrification continued in the
BOD bottle resulting in higher BOD concentrations than would be expected
for only carbonaceous material.
Wuhrmann on the other hand, used three pilot plants in parallel in the
experiments in Zurich, Switzerland. The detention times used were 25, 50,
73
-------
60
50
o
o
LU
ID
_J
30
LU
LU
m 20
=>
O
10
0
GOVALLE
HYPERION
0
0.5 1.0 1.5 2.0 2.5
SOLUBLE COD LOAD (Ib/lb/day)
FIG.3-6
SOLUBLE COD LOAD AND EFFLUENT
-------
50
40
30
I-
z
UJ
u_
UL
UJ
Q
O
CD
20
10
0
GOVALLE
HYPERION
WUHRMANN
A
A
ABA
0
0.5
1.0 1.5 2.0
BOD LOAD (Ib/lb/day)
FIG.3-7
BOD LOAD AND BOD EFFLUENT
75
-------
and 125 minutes, respectively. The mixed liquor suspended solids concen-
trations were maintained at 3300 and 6000 mg/1, respectively. The overflow
rates to the clarifiers were about 530 gallons/sq ft - day which was about
the same as that reported for the Hyperion experiments. Therefore, the data
reported by Wuhrmann compare favorably with that reported by Hyperion.
The data presented in Figure 3-5 indicated that the soluble effluent BOD
concentrations only increased slightly with increasing loading. Therefore,
it can be expected that the higher total effluent BOD concentrations are
caused primarily by the increased suspended solids concentrations in the
effluent. The effluent suspended solids are plotted versus the effluent
BOD in Figure 3-8. These data indicate that the effluent BOD concentration
increases as the suspended solids concentration increases. The scatter
in this particular plot is caused by the number of factors including nitrifi-
cation in the BOD bottle which results in a relatively high BOD at relatively
low suspended solids concentration and the activity of the suspended solids.
The suspended solids of a highly loaded plant will exert a higher BOD in the
effluent than those solids from a relatively lightly loaded plant. For these
data the ratio of the concentrations of total BOD to the suspended solids
(BOD/SS) is in the range of 0.55 to 1.10.
The concentration of suspended solids in the clarifier effluent is affected
by the hydraulic overflow rate, the solids loading rate to the clarifier, and
the settling and flocculation characteristics of the sludge. At a constant
overflow rate and initial suspended solids concentration, the settling and
flocculation of the sludge are the only factors which influence the effluent
concentration. These sludge properties can be related to sludge age. There-
fore, the effluent concentration of suspended solids were plotted against
sludge age for the data reported by Wuhrmann and for the experiments at the
Hyperion Plant since the overflow rate for these experiments was the same
(530 gpd/sq ft). These data are presented in Figure 3-9 and indicate that
the effluent suspended solids concentration increases as the sludge age
decreases. However, the effluent suspended solids concentration would
increase sharply for a longer sludge age typical of the extended aeration
systems. Temperature is also a factor as indicated by higher effluent sus-
pended solids concentrations at lower temperatures as reported by Wuhrmann.
The performance of the activated sludge process treating municipal wastewater
is controlled in some respect by the concentration of suspended solids in
the effluent. The suspended solids in the effluent of a well operated second-
ary clarifier are essentially small floe particles (pin floes) and single micro-
bial cells. As a concentration of ciliated protozoa increases, the effluent
suspended solids decrease. These protozoa are predators of bacteria and
consume the fine floe particles. The growth rate of the protozoa in most
cases is much lower than that for the bacteria. Therefore, as the sludge
age increases, there is a better chance for the protozoa concentration to
76
-------
50
40h
o
-J 30
o
Q
Ld
Q
20
10
0
0
A •
A VvJHRMANN
• GOVALLE
. HYPERION
i
10
20
30 40
BOD (mg/i)
FIG.3-8
BOD AND SUSPENDED SOLIDS
EFFLUENT
77
-------
50-v
40
30
UJ
ID
St 20
UJ
10
0
• HYPERION
A WUHRMANN, SUMMER
V WUHRMANN, WINTER
A A
0
6 8 10
SLUDGE AGE (days)
FIG.3-9
EFFLUENT SS AND SLUDGE AGE
78
-------
increase resulting in a reduction in the concentration of suspended solids
in the effluent and an improved effluent quality. Temperature also affects
the performance of a clarifier. As the temperature decreases, the viscosity
of the water increases and thereby reduces the settling rate of the floe
particles. Therefore, a higher concentration of suspended solids in the
effluent is observed at low temperatures .
Lower overflow rates in the final clarifier did not seem to be the only
solution to increasing the efficiency of the final settling process. Other
remedial procedures may be dictated by the characteristics of the sludge.
In some cases, flocculating agents may be required to produce a more
stable and more settleable floe. The gases resulting from denitrification
or from the capture of air bubbles may cause floating floe particles , and
some way of releasing the gas is required to improve settling. Final
clarifiers may be plagued with density currents which affect the efficiency
of solids removal by possibly carrying suspended solids from the sludge
blanket up over the effluent weirs .
The mathematical model presented in Equation 3-3 may be used to calculate
oxygen requirements if no nitrification takes place.
,
"= a
S - S*
,o a
3-3
In practice the oxygen requirements per unit volume may be more useful.
This equation may be modified to include the oxygen requirements for nitri-
fication .
in which:
R
t
S
S*
e
X
ANH,
24
- S*)
24ANH
3-4
oxygen uptake per unit volume (lb/1000 gal/day)
theoretical aeration time (hours)
influent BOD (mg/1)
soluble effluent BOD (mg/1)
MLSS (mg/1)
ammonia removal (mg/1)
79
-------
a' = constant relating to the oxygen required for sludge
synthesis (for municipal wastewater = 0 .53 Ib/lb)
b' - endogenous oxygen uptake rate (for municipal waste-
water = 0 .15 Ib/lb/day) .
In most municipal biological treatment plants the residual soluble effluent
BOD is three to seven mg/1 which in fact is about five percent of the influent
concentration, and S - S* = 0.95S . Therefore, a' = 0.53 Ib/lb, and
a' (S - S*) = 0.05S . Therefore, equation 3-4 may be rewritten as
Equation 3-5 .
8 34 T 24So
R = 1000 LP-5-T^ + 0.15X + 4.6 ~~J 3-5
These equations can be used to calculate the total oxygen requirements
per unit volume of aeration tank per day. However, the distribution of
oxygen required to satisfy the oxygen uptake rates which vary throughout
the aeration tank and throughout the course of the day is not provided.
The maximum oxygen uptake rate was observed in laboratory and full-scale
experiments to occur at the end of the daily peak loading. Therefore, the
oxygen uptake rate reaches a peak normally in late afternoon when the
rate of flow and the strength of the incoming wastewater begin to decrease.
The ratio of the peak oxygen uptake rate to the daily average oxygen uptake
rate is much higher in those plants which are highly loaded than in the
plants which have a low organic loading . In the low loaded plants , the
endogenous oxygen uptake rate tends to equalize the peak oxygen uptake
rates which may develop. The ratio of oxygen uptake rates is also lower
than the ratio of the peak organic loading to the average organic loading
for a given day. The results of laboratory-scale studies indicate that some
of the suspended solids which accumulate in the mixed liquor undergo
degradation after the peak organic loading has passed through the plant
(Berryhill , 1970) . This oxygen requirement caused by the degradation of
the insoluble organic material may be attributable to the fact that these
organic solids undergo degradation at a much lower rate than the dissolved
organic material. Therefore, for design purposes, it can be assumed that
the distribution of oxygen uptake rates at a treatment plant is similar to
the distribution of incoming wastewater flow. The oxygen uptake rate also
varies with the length of longitudinal tanks. The biodegradation of soluble
organic material is high compared to that for the nonsoluble material.
Therefore, the oxygen uptake rate in the aeration tank is almost always
higher in the first part of the tank. The oxygen uptake rate will then decrease
as the concentration of soluble organic material decreases. The results
of experimental, studies point to this variation in oxygen uptake rates and
the peaks that can occur. These data presented in Figure 3-10 indicate the
relationship of oxygen uptake rate to aeration time for a batch contact
80
-------
120
LU
CE
LU
CL
ID
LU
O
g
100
80
60
40
20
0
4LWASTEWATER
144 mg/l BOD
25° C
2L RETURN SLUDG
STABILIZATION
6800 mg/l ss
I
CONTACT
2350 mg/l SS
0
30
60 90 120
AERATION TIME (min.)
FIG.3-10
OXYGEN UPTAKE RATES (BATCH)
81
-------
stabilization system. After the wastewater is introduced, the oxygen
uptake rate increases sharply until a peak is reached and decreases mark-
edly with time. This peak coincides with the rate of utilization of the
soluble organic material. In the full-scale plant where the contact and
stabilization zones were not physically separated, this peaking of oxygen
uptake rate was reduced since some degree of backmixing was observed.
This backmixing causes mixing of the influent with a larger volume of mixed
liquor and tends to equalize the oxygen uptake rate.
The distribution of incoming wastewater to various points in the aeration
tank in the step aeration process also tends to maintain a relatively constant
oxygen uptake throughout the entire length of the tank. The results of oxygen
uptake profiles determined in the contact stabilization plant and the conven-
tional activated sludge system at the Govalle Treatment Plant in Austin,
Texas are shown in Figure 3-11. These results indicate a similar pattern,
namely that the oxygen uptake rate is highest shortly after the influent
wastewater is mixed with return sludge and decreases rapidly to an equil-
ibrium level. After this point, the oxygen uptake rate decreases slightly
to a minimum rate at the effluent of the aeration basin.
AERATION
The oxygen required for biodegradation of organic material in wastewaters
is usually introduced into the aeration tank as diffused air or by mechanical
equipment. The aeration equipment is normally evaluated under standard
conditions of a water temperature of 20 C and an oxygen concentration of
zero. The aeration equipment is classified according to oxygenation cap-
acity (OC) expressed in units of pounds of oxygen per unit volume of aeration
tank (Ib O-/1000 gal-day). At times, the aeration efficiency (N ) is also
provided in terms of pounds of oxygen per horsepower hour (lb/I$-hr) . How-
ever, the oxygen transfer is different under process conditions than under
standard conditions. An oxygen concentration of about two mg/1 is generally
maintained in the aeration basin although it may be possible to operate at
lower dissolved oxygen concentrations.
The oxygen transfer rate constant (k a) is related to the oxygenation
capacity by Equation 3-6.
OC = (k a) (C ) 3-6
J_i S
in which:
C = concentration of oxygen in water at saturation
S
82
-------
150
100
o>
JE
Ul
h-
<
(T
LU
N£
<
Q_
UJ
o
X
o
50
0
CONTACT STABILIZATION
AVG. 4700 mg/l SS
STABILIZATION
CONTACT
50-
0
ACTIVATED SLUDGE
0 100 200 300
DISTANCE ALONG TANK (ft)
FiG.3-11
OXYGEN UPTAKE PROFILES
83
-------
The oxygen transfer rote constant increases as the temperature increases,
but the oxygen concentration at saturation decreases at higher temperatures.
Therefore, although these two parameters are affected by temperature, the
oxygenation capacity is relatively independent of temperature. The oxygenation
capacity decreases by about six percent when the temperature increases from
five degrees to 30 C.
The oxygen transfer is different for wastewater, for mixed liquor and for tap
water under comparable conditions. The actual oxygen transfer rate, there-
fore, can be expressed by Equation 3-7:
OTT=oc_lr al.02(2°-T) 3-7
S20
in which:
OT = oxygen transfer rate at temperature T in C
(lb/1000 gal-day)
OC = oxygenation c pacity at T = 20°C (lb/1000 gal-day)
C = concentration of oxygen in liquid (mg/1)
C = concentration of oxygen in liquid at saturation at
ST T°C '
C = concentration of oxygen in liquid at saturation at
S20 T=20°C
or = ratio of oxygen transfer under process conditions
to the oxygen transfer in tap water at the same
temperature
The effect of temperature on the oxygenation capacity is presented in Figure
3-12. The curve shows that temperature changes have a more noticeable
effect at the lower temperatures than at the higher temperatures.
The concentration of oxygen at saturation in a diffused air aeration system
must be corrected for the increased pressure that occurs as the air is re-
leased at the bottom of the tank. The saturation concentration of dissolved
oxygen can be adjusted for the partial pressure at the mid-depth of the tank
by using Equation 3-8.
°t.
3-8
eh
-------
o
o
o
CVJ
0.96-
0 10 20 30
WATER TEMPERATURE °C
FIG.3-12 INFLUENCE OF TEMPERATURE ON
OXYGENATION CAPACITY
85
-------
in which:
concentration of oxygen at saturation at mid-depth
sm , /,,
(mg/1)
C - concentration of oxygen at saturation under standard
conditions (mg/1)
P, = absolute pressure at depth of air release (Ib/sq in)
b
O = oxygen in air leaving the aeration tank (percent)
The concentration of oxygen at saturation at mid-depth (C ) therefore
depends on the depth of the diffuser as well as the percentage of oxygen
in the air leaving the aeration tanks. The effect of the depth at which
the diffusers are located on the ratio of the concentration of oxygen at
saturation at mid-depth to the concentration of dissolved oxygen at sat-
urations under standard conditions is presented in Figure 3-13. This curve
was calculated for an oxygen transfer efficiency of five percent and indi-
cates a linear relationship between the ratio C /C and depth.
sm v
Various values of alpha used in Equation 3-7 have been reported for mun-
icipal wastewater. In bubble aeration systems, alpha has been reported
to be equal to between 0.60 and 0.80 whereas in surface aeration systems,
this value ranged from 0 . 80 to 1 .10 .
The interaction between oxygen transfer, depth and rate of air flow are pre-
sented schematically in Figure 3-14. A linear relationship between the
oxygen transfer and the rate of air flow at a constant depth exists. However,
at the extremely high air flow rates, the rate of oxygen transfer tends to
level off at some equilibrium value. There is also a linear relationship
between oxygen transfer and depth at a constant air flow rate. The exact
value of the oxygen transfer at different depths and air flow rates is con-
trolled to a large extent by the type of diffuser or bubble introduction device
used. The aerator configuration and tank geometry influence the oxygen
transfer rate. In a spiral flow tank, the actual velocity of the water increases
with increasing air flow rates; therefore, the water air bubble contact time
and the shear of the bubbles decrease resulting in a lower oxygen transfer
per unit of volume of air. The ratio of the depth to width of the aeration
tank also influences the oxygen transfer rate. The aeration device controls
the diameter of the bubbles which are released. The oxygen transfer de-
creases as the diameter of the bubble increases. The oxygen transfer for
a specific aeration device can be expressed as the oxygen transfer per 1000
86
-------
1.3
1.2
1.0
0
FIG.3-13 OXYGEN
OF AIR
10 20
DIFFUSER DEPTH (ft)
SATURATION AND DEPTH
RELEASE
87
-------
DEPTH CONSTANT
AIRFLOW
r
o:
LU
u_
C/)
a:
h-
:z
LU
>
X
o
/ AIRFLOW CONSTANT
FIG.3-14
BUBBLE AERATION
(SCHEMATIC)
DEPTH
- OXYGEN TRANSFER
-------
cubic feet of air flow per foot of depth of the air release. Each aeration
device has optimum air flow rate beyond which the aeration efficiency de-
creases .
Mechanical aerators which are located at. or near the surface of the liquid
in the aeration tank are being used to a large extent. These aerators can
be categorized as:
(a) cone aerators with or without draft tubes
(b) pump-type aerators which operate at high speeds, and
(c) brush aerators.
Schematic diagrams of the various types of surface aerators are shown in
Figure 3-15.
Pump-type aerators are directly coupled to the motor and no gear box for
speed reduction is required. The oxygen transfer can be controlled by intro-
ducing air into the suction portion of the aerators. However, less water
is pumped.
The oxygen transfer of cone aerators increases within a given range with
the depth of immersion of the aerator and with increasing peripheral velocity.
Available data indicate that the increase in oxygen transfer is proportional
to the third power of the peripheral velocity. The oxygen transfer also
increases as the tank volume is reduced. Therefore, the oxygenation effici-
ency expressed as Ib/HP-hr increases with increasing power level expressed
as HP/1000 gallons. The general relationships between oxygen transfer and
depth of immersion and the peripheral speed are illustrated in Figure 3-16.
The relationship between oxygen transfer efficiency and power level is also
illustrated schematically. The characteristics of aeration device determine
the need for a draft tube. Draft tube aerators usually can be used only in
very shallow tanks,
Sludge deposits may develop in the aeration tank when mechanical aerators
are used. Knop and Kalbskopf (1968) indicate that the velocity at the bottom
of an aeration tank in which cone or turbine-type aerators were used is a
function of a power level. It has been observed that sludge deposits could
not be found in tanks where the bottom velocity was much lower than one
ft/sec. These reports indicate that turbulent conditions exist in the bottom
of the tanks and that the velocities cannot be effectively measured by the
common propeller-type meters.
Brush aerators which were developed by Kessener were initially mounted on
the walls of aeration tanks in which spiral flow was maintained. However,
89
-------
BUBBLE AERATION
WITHOUT 'DRAFT TUBE
CONE
WITH DRAFT TUBE
AERATION
MAMMOTH ROTOR AERATION
FIG. 3-15
TYPICAL AERATION TANKS
-------
cc
bJ
ct:
UJ
e>
g
SPEED CONSTANT
o:
LU
U_
ct:
LJ
e>
g
IMMERSION DEPTH
DEPTH CONSTANT
PERIPHERAL SPEED
CO
O
O
UJ
O
U-
u_
UJ
POWER LEVEL (hp/IO qal)
FIG.3-16
MECHANICAL AERATOR CHARACTERISTICS
91
-------
the development of the cage-rotor and the mammoth rotor has eliminated the
use of wall mounted brushes. Brushes are now mounted in a horizontal
position perpendicular to the direction of flow. If more than one pair of
rotors is operated in a single tank, baffles must be installed near the rotors
to enhance mixing and maintain higher oxygen transfer rates. The oxygen
transfer for brush aerators increases linearly with the depth of immersion
and with the peripheral speed raised to the 2.5 power. The brushes are
usually operated at a fixed speed and the oxygen transfer can be controlled
by the depth of immersion. The rate of oxygen transfer to an aeration sys-
tem of a specific surface area can be reported as pounds of oxygen per
hour per unit or pounds of oxygen per HP/hour. The operating range must
be given and the dimensions of the tank must be provided in order to effect-
ively design a system using brush aerators.
Von der Emde and Kayser (1969) reported the oxygenation capacity and oxygen
transfer efficiency of various aeration devices under process conditions.
These data indicate that for bubble aeration systems in which nozzles with
0.28-inch diameter orifices were used, the value for OC = 0.95 Ib O /1000
cu ft air/ft of depth of air release and a value of N = 1.30 Ib O2/HP-hr. The
data reported for ceramic tube-type diffused aeration systems was OC = 0...2
Ib O2/1000 cu ft air/ft of depth of air release and N = 1.85 Ib CL/HP-hr.
The oxygen transfer efficiency fc, surface aerators of the cone-type was re-
ported asN = 2.0to3.51b C>2/HP-hr.
The above discussion of aeration equipment indicates that each aeration
device requires a specific and different tank geometry, and the selection of
aeration equipment must include the cost of the tanks as well as the aeration
efficiency under process conditions. More detailed discussion of the design
of aeration equipment may be found in Gloyna and Eckenfelder (1968) and in
Eckenfelder and Ford (1970).
Recently the use of pure oxygen for aeration has been investigated by
Albertson (1970). Two full-scale activated sludge treatment plants were
operated in parallel in which one system was aerated with pure oxygen and
the other with diffused air. However, some of the operating conditions
such as sludge age and mixed liquor volatile suspended solids concentra-
tion were different in the air and oxygen systems operating in parallel. The
oxygen system operated at a higher sludge age and mixed liquor volatile
suspended solids concentration than the air system. The increased sludge
age results in a lower production of cells, a greater destruction of volatile
solids and an overall lower sludge yield. The quantity of soluble BOD re-
moved per unit tank volume increases as the concentration of mixed liquor
volatile suspended solids increases. These factors partially explain the
higher rate of BOD removal and a lower sludge yield reported for the oxygen
system when compared with the aerated activated sludge system. Ball and
92
-------
Humenick (1971) prepared a review and analysis of oxygen systems for
municipal wastewater treatment and concluded that although oxygen systems
may provide sorno real advantages over air systornr, , the benefits may not
outweigh tho costs.
The greatest savings for the pure oxygen system in total annual cost were
based on the low production of solids and the reduced sludge handling costs
A comparison of the pure oxygen system with surface aerators which in fact
have on the average about a 50 percent higher efficiency of oxygen transfer
than the diffused air system may have resulted in different results.
WASTE SLUDGE AND RETURN SLUDGE
The quantity of sludge produced during the activated sludge process is a
function of the sludge age as well as the quantity of BOD removed and can
be expressed mathematically as:
S - S* X
-L-.M^-jJLl _b + -2Q
Typical values of a* and b are 0 o3 and 0.075, respectively. The soluble
BOD in the effluent is equal to about five percent of the initial BOD or
S* = 0.05 S . The results of two studies using municipal wastewater in-
dfcate that ftie nonbiodegradable influent suspended solids represented 60
percent of the influent suspended solids. Introducing these values into
Equation 3-9 results in Equation 3-10 which can be further reduced to Equa-
tion 3-11.
0.95 S 0.6 X
-S- 3-JO
0.6 (S +X )
1 ° °
ir — » - -°-075 3-n
The relationship presented in Equation 3-11 between the sludge age and
the summation of the influent BOD and suspended solids is presented in
Figure 3-17. The data points are those reported by Wuhrmann (1964) for
pilot plant studies as well as for the Hyperion and Govalle full-scale exper-
iments . The line drawn through the points has a slope of 0 . 60 and provides
a basis for estimating the amount of sludge produced in a somewhat better
fashion than the growth models which depend on the composition of the
wastewater and for which different values of the coefficients a and b must
be determined. The scatter in the data, especially in the results reported
93
-------
for the operations at the Govalle Treatment Plant may be explained partially
by the fact that the quantity of sludge wasted at this plant is merely an
estimate and is not based on any specific flow measurements. There is
also a greater percentage of nonbiodegradablc solids in the effluent to the
Govalle Plants, since no primary clarification proceeds the aeration tank.
However, the data reported by Hyperion are described quite well by the
line shown on Figure 3-17. This closeness of fit may be explained by the
fact that at the Hyperion Plant, primary treatment preceeds the aeration
tank and much of the nonbiodegradable suspended solids are removed in
the primary clarifier. The scatter of the Wuhrmann data can be attributed
to difficulties encountered jn attempting to measure waste sludge flows in
pilot plant operations .
The excess sludge calculated as the reciprocal of sludge age is the total
mass of solids leaving the activated sludge system. Therefore, the flow
of waste sludge may be calculated.
