jTTW
L/A-625/1-87-001
Center for Environmental
Research Information
Cincinnati OH 45268
Technology Transfer
Research Laboratory
ti OH 45268
EPA 625 1-87 001
&EPA Design Manual
Phosphorus Removal
-------
EPA/625/1-87/001
September 1987
Design Manual
Phosphorus Removal
U.S. Environmental Protection Agency
Office of Research and Development
Center for Environmental Research Information
Cincinnati, OH 45268
Water Engineering Research Laboratory
Cincinnati, OH 45268
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, IZth Floor
Chicago. IL 60604-3590
-------
Notice
This document has been reviewed in accordance with the U.S. Environmental Protection
Agency's peer and administrative review policies and approved for publication. Mention of trade
names or commercial products does not constitute endorsement or recommendation for use.
This document is not intended to be a guidance or support document for a specific regulatory
program. Guidance documents are available from EPA and must be consulted to address
specific regulatory issues.
-'C
-------
Contents
Chapter Page
1 INTRODUCTION 1
1.1 Purpose 1
1.2 Scope 1
1.3 Using the Manual 1
1.4 Reference 2
2 SELECTING A PHOSPHORUS REMOVAL STRATEGY 3
2.1 Description of Approach 3
2.2 Information and Monitoring Data Required 3
2.2.1 New System Data Requirements 3
2.2.2 Existing System Data Requirements 5
2.3 Possible Phosphorus Removal Alternatives 5
2.3.1 Chemical Addition Alternatives 5
2.3.2 Biological Phosphorus Removal Alternatives 6
2.4 Phosphorus Removal System Selection Strategy 8
3 PHOSPHORUS REMOVAL BY BIOLOGICAL PROCESSES 15
3.1 Introduction and Theory 15
3.1.1 Biological Phosphorus Removal Mechanism 16
3.2 Applications 18
3.2.1 Process Descriptions 19
3.3 Performance 25
3.3.1 Phostrip Performance 25
3.3.2 Modified Bardenpho Process Performance 26
3.3.3 A/O Process Performance 27
3.3.4 Operationally Modified Activated Sludge Process Performance 29
3.3.5 Factors Affecting Performance 30
3.4 Equipment Requirements 36
3.5 Design Methodology 36
3.5.1 Phostrip Process 36
3.5.2 Mainstream Biological Phosphorus Removal Processes 38
3.6 Process Modifications to Improve Performance 43
3.7 Retrofit Considerations 44
3.8 Case Histories 45
3.8.1 Phostrip Process - Little Patuxent, Maryland 45
3.8.2 Modified Bardenpho Process - Kelowna, Canada 46
3.8.3 A/O Process - Pontiac, Michigan 47
3.9 Costs 47
3.10 References 50
in
-------
Contents (continued)
x Chapter Page
4 PHOSPHORUS REMOVAL BY CHEMICAL ADDITION 55
4.1 Introduction and Theory 55
4.1.1 Aluminum Compounds 55
4.1.2 Iron Compounds 56
4.2 Application Points 57
4.2.1 Mineral Addition Before Primary Clarification 58
4.2.2 Mineral Addition to Secondary Processes 59
4.2.3 Mineral Addition at Multiple Points 60
4.3 Performance 60
4.4 Equipment Requirements 60
4.4.1 Chemical Handling and Storage 60
4.4.2 Dry Chemical Feeding and Dissolving 64
4.4.3 Liquid Chemical and Solution Feeding 65
4.4.4 Chemical Dosage Control 66
4.4.5 Chemical Mixing and Flocculation 68
4.4.6 Clarification 69
4.5 Design Methodology 69
4.5.1 Wastewater Characterization 69
4.5.2 Determination of Chemical Doses 70
4.5.3 Selection of Chemicals 70
4.5.4 Evaluation of Existing Unit Processes 70
4.5.5 Conduct of Full-Scale Trials 71
4.5.6 Design of Chemical Handling System 71
4.5.7 Design of Liquid Processes 73
4.5.8 Design of Sludge Handling System 74
4.5.9 Design Example 75
4.6 Retrofit Considerations 77
4.7 Case Histories 77
4.7.1 Orillia, Ontario 77
4.7.2 Elizabethtown, Pennsylvania 78
4.7.3 Little Hunting Creek, Virginia 79
4.8 Costs 80
4.9 References 80
5 SLUDGE HANDLING 83
5.1 Introduction 83
5.2 Current Practice for Handling Chemical Sludges 83
5.2.1 Introduction 83
5.2.2 Points of Chemical Addition and Methods of Combining Sludges 83
5.2.3 Sludge Generation Rates and Characteristics 84
5.2.4 Prevalence of Various Sludge Treatment and Disposal Options 87
5.3 Sludge Derived from Addition of Aluminum Salts 87
5.3.1 Sludge Characteristics 87
5.3.2 Sludge Generation Rates 88
5.3.3 Thickening 92
5.3.4 Stabilization 93
5.3.5 Conditioning 95
5.3.6 Dewatenng 95
5.3.7 Incineration 99
5.3.8 Disposal 99
IV
-------
Contents (continued)
Chapter Page
5.4 Sludge Derived from Addition of Iron Salts 100
5.4.1 Sludge Characteristics 100
5.4.2 Sludge Generation Rates 100
5.4.3 Thickening 101
5.4.4 Stabilization 103
5.4.5 Conditioning 104
5.4.6 Dewatering 104
5.4.7 Incineration 107
5.4.8 Disposal 107
5.5 Sludges Derived From Biological Phosphorus Removal Processes 108
5.5.1 Characteristics 108
5.5.2 Sludge Generation Rates 108
5.5.3 Thickening 108
5.5.4 Stabilization 108
5.5.5 Conditioning 109
5.5.6 Dewatering 109
5.5.7 Disposal 109
5.6 Sludges Derived from Addition of Lime 109
5.7 Case Histories 110
5.7.1 Baltimore, Maryland 110
5.7.2 Lorton, Virginia 110
5.8 Costs 112
5.9 References 112
-------
Figures
Number Page
2-1 Phosphorus Removal System Selection Strategy 9
3-1 Biological phosphorus and BOD removal due to anaerobic-aerobic
contacting 16
3-2 Acetate assimilation and phosphorus release vs. anaerobic time 17
3-3 Schematic of biological phosphorus removal mechanism 19
3-4 Commercial biological phosphorus removal processes - Phostrip,
Modified Bardenpho, A/0 20
3-5 Biological phosphorus removal - Sequencing Batch Reactor 23
3-6 UCT process flow schematics 24
3-7 Operationally modified activated sludge system for
biological phosphorus removal 24
3-8 Combination biological phosphorus removal system 24
3-9 Maximum effluent soluble P concentration for total effluent P < 1.0 mg/l 31
3-10 Effluent soluble P concentration vs. influent TBOD:TP ratio 32
3-11 Reported effects of pH on biological phosphorus removal 35
3-12 TBOD:TP removal vs. solids retention time (SRT) 40
3-13 Pre-denitrification and post-denitrification schemes in biological
phosphorus removal systems 41
3-14 Specific denitrification rate 42
3-15 Primary sludge fermentation design schemes 44
4-1 Alternative schemes for mineral addition for phosphorus removal 58
4-2 Impact of point of addition on effectiveness of phosphorus removal
using aluminum 59
4-3 Typical dry chemical feed system 64
4-4 Liquid chemical feed alternatives for elevated storage 67
4-5 Liquid chemical feed alternatives for ground storage 67
4-6 Typical flash mix tank 69
4-7 Typical mechanically mixed flocculating clarifier 70
5-1 Volumetric sludge production from EPA survey 85
5-2 Mass sludge production from EPA survey 86
5-3 Range of thickener operating periods for ferric-primary sludge 102
5-4 Range of thickener operating periods for alum-primary sludge 102
5-5 Comparison of sludge quantities, conditioning costs, and
hauling costs before and after phosphorus removal; Baltimore, MD 111
VI
-------
Tables
Number Page
2-1 Information Required for Discharge Limits 4
2-2 Information Required for Influent Wastewater Characteristics 4
2-3 Advantages and Disadvantages of Metal Salt Addition for Phosphorus
Removal 7
2-4 Advantages and Disadvantages of Lime Addition for Phosphorus Removal 7
2-5 Advantages and Disadvantages of Biological Phosphorus Removal Processes ... 8
2-6 Application Criteria Matrix - New or Existing Facility; Type of Nutrient
Removed 9
2-7 Application Criteria Matrix - Ability of Process to Meet Effluent Phosphorus
Limitations 10
2-8 Application Criteria Matrix - Ability of Process to Remove NH4 or TN 10
2-9 Application Criteria Matrix - Effect of Plant Size on Process Applicability 12
2-10 Application Criteria Matrix - Effect of TBOD:TP Ratio <20 on Process
Applicability 12
2-11 Application Criteria Matrix - Effect of High Wastewater Alkalinity
on Process Applicability 13
2-12 Application Criteria Matrix - Effect of Sludge Production on Process
Applicability 13
2-13 Application Criteria Matrix - Effect of O&M Requirements on Process
Applicability 14
2-14 Application Criteria Matrix - Effect of Short-Term Life on Process
Applicability 14
3-1 Molar Ratios of Ions Co-Transported with Phosphorus 18
3-2 Typical Operating Conditions for Biological Phosphorus Removal Processes .... 21
3-3 SBR Operating Sequence - Biological Phosphorus Removal 23
3-4 Basic Design Information for Full-Scale Phostrip Plants 26
3-5 Performance Data Summary for Full-scale Phostrip Plants 26
3-6 Modified Bardenpho Process Full-scale Plant Design Summary 27
3-7 Palmetto, FL Modified Bardenpho Performance
(April 1981 - March 1982Monthly Averages) 28
3-8 Kelowna, Canada Modified Bardenpho Process Performance Results
of Cumulative Frequency Plot of Data 28
3-9 Summary of A/O Process Average Effluent Quality, Largo, FL 29
3-10 Pontiac, Ml A/O System Operating Conditions 29
3-11 Full-scale A/O Process Performance, Pontiac, Ml 29
3-12 Operating Conditions for Operationally Modified Activated Sludge Systems 30
3-13 Average Performance of Operationally Modified Activated Sludge Systems 30
3-14 Approximated Mass of Phosphorus Storage Compounds 39
3-15 Summary of Phostrip Process Treatment Performance for Little
Patuxent Plant - April 1985 46
3-16 Cost comparison - Case 1: Phosphorus removal (effluent TP = 1 mg/l) 49
3-17 Comparison - Case 2: Phosphorus removal (effluent TP = 2 mg/l) 49
VII
-------
Tables (continued)
Number Page
3-18 Cost comparison - Case 3: Phosphorus removal plus nitrification
(effluent TP = 2 mg/l; NH4-N = 1 mg/l) 49
3-19 Cost comparison - Case 4: Phosphorus removal plus nitrification
(effluent TP = 2 mg/l; TN = 3 mg/l) 49
4-1 Characteristics of Aluminum and Iron Salts 57
4-2 Distribution of Selected Phosphorus Removal Facilities by Point of Chemical
Addition and Cation Used 59
4-3 Potential Effectiveness of Primary and Secondary Treatment With and
Without Mineral Addition for Phosphorus Removal 59
4-4 Performance of Facilities Using Mineral Salts for Phosphorus Removal 61
4-5 Chemical Dosage Summary from Ontario Treatability Studies 63
4-6 Types of Chemical Feeders 66
4-7 Values of Ky for Determining Impeller Power Requirements 73
4-8 Recommended Overflow Rates for Conventional Clarifiers
Receiving Wastewater Coagulated with Mineral Salts 75
4-9 Summary of Plant Data During Single-Point and Dual-Point Alum Addition ... 78
4-10 Summary of Performance at Elizabethtown, PA 79
4-11 Summary of Performance at Little Hunting Creek, VA 80
4-12 Cost Estimates for Phosphorus Removal by Mineral Addition 80
5-1 Combination of Chemical Sludges with Other Sludges for Processing as
Practiced by Plants in EPA Survey 84
5-2 Combination of Chemical Sludges with Other Sludges for Processing as
Practiced by Plants in EPA Survey 88
5-3 Prevalence of Treatment and Disposal Processes for Chemical Sludges
Among Plants in EPA Survey 89
5-4 Typical Gravity Thickener Design Criteria 92
5-5 Summary of Vacuum Filter Performance for Alum-Primary Sludge 97
5-6 Design vacuum filtration rates for conventional sludges 97
5-7 Component vacuum filtration characteristics 97
5-8 Design Factors for Vacuum Filtration of Conventional Plus Aluminum
Sludges 98
5-9 Potential Problems with Chemical Sludge Incineration 99
5-10 Typical Gravity Thickener Design Criteria 102
5-11 Effect of Phosphorus Removal on Gravity Thickening Properties of
Alum-Primary and Ferric-Primary Sludge 102
5-12 Performance of Flotation Thickeners for Treating Iron Sludges 103
5-13 Summary of Vacuum Filter Performance for Iron-Primary Sludge 105
5-14 Design Factors for Vacuum Filtration of Conventional Plus Iron Sludges 106
VIII
-------
Acknowledgments
Many individuals contributed to the preparation and review of this handbook. Contract
administration was provided by the Center for Environmental Research Information, Cincinnati,
Ohio.
Major Authors:
Robert P. G. Bowker - J. M. Smith & Associates, PSC, Cincinnati, Ohio
H. David Stensel - University of Washington, Seattle, Washington
Contributing Authors:
George L. Hartmann - J. M. Smith & Associates, PSC, Cincinnati, Ohio
John M. Smith - J. M. Smith & Associates, PSC, Cincinnati, Ohio
Reviewers:
Walter Gilbert - EPA-OMPC, Washington, DC
Wen H. Huang - EPA-OMPC, Washington, DC
James Wheeler - EPA-OMPC, Washington, DC
Richard C. Brenner - EPA-WERL, Cincinnati, Ohio
Francis Evans III - EPA-WERL, Cincinnati, Ohio
Joseph B. Farrell - EPA-WERL, Cincinnati, Ohio
James F. Kreissl - EPA-WERL, Cincinnati, Ohio
James A. Heidman - EPA-WERL, Cincinnati, Ohio
Sherwood Reed - USACOE, Hanover, New Hampshire
Glen T. Daigger - CH2M/HMI, Denver, Colorado
Contract Project Officer:
Denis J. Lussier - EPA-CERI, Cincinnati, Ohio
IX
-------
Chapter 1
Introduction
1.1 Purpose
The Environmental Protection Agency has sponsored
many research and demonstration studies at several
cities in the past few years to advance the knowledge
of phosphorus removal. Local and state governments
and private industries have also contributed to this
work. This manual is intended to summarize process
design information for the best developed
phosphorus-removal methods that have resulted
from this governmental and private effort.
1.2 Scope
The sources and quantities of phosphorus in
domestic wastewaters vary significantly. Industrial
contribution, non-point source runoff, the use or
nonuse of phosphate-bearing detergents, and other
factors make generalization of expected wastewater
phosphorus concentration impossible. The manual
user should ascertain the phosphorus concentrations
(actual or expected) for the specific wastewater in
question.
This manual discusses several proven phosphorus-
removal methods, including phosphorus removal
obtainable through biological activity as well as
chemical precipitation techniques.
Biological phosphorus removal was not included in
the previous version of this manual and represents a
major addition.
Appropriate chemistry for phosphorus removal by
chemical addition is presented where appropriate for
illustrative purposes.
The use of lime as a chemical precipitant for
phosphorus removal, which received major treatment
in the previous version, is not covered in this manual
due to its loss of popularity as a phosphorus-
removal technique. The reasons for current infrequent
use of lime are discussed in Chapter 4. The user can
refer to several other documents for detailed
information and design criteria for lime addition,
including the previous version of this manual (1).
Treatment methods in which phosphorus removal
occurs, but is not a principal objective, are also
omitted. The latter group of processes includes such
technologies as ion exchange, reverse osmosis and
other demineralization treatments which at present
are more closely associated with wastewater
renovation and reuse than with pollution control.
These will be included in updated versions of the
manual when appropriate.
The information included was obtained from the
available literature, progress reports from
demonstration studies, and private communications
with investigators actively working in the field. Design
guidelines have been developed from these sources.
The information contained in this manual is oriented
toward design methods and operating procedures for
removal of phosphorus from wastewater.
Cost information from actual phosphorus-removing
installations is presented when available. Planning
level cost estimates are also included.
1.3 Using the Manual
Chapter 2 presents a recommended approach to
selecting a phosphorus removal strategy. This
approach identifies the required effluent phosphorus
limits and screens potential phosphorus-removal
techniques to identify those processes capable of
meeting the specified requirements.
The screening process is a step-by-step procedure
that identifies the information required to make the
engineering judgements necessary at each step of
the process. After applying the screening
methodology, the manual user can go to the specific
chapter in the manual dealing with the potential
phosphorus-removal processes available for his
specific case:
Chapter 3 - biological phosphorus removal.
Chapter 4 - phosphorus removal by mineral
addition.
-------
Chapter 5 discusses all aspects of sludge handling
associated with sludges generated from
phosphorus-removing facilities.
1.4 Reference
When an NTIS number is cited in a reference, that
reference is available from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
(703) 487-4650
1. Process Design Manual for Phosphorus
Removal. EPA 625/1-76-001 a, NTIS No. PB-
259150, U.S. Environmental Protection Agency,
Center for Environmental Research Information.
Cincinnati, OH, 1976.
-------
Chapter 2
Selecting a Phosphorus Removal Strategy
2.1 Description of Approach
The approach described herein for selecting a
phosphorus removal system identifies the required
effluent phosphorus limits, and then screens
phosphorus removal processes to identify those
processes capable of meeting specified requirements.
All alternative phosphorus removal technologies are
considered initially, but non-applicable technologies
are eliminated through a sequential selective
screening process.
The selective screening methodology is a step-by-
step procedure that identifies information required to
make the necessary engineering judgements at each
step of the process. Brief descriptions of the various
chemical and biological phosphorus removal
processes and the advantages and disadvantages of
each process have been provided. The effectiveness
of the selective screening process is dependent upon
the amount and level of detail of the initial information
used.
For each specific phosphorus removal alternative
considered, the sections of the manual dealing with
that process should be read carefully before final
selection of a phosphorus removal system. Other
sources of background information include
manufacturers or proprietors of phosphorus removal
technologies, and site visits to operating phosphorus
removal plants. The process should not be
considered completely rigorous, since some
subjective judgements are required. It is intended as
a comprehensive overall guide to the selection of an
appropriate phosphorus removal process.
The selection procedure must consider all aspects of
the phosphorus removal process including its impact
on plant performance, operations and maintenance.
Important factors are: a) degree of phosphorus
removal required, b) size of plant, c) impact on sludge
handling, d) permanent or temporary nature of
phosphorus removal requirement, e) total cost and f)
impact on operation and maintenance.
The system screening process is presented in
Section 2.4. The basic information needed to apply
the screening procedure is discussed in Section 2.2,
and a brief description of the state-of-the-art
phosphorus removal technologies is presented in
Section 2.3.
2.2 Information and Monitoring Data
Required
The information and monitoring data required for
selecting a phosphorus removal system are described
below. Information required for new plants is
somewhat different than for existing facilities. Most
phosphorus removal systems will be retrofitted to
existing systems due to imposition of new permit
requirements. In either case, knowledge of what
information is required to make a cost effective
selection is critical in being able to utilize the
screening methodology presented later.
2.2.7 New System Data Requirements
2.2.1.1 Effluent Discharge Requirements
The first step taken in evaluating alternative
phosphorus removal processes for a new facility is to
determine the phosphorus discharge limitations.
These limits should be determined in as detailed a
manner as possible to define the daily, weekly,
monthly and possibly seasonal phosphorus limits.
Determination of whether an existing facility will be
phased out due to regionalization or plant
consolidation should also be ascertained to ensure
that the time frame used in the cost analysis reflects
the expected plant lifetime.
Phosphorus limits may be set as a minimum percent
removal, as a specific effluent concentration or on a
mass per day basis. In addition to the phosphorus
discharge limitations, the permit limits for BODs, TSS,
pH, NH4-N or total N must be known. The necessity
to remove ammonia and/or total-nitrogen from the
effluent can have a significant impact on the selection
of the phosphorus removal process. The information
normally required is shown in Table 2-1.
2.2.1.2 Wastewater Characteristics
Once effluent limitations have been determined, the
next step is to develop the information on wastewater
characteristics which may affect the choice of
phosphorus removal alternatives. For a new facility,
-------
Table 2-1. Information Required for Discharge Limits
Parameter Weekly Monthly Seasonal
Total P, mg/l
Total P, kg/d
Total P, % removal
BOD5, mg/l
TSS, mg/l
NH4-N, mg/l
Total N, mg/l
pH, units
Table 2-2. Information Required for Influent Wastewater
Characteristics
Parameter Daily1 Weekly Monthly
Flow
Average, m3/d
Maximum, m3/d
Peak/Average Ratio
Total P, average, mg/l
Soluble P, average, mg/l
BOD5, average, mg/l
SBOO, average, mg/l
TSS, average, mg/l
NH4-N, average, mg/l2
Total N, average, mg/l2
pH, units
Temperature
Alkalinity, average, mg/l
TBOD:Total P ratio
1 Seasonal column should be added for biological phosphorus
removal facilities.
2 Where effluent discharge permit limits Total N or NH4-N.
the most important wastewater characteristics are
listed in Table 2-2. In addition, any significant
contribution from individual sources should be
characterized. The effect of these variables on
alternative phosphorus removal processes is
summarized in Section 2.4. Detailed explanations are
provided in Chapters 3 and 4 for the specific
processes.
The best data on anticipated plant flows and
wastewater characteristics are from plant influent
monitoring data. If data on nutrients, alkalinity, etc.
are not available, it is recommended that data from
the literature be used for facilities of similiar size and
service characteristics. Actual characterization data
from the wastewaters to be treated are much
preferred, and every effort should be expended to
collect representative information. The importance of
adequate wastewater characterization cannot be
over-emphasized. The cost of sampling and analysis
required for proper wastewater characterization is
almost always offset by more efficient system design
and lower total costs for phosphorus removal.
Alkalinity and pH data are site specific and must be
determined for each facility. Where plant influent data
are not available, alkalinity can be best obtained from
the water supply sources serving the contributing
wastewater generators.
2.2.1.3 Other Information Required
Other information that may have a significant
influence on the phosphorus removal alternatives are:
1. Sludge disposal alternatives.
2. Service area characteristics including industrial
contributions.
3. Plant size and location; available land.
4. Facility design lifetime.
5. Local availability and cost of chemicals.
Smaller plants in more rural settings generally offer a
wider range of alternatives because of the reduced
constraints of sludge handling. Chemical alternatives
generally require less space but exhibit higher
operation and maintenance costs. Biological
alternatives require more careful analysis of total
treatment requirements including the proper selection
of a basic BOD removal process.
Sludge generation can be a significant factor
influencing the phosphorus removal alternative
selected. Many communities are limited in their
choice of sludge disposal options due to local or
state regulations, air quality requirements or biases on
the part of local decision makers and/or local
concerned citizens. Because of the importance of
sludge handling cost as a part of the total treatment
costs, new phosphorus removal projects must
consider sludge handling as an intergal part of the
total system analysis.
The typical planning period for new wastewater
treatment facilities is 20 years. However, phosphorus
removal standards may be established for different
time frames. Chemical alternatives may be more cost
effective for shorter term design periods, while
biological alternatives with higher capital costs may
be more appropriate for longer term projects.
Wastewater characteristics are determined to a large
degree by the composition of the contributing sewer
dischargers. The breakdown of the system
wastewater contributors into residential, commercial
and industrial discharges can be made from
community records and/or planning studies. Where
significant industrial contributors are identified, the
impact of their wastes on the wastewater composition
-------
should be reviewed for its effect on the various
phosphorus removal alternatives selected for
consideration.
2.2.2 Existing System Data Requirements
Incorporation of phosphorus removal technology into
existing facilities presents a somewhat more
complicated initial analysis than that required for new
systems. Existing systems may offer opportunities for
cost effective incorporation of some of the available
phosphorus removal technologies, but at the same
time may present severe constraints to the use of
others. The more important considerations for existing
facilities are:
1. Whether or not the existing facility is meeting or
can be easily upgraded to meet the non-
phosphorus effluent limits.
2. Whether or not the existing hydraulic capacity is
adequate for the proposed design flows.
3. Whether or not the age and condition of the
existing plant justifies the inclusion of a new
phosphorus removal technology.
4. Whether or not the existing facility must be
upgraded (to improve effluent quality) or expanded,
or both.
5. Whether or not the basic form of biological
treatment, i.e., fixed film or suspended growth, is
compatible with phosphorus removal technologies
capable of meeting the required effluent
phosphorus limitations.
Once these basic questions have been addressed,
the analysis may proceed in the same manner
described earlier for new facilities.
1. Type, capacity and efficiency of existing unit
operations in service:
a. Liquid stream
hydraulic capacity
biological capacity
aeration capacity
modifications planned or underway
b. Solids stream
thickening and dewatering capacity
stabilization capacity and efficiency
ultimate disposal capacity
compatibility of sludge handling with applicable
phosphorus removal technologies
2. Unit operation performance for specific parameters,
including:
Total P
TSS
BOD5
Total N
NH4-N
3. Unit operations known to be obsolete or to require
replacement.
4. Unused or abandoned plant capacity that could be
brought into service.
5. Discharge points of recycle streams from existing
sludge handling systems; and loadings associated
with these streams for the above parameters.
6. Sludge disposal options available with existing
sludge handling system; loading capacities and
efficiency of all sludge handling processes.
Capacity analysis of solids handling systems
should note the percent of time units are in service
and the percent of existing operating capacity that
is being used.
Personnel requirements for operation of existing
equipment should be developed so that effects of
adding new operations or expanding operating hours
of existing equipment can be used in evaluating the
additional costs of various phosphorus removal
alternatives. The level of sophistication of operations
for the applicable phosphorus removal technologies
should be compared to the existing plant operation
requirements. For example, if existing plant
performance is inadequate due to deficient
operations, the incorporation of a biological
phosphorus removal technology requires substantial
upgrading of operator staffing and competence level.
2.3 Possible Phosphorus Removal
Alternatives
Detailed descriptions of phosphorus removal
technologies and the attendant sludge handling
alternatives are presented in Chapters 3 through 5.
This section briefly summarizes the various
phosphorus removal technologies. Simplified process
flow schematics, along with expected performance
ranges and the major advantages and disadvantages
of each process, are discussed below.
2.3.7 Chemical Addition Alternatives
2.3.1.1 Metal Salt Addition (Aluminum and Iron)
Metal salts of aluminum and iron added to wastewater
react with phosphates to form insoluble aluminum or
iron phosphate precipitates. These compounds
include aluminum sulfate (alum), sodium aluminate,
ferric chloride, ferrous chloride, and ferrous sulfate.
They are generally added upstream of either the
primary clarifier or the secondary clanfier. In some
cases metal salts are added to both primary and
secondary processes. Figure 4-1 shows alternative
-------
metal salt addition flow schematics. The chemicals
may also be added separately to a tertiary clarifier.
The quantity of metal salt added is determined by the
concentration of phosphorus species in the influent
wastewater and effluent discharge permit
requirements.
Systems with metal salt addition can achieve 80-95
percent total phosphorus removal. For effluent
limitations of 1.0 mg/l total phosphorus, metal salt
addition with conventionally designed clarifiers is
acceptable. Effluent limitations of 1.0 mg/l total
phosphorus can be met with metal salt addition and
efficient clarification to assure effluent TSS of less
than 15 mg/l. To consistently meet total phosphorus
discharge limitations of 0.5 mg/l, filtration of
secondary effluent will most likely be necessary.
Determination of the best point or points of chemical
addition is best determined by full-scale plant
testing. Jar tests using the different metal salts can
provide sufficient information needed to conduct cost
estimating and evaluation of the relative impact of
sludge production on sludge handling and disposal
processes. Pilot or full-scale testing is
recommended to develop detailed design criteria
where stringent discharge requirements (e.g., 0.5
mg/l total phosphorus) are imposed.
Advantages and disadvantages of metal salt addition
for phosphorus removal are summarized in Table 2-
3.
2.3.1.2 Lime Addition
Lime is used to remove phosphorus by addition either
to the primary clarifier or to the effluent from the
secondary clarifier in a separate tertiary unit.
Phosphorus removal with lime is basically a water
softening process and the quantity of lime required to
remove the phosphorus is dependent on the alkalinity
of the wastewater rather than on the phosphorus
content.
Lime removal systems are either low-lime (single
stage) where pH is kept below 10.0 and which can
achieve 1.0 mg/l effluent total phosphorus levels or a
two-stage high-lime process that raises the pH to
11.0-11.5 and is used to achieve very low (<1.0
mg/l) effluent total phosphorus concentrations. High-
lime treatment uses more lime and also requires
recarbonation to reduce the wastewater pH before
discharge to a downstream biological unit or
discharge from the plant. The effluent phosphorus
levels from a two-stage tertiary lime process with
filtration can meet total phosphorus effluent limits as
low as 0.1.
The lime addition process produces a substantial
amount of additional sludge, even greater than with
metal salt addition.
Lime addition systems are usually pH controlled and
entail lime storage, feed, and mixing units. This
equipment often requires considerable maintenance.
Lime can be reused by calcining the lime sludge. This
step is only applicable at larger plants due to the high
capital and operating costs for the recalcining
process. Even with recalcining, about 20-30 percent
of make-up lime is required.
Lime addition is seldom practiced in the United States
today due to the high chemical usage, problems
associated with handling lime, and the large volume
of sludge generated from lime addition systems.
Advantages and disadvantages of lime addition for
phosphorus removal are summarized in Table 2-4.
2.3.2 Biological Phosphorus Removal Alternatives
Biological phosphorus removal is a recently
developed technique of designing suspended growth
activated sludge systems to remove soluble
phosphorus from wastewater. Six variations on this
phenomenon are described in Chapter 3 of this
manual. These alternatives are:
Phostrip process
Modified Bardenpho process
A/0 process
UCT (University of Capetown) process
Sequencing Batch Reactor (SBR) process
Operationally modified activated sludge processes
Figures 3-4, 3-5, 3-6 and 3-7 are flow diagrams
of these processes. Detailed descriptions of the
various biological processes and the theories and
mechanisms of operation can be found in Chapter 3.
Case histories of various facilities are also presented.
Phostrip is the only biological phosphorus removal
process that incorporates an anaerobic zone in the
sludge recycle system. The Phostrip process takes a
sidestream from the return activated sludge and
subjects it to anaerobic conditions in a separate
anaerobic tank before returning the sludge to the
aeration basin. Maintenance of the activated sludge in
the anaerobic state leads to phosphorus release, and
when the sludge is returned to the aeration basin or
reaerated, the activated sludge biota take up an
excess amount of phosphorus during the growth
process. Lime is used to precipitate the phosphorus
released in the anaerobic tank. Since only 20 to 30
percent of the plant flow passes through the
anaerobic tank, the quantity of lime required is much
less than in a mainstream lime addition system, and
less sludge is produced.
The Modified Bardenpho process and the UCT
process are designed to remove both nitrogen and
phosphorus.
-------
Table 2-3. Advantages and Disadvantages of Metal Salt Addition for Phosphorus Removal
Advantages
Disadvantages
1. Reliable, well documented phosphorus removal technique.
Most popular process used in the United States.
2. Chemical costs can be reduced substantially if wasle pickle
liquors (ferrous chloride or ferrous sulfate) are available and
can be used.
3. Chemical usage requirement is basically dependent on total
phosphorus concentration of wastewater and required effluent
levels.
4. Controls required for phosphorus removal are fairly simple and
straightforward.
5. Relatively easy and inexpensive to install at existing facilities
6. Sludge produced can be processed in same manner as in
non-phosphorus removal systems.
7. Primary clanfier metal addition can reduce organic load to
secondary unit by 25-35 percent.
8. Effluent phosphorus levels can be controlled by metal salt
dosages to maximum efficiency levels.
1. Chemical costs higher than for biological phosphorus removal
systems which require little or no metal salt addition
2. Significantly more sludge produced than with wastewater
treatment process without metal addition; may overload
existing sludge handling equipment; higher sludge treatment
and disposal costs.
3. Sludge produced generally does not dewater as well or as
easily as conventional wastewater treatment plant sludges
where metal salts are not added.
Table 2-4. Advantages and Disadvantages of Lime Addition for Phosphorus Removal
Advantages Disadvantages
1. Simple process control, as lime dosage paced by pH control
Lime dosage required does not vary with phosphorus
concentrations, only alkalinity of wastewater.
2. Very high phosphorus removals achievable with high lime
process.
3. Many heavy metals, such as chrome, nickel, etc., are
effectively removed.
4. Primary lime addition reduces organic load to biological
treatment units.
1 High chemical costs for wastewater facility with hard (high
alkalinity) waters.
2. More sludge produced than for any other phosphorus removal
process.
3. Equipment requirements and maintenance costs for lime
storage, feeding, and handling equipment are extremely high.
4. High capital and operating costs; not widely used in the United
Stales today.
5. Additional recarbonation step required for high lime process.
The A/O process is primarily designed for phosphorus
removal but it can also be designed to accomplish
both phosphorus removal and nitrification.
Biological phosphorus removal has been achieved in
existing plug flow activated sludge plants by modifying
the aeration practice at the head end of the aeration
tanks. This modification consists of shutting off the air
or aerators at the head of the tank to promote an
anoxic or anaerobic zone before the wastewater and
return sludge are aerated.
Phosphorus removal in an SBR has been
accomplished in limited full-scale tests by modifying
the sequencing periods and aeration schedules to
provide the needed anaerobic/aerobic mixing periods.
The sludge quantities produced by all the biological
phosphorus removal processes, with the exception of
Phostrip, are not any greater than those produced
from conventional suspended growth systems.
Processing of the biological phosphorus sludges,
however, requires care that the sludge processing
steps do not result in phosphorus resolubilization, and
return of soluble phosphorus to the system.
The degree of phosphorus removal from biological
phosphorus removal processes (except Phostrip) is
dependent on the influent BOD:P ratio of the
wastewater. Within the proper BOD:P ranges, the
Modified Bardenpho, A/0, UCT and operationally
modified activated sludge processes can achieve
1.0-2.0 mg/l total phosphorus (TP) level in the final
effluent. A TBOD:TP ratio of at least 20:1
(SBOD:soluble P of 12:1 to 15:1) is usually required
to meet these limitations. Achievement of lower
effluent total phosphorus concentrations normally
requires efficient clarification to achieve less than 20
-------
Table 2-5. Advantages and Disadvantages of Biological Phosphorus Removal Processes
Advantages
Disadvantages
1. Sludge quantities generated by biological phosphorus removal
processes are comparable to sludge production from
conventional activated sludge systems.
2. Can be implemented directly at existing plug flow activated
sludge plants with little or no equipment changes or additions,
provided that the plant has sufficient capacity.
3. Can utilize existing sludge handling equipment for plants
retrofitted with biological phosphorus removal process if
phosphorus is not solubilized and returned to the plant during
sludge handling.
4. Little or no chemicals or chemical handling equipment required
except for Phostnp process or for effluent polishing.
5. Phosphorus removal can be accomplished together with
ammonia nitrogen or total nitrogen removal at virtually no
additional operating cost with some of the processes.
6. For some of the processes, better control of filamentous
organisms in the activated sludge system is possible.
1. In all but Phostrip, phosphorus removal performance is
controlled by the BOD:TP ratio of the wastewater.
2. Requires highly efficient secondary clanfier performance to
achieve 1 mg/l total phosphorus.
3. Not easily retrofitted into fixed film biological systems.
4. Potential for phosphorus release in sludge handling system.
Recycle streams must be low in phosphorus content.
5. Standby chemical feed equipment may be necessary in case
of loss of biological phosphorus removal efficiency.
mg/l TSS concentrations or the use of tertiary filters.
Discussion of the effect of BOD:P ratios and other
factors on phosphorus removal is contained in
Section 3.3.5.
In all these systems, pilot testing is highly
recommended to determine what performance levels
can be achieved for the specific wastewater to be
treated. Advantages and disadvantages of biological
phosphorus removal systems are summarized in
Table 2-5.
2.4 Phosphorus Removal System
Selection Strategy
The strategy that is described below for selecting a
phosphorus removal system is a selective screening
process whereby all phosphorus removal alternatives
are initially considered, and then are screened against
various sets of criteria that impact the choice of a
system. At each step in the four step process,
alternatives are evaluated against a corresponding
application matrix (or set of matrices) which shows
the applicability of the process for various sets of
conditions. Non-applicable technologies are rejected
at each step with the selected alternative being
chosen as a result of the fourth step of the process,
which is the final cost-effectiveness analysis. A flow
diagram of the overall selection strategy is depicted in
Figure 2-1.
It should be emphasized that this approach is
intended as a guide for assisting the engineer in the
overall selection process. The approach is not
rigorous, since subjective information is used in the
selection/rejection process and procedures for
reiterating through the process are not provided.
The four steps of the selective screening process are
described below.
Step 1:
Categorize the facility as to whether it is an existing
or a new plant, and whether the effluent nutrient
discharge limitations are for phosphorus only, or for
phosphorus plus nitrogen. In general, for new
facilities, all possible alternatives should initially be
considered, although some may have greater
apparent applicability than others. For existing
facilities, some alternatives can justifiably be removed
from further consideration.
For example, for a trickling filter plant with sufficient
design capacity that must meet a new phosphorus
limitation, it is unlikely that a new biological
phosphorus removal system would be competitive
with metal salt addition to the primary clarifier.
However, if the same plant were required to meet
both phosphorus and nitrogen limitations, a new
biological suspended growth system designed for N
and P removal may be more cost effective than
modifying the existing plant to achieve nutrient
removal. The applicability criteria matrix for Step 1
screening is shown as Table 2-6.
Step 2:
Apply P-removal process capabilities and determine
which processes can meet phosphorus limitations.