1 VY
W =~- ( -~ - QX ) 3-13
XRS ' 6
The sludge wasting facility should be able to discharge the total amount
of sludge wasted per day during a 10 - 14 hour period if we assume that
the effluent suspended solids concentration is zero and that the concen-
tration of return sludge is the lowest which can be obtained with the total
return sludge pump capacity and a mixed liquor suspended solids concen-
tration of 2,000 mg/l. The return sludge concentration depends on the
flow rates of return sludge and of incoming wastewater as well as the
concentration of mixed liquor suspended solids maintained in the aeration
tank. This relationship can be expressed in Equation 3-14 and 3-15.
VRS • x(0 + V 3-M
QRS
V — V A, p
X~XRS (oTo) 3-15
in which:
X = mixed liquor suspended solids concentration (mg/l)
X _ = concentration of suspended solids in the return sludge
^ (mg/l)
-------
2.5'
LU
CD
O
=>
_J
CO
CO
CO
LU
O
X
Ld
_J
O
2.0
1.5
1.0
0.5
0
0
FIG. 3-17
EXCESS
A WUHRMANN
• GOVALLE
• HYPERION
1.0 2.0 3.0 4.0
(SS + BOD) -LOAD (Ib /Ib /day)
SLUDGE AND (SS + BOD) - LOAD
-------
Q = return sludge flow rate (1000 gal/hr)
RS
Q = wastewater flow rate (1000 gal/hr)
The concentration of suspended solids in the mixed liquor in the aeration
tank will decrease as the incoming wastewater flow increases, if the rate
of flow of return sludge and the concentration of suspended solids in the
return sludge remain constant. However, the variations in the concentra-
tion of mixed liquor suspended solids will decrease as the ratio of return
sludge flow to incoming wastewater flow increases. It is from this point
of view that the return sludge flow rate should be maintained at some
reasonable level or not be permitted to be very low.
The concentration of suspended solids in the return sludge is generally
estimated by using the concentration of activated sludge which results after
30 minutes of settling. The reciprocal of this concentration of solids in
the settled sludge is the sludge volume index. Therefore, the concentra-
tion of suspended solids in the return sludge may be expressed as Equa-
tion 3-16.
3-16
in which:
X = concentration of suspended solids in the return
sludge in mg/1
SVI = sludge volume index (milliliters per gram) and re-
presents the volume occupied by one gram of sludge
after settling for 30 minutes in a one liter cylinder.
The sludge volume index for a municipal activated sludge plant is about 150.
The limiting concentration of suspended solids in the return sludge will
therefore be about 6,700 mg/1. At a return sludge flow equal to the incom-
ing wastewater flow, a mixed liquor suspended solids concentration of
about 3500 mg/1 could be maintained. Therefore, in general the capacity
of the return sludge pumps should be about equal to the average wastewater
design flow. The aeration tank should be designed based on a concentration
of suspended solids in the mixed liquor of about 3,000 mg/1.
The energy required for pumping the return sludge is negligible compared to
the energy required for aeration. Therefore, the return sludge pumps could
be operated at a constant flow rate of the entire 24-hour period. By operating
continuously at a predetermined flow rate, the return sludge system would
96
-------
not require any elaborate flow control devices . Two or three return sludge
pumps having the same capacity should provide enough flexibility to handle
the variations of return sludge flow required. In most municipal treatment
plants , it is common practice to have more than one return sludge pump.
Screw pumps are preferred in the newer plants in Europe. In many cases, a
single screw pump is installed and is capable of returning the quantity of
sludge required at the large treatment plants.
It is difficult to maintain and control the concentration of suspended solids
in the mixed liquor by changing the rate of return sludge flow. A more
effective way of controlling the concentration of mixed liquor suspended
solids is by wasting sludge. Unfortunately, this method provides for only
a decrease in the concentration of the mixed liquor suspended solids.
Therefore, wasting sludge should be carefully controlled. An alternate
system is to waste sludge continuously; however, if semi-continuous
wastage is required, this waste can be controlled by pumps which are
operated on a time cycle or by using some other suitable flow control device.
EXTENDED AERATION PROCESS
The extended aeration process i:. a modification of the activated sludge
process in which the sludge age is maintained at a relatively high value
providing time for a portion of the sludge to be stabilized. An extended
aeration plant normally consists of an aeration tank, followed by a final
clarifier and the necessary pumps for sludge return. The excess sludge
is generally wasted in the form of suspended solids in the effluent.
The extended aeration process is widely used for relatively small communities
in the United States and Europe. The basic difference in the operation of
extended aeration plants in the United States and Europe is that in Europe
the excess sludge is removed from the system in order to maintain effluent
which contains a BOD of five to 15 mg/1. At the small plants serving com-
munities of 100 to 500 people, the sludge is dewatered on the sludge drying
beds. At the larger plants, the excess sludge is stored and subsequently
trucked to the agriculture areas. The basic design parameter is the degree
of stabilization of the waste sludge. At organic loadings of about 0.05 Ib
BOD/lb of MLSS-day, the sludge is reported to be well stabilized. Storage
of this material under anaerobic conditions does not cause any severe odor
problems. A number of plants are operated at organic loadings of 0 .1 Ib
BOD/lb MLSS-day in order to reduce the volume of aeration that is required.
The excess sludge from these plants is partially stabilized. The organic
loading rate expressed in terms of initial suspended solids concentration as
well as initial BOD concentration can be calculated using Equation 3-11. If
one assumes that the wasted sludge is well stabilized at a sludge age of
30 days , this equation can be modified.
97
-------
X I iS
"
therefore
X + S
X.L
= 0.18 3-18
The BOD loading rate would therefore be 0.065 Ib BOD/lb MLSS-day based
on the assumption that the ratio of influent suspended solids to BOD is
about 1.75. The volume of the aeration tank for an extended aeration
plant can be based on a load of influent BOD and suspended solids of 0.15.
This value would give a somewhat larger aeration volume compared to the
value calculated from the theoretical equations .
FINAL CLARIFIERS
The final clarifiers represent an important unit process in the activated
sludge plant. The suspended solids are separated from the effluent liquid
in the final clarifier and the degree to which the sludge concentrate in the
final clarifier will control to some extent the rate at which return sludge
must be pumped to the aeration tank.
Clarifiers can be divided into two characteristic groups depending on the
hydraulics, namely upflow clarifiers and horizontal clarifiers. The clari-
fiers may be circular or rectangular and exhibit either flow pattern. Deep
clarifiers with a relatively small surface area are generally the upflow
types whereas the shallow basins with large surface areas are more likely
to be horizontal flow types. The design of final clarifiers is generally
based on the overflow rate and the detention time. Theoretical conditions,
however, indicate that the overflow rate is far more important than the
detention time.
The method by which the settled sludge is withdrawn from the clarifier also
affects the performance of the clarifier. The preferred sludge collection
mechanism is a vacuum or suction type draw off. The plow type collectors
with the chain and flight mechanism in rectangular basins or the bridge
with attached plows in circular basins are not very effective in concentrating
the waste activated sludge. The sludge tends to flow over the plow as the
plow moves through the sludge blanket and very little movement of the sludge
to the concentrating compartment takes place. The vacuum or suction type
draw off actually draw the solids into the collection pipe under the force of
the hydraulic head in the tank. Schematic diagrams of center feed and
peripheral feed circular clarifiers and rectangular clarifiers are illustrated
-------
in Figure 3-18. The center feed clarifier and the rectangular clarifier
illustrate the plow-type collectors whereas the peripheral feed clarifier
indicates the suction draw off. Suction draw off equipment can also be
installed in the center feed circular clarifier or a rectangular clarifier.
The results of recent investigations by Pflanz (1968) indicate that the
effluent suspended solids concentration from final clarifiers operated at
constant overflow rates, increased as the concentration of mixed liquor
suspended solids increased. The results also indicate that at a constant
overflow rate and a constant concentration of mixed liquor suspended
solids, the effluent suspended solids concentration increased as the
sludge volume index increased. The effluent suspended solids also in-
creased as the temperature decreased.
The interdependency of the mixed liquor suspended solids and the sludge
volume index indicate that a relationship also exists between effluent
suspended solids concentration and the sludge volume surface load. The
sludge volume surface load is the product of the overflow rate (gallons
per square foot per day) and the fraction of the initial volume occupied
by the sludge solids after settling for 30 minutes. Therefore, the sludge
volume surface load in a final cMrifier with an overflow rate of 1000 gal/
sq ft-day, and a sludge which occupies 1/10 the volume of the initial
sludge after settling 30 minutes would be 100 gal/sq ft-day.
The relationship between the sludge volume surface load and the effluent
suspended solids concentration are presented in Figure 3-19. This plot
is based on the data presented by Pflanz (1969) . The data show consid-
erable scatter primarily at the higher values of sludge volume surface
loading, and effluent suspended solids concentration. However, at the
lower range of values, there is a fairly good correlation. The scatter
seems to be the result of errors in determining the sludge volume index,
which is generally determined under quiescent conditions. However, the
conditions in the settling tank are far from quiescent and a good degree
of the turbulence in the full-scale tank may actually promote flocculation
of the sludge particles. Therefore, in the laboratory determination of the
sludge volume index a slow stirring device which is placed in the cylinder
would more closely simulate the conditions in the field and may give much
more reliable results.
Much of the sludge settles near the inlet zone of the final clarifier under
normal overflow flow rates of about 500 gal/sq ft-day and a mixed liquor
suspended solids concentration of about 3000 mg/1. A thin layer of diluted
sludge may result in a density current. This diluted layer will move along
the bottom of the tank, rise to the effluent weirs and result in high effluent
suspended solids concentrations. Prevention of these density currents
99
-------
SLUDGE
CIRCULAR CLARIFIER WITH CENTER FEED
O Q Q_Q_ O Q 0^3 Q
SLUDGE
CIRCULAR CLARIFIER WITH PERIPHERAL FEED
t i
SLUDGE
RECTANGULAR CLARIFIER
FIG.3-18
DIFFERENT TYPES OF FINAL CLARIFIERS
100
-------
o
JZ
I
O
O
LU
O
ti-
er
ID
CO
LU
^
ID
_J
O
LU
CD
Q
CO
14
12
10
8
RECTANGULAR SETTLING TANK
DATA AFTER PFLANZ (1969)
svi = 155 ml/g
= 20i
= 306
0 10 20 30 40 50 60
SUSPENDED SOLIDS EFFLUENT (mg/l)
F1G.3H9
SLUDGE VOLUME SURFACE LOAD
AND EFFLUENT SUSPENDED SOLIDS
101
-------
would reduce the; concentration ol suspended solids in the effluent. Density
currents were minimized at rectangular finai ciorifiers which were eight to
ten feet deep at three plants in Europe. The method used for minimizing the
density currents are illustrated in Figure 3-20 .
The settling tanks at the treatment plant at Nordhorn, Germany, had a single
effluent weir across the entire width of the tank. However, high effluent
suspended solids concentrations were observed and attributable primarily
to density currents. A vertical plastic sheet was placed at a point about
1/4 the total tank length and essentially divided the tank into two separate
parts. Most of the solids settled in the first part. The increased velocity
of flow over the top of the sheet caused the remainder of the suspended
solids to become well mixed. Therefore, no density currents were observed
in the second part of the tank. This procedure reduced the concentration
of suspended solids observed in the effluent.
The results of experiments conducted at the treatment plant at Baden, Austria,
indicated that if the weirs at the end of the tank were closed and the efflu-
ent was only taken off by the weirs along the side of the tank as illustrated
in Figure 3-20, the concentration of suspended solids in the effluent was
much less than when all weirs were operated. Similar results were reported
at the treatment plant in Ergolz, Switzerland. In this case, six weirs were
originally used as those illustrated in Figure 3-20 . However, when the weirs
1,3, and 5 were closed, the concentration of suspended solids in the efflu-
ent was markedly reduced.
The withdrawal of sludge from secondary clarifiers for return to the aeration
basin takes place at a point very near the inlet to the tank with the excep-
tion of the peripheral feed tanks. Therefore, the required area of the final
clarifier is calculated using the incoming wastewater design flow and the
return sludge flow rate is not included. The overflow rate for a clarifier may
be calculated based on the sludge volume surface load and desired effluent
suspended solids concentration. For example, for an effluent suspended
solids concentration of 25 mg/1 the sludge volume surface load is five gal/
sq ft-hr (120 gal/sq ft-day) based on the curve in Figure 3-19. At a mixed
liquor suspended solids concentration of 3000 mg/1 and a sludge volume
index of 100 ml/g , the volume occupied by the settled sludge after 30 min-
utes of settling is 30 percent of the initial volume. This calculation is
shown in Equation 3-19.
3000 mg/1 - x — x ^_ = 0>3 3
gm 1000 mg 1000 ml
Therefore, the required overflow rate is equal to 400 gal/sq ft-day as
shown by the calculation in Equation 3-20.
120 gal/sqft-Iday = 40Q gaj/sq f._day
102
-------
NORDHORN
PLASTIC SHEET
-J ------- . .1 ------- J ------- *. ------- *.-
V
BADEN
t t
1 1
! 7 2//
1 3
1 \ \ V
1 1 ^ A ^
« 1
I
V ^
\
A ^
'
<
>
4
-
••
r I
>
>
ERGOLZ 5 3
I' ^'
FIG.3-20
IMPROVEMENTS OF FINAL CLARIFIERS
103
-------
The calculated overflow rate is somewhat lower than that frequently used
for design valves of 600 to 800 gal/sq ft-day. This lower value based on
the data presented by Pflanz (1968) may be attributed to the fact that in the
particular secondary clarifier used by Pflanz only one effluent weir across
the entire width of the rectangular basin was used at relatively low temper-
atures, namely between 13 -15 C. Therefore, at newly designed plants ,
the overflow rate of the secondary clarifier of 600 gal/sq ft-day can be
used for smaller plants up to a total of one MGD, and up to 700 gal/sq ft-
day for the larger plants. The design calculations should not be based
on the daily design flow but rather on the peak flow. Therefore, if the design
is based on the peak hourly flow, the overflow rates for smaller plants should
be 25 gal/sq ft-hour and 30 gal/sq ft-hr for the larger plants.
NITRIFICATION AND DENITRIFICATION
Nitrification and denitrification have been reported to occur in the activated
sludge process. These processes remove nitrogen from the effluent from
the activated sludge process.
Nitrification involves a conversion of ammonia to nitrite and nitrate. The
reactions are presented in Equations 3-21 and 3-22.
New
2NH4 + 302 Nitrosomonas > - +
Cells
3-21
2N02 f 02 Nltr0bdCter ) 2NO~ + New Nitrobacter cells
3-22
The conversion of ammonia to nitrite is accomplished by the microorganism
Nitrosomonas, and the conversion of nitrite nitrogen to nitrate nitrogen is
accomplished by Nitrobacter organisms. The rate of growth of Nitrosomonas
is described by Equation 3-23 for nitrification.
(NH41
W ~ Mmax k + (NH.) 3"23
m 4
in which:
H = growth rate of Nitrosomonas (day" )
H = maximum growth rate (day )
ITlckX
(NH ) = concentration of ammonia (mg/1)
-------
k = concentration of ammonia when ^= 0.5u (mg/1)
m Tnax
Downing (1964) reported the growth constants for Nitrosomonas at 20 C
to be y. = 0.33/day and k =1.0 mg/1 of ammonia nitrogen. The maxi-
mum growth rate for the Nitrobacter at 20 C reported by Downing was
p = 0.14/day. These data indicate that the maximum growth of Nitro-
somonas and Nitrobacter probably occur at an ammonia concentration of
about 3.0 mg/1. At concentrations of ammonia above this minimum con-
centration, the rate of nitrification is independent of the ammonia con-
centration and the nitrification reactions are zero order. The growth rate
constants for the nitrifying bacteria are relatively low compared to those
reported for heterotrophs . For example , Schulze (1964) reported that the
maximum growth rate for Escherichia coli and a glucose substrate was 18
per day. Therefore, the detention time in the activated sludge system
must be greater than the generation time of the nitrifying bacteria , other-
wise these bacteria would be washed out of the system. Since the nitri-
fying bacteria are associated with the mixed liquor the sludge age should
be sufficiently long to provide active growth of the nitrifying bacteria if
nitrification is to take place. On the other hand, the amount of nitrifica-
tion which occurs can be calculated from the sludge age.
The maximum growth rate of the nitrifying bacteria is markedly affected by
changes in temperature. The result of investigations at the Water Pollution
Laboratory (Anon. , 1967) indicate that the rate constant increases about
seven percent per degree centigrade. The effect of temperature on the growth
of nitrifying bacteria is presented in Figure 3-21 Nitrification will take
place if the reciprocal sludge age (1/G) is lower than the maximum growth
rate of the nitrifying bacteria.
Aerobic conditions must be maintained for nitrification. Downing (1964)
and Wuhrmann (1964) reported that a minimum concentration of dissolved
oxygen of one mg/1 is required in order to achieve "litrification. The maxi-
mum rate of nitrification requires dissolved oxygen concentrations of more
than 200 mg/1.
Nitrification will occur to some extent in activated sludge treatment plants
operated at organic loading rates of about 0.25 Ib BOD/lb MLSS-day or
lower. If nitrification does occur, a higher oxygen uptake rate will be
exerted than that calculated for the removal of carbonaceous matter only.
Therefore, nitrification should be included in calculating oxygen requirements.
Partial nitrification is possible in the contact stabilization process, since
the sludge age is relatively high in this process. However, since the contact
time is relatively short, only a partial nitrification is possible. In the
105
-------
o
^ 30
ui
QC
< 20
cr
LJ
o_
UJ
10
1
1
i
1
1
0.4
MAX
0.5
0.6
0.1 0.2 0.3
MAXIMUM GROWTH RATE,
FIG.3-21
TEMPERATURE EFFECTS ON GROWTH RATE OF
NITRIFYING BACTERIA (Downing 1964)
106
-------
stabilization tanks where the concentration of nitrifying bacteria is high,
the concentration ammonia is the limiting factor. The ammonia removal in
the contact stabilization process will be increased at higher return sludge
flow rates and lower contact times.
Denitrification can take place once the ammonia has been converted to the
nitrite or nitrate form. If the dissolved oxygen concentration in the mixed
liquor is at a concentration less than one mg/1 heterotrophic bacteria can
utilize the oxygen associated with the nitrites and nitrates for metabolism,
and the nitrogen is released as nitrogen gas. A denitrified effluent can be
obtained by placing a denitrification tank between the aeration basin and
the final clarifier. The denitrification tank should be equipped with a mix-
ing device to maintain the solids in suspension but not to introduce any
oxygen. The time required for total denitrification can be estimated from
the initial dissolved oxygen concentration in the mixed liquor and the
quantity of oxygen present in the form of nitrites or nitrates as well as the
oxygen uptake rate of the mixed liquor. Wuhrmann (1964) indicated that
the time required for denitrification can be calculated from the mass balance
which is presented in Equation 3-24:
D Q R^ 3-25
in which:
t = detention time of the mixed liquor in the denitrification
D basin (hrs)
C = dissolved oxygen concentration in the mixed liquor
entering the denitrification basin (mg/1'
C = oxygen in the form of nitrite and nitrates (mg/1)
R = oxygen uptake rate of the mixed liquor in denitrification
basin (mg/l-hr)
The oxygen uptake rate in the denitrification basin is very similar to the
endogenous rate. If nitrification takes place in the aeration tank an no de-
nitrification is provided, partial denitrification may take place in the final
clarifier and the released nitrogen gas may cause a portion of the settled
sludge to float. Therefore baffles should be located near the effluent weirs
in the final clarifiers to reduce the carryover of suspended solids into the
effluent.
107
-------
PROCESS CONTROL
The practical control of the activated sludge process involves:
(a) control of the concentration of dissolved oxygen in the aeration
tank, or
(b) control of the mixed liquor suspended solids concentration.
The dissolved oxygen concentration in the aeration basin should usually be
maintained between 1.0 and 2.0 mg/1. The amount of oxygen transferred in
the aeration basin therefore should be capable of satisfying variations in the
oxygen uptake rate. The automatic control of the oxygen transfer to the sys-
tem may be realized by monitoring the concentration of dissolved oxygen at
the point in the aeration tank where the oxygen uptake rate is at a maximum.
If the concentration of dissolved oxygen in the aeration tank falls below a
minimum desirable level the aeration equipment can be called upon to provide
additional oxygen. On the other hand, if the dissolved oxygen concentration
in the aeration basin exceeds the desired level, the amount of oxygen intro-
duced can be reduced. The savings in power costs over a number of years
of operation would probably be sufficient to cover the capital cost of the
automatic control system. A single oxygen probe would be required for a
completely mixed aeration tank or the tank in a step aeration system, since
the oxygen uptake rate should be the same throughout the entire volume of
the tank. In longitudinal spiral flow tanks generally used for the conven-
tional activated sludge process, the oxygen probe should be installed near
thewastewater inlet since it is at this point that the oxygen uptake rate is
the highest.
The concentration of mixed liquor suspended solids in the aeration tank can
be decreased by increasing the quantity of sludge wasted. An increase in the
mixed liquor suspended solids concentration is practically impossible; there-
fore, the quantity of sludge wasted should be carefully controlled. Generally,
continuous wasting at a predetermined rate is preferred to the practice of
wasting sludge periodically.
Guarino and Carpenter (1970) indicated that the control of the loading rate is
possible, by storing return sludge to provide the necessary sludge to main-
tain a constant loading rate. This procedure is essential since the incoming
BOD is very difficult to control. Therefore, by controlling the concentration
of mixed liquor suspended solids, the loading rate can be maintained relative-
ly constant. Sludge is pumped to storage tanks during the night time hours
when the loading rates are relatively low and introduced into the system during
the day time when the organic loading roaches its peak. This system would
provide savings in power roquirod for accommodating tho oxygen uptake rate.
108
-------
All municipal wastewater treatment plants operate under transient loading
conditions. The ratio of the daily peak to the minimum flow is in a range
of 3-1 to 10-1 or more. In most treatment plants , the effluent BOD is
practically not affected by the fluctuating transient conditions provided
the organic loading was equal to or less than 0.5 Ib BOD/lb MLSS-day.
Observations at treatment plants indicate that the biological processes
require some time for recovery. Generally the recovery occurs during the
night time hours when the organic load to the plant is low or on the weekends
when loadings are also lower than during the work week. The recovery time
could be required for stabilization of the biological solids and/or for the
biodegradation of other particulate material which accumulate in the mixed
liquor.
DESIGN FACTORS
Design and layout of an activated sludge plant for the treatment of municipal
wastewater is based on the following design data:
(a) wastewater flow (MGD) daily peak wastewater flow (1000 gal/hr)
(b) average influent BOD (mg/1)
(c) average influent suspended solids (mg/1)
The basic data should also include any industrial waste and information re-
lating to type, source of wastewater, and quantity of discharge. This inform-
ation is required to minimize the possibility of shock loadings to the activated
sludge system. If more than 1/2 the total BOD load to the plant is contributed
by a single industry, it is conceivable that laboratory or pilot-scale treata-
bility studies should be performed to effectively select the proper design para-
meters .