For those plants with phosphorus and nitrogen
limitations, the ability of the process to meet nitrogen
limitations will also be considered. In conducting a
study of plants having both phosphorus and nitrogen
effluent limitations, an additional screening step
should be made to include other nitrogen removal
processes that can be used independent of the
-------
Figure 2-1. Phosphorus Removal System Selection Strategy.
All Alternatives
Selected
Alternatives
CRITERIA:
Evaluate according to
criteria in Table 2-6
Evaluate according to
criteria in Tables 2-7
and 2-8
Evaluate according to
criteria in Tables 2-9
thru 2-11
Evaluate according to
cost-effectiveness
analysis
STEP: 1
Application criteria
New plant
Existing plant
Suspended growth
Fixed film
P removal only
P and N removal
Application criteria
Effluent P limitations
Effluent P and N
limitations
Application criteria
BOD:TP ratio
Sludge production
O&M
Application criteria
Total present worth
Site requirements
Reliability
Environmental impact
Table 2-6. Application Criteria Matrix - New or Existing Facility; Type of Nutrient Removed
P Removal Only
P plus NH4 or N Removal
Process
(A) Phostnp
(B) Mod. Bardenpho
(C) A/O
(D) SBR
(E) UCT
(F) Mod. A.S.
(G) Metal Salt
(H) 1 -Stage Lime
(I) 2-Stage Lime
Y - Applicable
N - Not Applicable
Existing Susp.
New Facility Growth
Y
E
Y
Y
E
Y
Y
Y
Y
E
M
Y*
Y
E
Y
Y
E
Y
Y
Y
Y
Existing Fixed
Growth
N
N
N
N
N
N
Y
Y
Y
New Facility
Y
Y
Y
Y
Y
Y
Y
Y
Y
Existing Susp.
Growth
M
Y
M
Y
Y
M
M
M
M
Existing Fixed
Growth
Y*
N
Y"
N
N
Y*
M
Y"
Y*
- May be Applicable, but exceeds treatment requirements.
- May be Applicable, but N or NH4 removal step required.
- Applicable where existing fixed film unit can be used for nitrification.
-------
phosphorus removal process, i.e. consider alum for
phosphorus removal and breakpoint chlorination, or a
separate nitrification tower to meet the P and N
limitations.
In addition, effluent limitations such as BOD5 and
suspended solids as well as other site specific
requirements not specifically listed as "application
criteria" may lead to rejection or inclusion of some
P-removal processes.
To help select the processes which can meet the
various phosphorus, or phosphorus and nitrogen
limitations, Tables 2-7 and 2-8 are used. These
tables are matrices showing the ability of the systems
to meet various phosphorus limits as well as being
able to meet effluent nitrogen limits. These matrices
are intended to be used as guidelines and are not
intended for rigid process selection. Specific data on
the nitrogen removal/conversion characteristics of the
biological phosphorus removal processes must be
studied carefully before determining whether a
specific process should be included or eliminated as
a possible alternative.
Step 3:
Step 3 consists of screening the alternative
processes that have been found to be capable of
meeting the effluent phosphorus or phosphorus and
nitrogen limitations with the applicability criteria in
Tables 2-9 through 2-11. The objectives of this
screening are to eliminate those processes which are
not suitable for further consideration. Where a
process is shown to be marginal and other processes
are shown to be capable of meeting the effluent
limitations, the marginal processes should also be
carried through to step 4.
The Application Criteria Matrices shown as Tables
2-9 through 2-11 have been developed to provide
a means of quickly determining the factors which are
pertinent for screening the different phosphorus
removal processes. As in the previous tables, these
matrices are intended to provide a means of
evaluating the impact of various factors on the
different phosphorus removal processes. In using the
matrices the user should read the applicable sections
of the manual which describe the processes in detail.
A brief description of the use of the matrices follows.
Table 2-9, Effect of Influent TBOD:TP Ratio on
Process Applicability, is based on research that
indicates plants with a TBOD:TP ratio less than 20
may have difficulty achieving final effluent total
phosphorus concentrations of 1.0-2.0 mg/l. Specific
sampling data on the variation of TBOD:TP ratio
and/or pilot-plant studies for a specific facility may
be necessary to define the capabilities of the
biological phosphorus removal process under
consideration.
Table 2-10, Sludge Production, indicates the effect
on sludge production of the different phosphorus
removal systems for typical process applications.
Operation and maintenance requirements for
increased sludge production are included in the
ratings for the processes. More specific data on
sludge production for the specific processes are
presented in Chapter 5.
Table 2-11, O & M Effects, shows the difference in
O & M requirements for the different processes. The
0 & M comparison does not include sludge handling
and disposal since this aspect is covered in Table 2-
10.
Step 4:
Step four consists of developing the capital, operation
and maintenance, and total present worth costs for all
applicable alternatives. Non-monetary factors are
also considered at this point, including:
1. Site requirements.
2. Reliability.
3. Environmental Impacts.
4. Operator skill level required for successful
operation.
The results of the comprehensive cost effectiveness
analysis, which evaluates costs of screened
alternatives against appropriate non-monetary
factors, results in the selection of a system which
meets project objectives at the lowest present worth
cost.
10
-------
Table 2-7. Application Criteria Matrix - Ability of Process to Meet Effluent Phosphorus Limitations
Process
(A) Phostrip
(B) Bardenpho
(C) A/O
(D) SBR
(E) UCT
(F) Mod. A.S.
(G) Metal Salt
(H) 1 -Stage Lime
(I) 2-Stage Lime
Alone
N
N
N
N
N
N
N
N
N
0.5 mg/l 1 .0 mg/l 2.0 mg/l
Effluent TP Effluent TP Effluent TP
w/M.S. w/M.S. w/M.S.
w/M.S. w/F & F Alone w/M.S. w/F & F Alone w/M.S. w/F & F
NNYYYYYYYYY
NNYMY'MYYYYY
NNYMY-MYYYYY
NNYMY'MYYYYY
NNYMY*MYYYYY
NNYMY*MY Y YYY
Y - Y" - Y - Y - Y -
NYYY'MYY Y YYY
MYYY'MYYY- Y-
Ability of alternatives (B)-(F) to meet effluent limits alone based on TBOD:TP ratio being above 20.
N - Can not meet effluent limits.
M - Marginal for meeting effluent limits.
Y - Will meet effluent limits.
Y* - Will meet effluent limits with highly efficient clarification.
M.S. - Metal salt addition to secondary clanfier.
F - Filtration of secondary clanfier effluent.
M.S. & F - Metal salt addition to secondary clanfier and secondary clanfier effluent filtration
Table 2-8.
Application Criteria Matrix - Ability of Process
to Remove NH4 or TN
Process
(A) Phostrip
(B) Bardenpho
(C) A/O
(D) SBR
(E) UCT
(F) Mod A.S
(G) Metal Salt
(H) 1 -Stage Lime
(I) 2-Stage Lime
0.5 mg/l
Effluent TP
NH4 TN
N/A N/A
N/A N/A
N/A N/A
N/A N/A
N/A N/A
N/A N/A
N N
N N
N N
1 .0 mg/l
Effluent TP
NH4 TN
Y N
Y Y
Y N
Y Y
Y Y
Y N
N N
N N
N N
2.0 mg/l
Effluent TP
NH4 TN
Y N
Y Y
Y
Y Y
Y Y
Y N
N N
N N
N N
Y - Can remove NH4 or total N.
N - Need separate process or modification for NH4 or total N
removal.
N/A - Not applicable for effluent P limitation shown.
11
-------
Table 2-9. Application Criteria Matrix - Effect of TBOD:TP Ratio < 20 on Process Applicability
Process
(A) Phostnp
(B) Mod. Bardenpho
(C) A/O
(D) SBR
(E) UCT
(F) Mod. A.S.
(G) Metal Salt
(H) 1 -Stage Lime
(I) 2-Stage Lime
0.5 mg/l
Effluent TP
Application
N/A
N/A
N/A
N/A
N/A
N/A
L
L
L
1 .0 mg/l
Effluent TP
Application
L
N
N
N
N
N
L
L
L
2.0 mg/l
Effluent TP
Application
L
M
M
M
M
M
L
L
L
L - No effect.
M - Marginal; may not meet effluent TP limitations without metal salt addition.
N - Cannot meet effluent phosphorus limits without metal salt addition.
N/A - Not applicable for effluent TP limitation shown.
Table 2-10. Application Criteria
Process
(A) Phostnp
(B) Mod. Bardenpho
(C) A/O
(D) SBR
(E) UCT
(F) Mod. A.S.
(G) Metal Salt
(H) 1 -Stage Lime
(I) 2-Stage Lime
Matrix - Effect of Sludge
0.5 mg/l
Effluent TP
Application
M
L
L
L
L
L
H
H
H
Production on Process Applicability
1 .0 mg/l
Effluent TP
Application
M
L
L
L
L
L
M
H
H
2.0 mg/l
Effluent TP
Application
M
L
L
L
L
L
M
H
H
L - Little or no increase in sludge production or handling (< 30%).
M - Increased sludge production and/or handling (30 - 100%).
H - Substantial increase in sludge production and handling (> 100%).
12
-------
Table 2-11. Application
Process
(A) Phostrip
(B) Mod. Bardenpho
(C) A/O
(D) SBR
(E) UCT
(F) Mod. A.S.
(G) Metal Salt
(H) 1 -Stage Lime
(I) 2-Stage Lime
Criteria Matrix • Effect of O&M Requirements
0.5 mg/l
Effluent TP
Application
N/A
N/A
N/A
N/A
N/A
N/A
M
S
S
on Process Applicability1
1.0 mg/l
Effluent TP
Application
M
M
M
M
M
M
L
M
S
2.0 mg/l
Effluent TP
Application
M
L
L
L
L
M
L
M
S
L - Little or no increase in O&M (0 - 10%).
M - Some increase in O&M (10 - 30%).
S - Substantial increase in O&M (> 30%).
N/A - Not applicable for effluent TP limitation shown.
1 O&M - For process control, monitoring and operations excluding sludge handling.
13
-------
CHAPTERS
Phosphorus Removal by Biological Processes
3.1 Introduction and Theory
Conventional secondary biological treatment systems
accomplish phosphorus removal by using phosphorus
for biomass synthesis during BOD removal.
Phosphorus is an important element in
microorganisms for energy transfer and for such cell
components as phospholipids, nucleotides, and
nucleic acids. Attachment of a phosphate radical
bond to adenosine triphosphate (ATP) results in the
storage of energy (7.4 Kcal/mole P), which is
available upon conversion of ATP back to adenosine
diphosphate (ADP). Phosphorus is also contained in
nucleotides such as nicotinamide adenine
dinucleotide (NAD) and flavin adenine dinucleotide
(FAD) which are used for hydrogen transfer during
substrate oxidation-reduction reactions. Ribonucleic
acid (RNA) and deoxyribonucleic acid (DMA) are
composed of a deoxyribose sugar structure with
attached amino acids of adenine, cytosine, guanine,
and thymine or uracil. The deoxyribose molecules are
attached by phosphorus bonds. Phosphorus may
account for 10-12 percent of the RNA or DNA mass.
A typical phosphorus content of microbial solids is
1.5-2 percent based on dry weight. Wasting of
excess biological solids with this phosphorus content
may result in a total phosphorus removal of 10-30
percent, depending on the BOD-to-phosphorus
ratio, the system sludge age, sludge handling
techniques, and sidestream return flows.
In 1955, Greenburg et a/. (1) proposed that activated
sludge could take up phosphorus at a level beyond its
normal microbial growth requirements. In 1959,
Srinath (2) reported on batch experiments to
conclude that vigorous aeration of activated sludge
could cause the concentration of soluble phosphorus
in mixed liquor to decrease rapidly to below 1 mg/l. In
1965, Levin and Shapiro (3) reported on enhanced
biological phosphorus removal using activated sludge
from the District of Columbia activated sludge plant.
Over 80 percent phosphorus removal was observed
by vigorous aeration of the sludge and without the
addition of chemicals. They termed the high
phosphorus removal "luxury uptake" by the
microorganisms. In some experiments, a small
amount of 2-4 di-nitrophenol was added that
inhibited phosphorus uptake, indicating the removal
was of biological origin. They also observed volutin
granules in the bacterial cells, which are reported in
the microbial literature to contain polyphosphates.
Acidification of the sludge resulted in the release of
phosphorus, which led to a proposed treatment flow
scheme of exposing return sludge to acidic conditions
and stripping of phosphorus. Shapiro et a/. (4) later
observed high phosphorus uptake at the Baltimore
sewage treatment plant and release in the bottom of
the secondary clarifiers under conditions of zero or
low dissolved oxygen (DO). Theyt proposed that the
return sludge could be intentionally exposed to such
conditions prior to return to the aeration basin to strip
out phosphorus. This work led to the development of
the Phostrip process (5,6).
High levels of phosphorus removal were observed at
various full-scale activated sludge plants in the
United States, including the Rillings Road plant in San
Antonio, Texas (7), the Hyperion plant in Los
Angeles, California (8), and the Back River plant in
Baltimore, Maryland (9). The three plants reported
total phosphorus removals of 85-95 percent, and the
phosphorus content of the waste sludge was 2-7.3
percent on a dry weight basis. All of the plants were
of the plug flow configuration using diffused aeration,
and the following operating characteristics were
judged important in all or some of the plants to
maximize phosphorus removal:
1. Require a DO concentration of 2.0 mg/l or greater
from the middle to end of the plug flow aeration
basins.
2. Prevent the recycle of phosphorus back to the
activated sludge system via sludge handling
streams.
3. Maintain aerobic conditions in the secondary
clarifiers to prevent the release of phosphorus into
the effluent.
Both Vacker (7) and Milbury (9) noted phosphorus
release by the mixed liquor and an increase in soluble
phosphorus concentration near the inlet of the
activated sludge tanks at the Rillings Road and Back
River plants, respectively. During the late 1960s to
15
-------
early 1970s there were varying opinions as to
whether the excess phosphorus removal observed at
these plants was due to chemical precipitation. In an
attempt to provide a rational chemical removal
explanation, one hypothesis was that the high pH, as
a result of high aeration rates and carbon dioxide
stripping at the end of the plug flow basins,
encouraged the formation of a calcium hydroxyapatite
precipitate (10).
In addition to the development of the Phostrip
biological phosphorus removal process in the early
1970s in the United States, biological phosphorus
removal was observed during the development of the
Bardenpho four-stage biological nitrification-
denitrification system by Barnard (11). The system
consists of sequential anoxic-aerobic-anoxic-
aerobic stages with an internal mixed liquor recycle
from the first aerobic stage to the first anoxic stage.
During a period of high phosphorus removal in a
100-m3/d (18-gpm) pilot-plant operation, Barnard
observed a soluble phosphorus concentration of 0.3
mg/l in the final aerobic basin. He recognized that
phosphorus was being released in the designated
second "anoxic" basin, which was actually
experiencing anaerobic conditions (absence of both
nitrate nitrogen and DO) and that it was being taken
up in the final aerobic stage. This led him to conclude
that biological phosphorus removal was possible in
activated sludge systems provided that an aerobic
stage was preceded by an anaerobic stage where
phosphorus release occurred. It was also noted that
when a high level of phosphorus removal was
reported in plug flow U.S. plants, phosphorus release
occurred near the inlet of the aeration basin followed
by phosphorus uptake along the length of the basin
where the DO concentration increased.
In a later paper, Barnard (12) proposed the use of a
separate anaerobic basin ahead of the Bardenpho
nitrogen removal system or ahead of aerobic basins
when nitrogen removal was not necessary. The
former was called the Modified Bardenpho process
and the latter the Phoredox process. Phoredox was
derived from "phosphorus" and "redox potential,"
which is at a lower level in the anaerobic phosphorus
release zone. Figure 3-1 shows phophorus release
and uptake characteristics of such biological
phosphorus removal systems that employ sequential
anaerobic-aerobic contacting.
Following Barnard's pilot-plant work, full-scale
plants were modified at Johannesburg, South Africa,
to investigate the feasibility of biological phosphorus
removal. At the Alexander plant, surface aerators near
the inlet of an activated sludge basin were turned off
to create an anaerobic zone (13). An overall nitrogen
removal of 85 percent and a total phosphorus
removal of 46 percent were reported. At the
Olifantsvlei plant, various combinations of surface
aerators were turned off in the four-stage system
Figure 3-1. Biological phosphorus and BOD removal due to
anaerobic-aerobic contacting.
Return Sludge
Time
and an effluent soluble phosphorus concentration of
0.9 mg/l was reported (14). Based on this work, a
modified Bardenpho system was designed for a
I50,000-m3/d (39-mgd) facility at the
Johannesburg Goudkuppies wastewater plant that
became operational in 1978 (15). In the late 1970s, a
modified Bardenpho plant was started up at Palmetto,
Florida (16), and a portion of the Largo, Florida facility
was converted to the A/O process (17), an
anaerobic-aerobic biological phosphorus removal
system to be discussed later.
3.1.1 Biological Phosphorus Removal Mechanism
The generally accepted theory for biological
phophorus removal is that anaerobic-aerobic
contacting results in a competitive substrate utilization
and selection of phophorus-storing microorganisms
(18,19). An understanding of the steps involved in the
biological phosphorus removal mechanism provides a
useful insight into the factors that can affect the
performance of biological phosphorus removal
systems. The following observations by various
investigators are presented as a background to the
proposed mechanism.
Funs and Chen (20) examined activated sludge from
the Baltimore Back River and the Seneca Falls, New
York treatment plants when the plants were exibiting
high levels of phosphorus removal. They concluded
that the organism asssociated with phosphorus
removal belonged to the Acinetobacter genus. These
bacteria are short, plump, gram-negative rods with a
size of 1-1.5 pm. They appear in pairs, short chains,
or clusters. They also subjected a pure culture to
16
-------
batch anaerobic-aerobic cycles and noted excess
phosphorus removal when acetate was fed to the
system. They postulated that the anaerobic phase in
excess phosphorus removal systems was important
for the production of simple carbohydrates such as
ethanol, acetate, and succinate, which serve as
carbon sources for Acinetobacter. Contrary to later
findings, they felt that the simple carbohydrates were
assimilated by the Acinetobacter in the aerobic phase
of the cycle. Fuhs and Chen also found that a
significant phosphorus release rate could be
promoted by the addition of carbon dioxide during the
anaerobic phase, which also lowered the pH. This
was also observed by Deinema (21).
Other investigators also reported observing significant
levels of Acinetobacter in biological excess
phosphorus removal systems (22-24). Letter (24)
also found significant levels of Aeromonas and
Pseudomonas which are capable of polyphosphate
accumulation. Hascoet et al. (25) also noted the
presence of Bacillus cereus in addition to
Acinetobacter and Suresh et al. (26) found small
amounts of Pseudomonas vesiculcris, besides
Acinetobacter, in samples cultured from an
anaerobic-aerobic phosphorus removing pilot plant.
Brodich (27) noted that the removal of phosphorus in
a system containing Acinetobacter became significant
only after the development of an Aeromonas
population. He postulated that the Aeromonas
bacteria served the important function of producing
fermentation products in the anaerobic phase for the
Acinetobacter. Letter and Murphy (28) noted an
increase of Pseudomonas and Aeromonas in
biological phosphorus removal systems. They also
found that these species of bacteria and a species of
Acinetobacter accomplished denitrification in anoxic
zones of biological nitrogen removal systems. Osborn
and Nicholls (15) reported rapid biological phosphorus
uptake during nitrate reduction in the absence of DO,
indicating that phosphorus uptake may be occurring
with denitrifying bacteria. Hascoet (25) also reported
phosphorus release in anoxic zones by Acinetobacter
provided that there was a relatively high level of
substrate availability.
Various investigators have observed a decrease in
soluble substrate and an increase in orthophosphate
concentrations in the anaerobic zone of anaerobic-
aerobic sequenced biological phosphorus removal
systems. Hong (29) showed a soluble BODs (SBOD)
concentration decrease from 45 to 15 mg/l and an
orthophosphorus concentration increase from 6 to 24
mg/l in the anaerobic zone. Ekama et al. (30) related
phosphorus release in the anaerobic zone to the
presence of a soluble, readily biodegradable
substrate. The concentration of a soluble readily
biodegradable substrate can be determined from the
increase in the oxygen uptake rate measurements of
a batch activated sludge sample after the addition of
influent. Rensink's (31) work on Acinetobacter led
him to investigate the change in acetate and soluble
phosphorus concentrations in the anaerobic phase.
Figure 3-2 shows the decrease in acetate
concentration and increase in orthophosphate
concentration as a function of the anaerobic time.
The molar ratio of acetate utilization to phosphorus
release was 1.3.
Figure 3-2. Acetate assimilation and phosphorus release
vs. anaerobic time (31).
Orthophosphorus *
20 40
Anaerobic Time, min.
60
Fukase (32), using fill-and-draw reactors, observed
an acetate utilization to phosphorus release molar
ratio of 1.0. Arvin (33) reported 0.7, Rabinowitz (34)
0.6, and Wentzel (35) 1.0 from batch studies using
sludge from excess biological phosphorus removal
systems. Rabinowitz (34) also found that the
phosphorus release magnitude and rate were affected
by the type of substrate. The amount of phosphorus
release for each substrate in decreasing order was
sodium acetate, propionic acid, glucose, acetic acid,
and butyric acid. On the other hand, Jones (36)
observed a greater phosphorus release in declining
order of butyric acid, ethanol, acetic acid, methanol,
and sodium acetate.
Release and uptake of metal ions have been
observed during phosphorus release in biological
phosphorus removal systems. A summary of the data
has been presented by Comeau (37) and is shown in
Table 3-1.
The understanding of the biological phosphorus
removal mechanism was significantly advanced with
the observations on storage of carbohydrate products
within biological cells in the anaerobic zone and
phosphorus-containing volutin granules in the
aerobic zone. The most commonly reported
17
-------
Table 3-1. Molar Ratios of Ions Co-Transported with
Phosphorus.
Ref. 38 Ref. 39 Ref. 37 Ref. 37 Ref. 37
Mg + 2/p
K + /P
Ca + 2/P
Na+/P
charges/P
0.26
0.27
0.00
-
0.79
0.32
0.23
-
-
0.87
0.24
0.34
0.06
0.00
0.94
0.28
0.20
0.09
0.00
0.94
0.27
0.23
0.12
0.00
1.01
anaerobic intracellular storage product has been
polyhydroxybutyrate (PHB).
PHB has been found in biologically-removed
phosphorus sludges by Timmerman (40) and in
Acinetobacter by Nicholls and Osborn (41). Lawson
and Tonhazy (23) isolated Acinetobacter and showed
that these bacteria could accumulate PHB and
polyphosphates. Deinema (42) also observed PHB in
a strain of phosphorus-removing Acinetobacter.
Senior (43) hypothesized that certain bacteria will
accumulate PHB during temporary deprivation of
oxygen. Buchan (22) reported that PHB increased in
bacterial cells while polyphosphate granules
decreased in size or disappeared in the anaerobic
zone of biological phosphorus removal systems.
PHB synthesis and degradation are described by
Gaudy (44). PHB is formed in the cell under
anaerobic conditions from acetoacetate serving as a
hydrogen acceptor. Acetate entering the bacterial
cells under anaerobic conditions can be converted to
acetyl-COA provided energy is available, and
acetyl-COA can be converted to acetoacetate since
the cell has a limited supply of the enzyme COA.
During oxidative conditions, PHB is oxidized to
acetyl-COA, which enters the Krebs cycle. PHB
oxidation does not occur until nearly all of the
endogenous carbon is used up according to Gaudy.
As described previously, Levin (3) reported finding
volutin granules in sludge samples during biological
phosphorus removal. In an extensive evaluation of
biological phosphorus removal, Harold (45) stated that
phosphorus was likely stored as polyphosphates
within volutin granules. Volutin granules contain lipids,
protein, RNA, and magnesium in addition to
polyphosphates. The granules are visible under the
light microscope and can also be identified by staining
with either toludine dye, which results in a reddish-
purple color, or with a methylene blue technique,
which results in a dark purple color. A high electron
beam directed on the microorganisms will also
volatilize the polyphosphate contained in the volutin
granules leaving holes in the cells.
Sell (46) photographed large masses of
polyphosphate granules contained in phosphorus
removing sludge during a cold temperature laboratory
investigation of biological phosphorus removal.
Buchan (22) analyzed the biological species obtained
from aerobic zones of various South African activated
sludge plants accomplishing biological phosphorus
removal. His analysis showed that the intracellular
polyphosphate granules contained an excess of 25
percent phosphorus. In the anaerobic zone, the large
polyphosphate granules had dispersed into smaller
granules and some cells had released virtually all of
their accumulated phosphorus.
The proposed biological phosphorus removal
mechanism (18,19) is summarized in Figure 3-3.
Acetate and other fermentation products are
produced from fermentation reactions by normally-
occurring facultative organisms in the anaerobic zone.
A generally accepted concept is that these
fermentation products are derived from the soluble
portion of the influent BOD and that there is not
sufficient time for the hydrolysis and conversion of
the influent paniculate BOD. The fermentation
products are preferred and readily assimilated and
stored by the microorganisms capable of excess
biological phosphorus removal. This assimilation and
storage is aided by the energy made available from
the hydrolysis of the stored polyphosphates during
the anaerobic period. The stored polyphosphate
provides energy for active transport of substrate and
for formation of acetoacetate, which is converted to
PHB. The fact that phosphorus-removing
microorganisms can assimilate the fermentation
products in the anaerobic phase means that they
have a competitive advantage compared to other
normally-occurring microorganisms in activated
sludge systems. Thus, the anaerobic phase results in
a population selection and development of
phosphorus-storing microorganisms. Rensink (31)
has pointed out that Acinetobacter are relatively slow
growing bacteria and that they prefer simple
carbohydrate substrates. Thus, without the anaerobic
phase, they may not be present at significant levels in
conventional activated sludge systems.
During the aerobic phase, the stored substrate
products are depleted (22) and soluble phosphorus is
taken up, with excess amounts stored as
polyphosphates. An increase in the population of
phosphorus-storing bacteria is also expected as a
result of substrate utilization. The above mechanism
indicates that the level of biological phosphorus
removal achieved is directly related to the amount of
substrate that can be fermented by normally-
occurring microorganisms in the anaerobic phase and
subsequently assimilated and stored as fermentation
products by phosphorus-removing microorganisms,
also in the anaerobic phase.
3.2 Applications
The recent developments leading to a better
understanding of the conditions causing excess
18
-------
Figure 3-3. Scxhematic of biological phosphorus removal
mechanism.
Substrate
Facultative
Bacteria
Anaerobic
Acetate plus
Fermentation
Products
Phosphorus-
Removing
Bacteria
CO2 + H2O
biological phosphorus removal help to explain the
earlier observations on excess phosphorus removal
reported for full-scale facilities. It is apparent now
that sufficient BOD was present and oxygen was
limiting so that fermentation conditions likely occurred
at the front end of the relatively long, narrow aeration
basins of these plants. Since these observations,
three major proprietary biological phosphorus removal
processes that employ more definitive anaerobic
fermentation zones have been commercialized. These
processes are, in order of development, the Phostrip
process, the modified Bardenpho process, and the
A/O process. These processes as well as other non-
proprietary systems will be described in this section.
Other options used are the UCT process, sequencing
batch reactors (SBRs), and operationally modified
activated sludge systems.
3.2.7 Process Descriptions
The three commercial biological phosphorus removal
processes are shown in Figure 3-4. The Phostrip
process was first proposed by Levin in 1965 (3). Pilot
plant data were collected at a number of municipal
plants from 1970 to 1973 and demonstrated high
levels of phosphorus removal. In 1973, the Seneca
Falls, New York activated sludge plant was converted
to the Phostrip process and evaluated (6). The
process combines both biological and chemical
phosphorus removal and has been referred to as a
sidestream process since a portion of the return
activated sludge flow is diverted for phosphorus
stripping and subsequent precipitation with lime. An
advantage of the Phostrip process is that an effluent
concentration of less than 1 mg/l total phosphorus
can be achieved with less dependence on the BOD
strength of the influent wastewater. A large
percentage of the phosphorus removed is tied up as
a lime sludge, which causes less concern than
handling a phosphorus-rich waste biological sludge.
Compared to chemical addition to an activated sludge
aeration basin for phosphorus precipitation, the
Phostrip process may require a lower chemical
dosage and cost, since the lime dosage is a function
of the alkalinity and not the amount of phosphorus to
be removed, as is the case for alum and iron salts.
This potential advantage is a function of wastewater
alkalinity, phosphorus concentration, and relative
chemical costs. The process may require more
operator skill and control relative to the stripper tank
operation and lime feeding. In addition, as discussed
in Chapter 4, significant problems have been reported
for lime storage and handling systems.
The sidestream flow diverted to the anaerobic
phosphorus stripper tank is normally 10-30 percent
of the influent flow. The stripper tank also functions
as a gravity sludge thickener. The average solids
detention time (SDT) in the stripper tank can be 5-
20 hours, with 8-12 hours being typical (47,48). The
SDT equals the mass of solids in the sludge blanket
divided by the mass of solids removed per day in the
tank underflow. Soluble phosphorus is released in the
anaerobic stripper tank. It is not known if the
phosphorus release is due to the exact same
mechanism as described for the anaerobic-aerobic
activated sludge contacting sequence. Fermentation
products for the biological phosphorus-removing
organisms may be derived from the metabolism of
hydrolyzed solids and from organics released from
lysed bacteria in the stripper. The released soluble
phosphorus may be from biological phosphorus-
removing microorganisms and from lysed bacteria.
Soluble phosphorus is transferred to the supernatant
either by recycling the stripper underflow to the
stripper influent or by passing an elutriation stream
through the stripper. The elutriation stream may be
primary effluent, secondary effluent, or supernatant
from the lime precipitation reactor.
The overflow from the stripper tank is continuously
fed to the chemical treatment tank where lime is
added for phosphorus removal. Two approaches have
19
-------
Figure 3-4. Commercial biological phosphorus removal processes.
Influent
Aeration
Basin
Effluent
_Direct_Retu_rn_ Sludge |_ > waste AS
Sidestream Feed Sludge
Phosphorus Stripped Return Sludge
Reactor-clarifier for
chemical precipitation
Phosphorus Rich Supernatant
Supernatant Return
Anaerobic \ |
Phosphorus H-
Stripper
Elutnant:
Stripped Sludge Recycle
Primary Effluent
Secondary Effluent
Supernatant Return
Reactor-Clarifier for
Chemical Precipitation
Waste Chemical Sludge
Phostrip Process
Influent f "
Anoxic
t
Aerobic
t
Anoxic
Aerobic
Internal Recycle
_ Return jSludge
Modified Bardenpho Process
Effluent
Waste AS
Influent '
\ '
i
i
Anaerobic
Stages
i
i
i
Oxic Stages
(Aerobic)
^
Return Sludge
A/O Process
Effluent
Waste AS
been proposed for the precipitation and removal of
the chemical sludge. The first, as shown in Figure 3-
4, is to use a separate reactor-clarifier unit for
treatment of the stripper overflow. The second is to
add the lime to the overflow but to settle the chemical
precipitate in the primary clarifier. The separate
treatment is the more common approach. The
underflow solids from the stripper tank are returned to
the aeration tank where biological uptake of
phosphorus occurs. Control of the sidestream feed
rate to the stripper tank affects the distribution
between phosphorus removal by chemical
precipitation or in the waste biological sludge.
A summary of typical recommended design criteria
for the Phostrip, Modified Bardenpho, and A/O
processes is shown in Table 3-2. A significant
design feature for the three processes is the
operating organic loading. The Phostrip process is not
confined to a narrow range of loadings as are the
other two processes and has been recommended for
a wide range of activated sludge operations. This is
20
-------
Table 3-2. Typical Operating Conditions for Biological Phosphorus Removal Processes (22,44,49,50,).
Phostrip
Parameter Value
AS System
F/M, kg TBOD/ --1
kg MLVSS/d
SRT, days2 --1
MLSS, mg/l 600-5,000
HRT.hr3 1-10
Modified
Parameter
F/M, kg TBOD/
kg MLVSS/d
SRT, days2
MLSS, mg/l
HRT, hr3
Anaerobic
Anoxic 1
Nitrification
(Aerobic 1 )
Anoxic 2
Aerobic 2
Bardenpho
Value
0.1-0.2
10-30
2,000-4,000
1-2
2-4
4-12
2-4
0.5-1.0
A/O
Parameter
F/M, kg TBOD/
kg MLVSS/d
SRT, days2
MLSS, mg/l
HRT, hr3
Anaerobic
Aerobic
Value
0.2-0.7
2-6
2,000-4,000
0.5-1.5
1-3
A/O plus
Parameter
F/M, kg TBOD/
kg MLVSS/d
SRT, days2
MLSS, mg/l
HRT, hr3
Anaerobic
Anoxic
Nitrification
Nitrification
Value
0.15-0.25
4-8
3,000-5,000
0.5-1.5
0.5-1.0
3.5-6.0
Phostrip Stripper
Feed,
% of inf. flow
SDT, hr
Sidewater
Depth, m
Elutriation Flow,
% of stripper
feed flow
Underflow,
% of inf. flow
P Release,
g P/g VSS
Reactor-Clarifier
Overflow Rate,
m3/m2/d
PH
Lime Dosage,
mg/l
20-30 Return Sludge, 100 Return Sludge, 25-40 Return Sludge, 20-50
% of inf. flow % of inf. flow % of inf. flow
5-20 Int. Recycle, 400 Int. Recycle, 100-300
% of inf. flow % of inf. flow
6.1
50-100
10-20
0.005-0.02
48
9-9.5
100-300
1 Based on activated sludge system design.
2 Average mass of solids in the system divided by average mass of solids wasted daily.
3 Hydraulic retention time, volume divided by influent flow rate.
due to the fact that Phostrip performance is related
more to the stripper operation and chemical treatment
step. The A/O process is generally designed as a
high-rate activated sludge system, while the
Modified Bardenpho process is generally designed at
relatively low overall loading rates due to the
detention time required for nitrification and
denitrification.
The Modified Bardenpho process, marketed by the
Eimco Process Equipment Company of Salt Lake
City, Utah, is both a nitrogen and a phosphorus
removal system. As Figure 3-4 illustrates, the
influent and return sludge are contacted in an
anaerobic tank to promote fermentation reactions and
phosphorus release prior to passing the mixed liquor
through the four-stage Bardenpho system. The
original development of the four-stage Bardenpho
process was described by Barnard (50) to provide
more than 90 percent nitrogen removal without using
an exogenous carbon source. In the first anoxic
stage, nitrate nitrogen contained in the internal
recycle from the nitrification stage is reduced to
nitrogen gas (denitrification) by metabolizing influent
BOD using nitrate oxygen instead of DO. About 70
percent of the nitrate nitrogen produced in the system
is removed in the first anoxic stage. In the nitrification
(first aerobic) stage, BOD removal, ammonium
nitrogen oxidation, and phosphorus uptake occurs.
The second anoxic stage provides sufficient detention
time for additional denitrification by mixed liquor
endogenous respiration, again using nitrate oxygen
instead of DO. The final aerobic stage provides a
short period of mixed liquor aeration prior to
clarification to minimize anaerobic conditions and
phosphorus release in the secondary clarifier.
21
-------
The resultant Modified Bardenpho design solids
retention time (SRT), based on the solids inventory in
all the aerobic and anoxic stages, is typically 10-20
days depending on the wastewater temperature and
influent nitrogen concentration. In some designs, the
tank volumes have been increased above the
nitrogen removal requirements to provide an extended
SRT of 20-30 days for the purpose of sludge
stabilization. In this way, further sludge digestion is
not included in the facility design. As will be described
in Section 3.3, this design approach results in less
sludge production and phosphorus removal per unit of
influent BOD removed.
The A/O process shown in Figure 3-4 is marketed in
the United States by Air Products and Chemicals, Inc.
(17) and is similar to the Phoredox concept described
by Barnard (12), except that the anaerobic and
aerobic stages are divided into a number of equal
size complete mix compartments. Typically, three
compartments have been used for the anaerobic
stage and four for the aerobic stage. The key features
of the A/O process are its relatively short design SRT
and high design organic loading rates. Compared to
the Modified Bardenpho process, this results in
greater sludge production and more phosphorus
removal per unit of BOD removed in the system.
However, the use of further sludge stabilization
methods, such as anaerobic or aerobic digestion,
must consider the amount of phosphorus released
during stabilization and the effect of recycle streams
from the stabilization units on facility performance.
As shown in Table 3-2, the A/O process can also be
used where nitrification and/or denitnfication are
required. The modified flow scheme incorporates an
anoxic stage for denitrification between the anaerobic
and aerobic stages and is called the A2/0 process.
The anoxic stage is also divided into three equal-
size, complete mix compartments. Mixed liquor is
recycled from the end of the nitrification stage to feed
nitrate nitrogen into the anoxic stage for
denitrification. Internal recycle flows of 100-300
percent have been used. Nitrate nitrogen removals of
40-70 percent can be accomplished this way.
The use of SBR systems for secondary treatment has
gained increased popularity in the United States
during the late 1970s and early 1980s. An evaluation
of SBR treatment capabilities, design aspects, full-
scale installations, and advantages has been
documented for conventional activated sludge
treatment applications (51). Though not a new
treatment concept, with reported operations dating
back to the early 1900s, the recent surge of interest
has been related to new and improved hardware
devices and to the successful EPA-funded, full-
scale, 20-month demonstration and evaluation of a
1,330-m3/d (0.35-mgd) facility at Culver, Indiana
(52). Unique hardware for the system consists of
motorized or pneumatically-actuated valves, level
sensors, automatic timers, microprocessor
controllers, and effluent withdrawal decanters. The
SBR treatment concept and operational flexibility
makes it an obvious candidate for employing
anaerobic-aerobic contacting for biological
phosphorus removal. Biological phosphorus removal
was demonstrated in the full-scale Culver, Indiana
facility during June and July 1984 (53).
A schematic of an SBR operation for biological
phosphorus removal is shown in Figure 3-5. The
SBR system is a fill-and-draw activated sludge
system. A single tank provides for activated sludge
aeration, settling, effluent withdrawal, and sludge
recycle. Biological phosphorus removal was
accomplished in two SBR basins at the Culver facility
that were operating at substantially different average
food-to-microorganism (F/M) loadings of 0.16 and
0.42 kg total BODs (TBOD)/kg MLVSS/d,
respectively. The operation steps consist first of a fill
period where flow is diverted to one of the SBR tanks
while the other tank(s) operates in the reaction, settle,
effluent withdrawal, or idle operation sequences. After
the fill period, the reactor contents are mixed but not
aerated to provide the anaerobic fermentation period
for phosphorus release and uptake of soluble
fermentation products. The next step is the react or
aeration period followed by a selected settling time
when both aeration and mixing are stopped. The
effluent is then withdrawn and, depending on the
influent flow rate, a variable length idle time may
occur. The operating times for this biological
phosphorus removal sequence in the two differently-
loaded SBR basins at Culver, Indiana, are shown in
Table 3-3 (53).
Figure 3-6 shows a further modification of the
Modified Bardenpho process. This modification was
developed at the University of Capetown in South
Africa (30) and has been termed the UCT process.
As shown, the return activated sludge is directed to
the anoxic stage instead of the anaerobic stage as in
the Modified Bardenpho process. The basis for this
development was previous work with biological
phosphorus removal systems that indicated initial
phosphorus removal efficiency could be negatively
affected by nitrate nitrogen entering the anaerobic
stage (16,54,55). Nitrate will serve as an electron
acceptor during the biological oxidation of BOD
entering the anaerobic stage. This results in
competition for the soluble, readily biodegradable
BOD that would normally be converted to
fermentation products for use by the biological
phosphorus-removing bacteria in the anaerobic zone
in the absence of nitrate nitrogen. The relative ratio
between the nitrate nitrogen in the sludge recycled to
the anaerobic stage in a Modified Bardenpho or A/O
process and the available, readily degradable soluble
BOD in the influent to that zone will determine if
sufficient BOD will remain after denitrification
reactions occur to produce a necessary level of
22
-------
Figure 3-5. Biological phosphorus removal using a Sequencing Batch Reactor.
Time
i
Fill
Anaerobic Mix
Aerate
Settle
Withdraw
Table 3-3. SBR Operating Sequence - Biological
Phosphorus Removal.
Period Low Loaded High Loaded
Fill and Anaerobic Mix
Aerate
Settle
Withdraw
Idle
hr
1.8
1.0
1.0
0.4
0.6
hr
3.0
0.4
0.7
0.7
0.0
fermentation products for biological phosphorus
removal. For wastewaters with a relatively high
TKN:BOD ratio, the effect of nitrate nitrogen in the
return sludge on anaerobic zone fermentation may be
significant for these two processes.
In contrast, the anoxic stage of the UCT process is
designed and operated to produce a very low nitrate
nitrogen concentration. The recycle of mixed liquor
from the anoxic stage to the anaerobic stage thereby
provides optimum conditions for conversion of
available soluble BOD to fermentation products. The
mixed liquor recycle from the aerobic stage to the
anoxic stage (recycle 2) can be controlled to assure a
minimal nitrate nitrogen concentration in recycle 1,
while acheiving some level of nitrogen removal in the
anoxic zone. The process has generally been
recommended for wastewaters with influent TKN:COD
ratios of greater than 0.08 or influent COD:TKN ratios
of less than 12.0 for South African applications.
Gerber et al. (56) compared UCT and Modified
Bardenpho process performance in pilot-scale
studies. At a COD:TKN ratio of 9.5, he found no
performance difference.
A modified UCT process is also shown in Figure 3-
6. In this case, the first anoxic zone is designed to
reduce only the nitrate nitrogen in the return activated
sludge. The second anoxic zone is designed for a
much higher quantity of nitrate nitrogen removal as
mixed liquor is recycled to it from the nitrification
zone.
Another approach to accomplish biological
phosphorus removal is to make operational changes
in existing activated sludge systems to create an
anaerobic fermentation zone ahead of the aeration
zone. Figure 3-7 indicates this approach. In practice,
it typically involves turning off air flow or aerators in
the front of the activated sludge basin. As described
in Section 3.1, this technique was demonstrated
during the earlier investigations of phosphorus
removal with the Bardenpho process. The plug flow
plants in the United States, for which high levels of
phosphorus removal were reported, likely had
insufficient aeration at the front end of the aeration
basins and inadvertently promoted the anaerobic-
aerobic contacting sequence. Full-scale U.S. plant
operation modifications that have been shown to
accomplish biological phosphorus removal are the
18,130-m3/d (4.8-mgd) Walt Disney World resort
complex wastewater treatment facility in Orlando,
Florida, and the 13,250-m.3/d (3.5-mgd) DePere,
Wisconsin wastewater treatment facility (47). Both
plants had about half of their original aeration volume
converted to non-aerated zones.
In addition to the designs presented, other
modifications have been proposed that combine
chemical treatment with anaerobic-aerobic staged
activated sludge systems. Alum can be added to
biological phosphorus removal systems as a polishing
step to reduce the total phosphorus concentration to
less than 1 mg/l where insufficient biological
phosphorus removal occurs. Alum is added to the
mixed liquor prior to the secondary clarifier at the
Palmetto Bardenpho facility (49). The addition results
in an effluent total phosphorus concentration of less
than 1 mg/l vs. 2-3 mg/l when no alum is added.
Figure 3-8 shows a combination biological
anaerobic-aerobic system used by Rensink (54) that
also incorporates a stripper for phosphorus removal.
The stripper consisted of a complete mix tank for
anaerobic contacting of a sidestream of return
activated sludge followed by a clarifier for separation
of the stripped sludge. This combination achieved
23
-------
Figure 3-6. UCT process flow schematics.
Recycle 1 Recycle 2
Influent
Anaerobic
Aerobic
_ ReturnJ5ludge
UCT Process
Effluent
Recycle 1
Recycle 2
Influent "^
|
Anaerobic
i
Anoxic
T
Anoxi
r
Aerobic
W Clanfier j-
[ _ r^eturn_SJudge ^
Modified UCT Process
Figure 3-7. Operationally modified activated sludge system for biological phosphorus removal.
Influent ^ ""
Anoxic/
Anaerobic
Return Sludge
f~^
x__^
T
1
1
Effluent
Effluent
Figure 3-8. Combination biological phosphorus removal system.
Anaerobic Aerobic
Influent "
>
L
1
2
3
4
5
1
2
3
4
5
4 L__
Return Sludge j
_^. Waste
Sludge
Clarifier
v
1
1
1 ~ 1
Phosphorus
Rich
Supernatant
Settling
Stripped Return Sludge
Mixed
XX
Stripper
Effluent
24
-------
more than 97 percent total phosphorus removal
compared to 40-50 percent removal for the
anaerobic-aerobic sequence without the stripper.
The activated sludge system was operated with a
relatively low organic loading and nitrification was also
occurring. High nitrate production may have affected
the phosphorus removal efficiency of the anaerobic-
aerobic system without the stripper.
3.3 Performance
An inventory of full-scale biological phosphorus
removal facilities was identified as of April 1984 by
Tetreault et a/. (47). Thirty biological phosphorus
removal facilities were identified as being in operation,
construction, or design with 28 of these being either a
Phostrip, Modified Bardenpho, or A/0 process. At that
time, 11 of the facilities were in operation with five
being Phostrip installations. Three were modified
Bardenpho installations, two were operationally
modified activated sludge systems, and one was an
A/O system.
3.3.1 Phostrip Performance
Table 3-4 summarizes the basic design information
for full-scale Phostrip plants, and Table 3-5
provides performance data summaries for phosphorus
removal by the Phostrip plants. The Seneca Falls,
New York plant was a full-scale demonstration
project, and the performance data are for one month
of intensive plant monitoring. The Phostrip process is
used in the first-stage activated sludge system of a
two-stage nitrification system at the Lansdale,
Pennsylvania plant. The second-stage nitrification
system consists of trickling filters and clarification.
Primary treatment is not provided at this plant, but a
24-hour detention, in-line flow equalization basin is
used. For the June to August 1984 test period, the
Lansdale effluent contained an average of 0.8 mg/l
total phosphorus. This performance illustrates the
ability of the Phostrip process to produce low effluent
phosphorus concentrations with weak wastewaters,
as the influent TBOD averaged only 41 mg/l. Reactor
clarifier overflow from the lime treatment step and
secondary effluent were used as the stripper tank
elutriant sources. The secondary effluent was nitrified
and the stripper overflow contained up to 3.0 mg/l
nitrate nitrogen (47). The orthophosphorus
concentration in the stripper overflow averaged 20.0
mg/l, and the stripper SDT was 20 hours. Lime
scaling occurred in the reactor-clarifier elutriation
return line.
The effluent phosphorus concentrations shown for
Adrian, Michigan, are for samples after the second-
stage nitrification step and filtration. A first-stage
activated sludge system incorporating the Phostrip
process is preceded by primary treatment and in-
line equalization. For the performance data shown in
Table 3-5, the plant influent TBOD averaged only 78
mg/l. The upset problems noted were associated with
mechanical problems with the anaerobic digesters.
During these periods sludge was temporarily stored in
the treatment system and the stripper tank was
overloaded.
The Savage, Maryland plant (also referred to as the
Little Patuxent plant) is a two-stage nitrification
activated sludge system following primary clarification.
The Phostrip process is operated within the high-
rate first-stage activated sludge system. Partial
nitrification occurs in this stage. The nitrification stage
is followed by filtration and chemical polishing to
consistently produce effluent total phosphorus
concentrations below 1.0 mg/l. The effluent
phosphorus concentration values shown in Table 3-
5 are for samples taken after the first-stage
activated sludge system. The improved effluent
phosphorus concentration for April 1985 reflects
operating changes made at the plant. These
consisted of changing the first-stage activated
sludge system from a step feed to a plug flow mode
and reducing the reactor-clarifier elutriation overflow
rate to the stripper tank. Stripper underflow was used
to supplement the reduced reactor-clarifier
elutriation flow. The reduced reactor-clarifier elutriant
flow rate was thought to be responsible for improved
stripper performance, resulting in an overflow
orthophosphorus concentration of 17.6 mg/l in April
1985 compared to 7.2 mg/l for the July 1984
operation.
The Southtowns, New York facility has a relatively low
influent phosphorus concentration, and the Phostrip
system is reportedly not operated continuously when
the effluent phosphorus limit is met by secondary
treatment and filtration alone (59). The Amherst, New
York facility is not operating its Phostrip system, as
two-point ferric chloride addition has been found to
effectively treat a presently lower influent total
phosphorus concentration of 3-4 mg/l. Prior to this
the plant experienced a number of mechanical and
operating problems. Mechanical problems included
scaling in the lime feed line, malfunction of lime feed
pumps, and freezing in the lime slurry line.
Operational problems were encountered when the
activated sludge system MLSS concentration was
increased to provide extended aeration conditions,
which resulted in overloading the stripper tank.
The full-scale Reno-Sparks, Nevada facility used
the Phostrip process for phosphorus removal as part
of a plant expansion design after successful results
with a 23,000-m3/d (6-mgd) plant-scale evaluation
(61). The 113,700-m3/d (30-mgd) expanded facility
employs five parallel anaerobic stripper units.
Elutriation is accomplished by recycling sludge from
the bottom of the stripper to the stripper inlet to
release phosphorus into the stripper supernatant.
During portions of the initial startup phase, the
effluent total phosphorus exceeded 1-3 mg/l. This
was attributed to colloidal carry-over of lime-
25
-------
Table 3-4. Basic Design Information for Full-Scale Phostrip Plants
Seneca Falls, Landsdale,
NY PA Adrian, Ml
Design flow, m3/d
Final eff. TP std., mg/l
Aeration by Oxygen or Air
Aeration mode
Nitrification, 1 - or 2-
stage sec. treatment
Equalization
Final filtration
Sludge handling
Strippers, no.
Reactor-Clanfiers or
Mxer/Flocculators, no
Elutriation source2
3,400
1.0
A
Complete Mix
1
No
No
Thickening,
Anaer. Dig.
1
MF 1
SR
9,500
2.0
A
Plug Flow
2
Yes
No
Thickening,
Vac Fill.
1
RC 1
RC/SEC
26,500
1.0
A
Conv.
2
Yes
Yes
Thickening,
Anaer. Dig.
1
MF 1
PRI
(57).
Savage, MD
56,800
0.31
A
Plug Flow or
Step Feed
2
Yes
Yes
Thickening,
Anaer. Dtg
2
RC 2
RC
Southtowns,
NY
60,600
1.0
O
Plug Flow
1
No
Yes
Filter Press,
Incineration
4
RC4
RC
Amherst, NY
90,900
1.0
O
High Rate
2
Yes
Yes
DAF
Thickening
2
RC 1
RC
Reno-
Sparks, NV
113,700
0.51
A
Plug Flow
1
No
Planned
Anaer Dig.
5
MF 2
SR
1 With final final filtration; chemical polishing available but not utilized
2 Sludge Recycle elutnation; Reactor-Clanfier overflow elutnation, PRInnary effluent supplement; SECondary effluent.
Table 3-5. Performance Data Summary for Full-scale Phostrip Plants.
Total Phosphorus, mg/l
Plant
Seneca Falls, NY
Landsdale, PA
Adrian, Ml
Savage, MD
Southtowns, NY
Amherst, NY
Reno, NV
Design
Flow
m3/d
3,400
9,500
26,500
56,800
60,600
90,900
113,700
Startup
Date
1973
1982
1981
1982
1982
1982
1981
Data
Period
mo.
1
12
11
6
1
1
4
12
4
Influent Averages
mm. mo. ave. mo. max. mo.
6.3
4.0 5.2 6.4
3.4 4 4 5.3
5.7 8 1 9.3
66
7.0
2.3 32 4.1
2.9 5.2 14.3
70-73
Effluent Averages
mm. mo. ave. mo. max. mo.
0.6
0.6 1.2 2.0
<0.1 0.4 0.6
05 1.2 1.7
1.7
0.5
0.3 0.5 0.9
04 1.3 2.5
0.8 1.1
Notes
Full-scale demo
Excludes periods
of upset due to
other plant
problems
July 1984
April 1985
9/82-12/82
Ref.
6
58
59
58
47
47
59
59
60
precipitated solids from a sludge lagoon (58).
Advanced control instrumentation and problems with
the lime feeder also added to startup difficulties. The
method of introducing the recycle sludge to the
stripper tank inlet was modified to prevent the
entrainment of air that would counteract anaerobic
conditions in the stripper. A distributor was devised to
introduce the return sludge at a depth of 1.8 m (6 ft.).
3.3.2 Modified Bardenpho Process Performance
Table 3-6 summarizes basic design parameters for
the first two full-scale North American Modified
Bardenpho facilities operating in the United States at
Palmetto, Florida, and in Canada at Kelowna, British
Columbia, respectively. Both plants are operated with
relatively long detention times to provide sufficient
volume for nitrification and denitrification as well as
biological phosphorus removal. The design SRT for
the Kelowna plant is 1.5-2.0 times the Palmetto
plant due to the need to operate at much colder
wastewater temperatures. Both plants have polishing
filters to meet stringent effluent requirements for BOD
and suspended solids as well as nutrients. Required
effluent limits for the Palmetto plant are equal to or
less than 5, 5, 3, and 1 mg/l for TBOD, total
suspended solids (TSS), total nitrogen, and total
phosphorus, respectively. The allowable limits for
TBOD, TSS, total nitrogen, and total phosphorus for
26
-------
Table 3-6.
Modified Bardenpho Process Full-Scale Plant
Design Summary.
Palmetto, FL
(64)
Kelowna,
Canada
(62,63)
Startup date 10/79 5/82
Flow, rrvVd 5,300 22,700
Detention time, hr (no. cells)
Anaerobic zone i.O(i) 2.0(1)
Anoxic 1 zone 2.7 (1) 4.0 (4)
Nitrification zone 4 7 (1) 9.0 (9)
Anoxic 2 zone 2.2 (1) 40 (4)
Reaeration zone 1.1 (1) 2.0(2)
Total, hr 116 21.0
SRT, days 20 30-40
MLSS, mg/l 3,500 3,000
Temperature, °C 18-25 9-20
Sec. clanfier application rate, 22.3 14.0
m3/m2/d
Polishing filter application rate, 93 7 23.4
m3/m2/d
Primary treatment No Yes
Biological sludge handling Drying Beds DAF Thick.,
Composting
the Kelowna plant are 8, 7, 6, and 2 mg/l,
respectively.
By coincidence, both plants use submerged turbine
devices for aeration. Other Modified Bardenpho plants
are using different aeration devices such as fine
bubble diffusion at Payson, Arizona, and at some
South African plants (57,65) . The Carrousel oxidation
ditch system has been used in the Modified
Bardenpho designs at Fort Meyers and Orange
County, Florida. The Kelowna plant was designed with
submerged turbine mixers and/or aerators in each of
the multiple cells used in the various anaerobic,
anoxic or aerobic zones to provide maximum
operating flexibility. The nitrification zone, for
example, has nine cells with each stage having a
design detention time of 1 hour. The last four cells
can be operated with or without air addition with the
turbine mixers, so that the nitrification detention times
and second anoxic zone detention times can be
varied. The second anoxic zone also can be operated
with air addition to provide additional nitrification time,
if needed.
Both plants shown in Table 3-6 use low-head,
automatic backwash filtration to meet effluent
suspended solids requirements. Both plants were
designed with primary clarification, but the Palmetto
plant modified its operation after startup (16) by
introducing the return activated sludge to the primary
tank to effectively use the tank to provide additional
anaerobic fermentation time. At the Kelowna plant
some of the organic material in the solids removed in
primary treatment can also be used to support the
biological phosphorus removal process. The
supernatant from the primary sludge gravity thickener
can be returned to the anaerobic fermentation zone
or first anoxic zone. This supernatant contains
fermentation products when the solids are held in the
thickener long enough.
Another similarity with the two plants is that they do
not employ either aerobic or anaerobic digestion of
the waste activated sludge. The practice of sludge
digestion has not been recommended for Modified
Bardenpho biological phosphorus removal plants (12)
since the recycle of phosphorus expected to be
released during sludge destruction would overload the
phosphorus removal capacity of the activated sludge
system. At Palmetto, the sludge is wasted from the
sludge recycle line directly to drying beds. The
recycle flow from the drying bed underdrains results
in a minimal additional phosphorus load to the plant
(66). The ultimate disposal of the drying bed sludge is
by land application. At Kelowna, dissolved air flotation
thickening of the waste activated sludge is used to
prevent anaerobic conditions and phosphorus release.
The sludge is further composted prior to ultimate
disposal.
One year of plant performance data (April 1981-
March 1982) is shown for the Palmetto facility in
Table 3-7. The data are monthly average summaries
of influent wastewater and filtered effluent quality
obtained from plant records. Due to the relatively
weak influent wastewater, biological phosphorus
removal capacity was limited. As shown, alum was
added prior to the secondary clarifier during 5 months
of the 1-year period to reduce the orthophosphorus
concentration and to achieve an effluent total
phosphorus concentration generally less than 1 mg/l.
The plant also consistently achieved effluent total
nitrogen concentrations of less than 3 mg/l.
Table 3-8 shows a 2-year performance summary
for the Kelowna plant from January 1983 through
December 1984. During this period, a number of
operating conditions were investigated and the
influent load was at about 54 percent of the design
load. The final filter effluent summary for the 2-year
operation includes results for a two-train operation, a
single-train operation, thickener supernatant feeding,
and splitting of the return sludge between the
fermentation zone and first anoxic zone. The plant
effluent was well within the treatment requirements for
nitrogen and phosphorus removal. During the one-
train operation period (May - November, 1984), the
single train was loaded at about 120 percent of its
design loading and effluent requirements were met.
3.3.3 A/O Process Performance
The first full-scale demonstration of the A/0
biological phosphorus removal process occurred with
the conversion of one-third of the 34,000-m3/d
(9-mgd) Largo, Florida contact stabilization system
27
-------
fable 3-7. Palmetto, FL Modified Bardenpho Process Performance (April 1981
April May June July Aug. Sept. Oct
Influent
Flow, m3/d
TBOD, mg/l
TSS, mg/l
Temperature, °C
TKN, mg/l
NH4-N, mg/l
Total P, mg/l
Ortho P, mg/l
Alkalinity, mg/l
Filtered effluent
TBOD
TSS, mg/l
Total N, mg/l
NO3-N, mg/l
NH4-N, mg/l
Total P, mg/l
Ortho P, mg/l
3,200
164
155
25
31.8
25.0
9.2
6.5
174
2
3
2.1
1.0
0.4
2.5
2.2
3,000
159
157
27
40.8
25.2
6.4
6.1
169
1
2
2.1
1 3
0.3
3.4
1.4
3,500
124
144
29.5
30.1
20.4
7.0
5.3
156
1
2
1.9
1.0
0.3
2.6
2.5
3,600
104
112
30.5
25.0
18.7
5.6
4.5
154
1
2
2.8
1.9
0.2
1.8
1.7
5,500
74
76
29.5
19.7
12.7
4.1
2.8
143
1
2
2.0
1.2
0.2
1.5
1.1
5,900
67
76
29
21.9
12.7
4.9
35
140
1
1
1.7
1 1
0.2
1.2
1.3
3,700
113
116
28
28.1
17.8
6.3
4.7
144
1
2
1.9
1.1
0.2
1 1
0.9
- March 1982 Monthly Averages) (49).
Nov. Dec. Jan. Feb. March
3,300
157
160
27
40.0
22.6
8.5
5.9
171
1
2
2.1
1.3
0.2
0.7
0.7
3,200
182
182
24
38.2
27.2
8.8
5.9
198
1
2
2.5
1.5
0.4
1.6
1.0
3,600
160
141
23
37.7
28.0
8.7
5.4
191
1
1
2.7
1.9
0.3
0.6
0.5
3,700
163
167
23
42.4
25.8
8.0
5.2
201
1
2
2.6
1 8
0.1
0.8
0.7
4,500
150
128
23
32.4
23.7
6.6
4.4
187
1
3
2.8
2.1
0.2
0.9
0.8
' Minimal alum dose applied prior to secondary clarification.
Table 3-8. Kelowna, Canada Modified Bardenpho Process
Performance Results of Cumulative Frequency
Plot of Data (62).
Median Lower 5% Upper 5%
Influent
(1/83 - 12/84)
Flow, m3/d
COD, mg/l
TKN, mg/l
NH4-N, mg/l
Total P, mg/l
Ortho P, mg/l
Final Effluent
(1/83 - 12/84)
TKN, mg/l
NO3-N, mg/l
NH4-N, mg/l
Total P, mg/l
Ortho P, mg/l
Final Effluent - 1 Train
(5/22/84-11/9/84)
Flow, m3/d
NO3-N, mg/l
NH4-N, mg/l
Ortho P, mg/l
12,400
3.3
24.5
17.5
4.5
3.8
1.5
1.8
0.1
0.8
0.77
14,000
2.0
<0.1
1.1
10,400
2.7
19.0
15.0
3.3
3.0
0.2
0.8
<0.1
0.2
0.15
12,000
1.2
<0.1
0.08
10,000
2.6
33.5
21.1
5.8
4.3
1.8
4.2
6.0
1.8
2.25
17,000
3.4
0.75
1.75
to an A/0 system treating primary effluent. The total
average hydraulic detention time was 4.1 hours with
0.9 hour in three anaerobic stages, 0.6 hour in two
anoxic stages, and 2.6 hours in five aerobic stages
(29). The A/O system had its own secondary clarifier.
During nitrification operating conditions, mixed liquor
from the fifth stage of the aerobic zone was recycled
back to the first anoxic stage. The MLSS
concentration and SRT were decreased to provide a
non-nitrifying operation. At that condition, the two
anoxic stages were operated as additional anaerobic
stages and internal recycle of mixed liquor was not
practiced.
Table 3-9 summarizes effluent concentrations
reported for nitrifying and non-nitrifying conditions.
Slightly lower TBOD and higher TSS concentrations
are shown for the nitrification operating period. The
orthophosphorus and total phosphorus effluent
concentrations were also higher. Unfortunately,
insufficient data were presented to allow a
determination of whether the difference in
performance was related to the nitrification operation
or changes in the primary effluent fed to the A/O
system. Data on influent characteristics for another
time period (August 1979-July 1980) indicated that
the TBOD, total phosphorus, and TKN ranges were
81-127 mg/l, 7.0-9.4 mg/l, and 21.4-30.5 mg/l,
respectively (29). The influent SBOD was about 50
percent of the influent TBOD.
28
-------
Table 3-9.
Parameter
Summary of A/O Process Effluent Quality
(Average Monthly and Monthly Range), Largo,
FL (29).
Non-Nitrification
Nitrification
Data period
TBOD, mg/l
TSS, mg/l
Total P, mg/l
Ortho P, mg/l
2/81 - 6/81
7 (6-8)
10(8-13)
1.4 (1.2-1.5)
0.6 (0.5-0.8)
9/81 - 2/82
5 (4-7)
18(10-22)
1.7 (1.3-2.2)
1.0(0.5-2.0)
The phosphorus content measured for the waste
activated sludge was reported to be 4.2-6.0 percent
(17). The sludge handling method proposed for the
Largo facility consisted of dewatering, drying, and
palletizing for use as a soil conditioner/fertilizer.
Table 3-10 describes operating conditions for a
two-train A/0 process at a Pontiac, Michigan EPA
demonstration site. The original activated sludge
system consisted of four plug flow trains with coarse
bubble diffusion. Two of the trains were converted to
the A/O process to allow a performance comparison
of biological phosphorus removal to conventional
activated sludge treatment. The existing tankage was
divided into desired stages by the installation of
wooden baffles. Side-mounted submersible mixers
were installed to provide mixing in the anaerobic
stages after plugging the diffuser lines. The
demonstration project also allowed an evaluation of
the A/O process under cold temperature operation,
during nitrification, and with anaerobic digestion
supernatant return (67).
Table 3-10. Pontiac, Ml A/O System Operating Conditions
(67).
Operating dates
Average flow, m3/d
Average detention time, hr (no. cells)
Anaerobic zone
Aerobic zone
SRT, days
Temperature, °C
Primary treatment
Biological sludge handling
7/13/84 -3/31/85
12,200
1.8(3)
6.7 (4)
16-24
10-17
Yes
Anaerobic Digestion
The treatment performance for different operating
phases is summarized in Table 3-11. Effluent total
phosphorus concentrations averaged less than 1.0
mg/l during the study period, with complete or partial
nitrification. Influent phosphorus concentrations were
relatively low, and a comparison of performance with
a parallel conventional activated sludge system
showed an increased total phosphorus removal of
1.2-1.6 mg/l for the A/0 process operation. Other
factors that could have potentially limited biological
phosphorus removal performance at Pontiac were the
relatively long SRT for an A/O system and the recycle
of nitrate nitrogen in the return sludge to the
anaerobic zone. It appears there was sufficient BOD
in the influent to offset these considerations.
Evaluation of the digester supernatant during this
study indicated a minimal level of soluble phosphorus
was released during anaerobic digestion.
Consequently, the impact of supernatant recycle on
system performance was minimal. One explanation
offered for this was the possibility of the formation of
a magnesium ammonium phosphate precipitate, but
further study was suggested. Average effluent TSS
concentrations from the secondary clarifier ranged
from 6 to 10 mg/l, thereby minimizing the
concentration of particulate phosphorus in the final
effluent.
Table 3-11. Full-scale A/O Process Performance, Pontiac,
Ml (67).
Influent
Influent
Flow, m3/d
TBOD, mg/l
SBOD, mg/l
NH4-N, mg/l
Total P, mg/l
Soluble P, mg/l
Temperature, °C
Reactor
MLSS, mg/l
MLVSS, mg/l
SRT, days
Effluent
TBOD, mg/l
SBOD, mg/l
NH4-N, mg/l
NO3-N, mg/l
Total P, mg/l
Soluble P, mg/l
TSS, mg/l
VSS, mg/l
Phase I
11,300
110
65
15.2
3.2
1.9
17
2,820
1,800
24
6.2
1.8
0.9
10.4
0.8
0.7
6
4
Phase II*
10,800
137
65
17.8
4.1
2.2
16
2,410
1,670
21
9.4
3.0
2.8
11.6
0.7
0.6
7
4
Phase III
12,070
143
87
16.1
3.7
2.2
11
2,340
1,640
19
12.9
2.6
5.9
6.7
0.4
0.3
8
5
Phase IV"
14,680
112
65
18.5
3.0
1.6
10
2,360
1,590
16
12.7
2.0
4.5
8.8
0.7
0.5
10
6
* Anaerobic digester supernatant returned during these phases.
3.3.4 Operationally Modified Activated Sludge
Process Performance
An operationally modified activated sludge system
involves turning off aerators at the front end of the
activated sludge basin to create anaerobic
fermentation conditions to provide the preferred
substrate for the phosphorus-removing bacteria. The
early reports on "luxury uptake" of phosphorus were
for plug flow plants at Baltimore, Maryland; San
Antonio, Texas; and Los Angeles, California. Oxygen
transfer with the diffused aeration systems at the front
end of these activated sludge systems was apparently
sufficiently limited to stimulate anaerobic fermentation
conditions. Unaerated conditions were purposefully
29
-------
created at the front end of activated sludge aeration
basins at DePere, Wisconsin, and at the Reedy Creek
Improvement District main plant in Lake Buena Vista,
Florida. The performance of these operationally
modified plants has been evaluated and the results
reported (47).
The Reedy Creek plant serves the Walt Disney World
resort complex. Operating conditions for the plant are
shown in Table 3-12. The plant uses four parallel
plug flow aeration basins following primary treatment.
The initial third of each basin is unaerated.
Backmixing provides sufficient agitation to maintain
suspension of solids in the unaerated zone.
Nitrification and denitrification also occur in the
system.
Table 3-12. Operating Conditions for Operationally Modified
Activated Sludge Systems (47).
Table 3-13. Average Performance of Operationally Modified
Activated Sludge Systems (47).
Design flow, m3/d
Detention time, hr
Unaerated zone
Aerated zone
Sec. clar. overflow rate,
m3/m2/d
SRT, days
MLSS, mg/l
Return sludge ratio
Primary treatment
Sludge handling
Reedy
Creek
22,700
3.0
6.0
14.7
7.2
2,100
0.59
Yes
DAF Thick.,
Aerobic Dig.,
Land Spread
DePere
53,750
7.5
15.0
17.9
10.6
3,000
0.81
No
DAF Thick.,
Filter Press,
Incineration
The DePere operation involved modifying a contact-
stabilization activated sludge system. The stabilization
tank air supply was stopped, and mixing was
accomplished in the basin with turbine aerator mixers.
A complete mix aerated contact basin was then used
for the aerobic treatment step. The detention times
are given in Table 3-12. The plant was operating at
about 50 percent of its design capacity, resulting in
the relatively long detention times shown.
Table 3-13 indicates that both plants achieved
relatively low effluent phosphorus concentrations
during the three-month summer test period. Effluent
suspended solids concentrations were low enough to
minimize the contribution of paniculate phosphorus to
the final effluent. Nitrification was occurring in both
plants, but apparently the biological phosphorus
removal levels were not limited by the nitrate nitrogen
present in the return activated sludge. This may have
been due to the level of BOD in the influent, warm
wastewater temperatures, and relatively long
unaerated detention times.
Test dates: June - August 1984
Influent
TBOD, mg/l
SBOD, mg/l
Total P, mg/l
Ortho P, mg/l
Effluent
TBOD, mg/l
TSS, mg/l
Total P, mg/l
Ortho P, mg/l
NH4-N, mg/l
Reedy
Creek
155
85
6.7
5.3
3
13
0.9
0.4
0.7
DePere
150
86
5.1
1.9
7
7
0.3
0.1
1.4
3.3.5 Factors Affecting Performance
3.3.5.1 Phostrip Process Solids Detention Time
and Elutriant Quality
In evaluating effects on performance, one must
distinguish between the sidestream operation of the
Phostrip system and the mainstream systems. The
Phostrip system has shown the highest degree of
treatment flexibility and treatment effectiveness with
low organic strength wastewaters because a
substantial amount of the total phosphorus removal
can occur via the stripper and chemical precipitation
operations. For example, effluent total phosphorus
concentrations of less than 1 mg/l were obtained at
Lansdale, Maryland, in spite of an average influent
TBOD of only 41 mg/l and an influent total TBOD:total
phosphorus (TP) ratio of only about 8.1. Under such
conditions the mainstream biological phosphorus
removal processes would be expected to achieve less
efficient phosphorus removal.
The critical design and operating parameters that
affect performance in the Phostrip process are the
stripper SDT, the elutriation rate, and the elutriant
source. The proposed mechanism for biological
phosphorus removal suggests that sufficient SDT is
needed in the stripper to form substrate fermentation
products from lysed bacteria. Successful performance
has been observed for stripper SDT values in the
range shown in Table 3-2. Longer SDT values are
suggested for operations with a significant quantity of
oxidized nitrogen entering the stripper, either via the
return activated sludge stream or the elutriant stream.
For such cases, a 50-percent increase in the
stripper SDT has been recommended (48).
The elutriant source can affect the SDT design of the
stripper and overall performance. The least desirable
elutriant source would be a nitrified secondary effluent
with a significant DO level. Some of the available
substrate in the stripper operation would be needed to
reduce the DO and nitrate oxygen before the
30
-------
necessary organic fermentation activity could occur.
As conditions change in the plant, the SDT may be
adjusted in the stripper by varying the sludge blanket
depth. The chemical treatment system overflow has
been frequently used as a stripper elutriant source
because of its low phosphorus content. Another
potential elutriant source is primary effluent. The
readily available organic material in primary effluent
could result in a lower SDT, since less organic
material is needed from the lyzing of biological solids
in the return activated sludge. This elutriant source
should contain little, if any, DO and no oxidized
nitrogen.
3.3.5.2 Effluent Suspended Solids
An important plant performance consideration for both
the Phostrip and mainstream processes is the
secondary effluent suspended solids concentration
and phosphorus content of those solids. This is even
more critical for the mainstream processes because
they normally produce mixed liquor suspended solids
higher in phosphorus content than the Phostrip
process. Phosphorus contents of MLSS on a dry
solids basis of 2.3-5.8 percent have been reported
for Phostrip and mainstream systems (47), with
values in the lower portion of this range reported for
the Phostrip process.
Many of the nutrient removal facilities, including
Palmetto, Kelowna, and Payson, have required final
filtration to meet very low effluent suspended solids
and BOD limits as well as nutrient removal
requirements. In other cases, low effluent suspended
solids may not be required. In these cases, the
necessity of effluent filtration may be evaluated as a
means to meet the required effluent total phosphorus
concentration. This consideration is illustrated in
Fiqure 3-9. An effluent total phosphorus
concentration requirement of 1.0 mg/l is assumed. If
the effluent soluble phosphorus concentration is 0.5
mg/l and the phosphorus content of the MLSS is 5
percent, the effluent TSS concentration has to be 10
mg/l or less to meet the 1.0-mg/l effluent total
phosphorus limit. If the solids phosphorus content
were 3 or 4 percent, the effluent TSS would have to
be equal to or less than 17 or 12.5 mg/l, respectively.
Thus, unless excellent secondary clarifier
performance is achieved or the effluent soluble
phosphorus concentration is very low (e.g., 0.2 mg/l
or less), a polishing filter would be required to meet
an effluent total phosphorus level of 1.0 mg/l. With
the exception of the Largo nitrification operating
condition, the full-scale plants described in the
performance section were able to produce effluent
TSS concentrations of less than 15 mg/l. The effluent
suspended solids data shown for Palmetto were after
filtration, but effluent TSS concentrations of 4-8 mg/l
from the secondary clarifier have been reported (16).
It appears that a conservatively designed secondary
clarifier could produce effluent suspended solids
concentrations low enough to meet a typical effluent
total phosphorus requirement of 1.0 mg/l, provided
the soluble phosphorus concentration in the effluent
stream does not exceed 0.4-0.6 mg/l.
Figure 3-9. Maximum effluent soluble P concentration for
effluent total P < 1.0 mg/l.
Eff. Sol. P, mg/l
1.0
0.8
0.6
0.4
0.2
2 % P in solids
10 15 20
Effluent TSS, mg/l
25
30
3.3.5.3 Available Organics for Phosphorus
Removal
Effluent soluble phosphorus concentrations as low as
0.1-0.2 mg/l have been achieved in the Modified
Bardenpho, A/O, and operationally modified activated
sludge processes (47,62,68). However, this is not
achieved at all plants and for all operating conditions
because of the dependence on the availability of
fermentation substrate products needed by the
phosphorus-storing bacteria relative to the amount
of phosphorus that must be removed in the system.
In addition, as will be discussed further, the required
ratio of fermentation substrate per unit of phosphorus
removed is affected by the amount of nitrate nitrogen
entering the fermentation zone and also by the SRT
of the system.
Based on the biological phosphorus removal
mechanism described in Section 3.1, a given amount
of fermentation products, such as acetate, consumed
by the phosphorus-removing organisms will yield a
certain quantity of new organisms. A significant
fraction of the dry weight of these organisms will be
phosphorus, and phosphorus removal eventually
occurs in the system by the wasting of these
organisms. Fundamental studies working with pure
cultures of Acinetobacter have determined a
synthesis yield of 0.42 g solids/g acetate and a
phosphorus content of 6-10 percent (69,70). Thus, if
31
-------
the amount of acetate or similar fermentation
products that could be consumed in a biological
phosphorus system were known, the quantity of
phosphorus that could be removed could be
estimated. Unfortunately, the complexity of the
process has thus far prevented the determination of
the amount of fermentation products produced, and
then consumed, by the phosphorus-storing
organisms. The fermentation products used by the
phosphorus-storing organisms will be generated in
the anaerobic zone and some may be present in the
influent of more septic wastewaters. Due to the rapid
assimilation of the fermentation products in the
anaerobic zone, it has not been possible to measure
their production rate (70).