A number of other factors must be considered in the design of a biological
treatment plant since other variables might affect the performance and oper-
ation of the overall plant. The need for primary sedimentation prior to bio-
logical treatment is sometimes questioned. The characteristics of the
wastewater and the manner in which the solids are to be ultimately disposed
of should be considered carefully prior to deciding on whether a primary clar-
ifier is required. The removal of suspended solids in the primary sedimenta-
tion tank would minimize the load on the biological treatment process since
some of the biodegradable solids would be removed and the oxygen uptake
requirements would be considerably lower. In this case, the primary solids
would more than likely be treated in an anaerobic digestion system prior to
disposal on the sludge drying beds. However, it is also possible for the
primary solids to be mixed with the waste activated sludge and treated by
109
-------
aerobic digestion provided the quantity of waste activated sludge is at
least equal to the quantity of primary sludge requiring disposal.
Generally, the final decision is based on a cost estimate. However, other
factors should also be included before a final decision. Primary sedimenta-
tion is preferable in those cases where the incoming suspended solids are
mainly inorganic such as might be found in a treatment plant handling com-
bined wastewater. Primary clarification also permits the removal of floating
materials. Therefore, if primary clarification is not installed prior to the
aeration tank, it is essential that the secondary clarifiers be equipped with
skimming equipment.
The layout of the biological treatment plant is also important. Any layout of
the processes should provide for flexible operation under a variety of condi-
tions . Most municipal waste treatment plants have a number of aeration
basins and final clarifiers operating in parallel. This requirement is gener-
ally established by the regulatory agency in the particular state or region.
However, pipes and channels connecting the various units should be so
designed that taking any one unit process out of operation does not affect
the use of the remaining unit processes. For example, it should be possible
to take an aeration tank out of £arvice and operate all clarifiers by distrib-
uting the incoming wastewater, and return sludge to the remaining aeration
units. On the other hand, if a final clarifier is out of operation, it should
be possible to continue operating the plant with all unit aeration tanks being
utilized. With this type of design and plant layout a single blower station
and a single sludge return pumping station could be included in the design.
Some additional piping might be required to maintain this flexibility, but the
cost of the piping is recovered in terms of the ability of having continuous
operation of the plant.
An estimate of the ultimate organic load and wastewater flow must be determined
at the time of the initial design, although it is extremely difficult to estimate
what might happen 25 to 50 years after the plant is in operation. The future
program of operation of a treatment plant is helpful in the design of the various
components so that the plant can be expanded as organic and hydraulic loads
increase. The initial construction phase of the treatment plant should be de-
signed for ease of expansion.
DESIGN FORMULATIONS AND EXAMPLES
The performance of the activated sludge process in terms of BOD concentration
in the effluent can be used for design of the process. The relationship between
the effluent BOD and the BOD loading was presented in Figure 3-7. The design
BOD effluent concentration is determined by taking the maximum effluent BOD
at a given organic loading and increasing that value by about 50 percent. This
110
-------
increase is necessary because the dat.i rapi<;scnt dveroqe values and the
standard deviation for these data was about 0.5 of the average. The effluent
suspended solids concentration was derived from the relationship developed
in Figure 3-9 which relates the effluent suspended solids concentration and
the sludge age.
The basic design data are listed in Table 3-1. Columns one and two
represent the particular parameter in question and the dimensions of parameter,
respectively. Column 3 represents the calculations for the extended aeration
process. The effluent qualities for the activated sludge process at various
loadings are summarized in Columns 4, 5, 6, 7, and 8, respectively. Two
sets of calculations for the oxygen uptake rate and the oxygen transfer are
presented for the activated sludge process for the loading rate of 0.25 Ib
BOD/lb MLSS-day. One set of calculations represents the oxygen require-
ments for biodegradation of only carbonaceous material while the other set
of calculations includes the oxygen requirements for nitrification of ten mg/1
in addition to the carbonaceous oxygen requirements.
The incoming wastewater contains 150 mg/1 of BOD and 150 mg/1 of suspended
solids. These design calculations can be applied to wastewaters which have
an incoming BOD which ranges from 100 to 200 mg/1 provided the ratio of in-
coming BOD to incoming SS is l:i. The information presented in the various
lines in Table 3-1 will be discussed below where discussion is necessary.
Line 6 Design MLSS:
3000 mg/1 has been assumed for all processes and process modi-
fications; however, in practice a MLSS concentration up to 4000
mg/1 can be applied.
Line 7 BOD Volume Load:
, / /, x BOD Load 8.34
(lb/1000 gal/day) =
Ib BOD mg/1 8.34 Ib.
Ib MLSS-day 6 gal
Line 8 Excess Sludge
0 . 6 (S + X )
X
in this example, SQ = XQ or SQ/XO = 1
111
-------
ro
TABLE 3-1
ACTIVATED SLUDGE PROCESS DESIGN CHART
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17 .
18.
19.
Design BOD Effluent
Average BOD Effluent
Design Suspended Solids Effluent
Average Suspended Solids Effluent
Design BOD Load
Design MLSS
BOD Volume Load
Excess Sludge
Oxygen Uptake Rate
per gallon tank
Oxygen Concentration, ml .
Oxygen Transfer
Airflow/gallon tank
Airf low/lb BOD Appl .
Peak (24/18)
Design Airflow
Detention Time
Airflow/Gallon Sewage
Waste Sludge/gallons
Dimensions
mg/1
mg/1
mg/1
mg/1
Ib/lb/day
mg/1
lb/103 gal/day
Ib/lb MLSS/day
Ib/lb MLSS/day
lb/103 gal/day
mg/1
lb/103 gal/day
cu ft/gal/day
cu ft/lb BOD
cu ft/lb BOD
cu ft/lb BOD
hours
cu ft/gal
lb/103 gal
Extended
Aeration
(15)**
(10)**
(-65)*
(-15)**
0.065
3000
1.60
0.03
0.185
4.40
1
5.20
3.60
2250
2250
2800
(19)
(2.90)
(0.42)**
Activated Sludge
20
10
25
15
0
.25
3000
0.370
9.30
2
12.20
8.60
1360
1820
2400
(1.
1.
6
0
(4
3
70)
37
(0
0
.25
.22
0
6
2
9
6
1000
1350
2000
.80)
.85
(1
1
.97)
.73
25
12
30
16
0
3000
12
0
.275 0
.90 10
2
.10 13
.40 9
740
1000
1500
(2
1
.27) (0
.02 0
(1
0
.50
.5
.52
.40
.0
.20
.30
.40)
.90
.93)
.74
.15)
.92
35
15
35
18
1.
3000
25
1.
0.
16.
2
21.
15.
600
800
1200
(1.
0.
(0.
0.
(1.
0.
0
12
65
20
50
00
20)
95
75)
60
20)
96
50
22
45
22
2.:
3000
50
2.32
1.15
28.73
2
38.0
26.5
550
730
1100
(c.6:x
0 . ? 0
(6!c9)
0.55
(1.25'-
i . o :•
* If no sludge is wasted (all sludge leaves through effluent)
**If sludge is wasted
-------
therefore
-V - 0.6 (2) BOD Load - 0.075
C<
Line 9 Oxygen Uptake Rate:
(Ib O /lb MLSS - day)
LJ
S - S
|-= a' ( ° 6) + b1 (eq 3-3)
A At
— = 0.5 (BOD Load) + 0.15
X
but for nitrification at BOD Loads of 0.25 Ib/BOD/lb MLSS - day,
Oxygen uptake for nitrification is:
R 4 fi (10) (24) g
F = 4'53000 (3.85) -°'°95
Total uptake rate including nitrification
0.370 lb O2
(0. 5 (0. 25) + 0.15) +0.095 = lb MLSS_day
Line 10 Oxygen Uptake Rate per 1000 gal of Tank Volume:
R_ _X(8.34) ™~2 ^ *'* }
X 1000 lb MLSS -day 10° gal.
Line 11 Dissolved Oxygen Concentration in Mixed Liquor:
mg/1
Line 12 Oxygen Transfer:
lb O9/1000 gal - day
°SM
Oxygen Uptake (— _ r )
113
-------
C , at standard conditions H.O mq/1
o
o. 0.')
Compensation for 13 foot depth of air release from Figure 3-13
CQ /C_ =1.16
M C
. ' . Cq = C_ a SM
&IK ^ r-
M Cs
C0 = 8.0 (0.9) 1.16= 8.3 mgA
SM
Line 13 Airflow Rate:
(Cu ft/gal - day)
Specific oxygen transfer = 0.11 Ib O /1000 cu ft/ft at 13 foot
depth of air release. 0 .11 (13) = 1.41 Ib O /1000 cu ft of air.
£*
Oxygen Transfer/1000 gal - day
1.43 Ib O /1000 cu ft air
Line 14 Average Airflow per Ib BOD Applied:
Cg
Oxygen uptake rate M . 1000
BOD Load (1.43) C_ - C
SM
Line 15 Airflow at Peak BOD:
(cu ft/lb BOD)
Peak hour = 1/18 of total daily
24
Peak airflow = — Average Airflow
lo
Line 16 Design Airflow:
(cu ft/lb BOD)
The design airflow is required for the layout of the blowers and
includes a factor of safety of 1.25 for extended aeration plants and
1.50 for the activated sludge plants. Design Airflow = peak airflow
(1.25) or Peak airflow (1.50).
LlV
-------
Line 17 Detention Time:
(hours)
S (24) (8.34)
o
(BOD Volume load) 1000
The numbers in ( ) in Table 3-1 are based on S = 150 mg/1.
The other values of detention time are based on S - 120 mg/1
o
Line 18 Airflow per Gallon of Wastewater:
(cu ft/gal)
(cu ft/gal of tank volume) (Detention time)
24 hrs/day
0 = V/t . ' . t/V - 1/0
Line 19 Waste Sludge:
(Ib sludge/1000 gallons wastewater)
0.6 (So +Xo) - (0.075f|) -Xe
X
o
The data in Table 3-1 will be used for this design example for an activated
sludge plant with primary clarification.
Sewage flow 20 MGD
Peak Hour 26.5 MGD
Primary Effluent 120 mg/1 BOD, 120 mg/1 SS
Total BOD after primary clarification is:
20 MGD (120 mg/1) 8.34 Ib/gal - 20,000 BOD/day
Design for effluent BOD 20 mg/1
Design load 0 .25 Ib BOD/lb MLSS - day
Design MLSS 3000 mg/1
BOD volume load (from Table 3-1) 6.25 Ib BOD/1000 gal/day
115
-------
Aeration tank volume:
(20,000 Ib BOD/day) 1000 gal „ n ,fi
6.251bBOD/day = 3.2x10 gallons
Detention Time:
3.2 x 10 gal.
20 x 106 gal
day
24 hours
day
= 3.84 hours
Excess sludge = 0.22 Ib/lb MLSS - day
Waste Sludge:
Total MLSS = 3 . 2 x 10 gal
(80'000 lb
- 80 ,000 Ib
= 17 ,600 Ib/day
Return sludge concentration at a return sludge flow rate equal to the flow
rate of the incoming wastewater and MLSS concentration, X = 3000 mg/1.
= X (° f V
Q
RS
RS
RS
= X
RS
Q
= 3000 (2/1)
= 6000 mg/1
Flow Rate of Waste Sludge
17,600 Ib/dayx 106
(6,000 mg/1) 8.34 Ib/gal
Oxygen Transfer (Average):
Total oxygen transfer per day
9.101b02
1000 gal-day 3-2x10 gal = 29,000 lb O,/day
= 352 ,000 gallons/day
9.10 Ib02/1000 gal/day
116
-------
At peak hour the oxygen transfer Kilo is:
29'000 IkOa <*§*- (|1) = i
day 24 hr 18
Including the Safety factor for design, the peak oxygen transfer rate
is
1,600 —(1.5) = 2,4001b/hour
hr
Aeration Equipment
Cone type aerators could be used and each aerator has an oxygen-
ation capacity of 240 Ib O2/hour; therefore, the number of aerators
required is
2400
240 = 10 aerators
The variation in oxygen transfer can be accomplished by changing
the depth of immersion by the ratio of 1:3; therefore, the minimum
oxygen transfer rate is about 800 Ib O2/hour.
Diffused Aeration System
Specific oxygen transfer rates = 0.11 Ib O2/1000 cu ft per foot
of depth of air release
Therefore, at a diffuser depth of 13 feet, the oxygen transfer rate is
0.11 Ib O,
2 13 ft = 1.43 Ib O0/1000 cu ft
10 ±L ~ 1 . IJ 1.JJ >^0
1000 cu ft-ft 2
The design airflow rate is
2400 Ib/hr 1000 cu ft = 28/000cfm
(1.43 Ib) 60 min/hr
The design airflow per Ib BOD is
28,OOP cu ft/nun ( 144° min )= 2000 cu ft/lb BOD
20,000 IbBOD/day v day '
117
-------
The airflow at the daily peak is
2000 cu ft/lb BOD ^ 1/35ocuft/lbBOD
118
-------
AERATED LAGOON PROCESS
PROCESS DESCRIPTION
Aerated lagoons are ponds in which diffused air or mechanical aeration
systems are installed. Detention times of less than two days to more than
30 days have been employed.
An aerated lagoon is generally visualized as a completely mixed tank.
The biodegradation process is controlled only by maintaining aerobic con-
ditions in at least part of the basin. The microbial population is controlled
by the concentration of substrate entering the lagoon and the detention time.
However, there is almost no control of the influent BOD after the aerated
lagoon system is put into operation. The power introduced into the aerated
lagoon for aeration and mixing establishes the extent to which sedimentation
of the solids takes place in the ponds. At high power levels almost all of
the solids are maintained in suspension and little settling occurs . These
systems are completely mixed and can be considered to be aerobic lagoons.
At lower power levels there is some settling of solids to the bottom of the
pond where the accumulated biodegradable solids undergo anaerobic decom-
position. These lagoons are facultative with an aerobic zone overlying the
anaerobic area.
In general, the suspended solids concentration in the effluent of aerated
lagoons is considerably higher than for the activated sludge process. The
effluent suspended solids concentration can be reduced somewhat by baffling
a segment of the pond to permit a quiescent zone where the suspended solids
may settle. An unmix pond may be installed following the aerated lagoon
to permit the removal of the settleable suspended solids. Algae may de-
velop and the effluent suspended solids concentration may not change but
the composition will be different.
PERFORMANCE OF THE AERATED LAGOON PROCESS
The mathematical model which is used to describe the kinetics of the
aerated lagoon process does not include the active mixed liquid suspended
solids since the concentration of suspended solids can be controlled only
by controlling the power level. The equation may be written as:
o 1 b
akt + ak 4-1
119
-------
Equation 4-1 indicates that the substrate concentration in the effluent BOD
decreases with increasing detention time and effluent BOD is independent
of the influent BOD. The concentration of active mixed liquor suspended
solids in an aerated lagoon operated at a given detention time and power
level will increase as the influent BOD increases. Therefore, the effluent
BOD concentration remains relatively constant. The substrate concentra-
tion in the lagoon, S, includes both soluble and insoluble organic material.
However, only the BOD of the soluble fraction is measured. In general,
it can be assumed therefore that the insoluble fraction, if it is measured
as BOD is relatively low since the concentration of microorganisms in the
lagoon is much higher than that which is found in the BOD bottle.
Equation 4-1 can be evaluated graphically ±>y plotting the substrate concen-
tration, S, versus the reciprocal of time (—). The straight line through
the data will have a slope of 1/ak and the intercept will be equal to b/ak.
However, there is some question whether the total or soluble BOD should
be used. One relationship will be developed for only the soluble BOD in
which it is assumed that the fraction of the total BOD removed is the same
as the fraction of the soluble BOD removed or
S*/S* = S /S 4-2
e o e o
The results of laboratory-scale aerated lagoon experiments at very short
detention times are presented in Figure 4-1. There is some scatter among the
data points; however, the slope of the line through the data is 7.0 (mg/1
of BOD) (days). Therefore, the product ka = 1/7.0 = 0.143. For municipal
wastewater a = 0.65; therefore, k = 0.22 (day ) (mg/1) . The intercept
in Figure 4-1 is difficult to determine. However, by assuming an endogenous
respiration rate, b- 0.15 per day, the intercept can be calculated as 0.15/7 .0
0.02 mg/1 of BOD. Comparison of this value to the measured BOD and con-
sidering the scatter of the data, the line describing the data could have been
drawn through the origin as well.
The rate constant for aerated lagoon systems can also be calculated by
Equation 4-3.
S - S
= kS 4-3
X t
a
The terms can be rearranged and the rate of substrate removal can be expressed
as:
S - S
= kX S 4-4
t a
120
-------
60 -
en
\
o
CO
*
-------
The terms k and Xa have frequently been combined as a single constant
K - (kXa) . This equation has been used to correlate aerated lagoon data.
A plot of the removal rate (SQ - S/t) versus the effluent BOD, S, generally
results in a straight line the slope of which is K = (kXa) . Typical data are
plotted in Figure 4-2 and the straight line describing these data has a
slope of 8.0. The rate constant was determined in these experimental
studies and k = 0.21; therefore, the active mixed liquids suspended solids
concentration Xa = 8/0.21 or approximately 40 mg/1. The data plotted in
Figure 4-2 are the results of analyses of composite samples collected from
laboratory units which were loaded at a constant rate. These laboratory
units were operated under nearly steady state conditions.
In the full-scale operation the influent wastewater flow varies; therefore,
the detention times in the aerated lagoon and the influent BOD also vary.
The equation for the unsteady state conditions can be derived from a mass
balance expressed as Equation 4-5.
QS0At - QSeAt = kXaVSeAt + VAS 4-5
The mixed liquid suspended solids concentration (Xa) can be expressed as
Equation 4-6.
a(S0 - S^
Xa ai+bt 4'6
The right hand side of Equation 4-6 can be used in Equation 4-5 in place
of X., and rewritten as Equation 4-7.
a
a (S - S )
QS At = QS At + kVS At ( , ° , 6 ) + VAS
o e e l + bt
4-7
Equation 4-7 can be rearranged as Equation 4-8 and used to calculate the
change in substrate concentrations.
, kaS
AS - t
-------
600
-------
-150
b_0
?m 50
0
t = 0.8 DAYS
J I I I I
t = 1.2 DAYS
0.4
0.6
0.8
1.0
0 0.2
TIME (days)
FKV4-3
AERATED LAGOON TRANSIENT CONDITIONS
-------
effluent BCD concentrations in Figure 4-3 varied from a minimum concentra-
tion of 7 . 6 mg/1 at the end of a nighttime period and increased to a maximum
concentration at the end of the 12-hour daylight period of 11.6 mg/1. As an influent
BOD concentration decreased during the nighttime hours, the effluent BOD con-
centration also decreased and reached a minimum concentration at the early
morning at a value of 7. 6 mg/1. However, the calculated effluent BOD using
a one-day detention time and the same rate constant was 8.85 mg/1. This
value compares closely to the effluent BCD of 9.1 mg/1 based on the com-
posite samples taken throughout the day. It should be pointed out at this
time that the variations in the effluent BOD concentration is caused primarily
by the changes in the detention time in the aerated lagoon system. The
effects of the changes of the influent BOD concentration are slight and only
affect the slope of the curve. Therefore, decreasing the detention time ac-
companied by an increase in the influent BOD results in a high effluent BOD
under steady state conditions.
The removal of soluble BOD is of interest for design purposes. The results
of laboratory-scale and full-scale experiments at the Williamson Creek
Wastewater Treatment Plant in Austin, Texas are plotted in Figure 4-4. The
detention time in the laboratory-scale units ranged from 0.09 to 0.54 days
(2.2 hours to 12.6 hours) while 'he detention time in the lagoons at the plant
varied from 0 .7 to 1.0 days , and 3-8 days for the C plant and B plant,
respectively. The data in Figure 4-4 indicate that the laboratory results
cluster at one part of the graph and the full-scale data cluster in another
area of the graph. The line through the laboratory-scale data has a slope
of minus 1 which indicates that the effluent BCD concentration is proportional
to reciprocal time (1/t) . These results confirm the theoretical considerations
which were discussed earlier. A line through the points which cluster around
the one-day detention time for the field-scale aerated lagoons also has a
slope of minus 1. The detention time does not seem to affect the effluent BOD
beyond the detention time of 2.5 days. At these long detention times the
effluent BOD was relatively constant and the soluble BOD ranged from three
to seven mg/1. The data indicate that the soluble BCD removal rate constant
in the full-scale process was less than that observed for the laboratory-scale
experiments .
It is doubtful that the effluent soluble BOD concentration could ever be reduced
to below four to five mg/1 even at longer detention times. The intersection
of the line describing the laboratory data and a horizontal line at four to five
mg/1 effluent soluble BOD occurs at a detention time of about 0.6 days. This
detention time is approximately equal to that at which the maximum active
mixed suspended solids was reported in the laboratory-scale lagoons. In-
creasing the detention time beyond the point at which the maximum active
mixed liquids suspended solids concentration occurs will not reduce the
125
-------
^
E
u_
o
o
CD
••
K
UJ
^
U.
U_
LU
UJ
CD
i
O
C/>
30
20
10
2
1
+ LABORATORY SCALE
FULL SCALE «
* TANK A
• TANK B
• TANK C
^.
22. •:
-.^ w
m
~ B •
- NUMBERS' °C
ill 1 i 1 i 1 1 III ! 1 i 1 1 IN
O.I 1.0 10
DETENTION TIME t (days)
FIG.4-4
SOLUBLE EFFLUENT - BOD VERSUS
DETENTION TIME
126
-------
effluent BOD concentration since
-------
200
CP
JE
CO
Q
_J
O
CO
O
LU
Q
LU
ID
CO
ID
_J
Lu
Lu
Ld
!60r-
120
80
40
0
POWER LEVELS
• BML,l8hp/IMG
• CML,32hp/IMG
• BML,llhp/IMG
20 40 60 RO 100
BOOT EFFLUENT (mg/l)
120
FIG.4-5
EFFLUENT SUSPENDED SOLIDS VERSUS
EFFLUENT BODT
128
-------
Published data which describe the performance of aerated lagoons treating
municipal wastewater are relatively sparse. McKinney and Benjes (1965)
reported that the effluent of an oonHod lagoon treating municipal wastewater
contained SO mg/1 of BOJ)
-------
optimum detention time for design purposes would be two days for an aerated
lagoon operating at a temperature of 20°C (68°F) . Longer detention times
would be required at lower temperatures since the rate constant k decreases
with decreasing temperature. The detention time required for operation at
temperatures less than 20°C can be calculated from Equation 4-11.
It is interesting to note that the curves in Figure 4-4 indicate that tempera-
ture has very little influence on the effluent BOD at detention times longer
than 2.5 days. However, the scatter of the data around the detention time
of one day for the field lagoons is partially caused by temperature variations
The temperature of the liquid in an aerated lagoon is generally different than
the temperature of the influent was tewater. Bishop, Malina , and Ecken-
f elder (1971) derived the following mathematical model for predicting temper-
atures of the lagoon, and this model is presented as Equation 4-12.
8.34 QT. + 145A (T - 2)
T _ - i - £ - 419
L 145A + 8.34 Q
in which:
T = weekly average lagoon temperature ( F)
J_t
T. = weekly average influent temperature ( F)
T = weekly average air temperature ( F)
f\
Q = wastewater flow (gal/day)
A = surface area of the lagoon (sq ft)
145 = average heat exchange coefficient
(BTU/sq ft - day - °F)
This model can be converted to a dimensionaless form expressed in Equation
4-13.
145 (-
T
4-13
i 145 + 8.34 (
130
-------
A nomograph has been deyo loped rrluLinq the ratio of Ihf: air temperature to
the influent temperature L(T - /0/T.J as well as the hydraulic surface
loading (Q/A) expressed as (gal/sq ft/day) . The relationship of detention
time is also included in Figure 4-6 for a lagoon depth of ten feet. The
curves indicate that at a short detention time the temperature has a rela-
tively small influence. Bishop, Malina , and Eckenfelder (1971) have shown
that this model was relatively insensitive with respect to the heat exchange
coefficient. The term (T^ - 2) approximates the equilibrium temperature
which is a temperature that a mixed water body without influent or effluent
reaches when exposed to air.
Oxygen Uptake
The oxygen uptake for a completely mixed aerated lagoon will follow the
basic relationship expressed as
D S - S
=a' + b<
The value of S in Equation 4-14 is the nonbiodegraded fraction of the
influent, and it is not the measured effluent BOD concentration. It can be
assumed that at a theoretical detention time longer than one day about 90
to 95 percent of the initial BOD has been degraded biologically and that
a1 = 0.53 Ib/lb. Therefore, 0 .53 (So - 0 .05 SQ) = 0 . 53 (0 . 95) SQ= 0 . 5 S
and Equation 4-14 can be rewritten in a similar manner to that for the
activated sludge process as Equation 4-15.
=°-5 + b'
Since the active mixed liquor suspended solids concentration is not readily
measurable in the aerated lagoon system. Equation 4-15 can be simplified.
S
R = 0.5—^ + b'X 4-16
However, the active mixed liquor suspended solids concentration can be
assumed to be approximately equal to 0.5 S . This value is a function of
the power level and for a completely mixed aerated lagoon Equation 4-16
can be rewritten.
R=°-5So(f+b'
131
-------
LU
§ 1.0
H
-------
in which:
R = oxygen uptake per unit volume (lb/1000 gal - day)
S = total BOD concentration in the influent (mg/1)
o
b1 = endogenous uptake rate (day ) (b1 = 0.15 for municipal
wastewaters)
In aerated lagoons in which sludge deposits undergo anaerobic decomposi-
tion, the soluble organic compounds released during anaerobic decomposi-
tion will be introduced into the bulk liquid and exert an additional oxygen
demand. The rate of anaerobic decomposition is higher at the higher temp-
eratures during the summer months and relatively low during the winter
months. Eckenfelder (1970) introduced factors of 1.05 and 1.2, to account
for the feedback of oxygen demanding substances from the anaerobic de-
composition of the settled sludge for winter and summer conditions, respec-
tively. Therefore, Eguation 4-17 can be rewritten for summer conditions
and winter conditions, respectively.
R = 0. 5 S (^ +b) (1.2) (Summer) 4-18
= 0.5 SQ (+b) ~- (1.05) (Winter) 4-19
Mixing and Power Levels
The degree of mixing has the greatest influence on the overall performance
of aerated lagoons. At high mixing levels all of the suspended solids are
maintained in suspension; therefore, the effluent suspended solids and
total BOD concentrations are relatively high. At the lower mixing level ,
a portion of the solids settle to the bottom of the lagoon, and the concen-
tration of suspended solids in the effluent is much lower than that observed
at the higher mixing levels. Therefore, the effluent concentration of total
BOD is also considerably lower. The degree of mixing in an aerated lagoon
is a function of a type of aeration equipment employed and the power level
expressed as HP/1000 gal.
The relationship between the concentration of suspended solids in the
aerated lagoon and the power level was evaluated at the aerated lagoons
at the Williamson Creek Wastewater Treatment Plant in Austin, Texas.
The aerated lagoon was operated at the highest possible power level for
a period of time. After the suspended solids concentration in a lagoon was
determined the power level was reduced in the evening. The suspended
solids concentration in the lagoon were determined at two different times
133
-------
of the next day at several sampling points. This procedure was repeated
until the lowest power level was applied.
The results of this investigation are presented in Figure 4-7 which relates
the concentration of miked liquor suspended solids to power level. The
data indicate that there is a sharp decrease in the suspended solids con-
centration as the power level decreases. However, at a minimum power
level the concentration of a mixed liquor suspended solids reached an
equilibrium value. The concentration of mixed liquor suspended solids
was about 55 mg/1 at power levels, between 9 and 13 horsepower per million
gallons. The data also indicate that the highest possible power level for
the aerator and pond geometry was 20 horsepower per million gallons. At
this power level all the solids in the system were not in suspension,
since theoretically, the maximum mixed liquor suspended solids concen-
tration would have been approached asymptotically. However, in this
particular aerated lagoon system, the maximum mixed liquor suspended
solids concentration could have been only slightly higher than the value
of 130 mg/1 indicated in Figure 4-7 since the influent suspended solids
concentration was not significantly greater in this value.
The mixed liquor suspended solids concentration under steady state
operating conditions for an aerated lagoon is affected by the influent
suspended solids, the detention time and the degree of mixing. The data
observed in field-scale ponds and presented in Figure 4-8 indicate that
a wide range of possible effluent suspended solids concentrations can be
observed at a given power level. There is a slight increase in the effluent
suspended solids concentration as the power level increases; however,
this increase is not markedly obvious. A comparision of the suspended
solids concentration in the aerated lagoon as a function of the power level
with results of mixing studies reported by Bishop, Malina , and Eckenfelder
(1971) provides an interesting insight into the problem of mixing aerated
lagoons. The degree of mixing is expressed as a number of completely
mixed tanks in a series as discussed earlier. A tank is completely mixed
as the calculated number of tanks in series is one. An infinite number of
tanks in series represents a plug flow system. The relationship between
the number of completely mixed tanks in series and the power level for the
aerated lagoons at the Williamson Creek Treatment Plant is presented in
Figure 4-9. The data indicate that the number of completely mixed tanks
in series is approximately one at power levels between 30 and 50 horse-
power per million gallons. As the power level decreases below 30 horse-
power per million gallons, the number of completely mixed tanks in series
increases and at power levels below 20 horsepower per million gallons
the number of completely mixed tanks in series increases sharply. The
results of tracer studies as well as the relationship between mixed liquor
-------
!40
120
100
-------
200
150
100
50
0
TANK A
TANK B
TANK C
WILLIAMSON CREEK
AUSTIN
OCT 1969 - JULY 1970
1
0 10 20 30
POWER LEVEL (hp/!06gal)
FIG.4-8
MLSS AND POWER LEVEL
40
136
-------
5
en
I
Q 4
LU
X
««>
*z.
\ ^
— \ 7L
LU °
LU
Q_
0 2
o
1
I
\
\
\
\
\
\
0 \
u_
o
cc
m 1
Z)
0
V.
i i i i I
0 10 20 30 40 50
POWER LEVEL (hp/IOsgol)
FIG.4-9
POWER LEVEL AND DEGREE OF MIXING
137
-------
suspended solids concentration and power level substantiate the assumption
by Eckenfelder (1970) that aerated lagoons can be considered completely
mixed tanks at power levels greater than 30 horsepower per million gallons.
The power level required to provide sufficient oxygen transfer to satisfy
the oxygen uptake rates can be calculated based on the influent BOD and
the detention time using Equation 4-17 . The oxygen transfer rate must be
equal to the oxygen uptake rate and may be calculated using Equation 4-20.
C — C
S20
(NQ) (Pv) (24) (1.02") 4-20
in which:
OT = oxygen transfer rate (lb/1000 gal )
C = concentration of dissolved oxygen at saturation (mg/1)
s
C = concentration of dissolved oxygen in the lagoon (mg/P
N = oxygen transfer efficiency (Ib/HP-hr)
o
Combining Equations 4-17 and 4-20 which are equal and solving for the
power level results in Equation 4-21 which can be used for different deten
tion times and for different influent BOD loadings .
P =1.73xlO(--.
v C - C NQ t
4-21
The curves presented in Figure 4-10 are based on an oxygen transfer
efficiency No = 1.7 lb of O2 per hp-hr, a dissolved oxygen concentration
under summer conditions of 7 .0 mg/1 and a requirement of a minimum dis-
solved oxygen concentration in a lagoon of 2 .0 mg/1 . The power level
calculated using Equation 4-21 is that required to maintain oxygen transfer
at a rate sufficient to satisfy the oxygen uptake requirements . The series
of curves presented in Figure 4-10 indicate that the power level required
to satisfy the oxygen uptake requirements in a lagoon operating at a one
day detention time and treating municipal wastewater with BOD concentra-
tions between 100 and 150 mg/1 is approximately 35 to 40 horsepower per
million gallons. This power level is also sufficient to maintain all sus-
pended solids in suspension. Therefore, the lagoon is completely mixed
and aerobic. However, if the detention time in a lagoon treating the same
wastewater is increased to two days , the power level required to provide
138
-------
0
en
to
O
X.
^s.
Q.
_J
LU
UJ
_J
or
LU
o
Q_
50
40
30
20
10
0
\ AEROBIC LAGOON
\\
i *
150
% 200 mg/l BOD
__\_\\
I ^ \ AEROBIC -
\ \ v ANAEROBIC LAGOON
\ \ \
V V v
~" » ^ x
\ 100 v^ ^^ ^* *^ ^. ^
1 1 1 — ~ — — —
01 2345
DETENTION TIME (days)
FIG.4-10
POWER LEVEL FOR OXYGEN TRANSFER
139
-------
oxygen to meet the oxygen uptake requirement is reduced to between 10 and
15 HP/million gallons. At this lower power level in the larger lagoon the
intensity of mixing would not be sufficient to completely mix the basin and
settling of the suspended solids will occur.
The results of the mixing studies indicate that a certain degree of bypassing
occurred in the lagoons at the Williamson Creek Plant. Relatively high con-
centrations of tracer were measured in the effluent consistently at about
10 to 20 minutes after the tracer input. Therefore, to minimize the short-
circuiting or bypassing in a square tank in which a single surface aerator
is located in the center of the tank, it would be advantageous to introduce
the influent wastewater directly below the aerators .
Effluent Suspended Solids
The suspended solids concentration in the effluent of a completely mixed
aerated lagoon can be calculated using a modification of Equation 2-22.
(S + X )
. ° v,
G Xt
-±=0.6 - 0.075 4-22
However, in the aerated lagoon process the sludge age (G) is approximately
equal to the detention time in the basin; therefore, Equation 4-22 can be
modified
X=0.6(S +X)-0.075Xt 4-23
o o
The concentration of mixed liquor suspended solids can be calculated from
Equation 4-24.
0.6 (S +X )
x = - o - o_ 4_24
1 + 0.075 t
in which:
S = total influent BOD concentration (mg/1)
o
X = influent suspended solids concentration (mg/1)
o
t = theoretical detention time (days)
X = effluent suspended solids concentration (mg/1)
The results of laboratory-scale experiments in which completely mixed aerated
lagoons were operated verify the applicability of Equation 4-24 to predict
the effluent suspended solids concentration.
-------
The effluent suspended solids concentration in the aerated lagoons in which
a portion of the suspended solids settle is a function of the power level, the
location of the effluent, the construction of the effluent structure (baffles,
weirs, stilling basins) and the detention time. A rough estimate of the
effluent suspended solids concentration for .this type of lagoon can be ob-
tained from the curve in Figure 4-7 which relates the effluent suspended
solids concentration to the power level.
DESIGN FACTORS
The basic data required for the design and layout of an aerated lagoon system
to treat municipal wastewater include:
(a) wastewater flow (MGD)
(b) average influent BOD concentration (mg/1)
(c) average influent suspended solids concentration (mg/1)
(d) the temperature of the wastewater including the lowest
and highest average weekly liquid and air temperatures .
Experimental results and published data indicate that a high reduction in
the concentration of soluble BCD can be achieved in aerated lagoons operat-
ing at a relatively short detention time of about two days and at a temper-
ature of about 20°C. The effluent total BOD and suspended solids of aerated
lagoons operated at detention times of one to eight days are generally high
when compared with the usual effluent requirements. The effluent suspended
solids generally settle well. An aerated lagoon with a three day detention
time followed by a clarifier will result in a final total BOD of 15 mg/1 or
less. However, the total BOD of the aerated lagoon effluent which contains
the biodegradable settleable suspended solids will be considerably higher
than the 15 mg/1. Aerated lagoons operated at detention time in excess of
20 to 30 days, are very similar to waste stabilization ponds.
Proper design and location of the aeration equipment will enable the system
to produce an acceptable effluent. However, land requirements for this
type of system are greater than for the activated sludge system. Aerated
lagoons can effectively treat municipal wastewater in relatively short de-
tention times; however, a portion of the effluent suspended solids must be
removed to reduce the total BOD to the acceptable levels. An aerated lagoon
can be followed by a waste stabilization pond, by another aerated lagoon
operated at a very low power level, or by using a final clarifier from which the
concentrated sludge may be pumped to a disposal site. A schematic repre-
sentation of these various alternatives is presented in Figure 4-11.
-------
AERATED
LAGOON
i
SPONDE
STABILIZATION
POND
AERATED LAGOON
LOW POWER-
LEVEL
FIG.4-11
AERATED LAGOON SYSTEMS
-------
DESIGN
Thf.- following procedure for tho dosign of an aerated lagoon system should
result in effective treatment of municipal wastowater:
(a) Select a detention time which will result in a low residual
soluble BOD. A detention time of two days at an operating temperature
of 20°C should be satisfactory for the treatment of municipal wastewater
at an initial total BOD of 100 to 150 mg/1.
(b) The selected detention time should be modified for temperature
changes . The minimum lagoon temperature can be computed from the curves
in Figure 4-6 using the minimum expected air and wastewater temperatures.
The required detention time can be determined based on Equation 4-11.
This approach requires a trial and error procedure.
(c) The volume of pond required can then be calculated (V = Qt) .
(d) The power level required for oxygen transfer to satisfy the
oxygen uptake requirements can be calculated from Equation 4-21. It
should be pointed out that this calculated power level in most cases only
is sufficient to satisfy the oxygen requirements for biodegradation.
(e) Compare the power level required for oxygen transfer with
that required to maintain an effluent suspended solids concentration (mg/1)
about equal to the influent BOD concentration; e.g. if the influent BOD is
equal to 100 mg/1, the effluent suspended solids concentration can also
be about 100 mg/1. The curve in Figure 4-7 indicates that a power level
of about 20 horsepower per million gallons would be sufficient to maintain
an effluent suspended solids concentration of about 100 mg/1. If a com-
pletely mixed aerobic lagoon is desired, the required power level for mix-
ing should be equal to or more than 30 horsepower per million gallons.
(f) In the case of the aerated lagoon in which some settling will
take place, the design of the aerators should be based on the higher of
the two required power levels. The total power requirement for the aerators
can be calculated based on the total volume of the aerated lagoon.
(g) Select the sludge handling system based on the method of
removal of suspended solids from the effluent.
-------
TRICKLING FILTER PROCESS
PROCESS DESCRIPTION
A trickling filter consists of a bed of coarse material such as broken
stones, clinkers, wood slats, plastic tubes, corrugated plastic sections,
or other material over which wastewater is distributed. Wastewater flows
over the medium on which a zoogleal slime develops. Dissolved organic
material in the wastewater is transported into the slime where biological
oxidation takes place and the effluent liquid is collected in an underdrain
system. Air passes through the void spaces in the medium and supplies
the oxygen required to maintain an aerobic environment. A trickling filter
will operate properly as long as the void spaces are not clogged by the
nrjuent solids or by excessive slime growth. In a well operating filter
the wastewater flows in a vertical direction with very little crossflow.
The wastewater is applied in such a way that the zoogleal mass is alter-
nately in contact with the wastewater and air.
Trickling filters have been operated over a wide range of hydraulic and
organic loadings. The classical definitions of a high-rate and a low-rate
filter are presented in the Mnnual for Sewage Treatment Plant Design (1967) .
However, the classical definitions are not related to the performance of
the trickling filter; therefore, these relationships are not very applicable
to the design of trickling filter plants.
The need for primary sedimentation prior to the trickling filter is a function
of the concentration of suspended solids in the influent wastewater as well
as the effective size of the medium. Stone medium generally requires
effective primary clarification to minimize problems with clogging. Primary
sedimentation is not necessary for those systems in which the medium con-
sists of corrugated plastic or other media which have large voids.
Various alternate flow diagrams for trickling filter plants are illustrated
in Figure 5-1. Trickling filters may be operated with recycling of part of
the effluent as well as by having a number of filters in series. The sus-
pended solids in the effluent of the trickling filter are biological materials
which slough off the medium. This sloughing action takes place as the
slime grows to such an extent that the hydraulic shear force resulting from
the downward flow of the wastewater actually separates the slime from the
surface of the medium and carries it to the underdrain system.
Trickling filters are relatively simple to operate, since the primary control
of the performance involves the rate of recirculation of filter effluent.
-------
Trickling filters hove been reported to recover readily from shock loads,
however, the performance of trickling filters is affected much more notice-
ably by shock loadings than that of the activated sludge process . The
applicability of the trickling filter process may be summarized by the
following statement from the Manual of Design of Sewage Treatment Plants
(1967): "They are capable of providing adequate treatment of such wastes
where the production of a plant effluent of 20 to 30 milligrams per liter
of BOD is acceptable or where partial treatment is acceptable."
PERFORMANCE OF TRICKLING FILTERS
The flow pattern in a trickling filter is generally assumed to be plug flow.
However, the results of tracer studies indicate that some mixing takes
place, but no backmixing is possible. Some particles of liquid will pass
through the filter at a much more rapid rate than others. In spite of this
mixing the mechanism of biodegradation of the soluble BOD can be described
by first order reaction kinetics applied to a plug flow process.
The overall performance of a trickling filter involves a number of process., s
including coagulation, flocculation, and/or biodegradation of the particu-
late organic matter as well as the efficiency of the final clarifier in sep-
arating the suspended solids from the liquid. Eckenfelder (1961) correlated
trickling filter data using a retardant reaction which is a first order reaction
applied to a completely mixed system.
The basic mathematical model describing the biodegradation of soluble
substrate in a plug flow system was presented as Equation 2-13, which
is rewritten as Equation 5-1:
Se -kXt , ,
= e 5-1
S*
o
in which
S* = effluent concentration of soluble substrate (mg/1)
e
S* = influent concentration of soluble substrate (mg/1)
o
k = rate constant (time )
X = concentration of active microorganisms (mg/1)
t = time of flow through the filter
-------
o
FIG.5-1
TRICKLING FILTER; FLOW DIAGRAMS
-------
If it is assumed that the entire filter medium is covered with a uniform
layer of microorganisms, the total microbial population is a function of
the available surface and can be expressed as Equation 5-2.
X=c,A 5-2
1 v
in which
X = active microbial population per unit volume of
filter
A = specific surface of medium (sq ft/cu ft)
v
c. = constant
The factor Cj is a function of the thickness of the film, the density of the
microbial population, the penetration of oxygen, and the transport of sub-
strate. Oxygen and substrate are transported into the film by means of
diffusion processes. The aerobic surface of a thin film is normally under-
lain by an anaerobic layer. Therefore, a very thick film does not nec-
essarily indicate that the number of active aerobic microorganisms is very
large. The soluble organic compounds released during anaerobic degrada-
tion are transported into the aerobic film.
The flow-through time is a function of the real flow per unit area of nominal
surface area of the filter medium and may be calculated from Equation 5-3,
if the flow velocity is assumed to be independent of depth.
t = c2 DQ~n 5-3
in which
t = time of flow to travel the depth of the filter
Q = hydraulic surface loading (million gallons per acre
per day (MGAD))
D = filter depth (feet)
Cj , n = constants which characterize the medium
The constants n, and c^, are a function of the size and configuration of
the medium. The value of n is relatively constant for several filter media,'
however, the coefficient 02 is a function of the specific surface areas
-------
(c£ = c~ A ) . In practice it is very difficult to independently evaluate
the exponents m and n. Equation 5-3 can be rewritten as Equation 5-4.
t = c0 D Am Q'n 5-4
3 v
The biodegradation rate constant, k, the coefficient, c , and the specific
surface area are interdependent and cannot be effectively evaluated separ-
ately. The coefficient, c2, which relates the time of flow through the
filter with the depth and the hydraulic loading can be determined from
tracer studies using tracer response techniques. However, when these
coefficients are determined from biological experimentation, a single con-
stant is developed which can be described by Equation 5-5 .
k* = k GI c2 5-5
Equation 5-1 can be modified and used to calculate the effluent substrate
concentration.
8* - S* e -
-------
The rate constant kg is the slope of the line resulting from a plot of the
fraction of BOD remaining (Se/SQ) at various depths of filter medium. This
particular plot is shown in Figure 5-2 . The data indicate that the rate
constant, kg, is a function of the hydraulic loading since the slope of the
line is different for different flow rates. The relationship between the rate
constant and the hydraulic loading rate can be obtained by rearranging and
rewriting Equation 5-7.
k* A
log kQ = log ( ) - n log Q 5-10
The exponent, n, is the slope of the line resulting from the plot of kg and
the hydraulic loading, Q, shown in Figure 5-2. The rate constant of k*
can be calculated from Equation 5-10 since, at a hydraulic loading of one,
log Q is zero and
k* A
log kg = log ( ) 5-11
Therefore, by extrapolating the line to a hydraulic loading of one the
numerical value for kg can be graphically determined. The specific sur-
face area is constant for the particular medium; therefore, the rate constant,
k*, can be determined from Equation 5-11 or graphically. The graphical
method is shown by the third plot in Figure 5-2 in which the fraction of
substrate remaining (Se/So) is plotted versus the quantity (A^DQ"11) . The
slope of the line resulting from this plot is the rate constant, k*.
The exponent, n, is dimensionless; however, the rate constant, k*, is
dependent upon the term Q~n and is markedly affected by the units of the
hydraulic loading. For example, a different numerical value will result
if the hydraulic loading is expressed in terms of gallons per minute per
square foot (gpm/sq ft) or million gallons per day per acre (MGAD) or cubic
meters per square meter day (rnVm^ - day) . The conversion from one set
of units to the next is presented in Equation 5-12.