Since the amount of fermentation products produced
in the system can not be measured, other indirect
methods have been proposed in an attempt to
quantify the phosphorus removal potential of a
system. Siebritz et al. (71) recognized that municipal
wastewater is made up of slowly biodegradable
subtrate and a subtrate fraction that is biodegraded
more rapidly. They measured the immediate oxygen
uptake of a mixed liquor upon addition of a
wastewater sample to quantify what they termed the
readily degradable portion of the substrate. They
established a minimum influent concentration of
readily degradable substrate of 25 mg/l for biological
phosphorus removal to proceed. Nicholls et al. (70)
had difficulty relating such measurements to biological
phosphorus removal performance. They proposed an
alternate method using nitrate as the electron
acceptor. In this method, the initial zero-order nitrate
reduction rate after addition of a wastewater sample
to the mixed liquor is measured. The amount of
nitrate nitrogen used during the time that a zero-
order reduction occurs is then related to a readily
biodegradable substrate by a stoichiometric
conversion. The basic concept in both methods is
that the more readily biodegradable substrate may be
the source of the fermentation products in the
anaerobic zone of biological phosphorus removal
systems.
As an alternate to these special tests, Hong et al. (29)
have used the soluble BOD concentration of the
influent wastewater as an indication of the amount of
substrate readily available for the formation of
fermentation products. They have recommended an
influent SBOD: soluble phosphorus (SP) ratio of at
least 15 to produce an effluent soluble phosphorus
concentration below 1.0 mg/l for A/0 systems
operating at F/M loadings above 0.15 kg TBOD/kg
MLVSS/d. Data presented by Tetreault et al. (47)
from the full-scale Largo A/0 system operation
supported this recommendation. At influent SBOD:SP
ratios below 12, effluent soluble phosphorus
concentrations varied from 0.5 to 4.5 mg/l.
Influent SBOD has not been measured at many of the
full-scale biological phosphorus removal plants for a
variety of possible reasons. Such reasons include the
facts that the mechanism of biological phosphorus
removal has only recently begun to be unraveled and
the relative degrees of soluble and particulate BOD
fermentation are not known. It has been recognized
that more phosphorus removals and lower effluent
soluble phosphorus concentrations occur for
wastewaters with higher influent TBOD:TP ratios.
Figure 3-10 summarizes data showing effluent
soluble phosphorus concentrations and TBOD:TP
ratios. Tetreault et al. (47) have recommended a
TBOD:TP ratio of greater than 20-25 to achieve an
effluent soluble phosphorus concentration below 1.0
mg/l.
Figure 3-10. Effluent soluble P concentration vs. influent
TBOO-.TP ratio.
Eff. Sol. P, mg/l
6
• Palmetto (49)
O Kelowna (62)
X Tokyo Pilot-Plant (72)
D Pontiac (67)
A Baltimore Pilot-Plant (68)
A A/O Pilot-Plants (29)
• De Pere (47)
o Reedy Creek (47)
vl
10 20 30 40
Effluent TBOD:TP Ratio
50
60
3.3.5.4 Effect of Solids Retention Time
Fiqure 3-10 shows that higher effluent soluble
phosphorus concentrations occurred for the Modified
Bardenpho facilities operated within the same range
of influent TBOD:TP ratios as other biological
phosphorus removal designs. The Bardenpho
systems were operated, as expected, at longer SRTs
to accomplish nitrification and denitrification. Lower
sludge yields associated with the longer SRTs would
logically decrease the phosphorus removal capacity
for the system.
Assuming a 4.5-percent waste activated sludge
phosphorus content, Barth and Stensel (66)
suggested a TBOD removal:TP removal ratio of 33 at
an SRT of 25 days and a ratio of 25 at an SRT of 8
days. Fukase ef al. (72) found, in an A'O system
32
-------
pilot-plant study treating municipal wastewater, that
the TBOD removal:TP removal ratio increased from
19 to 26 as SRT was increased from 4.3 to 8.0 days.
At the same time, the phosphorus content of the
activated sludge decreased from 5.4 to 3.7 percent.
Maier et al. (73) found in pilot-plant studies that the
rate of phosphorus uptake per unit of mixed liquor
solids decreased by a factor of 2.6 as the F/M loading
was decreased from 0.2 to 0.1 kg TBOD/kg
MLVSS/d. Tracy and Flammino (74) showed that for
identical influent TBOD:TP ratios of 16, the rate of
phosphorus uptake in the aerobic zone decreased by
a factor of 3 as the F/M loading was decreased from
0.44 to 0.24 TBOD/kg MLVSS/d in bench-scale
studies.
These results indicate that operation at longer SRT
values will decrease the efficiency of phosphorus
removal per unit of BOD removed. To maximize
biological phosphorus removal, systems should not
be operated with SRT values in excess of that
required for overall treatment needs. Systems that
require nitrification and denitrification, such as the
Modified Bardenpho system or extended aeration
systems promoting sludge stabilization, will require
much higher influent TBOD:TP ratios to produce
soluble phosphorus concentrations below 1.0 mg/l.
3.3.5.5 Nitrate Nitrogen in the Anaerobic Zone
Barnard (12) was the first to point out that nitrate
nitrogen entering the anaerobic zone of biological
phosphorus removal systems could reduce the
phosphorus removal capability of the system. He
attributed this to an increase in the redox potential of
the reactor and a reduction in the degree of anaerobic
stress to induce phosphorus release. However, an
improved understanding of the phosphorus removal
mechanism indicates that nitrate reduction in the
anaerobic zone utilizes substrate that would otherwise
be available for assimilation by the phosphorus-
storing organisms. Thus, nitrate has the effect of
reducing the net influent BOD/P ratio for the system.
Because of this, variable results have been observed
for systems with nitrate nitrogen present. The degree
of variability depends on the system influent BOD and
phosphorus concentrations and the system SRT. The
return activated sludge recirculation ratio is also
important as this affects the amount of nitrate
nitrogen fed to the anaerobic zone.
Simpkins and McClaren (55) reported a total
phosphorus removal efficiency reduction from 90 to
55 percent when the effluent nitrate nitrogen
concentration increased from 4.0 mg/l to 6.7 mg/l in a
Modified Bardenpho pilot-plant study. During the
Palmetto Modified Bardenpho operation (16), the
internal recycle pumps were stopped, causing the
effluent nitrate nitrogen concentration to increase to
about 10 mg/l and the effluent total phosphorus
concentration to increase from 2.3 to 7.1 mg/l. These
two examples illustrate the sensitivity of phosphorus
removal efficiency to nitrate nitrogen entering the
anaerobic zone of long-SRT biological phosphorus
removal systems treating relatively weak wastewaters.
Similar effects of nitrate nitrogen on biological
phosphorus removal have been reported in an
operationally modified activated sludge system (13).
Vinconneau et al. (75) also showed that nitrate could
significantly affect biological phosphorus removal
performance for a lightly loaded A/O system. At
similar influent BOD:P ratios and operating F/M
loadings, the effluent total phosphorus concentration
decreased from 2.0 to 0.9 mg/l as the effluent nitrate
nitrogen concentration decreased from 3.4 to 0.6
mg/l. The effluent soluble phosphorus conentration at
Pontiac (67) was consistently below 1.0 mg/l even
though the average effluent nitrate nitrogen
concentration ranged from 6.7 to 11.6 mg/l for the
four study periods reported. The low effluent
phosphorus concentration was attributed to a
relatively high influent BOD:P ratio so that excess
BOD was available to reduce the nitrate. The
operationally modified activated sludge systems at
Reedy Creek and De Pere (47) also had a relatively
high influent BOD:P ratio and produced low effluent
soluble phosphorus concentrations in spite of the
occurrence of nitrification. In the Reedy Creek
system, nitrate nitrogen was also fed to the anaerobic
zone by internal circulation of the nitrified mixed liquor
as well as via the return activated sludge. Nitrification
and denitrification were occurring concurrently as
indicated by effluent nitrate nitrogen concentrations of
less than 4 mg/l and ammonium nitrogen
concentrations of less than 1 mg/l.
Rabinowitz (34) studied the effect of nitrate nitrogen
concentration on phosphorus release in batch tests
using activated sludge developed in a UCT system
pilot plant. Sodium acetate was used for the substrate
source. He found that with excess substrate available,
the phosphorus release during anaerobic contacting
was inversely proportional to the amount of nitrate
nitrogen present. He further found that the
denitrification of nitrate in the anaerobic batch tests
had the effect of reducing the availability of substrate
for phosphorus release. The substrate consumption
for denitrification was found to be 3.6 mg COD/mg
nitrate nitrogen reduced. This ratio is in close
agreement with a ratio of 3.53 developed by McCarty
(76) for denitrification using acetate. The ratio in
actual wastewater treatment systems will depend on
the characteristics of the substrate used for
denitrification in the anaerobic zone. With nitrate
nitrogen present, substrate would not necessarily be
converted to volatile fatty acids by fermentation but
could be used for denitrification directly. A substrate
consumption ratio determined from an anoxic-
aerobic pilot-plant system treating domestic
wastewater was about 5.0 mg soluble COD/ mg
nitrate nitrogen reduced for complete denitrification
33
-------
(49). The same reference reported on substrate
consumption for anoxic-aerobic system
denitrification using eleven different industrial
wastewater substrate sources. The mean substrate
consumption ratio was 5.3 mg COD/mg nitrate
nitrogen reduced, with reported values of 2.2-10.2.
Some investigators have also reported on
experiments that suggested biological phosphorus-
removing organisms are capable of denitrification.
Under substrate limiting conditions, phosphorus
uptake and nitrate reduction occurred simultaneously
in an anoxic reactor (15,34). Nitrate reduction was
also observed in a pure culture experiment with
Acinetobacter 21OA (69). The nitrate nitrogen was
reduced only to nitrite nitrogen, which was toxic at
about a 500 mg/l concentration.
3.3.5.6 Wastewater Temperature
The Phostrip, Modified Bardenpho, and A/O
processes have been applied successfully for both
cold and warm wastewater temperature conditions.
Reported data on the operation and performance of
operationally modified activated sludge systems for
biological phosphorus removal during cold wastewater
temperature conditions are limited.
Peirano et al. (77) reported that wastewater
temperature had no significant effect on Phostrip
process efficiency during plant-scale testing at
Reno-Sparks. This is likely the result of having an
adequate size stripper to handle lower activity levels
at cold temperatures. Shapiro et al. (4) showed
specific phosphorus release rates for activated sludge
ranging from 0.63 mg/l-hr/g of volatile suspended
solids at 10°C (50°F) to 3.15 at 30°C (86°F).
Modified Bardenpho systems have been designed at
about twice the total hydraulic detention time for
treatment of 10°C (50°F) wastewater vs. 2Q°C
(68°F) wastewater (63,64). This difference is due to
the effect of temperature on the nitrification-
denitrification design and is not related to the
phosphorus removal design. Prior to the Kelowna
plant design, bench-scale studies showed that 90
percent biological phosphorus removal was possible
over a temperature range from 18°C (64°F) down to
6°C (43°F) (78). The study did exhibit a decreased
nitrogen removal efficiency below 10°C (50°F),
however.
The full-scale A/0 system operation demonstrated at
Pontiac, Michigan, revealed that biological
phosphorus removal was not affected by wastewater
temperatures as low as 10°C (50°F) (67). Biological
phosphorus removal was studied in laboratory batch
units over a temperature range of 5-15°C (41-
60 °F) by Sell et al. (46). The amount of phosphorus
removed at 5°C (41 °F) vs. 15°C (60°F) was greater
by more than 40 percent. The improvement was
credited to a population shift to more slow growing
psychrophihc bacteria with a higher cell yield.
Groenestijn and Deinema (69) reported that the
phosphorus content of a pure culture of Acinetobacter
decreased from 10.1 percent at 5°C (41 °F) to 1.4
percent at 35°C (95°F). An A/0 system operating at
a low organic loading rate of 0.032 kg COD/kg
MLSS/d produced its lowest effluent soluble
phosphorus concentrations of 0.9 mg/l during the
coldest operating month when wastewater
temperature was 5°C (41 °F). The phosphorus
content of the sludge was 4.7 percent compared to a
range of 3.5 to 4.9 percent for five other months (75).
3.3.5.7 pH
Two different laboratory experiments using synthetic
wastewater to evaluate the effect of pH have been
reported. Groenestijn and Deinema (69) studied the
effect of pH at a wastewater temperature of 25 °C
(77°F) on the maximum specific growth rate of a
strain of Acinetobacter. As shown in Figure 3-11,
the maximum specific growth rate was 42 percent
higher at a pH of 8.5 compared to that at a pH of 7.0.
Below a pH of 7.0, a steady decline in the maximum
specific growth rate occurred. Below a pH of 6.0 the
organisms did not grow. Between pH values of 6.5-
8.0, the phosphorus content of the culture remained
constant at about 6.0 percent. It increased to 7.5
percent at a pH of 6.0. Tracy and Flammino (74)
studied the effect of pH on the specific phosphorus
uptake rate (g P/g VSS/hr) in the aerobic phase of an
anoxic-aerobic lab reactor. Their results are also
normalized to a pH of 7.0 in Figure 3-11. They
claimed little difference in the phosphorus uptake rate
from a pH of 6.5 to a pH of 7.0. Below a pH of 6.5,
the phosphorus uptake rate declined steadily. They
further stated that all activity was lost at a pH of 5.2.
As they increased the pH, they claimed that the
phosphorus uptake activity was essentially duplicated.
Nagashima et al. (79) found that total phosphorus
removal in the Modified Bardenpho process was
improved from 42 to 92 percent as the pH was
increased from 5 to 8. These results suggest that the
efficiency of biolgical phosphorus removal may
decline significantly below a pH of 6.5.
3.3.5.8 DO Concentration in the Phosphorus
Uptake Zone
No specific studies have been reported that address
the effect of the DO concentration on biological
phosphorus removal. The biological phosphorus
removal mechanism suggests that the DO
concentration may affect the rate of phosphorus
uptake in the aerobic zone, but not the amount of
phosphorus removal possible, provided that sufficient
aerobic time is available. The mechanism teaches
that the oxidation of stored or exogenous
carbonaceous materials produces energy for the
incorporation of soluble phosphorus into cellular
polyphosphate compounds. In the treatment of an
acetate wastewater in an anaerobic-aerobic fill-
34
-------
Figure 3-11. Reported effects of pH on biological
phosphorus removal.
Ratio of Parameter Rate to the
Same Rate at pH = 7.0
1.6 r
1,2 -
Maximum Specific
Growth Rate (69)
Aerobic Zone Specific
Phosphorus Uptake
Rate (74)
6.5
7.0
pH
7.5
8.8
8.5
and-draw system, Fukase et al. (32) showed that
aerobic detention times required for maximum
phosphorus uptake were 1-2 hours. Tracy and
Flammino (74) showed that 80 minutes was required
for the A/0 process treating a municipal wastewater
and 160 minutes for treating a municipal-food
processing wastewater combination.
Miyamoto-Mills et al. (38) obtained effluent total
phophorus concentrations below 1 mg/l in a Phostrip
system pilot-plant study with the aerobic stage
operating at DO concentrations of either 2.5 or 0.5
mg/l. The lower DO concentration was set to limit
nitrification during the study. At the higher DO
concentration, the system was operated with
nitrification occurring.
Ekama ef al. (80) state that biological phosphorus
removal will be adversely affected in biological
combined nitrogen and phosphorus removal systems
unless the DO concentration in the aerobic zone
remains 1.5-3.0 mg/l. If the DO is too low, they
claim that phosphorus removal may be reduced,
nitrification will be limited, and a poor settling sludge
may be developed. If too high, denitrification
performance could be limited due to the increase in
DO recycled to the first anoxic zone. A resultant
higher nitrate nitrogen concentration could then affect
the phosphorus release performance of the anaerobic
zone.
3.3.5.9 Anaerobic Fermentator Zone
Considerations
The anaerobic zone contact time for Modified
Bardenpho and A/0 systems has ranged from 0.9
hour for the Largo A/O facility to 2.0 hours for
Modified Bardenpho facilities at the Payson and
Kelowna plants. Early full-scale plant investigations
at Palmetto, Florida, found that increasing the
anaerobic detention time from 1.1 to 2.6 hours
increased the percent total phosphorus removal from
59 to 71 percent (16). In a Bardenpho pilot-plant
study, McLaren and Wood (81) found that the
effluent soluble phosphorus concentration decreased
from 3 to less than 1 mg/l as the anaerobic detention
time was doubled from 2 to 4 hours. However, after
establishing removal at the 4-hour detention time, an
effluent soluble phosphorus concentration of less than
1 mg/l was maintained under variable anaerobic
detention times. During investigations using an A/O
pilot plant at the Saint Mars La Jaille, France
wastewater facility, the anaerobic contact zone mixers
were periodically turned off during the day and
improved phosphorus removal was reported (75).
With the mixers off, improved performance was
attributed to a greater SDT. In these cases, it appears
that the longer contact time results in the
fermentation of particulates or materials that are more
slowly converted to fatty acids. The necessity for and
success of longer anaerobic contact times may vary
depending on the strength and nature of the
wastewater. Methods to increase the production of
fatty acids to improve biological phosphorus removal
performance will be discussed in Section 3.6.
Another important aspect of anaerobic contactor
design and performance is to limit the amount of
oxygen entering the zone. Any DO present will
deplete readily-available substrate and thus reduce
the amount of fatty acids that will be produced for
biological phosphorus removal. The presence of
excess DO was identified as causing poor
performance for biological phosphorus removal at a
number of full-scale South African facilities (57).
This was also suspected, in combination with a weak
wastewater, of causing poor phosphorus removal and
filamentous sludge growth during a portion of the
operating period of a U.S. Modified Bardenpho
system (65). Possible sources of high DO input to the
anaerobic zone have included high influent DO
concentrations in the wastewater associated with
infiltration, the use of Archimedes screw pumps for
the return sludge or influent feed, cascading of
wastewater through the influent channel flow
measurement or grit removal systems, and vortices
created by stirrers in the anaerobic basins. Such
conditions should be avoided as much as possible in
the design of biological phosphorus removal systems
using the anaerobic fermentation step.
Anaerobic fermentation zones have been designed as
single-stage, complete mixed basins or three to four
35
-------
basins in series as in the A/O process designs.
Experiments by Ekama et al. (30) support a multiple-
stage design. They claimed improved phosphorus
release and improved phosphorus removal for a
multiple-stage system compared to a single-stage
fermentation reactor operation. They explain this
difference by a model that describes fermentation of
readily-available substrate as a first-order reaction.
3.4 Equipment Requirements
Three major areas of equipment requirements for the
Phostrip system are the stripper tank, the lime feed
system, and the chemical precipitation tank. Piping
and necessary pumping designs are also required to
route a portion of the return sludge to the stripper, to
provide elutriant to the stripper, to transport stripper
underflow sludge to the aeration basin, to convey
stripper overflow to the chemical treatment unit, and
to feed lime to the chemical treatment step.
Stripper tanks are typically sludge thickening tanks
with modifications for sludge inventory control and
elutriation. The tank has a center well for sludge
feeding, a scum baffle and overflow weir, a sludge
rake mechanism, and sludge blanket level indicators.
Underflow solids density probes have also been
recommended by the Phostrip process supplier.
The lime feed system will normally include a lime
storage tank, a slaking operation, and a lime slurry
feed and control system. Mixing of the lime with the
stripper supernatant may be accomplished by using
static in-line mixers or a flash mix chamber with a
mechanical stirrer. The flash mix chamber should
have about a 1-minute detention time.
The chemical treatment units in the Phostrip process
are typically solids contact units. In these units,
relatively large, cone-shaped skirts form mixing
zones in the center of the circular tanks to promote
flocculation. With lime treatment, previously
precipitated solids provide a seed for newly formed
precipitates and floe growth. Heavier floe solids that
fall out of the mixed zone settle out in the clarification
area below and around the skirt. A raking mechanism
is used to move settled solids to a center withdrawal
point.
Equipment requirements for the mainstream biological
phosphorus removal system are minimal and
relatively simple. Mixers are needed to suspend the
mixed liquor solids in the anaerobic and/or anoxic
zones of the various designs. Such mixers have
typically been designed with an energy input per unit
volume of about 10 W/m3 (0.4 hp/1,000 ft3). Lower
values than this may be desirable to minimize
induced air entrainment by the mixers (57). Anti-
vortex baffles are also used for this purpose. Internal
recycle of sludge to anoxic or anaerobic zones is
accomplished with low head, high capacity pumps.
3.5 Design Methodology
3.5.1 Phostrip Process
The major design considerations for the Phostrip
process are the size of the stripper and solids contact
tanks and the lime feed rate. The size of the solids
contact tank will be a function of the stripper tank
supernatant overflow rate. This will be determined by
the return sludge feed rate to the stripper, the degree
of solids thickening achieved, and the elutriation rate
if the elutriant is composed of an outside flow instead
of recycled stripper sludge. The lime feed rate will be
affected by the stripper tank supernatant
characteristics, which impact the ability to raise the
pH for phosphorus precipitation, as well as the
stripper tank supernatant overflow rate. Typical values
for stripper and reactor-clarifier design were given in
Table 3-2.
The stripper design procedure involves the following
steps:
1. Determine or select the amount of return sludge
that will pass through the stripper.
2. Select the stripper underflow sludge concentration.
3. Select the stripper SDT.
4. Based on the above, calculate the volume of
sludge necessary in the stripper.
5. Using a solids flux analysis (82) or appropriate
solids loadings, calculate the stripper area
requirements.
6. Using information from steps 4 and 5, determine
the sludge depth in the stripper.
7. Provide a selected supernatant water depth to
obtain the total stripper sidewater depth. A
supernatant water depth of 1.5 m (5 ft) has been
recommended (77). The stripper depth may be
increased to provide additional sludge inventory
and operating flexibility.
The amount of the return sludge passing through the
stripper is usually selected based on pilot-plant work
or previous full-scale plant operating information.
Another approach, presented by Peirano et. al. (77),
was developed from plant-scale Phostrip
performance testing at Reno-Sparks. The
phosphorus removal efficiency was correlated with
three main operating parameters: the amount of
return sludge passing through the stripper relative to
the plant flow, the stripper SDT, and the stripper
supernatant flow. The correlation developed can be
expressed as follows:
1.85 - [log (100 - E)]/2.11 = (SL x D)1/2 (SU)
(3-1)
36
-------
where,
E = percent phosphorus removal
SL = return sludge passing through stripper tank,
100 Ib dry solids/mil gal of system influent
flow
D = SDT, hr
SU = stripper supernatant flow as ratio of influent
flow
This relationship indicates that phosphorus removal is
affected by the solids loading to the stripper and the
stripper SDT. The following example illustrates the
design procedure.
Wastewater and Plant Design Assumptions:
Influent flow = 10,000 m3/d (2.6 rngd)
Primary effluent TBOD = 120 mg/l
Primary effluent TP = 8 mg/l
Activated sludge recycle flow rate as percentage
of influent flow rate = 80 percent
Activated sludge recycle solids
concentration = 6,000 mg/l
Stripper Design Assumptions:
SDT = 10 hr
Underflow solids concentration = 9,000 mg/l
Return sludge flow rate to stripper
as percentage of influent flow rate = 25 percent
Design Steps:
1. Return sludge flow rate to stripper:
0.25 (10,000 m3/d) = 2,500 m3/d
Recycle sludge passed through stripper:
(0.25/0.8) x 100 = 31 percent (based on total
return sludge flow
rate)
2. Stripper underflow solids
concentration = 9,000 mg/l
3. Stripper SDT = 10 hr
4. Stripper sludge volume produced/day (i.e., stripper
underflow rate):
(2,500 m3/d) (6,000 mg/l -f 9,000 mg/l)
= 1,667m3/d
Net stripper tank sludge volume:
(1,667 m3/d -f 0.8 ) (10 hr) (d/24 hr)= 868 m3
(30,650 ft3)
This net estimate of stripper tank sludge volume
required is based on the stripper underflow solids
rate (or concentration) and an assumed density
factor of 0.8 to account for the possibility of a
lower thickened sludge concentration due to
variations in the stripper operation.
5. Solids loading to stripper:
(6,000 mg/l) (2,500 m3/d) (0.001 kg/mg)
= 15,000 kg/d (33,070 Ib/d)
Assume allowable solids flux rate for 9,000-mg/l
underflow solids concentration = 50 kg/m2/d
Stripper area req'd = 15,000 kg/d -r 50 kg/m2/d
= 300 m2 (3,230 ft2)
Overflow rate = (2500 m3/d * 300 m2)(d/24 hr)
= 0.35 m/hr (205 gpd/ft2)
6. Stripper sludge depth:
868 m3 * 300 m2 = 2.9 m (9.6 ft)
7. Minimum stripper depth:
1.5 m + 2.9 m = 4.4m (14ft.)
[use total stripper depth of 5.5 m (18 ft) for added
inventory flexibility]
8. Supernatant flow assuming primary effluent
elutriation at 50 percent of stripper feed flow:
(2,500 m3/d)(0.50)
+ (2,500 m3/d)[1 - (6,000 mg/l 4 9,000 mg/l)]
= 2,083 m3/d
9. Solids contact unit for lime precipitation:
Assume: Overflow rate = 49 m3/m2/d
Area = 2,083 m3/d •=• 49 m3/m2/d
= 42.5 rn2
Diameter = 7.4 m (24 ft)
Lime feed at dose of 200 mg/l:
(2,083 m3/d) (200 mg/l) (0.001) =417 kg/d
(919 Ib/d)
10. Check phosphorus removal; use phosphorus
release in stripper of 0.01 g P/g VSS (see Table
3-2); assume 70 percent volatile solids:
(15,000 kg/d) (0.70) (0.01 g/g) = 105 kg/d
(236 Ib/d)
Phosphorus removed by stripper supernatant
treatment:
37
-------
(105 kg/d x 2,083 m3/d) -5- [2,500 m3/d + 2,500
m3/d (0.5)] = 58.3 kg/d (126 Ib/d)
11. Determine phosphorus content of waste sludge:
Total phosphorus in influent to activated sludge
system:
(10,000 m3/d)(8 mg/l) (0.001) = 80 kg/d (176 Ib/d)
Assume TP in effluent = 0.5 mg/l
Phosphorus in activated sludge waste solids:
(80 - 58.3) kg/d - 0.5 mg/l (10,000 rr»3/d) (0.001)
= 16.7 kg/d (37 Ib/d)
Assume net sludge yield for biological system
following primary treatment = 0.55 g TSS/g TBOD
removed
TBOD removed = 120 mg/l - 10 mg/l = 110 mg/l
Net sludge produced
= (110 mg/l) (0.55 g/g) (10,000 m3/d) (0.001)
= 605 kg/d (1,334 Ib/d)
P in waste sludge
= (16.7 kg/d -f 605 kg/d) x 100 = 2.8 percent
This is a relatively low sludge phosphorus content
compared to mainstream biological phosphorus
removal systems.
3.5.2 Mainstream Biological Phosphorus Removal
Processes
A variety of process configurations have been
presented for mainstream biological phosphorus
removal. While the aerobic zones may be designed
for different treatment objectives or the internal
recycle and nitrate reduction schemes may be
different, there are common design considerations
that apply to all of these systems. These
considerations include the design of the anaerobic
zone, the need for sufficient time and DO in the
aerobic zone, the denitrification reactor design when
needed, and sludge handling. Another major
consideration is the effluent phosphorus level that can
be achieved and whether chemical addition and/or
effluent filtration is necessary to meet required
treatment levels. In many cases, pilot-plant or
bench-scale studies should be recommended to
determine the final design since treatment
performance is very sensitive to individual wastewater
characteristics.
The anaerobic zone contact time is presently based
on pilot-plant studies or previous experience and
has ranged from 0.9 to 2.0 hours. Staging of the
anaerobic zone should theoretically decrease the
required detention time for the fermentation of soluble
organics. This has to be balanced against the higher
cost for an increased number of mixers and the use
of more divider walls. A DO concentration of greater
than 2.0 mg/l is commonly recommended in the
aerobic zone. The size of the aerobic zone should be
kept as small as is consistent with the overall
treatment objective of maximizing phosphorus
removal. Sufficient aerobic time is needed for
phosphorus uptake.
Waste sludge from biological phosphorus removal
systems is handled in a manner to minimize any
recycle of released phosphorus back to the activated
sludge system. As gravity thickening normally results
in substantial release of phosphorus from the sludge,
dissolved air flotation thickening has been used where
sludge thickening is needed. Provisions may be
required to chemically treat any recycle streams from
digestion of the phosphorus laden sludge. As solids
are lysed and destroyed in aerobic digestion, a
proportional amount of phosphorus is released to the
liquor. The same phenomenon is expected with
anaerobic digestion, but the Pontiac, Michigan study
(67) did not note significant levels of phosphorus in
the digester supernatant. An ammonium-
magnesium-phosphate precipitate may have formed
in the digester, and further studies are needed in this
area.
3.5.2.1 Sludge Production
No significant differences in sludge production for
biological phosphorus removal systems compared to
typical sludge yield values are reported in the
literature for the operating conditions employed.
However, if mixed liquor solids are capable of storing
phosphorus, some increase in net sludge yield should
be expected. To calculate the increase, an estimate
of the mass of the associated chemical constiutents
is needed. This may be approximated from the
constiutents reported in solution during phosphorus
release (Section 3.1.1). Table 3-14 summarizes the
total expected stored mass/unit of phosphorus stored.
The following example illustrates the increase in
sludge mass associated with biological phosphorus
removal:
Assume:
Net solids yield = 0.70 g TSS/g TBOD removed
Normal phosphorus content = 2 percent
Biological phosphorus removal raises the
phosphorus content in the dry solids to 4 percent
Calculate:
Solids yield/100 mg TBOD removed = 70 mg
Normal P removed/100 mg TBOD removed
= 0.02(70) = 1.4mg
38
-------
Table 3-14. Approximated Mass of Phosphorus Storage
Compounds.
function of the net sludge production, the phosphorus
content of the sludge, and the amount of BOD
removed. This is shown as follows:
Constituent
Mg
K
Ca
O
P
Total
Molecular
Weight
24.3
39.1
40
16
31
Mole per
Mole of P
0.28
0.20
0.09
4
1
g/gP
0.22
0.25
0.12
2.06
1.0
3.65
PB = additional P removed by biological
phosphorus removal
(1.4 + PB) -5- (70 + 3.65 PB) = 0.04
PB = 1.64 mg/100 mg TBOD removed
Amount of sludge produced with biological
phosphorus removal/100 mg TBOD removed:
70 + 3.65 (1.64) = 76 mg
Ratio of biological phosphorus process sludge
production to conventional process sludge
production:
76/70 = 1.085
The sludge yield increased by 8.5 percent for this
example. If the phosphorus content of the waste
sludge increased to 5 percent, the net sludge yield
increase is estimated to be 13 percent. Thus, an
increase in the mass of waste solids is expected for
biological phosphorus removal systems. However, the
overall impact on sludge handling may not be
negative due to the excellent sludge thickening and
dewatering characteristics reported for these
systems. Mixed liquuor SVI values generally less than
80 ml/g have been reported for Modified Bardenpho
and A/O systems (12,16,29).
3.5.2.2 Phosphorus Removal Efficiency
In the absence of bench-scale or pilot-plant
studies, estimates of biological phosphorus removal
efficiency can be made to determine if chemical
addition and/or filtration may be necessary to meet
the effluent requirement. The selection of filtration
during the design phase may also be affected by the
assumed secondary clarification efficiency. If an
effluent TSS concentration of less than 10-12 mg/l
must be attained to achieve an effluent total
phosphorus concentration of less than 1 mg/l,
filtration would generally be required.
The amount of phosphorus that may be removed in a
biological phosphorus removal system will be a
(Yn) (Fp) = DP/DBOD
(3-2)
where,
Yn = net solids yield, g TSS/g TBOD
removed
Fp = fraction of P in dry solids, g P g TSS
DP/DBOD = total phosphorus removed/unit of
TBOD removed, g TP/g TBOD
The net solids yield is a function of the system
operating SRT value and influent wastewater
characteristics. The use of primary treatment will also
lower the net solids yield since much of the influent
inert solids will be removed during primary settling.
The fraction of phosphorus in the solids has been
shown to be quite variable depending on influent
wastewater characteristics and operating conditions.
A value for Fp can be selected based on results
reported for other facilities. Figure 3-12 shows the
TBOD:TP removal ratio as a function of SRT and an
assumed Fp value of 0.05 for a system with no
primary treatment (83). The net solids yield as a
function of SRT was taken from a curve given for
municipal wastewaters in WPCF MOP 8 (84). A
system with primary treatment would have a lower Yn
but a higher Fp value since less inert material would
be contained in the mixed liquor solids.
Example:
Assume:
Fp = 0.05 g P/g TSS
Influent TBOD = 160 mg/l
Influent TP = 7.5 mg/l
Influent TBOD:TP = 21.3
Effluent TSS = 12 mg/l
Effluent TBOD = 5 mg/l
No primary treatment
T = 20°C
Calculate:
Design SRT, days 5 10 20
Yn (MOP 8), g/g 0.92 0.81 0.70
DP/DBOD (Eq. 3-2) 0.046 0.041 0.035
DBOD, mg/l 155 155 155
DP, mg/l 7.1 6.4 5.4
Eff. SP
(7.5 - DP), mg/l 0.4 1.1 2.1
Eff. part. P, mg/l 0.6 0.6 0.6
Eff. TP, mg/l 1.0 1.7 2.7
TP removal, percent 87 77 64
The above example shows the effect of SRT on
estimated total phosphorus removal efficiency. If an
39
-------
Figure 3-12. TBOD:TP removal vs. solids retention time
(SRT).
TBOD:TP
35 i—
30
25
20
15
Fp = 0.05 g P/g TSS (assumed)
I
10
SRT, days
15
20
effluent total phosphorus concentration of 1.0 mg/l is
required, the 5-day SRT operation may meet
performance standards without chemical addition or
filtration. The longer SRT operating systems would
require chemical addition to further reduce effluent
soluble phosphorus concentrations. The example is
based on a relatively low influent TBOD:TP ratio
compared to some of the ratios observed in the plant
performance section. However, the ratio used is
within the range of values that can be expected for
some domestic wastewaters (82).
The effect of nitrate nitrogen can be estimated by
calculating the reduction in available BOD due to
dentrification.
Assume:
Effective nitrate nitrogen concentration to
anaerobic zone = 5 mg/l after mixing of recycle
sludge and influent.
TBOD consumed for denitrification:
(4 mg TBOD/mg NOa-N) (5 mg/l) = 20 mg/l
For SRT = 5 days, remaining influent TBOD (i.e.,
TBOD available for biological phosphorus removal)
= 160 mg/l - 20 mg/l = 140 mg/l
DBOD = 140 mg/l - 5 mg/l = 135 mg/l
DP = 0.046 (135 mg/l) = 6.2 mg/l
Effluent SP = 7.5 mg/l - 6.2 mg/l
= 1.3 mg/l vs. 0.4 mg/l previously
Thus, for She low influent TBOD:TP ratio used for this
example, nitrification and the presence of nitrates in
the return sludge could significantly affect the effluent
soluble and total phosphorus concentrations.
3.5.2-3 iJ|t??t.e Mitrogsn Rernov?! Design
Figure 3-13 illustrates the two modes of
denitrification operation used in biological phosphorus
removal systems. Nitrified mixed liquor is recycled to
a pre-denitrification ?one in the Modified Bardenpho
process and also in ihe A/O process when nitrification
occurs. The recycle ratio is generally in the range of
4:1 based on the influent flow. In this zone, the
incoming substrate drives the denitrification reaction
as the facultative organisms use nitrate-released
oxygen as the electron acceptor in lieu of DO. The
oxygen equivalent of the nitrate radical is 2.86 g O2/g
NOa-N. The Modified Bardenpho process has a
second enoxic tank, or post-denitrification zone, in
addition to the pre-denitrification zone. In the second
anoxic zone, the denitrification rate is driven by the
endogenous respiration oxygen demand of the mixed
liquor since the influent substrate is depleted after the
nitrification step.
The design objectives for biological phosphorus
removal systems incorporating denitrification are to
first determine the amount of nitrate nitrogen entering
the pre-denitrification and post-denitrification zones
and (hen to determine the volume of the anoxic
zones. A critical design aspect is the mixed liquor
denitnficfition ie>to occurring in each type of anoxic
zone. A design approach for denitrification will be
bnefiy presented here, since nitrate reduction can be
an important consideration in biological phosphorus
removal systems.
The first step in the design is the preparation of a
mass balance to determine the amount ot influent
nitrogen that will be oxidized to nitrate nitrogen. It is
generally assumed that the distribution of influent
nitrogen is to nitrate nitrogen, effluent ammonium
nitrogen, and solids synthesis:
NO
;.> - NHe - N
syn
(3-3)
NO ™ amount of influent nitrogen converted to
oxidized nitrogen, rng/l
N0 - influent total nitrogen, mg/l
NHe " effluent ammonium nitrogen, mg/l
Npyn - smount of influent nitrogen used in solids
synthe?is, mg/l
The amount of nitrogen used in synthesis can be
estimated from the amount of BOD removed, the net
solids yield as a function of SRT, and the nitrogen
content of Ihe mixed liquor. The nitrogen content of
becteria is 10-12 percent, but a lower value will
f'-p"~- -Uy l-e- me^'Torl 'f"-r mixed liquor solids because
40
-------
Figure 3-13. Pre-denitrification and post-denitrification schemes in biological phosphorus removal systems.
Q,S0,N0 ^
i
t
i
Anaerobic
nii '»
r
Pre-
Denitrification
>•
Anoxic 1
Ne
Nitrification
N
-> —
Return Sludge
(from Sec. Clanfier)
•^ Bardenpho Addition >
Posl-
Denitrification
>
Anoxic 2
V2
Aerobic
>
Or to Clanfier
To Clanfier
of the presence of inerts and non-biological solids.
Values of 5-8 percent may be more appropriate:
Nsyn = Yn (DBOD) Fn
(3-4)
where,
Fn = fraction of nitrogen in mixed liquor solids,
9/9
Once NO is determined, the next step is to perform a
mass balance describing the distribution of the nitrate
produced in the nitrification zone, which results in the
following:
N = NO/(R + r + 1)
(3-5)
where,
N
R
r
= nitrate nitrogen concentration in the
nitrification zone, mg/l
= ratio of internal recycle flow (to the pre-
denitrification zone) to influent flow
= ratio of return sludge flow to influent flow
Equation 3-5 is applicable to both A/0 and Modified
Bardenpho system designs. The rate of nitrate
nitrogen addition to either denitrification zone can be
calculated once the value of N is determined. The
design approach assumes that all of the nitrate
nitrogen entering the anoxic zones is completely
reduced even though a residual nitrate nitrogen
concentration of 0.3-0.5 mg/l may exist. The volume
of the denitrification zones is then determined based
on the amount of nitrate nitrogen entering the zone
and the specific denitrification rate:
Anoxic 1 Volume (applies to both A/0 and Modified
Bardenpho):
(3-6)
(3-7)
V! = RQN/[(X)
Anoxic 2 Volume (Modified Bardenpho)
V2 = [{1 + r) NQ]/[(X) (SDNR2)]
where,
Vi = volume of pre-denitrification zone, m3
\/2 = volume of post-denitrification zone, m3
Q = influent flow, m3/d
X = MLSS concentration, mg/l
SDNRi = specific denitrification rate in pre-
denitrification zone, g NOs-N/g X/d.
SDNR2 = specific denitrification rate in post-
denitrification zone, g NOa-N/g X/d.
The SDNR has been predicted from the specific
oxygen uptake rate as follows:
SDNR = Fd SOUR/2.86
(3-8)
where,
Fd = fraction of substrate reaction rate when
nitrogen-released oxygen is the
electron acceptor vs. when DO is the
electron acceptor, g/g.
SOUR = specific oxygen uptake rate, g 02/g
TSS/d
Previous investigators have found values for F^ of
0.41-0.55 for systems in which the mixed liquor was
subject to both aerobic and anoxic conditions (85-
87). The reduced reaction rate during nitrate
reduction is attributed to the possibility that the entire
biological population cannot use nitrate as an electron
acceptor in the absence of DO, and that the biological
reaction rate may be slower when nitrate is the
electron acceptor.
With this approach, good agreement has been
observed for pre-denitrification SDNR reaction rates
predicted from treatment of a tannery wastewater in
the presence of plentiful substrate (87) and post-
denitrification SDNR reaction rates predicted from
endogenous respiration for domestic wastewater with
limited available substrate (85). The SDNR prediction
for pre-denitrification with domestic wastewater has
not been demonstrated and is limited by unknown
reaction rates for paniculate and soluble BOD. An
41
-------
SDNR relationship based on the F/M loading to the
pre-denitrification zone has been demonstrated for
piiot- and full-scale plant data (64):
SDNR, = 0.03 (F/M)! + 0.029
= QSo/XVt
(3-9)
(3-10)
where,
(F/M)i = food-to-mass loading in pre-
demtrification zone, g TBOD/g MLSS/d
S0 = influent TBOD, mg/l
The (F/M) | value can be calculated for staged or
completely mixed pre-denitrification zone as a
function of the volume (Vi ) selected for the zone.
The SDNR for the post-denitrification zone can be
calculated as follows using an F
-------
Predicted SDNRi = 0.03 (0.57) + 0.029
= 0.046 g NOa-N/g TSS/d
OK - Assumed SDNRi lower than predicted
Try: V! = 800 m3
Assumed SDNRi = 40 •=- 800
= 0.05 g NOa-N/g TSS/d
(F/M)i = 571.4 -5- 800
= 0.71 g TBOD/g MLSS/d
Predicted SDNR! = 0.03(0.71) + 0.029
= 0.05 g NOa-N/g TSS/d
- OK
= (800 m3 * 10,000 m3/d)(24 hr/d)
= 1.92hr
Step 3: Determine \/2 using Equations 3-7 and 3-
12.
V2 = [(1 + 1)(3.5mg/l)(1 0,000 m3/d)]
4- [(3,500 mg/l)(SDNR2)]
= 20/SDNR2
SDNR2 = 0.175 (1.25 g/g^0.70 g/g)(!/20 d)
= 0.0156 g NOa-N/g TSS/d
V2 = 20 4- 0.0156 = 1,280 m3
V2/Q = (1,280m3 -f 1 0,000 m3/d)(24 hr/d)
= 3.1 hr
The design procedure should also check the
TBOD:NO ratio to determine that there is sufficient
TBOD available for the amount of nitrate nitrogen to
be reduced. A ratio of at least 4:1 is recommended.
In this case, there is sufficient TBOD available for
denitrification.
3.6 Process Modifications to Improve
Performance
The major performance limitation of biological
phosphorus removal systems is the amount of volatile
fatty acids available relative to the amount of
phosphorus that must be removed by the
phosphorus-storing microoganisms. As described in
Section 3.3.5.3, wastewaters with lower influent
BOD-to-phosphorus ratios may not produce a
sufficient level of volatile fatty acids (VFAs) in the
fermentation zone to fully trigger the biological
phosphorus removal mechanisms. In some
installations, it has been necessary to add chemicals
to reduce the effluent total phosphorus concentration
to discharge requirement levels. Another means of
improving performance is to increase the amount of
VFAs available to the microorganisms.
It is generally accepted that the major contribution of
VFAs produced in mainstream biological phosphorus
removal systems is from readily-degradable soluble
BOD entering the fermentation zone. For many
wastewaters this may only represent 30-60 percent
of the influent TBOD. Osborn and Nicholls (15),
suspecting that VFAs were important substrates for
biological phosphorus removal, operated a primary
sludge treatment digester at a high loading to
encourage only acid fermentation. The fermentated
sludge was then fed to the anaerobic zone of a
Modified Bardenpho system. The phosphorus removal
was excellent, but the sludge addition increased the
aeration energy requirements of the plant. Eventually
methane fermentation developed in the digester and
phosphorus removal efficiency declined.
Oldham and Stevens (62) presented data showing the
benefits of using primary sludge fermentation
products at the Kelowna Modified Bardenpho facility.
The primary sludge at this facility is directed to a
gravity thickener where it is held long enough for acid
fermentation to develop. The thickened sludge was
passed through a 2.5-mm screen, and the screened
liquid containing fine solids was directed to the plant
fermentation zones. The fermentation stream was
alternately directed to both and one of the two
modules to compare performance with and without
the fermentation product addition. When added to
both modules, the effluent soluble phosphorus
concentration was generally below 0.5 mg/l. During
the alternating operation, the module receiving the
fermenter liquor immediately exhibited a higher
degree of phosphorus release in the fermentation
zone and effluent soluble phosphorus concentrations
well below 0.5 mg/l within about 3 days. The module
not receiving the fermenter liquor produced effluent
soluble phosphorus concentrations generally between
2 and 3 mg/l. During these tests, the VFA
concentration of the thickener liquor was 110-140
mg/l. Since the fermenter liquor flow rate was only
8-10 percent of the influent flow rate, the increased
VFA concentration in the influent was 9-10 mg/l. The
sludge depth of the thickener was later increased to
promote additional solids detention time and VFA
production. The VFA concentration of the fermenter
liquor increased to 200-300 mg/l, but this was
followed by a decline in phosphorus removal
efficiency. The pH of the thickener liquor also
decreased. Barnard (88) postulated that the lower pH
resulted in fermentation products that were not readily
available to the phosphorus-storing microorganisms.
Rabmowitz and Oldham (89) carried out UCT pilot-
plant studies that were also fed settled supernatant
from primary sludge fermentation. The sludge was
fermented in a two-stage, completely mixed reactor
followed by a clarifier for solids separation and sludge
return to the first-stage sludge fermenter. The
settled liquid contained VFAs in the range of 150-
185 mg/l. The phosphorus removal in two sets of
experiments increased by 100 percent and 47 percent
43
-------
after the fermented liquid addition. The VFA
production averaged 0,09 mg/mg COD applied to the
fermenter.
Figure 3-15 shows possible design schemes for
primary sludge fermentation. The first is termed the
activated primary sedimentation tank by Barnard (90).
The recycling of thickened fermented solids serves a
number of purposes. First, it provides mixing of the
newly settled solids with the fermentation organisms.
The acids produced in the thickener are also
elutriated in the primary tank and then directed to the
activated sludge process. In this way, primary
treatment is also used to decrease the loading to the
secondary treatment step and the size of the
activated sludge system. Another advantage claimed
by Barnard is that the pH of the solids in the
fermenter is better buffered, resulting in the
production of the preferred VFAs. The
settler/thickener (deep tank) design can accomplish
the same objectives as the activated primary
sedimentation tank. Deep tank designs have been
used previously to provide both settling and
thickening. Rabinowitz (34) proposed a design similar
to the activated primary sedimentation tank, but the
pumping rate out of the thickener is controlled to
maintain the desired sludge detention time for
fermentation.
3.7 Retrofit Considerations
The sidestream treatment feature of the Phostrip
process makes it readily adaptable for retrofitting
existing facilities. Separate tankage and piping are
added to strip phosphorus from a portion of the return
sludge, return the stripped sludge to the activated
sludge system, and lime treat the supernatant from
the stripper tank. The design features for retrofitting
are thus similar to the design aspects discussed for a
new facility. The source of the elutriant flow and the
organic loading to the activated sludge system are
important design considerations. An elutriant source
that is high in nitrates will require a longer stripper
solids detention time and could negatively impact
stripper performance. Activated sludge systems
operated at longer SRTs will have a less active
sludge, which can impact the stripper detention time
and/or performance. For example, the Phostrip
process has not been applied to systems operating
with an extended aeration sludge age. As discussed
in Chapter 5, additional sludge resulting from lime
treatment of the stripper supernatant must be
considered.
For mainstream retrofit alternatives, the design choice
could be an operationally modified activated sludge
system, an A/O system, or a Modified Bardenpho
system. In all three cases, an anaerobic fermentation
zone must be provided at the head end of the
activated sludge facility. The retrofit design involves
determining the volume requirements of the anaerobic
Figure 3-15. Primary sludge fermentation design schemes.
Influent
I
Primary
Settling
Primary
Effluent
Thickener/
Fermenter
Supernatant
' Waste Sludge
Fermented_ Sludge_Recy_cl_e ^ tto digester)
Activated Primary Sedimentation Scheme
Influent
+
1
T
Primary
Settling
(deep
tank)
k Primary
Effluent
Fermented Sludge Recycle
Waste Sludge
(to digester)
Settlern'hickener Scheme
zones and anoxic zones where used. Some or all of
the additional volume requirements may be available
in the existing plant tankage or they may have to be
added. In the latter case, the hydraulic and physical
arrangements of the specific plant will be a factor in
selecting the most economical modification.
Excess tank volume may be available for retrofitting
for the following reasons:
1. The plant is underloaded, and anticipated future
loading increases are less than originally expected.
2. The better SVI associated with biological
phosphorus removal will allow operation with a
much higher MLSS concentration than the present
operation.
3. The system operating SRT can be reduced from
the original design without a loss in effluent quality.
A lower SRT value has been shown to improve the
performance of biological phosphorus removal
processes.
44
-------
In the case of the A/O process without nitrification,
the retrofit requirement may require additional tank
volume of only 45-minutes detention time. This
small volume may frequently be available in the
existing system, especially when the improved sludge
thickening characteristics are considered.
For all of the mainstream processes, the retrofit
design must consider the processing of the waste
activated sludge and the potential release and recycle
of phosphorus to the activated sludge system. Further
work is needed to investigate the fate of released
phosphorus in anaerobic digesters. Aerobic digestion
will result in a phosphorus release that is proportional
to the sludge mass reduction. Removal of the sludge
for land application is a favored alternative that also
takes advantage of the higher nutrient content of the
waste sludge.
The choice of the retrofit system will depend on
treatment objectives, wastewater characteristics, and
economics. In all cases, the biological phosphorus
removal retrofit design should be compared to the
chemical treatment alternative for phosphorus
removal. Compared to the biological methods,
chemical alternatives will have a higher operating cost
due to chemical addition and increased sludge
handling. As discussed in Section 3.5.2.1, sludge
production from a biological phosphorus removal
system is expected to be only slightly greater than
that from a conventional activated sludge system. The
biological phosphorus removal alternatives may
require a higher initial capital investment for facility
modifications but, in the long term, a net savings may
result from the savings in operating costs.
Retrofit economic comparisons are site specific.
Numerous factors are involved, and it is difficult to
make general statements concerning these
comparisons. As shown at Pontiac, Michigan (67), it
was extremely simple to modify the existing system to
operate in the A/O process mode and the capital cost
was minimal. The high influent BOD:P ratio also
favored the selection of a mainstream biological
phosphorus removal process. On the other hand,
weak wastewaters with a low influent BOD:P ratio
generally favor the selection of a chemical treatment
alternative, the Phostrip process, or perhaps a
mainstream biological phosphorus removal process
coupled with primary sludge fermentation.
Treatment needs will also affect the retrofit process
selection and design. If a high level of nitrogen
removal in addition to BOD and phosphorus removal
are required, the Modified Bardenpho process is a
prime candidate. If a lesser degree of nitrogen
removal is desired along with BOD and phosphorus
removal, the A2/O and UCT processes should also be
considered. If only nitrification and BOD removal are
required, the A/O process with or without an anoxic
zone, the UCT process, an operationally modified
activated sludge system, and the Phostrip process
are all candidates. If BOD removal only is the
objective, all of the above processes with the
exception of the Modified Bardenpho, UCT, and A2/O
processes should be considered.
An operationally modified activated sludge system
may be considered a higher risk alternative for
retrofitting since it generally lacks the well defined
anaerobic-aerobic zones of the UCT, A/O, and
Modified Bardenpho processes. However, with
favorable wastewater characterisitics and a relatively
large anaerobic zone, such systems have been able
to achieve effluent phosphorus concentrations
equivalent to those of the staged systems. The plug
flow systems reporting "luxury uptake" in the early
literature also showed that operationally modified
systems could achieve good phosphorus removal
even though coarse bubble aeration was applied to
the anaerobic fermentation zone.
Many plants can be easily modified to create
fermentation zones. This can be done by turning off
selected aerators, decreasing the air supply to
sparged air headers at the head of the aeration tank,
turning off aerators and adding mixers at the head of
the aeration tank, or by recycling the return activated
sludge though existing primary clarifiers for anaerobic
contacting with raw wastewater. In the latter case, the
entire primary solids inventory would then be directed
to the activated sludge aeration tank. For a modified
operation of this type, existing oxygen transfer and
organic treatment capacities would have to be
carefully evaluated.
Operationally modified activated sludge systems may
also be used with chemical treatment phosphorus
removal systems, where appropriate, to decrease
chemical treatment costs. An advantage of the
operationally modified activated sludge alternative is
that it can usually be easily tested in the existing plant
before final process design decisions are developed.
In summary, biological phosphorus removal process
options generally are easily adaptable for plant
retrofitting. The process choice, design, and
economics, however, will be extremely site specific.
All the design considerations described for new
facilities apply also to retrofit designs.
3.8 Case Histories
3.8.1 Phostrip Process • Little Patuxent,
Maryland
The Little Patuxent (Savage, Maryland) wastewater
treatment plant has a design treatment capacity of
56,800 m3/d (15 mgd) and was started up in 1982.
The plant treatment scheme includes primary
treatment; a first-stage, high-rate activated sludge
process; separate-stage nitrification; and chemical
coagulation/flocculation with filtration for residual
45
-------
phosphorus and suspended solids removal. Primary
and waste biological sludges are aerobically digested
before being gravity thickened with the waste
chemical sludge. The thickened sludge is dewatered
with a belt filter press before final disposal.
The Phostrip process is incorporated within the first-
stage activated sludge system. Partial nitrification also
occurs in the first-stage system. System operation
and performance were studied in 1984 and 1985 as
part of a U.S. EPA-sponsored evaluation of full-
scale biological phosphorus removal installations. The
information given here is from that study (48).
Operating changes were made in the first-stage
activated sludge system in April 1985 to improve
treatment performance. The first-stage activated
sludge operation was changed from step feed to plug
flow to improve phosphorus uptake. During the step-
feed operating mode, the mixed liquor DO
concentration was normally less than 2 mg/l and
phosphorus release occurred in the secondary
clarifiers. During these operating conditions, the
effluent total phosphorus concentration averaged 2
mg/l.
Other changes made concurrently with the switch to a
plug flow operating mode were to increase the mixed
liquor DO concentration to 4 mg/l at the effluent end
of the aeration basin to prevent phosphorus release in
the secondary clarifiers and to increase the return
sludge ratio to decrease the secondary clarifier
blanket level. Other changes were also
simultaneously made in the stripper operation. During
April 1985, the first-stage activated sludge operating
parameters were reported as follows:
F/M loading = 0.5 kg TBOD/kg MLVSS/d
MLSS = 2,000 mg/l
SRT = 4.6 days
HRT = 3.0 hr
Return sludge ratio = 0.57
The elutriation rate to the stripper was decreased in
April 1985 from 121 to 50 percent of the stripper feed
flow rate. Recycled stripper underflow was also used
as an elutriant source in addition to the previously-
used reactor-clarifier overflow. The SDT in the
stripper was maintained at 7 hours. The return sludge
flow to the stripper was increased from 22 to 34
percent of the system influent flow rate.
The orthophosphorous concentration of the stripper
liquor increased from 7.2 to 17.6 mg/l after the above
changes.As shown previously in Table 3-5, the
average monthly effluent phosphorus concentration
from the first-stage activated sludge system
decreased from 1.7 to 0.5 mg/l.
The overflow rate of the reactor-clarifier used for
chemical treatment of the stripper overflow was 34
m.3/d/m2 (840 gpd/ft2). The lime dosage was about
100 mg/l to maintain a pH of 9.5 for phosphorus
precipitation. About 50-60 percent of the total
amount of phosphorus removed in the system was
removed in the stripper operation. The remainder was
removed via the waste activated sludge. The
phosphorus content of the waste solids on a dry
weight basis was about 3.9 percent.
April 1985 treatment performance for the first-stage
activated sludge system is summarized in Table 3-
15. Total phosphorus removal efficiency averaged 94
percent in spite of a relatively weak influent TBOD
concentration that resulted in an influent TBOD:TP
ratio of 13.1. Effluent TBOD and TSS concentrations
were also very good.
Table 3-15. Summary of Phostrip Process Treatment
Performance for Little Patuxent Plant - April
1985.
Flow, m3/d
Influent
Total P, mg/l
1 st-Staqe AS Influent
TBOD, mg/l
Total P, mg/l
Ortho P, mg/l
TBOD:TP
1 st-Staqe AS Effluent
TBOD, mg/l
TSS, mg/l
Total P, mg/l
Ortho P, mg/l
NH4-N, mg/l
Total N
40,900
8.8
92
7.0
4.8
13.1
6
11
0.5
0.1
9.5
21
3.8.2 Modified Bardenpho Process - Kelowna,
Canada
The city of Kelowna, Canada, is located in the
Okanagen Valley in the Province of British Columbia.
Okanagen Lake is generally considered to be in an
oligotrophic state, and, in the late 1960s, algal blooms
on several areas of the lake caused concern. A water
basin study recommended 80 percent total
phosphorus reduction from sewerage system
discharge points. In addition, nitrogen was also
considered limiting in certain areas of the lake.
Several treatment schemes were considered
including land application and advanced physical-
chemical treatment before selecting the Modified
Bardenpho process on the basis of lowest cost.
The existing facility had a design capacity of 11,400
m.3/d (3.0 mgd) and consisted of primary treatment,
46
-------
conventional activated sludge, and trickling filters for
a portion of the flow. The existing inlet works, primary
clarifiers, and secondary clarifiers were incorporated
in the new plant. The upgraded plant design flow is
22,700 m3/d (6.0 mgd), with a maximum hourly flow
of twice the average flow. Raw wastewater flows
through conventional headworks consisting of a
barminnutor, grit chambers, and Parshall flumes for
flow measurement. The three existing primary
clarifiers are now operated in a high-rate mode prior
to the Modified Bardenpho system reactor.
During the design phase, the Modified Bardenpho
process was in its early period of application and an
effort was made to provide maximum operational
flexibility. The process was divided into two modules
with 21 square cells in each module. The design
loadings have been given in Table 3-6 in Section
3.3.2. Following secondary clarification, polishing
filters and chlorination are employed before effluent
discharge. The primary sludge is directed to a gravity
thickener, and the waste activated sludge is thickened
with dissolved air flotation to prevent phosphorus
release. The thickened sludge is combined in a
sludge storage vault prior to disposal by composting.
As discussed in Section 3.6, the thickened sludge
liquor has been used to enhance biological
phosphorus removal performance.
Turbine aerators were selected for the process
reactor. Air is supplied to these turbines by centrifugal
blowers. The aerators are equipped with two-speed
motors so that certain cells can be operated as
anoxic cells (mixing only) or aerated cells. The
maximum design power consumption is about 280 kW
(375 hp) for the blowers and an additional 250 kW
(335 hp) for the turbine aerators, stirrers, and recycle
pumps. DO probes are used to control the air supply.
The startup date for the Kelowna plant was May
1982. The total construction cost for the plant
installation was about $12,000,000 (U.S. $). The
cost reflects some unique geotechnical problems at
the site, which is underlain by sedimentary and deltaic
deposits.
The floors of the new structures were laid in a 3-m
(10-ft) thick layer of soft silt that is contained in sand
and gravel. Since a high water table was also
present, extensive dewatering was required and sheet
piling was used during construction (63).
The treatment performance of the Kelowna plant,
described previously in Section 3.3.2, indicates the
facility is able to meet its discharge requirements of
8, 7, 6 and 2 mg/l for TBOD, TSS, total nitrogen, and
total phosphorus, respectiveley. In Section 3.6, the
benefits of using primary sludge thickener liquor to
enhance fermentation were described, showing that
this operating procedure resulted in much lower
effluent soluble phosphorus concentrations.
3.8.3 A/O Process - Pontiac, Michigan
The process design and performance of the A/O
system installed at the Pontiac, Michigan East
Boulevard Plant have been described in Section
3.3.3. The existing facility provided many advantages
for the investigation of an A/O system in a full-scale
facility. Two of four activated sludge trains were
converted to the A/0 process mode to allow a
comparison of treatment performance with the
conventional system. The A/O system was operated
at cold wastewater temperatures and under nitrifying
conditions, contrary to the Largo, Florida A/O
operation. However, the system was also operated at
a relatively long SRT for an A/O system. Waste
activated sludge was anaerobically digested, and this
provided an opportunity to evaluate potential
phosphorus release back to the digester recycle
stream. The design and performance summaries for
the facility were shown previously in Tables 3-10
and 3-11, respectively.
Significant conclusions from this full-scale evaluation
were that an effluent total phosphorus concentration
of less than 1 mg/l could be achieved without effluent
filtration and that the anaerobic digester supernatant
did not present a significant recycle phosphorus load
to the plant. The latter needs further investigation to
explore why phosphorus was not solubilized during
digestion.
It should be noted that the wastewater characteristics
were very favorable for biological phosphorus
removal, which tended to offset the relatively long
SRT needed for nitrification. The influent TBOD:TP
ratio was generally between 30 and 40.
The project also illustrated that an existing activated
sludge facility could be easily upgraded for biological
phosphorus removal at minimal cost and within a
relatively short time period. The retrofit effort involved
the allocation of 21 percent of the existing aeration
basin to a three-stage anaerobic fermentation zone.
This involved the addition of wooden baffles and
mechanical mixers. The retrofit operation was
completed in 2 months at a cost of approximately
$57,000 (1984 $) for conversion of 13,250 m3/d (3.5
mgd) of treatment capacity to biological phosphorus
removal (67). The existing aeration basin treatment
capacity and air supply were sufficient to
accommodate the plant modification.
3.9 Costs
The costs for biological phosphorus removal
processes are sensitive to wastewater characteristics,
treatment level needs, and existing equipment and
site considerations. Where only phosphorus removal
is required and nitrification is not occurring,
reasonable retrofit treatment alternatives include
chemical addition to existing biological systems and
the Phostrip and A/O processes. If nitrification is
47
-------
occurring or is required, the UCT process and the
A/0 process with an anoxic zone and internal recycle
(i.e., the A2/O process) are candidates as well. An
anoxic zone for partial denitrification is not strictly
needed to achieve nitrification with the A/0 process,
but it is recommended to minimize the amount of
nitrate nitrogen recycled to the anaerobic zone in the
return sludge and its adverse effect on biological
phosphorus release in that zone. For both
phosphorus removal and a high level of nitrogen
removal, the Modified Bardenpho process is a viable
alternative along with a variety of advanced treatment
designs using chemical addition for phosphorus
removal.
An analysis was performed to compare the cost of
biological phosphorus removal to that for chemical
addition to activated sludge for retrofitting existing
facilities (91). Effluent total phosphorus limits of 1.0
and 0.3 mg/l were considered. Nitrogen removal was
not a requirement in the analysis. The analysis
concluded that chemical addition to activated sludge
was more cost-effective to meet a 1.0-mg/l total
phosphorus effluent and was also more cost effective
for meeting an effluent total phosphorus
concentration of 0.3 mg/l for flows of up to 4,500
m3/d (1.2 mgd). The A/0 process was determined to
be more cost effective for flows of 13,600 m3/d (3.6
mgd) or more.
On the other hand, the cost of the A/0 retrofit for
Pontiac, Michigan, was well below these cost
predictions. The Phostrip process was selected for
the Reno-Sparks 150,000-m3/d (40-mgd)
phosphorus removal retrofit after it was estimated that
a total annual cost savings of $500,000 would be
realized compared to chemical addition to activated
sludge (77). It appears, therefore, that the potential
for realizing retrofit cost savings with biological
phosphorus removal will likely be very site specific.
A key economic factor in the above cost analysis and
in other cost analyses is the decision to include
polishing filters to meet effluent phosphorus
concentrations of less than 1.0 mg/l for the A/0, UCT,
and Modified Bardenpho systems. This would also
apply to operationally modified activated sludge
processes. Some chemical addition may be required
in the above processes where unfavorable BOD:P
ratios exist. Cost considerations for external acetate
production may also have to be developed. Previous
plant performance data indicate that effluent filtration
may not always be required. This will be a function of
the influent BOD:P ratio or availability of fermentation
products, the secondary clarifier design, the system
SRT, and other parameters that affect activated
sludge flocculation and clarification properties.
Cost curves for new plants have been presented in a
report entitled Emerging Technology Assessment of
Biological Removal of Phosphorus (48). Tables 3-16
through 3-19 summarize the updated capital and
O&M costs developed for four basic cases. The
updated capital costs are based on an Engineering
News Record Construction Cost Index of 4367 (May
1987). The updated O&M costs are based on an
EPA Escalation Index of 3.83. The four cases are:
Case 1:
Phosphorus removal only with a required effluent total
phosphorus concentration of 1 mg/l. A comparison is
made between a single-stage activated sludge
system with alum addition, a Phostrip system, and an
A/0 system. Effluent filtration is assumed with the
A/0 system.
Case 2.
Same as Case 1 except the required effluent total
phosphorus concentration is 2 mg/l. Without effluent
filtration, in this case, the A/O system is shown to be
most cost-effective.
Case 3:
Same as Case 2 with the addition of nitrification. In
this case, a two-stage nitrification system with alum
addition is assumed for the conventional alternative
and is compared to a single-sludge A/O system. The
two-stage system has a much higher capital cost.
Case 4:
Same as Case 3 with the addition of denitrification to
achieve an effluent total nitrogen concentration of 3
mg/l. In this case, a three-stage activated sludge
system with alum addition is compared to a Modified
Bardenpho system. The three-stage system has
significantly higher capital and operating costs.
In summary, the cost comparisons illustrate that the
biological phosphorus removal alternatives may be
competitive with conventional chemical methods. The
use of effluent filtration is a critical economic factor
and any final cost comparison will be extremely site
specific and affected by wastewater characteristics.
48
-------
Table 3-16. Cost comparison - Case 1: Phosphorus removal (effluent TP = 1 mg/l)
Alternative
1 -stage AS with alum addition
Phostnp
A/O (4-hr detention) with effluent
filters
Table 3-17. Cost comparison -
Alternative
1 -stage AS with alum addition
Phostrip
A/O (4-hr detention)
Table 3-18. Cost comparison -
Alternative
2-stage AS with alum addition
A/O (6-hr detention) for nitrification
and partial denitriflcation to
total N = 10 mg/l
Table 3-19. Cost comparison -
3 mg/l)
Alternative
3-stage AS with alum addition
Modified Bardenpho
Costs-
Capital, $
O&M, $/yr
Total present worth, $
Capital, $
O&M, $/yr
Total present worth, $
Capital, $
O&M, $/yr
Total present worth, $
Case 2: Phosphorus removal
Costs*
Capital, $
O&M, $/yr
Total present worth, $
Capital, $
O&M, $/yr
Total present worth, $
Capital, $
O&M, $/yr
Total present worth, $
Case 3: Phosphorus removal
Costs"
Capital, $
O&M, $/yr
Total present worth, $
Capital, $
O&M, $/yr
Total present worth, $
Case 4: Phosphorus removal
Costs*
Capital, $
O&M, $/yr
Total present worth, $
Capital, $
O&M, $/yr
Total present worth, $
1,890
2,774,000
218,000
4,782,000
3,801,000
273,000
6,315,000
3,370,000
227,000
5,461,000
(effluent TP = 2
1,890
2,762,000
213,000
4,724,000
3,801,000
273,000
6,315,000
2,813,000
197,000
4,627,000
plus nitrification
1,890
3,370,000
245,000
5,627,000
3,142,000
210,000
5,076,000
plus nitrification
1,890
3,869,000
296,000
6,595,000
3,321,000
205,000
5,209,000
Plant Size, nA'd
18,900
10,851,000
868,000
18,846,000
12,602,000
744,000
19,455,000
13,257,000
836,000
20,957,000
mg/l)
Plant Size, m3d
18,900
10,821,000
835,000
18,512,000
12,602,000
744,000
19,455,000
10,819,000
692,000
17,193,000
(effluent TP = 2 mg/l; NH4-N
Plant Size, m3d
18,900
12,820,000
921,000
21,303,000
1 1 ,942,000
764,000
18,979,000
and denitriflcation (effluent TP
Plant Size, m3d
18,900
14,553,000
1 ,200,000
25,605,000
13,553,000
756,000
20,516,000
189,200
55,568,000
5,611,000
107,248,000
59,073,000
3,956,000
95,509,000
63,472,000
4,545,000
105,333,000
189,200
55,350,000
5,276,000
103,944,000
59,073,000
3,956,000
95,509,000
52,314,000
3,820,000
87,498,000
= 1 ma/I)
189,200
63,381,000
5,793,090
116,737,000
59,169,000
4,264,000
98,442,000
= 2 mg/l; TN =
189,200
72,777,000
8,059,000
147,004,000
77,472,000
4,:>52,000
1 19,398,000
* Total present worth calculated assuming a 20-year life and a discount factor of 8-7/8 percent (PWF =- 9.2104)
49
-------
3.10 References
When an NTIS number is cited in a reference, that
reference is available from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
(703) 487-4650
1. Greenburg, A.E., Levin, G. and W.J. Kauffman.
Effect of Phosphorus Removal on the Activated
Sludge Process. Sewage and Industrial Wastes
27: 227, 1955.
2. Srinath, E.G. et al. Rapid Removal of Phosphorus
from Sewage by Activated Sludge. Experientia
(Switzerland), 15:339, 1959.
3. Levin, G.V. and J. Shapiro. Metabolic Uptake of
Phosphorus by Wastewater Organisms. JWPCF
37: 800, 1965.
4. Shapiro, J., Levin, G.V. and Z.G. Humberto.
Anoxically Induced Release of Phosphate in
Wastewater Treatment. JWPCF 39: 1810, 1967.
5. Levin, G.V., Topol, G.J., Tarnay, A.G. and R.B.
Samworth. Pilot Plant Tests of a Phosphate
Removal Process. JWPCF 47(6): 1940, 1972.
6 Levin, G.V., Topol, G.J. and A.G. Tarnay.
Operation of Full Scale Biological Phosphorus
Removal Plant. JWPCF 47(3): 1940, 1975.
7. Vacker, D. et al. Phosphate Removal through
Municipal Wastewater Treatment at San Antonio,
Texas. JWPCF 39:750, 1967.
8. Bargman, R.D. et al. Continuous Studies in the
Removal of Phosphorus by the Activated Sludge
Process. Chem. Engr. Prog. Symp. Ser., 67: 117,
1970.
9. Milbury, W.F. et al. Operation of Conventional
Activated Sludge for Maximum Phosphorus
Removal. JWPCF 43:1890, 1971.
10. Menar, A.B. and D. Jenkins. The Fate of
Phosphorus in Waste Treatment Processes: The
Enhanced Removal of Phosphate by Activated
Sludge. Proceedings of the 24th Purdue Industrial
Waste Conference, Lafayette, Indiana. 1969.
11. Barnard, J.L. Cut P and N Without Chemicals.
Water and Wastes Engineering. 7, 1974.
12. Barnard, J.L. A Review of Biological Phosphorus
Removal in the Activated Sludge Process. Water
SA, 2:136, 1976.
13. Nicholls, H.A. Full Scale Experimentation on the
New Johannesburg Extended Aeration Plants.
Water S.A., 1:121, 1975.
14. Venter, S.L.V. et al. Optimization of the
Johannesburg Olifantsvlei Extended Aeration
Plant for Phosphorus Removal. Prog, in Wat.
Technology 10:279, 1978.
15. Osborn, D.W. and H.A. Nicholls. Optimization of
the Activated Sludge Process for the Biological
Removal of Phosphorus. Int. Conf. on Advanced
Treatment and Reclamation of Wastewater,
Johannesburg, South Aafrica, June 1977.
16. Stensel, H.D. et al. Performance of First U.S. Full
Scale Bardenpho Facility. Proceedings of EPA
International Seminar on Control of Nutrients in
Municipal Wastewater Effluents, San Diego, Calif.,
September 1980.
17. Hong, S.N. et al. A Biological Wastewater
Treatment System for Nutrient Removal.
Presented at the 54th Annual WPCF Conference,
Detroit, Michigan, October 4-9, 1981.
18. Marais, G.vR., Loewanthal, R.E. and I.P. Siebritz.
Review: Observations Supporting Phosphate
Removal by Biological Excess Uptake. Selected
Papers on Activated Sludge Process Research,
University of Capetown, South Africa, April 1982.
19. Stensel, H.D. Fundamental Principles of
Biological Phosphorus Removal. Presented at the
Workshop on Biological Phosphorus Removal in
Municipal Wastewater Treatment, Annapolis,
Maryland, June 22-24, 1982.
20. Fuhs, G.W. and M. Chen. Microbial Basis for
Phosphate Removal in the Activated Sludge
Process for the Treatment of Wastewater. Microb.
Ecol., 2:119, 1975.
21. Deinema, H., Van Loosdrecht, M. and A.
Scholten. Some Physiological Characteristics of
Acinetobacter Spp Accumulating Large Amounts
of Phosphate. Enhanced Biological Phosphorus
Removal from Wastewater, Vol. I. IAWPRC Post
Conference Seminar, p. 154, Sept. 24, 1984,
Paris, France.
22. Buchan, L. The Location and Nature of
Accumulated Phosphorus in Seven Sludges from
Activated Sludge Plants which Exhibited
Enhanced Phosphorus Removal. Water SA, 7:1,
1981.
23. Lawson, E.N. and N.E. Tonhazy. Changes in
Morphology and Phosphate-Uptake Patterns of
50
-------
Acmetobacter Calcoaceticus Strains. Water SA,
6:105, 1980.
24. Letter, L.H. The Role of Bacterial Phosphate
Metabolisms in Enhanced Phosphorus Removal
from the Activated Sludge Process. Enhanced
Biological Phosphorus Removal from Wastewater,
Vol. I., IAWPRC Post Conference Seminar, p.
173, September 24, 1984, Paris, France.
25. Hascoet, M.C., Florent, M. and P. Granger.
Biochemical Aspects of Biological Phosphorus
Removal from Wastewater. Enhanced Biological
Phosphorus Removal from Wastewater, Vol. I.,
IAWPRC Post Conference Seminar, p. 33,
September 24, 1984, Paris, France.
26. Suresh, N., Warburg, M., Timmerman, M., Wells,
J., Coccia, M., Roberts, M.F. and H.O. Halvorson.
New Strategies for the Isolation of
Microorganisms Responsible for Phosphate
Accumulation. Enhanced Biological Phosphorus
Removal from Wastewater, Vol. I., IAWPRC Post
Conference Seminar, p. 131, September 24,
1984, Pans, France.
27. Brodisch, K.E.U. Interaction of Different Groups of
Micro-organisms in Biological Phosphate
Removal. Enhanced Biological Phosphorus
Removal from Wastewater, Vol. I., IAWPRC Post
Conference Seminar, p. 121, September 24,
1984, Pans, France.
28. Lotter, L.H. and M. Murphy. The Identification of
Heterotrophic Bacteria in an Activated Sludge
Plant with Particular Reference to Polyphosphate
Accumulation. Water SA, 11 (4): 172, October
1985.
29. Hong, S.N. et al. A Biological Wastewater
Treatment System for Nutrient Removal.
Presented at EPA Workshop on Biological
Phosphorus Removal in Municipal Wastewater
Treatment, Annapolis, Maryland, June 22-24,
1982.
30. Ekama, G.A., Marais, G.vR. and I.P. Siebritz.
Biological Excess Phosphorus Removal. Chapter
7, Theory, Design, and Operation of Nutrient
Removal Activated Sludge Processes, Water
Research Commission, Pretoria, South Africa,
1984.
31. Rensmk, J.H., Donker, H.J.G.W. and H.P. de
Vries. Biological P Removal in Domestic
Wastewater by the Activated Sludge Process.
Presented at 5th European Sewage and Refuse
Symposium, Munich, West Germany, June 1981,
Procs. 487-502.