5-12
MGAD = 0.016 gpm/sq f t = 0.939 mVm2 - day
The influence of the exponent, n, and the units in which the hydraulic
loading is expressed are summarized in Table 5-1.
150
-------
LOG
LOGKQ
LOG
Se
DEPTH
*o.
Q
Q.
Q
LOG HYDRAULIC LOADING
K
AVDQ
-n
FIG.5-2
TRICKLING FILTER EVALUATION
151
-------
Table 5-1 Influence of n and Q on Rate Constant k*
3 2
n/Q MGAD gpm/sq ft m /m - day
0.7
k*
1.0
k*
0.055
k*
0.955
0.4 1.0 0.19 0.974
Additional confusion is introduced when values of the rate constant, k*,
are presented without indicating the logarithmic base on which the values
were determined. The rate constant, k*, derived from the slope of the
plot is in units of base 10 logarithms; therefore, k * = 2.303 k* . In the
mathematical models describing the performance of trickling filters all
equations are presented as exponential functions to the base e. Therefore,
the rate constant, k*, should be expressed in terms of the base e along with
the units of the hydraulic loading. The numerical value of the exponent,
n, should also be included with the other data .
The influence of the exponent, n, on the rate constant, k*, is illustrated
graphically in Figure 5-3. Experimental data at two different hydraulic
loadings to a 20-foot deep filter are used to calculate the rate constant.
The exponent, n, was evaluated from tracer studies and n =0.70. Using
this value of the exponent the hydraulic loading did not markedly affect
the performance of the filter and the slope of the line through the data re-
sulted in a single rate constant, k* = 0.056. The exponent reported in the
literature is n =0.40. Separate lines fit the data observed at each of the
two different hydraulic loadings. These curves are also illustrated in
Figure 5-3. The slopes of the lines were k* = 0.0121 and O.C168, respectively
when the exponent was n = 0.40. This illustration may provide a partial
explanation for the relatively wide range of values reported in the literature
for the rate constant k*.
Data observed using a three-foot diameter pilot-scale trickling filter with
various media are presented in Figure 5-4. An exponent n = 0.50 was used
in calculating the data for presentation in this figure. The specific surface
area of the medium was 10 sq ft/cu ft of fresh limestone which had an effective
size of four to six inches, 27 sq ft/cu ft of corrugated plastic sheets and
30 sq ft/cu ft of plastic rings. The data observed in the experiments in which
the filter medium was either the corrugated plastic sheet or the plastic rings
compare quite well. However, it is interesting to note that the slope of
line drawn through the data will vary and depends on which data points might
be selected. The composite sample data were used to draw the line of best
152
-------
-z.
z
<
LU
cr
Q
o
CD
h-
^
PERCE
80
60
40
30
oo
C\J
c
i r\r\
IUU
80
60
40
30
^^"\
20
\ [n= 0.70 !
-\
^ • 20 mgad
_ \ +60 mgad i
\ k* =0.056
A
' , \, .
1 1 M 1
) 10 20 30 40
«^^
"^0 n=0.40
^*\
Ny^*. k*= 0.0121
v^
k*= 0.0168 TX. ^^^
4-.
^+
^-|-
0 20 40 60 80 100
AvDQ'n
FIG.5-3
INFLUENCE OF n ON k*
153
-------
100
o
80
_ * o
CORRUGATED PLASTIC
SHEETS
+ RUN I
• RUN 2
o RUN 3
PLASTIC RINGS
® RUN 6
Z
S 60
or
Q
f-\
CD 50
t-
UJ
o
cc 40
UJ
Q-
30
t-\ rwv,i\
\MO A RUN 74 9
_ ' T 0 COMPOSITE SAMPLES
\ jwk
'A V
I \"*/~\
~ \ K* 0.017 \
i / \
V \
1 \*
1 (jg) .A.
— • \
+" \
1 \
\ ®\
^1 \
\ \
1 ^ 1 1 1 1 1 i 1 1
0 10 20 30 40 50 60 70 80 90
AvDQ"a5
FIG.5-4 *
DETERMINATION OF k
PILOT PLANT DATA.GO/ALLE, AUSTIN
-------
fit. The average rate constant for the plastic media was k* - 0.017. The
data observed for the rock filter medium result in a line which had a slope
considerably different than those for the plastic media . The exponent n
for corrugated plastic medium was evaluated using data from three experi-
mental runs . The data reported by Gromiec and Malina (1970) indicate
that n = 0.73. The value of the exponent n = 0.70 is used for evaluating
the data observed when the filter medium was plastic rings. The data com-
pare very well with the data observed for corrugated plastic sheets. The
composite sample data observed for the rock medium will be comparable
to the plastic medium data only if the exponent n = 0.36. These data are
presented in Figure 5-5 and the slope of the line indicates that the rate
constant k* = 0.050. This rate constant is approximately three times that
calculated based on an exponent, n = 0.50.
These illustrations indicate the difficulty encountered when correlations
of data from different experiments are attempted. The two procedures dis-
cussed have limited applications because the exponent, n, is affected by
the hydraulic loading; however, the value does not reflect the effects of
the detention time. The detention time in the trickling filter was expressed
in Equation 5-3. The coefficient 02 used in this equation can be assumed
to be different for different filter media and is included in the overall rate
constant, k*. Therefore, the rate constant can also be expected to vary
with the type of filter medium.
The flow-through time is also a function of the configuration of the filter
medium and the hydraulic surface load. Therefore, the flow-through time
for different types of media can be characterized and the rate constant can
be calculated independent of the detention time.
FACTORS AFFECTING PERFORMANCE
The overall performance of trickling filters traditionally has been related to
the hydraulic surface loading and the organic loading. The performance
could be correlated to either parameter when the concentration of BOD of
the wastewater and the depth of the filter remain relatively constant.
The use of plastic media in place of rocks has led investigators to evaluate
other parameters which relate to the performance of trickling filters. Attempts
to correlate the quantity of filter slime to the actual surface area and in
turn determine the organic load in terms of pounds of BOD per unit area of
slime have been reported by Rincke and Wolters (1970) as well as by Lamb
and Owen (1970). The effects of depth on the overall performance of trick-
ling filters which were operated under identical conditions were reported
by Keefer and Meisel (1952), Audion et al. (1970), Chipperfield (1967), and
155
-------
100
LU
o:
Q
o
CD
UJ
o
a:
UJ
o_
80
^60
50
40
30
0
CORRUGATED PLASTIC
SHEETS n=0.73
-I- RUN I
• RUN 2
oRUN 3
PLASTIC RINGS
• RUN 6
10
-n
0
FIG.5-5
DETERMINATION OF K*
PILOT PLANT DATA'GOVALLE,AUSTIN
20
30
156
-------
Bruce and Merkens (1970) . Trickling filters were packed to a depth of up
to 25 foot with corrugated or tubular plastic medium having specific surface
areas of 12 .2 to 67 square feet per cubic foot. The influent BOD was 200
to 400 mg/1. The results of these studies indicated that the volumetric
BOD removal rate (Ib BOD/1000 cu ft-day) was dependent only on the volume-
tric BOD loading (Ib BOD applied/1000 cu ft-day) and was independent of
the depth.
The relationship between BOD removal and the specific surface area is
presented in Figure 5-6. The data indicate that the BOD removal increases
as the specific surface area increases. The results of tracer studies con-
ducted by Bruce (1970) are plotted in Figure 5-7. These data indicate that
the relationship between the detention time in the filter and the specific
surface area of the medium is approximately linear. The deviation from
linearity of the data observed at higher specific surface areas may be
partially attributable to the geometric configuration of the particular medium.
The performance of trickling filters is therefore related to the volumetric
BOD loading as well as the specific surface area of the filter medium. The
volume of filter medium required for effective treatment is considerably less
for plastic media which have higher specific surface areas than for rock.
Theoretically, the relationship between the volume required and the specific
surface area may be presented as Equation 5-13.
V A
x vx 5-13
V A
o vo
in which
V = required volume of plastic filter medium (cu ft)
x
V = required volume of rock filter medium (cu ft)
o
A = specific surface area of plastic medium (sq ft/cu ft)
vx
A = specific surface area of rock filter medium (sq ft/cu ft)
vo
This relationship has not been completely verified; however, some of the
results of recent experiments indicate that this relationship is valid. A
safety factor to reflect the effective slime area is incorporated when filter
medium other than rock is used. Equation 5-14 represents this relationship
in which it is assumed that only 70 percent of the difference in the specific
surface area of the plastic medium and the conventional rock medium is
157
-------
SPECIFIC SURFACE
(sqft/cuft)
FIG.5-6
BOD REMOVAL VERSUS SPECIFIC SURFACE
158
-------
o
LU
t-
LU
0
* i 5
or
0
(13.5 mgad)
20 40 60 80
SPECIFIC SURFACE (sqft/cu ft)
FIG.5-7
HYDRAULIC DETENTION TIME VERSUS
SPECIFIC SURFACE
159
-------
effective and that the specific surface area of rock filter medium, A - 12
sq ft/cu ft.
-
Vo " 1 + 0.7 *v
12
The relationship between the fraction of volume required for plastic filter
medium compared to conventional rock filter medium and the specific sur-
face area are presented in Figure 5-8. The required volume can be reduced
by 50 percent when filter medium which has a specific surface area of 30
sq ft/cu ft is used. Using a medium with a specific surface area of 60 sq
ft/cu ft results in a filter which occupies only 25 percent of the volume
required for conventional rock filter. However, increasing the specific
surface area beyond 60 sq ft/cu ft does not markedly reduce the volume
requirements. Therefore, the removal of BOD by trickling filters should
increase as the specific surface area increases. The results of experi-
ments by Bruce (1970) demonstrated this relationship.
The performance of trickling filters can be improved by increasing the specific
surface area of the filter medium. This relationship is illustrated in Figure
5-9. The series of curves indicate that the efficiency of BOD removal of
a high-rate rock trickling filter can be increased from 50 to 75 percent by re-
placing the rock medium with a plastic medium with a specific surface area
of 60 sq ft/cu ft. However, as the efficiency of the rock filter increases,
the improvement in the performance of the filter resulting from increasing the
specific surface of the medium is less. For example, increasing the specific
surface area of a high rate rock trickling filter operating at an efficiency of
95 percent would result in an efficiency of 98 percent. The specific surface
area of 60 sq ft/cu ft seems to be the limiting value since negligible changes
in the efficiency and performance of the filter are observed at higher values .
Other data indicate that problems with clogging are more common when the
filter medium has specific surface area of greater than 60 sq ft/cu ft.
Theoretically, the efficiency of BOD removal by low-rate trickling filters
can also be increased by using a filter medium which has a higher specific
surface area than rock. However, the results of investigations by Audion,
et al. (1970) indicate that the increase in efficiency was negligible at various
loading rates. This phenomenon may be explained by the fact that at low
hydraulic loadings the slime surface is not uniformly wet; therefore, only a
small portion of the slime is actually in contact with the wastewater. This
uneven wetting is more noticeable when corrugated or tubular medium is
used since the waste water flows vertically downward with very little lateral
distribution. The results of tracer studies reported by Bruce and Merkens
160
-------
i.O
> 0.8
Q
UJ
9=
S 0.6
-------
iOO
o
z
UJ
o
UJ
h-
z
LU
O
tr
LU
Q_
0 20 40 60 80
SPECIFIC SURFACE (sqft/cuft)
100
FIG.5-9
PERFORMANCE AND SPECIFIC SURFACE
162
-------
(1970) indicate that in a 7.5-foot deep trickling filter with corrugated
plastic medium, about 90 percent of the tracer was recovered within an
eight-inch diameter circle and that 95 percent of the tracer was recovered
within a 16-inch diameter circle. These data indicate that uniform wetting
of the surfaces of the filter medium is a very important design consideration.
Bruce and Merkens (1970) indicated that the minimum hydraulic loading for
a corrugated plastic medium for complete wetting of the surfaces was 10.5
gallons per square foot per hour.
The minimum hydraulic surface loading can be calculated based on the
assumption that the wetting or the thickness of the liquid film varies
linearly with the specific surface area. Equation 5-15 can be used to
calculate the minimum required hydraulic service loading, based on the
assumption that the specific surface area of rock medium AV = 12 sq ft/cu
ft and that the rock is uniformly wet at a hydraulic loading of 5 gal/sq
ft-hour.
qA = 5T7 = °'42Av 5-15
in which
q = hydraulic surface loading (gal/sq ft-hour)
f\
A = specific surface area of filter medium (sq ft/cu ft)
v
The removal of soluble BOD is a function of the depth of the filter medium;
however, the overall performance of a trickling filter was not affected by
depth beyond some minimal depth required for the removal of soluble organics.
The BOD of the effluent of the final clarifier following a trickling filter
is a function of the concentration of suspended solids in the effluent.
Bruce and Merkens (1970) reported that the nonsoluble BOD of the trickling
filter effluent was equal to 60 to 80 percent of the concentration of suspended
solids in the effluent. Therefore, the concentration of BOD in the effluent
of a trickling filter when the suspended solids concentration is 30 mg/1
would be at least 18 mg/1. Therefore, the overall performance of a trickling
filter plant can be improved by increasing the efficiency of removal of sus-
pended solids in the final clarifier. This relationship between the BOD and
suspended solids in the effluent of biological processes has also been
noted for the activated sludge process.
The minimum depth required for trickling filters should be based on the
desired effluent concentration of soluble BOD. The significance of depth
of the filter in connection with soluble BOD is illustrated in the following
example. Equation 5-6 can be used to calculate the hydraulic surface
163
-------
loading for , specific surface area,
A = 27 sq ft/cu ft and exponent, n = 0.73. Equation 5-6 can be reduced
to the relationship between hydraulic surface loading and depth.
5-16
The calculated hydraulic surface loading for a ten-foot and a 20-foot deep
filter are 8.0 MGAD and 20.8 MGAD, respectively. Therefore, the volume
of trickling filter required for 95 percent removal of soluble BOD for a
20-foot filter is approximately one-half the volume required for a ten-foot
filter. Applying a hydraulic load of 20.8 MGAD to a ten-foot trickling
filter would result in a soluble BOD removal efficiency of only 78 percent.
Therefore, trickling filters should be designed to be deep enough to pro-
duce an effluent which has the desired concentration of soluble BOD. The
overall performance of trickling filters is affected to some extent by re-
circulation of the effluent. Recirculation is recommended in high-rate
filters in order to minimize clogging and ponding caused by growth of the
slime. The problem of clogging is minimized to some extent by using
corrugated plastic filter media which have a relatively high fraction of
void spaces. The effect of recirculation on removal of soluble BOD can
be determined by developing a mass balance, as indicated in Equation
5-17.
Q S -i- Q (RC)S
0 e = S 5-17
Q (1 + RC) e
The rate of recirculation is included in the term RC which is the ratio of
the recirculated flow to the incoming flow. Equation 5-17 can be combined
with Equation 5-6 and rewritten as
S* [ 1 + (RC)] -kA D[(l + RC) Q]~n
6 =e V 5-18
S* + (RC) S*
o e
S* -kA D [ (1 + RC) Q]~n
v
o -kA D [(1 + RC) Q]~n
[ (1 + RC) - (RC) ] e V
-------
in which
S* , S* = concentration of soluble influent and effluent BOD
Q = hydraulic surface loading without recirculation (MGAD)
RC = Recirculation ratio = (rate of flow of recycle/rate of
flow of influent)
The efficiency of BOD removal using recirculation can be calculated using
Equation 5-19 and the parameters for corrugated plastic media. The ef-
ficiency for a ten-foot deep filter with a hydraulic surface loading without
recirculation of 20 .8 million gallons per day is approximately 78 percent.
At a recirculation ratio RG = 1 and RC = 2 the efficiency was about the same,
namely 74 percent. In this example, theoretically the performance of the
trickling filter is not improved by using recirculation. Similar results were
also reported for experiments using corrugated plastic medium at the Govalle
Wastewater Treatment Plant in Austin, Texas. These data were observed at
hydraulic surface loading rates of 94.5 MGAD without recirculation and with
100 percent recirculation. The effluent soluble BOD in both cases was ap-
proximately the same.
Recirculation can improve the soluble BCD removal efficiency of a trickling
filter in those cases where the exponent, n, is relatively low and when the
soluble effluent BOD is relatively high without recirculation. The quality
of the effluent of trickling filters has usually been reported to be improved
when recirculation is practiced. This improved effluent quality is more than
likely the result of more uniform wetting of the filter medium and the result
of washing out of the accumulated solids from the medium. Therefore, gen-
eral design practice should include recirculation pumps for conventional
(rock) trickling filters . The overall effect of recirculation on performance
of a trickling filter plant should be evaluated for each particular installation.
The effluent of the trickling filter rather than that of the secondary clarifier
should be recycled either directly to the filter or through the primary clarifier.
This practice is preferred, since in many plants chemical flocculating agents
would be added to the secondary clarifier in order to improve the efficiency
of solids separation.
The operating temperature has an effect on the overall performance of a
trickling filter plant in two ways. The biological processes are temperature
dependent and the settling velocity of the suspended solids is affected by
the change in viscosity resulting from temperature changes. The effect of
temperature on the rate of biodegradation is expressed in Equation 5-20.
165
-------
The tomporaturfj coefficient, 9, has been reported to vary from 1.035 to
1.072 (Kokonfclder , 1966). The discrepancy in the value of the temperature
coefficient may be attributable to the difference in the depth of penetration
of oxygen and the rate of oxygen uptake at different temperatures. Eckenfelder
(1966) indicated that the effects of temperature changes vould be much more
marked at higher loading. Bruce (1970) indicated that the temperature coef-
ficient, 9, for the overall performance of a trickling filter plant, ranged from
1.1 to 1.35 with an average value of 1.21. The average monthly temperature
during these experiments varied from 9°C to 18°C (48°F to 64°F) . During
these studies, the efficiency in some of the filters dropped from 80 percent
to 40 percent and in other filters from 55 percent to 20 percent as the temper-
ature dropped. The effect of temperature on high-rate trickling filters must
be considered in evaluating the performance of these treatment units.
ROCK TRICKLING FILTERS
The conventional trickling filter contains a rock medium which has a relatively
narrow range of specific surface areas. Various mathematical models have
been developed to describe the performance of the trickling filters in terms
of hydraulic surface load or organic loading rather than specific surface area.
Fairall (1966) developed the mathematical model expressed in Equation 5-21.
S S k V ~
o e
This equation is illustrated graphically in Figure 5-10 in which the rate of
substrate removal is related to the volumetric hydraulic loading. The slope
of the line through the data is the reciprocal of the rate constant, 1/k. The
wide range of the rate constant from k = 1 to k = 10 (day"1) indicates that
the rate constant is a function of the influent BOD or of the BOD volume
load. The data presented by Bruce (1970) also indicate that the BOD volume
load affects the rate constant. Therefore, the correlation of the BOD volume
load to the efficiency of BOD removal is the preferred method of evaluating
data.
The National Research Council (1946) compiled data for 34 operating trickling
filters over an eight month period. The range of BOD removal was between
75 and 95 percent for organic loading of up to 2000 Ib BOD/acre-day. These
data are presented graphically in Figure 5-11 and the average performance of
a trickling filter based on these data can be expressed by the model.
166
-------
5-
1.0
o>
CO
O>
CO
I
O
CO
0.
F-AIRALL1I956)
10 20 30
HYDRAULIC VOLUME LOAD (gal /cu ft /day)
FIG. 5-10
RANGE OF RATE CONSTANTS
167
-------
100
80
O
^
LU
60
NRC (1946)
LjJ
O
-------
So ~ Se 1
( g ) = fj-r- 5-22
o S Q
1 + 0.0085 ("—)
in which
SO = BOD applied (Ib/day)
V = volume of medium (acre-feet)
1 + RC
F = number of effective passes =•
[1 + 0.1 (RC)] 2
RC = recirculation ratio = (recycle flow rate/influent flow
rate)
The effect of BOD volumetric load on the efficiency of trickling filters has
also been reported by Keefer and Meisel (1952) and Rincke (1967) . These
data are presented graphically in Figure 5-12 and indicate that with increas-
ing organic loading rates, the overall performance of trickling filters de-
creases. The average performance of trickling filters with depths of ten
to 14 feet was developed by Rincke (1967) and is expressed in the form of
the mathematical model:
S - S QS
100 (- -)= 93-0.272 -£- 5-23
in which
QS
o
V
BOD volume load (Ib BOD/1000 cu ft - day)
The NRC (1946) data cover a much broader spectrum of operating conditions
than the data reported by Keefer and Meisel (1952) or by Rincke (1967) . The
data reported by Keefer and Rincke are much more comparable and cover a
much narrower range, since these data were collected under relatively
similar operating conditions .
EXCESS SLUDGE
The excess sludge produced in a trickling filter is a function of the rate of
conversion of substrate to cell material, the endogenous respiration rate and
the accumulation of non-biological solids in the filter slime. Sludge production
169
-------
100
80
UJ
a:
60
100
KEEFER(MEISEL(I952)
Ld
O
or
LU
80
60
40
RINCKECI967)
i
0 25 50
BOD-VOLUME-LOAD (Ib XIOOOcu f t/day)
FIG.5-12
REMOVAL AND BOD-VOLUME-LOAD
170
-------
in trickling filters is much similar to that developed in the activated sludge
process. The quantity of excess sludge in a trickling filter plant can be
determined by a mass balance presented as Equation 5-24.
QX = Q a (S - S*) - bVX + QX 5-24
e o e on
in which
0 = wastewater flow (MGD)
V = filter volume ( cu ft)
X = effluent suspended solids from filter (mg/1)
C
X = influent nonbiodegradable suspended solids (mg/1)
S = influent BOD (mg/1)
S* = effluent soluble BOD (mg/1)
ti
X = biological sludge mass per unit volume (Ib/cu ft)
a, b = constants
The sludge mass in a trickling filter can be related to the specific surface
area by Equation 5-25 .
VX = f (VAv) 5-25
Combining 5-24 and 5-25 and rearranging the terms results in
T^~~ * = -Q- [ a (S - S*) + X ] - b r 9K
YA e VA o e on b-zb
v v
The constants in this equation can be evaluated graphically as illustrated
in Figure 5-13 which represents a plot of the (BOD + SS) loading versus the
excess sludge. The data are based on the assumptions that the effluent
soluble BOD concentration is approximately zero and that the influent non-
biodegradable suspended solids equal the total suspended solids entering
the system. The data presented in Figure 5-13 represent experimental results
reported by Bruce and Merkens (1970) and data observed at the Govalle
Wastewater Treatment Plant in Austin, Texas,for rock and plastic ring media ,
respectively.