32. Fukase, T., Shibeta, M., and X. Mijayi. Studies on
the Mechanism of Biological Phosphorus
Removal. Japan Journal Water Pollution
Research 5, 309, 1982.
33. Arvin, E. Biological Removal of Phosphorus from
Wastewater. CRC Critical Rev. Environmental
Control, 15:25-69, 1985.
34. Rabinowitz, B. The Role of Specific Substrates in
Excess Biological Phosphorus Removal. Ph.D.
Thesis, The University of British Colombia,
Vancouver, British Columbia, Canada, October
1985.
35. Wentzel, M.C., Dold, P.L., Ekama, G.A. and
G.vR. Marais. Kinetics of Biological Phosphorus
Release. Enhanced Biological Phosphorus
Removal from Wastewater, Vol. I., IAWPRC Post
Conference Seminar, p. 89, September 24, 1984,
Paris, France.
36. Jones, P.H., Tadwalker, A. and C.L. Hsu. Studies
in the Enhanced Uptake of Phosphorus by
Activated Sludge: Effect of Substrate Addition.
Proceedings, New Directions and Research in
Waste Treatment and Residuals Management,
The University of British Columbia, Vancouver,
British Columbia, Canada, June 23-28, 1985, p.
324.
37. Comeau, Y., Hall, K.J., Hancock, R.E.W. and
W.K. Oldham. Biochemical Model for Enhanced
Biological Phosphorus Removal. Proceedings,
New Directions and Research in Waste
Treatment and Residuals Management, The
University of British Columbia, Vancouver, B.C.
Canada, June 23-28, 1985, p. 324.
38. Miyamoto-Mills, J. et al. Design and Operation
of a Pilot-Scale Biological Phosphate Removal
Plant at the Central Contra Costa Sanitary
District. Water Science Technology, 15:153,
1983.
39. Arvin, E. and G.H. Kristensen. Exchange of
Organics, Phosphate and Cations Between
Sludge and Water in Biological Phosphorus and
Nitrogen Removal Processes. Enhanced
Biological Phosphorus Removal from Wastewater,
Vol. I., IAWPRC Post Conference Seminar, p.
183, September 24, 1984, Paris, France.
40. Timmerman, M.W. Biological Phosphate Removal
from Domestic Wastewater Using
Anerobic/Aerobic Treatment. Chapter 26 in
Development in Industrial Microbiology, p. 285,
1979.
51
-------
41. Nicholls, H.A. and D.W. Osborn. Bacterial Stress:
Prerequisite for Biological Removal of
Phosphorus. JWPCF 51(3).557, 1979.
42. Deinema, M.H. et al. The Accumulation of
Polyphosphate in Acmetobacter Spp.
Microbiology Letters, Federation of Microbiological
Societies, 273-279, 1980.
43. Senior, P.J. et al. The Role of Oxygen Limitation
in the Formation of Poly B Hydroxybutyrate during
Batch and Continuous Culture of Azotobacter
beijerincku. Biochem, Jour., 128:1193, 1972.
44. Gaudy, A. and E. Gaudy. Microbiology for
Environmental Scientists and Engineers.McGraw
Hill, 1980.
45. Harold, P.M. Inorganic Polyphosphates in Biology:
Structure Metabolism and Function. Bacteriol.
Reviews, 30:772, 1966.
46. Sell, R.L. et al. Low Temperature Biological
Phosphorus Removal. Presented at the 54th
Annual WPCF Conference, Detroit, Michigan,
October 1981.
47. Tetreault, M.J., Benedict, A.H., Kaempfer, C. and
E.F. Barth. Biological Phosphorus Removal - A
Technology Evaluation. JWPCF 58(8):8232-837,
1986.
48. Emerging Technology Assessment of Biological
Removal of Phosphorus. EPA-600/2-85-008,
NTIS No. PB-85-165744, U.S. Environmental
Protection Agency, Cincinnati, Ohio.
49. Nutrient Control. Manual of Practice FD-7.
Facilities Design, Water Pollution Control
Federation, 1983.
50. Barnard, J.C. Biological Denitrification. Journal
International Water Pollution Control Federation,
72:6, 1973.
51. Arora, M.L., Barth, E.F. and M.B. Umphres.
Technology Evaluation of Sequencing Batch
Reators. JWPCF 57:807, 1985.
52. Irvine, R.L. et al. Municipal Application of
Sequencing Batch Treatment at Culver, Indiana.
JWPCF 55:484, 1983.
53. Irvine, R.L. et al. Organic Loading Study of Full-
Scale Sequencing Batch Reators. JWPCF
57:847, 1985.
54. Rensink, J.H., Donker, H.J.G.W. and T.S.J.
Simons. Phosphorus Removal at Low Sludge
Loadings. Enhanced Biological Phosphorus
Removal From Wastewater, Vol. I, IAWPRC Post
Conference Seminar, Sept. 24, 1984, p. 217,
Paris, France.
55. Simpkins, M.J. and A.R. McLaren. Consistent
Biological Phosphate and Nitrate Removal in an
Activated Sludge Plant. Progr. Water Technol.
(G.B.) 10 (5/6):433, 1978.
56. Gerber, A., Simpkins, M.J., Winter, C.T. and J.A.
Scheepers. Biological Nutrient Removal from
Wastewater Effluents: Performance Evaluation of
the Full PHOREDOX and UCT Processes.
Contract Report 52, Council for Scientific and
Industrial Research, Pretoria, South Africa, 1982.
57. Paepcke, B.H. Introduction to Biological
Phosphorus Removal. Proceedings of the
Seminar on Biological Phosphorus Removal in
Municipal Wastewater Treatment, Penticton,
British Columbia, April 17 and 18, 1985.
58. Irvine, R.L., Stensel, H.D. and J.F. Alleman.
Summary Report, Workshop on Biological
Phosphorus Removal in Municipal Wastewater
Treatment, Annapolis, Maryland, June 22-24,
1982. Sponsored by U.S. EPA Municipal
Environmental Research Laboratory, Cincinnati,
Ohio, September 1982.
59. Levin, G.V. Presented at the EPA Workshop on
Biological Phosphorus Removal in Municipal
Wastewater Treatment, San Francisco, Calif.,
1983.
60. Inventory of Full Scale Biological Phosphorus and
Nitrogen Removal Facilities. Report submitted to
the U.S. Environmental Protcetion Agency, Brown
and Caldwell Engineers, March 1984.
61. Peirano, L.E. Low Cost Phosphorus Removal at
Reno-Sparks, Nevada. JWPCF 49:1568, 1977.
62. Oldham, W.K. and G.M. Stevens. Operating
Experiences with the Kelowna Pollution Control
Centre. Proceedings of the Seminar on Biological
Phosphorus Removal in Municipal Wastewater
Treatment, Penticton, British Columbia, Canada,
April 17 and 18, 1985.
63. Leslie, P.J. Design of the Kelowna Pollution
Control Centre. Proceedings of the Seminar on
Biological Phosphorus Removal in Municipal
Wastewater Treatment, Penticton, British
Columbia, Canada, April 17 and 18, 1985.
64. Burdick, C.R., Refling, D.R. and H.D. Stensel.
Advanced Biological Treatment to Achieve
Nutrient Control. JWPCF 54(7):1078, 1982.
65. Stensel, H.D. et al. Evaluation of Nutrient
Removal at Payson, Arizona. Proceedings, New
52
-------
Directions and Research in Waste Treatment and
Residuals Management, The University of British
Columbia, Vancouver, B.C., Canada, June 23-
28, 1985, p. 476.
66. Barth, E.F. and H.D. Stensel. Internationa!
Nutrient Control Technology for Municipal
Effluents. JWPCF 53(12):1691, 1981.
67. Kang, S.J. et al. A Year's Low Temperature
Operation in Michigan of the A/0 System for
Nutrient Removal. Presented at the 58th Annual
Water Pollution Control Federation Conference,
Kansas City, Missouri, October 1985.
68. Deakyne, C.W., Patel, M.A. and D.J. Krichten.
Pilot Plant Demonstration of Biological
Phosphorus Removal. JWPCF 56(7):867, 1984.
69. Groenestijn, J.W. and M.H. Demema. Effects of
Cultural Conditions on Phosphate Accumulation
and Release by Acinetobacter Strain 210A.
Proceedings of the International Conference,
Management Strategies for Phosphorus in the
Environment, Lisbon, Portugal, July 1-4, 1985.
70. Nicholls, H.A., Pitman, A.R. and D.W. Osborn.
The Readily Biodegradable Fraction of Sewage;
Its Influence on Phosphorus Removal and
Measurement. Enhanced Biological Phosphorus
Removal from Wastewater, Vol. I, IAWPRC Post
Conference Seminar, September 24, 1984, Paris,
France, 105.
71. Siebntz, I.P., Ekama, G.A. and G.vR. Marais.
Biological Phosphorus Removal in the Activated
Sludge Process. Research Report W46,
Department of Civil Engineering, University of
Capetown, South Africa, 1983.
72. Fukase, T., Shibata, M. and V. Mayaji. Factors
Affecting Biological Removal of Phosphorus.
Enhanced Biological Phosphorus Removal from
Wastewater-, Vol. I, IAWPRC Post Conference
Seminar, Sept. 24, 1984, Pans, France, p. 239.
73. Maier, W. et al. Pilot Scale Studies on Enhanced
Phosphorus Removal in a Single Stage Activated
Sludge Treatment Plant. Enhanced Biological
Phosphorus Removal from Wastewater, Vol. II
IAWPRC Post Conference Seminar, S^pL 24,
1984, Pans, Fiance, p. 51.
74. Tracy, K.D. and A. Flammmo. Kinetics of
Biological Phosphorus Removal. Presented at the
58th Annual Water Pollution Control Federation
Conference, Kansas Citv, Missouri, October
1985.
75. Vinconneau, J.O., Haseoet, M.C. and M. Horentz.
The First Applications of Biological Phosphorus
Removal in France. Proceedings of the
International Conference, Management Strategies
for Phosphorus in the Environment, Lisbon,
Portugal, July 1-4, 1985.
76. McCarty, P.L., Beck, L. and P. St. Arnant.
Biological Denitrification of Wastewaters by
Addition of Organic Materials. Proceedings 24th
Industrial Waste Conference, Purdue Univ., West
Lafayette, Indiana, 1969.
77. Peirano, L.E. et al. Full Scale Experiences with
the Phostrip Process. Presented at the Workshop
on Biological Phosphorus Removal in Municipal
Wastewater Treatment, Annapolis, Maryland,
June 22-24, 1982.
78. Oldham, W.R. and H.P. Dew. Cold Temperature
Operation of the Bardenpho Process. Presented
at the 14th Canadian Symposium on Water
Pollution Research, 1979.
79. Nagashima, M. et al. A Nitrification/Denitrification
Recycling System for Nitrogen and Phosphorus
Removal from Fermentation Wastewater.
Fermentation Technology, 57:2, 1979.
80. Ekama, G.A., Siebritz, I.P. and G.vR. Marais.
Considerations in the Process Design of Nutrient
Removal Activated Sludge Processes. Selected
Papers on Activated Sludge Process Research at
the University of Capetown, Capetown, South
Africa, April 1982.
81. McLaren, A.R. and R.J. Wood. Effective
Phosphorus Removal from Sewage by Biological
Means. Water SA, 2, 47, 1976.
82. Wastewater Engineering: Treatment, Disposal,
Reuse. Metcalf & Eddy, Inc. McGraw-Hill, New
York, New York, 1979.
83. Stensel, H.D. Design of Biological Nutrient
Removal Systems by Graphical Procedures. Civil
Engineering for Practicing and Design Engineers
2(1), 1982.
84. Wastewater Treatment Plant Design. Water
Pollution Control Federation Manual of Practice
MOP 8, Washington, DC, 1977.
85. Stensel, H.D. Biological Nitrogen Removal
System Design. Water 70(209), 1980.
86. Wuhrman, K. Nitrogen Relationships in Biological
Treatment Processes - III. Denitrification in the
Modified Activated Sludge Process. Water
Research 3:177,1969.
87. Panzer, C.C. Substrate Utilization Approach for
Design of Nitrogen Control. Journal of the
53
-------
Environmental Engineering Division, ASCE, 110
(2):369, April 1984.
88. Barnard, J.L. The Role of Full Scale Research in
Biological Phosphate Removal. Proceedings,
New Directions and Research in Waste
Treatment and Residulals Management, The
University of British Columbia, Vancouver, B.C.,
Canada, June 23-28, 1985.
89. Rabinowizx, B. and W.K. Oldham. The use of
Primary Sludge Fermentation in the Enhanced
Biological Phosphorus Removal Process.
Proceedings, New Directions and Research in
Waste Treatment and Residuals Management,
The University of British Columbia, Vancouver,
B.C., Canada, June 23-28, 1985.
90. Barnard, J.L., Activated Primary Tanks for
Phosphate Removal. Water SA, 10 (3):July 1984.
91. Nutt, S.G. and N.W. Schmidtke. Technical and
Economical Feasibility of Retrofitting Existing
Municipal Wastewater Treatment Plants in
Canada for Biological Phosphorus Removal.
Proceedings of the Seminar on Biological
Phosphorus Removal in Municipal Wastewater
Treatment, Penticton, British Columbia, Canada,
April 17 and 18, 1985.
54
-------
Chapter 4
Phosphorus Removal by Chemical Addition
4.1 Introduction and Theory
Many wastewater treatment plants that are required to
remove phosphorus do so by adding chemicals to
precipitate the phosphate present in the wastewater.
As described in Section 4.2, chemicals may be added
to primary, secondary, or tertiary processes, or at
multiple locations in the plant. Chemicals used for
phosphorus precipitation include metal salts such as
ferric chloride and aluminum sulfate (alum), and lime.
This chapter: 1) describes the characteristics of
chemicals used for phosphorus removal and the
reactions which occur during phosphorus
precipitation, 2) discusses the alternative points for
chemical addition, 3) assesses the performance of
chemical addition systems in removing phosphorus,
and 4) provides design procedures for chemical
storage and feed facilities.
A variety of metal salts are used for removal of
phosphorus from municipal wastewater. The most
common chemicals are aluminum sulfate (alum) and
ferric chloride. Ferrous sulfate and ferrous chloride
solutions, which are available as byproducts of
steelmaking operations (pickle liquor) are also used.
Sodium aluminate addition is sometimes practiced at
facilities with low alkalinity wastewaters. Two other
chemical compounds that have been investigated for
phosphorus removal are aluminum chlorohydrate and
polyaluminum chloride. In many cases, anionic
polymers are used in addition to the mineral salt to
assist in solids separation.
Mineral salts are by far the most common chemicals
used for phosphorus removal. A number of plants
originally designed for use of lime are currently using
alum or ferric chloride. In a 1979 survey of 104 plants
removing phosphorus in the lower Great Lakes basin,
53 facilities used iron salts, 49 used aluminum salts,
and only 2 used lime (1).
The reasons for infrequent use of lime for phosphorus
removal include: 1) the substantial increase in the
mass of sludge to be handled compared to that from
use of metal salts, and 2) the operation and
maintenance problems associated with the handling,
storage, and feeding of lime. Due to the fact that
aluminum and iron salts have all but replaced lime as
a phosphorus precipitant, a detailed discussion of
lime use is not presented in this Manual. Detailed
information and design criteria for lime addition
facilities are contained in References 2 through 5.
The reactions between phosphorus and metal salts
are complex. For purposes of this discussion, it is
assumed that the primary mechanism of phosphorus
removal is interaction of the metal ion with
orthophosphate to form an insoluble precipitate. The
reactions presented below are for illustrative purposes
and may not represent the true mechanisms which
take place due to the variations in wastewater
characteristics and forms of phosphorus present. It is
recommended that any engineering examination of
chemical addition for phosphorus removal include a
jar test of the actual wastewater of concern (see
Section 4.5.2). This will avoid the common error of
assuming a required dosage when actual dosage can
vary substantially between facilities (see Section 4-
3) and at a given facility at different times of day and
season of the year (see Section 4.5).
4.1.1 Aluminum Compounds
Aluminum ions combine with phosphate ions to form
aluminum phosphate, as shown by:
AI3
P043~ -» AIPO4
(4-1)
On a mole basis, 1 mole of Al will react with 1 mole
of P04, or 1 mole of P. On a weight basis, 27 g of Al
will react with 95 g of P04 (or 31 g as P) to form 122
g of AIPO4. The AI:P weight ratio is thus 27 Al to 31
P or 0.87:1 (2).
The most common form of aluminum in use for
phosphorus precipitation is "alum" or "filter alum," a
hydrated aluminum sulfate with the approximate
formula Al2(S04)3»14H2O. Alum contains about 9.1
percent soluble aluminum as Al and 17 percent
soluble aluminum as
The reaction of alum with phosphate can be
described by:
AI2(S04)3»1 4H2O + 2P043--H.2AIP04 i + 3S042" + 1 4H20
(4-2)
55
-------
One mole (594 g) of alum will react with 2 moles (190
g) of phosphate containing 62 g phosphorus to form 2
moles (244 g) of AIPO4. Thus, the weight ratio of
alum to phosphorus is 594 to 62 or 9.6:1.
In practice, the quantities of alum required are higher
than the stoichiometry would predict. This is due to
competing reactions, which vary with the wastewater.
Among the most notable factors that affect the actual
quantity of alum required to attain a specific P
concentration are: the alkalinity and final pH of the
wastewater; ionic constituents such as sulfate,
flouride, sodium, etc.; quantity and nature of
suspended solids, e.g., kaolin vs. montmorillomite
clays; microorganisms, and other colloidal species;
the actual ratio of Al to P; and the intensity of mixing
and other physical conditions extant in the treatment
facilities. For the purposes of engineering design the
following stoichiometric reaction is provided:
AI2(SO4)3»14H20 + 6HC03--»2AI(OH)3 i + 6CO2 + 14H2O
+ 3S043-
(4-3)
The optimum pH for phosphorus removal using alum
is in the range of 5.5-6.5. The extent of pH
depression resulting from alum addition will depend
on the alkalinity of the wastewater and the alum
dosage. In unusual cases in which the buffering
capacity (alkalinity) of the wastewater is very low,
addition of alkaline chemicals may be required to
offset the pH depression resulting from alum addition.
Although strong acids could be used to lower the
wastewater pH to the optimum point, it may be
simpler to use a higher alum dosage to depress the
pH. The relative economics of the two approaches
should be evaluated by the engineer during the
planning process.
Alum can be purchased as dry alum in bags, drums,
or in bulk, or as liquid alum in tank cars or trucks.
Characteristics of alum are given in Table 4-1 (3,4).
Sodium aluminate is sometimes used for phosphorus
precipitation. The chemical formula for sodium
aluminate is Na2Al204 or NaAI02- The commercial
granular trihydrate is written as Na2O»Al203»3H2O.
The reaction between sodium aluminate and
phosphate may be expressed as (6):
Na2O«AI203 + 2P043- -» 2AIP04 i
2NaOH + 60H"
(4-4)
Note the presence of NaOH as a product of the
reaction, which will tend to increase pH rather than
lower it. This allows sodium aluminate to be used with
low alkalinity wastewaters in which use of alum would
cause excessive depression of pH.
The mole ratio of Al to P is 1:1. The weight ratio of
AI:P is 0.87:1, while the weight ratio of sodium
aluminate to phosphorus is approximately 3.6:1.
Characteristics of sodium aluminate are given in
Table 4-1 (3,4).
Aluminum chlorohydrate and polyaluminum chloride
are other potentially useful chemicals for phosphorus
precipitation. In jar test studies with raw wastewater
samples from four treatment plants, it was found that
polyaluminum chloride was superior to aluminum
sulfate in removing total phosphate, while aluminum
chlorohydrate gave poorer results than the other two
chemicals (7). If available at a cost competitive with
the more common chemicals used for phosphorus
precipitation, further investigation of their use may be
justified.
A "sewage grade" granular alum has been used in
Scandanavia for 15-20 years and has been tested at
several wastewater treatment plants in the United
States. This material is a mixture of aluminum and
iron sulfates, containing approximately 13.7 percent
aluminum as AI2O3 and 4.3 percent iron as Fe2C-3.
The wastewater treatment plant serving Geneva, New
York is currently using this chemical, which was
imported from Sweden. However, its distribution in
the United States has been discontinued.
4.7.2 Iron Compounds
Iron salts are commonly used in the precipitation of
phosphorus from municipal wastewater. Both ferrous
(Fe2 + ) and ferric (Fe3 + ) ions can be used in the
form of ferric chloride, ferrous chloride, ferric sulfate,
and ferrous sulfate. Ferrous chloride and ferrous
sulfate are also available as byproducts of
steelmaking operations (waste pickle liquor), although
these solutions may contain large quantities of free
hydrochloric or sulfuric acid which can cause
destruction of alkalinity and pH depression.
Characteristics of iron salts used for phosphorus
precipitation are given in Table 4-1 (3,4).
A typical reaction between ferric chloride and
phosphate can be approximated by (8):
FeCI3 + PO43- -» FePO4 i + 3CI'
(4-5)
The mole ratio of Fe to P is 1:1. 162.3 g of FeCIa will
react with 95 g of PC>4 to form 150.8 g of FePC-4.
Stoichiometric weight ratio of Fe:P is 1.8:1, while the
weight ratio of FeCIs to P is 5.2:1. As with alum, the
reaction mechanism is more complex than the
equation shown above.
The reaction between ferrous salts (ferrous chloride
and ferrous sulfate) and phosphate can be
approximated by:
3FeCI2 + 2P043- -» Fe3(PO4)2 i + 6Cf (4-6)
3FeS04 + 2PO43- -» Fe3(PO4)2 i + 3S042-
(4-7)
56
-------
Table 4-1. Characteristics of Aluminum and Iron Salts (3,4)
Common Name and
Formula
Dry alum
AI2(SO4)3»14H2O
Liquid alum
AI2(SO4)3»14H2O
Dry sodium aluminate
Na2AI204
Liq. sodium aluminate
Na2AI2O4
Liq. ferric chloride
FeCI3
Liq. ferrous chloride
FeCI2
Dry ferrous sulfate
FeSO4*7H2O
Shipping Data
Available
Forms
Lump
Ground
Rice
Powdered
Commercial;
waste pickle
liquor
Commercial;
waste pickle
liquor
Containers and
Requirements
Bags: 45,90 kg;
Bbl.: 135,180kg;
Drums: 11,45,110 kg;
Bulk - Car loads
Tank trucks
Store dry
Bulk - Car loads
Tank trucks
Bags: 23,45,68 kg;
Bulk: not available.
6 month max. storage.
Drums: 170 kg;
Tank truck;
Tank car;
2-3 month max. storage.
Carboys: 19,49 I;
Tank truks: 1 1 ,500-
15,000 I;
Tank cars: 1 5,000-
38,000 I
Drums: 1 90 I;
Tank trucks: 15,0001
Tank cars
Physical and Chemical Characteristics
Appearance and
Properties
White/cream color.
pH: 3.0-3.5 for 1-10%
solution.
Dust is irritant to mucous
membranes.
Will begin to crystalize @
-1°C;
Crystalizes @ -8°C.
Corrosive.
pH: 1 1 .9 for 1 % solution.
Non-corrosive.
Dust is irritant.
Strong alkali.
Handle as caustic.
Dark brown, oily.
pH: 2.0 for 1 % solution.
Very corrosive., stains
concrete and other
materials.
Dark brown, oily.
Free acid content typically
1 -1 .5% but may reach
10%.
Slightly less corrosive
than FeCI3
Acidic when dissolved.
Composition is variable.
Oxidizes in moist air.
Cakes @ storage temp.
above 20°C.
Bulk Density
(kg/cm)
600-1,200
1,330 @ 16°C
640-800
1,340-1,490
1,190-1,250
990-1,060
Commercial Strength
17% AI2O3 by wt.
8.3% AI2O3 by wt.
41-46% AI2O3 by
wt.
4.9-26.7% AI203
by wt.
35-45% FeCI3 by
wt.
20-25% FeCI2 by
wt.
55-58% FeSO4 by
wt.
In reality these reactions are more complex. The mole
ratio of Fe to P is 3:2. Weight ratio of ferrous ion to
phosphorus is 3.2:1.
Addition of iron salts will result in the destruction of
alkalinity as described by:
FeCI3 + 3HCO3' -» Fe(OH)3
+ 3CO2 + 3d'
(4-8)
Iron salts are most effective for phosphorus
precipitation within a certain pH range. For ferric
(Fe3 + ) ion, the optimum pH range is 4.5-5.0.
However, significant removal of phosphorus can be
achieved at higher pH. For ferrous (Fe2 + ) ion, the
optimum pH is approximately 8. Good phosphorus
removal can be achieved between pM values of 7 and
8. Canadian studies have shown that effective
phosphorus precipitation does not occur until the
ferrous ion is oxidized to ferric ion, and for this reason
do not support the use of ferrous salts in primary
treatment (9). However, both ferrous chloride and
ferrous sulfate have been used effectively in primary
treatment (3,10).
4.2 Application Points
The most common points for addition of aluminum
and iron salts for phosphorus removal are: 1)
immediately upstream of the primary clarifier, 2) in or
immediately after the aeration basins prior to final
clarification, and 3) at both points simultaneously.
Another option is use of separate, tertiary chemical
clarification. Such a scheme would only be justified
for very stringent effluent discharge standards, as for
reuse.
Figure 4-1 shows the common schemes for
chemical addition in an activated sludge plant. The
advantages of addition to primary treatment include
greater opportunity for adequate mixing and
flocculation, and reduced loadings to downstream
processes as a result of improved BOD and SS
removal. The major disadvantage of chemical addition
to the primaries is that incomplete phosphorus
precipitation may result because of the presence of
phosphorus forms other than orthophosphate that are
not easily precipitated.
57
-------
Figure 4-1. Alternative schemes for mineral addition for phosphorus removal.
Flocculant
Aid
Influent "^
Rapid
Mix
^ hi
Al or Fe I ±
^ Y.
Sludge to Processing
Aeration
Basin
a. Mineral addition in primary treatment
Influent
Flocculant
Aid
t '
A
l
Item
t
1
ative
Aeration
Basin
t
t
Ki
t
1 1
mineral addition points
H
Effluent
Sludge to Processing
Flocculant
Aid
b. Mineral addition in secondary treatment
Influent ""
Alor
Rapid
Mix
i
Fe i
T ^/ Pnmarv ] ^
"\ Clanfier / ^ i
i
k
t
1
[ Alternative
^ _t .
Aeration
Basin
t
1
t
1
J s
t
"\ Cl<
V
mineral addition points
Sec.
Effluent
Sludge to Processing
c. Multiple point mineral addition
In general, higher levels of phosphorus removal can
be achieved by chemical addition to the secondary
process or by addition at multiple points in the
treatment train.
As discussed in Chapter 2, adequate wastewater
characterization and jar testing are essential before
implementing a chemical addition program.
Table 4-2, compiled based on a survey of 104
plants in the U.S. and Canada, shows a breakdown of
plants by location of chemical addition and by cation
used for precipitation (1).
4.2.1 Mineral Addition Before Primary
Clarification
Phosphorus removal by addition of mineral salts in
primary treatment requires good mixing and
flocculation in order to ensure optimum results. With
proper design and operation of mixing and flocculation
systems, 70-90 percent phosphorus removal can be
achieved in primary treatment. Significant increases in
BOD and SS removal efficiencies can also be
expected from mineral salt addition. Table 4-3
provides a summary of potential removal efficiencies
of P, BOD, and SS in primary and secondary
treatment with and without mineral addition (3).
For mineral addition to primary treatment, provision of
a separate rapid mix tank may be necessary. As
discussed in Section 4.2.4, existing hydraulic
structures such as Parshall flumes and drop
manholes have been used, but may not provide
sufficient turbulence to ensure complete and intimate
mixing of the chemical with the wastewater. Good
performance has been reported in Canada by
injection into the discharge side of raw sewage
pumps, addition to aerated grit chambers, preaeration
58
-------
Table 4-2. Distribution of Selected Phosphorus Removal Facilities by Point of Chemical Addition and Cation Used (1)
Number of Plants
Point of
Addition
Primary
Secondary
Tertiary
Total
United States
Al
1
26
2
29
Fe
16
6
2
24
Total
17
32
4
53
Al
2
17
1
20
Canada
Fe
20
8
0
28
Total
22
25
1
48
Al
3
43
3
49
Total
Fe
36
14
2
531
Total
39
57
5
102
1 One plant does not specify point of addition.
Table 4-3. Potential effectiveness of Primary and Secondary Treatment With and Without Mineral Addition for Phosphorus
Removal
Phosphorus Removal (%)
SS Removal (%)
BOD Removal (%)
Primary Treatment
Secondary Treatment
Trickling Filter
Activated Sludge
Without
5-10
10-20
10-20
With
70-90
80-95
80-95
Without
40-70
80-90
80-95
With
60-75
85-95
85-95
Without
25-40
75-90
85-95
With
40-65
80-95
85-95
channels, or Parshall flumes, and addition to pipes or
channels between primary clarifiers and aeration
tanks if supplemented with mechanical or air mixing
(11,12).
Initial mixing intensity is very important for complete
dispersal of the coagulant and, thus, maximum
efficiency. Flocculation normally occurs naturally in
the primary clarifier, particularly in the center feed
well, which may preclude the need for a separate
flocculation basin. New clarifiers can be designed with
mixers and a center flocculation well with as much as
30 minutes detention time. If necessary, existing
clarifiers can be modified to provide a designated
flocculation zone.
Due to competing reactions and variation in coagulant
demand, more chemical will generally be required for
phosphorus precipitation in primary treatment than in
secondary or tertiary treatment. This can be seen in
Figure 4-2, which shows phosphorus removed per
mole of aluminum added for three different application
points (3).
4.2.2 Mineral Addition to Secondary Processes
Addition of aluminum or iron salts directly to the
aeration basins or between the aeration basins and
final clarifiers is common practice for chemical
precipitation of phosphorus. This alternative has
considerable flexibility in the point of chemical
addition, allowing modifications of the injection point
to ensure use of the best available conditions for
coagulation and flocculation to occur. Unfortunately,
the approach has some drawbacks, in that velocity
Figure 4-2. Impact of point of addition on effectiveness of
phosphorus removal using aluminum (3).
Moles Tot. Sol. P Rem. per
Mole Aluminum Added
1-2 i-
1.0
0.8
0.6
0.4
0.2
A Final Effluent
O Aerator
• Raw Wastewater
Moles Aluminum Added per
Mole Initial Tot. Sol. P
59
-------
gradients or turbulence levels are likely to be less
than ideal for proper mixing and flocculation to occur.
Since the optimal point of chemical addition will vary
depending on the choice of chemicals, velocity
gradients in the aeration basin and inter-basin
channels, and wastewater characteristics, full scale
experimentation with various points of addition will
likely be necessary. If a high degree of phosphorus
removal is required, mineral addition should occur
downstream of any return streams such as digester
supernatant. Addition of minerals to secondary
processes may result in an increase in dissolved
solids in the effluent, particularly when pickling liquors
or other "impure" chemical sources are used.
4.2.3 Mineral Addition at Multiple Points
Addition of mineral salts at multiple locations in the
treatment plant has been found to be an efficient and
cost-effective means of chemical addition for
phosphorus control. Advantages of this approach are
overall reduction in chemical requirements to achieve
a given effluent phosphorus objective and increased
operational flexibility. In design of new facilities,
provision of multiple chemical addition points is
recommended to allow optimization of the chemical
feed system to achieve the most economical and
reliable solution.
It should be noted that, when aluminum or iron salts
are used for phosphorus precipitation, addition of
small amounts of a coagulant aid such as anionic
polyelectrolytes may be necessary before the final
clarifier to assist in removing dispersed metal-
phosphate floe. A typical polymer dose when used as
a coagulant aid is 0.1-0.25 mg/l (8).
4.3 Performance
Table 4-4 shows performance data from selected
facilities using various mineral salts and points of
addition to achieve phosphorus removal (13). In
general, 70-90 percent phosphorus removal can be
expected in primary treatment, with removal
efficiencies in secondary treatment ranging from 80 to
95 percent.
Dosages can be expected to vary widely depending
on choice of chemical and wastewater characteristics,
as well as other factors such as degree of mixing
intensity at the point of addition and opportunity for
flocculation. Table 4-5 shows a summary of dosages
employed during full-scale studies at numerous
wastewater treatment plants (14). The dosages cited
are those required to achieve a total phosphorus
concentration of 1 mg/l in the plant effluent.
4.4 Equipment Requirements
4.4.1 Chemical Handling and Storage
As seen in Table 4-1, a variety of chemicals are
available for precipitation of phosphorus from
wastewater. Each chemical has characteristic
requirements regarding type and maximum duration of
storage, choice of piping and transport systems,
chemical feeding equipment, and safety precautions.
Chemical handling and storage requirements for
common chemicals used in phosphorus control are
described below.
a. Aluminum Salts
Aluminum sulfate is available in dry or liquid form. Dry
alum or "filter alum" can be purchased in lump,
ground, rice, or powdered grades. Water utilities
prefer ground or rice alum because of superior flow
characteristics. Bulk alum should be stored in mild
steel or concrete bins with dust collection equipment.
Gates should be provided on bulk storage vessels to
allow isolation of feed equipment.
A typical bulk storage tank for dry chemicals is shown
in Figure 4-3. Bulk dry alum can be transferred with
screw conveyors, bucket elevators, or pneumatic
conveyors. Bags and drums of alum should be stored
in dry locations. Day hoppers receiving alum from
bags or drums should have a minimum storage
capacity of eight hours at the maximum expected
feed rate. Hopper bottoms should have a minimum
wall slope of 60 degrees to prevent arching.
Dry alum is not corrosive unless it absorbs moisture.
Alum dust, however, can cause minor irritation of the
eyes and respiratory tract.
Liquid alum is available in 11-19 m3 (3,000-5,000
gal) tank truck lots or 26-68 m3 (7,000-18,000 gal)
tank car lots. Transportation costs are greater for
liquid alum since it is nearly half water by weight.
Liquid alum will generally be more economical than
dry alum if the point of use is within 160 km (100 mi)
of the manufacturing location. However, because of
the ease of handling, storage, and feeding in liquid
form, the practical limit for transport may be 320 km
(200 mi) or more (3).
Alum is typically stored without dilution at the shipping
concentration received at the plant. Storage tanks
located outside should be closed and vented, with
provisions for heating to maintain temperatures above
-4°C (25°F) to prevent crystallization. Materials of
construction for liquid alum storage vessels include
type 316 stainless steel, fiberglass-reinforced plastic
(FRP), or steel lined with rubber, polyvinyl chloride
(PVC), or lead. Liquid alum can be stored indefinitely.
Storage tanks should be sized to accommodate a
10-day to 2-week supply and should be capable of
handling 1-1/2 times the maximum quantity shipped.
Liquid alum is moderately corrosive, and hand and
face protection should be worn when working on
leaking equipment. Any spills should be immediately
flushed with water as liquid alum becomes very
slippery upon evaporation (3).
60
-------
Table 4-4. Performance of Facilities Using Mineral Salts for Phosphorus Removal (11)
Plant Type
and Location
Plug Flow AS
Waupaca, Wl
East Chicago, IN
Mason, Ml
Flushing, Ml
Appleton, Wl
Grand Ledge, Ml
Bowling Green, OH
Kenosha, Wl
Toledo, OH
Clintonville, Wl
Complete Mix AS
Thiensville, Wl
Two Harbors, MN
Escanaba, Ml
Sheboygan, Wl
Lima, OH
Miles, Ml
Crown Point, IN
Cedarburg, Wl
Contact Stabilization AS
Neenah, Wl
Neenah, Wl
Algoma, Wl
Grafton, Wl
Port Washington, Wl
Port Clinton, OH
Oberlin, OH
North Olmstead, OH
Pure Oxygen AS
Fon du Lac, Wl
Extended Aeration AS
Aurora, MN
Upper Allen, PA
Corunna, Ontario
Saukville, Wl
Plymouth, Wl
Trenton, OH
Seneca, MD
Design
Flow
m3/d
4,760
75,700
5,700
4,400
62,500
5,700
30,300
106,000
386,100
3,800
900
4,500
8,300
69,600
70,000
22,000
13,600
11,400
5,700
14,800
2,800
8,100
4,700
5,700
5,700
34,000
41,600
1,900
1,800
3,800
7,600
6,200
13,200
18,900
Average
Flow
m3/d
2,200
59,800
5,000
6,000
52,200
3,000
20,100
90,500
310,400
2,700
3,300
3,400
7,600
46,600
15,100
12,100
8,700
7,600
4,000
16,700
3,000
3,600
5,800
6,400
5,700
21,200
26,900
1,700
1,200
2,000
2,400
5,800
9,600
15,100
Chemicals
Alum
Alum
Polymer
Ferric Chloride
Polymer
Ferric Chloride
Ferrous Chloride
Ferrous Chloride
Ferrous Chloride
Polymer
Ferrous Sulfate
Ferrous Sulfate
Polymer
Ferrous Sulfate
Alum
Polymer
Alum
Feme Chloride
Polymer
Ferric Chloride
Ferrous Chloride
Polymer
Ferrous Chloride
Ferrous Chloride
Polymer
Ferrous Sulfate
Polymer
Alum
Alum
Ferric Chloride
Polymer
Ferrous Chloride
Ferrous Chloride
Ferrous Chloride
Ferrous Chloride
Sodium Alummate
Alum
Polymer
Alum
Alum
Polymer
Alum
Ferrous Chloride
Ferrous Chloride
Ferrous Chloride
Sodium Alummate
Polymer
Chemical
Feed Point
Sec. Clarifier
Sec. Clarifier
Sec. Clarifier
Prim. Clarifier
Prim. Clanfier
Sec. Biol. Process
Sec. Biol. Process
Plant Influent
Sec. Biol. Process
Sec. Clarifier
Sec. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Sec. Clarifier
Sec. Biol. Process
Sec. Biol. Process
Sec. Clarifier
Prim. Clanfier
Prim. Clanfier
Sec. Clarifier
Prim. Clarifier
Prim. Clarifier
Sec. Biol. Process
Sec. Clarifier
Sec. Clarifier
Sec. Clarifier
Sec. Clarifier
Prim. Clanfier
Sec. Biol. Process
Prim. Clanfier
Prim. Clanfier
Prim. Clarifier
Prim. Clarifier
Sec. Biol. Process
Prim. Clarifier
Sec. Biol. Process
Sec. Clarifier
Sec. Clarifier
Prim. Clarifier
Sec. Biol. Process
Sec. Biol. Process
Sec. Clarifier
Prim. Clanfier
Sec. Biol. Process
Sec. Biol. Process
Plant Influent
Sec. Clarifier
Chemical
Dosage
mg/l as
Metal Ion
24.6
7.7
1.0
9.1
0.05
5.3
0.15
16.8
5.6
5.2
5.35
3.6
5.3
9.3
0.82
9.6
4.7
0.35
10.2
13.2
0.07
10.9
11.0
0.94
9.9
7.7
4.1
33.0
0.07
16.2
8.5
10.2
6.4
8.3
8.5
0.75
16.9
8.2
0.37
5.0
10.3
7.7
2.56
4.3
2.4
Metal Ion:
Inf. TP
3.25
3.99
1.4
1.56
1.6
1.24
0.62
1.43
1.3
1.47
2.46
1.6
1.04
1.6
3.38
2.66
2.0
2.99
2.2
1.0
10.0
2.31
1.44
1.96
1.08
2.86
1.18
5.83
0.92
0.65
1.61
1.15
0.42
0.61
Inf.