171
-------
-I-BRUCE (1970)
• AUSTIN.ROCK
•AUSTIN,PLASTIC RINGS
O
8
30
Ld
g 20
ID
_J
co
CO
LU
O
X
LJ
10
0
10 20 30 40
(BOD+ SS) - LOAD (Ib/IOOOsqft/day)
50
FIG.5-13
EXCESS SLUDGE FROM TRICKLING FILTERS
172
-------
The slope for the data reported for the Austin plant is somewhat higher than
that for the other data . The wastewater at the Austin plant has a lower ratio
of BOD to suspended solids which might account for the higher slope. The
slope of the lines tend to decrease at the higher loading and is possibly the
result of a lower rate of solids biodegradation, at the higher loadings than
at lower loadings. An estimate of the excess sludge from a high-rate trick-
ling filter can be generally determined by Equation 5-27 which is derived
from the data plotted in Figure 5-13.
X - 0.5,-- (BOD + SS)-0.5 5-27
VA e VA
v v
The excess solids are expressed in terms of pounds of solids per thousand
square feet of filter medium per day in this equation. Equation 5-27 can
be written as
VA
X = 0.5 (BOD + SS) - 0.5 (— — ) 5-28
e Q
This equation indicates that 50 percent of the (BOD + SS) is converted to
suspended solids which remain in the filter effluent. This conversion
factor is somewhat lower than tnat reported for activated sludge systems.
The quantity of excess sludge produced in a low-rate trickling filter is
much lower than that reported for high-rate filters or for the activated
sludge process . The lower rate of solids accumulation may be attribut-
able to the grazing activities of protozoa. The activity of the protozoa
is reduced considerably at low temperatures. Therefore, slime tends to
accumulate in the trickling filter during winter operation and the filter tends
to unload the slime in the spring when the activity of the microorganisms
is once again increased.
DESIGN FACTORS
The data required for design of trickling filter plants is the same information
as is necessary for the design of activated sludge plants.
Conventional rock trickling filters require excellent solids removal in primary
clarification to avoid clogging of the filter medium and ponding of the waste-
water. Corrugated or tubular plastic materials provide a filter medium with
relatively high void spaces; therefore, more influent suspended solids can
be tolerated. A primary clarifier should precede any trickling filter treating
municipal wastewater.
173
-------
Recirculation of the effluent of the trickling filter to the primary clarifier
or to the trickling filter may be practiced. The primary advantage of re-
circulation is a uniform wetting of the slime on the filter medium and
possibly an increased removal of soluble BOD when high effluent concen-
trations result without recirculation.
Filter Volume and Filter Medium
The volume of the medium required and the cost can be calculated for
trickling filters. This cost of rock can be compared with the cost of other
filter media. The cost estimate should include the medium, the structure
to house the medium, and the pumping requirements. A conventional rock-
line material has specific surface area which ranges between ten and 20
sq ft/cu ft whereas the plastic media have specific surface areas that
range from 25 to 70 sq ft/cu ft. The filter medium can be classified as
follows:
(a) randomly packed material (rocks, stones, plastic rings , raschid
rings, etc.)
(b) self-supporting modules or corregated plastic sheets
(c) suspended corrugated sheets
(d) self-supporting plastic pipe or tubular materials
The main advantage of the plastic media is a high percentage of void spaces
which is between 90 and 97 percent of the total volume compared to 30 to
50 percent of the volume for conventional rock filters. The void space pro-
vides for circulation of air as well as minimizes the potential clogging
problems. As the specific surface area increases, the void space decreases.
The relationship between the clear space openings and the specific surface
areas for various types and configurations of synthetic media are presented
in Figure 5-14. It is interesting to note that square pipes, honeycomb pipes,
spheres, and corrugated sheets which have the same diameter of open space
or distance between sheets have about the same specific surface area .
Circular pipes exhibit the highest specific surface area if the inside and
outside surfaces of the pipe are included. However, at relatively small
pipe diameters, the void space between the pipes becomes clogged with
solids and the specific surface will decrease to about one-half the theoretical
value. Corrugated plastic sheets have specific surface areas which are
about twice that of a plain sheet. Calculation of the specific surface area
for corrugated plastic sheets which are placed 2.1 inches apart indicate a
theoretical specific surface area of 27.5 sq ft/cu ft. This value compares
favorably with the value of 27 sq ft/cu ft reported by the manufacturers.
-------
SPHERES
HONEYCOMBS
AND SQUARES
CORRUGATED SHEETS
FLAT SHEE
'0.2 0.3
0.5 1.0 2.0 3.0
DIAMETER (inches)
FIG.5-14
RELATIONSHIP OF SPECIFIC SURFACE AND
DIAMETER FOR VARIOUS MEDIA
175
-------
Therefore, the curves in J'Kjure 5-14 can be used to (jive preliminary
estimates of the relationship of specific surface arga and the distance
between sheets or clear diameters. A clear opening of 0.7 - 0.8 inches
is required to avoid clogging of the medium by slime. This distance
therefore limits the maximum specific surface area to about 70 sq ft/cu
ft.
Operating Factors
There are two operational techniques of improving the efficiency of a
trickling filter system which is already installed. Recirculation may be
employed or the dosing frequency can be changed. Recirculation can mini-
mize ponding and clogging by washing out accumulated slime and solids.
Hawks and Shepherd (1970) reported that the dosing frequency also can
affect the operation of a trickling filter. Dosing frequencies of 0.3 and
14 minutes were used over a five-year period of operation. The results
indicate that in the summer, the performance of the trickling filter was
better at high dosing frequency and in the winter the performance was
better at the low dosing frequency. At the lower dosing frequency, a high
hydraulic loading is applied to relatively small areas of the filter and the
washout effect is greater. At tha low dosing frequency in the wintertime,
less slime tends to accumulate than at the high dosing frequencies.
Variation of dosing frequency might be considered to improve the opera-
tion of low-rate trickling filters as well.
Design Formulation
Various design formulations have been developed and various standards
of design have been recommended. The Ten State Standards (1968) recom-
mended design loading rates of up to 50 Ib BOD/1000 cu ft - day. The
BOD removal efficiency at this range of organic loadings is between 67 and
75 percent. These efficiencies are at the lower end of the range reported
in Figures 5-9 and 5-10. These standards do not include any temperature
effects, and it seems that the design performance includes winter condi-
tions. Other published information indicates that the efficiency is limited
to about 80 to 85 percent regardless of the type of filter medium and depth
of medium used. Low-rate trickling filters can be operated at higher ef-
ficiencies than the high-rate trickling filters.
The results of studies conducted in England with smaller size graded rocks
and slags indicate that detention times in these filters are relatively long
and that high BOD removal efficiencies are possible. Eden (1964) reports
that for a five-foot deep" trickling filter, detention times of one to two hours
are possible depending on the rate of hydraulic application. Therefore, a
higher efficiency can be obtained at these longer detention times.
176
-------
Trickling filters which are packed with plastic media require uniform wetting
and the flow of the wastewater is essentially vertically downward. There-
fore, long detention times are not possible with this type of medium. In
England, two stage filtration is employed for many industrial wastes. The
first-stage trickling filter is operated as a high-rate filter with a plastic
medium. The second-stage trickling filter is operated at a low-rate using
conventional rock or slag media.
The performance of single stage conventional rock filters at different BOD
volume loads are presented in Table 5-2. The efficiency is expressed as
the average as well as the recommended design values. At the lowest BOD
volume loading, the effluent BOD concentration should average about 20
mg/1 although values of 25 mg/1 can be expected.
The required volume of filter medium can be reduced when plastic corrugated
or tubular material is used. The reduction in volume required can be cal-
culated using Equation 5-14, and the hydraulic surface load required to
maintain uniform wetting of the medium can be computed from Equation 5-15.
The excess sludge produced during high rate trickling filtration can be
calculated from Equation 5-27 .
TABLE 5-2
TRICKLING FILTER DESIGN
Conventional Filling Material
(A ~ 12 sq ft/cu ft)
Efficiency
BOD Volume Load Average Design
lb/103 cu ft/day
10 -82% 75%
50 -77% 67%
75 -70% 60%
177
-------
WASTi; STAIifT.IZATION PONDS
PROCESS DESCRIPTION
Waste stabilization ponds are biological waste treatment systems in which
algal photo synthetic and bacterial oxidation are effective in stabilizing a
portion of the organic material in wastewaters. The process is controlled
to a large extent by climatic conditions, primarily temperature and wind
action.
A waste stabilization pond system can include a single pond, a number of
ponds in series or parallel. The environment within the pond and the pur-
pose for which the pond is used are the bases for classification of ponds
as anaerobic ponds, facultative ponds and maturation ponds. These three
types of ponds are generally used for the treatment of wastewaters. High-
rate ponds have also been used in some cases for a growth of algae which
are to be harvested.
Anaerobic ponds are those ponds in which no dissolved oxygen is detected
immediately below the surface. The rate of oxygen utilization is much
greater than the rate of reoxygenation. Anaerobic ponds for the treatment
of municipal wastewater are essentially large sedimentation and digestion
basins. Long detention times are provided for the settled solids to undergo
anaerobic degradation. The effluent of an anaerobic pond contains a rela-
tively high concentration of soluble and colloidal organic material (BOD)
and requires additional treatment.
Facultative ponds are most commonly used for the treatment of wastewaters.
A benthic anaerobic zone of activity is overlain by an aerobic zone of
biological activity near the surface. A portion of the suspended solids
entering a facultative pond will settle to the bottom and undergo anaerobic
decomposition. The soluble organic material in the wastewater and those
soluble organics released by the anaerobic degradation of the organic solids
are used by heterotrophic bacteria and converted to carbon dioxide and new
bacterial cellular material. The oxygen required for aerobic activity is
provided by algal photosynthesis and to a lesser extent by diffusion of
atmospheric oxygen into the surface waters of the pond. The effluent of
facultative ponds will contain a relatively low concentration of soluble BOD
and varying concentrations of algal cells .
Maturation ponds are used to polish the effluent of facultative ponds or
of other biological treatment systems. A reduction in the bacterial content,
179
-------
suspended solids, and some nutrients are reali/.ed in maturation ponds.
The removal of BOD in maturation ponds is negligible.
KINETIC MODELS AND DESIGN CONSIDERATIONS
The number of mathematical models developed for the design of waste
stabilization ponds is limited, although pond systems have been widely
used and the operation of ponds investigated during the past twenty years.
Most of the mathematical models have not been developed in sufficient
detail to completely describe the reactions and microbial activity within
the pond. In other cases, the application of the mathematical models is
limited to a specific geographical region or to ponds of a specific design.
ANAEROBIC PONDS
The mechanism of anaerobic degradation of organic material involves the
sequential activity of facultative microorganisms and methane forming
bacteria. Complex organic material is hydrolyzed by the facultative
bacteria to organic acids , which can be converted by methane forming
bacteria to methane and carbon dioxide gases. Therefore, carbon can be
removed from the system by methane fermentation.
Anaerobic ponds are designed to maintain environmental conditions which
are favorable for the development of methane bacteria. The primary factors
affecting the growth of methane bacteria are temperature, pH, detention
time, and organic loading rate. Methane bacteria grow relatively showly
compared to facultative organisms and require much longer detention times
for development of an adequate population to carry out effective methane
fermentation. The presence of free dissolved oxygen in the environment
can be inhibitory to the methane bacteria. The facultative bacteria use
the dissolved oxygen and protect the methane bacteria from exposure to
oxygen. However, a thin algal surface layer in a pond will help minimize
odors. The minimal temperature for active growth of methane bacteria
is approximately 20°C. However, methane bacteria will continue to grow
but at a much reduced rate at temperatures as low as 15°C, but the rate of
gas production will become almost negligible. The pH range for effective
methane fermentation is between pH 6.6 and pH 7.2.
The hydraulic detention time in anaerobic ponds should be between two and
five days and is about the same as the generation time of the more rapidly
growing methane forming bacteria. Other methane bacteria have generation
times of 20 to 30 days. The more slowly growing methane forming bacteria
are more likely to be found in the sludge deposits in the anaerobic pond
than in the liquid phase. At these long hydraulic detention times most of
180
-------
the suspended solids will settle to the bottom. The organic fraction of
the solids in the benthic zone will undergo decomposition and some of the
organic material will be converted to methane gas. Aguirre and Gloyna
(1971) indicate that a 50 percent removal of BOD was observed for anaerobic
ponds treating municipal wastewater at the Govalle Treatment Plant in
Austin, Texas, at organic loadings of 900 pounds of ultimate BOD per acre
per day (six Ib BODU/1000 cu ft - day) and a detention time of 2 .3 days
at a temperature of 16°C (60°F) . No odors were noticeable at these con-
ditions . However, at an organic loading of 1600 Ib BODjj/acre-day (11
Ib BODU/1000 cu ft-day) and a detention time of 1.3 days, considerable
odors were noticeable although the removal of BOD was also approximately
50 percent. At the higher loading, the temperature was 28°C ( 82°F) , and
the production of odor was probably the result of active fermentation and
mixing rather than inadequate detention time.
White (1970) indicated that the average volumetric loading recommended
in various states was 12 to 15 Ib BOD/1000 cu ft - day. However, the
volumetric loadings varied from a minimum of three Ib BOD/1000 cu ft -
day for Georgia to a maximum of 1000 Ib BOD/1000 cu ft - day in Texas.
The Texas regulations also recommend that the solids loading to the pond
be between 100 and 400 Ib SS/1000 cu ft - day and detention time between
five and 30 days. A comparison of these recommended criteria with the
results reported by Aguirre and Gloyna (1971) indicate that the higher
loadings are for the treatment of industrial wastes in which the concentra-
tion of organic material and settleable solids are relatively high, and high
loading rates are possible at relatively long detention times. However,
with domestic wastewater in which the concentration of BOD is approxi-
mately 200 mg/1 loadings of 12 to 15 Ib BOD/1000 cu ft - day would be
achieved at hydraulic detention times of one to two days .
The depth of anaerobic ponds should be between 10 and 15 feet. Sufficient
volume should be provided for the storage of sludge resulting from the ac-
cumulation of settled solids . The introduction of the incoming wastewater
near the bottom of the pond and distributed as uniformly as possible over
the entire area of the pond is preferred.
Facultative Ponds
Facultative ponds are waste stabilization systems in which the removal
of organic material takes place via two mechanisms, namely anaerobic
degradation and facultative or aerobic biodegradation of the organic material
by bacteria with oxygen provided by photosynthesis by algae. The accum-
ulated organic solids in the sludge layer are degraded anaerobically and
result in the formation of methane, carbon dioxide and soluble compounds .
181
-------
Algal photosynthesis produces some of the oxygen required to maintain
aerobic, odor-free conditions. Algae utilize the inorganic carbon which
is made available from the aerobic respiration , the anaerobic decomposi-
tion of organic solids, the atmosphere and the influent wastewater. This
carbon is converted to cell material, and as such,is removed only when the
algae are harvested.
Aerobic respiration is responsible for the conversion of the organic material
to new cell material as well as the carbon dioxide which is then available
for algal photosynthesis. For facultative ponds the concentration of soluble
BOD of domestic wastewater under favorable environmental conditions may
be reduced to a residual level of approximately 15 mg/1.
The composition of algal cells includes carbon, nitrogen and phosphorus
in a ratio of about 106:16:1 expressed as millimoles per liter (Stumm and
Morgan, 1963) . The aerobic utilization of one mole of organic carbon
results in the conversion of approximately 65 percent of the carbon to
bacterial cell mass and approximately 35 percent of the carbon is released
as carbon dioxide (Wuhrmann, 1964) . In addition to the utilization of the
organic material, approximately 0.65 moles of carbon dioxide are produced
per mole of oxygen uptake. Algae release one mole of oxygen for each
mole of carbon dioxide utilized . Therefore , a net deficit of 0 .35 moles of
carbon dioxide exist. This deficit is in part satisfied by the anaerobic
degradation of cell material. If the dead cells undergo aerobic degradation,
all of the carbon is released as carbon dioxide and re-enters the carbon
cycle in the pond. However, since the dead cells generally settle to the
bottom of the pond and undergo anaerobic decomposition, about 70 percent
of the carbon is released as methane gas and 30 percent of the carbon re-
enters the carbon cycle, as carbon dioxide.
Two of the most important factors in facultative waste stabilization pond
design include:
(a) the dependence of algal growth on the total carbon in the
system
(b) The primary source of carbon is also derived from methane
fermentation
Therefore, the theoretical design of waste stabilization bonds should be
oriented toward a balanced system in which the algal population is main-
tained at the minimal level to insure aerobic conditions in the liquid zone.
However, it is practically impossible to control algal production at either
a maximum or minimum because of changes in the environment in the pond
as well as in the available nutrients. Therefore, the design should be
182
-------
based on maintaining a sufficient algal population to minimize anaerobic
conditions and to provide for efficient operation. The algae in the effluent
of a properly designed pond will probably represent 90 percent of the total
BOD leaving the pond. However, these cells will not exert an immediate
oxygen demand on the receiving stream, but do represent a suspended solids
load .
The various processes involved in the efficient operation of a facultative
stabilization pond are illustrated schematically in Figure 6-1. The inter-
action among the anaerobic decomposition in the benthic zone, the aerobic
respiration and the algal photosynthesis are shown.
The primary design criteria for facultative ponds involve maintaining a
minimum soluble organic concentration as well as a minimum algal concen-
tration in the pond effluent. The bacterial population developed in a pond
can be calculated from a mass balance illustrated as Equation 6-1 and
modified as Equation 6-2.
QX = a Q (S* - S*) - bVX 6-1
a o e a
a (S* - 5*)
v -
X ~
a ~ 1+ bt
in which
X = concentration of heterotrophic bacteria (mg/1)
a
S* = concentration of soluble BOD in the influent plus
° the BOD released by anaerobic decomposition (mg/1)
S* = concentration of soluble BOD in effluent (mg/1)
e
a = coefficient of conversion of BOD removed to bacterial
cells
b = coefficient endogenous respiration rate (day )
t = coefficient of detention time of liquid in aerobic
zone of a pond (day)
These equations are based on the assumption that the aerobic portion of the
ponds is completely mixed and that the removal kinetics of soluble organic
material are similar to those of the aerated lagoon process. The model
183
-------
LIGHT
4
METHANE
INORGANIC CARBON
NITROGEN + PHOSPHORUS
SOLUBLE ORGANICS
ALGAL PHOTOSYNTHESIS
SETTLEABLE ORGANICS
CO,
CO.
N '
P
AEROBIC DECOMPOSITION
SOLIDS
I
SOLUBLE ORGANICS
ANAEROBIC DECOMPOSITION
HUMUS
FIG.6-1
MECHANISMS OF DEGRADATIONS IN FACULATIVE PONDS
ALGAE
NITROGEN 4 PHOSPHORUS^
SOLUBLE ORGANICS.
INORGANIC CARBON
BACTERIA
-------
for the biological degradation of soluble organics in the completely mixed
reactor is presented in Equation 6-3.
S* - S*
Y t- = kS* 6~3
Xt e
a
in which
k = BOD removal rate constant
The soluble organic material in the effluent can be calculated by combining
Equations 6-2 and 6-3.
s* = 1 + b t _ _J + b 6_4
e akt akt ak
This mathematical model indicates that a residual soluble organic concen-
tration (S§ = b/ak) will remain in all systems. The concentration of organic
material in the effluent above this residual is a function of detention time
as well as the reaction rate constant. In this particular model, the con-
stants b and k are temperature dependent, and can be calculated for any
temperature by using either Equation 6-5 or 6-6.
, JT-20) ,. _
c K
6-6
The mathematical model implied bacterial growth. The concentration of
heterotrophic bacteria in the pond will be higher as the rate of substrate
removal [(S* - S|)/t] increases. This model can be applied to the entire
pond or to sections of the pond. At the inlet section the influent substrate
(S*) is relatively high, and the bacterial activity will be higher than at
the outlet section where the substrate concentration is relatively low.
However, the gross removal of soluble substrate will be dependent upon
the volume of the aerobic zone and the duration of aerobic conditions.
For example, if the depth of the aerobic zone of a pond during the daytime
hours is one half the pond depth and if the dissolved oxygen concentration
in the pond is zero during the night-time hours , the volume of the aerobic
zone would be one fourth the volume of the entire pond and the duration
of aerobic conditions would be only one fourth of the hydraulic detention
time. The aerobic zone can be represented as a specific volume of the
pond (V) . Therefore, the soluble effluent BOD should be a function of
185
-------
the organic load applied to the aerobic volume and the temperature of the
pond. This relationship can be written as Equation 6-7.
QS
Se = f ( ~\?~) ' (T) 6'7
However, the depth of the aerobic zone will fluctuate daily and seasonally.
Therefore, a more practical relationship includes the effluent BOD and
BOD surface load (Ib BOD/acre) as shown in Equation 6-8.
QS
S* =f (-T2-) , (T) 6-8
6 f\
Hermann and Gloyna (1958) proposed one of the first rational design
equations for stabilization ponds. A modified presentation of this mathe-
matical formulation is presented as Equation 6-9, which can be used to
calculate the required pond volume for the removal of 90 percent of the
influent BOD and is related to the fixed optimum detention time (t ) at
35°C. The optimum detention time is based on an average influent BOD
of municipal wastewater, SQ = 200 mg/1. This relationship also includes
a factor to compensate for any algal toxicity which might result from con-
stituents in the municipal wastewater (Hwang and Gloyna, 1967) and a sulfide
compensation factor (Espino and Gloyna, 1968) .
OS (35 - T)
V - K < 20§> (to> 9 (0 (n 6-9
in which
V = pond volume required (acre - ft) or (cu ft)
Q = wastewater flow rate (gal/day)
S - influent BOD concentration (mg/1)
t = optimum time (days)
0 = temperature coefficient
T = minimum monthly temperature (°C)
f = algal toxicity or compensation factor
f - sulfato, S O2- effect (f - 1 for S
186
-------
K = conversion factor to make units of the terms compatible
Marais and Shaw (1961) applied first order kinetics to the waste stabiliza-
tion pond performance and assumed completely mixed conditions. The
mathematical model which was developed is presented as Equation 6-10.
Se i
So 1 + k,9<35 -
in which
S , S = influent and effluent BOD (mg/1)
\J C
k = rate constant (day )
t = hydraulic detention time (day)
These mathematical models provide a reliable foundation for estimating
the required size of waste stabilization ponds. In general, these equations
consider the influence of temperature and the biodegradation rate. Marais
and Shaw (1961) recommend a value of the rate constant, k = 0.17 which
was calculated from data observed in full-scale stabilization ponds. Hermann
and Gloyna (1958) applied an optimum detention time (to) developed from
laboratory-scale work in baffled aquaria.