TP
mg/1
7.56
1.93
6.5
3.4
10.5
4.5
8.4
3.74
2.76
3.6
3.78
6.0
4.5
6.38
3.9
4.1
5.5
3.31
3.5
4.1
3.3
7.0
5.9
5.2
5.9
2.9
7.2
2.9
8.9
7.74
6.4
6.7
6.1
7.1
Eff.
TP
mg/l
0.86
0.38
0.88
0.48
0.8
0.7
0.75
0.36
0.35
0.75
0.29
0.25
0.82
0.9
0.5
0.7
0.7
0.67
0.7
0.8
0.23
0.69
1.0
0.5
1.0
0.7
0.73
0.76
2.0
0.36
0.59
0.77
0.65
1.6
61
-------
Table 4-4. Performance of Facilities Using Mineral Salts for Phosphorus Removal (continued)
Plant Type
and Location
Step Aeration AS
Fort Wayne, IN
East Lansing, Ml
Oak Creek, Wl
Elkhart, IN
2-Stage Nitrification AS
Piscataway, MD
High Rate TF
Geneva, OH
Coldwater, Ml
Oconto Falls, Wl
Kendalville, IN
Standard Rate TF
Willard, OH
Elizabethtown, PA
Durand, Ml
Saglnaw, Ml
Little Hunting Creek, VA
Bay City, Ml
Coloma, Ml
RBC
Romeo, Ml
Chesaning, Ml
Negaunee, Ml
Dexter, Ml
Hartford, Ml
St Johns, Ml
Charlotte, Ml
Oxidation Ditch
Lapeer, Ml
Portage, IN
Design
Flow
m3/d
227,100
71,200
454,200
75,700
113,600
7,600
8,700
1,900
10,100
5,100
11,400
3,000
16,700
17,000
75,700
8,300
6,100
2,200
6,100
2,200
1,300
7,200
4,500
7,000
13,200
Average
Flow
m3/d
170,100
42,800
340,650
60,200
54,900
3,900
7,400
1,400
7,600
4,800
6,500
2,700
6,400
14,400
33,300
5,300
3,300
2,000
3,300
800
800
6,300
2,700
7,200
8,400
Chemicals
Ferrous Chloride
Ferrous Chloride
Polymer
Ferrous Sulfate
Ferrous Sulfate
Alum
Polymer
Alum
Ferric Chloride
Polymer
Ferric Chloride
Feme Chloride
Polymer
Alum
Polymer
Alum
Polymer
Ferric Chloride
Ferric Chloride
Polymer
Ferric Chloride
Polymer
Ferric Chloride
Polymer
Ferrous Chloride
Alum
Polymer
Ferric Chloride
Polymer
Ferric Chloride
Polymer
Ferric Chloride
Polymer
Ferrous Chloride
Polymer
Ferrous Chloride
Polymer
Ferrous Chloride
Polymer
Ferric Chloride
Ferrous Chloride
Chemical
Feed Point
Sec. Biol. Process
Sec. Clarifier
Sec. Clarifier
Sec. Biol. Process
Sec. Clarifier
Sec. Clarifier
Sec. Clarifier
Sec. Clarifier
Sec. Clarifier
Sec. Clarifier
Sec. Biol. Process
Sec. Biol. Process
Sec. Biol. Process
Prim. Clarifier
Prim. Clarifier
Sec. Clarifier
Sec. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Sec. Clarifier
Sec. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Sec. Clarifier
Sec. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Prim. Clarifier
Sec. Clanfier
Sec. Clarifier
Sec. Clarifier
Chemical
Dosage
mg/l as
Metal Ion
4.3
5.9
0.05
4.4
1.6
8.8
3.8
12.1
8.3
0.1
8.81
14.7
0.25
6.3
0.14
12.8
0.4
11.2
9.6
0.1
42.5
2.8
9.5
0.29
4.1
7.1
0.77
9.0
0.4
7.5
1.0
10.2
0.5
13.0
0.6
5.01
0.04
13.7
0.18
4.65
9.9
Metal Ion:
Inf. TP
0.54
1.11
0.96
0.63
1.44
4.03
2.02
2.4
4.05
1.21
2.51
2.2
0.99
4.57
2.07
1.71
2.4
3.46
3.75
2.0
3.25
1.38
2.45
0.88
1.65
Inf.
TP
mg/l
7.9
5.3
4.6
2.56
6.13
3.0
4.1
3.67
3.63
5.2
5.1
5.1
9.7
9.3
4.6
2.4
2.96
2.6
2.0
5.11
4.0
3.7
5.6
5.3
6.0
Eff.
TP
mg/l
0.67
0.9
0.54
0.83
0.2
0.4
0.88
0.45
0.35
0.82
1.7
0.83
1.5
0.2
0.5
0.65
0.46
0.6
0.95
0.46
0.75
0.5
0.68
1.2
1.5
62
-------
Table 4-5.
Point of
Addition
Chemical Dosage Summary from Ontario
Treatability Studies (14)
Chemical
Ave. Metal
Number Ave. lon/TP
of Plants Dosage a.b Ratio
mg/l
Raw
wastewater
Mixed liquor
Feme chloride
Alum
Ferric chloride
Alum
7
5
20
15
14.2
10.3
9.5
7.5
2.7
1.7
1.5
1.6
a Dosage required to achieve effluent TP of 1 mg/l.
b Expressed as Fe or Al.
Sodium aluminate is available in both dry and liquid
forms (see Table 4-1). Dry sodium aluminate should
be stored for a maximum of six months at 16-32°C
(60-80° F) and will deteriorate with exposure to the
atmosphere. Hopper agitation may be required to
prevent caking and bridging. Storage vessels may be
mild or stainless steel, FRP, or concrete. Use of
copper and its alloys, rubber, and aluminum should
be avoided. Maximum recommended storage time for
liquid sodium aluminate is two to three months (4).
Sodium aluminate should be treated as a caustic
similar to sodium hydroxide. Body contact should be
prevented, and hand and face protection must be
worn when working on sodium aluminate storage or
feed equipment.
b. Iron Salts
Iron salts discussed here include ferric chloride,
ferrous chloride, and ferrous sulfate, which are the
most common iron compounds used for phosphorus
removal. Other iron salts such as ferric sulfate can
also be employed for precipitation of phosphorus.
Ferric chloride is available as a liquid in carboys and
in bulk, as shown in Table 4-1. Tank trucks and tank
cars are usually unloaded pneumatically. Designated
safety procedures should be closely followed during
loading and unloading operations. Ferric chloride
storage tanks may be constructed of steel lined with
rubber or plastic, FRP, or synthetic resins. In most
cases, ferric chloride should be stored in heated
buildings or in heated tanks to prevent crystallization.
Liquid ferric chloride can be stored indefinitely without
deterioration. Ten to 14 days of storage capacity is
recommended, with the ability to handle 1-1/2 times
the largest anticipated shipment (3).
Ferric chloride is a corrosive material. When working
on ferric chloride handling equipment, workmen
should wear rubber gloves, rubber aprons, and
goggles or a face shield. Any contact with eyes or
skin should be flushed with running water. If ingested,
vomiting should be induced.
Ferric chloride will stain concrete and other materials.
To prevent staining in areas where ferric chloride is
handled, rubber mats or resistant coatings should be
used.
Ferrous chloride, or waste pickle liquor, is a
byproduct of steelmaking operations and is available
in bulk tank car or tank truck lots. The free acid
content may vary from 1 to 10 percent but is usually
1-2 percent. Although slightly less corrosive than
ferric chloride, ferrous chloride generally has the
same storage and handling requirements. Since
pickle liquor may not be available on a continuous
basis, storage and handling facilities should be
suitable for accommodating ferric chloride as an
alternate chemical.
Ferrous sulfate is also a byproduct of steelmaking
operations, although the product is normally sold in
the dry form as granules, crystals, powder, or lumps.
Composition may be variable but typically contains
55-58 percent FeS04- Dry ferrous sulfate will
oxidize and hydrate in moist air and will cake at
temperatures above 20°C (68°F). In the dry form, it
should be stored in cool dry areas. Storage
containers may be constructed of concrete, synthetic
resin, or steel lined with asphalt, rubber, PVC, or
chemically resistant resins (3,4). Ferrous sulfate dust
is irritating to the eyes and respiratory tract.
Ferrous sulfate solution is acidic and should be
handled with the same precautions that apply to ferric
chloride. Construction materials for storage vessels
for ferrous sulfate solutions are the same as those for
ferric chloride solutions.
c. Polymer
While polymer is not used specifically for phosphorus
precipitation, its use in conjunction with aluminum and
iron salts is so common as to warrant a separate
discussion. Small quantities of polymer are typically
added just downstream of the metal addition point to
assist in the agglomeration and settling of the metal-
phosphate floe. A minimum lag time of 10 seconds
(flow time) is recommended between the point of
metal addition and the polymer injection point (14).
Some designers have suggested lag times of two to
five minutes (15).
A multitude of polymers are available in dry or liquid
form. Dry polymer is shipped in a variety of packages
and containers, depending on the manufacturer.
Bagged dry polymer should be stored in cool, low
humidity areas. Bags should be removed in proper
rotation to prevent excessive storage times. Polymers
are generally low in toxicity and are not irritating.
Polymer is added to a process in the solution form,
requiring blending of dry polymer with water to form a
stock solution, followed by an aging period. The stock
solution is usually diluted prior to use. Liquid polymers
63
-------
Figure 4-3. Typical dry chemical feed system.
Dust
Collector
n
Bulk
Fill Pif
(Pneume
Storage
Bin
Dust
Collector
Screen
with
Breaker
Flexible
Connection
Bag
/Fill
Day Hopper for
Dry Chemical
from Bags
or Drums
_LJ
Alternative Supplies Depending on Storage
Control Solenoid
Valve Valve
require no aging but are normally diluted with water
before application.
Polymer solutions are typically stored in FRP, type
316 stainless steel, or plastic lined steel tanks. To
prevent deterioration, polymer solutions should be
stored no longer than 1-3 days. Check with the
manufacturer for recommendations for the specific
product.
Instantaneous blending polymer systems may have
application for temporary or intermittent use
situations. These systems automatically meter, dilute,
activate, and feed liquid polymer and water and do
not require separate storage, holding, and mixing
tanks.
4.4.2 Dry Chemical Feeding and Dissolving
Dry chemical feed equipment can be of three types:
• volumetric
• loss-in-weight gravimetric
• belt gravimetric
Volumetric feeders are the least expensive and can
be used where cost is a concern, chemical delivery
rates are low, and great accuracy is not required.
Volumetric feeders generally employ a screw feed
mechanism.
Loss-in-weight gravimetric feeders provide a high
degree of accuracy (1 percent) and are
recommended where close control of chemical
64
-------
dosages can result in substantial savings in chemical
costs. Maximum feed rates of such units are
approximately 1,800 kg (4,000 lb)/hr.
Belt gravimetric feeders are intermediate in cost
between volumetric and loss-in-weight gravimetric
feeders, and can provide accurate and reliable
service.
In general, closed construction is preferable for
chemical feeders, since this exposes a minimum of
operating components to the corrosive vapors from
the dissolving or solution tank. The various types of
chemical feeders available are shown in Table 4-6
(4).
Gravimetric feeders offer the following advantages
over volumetric feeders:
1. Calibration is normally not required.
2. Greater accuracy and dependability.
3. Incorporation of totalizer to allow maintenance of
accurate records and inventories.
4. Automatic proportioning.
5. Low maintenance; simple operation.
When dry chemicals are used, a working solution is
made up by blending water with the chemical in a
mechanically-agitated dissolving tank or solution
tank. With bulk chemicals, such systems employ a
water meter in conjunction with a variable rate feeder
to achieve a continuous stream of the solution at the
proper strength. With bags or containers, the proper
solution is made up manually on a batch basis.
With alum or sodium aluminate, the recommended
minimum solution strength is 6 percent or 0.06 kg/liter
of water (0.5 Ib/gal). The detention time in the
dissolver should be 5 minutes at the maximum feed
rate. The same recommendations are made for
dissolution of dry ferrous sulfate. For ferric sulfate,
feed solutions are made up at a water to chemical
weight ratio of 2:1 to 8:1, with a typical ratio being 4:1
or 0.25 kg Fe2SC>4/liter of water (2.1 Ib/gal). Solutions
having strengths less than 1 percent are subject to
hydrolysis and deposition of ferric hydroxide.
A typical arrangement for the feeding and dissolving
of dry chemicals is shown in Figure 4-3. It should be
noted that the degree of automation in dry chemical
dissolution systems will depend on the size of the
plant and daily chemical usage.
For plants less than 3,785 m3/d (1 mgd), manual
preparation of the chemical solution on a batch basis
may be indicated. This is typically accomplished in a
day tank in which dry, bagged chemical is
mechanically mixed with water to reach the desired
concentration. For larger facilities, the chemical
solution is prepared automatically using a controller
which adjusts feed rate of dry chemical in proportion
to potable water flowrate (16). A system of this type is
depicted in Figure 4-3.
4.4.3 Liquid Chemical and Solution Feeding
Several alternatives are available for feeding liquid
chemicals or chemical solutions. The pressure head
often determines the type of system to be used.
Rotary dipper feeders or rotameters with control
valves are commonly used for gravity feed
applications, while metering pumps are used for
feeding chemical solutions under pressure.
Although provision is sometimes made for dilution of
liquid alum or ferric chloride prior to feeding into the
process, this is generally unnecessary, and may be
undesirable due to the occurrence of hydrolysis at
dilute concentrations. It has been found that feeding
undiluted liquid alum results in better coagulation and
settling (3). For polymers, dilution of the stock
solution is generally practiced to allow better
dispersion of the polymer in the wastewater.
Figure 4-4 shows typical chemical feed
arrangements for elevated chemical storage systems,
while Figure 4-5 shows alternatives for ground
storage. Rotary dipper feeders are reliable feeders
that are commonly used for gravity flow applications.
Feed rates can be varied based on a signal from a
mainstream flow meter (flow proportional control) as
discussed later in this section. Rotameters in
conjunction with control valves may also be used for
small applications where frequent variation in
chemical feed rate is not required. Rotameters should
not be used with ferric chloride or other iron solutions
since the sight glass will become stained and opaque.
Centrifugal transfer pumps, as shown in Figure 4-5a,
should be direct connected but not close-coupled to
prevent leakage into the motor. Pump components for
liquid alum service should be constructed of 316
stainless steel, FRP, or plastics. For ferric chloride,
graphite or rubber lined pumps with Teflon seals are
recommended. Metering pumps are typically of the
positive displacement type, either diaphragm or
plunger. Diaphragm pumps protected with internal or
external relief valves are preferred. A back pressure
valve is recommended to provide positive check valve
operation (3). Materials of construction for chemical
feed service include 316 stainless steel, FRP,
plastics, and rubber. Manufacturers recommendations
should be followed regarding selection of pump
materials for the specific chemical of interest.
For pipes transporting alum solution, use of FRP,
PVC, or other plastics is recommended. Valves
should be plastic, 316 stainless steel, or rubber-lined
iron or steel. For ferric chloride conveyance, pipes
should be constructed of steel lined with rubber or
Saran, FRP, or plastics. Valves should be rubber- or
resin-lined diaphragm valves, Saran-lined valves
with Teflon diaphragms, rubber-sleeved pinch
65
-------
Table 4-6. Types of Chemical Feeders (4)
Type of Feeder
Limitations
Use
Capacity
Range
Dry feeders
Volumetric:
Oscillating plate
Oscillating throat (universal)
Rotating disc
Rotating cylinder (star)
Screw
Ribbon
Belt
Gravimetric:
Continuous - belt and scale
Loss in weight
Solution feeders
Nonpositive displacement:
Decanter (lowering pipe)
Orifice
Rotameter (calibrated value)
Loss in weigth (tank w/control
valve)
Positive displacement:
Rotating dipper
Proportional pump:
Diaphragm
Piston
Any material, granules or powder.
Any material, any particle size.
Most materials including NaF, granules or powder.
Any material, granules or powder.
Dry, free flowing material, powder or granular.
Dry, free flowing material, powder, granular or lumps.
Dry, free flowing material up to 1.5-in size, powder or
granular.
Dry, free flowing granular material, or floodable material.
Most materials, powder, granular or lumps.
Most solutions or light slurries.
Most solutions.
Clear solutions.
Most solutions.
Most solutions or slurries.
Most solutions. Special unit for 5% slurries.1
Most solutions, light slurries.
liters/h
0.3 - 1,000
0.06 - 2,800
0.3 - 28
230 - 57,000
or
200 - 8,500
1.4 - 510
0.06 - 4.5
2.8 - 85,000
0.6 - 57
0.6 - 2,300
0.3 - 280
4.5 - 142
0.1 - 4.5
or
0.3 - 570
0.06 - 5.7
2.8 - 850
0.1 - 4.2
0.3 - 4,800
40 to 1
40 to 1
20 to 1
10 to 1
or
100 to 1
20 to 1
10 to 1
10 to 1
to
100 to 1
100to 1
10010 1
10010 1
10 to 1
10 to 1
30 to 1
100 to 1
100 to 1
20 to 1
1 Use special heads and valves for slurries.
valves, or plastic ball valves (3,4). Pipe selection for
polymer service should be made after the type of
polymer has been determined. Plastic pipe or type
316 stainless steel is normally used.
4.4.4 Chemical Dosage Control
Control of chemical feeding is a critical part of any
phosphorus removal system. Control of chemical
dosages is important not only to ensure that effluent
phosphorus requirements are consistently met, but
also to keep chemical use and operating costs to a
minimum. As discussed in Chapter 2, phosphorus
loadings to a plant can be expected to fluctuate
significantly on an hourly, daily, and even seasonal
basis.
The type and complexity of selected control systems
are dependent on plant size and sophistication of
operations, as well as daily chemical usage. For small
plants with low chemical requirements, a complex
automated control system cannot be justified on the
basis of either economics or practicality. For large
plants, however, close control of chemical dosage
can result in substantial savings in chemical costs,
and a more sophisticated control system would be
within the operation and maintenance capabilities of
the plant staff. Several control options are discussed
below.
a. Manual Control
Manual dosage control may be appropriate for plants
less than 3,785 m3/d (1 mgd). Manual operation of a
chemical feed system would involve: 1) daily
preparation of the chemical solution on a batch basis
(if dry chemicals are purchased), and 2) manual
setting of a control valve or the stroke of a diaphragm
metering pump to establish proper flowrate of
chemical solution. Although such a system is reliable
and requires little maintenance, it has a major
disadvantage in that chemical feed rate is constant
66
-------
Figure 4-4. Liquid chemical feed alternatives for elevated
storage.
Rotary dipper feeder
a. Rotary dipper feeder
Figure 4-5. Liquid chemical feed alternatives for ground
storage.
Recirculation
Ground
chemical
storage
tank
JS
r
*ft|4
Rotary dipper
feeder
To process
(gravity
feed)
a Rotary dipper feeder
Elevated
chemical
tank
Control
valve
To process
1
Rotameter (or mag meter)
b. Rotameter with control valve
Elevated
chemical
storage
tank
Relief
valve
Y
nr
,
Back pressure
valve
To process
Metering pump
c. Metering pump
Rotameter (or mag meter)
Control
valve
I—Xh-
To process
(pressure
feed)
Centrifugal transfer pump
Centrifugal pump with rotameter and control valve
Back pressure
valve
tx-
t
To process
(pressure
feed)
c. Metering pump
and plant flow is variable, resulting in variation in
dosages. This problem can be overcome to some
degree by manually varying the chemical feed rate at
specified intervals based on historical flow and
performance data (17). However, the potential for
overdosing or underdosing exists. At plants where
flows are generally predictable, this problem is
minimal. At others, such as those with significant
infiltration/inflow, operational schedules should reflect
the need to adjust rates during initial periods.
b. Flow Proportional and Programmed Control
Flow proportional control involves automatic
adjustment of chemical feed rate in proportion to plant
flow. This is accomplished by transmitting a signal
from a flow measuring device to a controller, which
then adjusts the speed of a rotary dipper feeder, the
stroke of a metering pump, or the opening of a
control valve. Such devices are well demonstrated
and reliable for a variety of chemical feed
applications. Unfortunately, the resulting chemical
feed rate will be proportional to plant flowrate, which
may bear little correlation to the actual phosphorus
loadings to the facility.
An alternative to flow proportional control is
programmed control in which chemical feed rate is
programmed in relation to historical data on mass
phosphorus loading patterns (product of plant flow
and phosphorus concentration). The use of
67
-------
programmable controllers is a relatively simple and
effective technique which has been successfully
implemented at full scale facilities. This requires an
accurate knowledge of incoming phosphorus
loadings, preferably on an hourly basis over a period
of one week. Chemical feed rates are thus
programmed to coincide with intervals of known
phosphorus loadings. For automated systems, this
approach is superior to flow proportional control since
it provides for more efficient and specific application
of chemicals. However, it still suffers from the fact
that chemical dose is based on historical flow and
phosphorus data. During wet weather flows, the
programmed dosing schedule may not be applicable.
c. Feed Forward and Feedback Control
A more sophisticated approach for control of chemical
dosage is the use of feed-forward or feedback
control, in which the chemical feed rate is controlled
in proportion to a combined signal of wastewater
flowrate and some wastewater characteristic such as
pH, conductivity, or phosphate concentration
(16,18,19). In Norwegian studies, it was found that
alkalinity was the predominant variable affecting alum
dosage. Since field proven instruments that provided
on-line analysis of alkalinity were not available, it
was decided to base chemical dosage on conductivity
measurements. Interestingly, good long-term
correlation was found between alkalinity and
conductivity. As a result, a conductivity feed forward,
open loop control system was implemented to allow
variation of chemical feed rate based on a combined
signal of wastewater flowrate and primary effluent
conductivity. This system was found to be effective,
resulting in savings in chemical costs over other
control alternatives. As of February, 1985, six plants
in Norway using alum for phosphorus control had
implemented the conductivity feed forward control
system (18).
In-line phosphate analyzers have been investigated
for use in automated chemical control systems (17).
Unfortunately, operation and maintenance
requirements are high, and skilled technicians must
be available to service such units. Particularly when
used for raw wastewater or primary effluent
applications, the devices are subject to plugging and
fouling. Overall, their use in chemical dosage control
applications has been unsuccessful.
4.4.5 Chemical Mixing and Flocculation
It is important that chemicals used for phosphorus
precipitation be intimately mixed with the wastewater
to ensure uniform dispersion and efficient application
of the chemicals. Where existing plants have been
modified to remove phosphorus by chemical addition,
it has been standard practice to use existing
structures and facilities to provide mixing of the
chemical with the wastewater. Points of addition have
included Parshall flumes, drop manholes, aerated grit
chambers, the discharge side of raw sewage pumps,
90° pipe bends, hydraulic jumps, and aeration basins,
where the turbulence levels were typically higher than
for other locations at the plant.
Although chemical addition at such locations has
been reported to be effective, it is unlikely that
optimum mixing conditions existed. Poor or
inadequate mixing can result in inefficient chemical
use and greater chemical consumption. Where
viscous chemicals such as polymers are added to the
waste stream, provision of intense mixing is essential.
It has been recommended that polymer be added at a
concentration of 0.01-0.05 percent, and never above
0.1 percent (14). This allows reduction in required
mixing power by providing for improved dispersion
upon injection into the main stream. Manufacturers'
recommendations should be followed for polymer
addition, since "overmixing" is possible, resulting in
reduced performance.
The most important parameters used in the design of
mixing and flocculation processes are detention time,
t, and velocity gradient, G. Velocity gradient is a
measure of the shear intensity imparted to a fluid.
The equation for velocity gradient is given in Section
4.5. For chemical mixing processes, detention times
are approximately 30 sec, and G values are on the
order of 300 m/sec/m although velocity gradients of
up to 1,000 m/sec/m have been recommended (20).
A typical flash mix tank is shown in Figure 4-6.
The flocculation process allows contact between
coagulated solids so that they agglomerate to form
solids with improved settling characteristics. Whether
for a new plant or for a retrofit application, separate
basins are seldom constructed specifically for
flocculation. Rather, existing plant components such
as aerated grit chambers, aerated distribution
channels, or feed wells of clarifiers are used, often
after some modification. Although use of existing
tankage may be effective for flocculation, tank size
and configuration should be carefully evaluated, and
velocity gradients calculated to ensure presence of
conditions that will promote good flocculation.
A common approach is the use of a flocculating
clarifier, in which an expanded center well provides
the desired detention time for flocculation. The
contents of the flocculation well can be agitated by
mechanical mixers or diffused air, although the
hydraulic regime in the center well may be such that
mechanical or air mixing does not provide additional
benefit (21). A typical flocculating clarifier is shown in
Figure 4-7.
Velocity gradients for flocculation processes generally
are 20-80 m/sec/m, depending on the chemicals
added and point of addition. Velocity gradients less
than 50 m/sec/m may yield floe particles with too
much trapped water, whereas velocity gradients
above 80 m/sec/m may cause excessive floe shear
68
-------
Figure 4-6. Typical flash mix tank.
Support Beams
Overflow
Impeller
Shaft
Impeller
I I I
Feed
and floe deterioration. In many cases, mineral salts
are added directly to the aeration basin to allow
mixing and flocculation to occur. Although this
practice is effective, it represents less than ideal
conditions for flocculation, as velocity gradients in
aeration basins result in floe shear. In one study, air
flowrates in the downstream end of the aeration basin
were reduced to achieve a velocity gradient of 60
m/sec/m, which was found to be optimum for
flocculation using ferric chloride (22). Addition of
anionic polymer prior to clarification assists in the
agglomeration of sheared floe.
Other parameters used in the design of flocculation
processes are Gt, the product of velocity gradient and
detention time, and GCt, where C is the ratio of
volume of floe to total volume of suspension. In water
treatment applications using aluminum or iron salts,
Gt values are typically 30,000-150,000, with GCt
values of 10-100 (23).
4.4.6 Clarification
Since clarifiers are used in virtually all wastewater
treatment plants and are not specific to phosphorus
removal systems, the following discussion will be
limited to how their application or design differs from
conventional practice when mineral salts are added to
precipitate phosphorus.
The two most common points of mineral addition are:
1) prior to primary clarification and 2) prior to
secondary clarification. Consequently, clarifier design
is critical to ensure adequate removal of suspended
solids and consistent achievement of phosphorus
discharge limitations.
Clarifiers used in chemical precipitation systems differ
little from those employed in conventional biological
treatment, although use of flocculation zones is
recommended to allow flocculation to occur after
addition of coagulants.
Provision of distinct flocculation zones is
recommended for either primary or secondary
clarifiers depending on the point of chemical addition.
This is particularly important for primary clarifiers,
since there may be little opportunity for flocculation to
occur in existing processes.
For secondary mineral addition, flocculation can occur
in aeration basins or channels preceding clarification,
but the use of flocculation zones in secondary
clarifiers is recommended practice to allow flexibility
in the point of chemical addition and to provide a
zone in which direct control can be exercised over
velocity gradients in order to achieve optimum
flocculation.
4.5 Design Methodology
The approach to design of facilities for phosphorus
removal using mineral salts follows the basic steps
outlined below.
1. Characterization of the wastewater (raw
wastewater, primary effluent, mixed liquor,
secondary effluent).
2. Determination of chemical doses required to meet
effluent phosphorus requirements based on
addition to primary or secondary treatment.
3. Selection of chemicals and points of addition.
4. Evaluation of existing unit processes (retrofit
applications).
5. Conduct of full-scale trials (retrofit applications).
6. Sizing and design of chemical storage, handling,
and feed systems.
7. Sizing and design of liquid processes such as rapid
mix and flocculation facilities, clarifiers, and
aeration basins.
8. Sizing and design of sludge handling facilities.
4.5.7 Wastewater Characterization
A thorough knowledge of wastewater characteristics
is required in order to proceed with chemical
selection and process design. The following
information is essential:
69
-------
Figure 4-7. Typical mechanically mixed flocculating clarifier.
Axial Flow -
" Slow Stirrers
t
Effluent
Flocculation
Chamber
(slow mix)
Influent
1. Average dry weather and wet weather flows.
2. Diurnal flow variation.
3. Characterization parameters.
a. BOD and COD
b. SS and VSS
c. Total and soluble phosphorus
d. Orthophosphate
e. pH
f. Alkalinity
4. Diurnal (hourly) variation in phosphorus loadings.
4.5.2 Determination of Chemical Doses
Jar tests should be conducted with the specific
wastewater to be treated if possible. Such tests
should be conducted over an extended period of time
to obtain representative wastewater characteristics.
Grab samples are collected at various times of the
day and days of the week, and a range of dosages
established for each chemical to be tested. A
minimum of 10 data points should be gathered for
each chemical dose on each waste stream
investigated. Jar test procedures are described in
Reference 24. Probability plots should then be
developed which show percent of time that effluent
phosphorus is less than a stated value for each
dosage. These plots are valuable in determining the
reliability of a particular coagulant over a range of
dosages.
4.5.3 Selection of Chemicals
Chemical selection is made based on: 1) performance
and reliability (results of jar tests) and 2) costs.
Factors included in the selection decision include
whether to purchase chemicals in dry or liquid form,
in bulk or in small lots. Consideration should be given
to whether secondary chemicals (e.g. polymers for
flocculation aids, bases for pH adjustment) will be
required, and in what quantities.
When mineral salts are used for phosphorus
precipitation, anionic polymers may be required, and
prudent design should include polymer addition
facilities. Adjustment of pH through addition of lime or
other alkaline materials may be necessary, particularly
with low-alkalinity wastewaters.
If waste products such as pickle liquor are available,
jar testing should include investigation of such
chemicals at various times to quantify variability of
these chemicals. Also, the engineer should quantify
their availability over long periods to determine
storage requirements.
Assuming that more than one chemical or
combination of chemicals yields acceptable results in
terms of performance and reliability, the selection of
chemicals is then an economic decision. A cost-
effectiveness analysis should be conducted which
addresses capital costs of chemical storage, handling,
and feeding equipment, considering both dry
chemicals and liquid chemicals in bulk and in small
lots; as well as operational costs, including costs of
chemicals, labor and power.
4.5.4 Evaluation of Existing Unit Processes
For retrofit applications the existing process train
should be evaluated with respect to:
1. Potential points of chemical addition.
2. Degree of mixing at identified points of addition
(e.g., raw sewage pumps, Parshall flumes or
hydraulic jumps, aerated grit chambers, etc.).
3. Opportunities for flocculation (e.g., preaeration
basins, clarifier feed wells, connection piping,
weirs, aeration basins, etc.).
70
-------
4. Adequacy of existing clarifiers (existing vs. required
design overflow rates).
Potential points of chemical addition include: 1) prior
to primary clarification, 2) prior to secondary
clarification, 3) in aeration basin at turbulent locations,
and 4) at several of these points simultaneously.
When evaluating existing structures or equipment to
provide mixing of the coagulant with the wastewater,
velocity gradients should be calculated where
possible to determine if the level of turbulence is
sufficient for adequate mixing. A minimum velocity
gradient of 300 m/sec/m is recommended. Canadian
studies showed that efficient mixing could be obtained
in full scale facilities by the following approaches (11):
1. Injection of the chemical into the discharge side of
raw sewage pumps.
2. Chemical addition into a preaeration tank or
aerated grit chamber.
3. Addition of chemical at a Parshall flume or similar
turbulent constriction.
4. Chemical addition to pipe or channel between
aeration tank and final clarifier with supplemental
mechanical or air mixing at the point of addition.
If existing structures are inadequate for providing
good mixing, separate mixing basins or modification
of existing structures will be required. For example,
existing preaeration basins can be modified to provide
for intense mixing in the upstream end, followed by
gentle mixing to promote flocculation. It is important
to note that phosphorus removal efficiency can suffer
significantly if intense mixing of the coagulant and
wastewater is not provided. During the Canadian
studies, it was found that, in several cases,
phosphorus removal efficiency was doubled by
increasing the intensity of mixing at the point of
chemical addition (11).
In most cases, existing structures of the facility can
be used, perhaps with some modification, to promote
flocculation of the coagulated wastewater. Some
flocculation will occur naturally during passage
through the clarifier feed well, although consideration
may be given to use of an enlarged center feed well
in the clarifier to provide approximately 20 minutes of
gently agitated flocculation.
The existing and design overflow rates in primary and
secondary clarifiers must be compared with
recommended designs for chemical precipitation
systems. If necessary, additional clarification capacity
should be provided. New primary clarifiers should be
equipped with center flocculation zones. For
secondary clarifiers preceded by chemical addition to
the aeration basins, specially designed flocculation
wells may be unnecessary, as air flowrates in the
downstream end of the aeration basins can be
adjusted to provide near optimum conditions for
flocculation.
4.5.5 Conduct of Full-Scale Trials
Where existing plants are to be retrofitted for
phosphorus removal, full scale trials are
recommended. Full-scale trials of 6 weeks for
primary treatment plants and 8 weeks for secondary
treatment plants have been used (25). Such trials
offer the following advantages:
1. Determination of optimum point(s) of chemical
addition.
2. Optimization of chemical dosages.
3. Evaluation of impacts on aeration requirements,
SRT, F/M and other key process variables.
4. Direct observation of impacts of variables on
effluent quality.
5. Estimation of the quantity (volumetric and mass) of
additional sludge to be processed.
6. Observation and determination of impacts on
sludge thickening, stabilization, and dewatering
characteristics; chemical requirements for sludge
conditioning; equipment modifications; and disposal
limitations.
4.5.6 Design of Chemical Handling System
The main components of a chemical handling system
include: 1) chemical storage facilities, 2) dry chemical
feed and dissolution equipment (if dry chemicals are
used), and 3) chemical solution feed systems.
a. Chemical Storage
Bulk storage facilities for either dry or liquid chemicals
should be sized for capacities of at least 50 percent
greater than the largest anticipated shipment, and
should provide a minimum capacity of a ten day
supply, and preferably a two week supply. Where
chemicals are delivered by rail, it may be prudent to
provide additional capacity to compensate for late or
intermittent deliveries.
Dry chemical storage vessels should be designed
with bottom hoppers having slopes of 60 degrees to
help present bridging of the chemical. Vibrators are
also used for this purpose. Dust collection equipment
is recommended to control fugitive dust emissions
during loading operations. Choice of materials is
dependent on the particular chemical to be stored.
For dry iron compounds, steel lined with rubber,
asphalt, or plastic, synthetic resins, or concrete may
be used. Dry alum can be stored in mild steel or
concrete bins. Where dry chemicals are purchased in
small lots such as bags or drums, the chemicals
should be stored in enclosed, dry areas.
71
-------
Liquid alum should be stored in tanks constructed of
FRP, 316 stainless steel, or steel lined with rubber or
PVC. Storage vessels for liquid sodium aluminate
should be constructed of mild or stainless steel, FRP,
or concrete. Use of copper and its alloys, rubber, and
aluminum should be avoided. If located outside, tanks
should be closed and vented, with heaters to maintain
temperatures above -4°C (25°F). Polyurethane
insulation between the tank and the pad is
recommended.
Liquid iron solutions such as ferric chloride and
ferrous chloride should be stored in vessels
constructed of FRP, synthetic resins, or steel lined
with rubber or plastic. Tank heaters and insulation
may be required, as ferric chloride will crystallize at
temperatures ranging from -12°C (10°F) for a 35
percent solution to 10°C (50°F) for a 45 percent
solution. To prevent freezing during shipment, weaker
solutions with lower freezing points are shipped
during the winter.
Liquid chemical storage tanks should be equipped
with a strip or float gauge to determine volume of tank
contents, an access manhole, a filling hose with quick
disconnect coupling, and a drain that is flush with the
tank bottom.
b. Dry Chemical Feed and Dissolution Systems
Dry chemical feeder selection will be dependent on
the chemical used. For dry alum, either open or
enclosed construction is permissible. However, with
dry iron compounds such as ferric or ferrous sulfate,
belt type feeders are rarely used because of their
open construction and exposure to corrosive vapors
from the dissolving tank. Provision of water jets for
vapor removal is recommended to protect equipment
and operating personnel.
Selection of a volumetric vs. gravimetric feeder is
dependent upon total chemical demand and degree of
accuracy required. For plants larger than 3,785 m3/d
(1 mgd), gravimetric feeders are recommended, since
the additional cost over a volumetric feeder is likely to
be offset by the savings in chemical costs during the
useful life of the equipment. Capacity ranges of
feeders should be a minimum of 10 to 1. Sufficient
capacity should exist to handle peak chemical
demands. For some feeders, the capacity range can
be significantly expanded by modification of the gear
box or by use of variable speed drives.