The relationship between the percent BOD remaining and detention time for
the Hermann-Gloyna equation are illustrated in Figure 6-2. The data re-
ported by Suwannakarn and Gloyna (1963) were for ponds operated at a temper-
ature of 35°C and the substrate was synthetic wastewater; however, the
data reported by Hermann and Gloyna were for municipal wastewater. The
optimum detention time for 90 percent BOD removal was about 3.8 days. The
optimum rate constant, kg5 = 0.60, was determined graphically from Figure
6-2. This rate constant can be adjusted to a temperature of 20°C using the
relationship shown as Equation 6-5 and the rate constant k2o = 0.18 is for
a temperature coefficient, 9 -- 1.085. This rate constant is very similar
to that reported by Marais and Shaw (1961) for a temperature of 20°C. The
temperature coefficient, 9, for stabilization ponds was developed by
Suwannakarn and Gloyna (1963) . The results of these investigations are
presented in Figure 6-3 in which the effect of temperature on BOD removal
is illustrated.
The optimum detention time for a single pond can be calculated from Equation
6-10 based on the assumption that the rate constant is the same in plug flow
and completely mixed systems. The optimum detention time (to) therefore
is 15 days at an operating temperature of 35°C based on the assumption that
187
-------
100
ui
cr
o
o
CD
z
UJ
o
tr
UJ
o.
10
o SUWARNNAKARN(I963) 35°C
+ HERMANN (1953) 25-33°C
024
DETENTION TIME (days)
FIG. 6-2
BOD REMOVAL IN LABORATORY
PONDS
188
-------
CD
LJ
or
Q
o
GQ
LOADING = 38.8 Ib BOD
acre-day
6 9 12 15
DETENTION TIME (days)
FIG.6-3
BOD REMAINING IN LAB SCALE PONDS
(SUWARNNAKARN -1963)
189
-------
90 percent of the soluble BOD is removed; therefore, Se/So = 0.1 and
on a rate constant k = 0.60. This apparently long detention time may be
explained by the fact that the equation is representative of a completely
mixed system while the rate constant was based on data observed in a plug
flow system. The substrate used in the experiments reported by Suwanna-
karn (1963) was almost completely soluble and readily biodegradable.
The hydraulic regime established in most stabilization pond systems is
somewhere between a plug flow and completely mixed systems . Therefore,
the recommended optimum detention time (to) of seven days was introduced
by Gloyna (1969) as a modification of the original relationship. This op-
timum time is applicable to those systems which closely approximate the
conditions in rectangular stabilization ponds with length to width ratios of
between 2:1 to 3:1.
Equation 6-9 can be modified and rearranged in terms of areal loading
n
200 o
1'085 6'12
0 (8.34) _ 0.326 (200) (8.34) d
A = 1.085(35-T)(t0)
QSo(8.34) M4d
-
1.085 (to)
in which
QS (8.34)
= areal BOD loading (Ib/acre - day)
A = surface area of the pond (acres)
d = depth of the pond (feet)
T = operating temperature, minimum monthly average
t = optimum detention time for 90 percent BOD removal
at 35°C (days)
0.326 = conversion from acre - feet to gallons
8.34 = conversion of gallons to pounds
190
-------
Equation 6-10 can be rearranged in a similar fashion to result in the same
equation as Equation 6-14.
A graphical presentation of Equation 6-14 is illustrated in Figure 6-4. The
curves represent the relationship between the areal loading and temperature
for 90 percent removal of BOD at three different detention times representing
plug flow system, a completely mixed basin and an intermediate case,
respectively.
The relationship between the detention time required for 90 percent removal
of BOD, the operating temperature, and the influent concentration of BOD
is illustrated in Figure 6-5. The detention time of seven days represents a
hydraulic flow regime somewhere between plug flow and completely mixed
system and was selected for calculation of these data. The curves in
Figure 6-4 and 6-5 are applicable over a temperature range some 5°C to 35°C.
The effect of temperature on the areal loading which may be applied to
ponds and result in a BOD removal of 90 percent is graphically presented
in Figure 6-6. Algal growth is generally inhibited in temperatures in excess
of 35°C and odors are likely to be produced from the ponds under these
conditions. At temperatures below 5°C , an ice cover generally develops
during winter conditions and tho rate of biodegradation is reduced to a
negligible rate. In this case the ponds should be designed to impound the
entire winter flow and effluent should be discharged only during the summer
and fall months.
The approximate range of effluent BOD concentration is useful for design
purposes and the average performance is only of interest over a long period
of time. Operating data from various installations were assembled by Aguirre
and Gloyna (1971) and are presented graphically in Figure 6-7. The wide
scatter in the data represents the influences of temperature, type of waste-
water treated, location of ponds, etc. The results of pilot-scale studies
of waste stabilization pond performance conducted at the Govalle Waste-
water Treatment plant in Austin, Texas are presented in Figure 6-8. These
data indicate that the effluent BOD observed at loadings between 30 and
150 Ib BOD/acre-day was lower than the maximum effluent BOD reported
in the literature and illustrated in Figure 6-7. The data presented in
Figure 6-8 indicate that the effluent quality expressed as total BOD is
dependent upon the areal loading rate. However, the soluble effluent
BOD is independent of the loading within the range of loading conditions
used in this study. Two theoretical effluent quality curves for facultative
ponds are also included in Figure 6-8. These curves are based on calcu-
lated values using Equation 6-14 for operating temperatures of 10°C and
20°C (50°F and 68°F) , respectively. The influent BOD (SQ) was assumed
to be 200 mg/1 and the relationship of optimum detention time, BOD removal
and the rate constant. The optimum detention (to) was calculated from the
relationship presented in Equation 6-15.
191
-------
10 15 20 25
TEMPERATURE (°C)
FIG. 6-4
AREAL LOADING RATE FOR 90% REMOVAL
192
35
-------
200
100
90
80
70
60
50
40
30
20
UJ
1
h-
O
H
LU
H-
LU
Q
10
9
8
7
6
5|
4
_s
l085
(35-T)
where t0= 7.0 days
1
1
10 15 20 25
TEMPERATURE (°C)
30
35
FIG.6-5
DETENTION TIME FOR 90% BOD REMOVAL
193
-------
120
j? 100
IT)
Q
o 80
CD
H
Z
LU
U_
U_
LU
Q
LU
(T
LU
60
40
20
i r
HERMANN-GLOYNA RELATIONSHIP
(MODIFIED)
I I
j I
0 50 100 150 200 250
AREAL LOADING (Ib BOD5/a/d)
FIG. 6-6
EFFLUENT QUALITY AS A FUNCTION
OF AREAL LOADING
191*
-------
200
250
50 100 150
BOD LOAD (Ib/acre/day)
FIG.6-7
RANGE OF LOAD AND EFFLUENT FROM WASTE
STABILIZATION PONDS
195
-------
120
100
c 80
\
o>
in
o
§ 60
LU
t 40
UJ
20
+ FILTERED
• TOTAL
T=IO°C
(Theoretical)
//T=28°Cto30°C
T=20°C
(Theoretical)
j_
50 100 150 200
AREAL LOADING (Ib/ocre/day)
250
FIG.6-8
EXPERIMENTAL AND THEORETICAL EFFLUENT
QUALITY AS A FUNCTION OF AREAL LOADING
196
-------
( So -1)
t -T 6-15
o
The two theoretical curves follow a similar pattern to the other data and
indicate that the constants developed are in the useful range. However,
for design purposes, these constants should not be over-emphasized since
the experimental data observed at pond temperatures of 20°C and above
resemble the performance of a pond operating at 10°C . This information
indicates that the rate constant in the pilot-scale stabilization pond was
much lower than the assumed theoretical value. The calculated rate
constants assuming completely mixed conditions are shown in Figure 6-9.
The rate constant calculated from the straight line indicates that k = 0 .11.
However, the dashed curve drawn through points indicates that the rate
constant might be a function of the loading and is similar to those reported
for other biological waste treatment processes.
Design of waste stabilization ponds is generally based on the surface
loading rate. In the United States, the recommended surface loading is
between 16 and 50 Ib BOD/acre-day (Englande, 1969) . The lower surface
loadings were used primarily in the northern states. However, under
favorable environmental conditions, waste stabilization ponds may be
effectively operated at much higher loading rates. Oswald (1968) operated
ponds in California at surface loadings in excess of 125 Ib BOD/acre-day.
Horning, et al. (1964) reported that ponds in Ohio could effectively treat
municipal wastewater at surface loadings of about 100 Ib BOD/acre-day.
Canter et al. (1969) & Mills (1961) reported that loadings of up to 200 Ib
BOD/acre-day have been applied to ponds and produced acceptable efflu-
ents in the southern part of the United States. Parker, et al. (1959) recom-
mended loadings of 100 Ib of BOD/acre-day and 60 Ib BOD/acre-day for
summer and winter conditions, respectively in Australia . Ponds of South
Africa have been operated successfully at surface loadings of 120 Ib
BOD/acre-day and as high as 250 Ib BOD/acre-day if recirculation was
included in the design (Meiring, et al., 1968) .
The effluent BOD concentration which can be expected at various loading
rates is important in the design of waste stabilization ponds. The maximum
effluent BOD based on published data can be related to the loading rate as
illustrated in Figure 6-7. The theoretical performance at temperatures of
10°C and 20°C (50°F and 68°F) , were used to develop average effluent con-
centrations and were related to design loading rates as illustrated in Table
6-1.
The soluble effluent BOD of stabilization ponds will range from 5 mg/1 to
15 mg/1. However, the effluent suspended solids may range from 40 to 200
mg/1 at higher loadings.
197
-------
CO
0 10 20 30 40 50 60 70 80
S(BOD5,mg/l)
FIG.6-9
CALCULATED BIODEGRADATION RATE
CONSTANT, ASSUMING COMPLETELY
MIXED CONDITIONS
198
-------
TABLE 6-1
LOADING RATES AND ErFLUENT BOD
EROM SINGLE FACULTATIVE WASTE STABILIZATION PONDS
5-8 FEET DEEP
LOADING RATE AVERAGE EFFLUENT BOD
Ib BOD/acre-day Temp. Temp.
5-15°C 15-25°C
12.5
25
50
100
150
10-20
15-25
20-40
30-60
40-80
10-15
10-20
10-30
20-50
25-60
MAXIMUM
EFFLUENT
BOD
30
40
55
75
90
199
-------
An acceptable effluent BOD can be produced by ponds in the southern states
at loadings of 50 to 150 Ib BOD/acre - day. In !.ho northern states, where
the ponds may be operated at low temperatures and under an ice cover
during part of the year, loadings of 20 to 50 Ib BOD/acre - day should be
used.
Sludge accumulation in facultative ponds is a function of the type of waste-
water and the quantity of settleable solids entering the ponds, the operating
temperature, and the rate of anaerobic degradation of the organic solids in
the benthic zone. The rate of synthesis of bacterial cells also affect the
solids accumulation in ponds.
MATURATION PONDS
Maturation ponds are usually used as tertiary treatment processes in which
the natural self-purification processes result in an improvement of the
quality of the effluent. Maturation ponds should not be used to provide
additional treatment to the effluent of under-designed conventional biolog-
ical processes. In maturation ponds, the concentration of bacteria, viruses
and nutrients, as well as suspended solids, can be reduced and thereby
improve the quality plant effluents.
The performance of maturation ponds is expressed primarily in terms of
reduction in the number of bacteria . The bacterial die-off rate is affected
by the detention times, sunlight, bactericidal agents released by algae,
temperature, dissolved oxygen, pH, predation, nutrient depletion, and
toxic materials (Davis and Gloyna, 1970) . Bacterial removal of between 90
and 99.9 percent have been reported in the literature. The total coliform
density of municipal wastewater expressed as the Most Probable Number
(MPN) is usually in a range of 10° to 10° per hundred milliliters. In most
cases the standards for discharge to receiving streams are set at an MPN
of 10^ per hundred milliliters. However, to meet these high effluent
standards , the reduction in the concentration of coliform organisms would
have to be about 10^ per 100 ml, or approximately 99.999 percent. These
high efficiencies cannot be guaranteed by a simple maturation pond even
at an extremely long detention time. In general, it is agreed that the
detention time for maturation ponds is between ten to 15 days .
PONDS IN SERIES
Waste stabilization ponds can be operated in series to minimize the
possibility of short circuiting and the discharge of partially treated efflu-
ent from stabilization ponds. In general, the effluent of the last pond of
a series is of better quality than that from a single pond which had the
identical detention time as the ponds in series. The results of pilot-scale
investigation at the Govalle Wastewater Treatment Plant in which a
200
-------
combination of ponds were operated in series and in parallel indicate mat
the system consisting of three ponds produced an effluent quality which is
better than that resulting from two ponds in series. A schematic diagram
of the pilot-scale stabilization pond systems are presented in Figure 6-10 .
One of the systems included an anaerobic pond, a facultative pond, and a
maturation pond in series . The second system included a facultative pond
which was deeper at the center to provide an anaerobic zone of the same
volume as the anaerobic pond in system one, followed by a maturation pond.
The third system included a facultative pond and a maturation pond. The
results of these investigations indicate that the three pond system produced
the best effluent. Approximately 50 percent of the BOD and organic carbon
in the influent wastewater was removed in the anaerobic pond. Therefore,
the BOD load to the facultative pond was reduced and the quantity of algal
cells produced in the facultative pond was also reduced, resulting in an
effluent of better quality. It might be advisable to install two anaerobic
ponds in order to provide sufficient time for the sludge accumulated in one
pond to undergo more complete anaerobic degradation and permit disposal
of the sludge on the land.
DESIGN FACTORS
The design of waste stabilization ponds for the treatment of municipal
wastewaters is based essentially upon the total quantity of wastewater to
be treated and in the influent BOD concentration. It is also essential that
any industrial wastes or toxic materials present in the incoming wastewater
be identified to minimize potential toxic effects on the algae. The basic
design information required includes:
(a) total wastewater flow per day (MGD)
(b) concentration of BOD (mg/1)
(c) ratio of five-day BOD to ultimate BOD
(d) the minimum and maximum air temperatures
(e) if freezing occurs, the duration of time during which an ice
cover can be expected
(f) the presence of any potentially toxic material
The design of waste stabilization pond systems is affected by the effluent
standards as well as the stream standards established by regulatory agencies
The design of waste stabilization ponds also is dependent upon the availa-
bility and location of suitable land which can be used as a plant site.
Waste stabilization pond systems are applicable for the treatment of
municipal wastewaters in those areas in which effluent standards have
201
-------
,s~- r~^7«
! X '/--
ANAEROBIC POND
DEPTH 10 FT.
MATURAT ON
PONDS D = 4 FT.
=
5—
D = 5.5'
100
X /•
-
'
80'
- 45'-
\
FACULTATIVE PONDS
81
(=5.5
)=9.5
8~
"0
100
•3
[•
i -_:
_P^
60'
230'
72'
100'
#3
FAC POND
MATUR POND
•
FIG.6-10
PILOT WASTE STABILIZATION PONDS
202
-------
been established which will permit an effluent BOD and suspended solids
(algal cells) concentration of more than 20 mg/1. The following design
procedure is proposed:
(a) The regulatory agencies accept the fact that reliable effluent
quality can be produced by waste stabilization ponds, but with
the understanding that the effluent quality will vary between
summer and winter operation. The pond system be designed to
meet the effluent standards as closely as possible.
(b) Relatively inexpensive land should be available for the
extension of the plant as population increases.
(c) The required volume for facultative ponds can be calculated
using Equation 6-9 and the areal loading can be calculated
from Equation 6-14. The curves presented in Figures 6-4,
6-5 , and 6-7 can be used to size the particular ponds.
(d) A maturation pond should follow a facultative pond to improve
the bacteriological quality of the plant effluent.
DESIGN EXAMPLE
Two waste stabilization pond systems are designed to treat municipal
wastewater. This example illustrates the difference between the perform-
ance of a single facultative pond followed by a maturation pond and a system
including an anaerobic pond followed by a facultative pond and a maturation
pond. The characteristics of the wastewater are:
Influent wastewater flow = 1 MGD
Influent concentration of BOD 5 =. 200 mg/1
Influent concentration of ultimate BOD = 300 mg/1
Influent concentration of suspended solids = 200 mg/1
The following climatological data are used in the design of the pond systems:
Average monthly temperature range = 10 to 30°C
Average temperature of the coldest month = 10°C
The rainfall is assumed to equal the evaporation
Two Pond System
The design procedure for the facultative pond is as follows:
The required detention time for the facultative pond can be
derived from Figure 6-5. At a temperature of 10°C and an
influent BOD of 300 mg/1, the required detention time is 82 days.
203
-------
The pond volume required is V = Q t
V = 1 x 106 S3! ( Cuft ) 82 day = 11 x 10& cu ft
day 7.48 gal
The surface area of the pond assuming a five-foot operating
depth is
sur d 5 it.
or
One foot of depth is added for sludge storage; therefore, the
volume is:
V=6ft (2.2 xlO6 sqft) 7-48gal = 99 x 106 gal
OV1 It
The organic load to the pond is:
1 x 106 gal , 300 mg> . 8.34 Ib* = 2500 Ib BODn
In6 , ( L ' ( gal j day
10 day
The areal load is therefore:
2500 Ib BODU/ day 49 . 5 Ib BODU
50. 5 acres ~ acre-day
The volumetric load is:
2500 Ib BODu/day 0.19 Ib BODU
13.2 x!0b cu ft 100 cu ft - day
Three Pond System
The required detention time of the anaerobic pond to minimize
odor problems is:
t = 5 days
The volume of the anaerobic pond must include sludge storage;
therefore:
Total Volume = Liquid Volume + Sludge Storage
-------
Liquid Volume = 5 days ( 1 x 10 -^~)( ? °" J = 0 . 67 x 10 cu ft
Sludge storage volume is based on the assumption that
100 percent of the solids are removed and concentrate
to three percent solids. The required detention time of
the solids for anaerobic degradation at 10°G is 90 days.
Dry solids removed:
The volume of sludge is:
1668 lb solids ( 100 lb sludge cu ft } Q d =0.8xl05cu f
day V 3 lb solids ' v 62.4 lb sludge'
Therefore , the Total Volume =
(7 x 105) + (0 . 8 x 105) = 7 . 8 x 105 cu ft , or 5 . 85 x 106 gal
5.85 x 1Q6 gal c oc ,
The total hydraulic detention time = j x 106 gal/day =
A depth of ten feet in anaerobic pond results in a surface area of:
= 0.78xl06cuft - acre =1.7 9 acres
sur 10 ft 43,560 sq ft
The loading to the anaerobic pond is:
2500 lb BOD /day 1395 lb BOD
Areal loading - 1.79 acres = acre - day
2500 lb BODu/day 3.2 lb BODu
Volume loading = 0. 78x10° cu ft = 1000 cu ft - day
The concentration of BOD entering the facultative pond assuming
50 percent BOD removal in the anaerobic pond is:
BOD =0.5 (300 mg/1) = 150 mg/1
u
The required detention time from Figure 6-5 for an operating
temperature of 10°C and BODu - 150 mg/1 is:
t = 41 days
205
-------
The required volume therefore is:
V-W daysll x!0- -SHJSt- = 5.48 x 106 Cu ft
At a pond depth of five feet, the surface area is:
Provision for sludge storage in the facultative pond following an anaerobic
pond is not necessary.
The organic load to the facultative pond is:
BOD = 1 x 106 gal ( 150^ Q^ U^ = 1250 lb/day
u 106day ' gal
The loadings to the facultative pond are:
1250 Ib BODU Ib BODU
Areal loading = — - - - — = 50 d
25 acre - day acre-day
1250 Ib BODu 0 . 28 Ib BODu
Volume loading = 5 >48 x 1Q6 ^ ft_day = 1000 cu ft-day
The comparison of the two systems is presented in Table 6-2. The area
requirements for the anaerobic-facultative-maturation pond system is almost
50 percent less than that required for the facultative-maturation-pond system
206
-------
TABLE 6-2
COMPARISON OF DESIGN REQUIREMENTS FOR A CONVENTIONAL
FACULTATIVE POND AND AN ANAEROBIC-FACULTATIVE POND SYSTEM
SYSTEM
Facultative
Maturation
ro — — __—__
° Total
Anaerobic
Facultative
Maturation
Total
DETENTION
TIME
(days)
82
15
97
5.85
41
15
61.85
VOLUME
(MG)
99
15
114
5.85
41
15
61.85
SURFACE
AREA
(acres)
50.5
12
62.5
1.79
25
12
38.79
AREAL
LOADING
,lb BODU ^
acre-day
49.5
1395
50
VOLUMETRIC
LOADING
Ib BODu
1000 cu ft - day
0.19
3.2
0.28
-------
ACKNOWLEDGMENTS
These guidelines were developed in connection with field-scale investiga-
tions of biological wastewater treatment processes at the various wastewater
treatment plants operated by the City of Austin, Texas. This manuscript
was prepared by Joseph F. Malina , Jr., Ph.D., P.E., Director of Environ-
mental Health Engineering Program, Professor of Civil Engineering at The
University of Texas at Austin, with the cooperation of Earnest F. Gloyna ,
Dr. Engr. , P.E. , Professor of Civil Engineering, Dean of the College of
Engineering, The University of Texas at Austin, W. Wesley Eckenfelder,
Jr. , P.E. , formerly Professor of Civil Engineering, University of Texas at
Austin and presently Professor of Water Resources in Sanitary Engineering
at Vanderbilt University, and Rolf Kayser, Ph.D. , Research Associate,
University of Texas at Austin and presently Research Engineer, University
of Braunschweig, West Germany.
W. R. Drynan, Ph.D. , Professor of Civil Engineering at The University of
Waterloo, Ontario, Canada, served as Project Manager during the first
year of the project and was instrumental in equipping the laboratory and
developing the operating schedule. Rolf Kayser, Ph.D. , served as the
Project Manager during the second year of the project and was instrumental
in completing the collection of experimental data and evaluating the per-
formance of the various processes.
The cooperation of the administrative and operating personnel of the
Department of Water and Wastewater Treatment of the City of Austin is
also acknowledged. Mr. Curtis E. Johnson, Assistant Director, Water
and Sewage Treatment, City of Austin, Texas, Mr. Mansel W. Smith, the
late Superintendent of the Wastewater Division, and Mr. D. F. Smallhorst,
Chief Engineer, provided technical and administrative inputs to the effec-
tive functioning of the project. Mr. Robert Pfaffman, Supervisor, Waste-
water Treatment Plant, City of Austin, was essential to the day-to-day
operation of the various processes at the treatment plants. Messrs. Bruce
E. Halbert, John F. Myatt, Richard K. Schmidt, Marek J. Gromiec, Frank
J. Fabre, Helmut R. Fleckseder, Donnie W. Berryhili, Neil E. Bishop,
Jorge Aguirre, John Nan-Chieu, and Walter Wen-Jo Chiang, who were
graduate students in Environmental Health Engineering Program contributed
to the evaluation of the various processes and the preparation of this
document.
The administrative assistance of Mr. George J. Putnicki, Acting Director,
Environmental Protection Agency, and Mr. Mac A. Weaver, as well as the
technical view of Mr. Robert Smith of the Water Quality Office, in Cin-
cinnati, Ohio, is gratefully appreciated.
209
-------
REFERENCES
1. Aguirre, J. M. , andGloyna, E. F. , "Nitrification and Denitrification
in a Model Waste Stabilization Pond," Technical Report EHE-05-6701,
CRWR 19, Environmental Health Engineering Laboratories, Center for
Research in Water Resources, The University of Texas at Austin (1967).
2. Aguirre, J. M. , and Gloyna, E. F., "Influence of the Anaerobic Sludge
Layer Design of Facultative Waste Stabilization Pond Performances,"
Technical Report EHE-71-3, CRWR-77, Environmental Health Engineering
Laboratories, Center for Research in Water Resources, The University of
Texas at Austin (1971).
3. Albertson, J. G., McWhirter, J. R., Robinson, E. K. , and Vahldieck, N.P..
"Investigations on the use of High Purity Oxygen Aeration in the Conven-
tional Activated Sludge Process, " Federal Water Quality Administration,
Department of The Interior, No. 17050 DNW, Washington, D. C. (1970).
4. Anonymous, "Sewage Treatment at Military Installations, " National
Research Council, Washington, D. C. (May, 1964).
5 . Anonymous, Design and Construction of Sanitary and Storm Sewers,
Manual of Practice-9, Water Pollution Control Federation, Washington,
D. C. (1969).
6. Anonymous, Water Pollution Research, Department of Scientific and Industrial
Research (British), Her Majesty's Stationery Office, London (1967).
7 . Anonymous, Water Pollution Research, Department of Scientific and Industrial
Research (British), Her Majesty's Stationery Office, London (1968).
8. Audoin, L. , Barabe, J. P., Brebion, G. , and Huriet, B.,"The Use of
Plastic Material as a Medium for Trickling Filters Treating Domestic
Sewage," Conference Papers 11-16, Fifth International Conference on
Water Pollution Research, San Francisco, California (1970).
9. Ball, J. E., and Humenick, M. J., "The Use of High Purity Oxygen in
Biological Treatment of Municipal Wastewater - A Review and Analysis, "
Paper presented at the Tenth Texas Industrial Water and Waste Conference,
The University of Texas at Austin (1971).
10. Berryhill, D. , Malina, J. F., Jr., and Kayser, R., "Laboratory Studies
of the Performance of the Contact Stabilization Process," Technical
Report EHE-70-16, CRWR-63, Environmental Health Engineering Research
Laboratories, Center for Research in Water Resources, The University of
Texas at Austin (1970).
211
-------
11. Billings, C. H., and Smallhorst, D. F., editors, Manual of Waste-
water Operations, Texas Water Utilities Association, Austin, Texas
(1971).
12. Bishop, Neil E., Malina, J. F., Jr., and Eckenfelder, W. W., Jr.,
"Studies on Mixing and Heat Exchange in Aerated Lagoons," Technical
Report EHE-70-21, CRWR-70, Environmental Health Engineering Research
Laboratories, Center for Research in Water Resources, The University of
Texas at Austin (1971).
13. Boyko, B. I., Mixing Studies on a Full-Scale Aeration Tank, Master's
Thesis, McMaster University (1968).
14. Boyko, B. I., and Murphy, K. L., "Longitudinal Mixing in Spiral Flow
Aeration Tanks, " Conference Paper, Second National Symposium oh
Sanitary Engineering and Research Development and Design, Cornell
University (1969).
15. Bruce, A. M., "Some Factors Affecting the Efficiency of High-Rate
Biological Filters," Conference Papers 11-14, Fifth International Confer
ence on Water Pollution Research, San Francisco (1970).
16. Bruce, A. M., and Merkeus, J. C., "Recent Studies of High-Rate
Biological Filtration," Water Pollution Control (England), pp 113-139
(1970).
17. Calvert, C. K., and Bloodgood, D. E., "Operational Experiments at the
Indianapolis Activated Sludge Plant," Sewage Works Journal, 6, p 1068
(1934).
18 Chiang, W., and Gloyna, E. F., "Biodegradation in Waste Stabilization
Ponds, " Technical Report EHE-70-24, CRWR-73, Environmental Health
Engineering Research Laboratories, Center for Research in Water Resources,
The University of Texas at Austin (1971).
19. Chieu, J. N. , and Gloyna, E. F., "Physical Characteristics of Pilot-Scale
Ponds," Technical Report EHE-70-15, CRWR-61, Environmental Health Engin-
eering Research Laboratories, Center for Research in Water Resources, The
University of Texas at Austin (1970).
20. Chipperfield, P. N. J., "Performance of Plastic Filter Media in Industrial
and Domestic Waste Treatment,Ir Journal Water Pollution Control Federation,
39, pp 1860-1976 (1967).
21. Clark, J. W., Viessman, W., Jr., and Hammer, M. J., Water Supply and
Pollution Control, Second Edition, International Textbook Company (1971).
212
-------
22. Gulp, Russell L. , "The Tahoe Process for Nutrient Removal, " Proceed-
ings Seventh Industrial Water and Waste Conference, Texas Water
Pollution Control Association, The University of Texas at Austin, pp 1-3-26
(1967).
23. Danckwerts, P. V., "Continuous Flow Systems, Distributions of Residence
Times," Chemical Engineering Science, 2, p 1 (1953).
24. Davis, E. M. , and Gloyna, E. F. , "Algal Influence on Dieoff Rates of
Indicator Bacteria," Proceedings of Industrial Waste Conference, Purdue
University (1970).
25. Downing, A. L., Painter, H. A., and Knowles, G. , "Nitrification in the
Activated Sludge Process, " Journal Institute of Sewage Purification, pp
130-151 (1964).
26. Eckenfelder, W. W. , Jr. , "Trickling Filtration Design and Performance,"
Journal Sanitary Engineering Division. Proceeding of the American Society
of Civil Engineers, 87, pp 33-45 (1961).
27. Eckenfelder, W. W. , Jr., Ind. strial Water Pollution Control, McGraw-
Hill Book Co. , New York, New York (1966).
28. Eckenfelder, W. W., Jr. , "A Design Procedure for Biological Nitrification
and Denitrification," Chemical Engineering Process Symposium Series,
pp 230-234 (1967).
29. Eckenfelder, W. W., Jr. , and Ford, D. L. , Water Pollution Control,
Experimental Procedures for Process Design, Pemberton Press, Austin
and New York (1970).
30 Eden, G. E. , Brendish, K. , and Harvey, B. R., "Measurement and
Significance of Retention," Proceedings Institute of Sewage Purification,
pp 513-522 (1964).
31 Espino, E., and Gloyna, E. F. , "Sulfide Production in Waste Stabilization
Ponds," Technical Report EHE-04-6802, CRWR-26, Environmental Health
Engineering Research Laboratories, Center for Research in Water Resources,
The University of Texas at Austin (1968).
32. Fabre, F. . Evaluation of Flexirings as Trickling Filter Medium, Master's
Thesis The University of Texas at Austin (1970).
33. Fair, G. M., Geyer, J. C., and Okun, D. A. , Water and Wastewater
Engineering, Volume 1, John Wiley & Sons, Inc. (1966).
213
-------
34. Fair, G. M., Geyer, J. C., and Okun, D. A., Elements of Water Supply
and Wastewater Disposal, Second Edition, John Wiley & Sons, Inc. (1971).
35. Fairall, J. M., "Correlation of Trickling Filters Data," Journal Sewage
and Industrial Wastes, 28, pp 1069-1074 (1956).
36. Garret, M. T., Jr., "Hydraulic Control of Activated Sludge Soluble Rate,"
Journal Sewage and Industrial Wastes, 30, pp 253-261 (1958).
37. Gloyna, E. F., Waste Stabilization Ponds - Basis for Wastewater Treat-
ment, Monograph Series, Geneva, Switzerland, WD/67.2, World Health
Organization (1971).
38. Gromiec, M. J., andMalina, J. F., Jr., "Verification of Trickling Filter
Models Using Surfpac," Technical Report EHE-70-13, CRWR-60, Environ-
mental Health Engineering Research laboratories, Center for Research in
Water Resources, The University of Texas at Austin (1970).
39. Guarino, C. F., and Carpenter, G. W., "Philadelphia's Plan Toward
Instrumentation and Automation of the Wastewater Treatment Processes,"
Conference Papers 11-24, Fifth International Conference on Water Pollution
Research, San Francisco, California (1970).
40. Halbert, B. E., and Malina, J. F., Jr., "Performance of the Activated
Sludge Process," Technical Report E HE-7 0-12, CRWR-59, Environmental
Health Engineering Research Laboratories, Center for Research in Water
Resources, The University of Texas at Austin (1970).
41. Hasseltine, T. R., "The Activated Sludge Process at Salinas, California
with Particular Reference to Causes and Control of Sludge Bulking,"
Sewage Works Journal, 4, p 461 (1932).
42. Herbert, D., Elsworth, R., and Telling, R., "The Continuous Culture of
Bacteria: A Theoretical and Experimental Study," Journal of General
Microbiology, 14. p 601 (1956).
43. Horning, W. B., Porges, R., Clarke, H. F., and Cooke, W. B.,
"Waste Stabilization Pond Study, Lebanon, Ohio," PHS Pub. No. 999-
WP-16, U. S. Public Health Service, Cincinnati, Ohio (1964).
44. Huang. J. C. and Gloyna. E. F. , "Effects of Toxic Organics on Photo-
synthetic Reoxygenation." Technical Report EHE-07-6701, CRWR-20,
Environmental Health Engineering Research Laboratories, Center for
Research in Water Resources, The University of Texas at Austin (1967).
-------
45. Kurd, C. H. , "Distinctive Characteristics of the Indianapolis Sewage
Treatment Plant, "Sewage Works Journal, 1, p 528 (1929).
46. Kalbskopf, K. H., "European Practices in Sedimentation," Chapter in
Water Quality Improvement by Physical and Chemical Processes, Volume
III, Center for Research in Water Resources, The University of Texas at
Austin, University of Texas Press (1970).
47. Keefer, C. E., and Meisel, J., "Remodeled Trickling Filters Exceeded
Expectations at Baltimore," Water and Sewage Works, pp 277-279 (1952).
48. Kehr, R. W. , "Detention of Liquids Being Mixed in a Continuous Flow
Tank." Sewage Works Journal, 8, pp 915-923 (1936).
49. Kessener, H. J. N. H., "Annual Report, Governmental Institute for
Purification of Wastewaters, 1933." Sewage Works Journal, 7, pp 135-
138 (1935).
50. Knop, E., and Kalbskopf, K. H., "Energy and Hydraulic Tests on
Mechanical Aeration Systems," In Advances in Water Pollution Research,
Pergammon Press, London (1969).
51. Lamb, R., and Owen, J. G. H., "A Suggested Formula for the Process of
Biological Filtration/' Journal Water Pollution Control (England) pp 209-
220 (1970).
52. Levenspiel, D., Chemical Reaction Refinery, John Wiley & Sons, Inc.,
New York (1962).
53. Levenspiel, O. , and Smith, W. K., "Notes on the Diffusion-type Model
for Longitudinal Mixing of Fluids in Flow." Chemical Engineering Science,
j>, pp 257-263 (1957).
54. Malina, J. F., Jr., Eckenfelder, W. W., Jr., Gloyna, E. F., Kayser, R.,
and Drynan, W. R., "Performance of Biological Treatment Processes -
Part of Design Guides for Biological Processes," Technical Report EHE-
71-2, CRWR-76, Environmental Health Engineering Research Laboratories,
Center for Research in Water Resources, The University of Texas at
Austin (1971).
55 McKinney, R. E. , and Beyers, H. H., Jr. , "Evaluation of Two Aerated
Lagoons," Proceedings American Society of Civil Engineers, Journal
Sanitary Engineering Division, 91, SAE6 (1965).
215
-------
56. Marais, G. v. R. , and Shaw, V. A. , "A Rational Theory for Design of
Sewage Stabilization Ponds in Central and South Africa," Transactions
South African Institute of Civil Engineers, 3, p 205 (1961).
57. Meiring, P. G. J. , Drews, R. J. L. C. , van Eck, H. , and Stander, G. J. ,
"A Guide to the use of Pond Systems in South Africa for the Purification of
Raw and Partially Treated Sewage," Special Report, National Institute for
Water Research Council for Scientific and Industrial Research, Pretoria,
South Africa (1968).
58. Merrel, J. C., Jr., et al., "The Santee Recreation Project," Final Report,
Federal Water Pollution Control Administration, Cincinnati, Ohio (1967).
59. Miyauchi, T., "Residence Time Curves," Chemical Engineering (Japan),
11(1953).
60. Monod, J. , "The Growth of Bacterial Cultures," Annual Review of Micro-
biology,^, p 371 (1949).
61. Myatt, J. F. , and Malina, J. F., Jr., "Performance of the Contact
Stabilization Process," Technical Report EHE-70-10, CRWR-58, Environ-
mental Health Engineering Research Laboratories, Center for Research in
Water Resources, The University of Texas at Austin (1970).
62. Oswald, W. J., "Advances in Anaerobic Pond System Design," Chapter
in Advances in Water Quality Improvement, Volume I, Center for Research
in Water Resources, The University of Texas at Austin, The University
of Texas Press (1968).
63. Oswald, W. J., "Metropolitan Wastes and Algal Nutrition," Transaction of
1960 Seminar on Algae and Metropolitan Wastes, U. S. Public Health Service,
Cincinnati, Ohio, pp 88-95 (1961).
64. Owen, R., "Removal of Phosphorus from Sewage Plant Effluent with Lime,"
Sewage and Industrial Wastes Journal, 25, No. 5, p 548 (1953).
65. Parker, C. D., Jones, H. L., and Greene, N. C. , "Performance of Large
Sewage Lagoons at Melbourne, Austrailia," Sewage and Industrial Wastes
Journal. 31, No. 2, p 133 (1959).
66. Peumrau, A. , et al., "Evaluation of Aerated Lagoons in Metropolitan
Winnipeg," Report Metropolitan Corporation of Greater Winnipeg (1970).
67. Pflanz, P., "Performance of (Activated Sludge) Secondary Sedimentation
Basins," Advances in Water Pollution Research, Pergammon Press, London
(1969).
216
-------
68. Rincke, G., "Ncuere Geischtspunkte zur Abwasscrreiniguncj mil Troplkor-
pern," Das Gas und Wasserfach, 108, No. 24, pp 667-673 (1967).
69. Rincke, G. , and Wolters , N. , "Technology of Plastic Medium Trickling
Filters," Conference Papers 11-15 , Fifth International Conference on Water
Pollution Research, San Francisco, California (1970).
70. Sawyer, C. N. , "New Concepts in Aerated Lagoon Design and Operation,"
Chapter in Advances in Water Quality Improvement, Volume 1, Center for
Research in Water Resources, The University of Texas at Austin, University
of Texas Press (1968) .
71. Schmidt, R. K., and Eckenfelder, W. W. , Jr. , "Parameter Responses on
Influent Suspended Solids in the Activated Sludge Treatment Process,"
Technical Report EHE-70-07, CRWR-56, Environmental Health Engineering
Research Laboratories, Center for Research in Water Resources, The Uni-
versity of Texas at Austin (1970) .
72. Schulze, K. L. , "The Activated Sludge Process as a Continuous Flow
Culture," Water and Sewage Works, 112 , No. 12, pp 526-538 (1964).
73. Smith, R. , and Eilers, R. G. , "Generalized Computer Model for Steady-
State Performance of the Activated Sludge Process," Federal Water Quality
Administration, Cincinnati, Ohio (1969).
74. Steel, E. W. , Water Supply and Sewerage, Fourth Edition, McGraw-Hill
Book Company, Inc. (1960).
75 . Suwannakarn , V. , and Gloyna , E. F. , Temperature Effects on Waste
Stabilization Pond Treatment, Technical Report, Environmental Health
Engineering Laboratories, The University of Texas at Austin (1963) .
76. Teissier, G. , "Les Lois Quantitatives de la Croissance," Annales de Physi-
ologie et de Physicochimie Biologigue, 12, p 527 (1936) .
77. Thomas, H. A., and McKee, J. E., "Longitudinal Mixing in Aeration Tanks,"
Sewage Works Journal, 16, No. 1, pp 42-45 (1944) .
78. Timpany, P. L., Variations of Axial Mixing in an Aeration Tank, Master's
Thesis, McMaster University (1966) .
79. Timpany, P. L., and Murphy, K. L., "Design and Analysis of Mixing for
an Aerobic Tank, Proceedings American Society of Civil Engineers, Journal
Sanitary Engineering Division, 93, SA5 pp 1-15 (1967) .
217
-------
«0. Van der Laan, E. T., "Notes on Diffusion Type Model for Longitudinal
Mixing in Flow," Chemical Engineering Science, 7, pp 187-191 (1958) .
81. Von der Emde, W., and Kayser, R., "Oxygen Transfer of Aerosol Aerobic
Systems Under Process Conditions," Unpublished Report, Vienna, Austria
(1969).
82. White, J. E., "Current Design Criteria for Anaerobic Lagoons," Proceedings
Second International Symposium for Waste Treatment Lagoons, Kansas City,
Mo. (1970).
83. Williamson. Paul E., Wastewater Treatment Facilities in Small Texas
Communities, Master's Thesis, The University of Texas at Austin (1971).
84. Wuhrmann, K., "Grundlagen fur die Dimensionierung der Beluftung bei
Belebtschlammanalagen," Schweizerische Zeitschrift fur Hydrologie^ 26,
pp 310-337 (1964).
85. Wuhrmann, K., "Hauptwirkungen und Wechsel Wirkungen einiger Betrieb-
sparameter im Belebtschlammsystem," Schweizerische Zeitschrift fur
Hydrologie, 26, pp 218-270 (1964).
86. Wuhrman, K., "Stickstoff-und Phosphorelimination," Schweizerische
Zeitschrift fur Hydrologie, 26, p 520 (1964).
87. Wuhrmann, K., "Research Developments in Regard to Concept and Base
Values of Activated Sludge System," In Advances in Water Quality Improve-
ment, Volume I, Center for Research in Water Resources, The University
of Texas at Austin, The University of Texas Press (1968).
88. Yousef, Y. A., and Malina, J. F. , Jr. , "The Fate of Coliform Organisms in
Waste Stabilization Ponds," Journal Water Pollution Control Federation, 36,
pp 1432-1442 (1964).
218
-------
SYMBOLS AND ABBREVIATIONS
a fraction of substrate converted to bacterial mass
a' oxygen consumed per unit of substrate removed
b substrate removed per unit of time, by endogenous respiration
b1 endogenous respiration rate
BOD biochemical oxygen demand
BODF filtered (assumed to be soluble) BOD5
BODT total BOD
°C degrees centigrade
c oxygen concentration
COD chemical oxygen demand
CODF filtered (assumption: soluble) COD
CODT total COD
°F degrees Fahrenheit
1C inorganic carbon
k reaction rate coefficient, related to MLSS
K reaction rate coefficient, MLSS included
k reaction rate coefficient, related to active MLSS
a
K reaction rate coefficient, active MLSS included
a
Kh heat transfer coefficient
k substrate-concentration at which the reaction-rate equals one-half
m k
max
k k divided by a
R max
k reaction rate coefficient, related to MLVSS
v
219
-------
k maximum removal rate, Monod - Michaelis - Mcntcn equation
max
K reaction rate coefficient, MLVSS included
v
K equals K, K
v a
K— reaction rate coefficient, at temperature T, °C
K reaction rate coefficient, at 20 C
£* U
MLSS mixed liquor suspended solids
MLVSS mixed liquor volatile suspended solids
NH_ ammonia
NO9 « nitrite + nitrate
£• T" O
OUR oxygen uptake rate
0 flow
R oxygen used, per unit weight of mixed liquor and per unit of time
S soluble substrate concentration, tank
S soluble substrate concentration, influent
s
ST total substrate concentration, supernatant after settling
e
ST total substrate concentration, influent
o
AS soluble substrate removed, S0 - S
SS suspended solids
t hydraulic detention time, t = V/Q
T temperature
TKN totak Kjelda hi nitrogen
TC total carbon
TOG total organic carbon
220
-------
TOCF filtered (assumption: soluble) TOG
TOCT total TOG
TPO. total phosphate
t variable time
v
X MLSS
X active MLSS
a
XQ SS in influent wastewater
X active SS in raw wastewater
oa
X volatile MLSS
v
V volume of reactor
9 tempera ture-correction-coefficient
When other symbols and abbreviations are used in the text, they are
explained.
221
-------
1
Accession Number
2
Subject Field & Group
0 5 D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
City of Austin, Texas
Center for Research in Water Resources
The University of Texas at Austin
Title
Design Guides for Biological Wastewater Treatment Processes
10
Authors)
Joseph F. Malina, Jr., P.E.,
Ph.D.
Rolf Kayser
W. W. Eckenfelder, Jr., P.E.
Earnest F. Gloyna, P.E., Dr.
Eng.
W. R. Drvnan. P.F.. . Ph.D.
16
21
Project Designation
EPA Project #11010 ESO
Note
22
Citation
23
Descriptors (Starred First)
'Wastewater Treatment*
Water Pollution Control*
Pollution Abatement*
Biological Treatment*
Sewage Treatment*
Biochemical Oxygen Demand
Organic Matter
Chemical Oxygen Demand
Oxygen
25
Identifiers (Starred First)
.Sewage Treatment Plant Design*
Biological Processes Design*
Activated Sludge*
Aerated Lagoons*
Stabilization Ponds*
Trickling Filters*
Hydraulic Loadings
Organic Loadings
Design Examples
27
Abstract
This report provides a set of guidelines for the design of biological processes for the
treatment of municipal wastewater. The equations and factors which must be considered
in the design of the activated sludge system, the contact stabilization system, trickling
filter plants, aerated lagoons, and waste stabilization ponds are identified. The appli-
cability and limitations of each system and mathematical model of each process are es-
tablished. Operating data from the Govalle Wastewater Treatment Plant, the Williamson
Treatment Plant, and the Walnut Creek Plant operated by the City of Austin, Texas and
other operating data from the treatment plants where sufficient applicable data were re-
corded were used to develop rate constants and other coefficients required for application
of the mathematical models and other design of treatment plants. The need for waste
characterization including variations in quantities in flow and composition of flow are
emphasized. The significant design considerations are discussed, design procedures are
outlined and design calculations are developed. This report contains 211 pages, 81
figures, ten tables, and 88 references.
Malina, Jr. , P.E,
Institution _
The University of Texas at Austin
WR;102 (REV. JULY 1969)
VYRSIC
Ph.D.
SEND TO- WATER RESOURCES SCIENTIFIC INFORI.
US DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C 2O240
4U.S. GOVERNMENT PRINTING OFFICE: 1972 484-483/72 1-3
------- |