For small plants using bagged chemicals, the feeder
can be attached to a "day hopper," which is filled
manually each day and which has a capacity for one
day's supply of chemical. In larger plants, the
chemical feeder is charged directly from the bulk
storage silo.
Dry chemicals must be added to water in a dissolving
tank to form a chemical solution prior to introduction
into the wastewater stream. For alum, dissolving
tanks should be constructed of FRP, plastics, or 316
stainless steel. The most dilute alum solution
recommended is a 6 percent solution (0.06 kg/I). For
dry sodium aluminate, dissolving chambers may be
mild steel or stainless steel. Preparation of a 6
percent solution is standard practice. Iron solutions
are actively corrosive, so care must be taken in
selecting materials. Ferric sulfate dissolvers should be
constructed of type 316 stainless steel, plastics,
rubber, or ceramics. Minimum solution strength for
ferric sulfate is one percent, and the typical water to
chemical weight ratio is 4:1. Ferrous sulfate
dissolution tanks should be constructed of type 304
stainless steel, plastics, rubber, or iron.
Recommended solution strength is 6 percent.
For most dry chemicals, dissolving chambers should
be designed for a detention time of 5 minutes at the
maximum feed rate. The exception to this is ferric
sulfate, for which a 20-minute detention time is
recommended (3). The dissolving tanks must be large
enough to provide the necessary detention time at the
maximum rate of feed of water and chemical.
Dissolvers should be mechanically mixed and should
be equipped with a water meter so that the proper
solution strength can be determined and maintained.
c. Chemical Solution Feeding
As discussed in Section 4.4, rotary dipper feeders or
metering pumps are commonly used to provide
positive control over chemical feed rates. Other
options are rotameters with control valves, or orifice
plates. Flow control devices should be sized for the
maximum expected flowrate of chemical solution.
Centrifugal pumps can be used for the transfer of
chemical solutions from dissolving tanks to other
vessels, or for direct feeding to the process through a
flow measurement device such as a rotameter. To
achieve uniform flow rates with centrifugal pumps,
pumping head must be constant. Centrifugal pumps
should be direct connected but not close-coupled to
prevent leakage into the motor. Pump components for
liquid alum service should be constructed of 316
stainless steel, FRP, or plastics. For ferric chloride,
graphite or rubber lined pumps with Teflon seals are
recommended.
Metering pumps are typically of the positive
displacement type, either diaphragm or plunger.
Diaphragm pumps protected with internal or external
relief valves are preferred. A back pressure valve is
recommended to provide positive check valve
operation (3).
Materials of construction for chemical feed service
include 316 stainless steel, FRP, plastics, and rubber.
Manufacturers' recommendations should be followed
regarding selection of pump materials for the specific
chemical of choice.
72
-------
For pipes transporting alum solution, use of FRP,
PVC, or other plastics is recommended. Valves
should be plastic, 316 stainless steel, or rubber-lined
iron or steel.
For ferric chloride conveyance, pipes should be
constructed of steel lined with rubber or Saran, FRP,
or plastics. Valves should be rubber- or resin-lined
diaphragm valves, Saran-lined valves with Teflon
diaphragms, rubber-sleeved pinch valves, or plastic
ball valves (3,4).
Pipe selection for polymer service should be made
after the type of polymer has been determined.
Plastic pipe or type 316 stainless steel is normally
used.
4.5.7 Design of Liquid Processes
Design of chemical phosphorus precipitation systems
requires consideration of not only chemical storage
and feed equipment, but also mainstream process
equipment such as aeration basins and clarifiers as
well as sludge handling equipment used for
thickening, stabilization, dewatering and disposal. For
example, where phosphorus is removed by chemical
precipitation in primary treatment, primary clarifier
design will likely vary from standard practice to
provide for flocculation. Since BOD and SS removal
efficiencies will be significantly improved, loadings to
downstream biological processes will be reduced,
while sludge generation rate will be increased. In
addition, sludge characteristics may be considerably
different than if no chemicals were added.
a. Rapid Mixing
In the design of new facilities incorporating chemical
phosphorus precipitation, provision of separate rapid
mixing basin(s) is recommended for complete contact
of the chemical with the wastewater. Rapid mix tanks
may be designed for detention times of 20-60
seconds, although 30 seconds is recommended for
metal salts. Velocity gradients (G) on the order of 300
m/sec/m are generally sufficient, although values as
high as 1,000 m/sec/m have been recommended
(15). The power required to maintain turbulent
conditions (Reynolds > 105) in a flash mix basin with
an impeller mixer can be calculated from (26):
P =pKTn3Da5/g
where,
P = power requirement, ft-lb/sec
p = mass density of the fluid, Ib/ft3
n = impeller revolutions per second, rps
Da = diameter of impeller, ft
g = acceleration due to gravity, 32.2 ft/sec2
KT = constant
Values of KT for various mixing devices are shown
in Table 4-7).
Table 4-7. Values of KT for Determining Impeller Power
Requirements (26)
Type of Impeller Kj
Propeller (square pitch, 3 blades)
Propeller (pitch of 2, 3 blades)
Turbine (6 flat blades)
Turbine (6 curved blades)
Turbine (6 arrowhead blades)
Fan turbine (6 blades)
Flat paddle (2 blades)
Shrouded turbine (6 curved blades)
Shrouded turbine (with stator, no baffles)
0.32
1.00
6.30
4.80
4.00
1.65
1.70
1.08
1.12
Velocity gradient can be calculated from the general
equation (26):
G = [P/Vp]l/2
where,
G = velocity gradient, ft/sec/ft
p = absolute fluid viscosity, Ib-sec/ft2;
approximately 2 x 10-5 ib-sec/ft2 @ 20 °C
V = basin volume, ft3
For an electrically driven mechanical mixer,
P = (WHP)(550)
where,
WHP = delivered water horsepower or
(KVA)x(Motor Eff.)x(Power Factor)/0.746
KVA = apparent power, kVA
b. Flocculation
As discussed in Section 4.4.5, numerous devices
have been used to promote flocculation, including
preaeration basins, interprocess channels, aeration
basins, and enlarged feed wells of clarifiers.
Maintenance of velocity gradients in the range of 50
to 80 m/sec/m for no more than 15 minutes is
recommended for good flocculation where metal salts
are used. Some have recommended high energy
flocculation for 5 minutes followed by gentle
flocculation for 15 minutes (15). Such "tapered
flocculation" can occur naturally in clarifiers and in
aeration basins in which air flows are decreased at
the downstream end.
The degree of flocculation is difficult to predict, since
it will vary with wastewater characteristics and choice
of chemicals, as well as with hydraulic characteristics
and energy dissipation in the basin. For this reason,
design of new facilities should provide for maximum
flexibility in order to achieve optimum conditions for
flocculation.
73
-------
Flocculation can be accomplished by mechanical
(paddle or turbine) mixers, air diffusers, or baffles.
Velocity gradient can be determined by substituting
the appropriate value of P into the general equation
below:
Paddles:
where,
G = [P/Vu]i/2
P = CouAv3/2g
P = power requirement, ft-lb/sec
CD = coefficient of drag of flocculator paddles,
dimensionless
co = mass fluid density, slugs/ft3
A = area of paddles, ft2
v = relative velocity of paddles in fluid, ft/sec
(about 0.7-0.8 of paddle tip speed)
Turbines:
where,
P =
P = power requirement, ft-lb/sec
Kj = constant
Da = diameter of impeller, ft
Air:
where,
P = 82 Qa log [(H + 34)734]
P = power imparted to water, ft-lb/sec
Qa = air supplied, scfm
H = head of water above air diffusers, ft
Baffles:
where,
P = Qtohf
P = power imparted to water, ft-lb/sec
Q = flow, ft3/sec
hf = head loss due to friction, ft
Important considerations in the design of flocculation
basins include:
1. Transport conditions -- minimize transport time
(<1 minute) from rapid mix tank to flocculator.
Keep velocities low and avoid turbulence from
flocculator to clarifier.
2. Flow distribution -- maintain good flow
distribution at flocculator inlet by designing for
headloss through inlet ports of 10 times transport
headless from first to last port.
3. Short circuiting - install baffle walls between
multiple paddle units in rectangular basins; in
circular units with rotary flocculators, place baffles
along wall to prevent entire tank contents from
rotating with mechanism.
4. Tapered flocculation - provide compartments
which increase in volume from the inlet end to the
outlet end of the tank. Reduce slow mix power with
each downstream stage.
5. Separate vs. combined units -- consider
combined flocculator-clarifier for lime precipitation
systems; compare costs, reliability, flexibility of
separate vs. combined unit processes.
When flocculation is carried out in an aeration basin
or separate structure, the velocity in conduits
conveying the floe to the clarifier should be kept
between 0.15 and 0.30 m/sec (0.5-1.0 ft/sec) so as
to prevent destruction of the floe (8).
c. Clarification
If not preceded by a flocculation process, clarifiers
receiving a chemically dosed effluent should be
designed with a flocculation zone. Several
manufacturers provide circular clarifiers with center
flocculation wells which incorporate a detention time
of 20-30 minutes. Generally, flocculation wells are
30-40 percent of the tank diameter. The walls of the
feedwell generally extend down to 60-75 percent of
the tank depth (14).
The principal design criterion for chemical clarification
is overflow rate. Table 4-8 provides a summary of
recommended design overflow rates for primary,
secondary, and tertiary clarifiers receiving wastewater
coagulated with mineral salts (14). These values are
generally conservative.
Minimum bottom slopes of clarifiers should be 8
percent. All clarifiers should be equipped with scum
removal mechanisms. A minimum sidewater depth of
3.6 m (12 ft) is recommended (14).
4.5.8 Design of Sludge Handling System
Addition of mineral salts for phosphorus precipitation
may significantly increase the quantity of sludge
generated due to the production of metal-phosphate
precipitates and metal hydroxides as well as the
improved removal of suspended solids.
Addition of metals upstream of the primary clarifier
will result in a primary sludge mass increase of 50-
100 percent. This increase is normally due almost
equally to improved capture of suspended solids and
additional chemical sludge (3). Overall plant sludge
74
-------
Table 4-8. Recommended Overflow Rates for Conventional
Clarifiers Receiving Wastewater Coagulated
with Mineral Salts (14)
Design Overflow Rates
(Average Flow)
Type of Clanfier
Primary
Secondary
Tertiary
w/o Polymer
m3/m2/d
(gpoVfl2)
24
(600)
24
(600)
24
(600)
w/Polymer
m3/m2/d
(gpd/ft2)
49
(1,200)
-
49
(1,200)
samples should be collected at various times of the
day and on various days of the week. Mixed liquor
samples should be collected at different operating
conditions (e.g. SRT) where possible. Evaluate
effectiveness of coagulants including ferric
chloride, alum, and ferrous sulfate over a
prescribed range of dosages. If waste pickle liquor
is available locally or at a reasonable cost, include
in jar tests.
Plot effluent total phosphorus vs. chemical dose on
a probability plot for each chemical. Determine
dosages that will meet effluent phosphorus
requirements at the desired level of probability
(e.g., 90 percent or 95 percent).
mass increase is much smaller owing to reduced
secondary sludge production from improved primary
removals, e.g., 60-70 percent increase across the
entire plant.
For metal addition to secondary processes, waste
mixed liquor sludge mass can be expected to
increase by 35-45 percent. Overall plant sludge
mass increase will be 5-25 percent (11). Metal
addition to either primary or secondary treatment will
not only increase sludge mass, but sludge volume will
be increased since settled sludge concentration in the
clarifiers may decrease by up to 20 percent (11). A
detailed discussion of characteristics, generation
rates, and treatment alternatives for sludges resulting
from chemical phosphorus removal systems is
presented in Chapter 5.
4.5.9 Design Example
Develop a preliminary process design for retrofitting
an existing 11,350 m3/d (3.0 mgd) activated sludge
facility for chemical precipitation of phosphorus. The
following effluent requirements are imposed:
BOD = 15 mg/l
SS = 15 mg/l
Total P = 0.8 mg/l
1. Characterize wastewater
BOD = 200 mg/l
SS = 210 mg/l
Total P = 7 mg/l
Ortho P = 4 mg/l
Alkalinity = 220 mg/l
pH = 7.1
Design ave. daily dry weather flow = 11,350 m3/d
Current ave. daily dry weather flow = 9,500 m3/d
Peak:average daily flow ratio = 2.5:1
2. Determine chemical dosages
Conduct jar tests using samples of both raw
wastewater and mixed liquor. Raw wastewater
3. Select chemical
Chemicals should be selected based on the
following criteria:
a. Performance and reliability (from probability
plots)
b. Unit costs
c. Reliability of supply
d. Operation and maintenance of chemical feed
equipment
e. Safety in handling
Based on analysis of the above criteria, assume
ferric chloride is selected as the most economical
and reliable chemical.
4. Evaluate existing liquid stream processes
An evaluation of existing liquid stream processes
resulted in the following observations and
conclusions:
a. Potential points of chemical addition
Two identified points of possible chemical addition
to the existing activated sludge facility are: 1)
upstream of the primary clarifiers and 2) in or
immediately after the aeration basis.
b. Degree of mixing
Inspection and evaluation of hydraulic structures
upstream of primary clarification indicate that
sufficient mixing and turbulence do not exist, thus
requiring construction of a separate rapid mix
basin. For chemical addition to secondary
treatment, it is concluded that direct chemical
addition to the aeration basins is feasible without
modification of the basins.
75
-------
c. Opportunities for flocculation
Evaluation of the hydraulic characteristics of the
existing primary clarifiers indicates that adequate
flocculation will occur in the feed well and other
portions of the clarifier. This will be confirmed in
full-scale trials.
For addition to the aeration basin, it is determined
that air flowrates through the diffusers in the
downstream portions of the basins can be readily
throttled to yield desired velocity gradients for good
flocculation. Existing velocity gradients are
approximately 120 m/sec/m. Full-scale trials are
necessary to determine optimum velocity gradients
in the aeration basin.
d. Adequacy of existing clarifiers
The existing facility has 2 primary clarifiers, each
15 m (50 ft) in diameter. Design overflow rate at
11,350 m3/d (3 mgd) average daily design flow is
32 m3/d/m2 (800 gpd/ft2), and approximately 27
m3/d/m2 (670 gpd/ft2) at present raw wastewater
flow. Comparison with design overflow rates for
chemical precipitation systems indicates that the
current overflow rate is only about 10 percent
greater than recommended [24 m3/d/m2 (600
gpd/ft2)]. Since the recommended rates are
conservative, expansion of existing clarifier
hydraulic capacity is not warranted at this time.
The existing final clarifiers were designed for an
overflow rate of 24 m3/d/m2 (600 gpd/ft2) at
average flow. Current overflow rates are
approximately 20 m3/d/m2 (500 gpd/ft2). These
rates are within recommended guidelines of 24
m3/d/m2 (600 gpd/ft2) for chemical clarification.
Thus, expansion of secondary clarifier hydraulic
capacity is not justified.
5. Conduct full-scale trials
Full-scale trials were conducted over period of
eight weeks using ferric chloride as a coagulant.
Key findings from the studies are as follows:
a. Split addition of ferric chloride to primary and
secondary treatment produced optimum results.
b. Optimum dosages were 8 mg/l to primary
treatment and 32 mg/l to secondary.
c. Addition of a rapid mix basin ahead of primary
treatment is necessary. For the trials, temporary
mixing facilities were constructed.
d. Good flocculation occurs naturally prior to the
primary clarifier.
e. Injection of ferric chloride at a point halfway
down the aeration basin was found to be
effective when air flows in the downstream third
of the basin were reduced to yield velocity
gradients of approximately 70 m/sec/m.
f. Effluent suspended solids were generally less
than 10 mg/l.
g. Effluent total phosphorus concentrations ranged
from 0.5 to 0.8 mg/l.
h. Modification of existng clarifiers is unnecessary.
6. Design chemical handling system
From full-scale trials, it was found that split
addition of ferric chloride to primary and secondary
treatment was the most economical and reliable
approach, with 20 percent of the chemical added
upstream of the primary clarifiers and 80 percent
added directly to the aeration basins. Optimum
dosages were approximately 8 mg/l and 32 mg/l to
primary and secondary treatment, respectively.
Ferric chloride is available locally in liquid form.
Total FeCIa requirements at average design flow:
40 mg/l x 11,350 m3/d x 0.001 = 454 kg/d
Average volumetric requirements for 40%
solution:
(454 kg/d) -5- (0.40 x 1,400 kg/m3) = 0.81 m3/d
(215gal/d)
Peak volumetric requirements:
2.5 x 0.81 m3/d = 2.02 m3/d (535 gal/d)
Size ferric chloride storage tanks based on 1-1/2
times largest anticipated shipment or 10 days
storage at maximum feed rate:
1.5 x 15 m3 = 22.5 m3 (5,940 gal)
10 days x 2.02 m3/d = 20.2 m3 (5,340 gal)
Use 2 storage tanks for total of 23 m3 (6,000 gal)
storage capacity.
Select 2 metering pumps (1 standby); positive
displacement diaphragm type. Size each pump at
twice the anticipated maximum chemical feed rate
- 85 l/hr (22.3 gph). Since data collected indicate
a relatively constant diurnal phosphorus
concentration in the raw wastewater, design for
flow proportional control of chemical metering
pumps.
7. Design liquid stream processes
Full-scale trials indicate that no existing hydraulic
components upstream of primary treatment provide
76
-------
sufficient turbulence to achieve adequate mixing.
Therefore, a separate rapid mix tank is required
upstream of the primary clarifiers. Existing primary
and secondary clarifiers do not require modification
to achieve desired performance.
Rapid Mix Tank
Size for detention time of 30 sec at average flow:
Volume = 3.3 m3 (870 gal)
Dimensions = 1.5 m x 1.5 m x 1.5 m deep
Use propeller mixer with Kj = 1.00 (Table 4-7)
Power required for mixing:
P = pKTn3Da5/g
Use p = 62.4lb/ft3
KT = 1.00
n = 600 rpm (10 rps)
Da = 1 ft
g = 32.2 ft/sec2
P = 62.4(1.00)(10)3(1)5/32.2
= 1,940ft-lb/sec
= 1,940 * 550 = 3.5 hp (use 4 hp)
Check velocity gradient, G:
G = [PA/p]l/2
= [(4 x 550) * (3.3 x 35.31 )(2 x 10-5)] 1/2
= 974 sec*1 (within the range discussed in
Section 4.5.7)
8. Design sludge stream processes
Although techniques are available to calculate
additional sludge production from chemical addition
(see Chapter 5), the most reliable method is
through full-scale trials. In addition, such trials
allow determination of the impact of chemical
addition on sludge thickening, stabilization, and
dewatering characteristics. A design example for
sludge handling in chemical phosphorus removal
systems is presented in Chapter 5.
4.6 Retrofit Considerations
In general, existing wastewater treatment plants can
be retrofitted for chemical phosphorus precipitation
relatively easily and inexpensively, provided that
sufficient hydraulic and solids handling capacity is
available. Design considerations include:
1. Existing vs. design flows and loadings (including
phosphorus); average and peak.
2. Existing and design overflow rates for primary and
secondary clarifiers, compared to recommended
values for chemical precipitation systems.
3. Current loadings to secondary biological
processes.
4. Solids loadings to sludge thickeners and
dewatering equipment, compared to expected
loadings with chemical precipitation.
5. Impacts of chemical addition on sludge thickening,
stabilization, and dewatering characteristics.
6. Impacts on volume of sludge for final disposal.
7. Availability of turbulent zones for chemical addition
points.
8. Availability of gently mixed zones for flocculation of
chemically coagulated wastewater.
The two variables which have the greatest impact on
the cost and complexity of retrofitting plants for
chemical phosphorus removal are the existing solids
handling and clarification capacity.
For any chemical precipitation system, sludge
production can be expected to increase substantially,
requiring a thorough capacity analysis of sludge
handling processes and equipment. In some cases,
existing equipment may be sufficient, requiring only
longer operating times for such sludge processing
operations such as dewatering. In other situations,
expansion of existing processing capability through
construction of additional tankage or installation of
new equipment may be necessary, involving
potentially large capital investments.
Design of clarifiers for precipitation of phosphorus by
mineral addition is typically conservative, although
when polymer is used as a flocculant aid, allowable
hydraulic loadings can often be significantly increased
without degradation of effluent quality.
In order to achieve low levels of phosphorus in plant
effluents, suspended solids concentrations must be
correspondingly low. It is generally true that, unless
effluent suspended solids concentrations can be
reduced to below about 15 mg/l, it is impossible to
achieve an effluent total phosphorus concentration of
less than 1 mg/l, even though soluble phosphorus
may be as low as 0.1 mg/l (11).
4.7 Case Histories
4.7.1 Orillia, Ontario
This case history is included to document the
capabilities of a dual point chemical addition system
to achieve low effluent phosphorus levels. The
information has been extracted from the final report of
a study funded by the Ontario Ministry of the
Environment (27).
77
-------
The Orillia, Ontario wastewater treatment plant
employs the conventional activated sludge process.
Unit operations include preliminary treatment, primary
clarification, aeration, secondary clarification, and
chlorination. Primary and secondary sludge is
pumped to a two-stage anaerobic digestion system.
The digested sludge is discharged to lagoons for
storage during winter months, and spread on
agricultural land during other times of the year.
Phosphorus removal is normally practiced by adding
alum to the outlet of the aeration tanks at a dosage of
approximately 65 mg/l. Target effluent phosphorus
levels are 1 mg/l. Design flow is 18,000 m.3/d (4.8
mgd). Plant performance data are summarized in
Table 4-9.
Table 4-9. Summary of Plant Data During Single-Point
and Dual-Point Alum Addition (27)
Parameter
Flow, m3/d
Raw sewage
BOD, mg/l
SS,mg/l
T-P, mg/l
Soluble P, mg/l
Primary effluent
BOD, mg/l
SS,mg/l
T-P, mg/l
Soluble P, mg/l
Final effluent, mg/l
BOD, mg/l
SS,mg/l
T-P, mg/l
Soluble P, mg/l
Alum dosage, mg/l
Primary process
Secondary
mg/l A|3+/mg/l P rem.
Single-Point
Alum Addition
14,000
115
197
8.4
2.9
62
89
4.5
2.0
11
10
0.65
0.18
0
64
1.3
Dual-Point
Alum Addition
17,000
138
182
5.2
1.8
52
64
2.7
0.9
17
14
0.36
0.06
16
32
1.1
A study was initiated in 1981 to assess the feasibility
of achieving effluent total phosphorus concentrations
of 0.3 mg/l. It was originally intended that this would
be achieved through chemical addition plus tertiary
filtration. However, initial jar tests indicated that the
effluent P objective of 0.3 mg/l might be obtained by
dual point chemical addition without tertiary filters.
The first phase of the study involved optimizing
existing operations. To improve mixing and
flocculation, the point of chemical addition was moved
from a manhole between the aeration basins and final
clarifiers to a point near the last air diffuser in the
aeration basin. This allowed reduction in the alum
dosage to 60 mg/l.
Full-scale performance of dual point alum addition
for phosphorus removal was then investigated. The
scheme involved simultaneous addition of
approximately 20 mg/l of alum to the raw wastewater
at the aerated grit chamber and approximately 40 mg/l
of alum to the tail end of the aeration basin. A
sampling program was conducted during the period
September 1981 through June 1982 to evaluate the
impact of dual chemical dosage points. Data collected
during this period are also summarized in Table 4-9.
Comparison of data for the two periods shows that
the use of dual point chemical addition improved the
removal of phosphorus significantly without increasing
the overall alum dosage. Average effluent P of 0.36
mg/l was achieved at an overall dosage of 48 mg/l
during dual point addition, whereas a dosage of 64
mg/l to the secondary process alone only achieved an
effluent P of 0.65 mg/l. With dual point addition,
average soluble P in the effluent was 0.06 mg/l, vs.
0.18 mg/l with single point chemical addition. Most of
the phosphorus leaving the system during either
period was associated with suspended solids. Note
that effluent SS and BOD were somewhat higher
during this period, possibly due to the fact that the
average wastewater flow was 21 percent greater.
Secondary clarifier average overflow rates were also
higher at 35 m3/d/m2 (860 gpd/ft2) vs. 30 m3/d/m2
(750 gpd/ft2) during single point addition. It was
concluded that the 0.3 mg/l P standard could be
consistently met with dual point, flow-paced
chemical addition with no tertiary filtration. Clarifier
overflow rates are critical in maintaining low effluent
SS and total P levels.
4.7.2 Elizabethtown, Pennsylvania (13)
The wastewater treatment plant serving the City of
Elizabethtown consists of screening, grit removal,
flow equalization, primary clarification, two stage
trickling filter system, rapid mix chambers, polymer
mix chambers, final clarification using clariflocculators,
and chlorination. Primary and secondary sludge is
blended, thickened, anaerobically digested, and
applied to agricultural land. Alum or ferric chloride is
added to the rapid mix chamber for phosphorus
removal. Polymer can be added immediately
downstream of the rapid mix chamber in a separate
mixing chamber. The NPDES permit requires the
plant to achieve the following effluent quality: BOD -
30 mg/l, SS - 30 mg/l, and total P - 2 mg/l.
Table 4-10 summarizes the performance of the plant
from July, 1985 through April, 1986. Alum and
polymer were used for phosphorus removal through
January of 1986, at which time plant personnel began
using ferric chloride because of its anticipated lower
cost. When ferric chloride addition began, polymer
was no longer used.
During the period of alum addition, average alum
dosage was 9.9 mg/l as Al, and average polymer
dosage was 0.4 mg/l. The ratio of aluminum applied
to phosphorus removed (AI:P) was 1.35. When alum
78
-------
Table 4-10. Summary of Performance at Elizabethtown, PA (13)
30-Day Average Values
BOD
SS
Month
July, 19851
August1
September1
October1
November1
December1
January, 19861
February2
March2
April2
Ave. July-Jan1
Ave. Feb-Apnl2
Flow
m3/d
5,870
5,830
6,130
6,020
6,960
7,420
7,270
10,600
8,590
7,000
6,510
8,740
In
mg/l
157
175
140
142
106
91
120
66
88
110
133
88
Out
mg/l
7
18
8
6
3
4
5
7
6
8
7
7
In
mg/l
189
220
202
243
279
249
179
119
171
207
223
166
Out
mg/l
14
32
10
9
12
15
14
16
15
12
15
14
In
mg/l
7.03
7.03
7.03
9.5
8.1
6.0
6.6
3.6
6.0
5.6
7.3
5.1
Out
mg/l
2.2
2.0
0.7
0.9
0.9
0.8
1.0
1.4
1.9
1.8
1.2
1.7
1 Alum and polymer used for P removal. Alum dose = 9.9 mg/l as Al (AI:P = 1.35). Polymer dose = 0.4 mg/l.
2 Ferric chloride used for P removal. FeCIa dose = 7.1 mg/l as Fe (Fe:P = 1.39).
3 Estimated values.
and polymer were replaced by ferric chloride in
February of 1986, average ferric chloride dosage was
7.1 mg/l as Fe, for an AI:P ratio of 1.39. Plant
performance was similar with alum and polymer vs.
ferric chloride. Concentrations of BOD, SS, and P in
the plant effluent were well within discharge
requirements.
Data on sludge production (dry solids basis) were not
available. It was estimated that approximately 40
percent of the sludge was chemical sludge when
alum and polymer were used, and approximately 30
percent of the sludge was chemical sludge when
ferric chloride was used.
Estimated unit costs for phosphorus removal,
including chemicals and chemical sludge disposal
were $3.52/kg ($1.60/lb) of P removed with alum and
polymer, and $2.73/kg ($1.24/lb) of P removed using
ferric chloride (1986 costs).
4.7.3 Little Hunting Creek, Virginia (13)
The Little Hunting Creek wastewater treatment plant
provides advanced secondary treatment using
single-stage trickling filters. Unit processes consist
of raw wastewater screens, grit chamber, primary
clarifiers, trickling filters, secondary clarifiers, and
chlorine contact tanks. Secondary sludge is returned
to the headworks for co-settling with primary sludge.
Waste sludge is conditioned with lime, dewatered by
vacuum filters, and hauled to the Lower Potomac
Pollution Control Plant for incineration. Phosphorus is
removed by ferric chloride addition to a manhole at
the headworks, and polymer addition: 1) just
downstream of the Parshall flume and 2) between the
trickling filters and final clarifiers. Effluent discharge
goals are 20 mg/l BOD, 20 mg/l SS, and 0.2 mg/l P
(voluntary).
Performance data for the period July 1984 through
June 1985 are shown in Table 4-11. The plant
effluent was consistently within NPDES permit
requirements for BOD and SS. Based on 30-day
average values, the plant met the voluntary effluent P
limit of 0.2 mg/l for nine of the twelve months.
Chemical requirements for the plant are high. Ferric
chloride dosage is approximately 125 mg/l (43 mg/l as
Fe), corresponding to a Fe:P ratio of 4.7. Total
dosage of polymer (two points of addition) is
approximately 2.8 mg/l. The high dosages are
indicative of difficulties involved in retrofitting the plant
for phosphorus removal.
Physical restrictions which prevent reduction of
chemical dosages include: 1) limited points for
chemical addition, 2) non-ideal conditions for mixing
and flocculation, 3) absence of a tertiary filter, and 4)
inability to accurately split the flow between the two
trickling filters. Dual point chemical addition is not
possible because the shallow depths of the
secondary clarifiers are insufficient to handle excess
chemical sludge.
Data on sludge production were not collected. It was
estimated that approximately 44 percent of the sludge
handled was chemical sludge from phosphorus
removal.
79
-------
Table 4-11. Summary of Performance at Little Hunting Creek, VA (13)
30-Day Average Values
BOD
Month
July, 1984
August
September
October
November
December
January, 1985
February
March
April
May
June
Average
Flow
m3/d
16,200
15,200
14,100
13,900
13,700
13,900
14,100
17,300
14,500
13,400
13,600
13,200
14,400
In
mg/l
169
167
183
180
186
196
202
153
160
164
162
169
175
Out
mg/l
10
10
8
8
11
11
11
14
14
11
9
13
11
SS
In
mg/l
190
183
200
188
191
169
197
147
149
149
179
177
177
Out
mg/l
8.6
6.6
7.6
7.0
6.5
6.8
8.1
9.3
8.8
9.3
10.0
10.3
8.3
P
In
mg/l
9.9
10.0
9.4
10.1
9.7
10.1
9.0
8.3
8.9
8.6
8.4
9.2
9.3
Out
mg/l
0.22
0.16
0.15
0.16
0.15
0.17
0.20
0.22
0.19
0.16
0.20
0.26
0.19
The unit cost of phosphorus removal including
chemicals, sludge conditioning, and chemical sludge
handling was estimated to be $13.39/kg P removed
($6.08/lb P). Approximately 68 percent of this cost
was attributed to sludge handling.
Table 4-12. Cost Estimates for Phosphorus Removal by
Mineral Addition.
Flow Rate
m3/d
3,785
37,850
Chemical
Liquid alum
Liquid FeCI3
Liquid alum
Liquid FeCI3
Chemical
Dose8
mg/l
65
50
65
50
Installed
Equipment
Costb.c
$
75,60Qd
65,700
122,800
116,800
Annual
O&M
Costb>d
$/yr
44,000
40,000
328,000
369,000
a Alum as AI2(S04)3«14H20
Dosages are average values from Reference 23 for mineral
addition to secondary processes to meet effluent TP = 1.0 mg/t.
bCost estimates from EPA report, Costs of Chemical
Clarification of Wastewater, 1976, updated to May 1987.
c Includes chemical storage and transfer facilities, chemical feed
pumps and piping, and rapid mix basin. Does not include
additional sludge handling equipment.
«J Includes O&M costs for labor, power @ $0.06/kWh,
maintenance materials and chemicals. Does not include O&M for
handling additional sludge generated. Chemical costs based on:
Liquid alum in bulk: $0.l7/kg
Liquid FeCI3 in bulk: $0.20/kg
4.8 Costs
Table 4-12 summarizes cost estimates for a
chemical addition system for phosphorus removal.
Capital and operation and maintenance costs are
presented for chemical feed systems for 3,785-m3/d
(1-mgd) and 37,850-m3/d (10-mgd) plants.
Sludge handling costs are not included. These costs
are planning level cost estimates (± 30 percent).
4.9 References
When an NTIS number is cited
reference is available from:
in a reference, that
2.
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
(703) 487-4650
DePinto, J.V., Switzenbaum, M.S., Young, T.C.
and J.K. Edzwald. Phosphorus Removal in Lower
Great Lakes Municipal Treatment Plants.
Proceedings of International Seminar on Control
of Nutrients in Municipal Wastewater Effluents,
Volume 1: Phosphorus. San Diego, CA.
September 9, 10, and 11, 1980, U.S.
Environmental Protection Agency, Cincinnati, OH,
1980.
Parker, D.S., de la Fuente, E., Britt, L.O.,
Spealman, M.L., Stenquist, R.J. and F.J. Zadick.
Lime Use in Wastewater Treatment: Design and
Cost Data. EPA-600/2-75-038, NTIS No.
80
-------
PB-248181, U.S. Environmental Protection
Agency, Cincinnati, OH, 1976.
Proceedings No. 1, Toronto, Canada, May 28-
29, 1973.
3. Process Design Manual for Phosphorus
Removal. EPA-625/1-76-001a, NTIS No.
PB-259150, U.S. Environmental Protection
Agency, Center for Environmental Research
Information. Cincinnati, OH, 1976.
4. Heim, N.E. and B.E. Burris. Chemical Aids
Manual for Wastewater Treatment Facilities.
EPA-430/9-79-018, NTIS No. PB-116816,
U.S. Environmental Protection Agency, Office of
Water Program Operations, Washington, DC,
1979.
5. Lime Handling Systems - Problems and
Remedies. U.S. Environmental Protection
Agency, Office of Municipal Pollution Control,
Washington, DC.
6. Rupke, J.W.G. Operational Experience in
Phosphorus Removal. In: Phosphorus
Management Strategies for Lakes, eds: R.C.
Loehr, C.S. Martin, W. Rast. Ann Arbor Science
Publishers, Inc., Ann Arbor, Ml, 1980.
7. Hatton, W. and A.M. Simpson. Use of Alternative
Aluminum Based Chemicals in Coagulation with
Particular Reference to Phosphorus Removal.
Environmental Technology Letters 6:225-230,
1985.
8. Gulp, R.L., Wesner, G.M. and G.L. Gulp.
Handbook of Advanced Wastewater Treatment,
Second Edition. Van Nostrand Reinhold
Company, New York, 1978.
9. Boyko, B.I. and J.W.G. Rupke. Phosphorus
Removal Within Existing Wastewater Treatment
Facilities. Canada-Ontario Agreement on Great
Lakes Water Quality, Project No. 71-1-1,
Ontario Ministry of the Environment, August,
1976.
10. Convery, J.J. Treatment Techniques for
Removing Phosphorus from Municipal
Wastewaters. NTIS No. PB-199072, U.S.
Environmental Protection Agency, Water Pollution
Control Research Series, 1970.
11. Black, S.A. Experience with Phosphorus Removal
at Existing Ontario Municipal Wastewater
Treatment Plants. In: Phosphorus Management
Strategies for Lakes, eds: R.C. Loehr, C.S.
Martin, W. Rast. Ann Arbor Science Publishers,
Inc. Ann Arbor, Ml, 1980.
12. Wilkes, W. Phosphorus Removal by Chemical
Addition Using Primary Treatment. In: Phosphorus
Removal Design Seminar, Conference
13. Retrofitting POTWs for Phosphorus Removal in
the Chesapeake Drainage Area.
14. Nutrient Control. WPCF Manual of Practice No.
FD-7, Water Pollution Control Federation,
Washington, DC, 1983.
15. Laughlin, J.E. Designing to Remove Phosphorus
by Using Metal Salts and Polymers in
Conventional Plants. Presented at EPA Design
Seminar for Wastewater Treatment Facilities,
Denver, CO, October 31-November 2, 1972.
16. Daniels, S.L. Instrumentation and Automatic
Control of Phosphorus Removal Processes. In:
Phosphorus Removal Design Seminar,
Conference Proceedings No. 1, Toronto, Canada,
May 28-29, 1973.
17. Culver, R.H. and D. Chaplick. Programming
Phosphate Treatment Saves Money. Water and
Sewage Works, March:84-87, 1978.
18. Damhaug, T. and K.T. Nedland. Alkalinity Feed
Forward Control of Alum Addition in Wastewater
Treatment-Field Experience. In: Management
Strategies for Phosphorus in the Environment,
Proceedings of the International Conference.
Lisbon, Portugal. Edited by J.N. Lester and
P.W.W. Kirk. Selper Ltd., London, 1985.
19. Damhaug, T., Ferguson, J.F. and H. Buset.
Control of Alum Addition in Wastewater
Treatment. Water Science Technology 13:531-
537, 1981.
20. Barth, E.F. and H.D. Stensel. International
Nutrient Control Technology for Municipal
Effluents. JWPCF 53(12):1691-1701, 1981.
21. Clarifier Design. WPCF Manual of Practice No.
FD-8, Water Pollution Control Federation,
Washington, DC, 19xx.
22. Singhal, A.K. Phosphorus and Nitrogen Removal
at Cadillac, Michigan. JWPCF 52(11):2762-
2770, 1980.
23. Process Design Manual for Suspended Solids
Removal. EPA-625/1-75-003a, NTIS No.
PB-259147, U.S. Environmental Protection
Agency, Center for Environmental Research
Information, Cincinnati, OH, 1975.
24. Simplified Procedures for Water Examination.
AWWA Manual M12, American Water Works
Association.
81
-------
25. Archer, J.D. Summary Report on phosphorus
removal. Canada-Ontario Agreement on Great
Lakes Water Quality, Project No. 75-1-42,
Ontario Ministry of the Environment, Toronto,
1975.
26. Wastewater Treatment Plant Design. ASCE
Manual No. 36, WPCF Manual of Practice No. 8,
American Society of Civil Engineers, New York,
NY and Water Pollution Control Federation,
Washington, DC, 1977.
27. High Level Phosphorus Removal from W.P.C.P.
Effluent; City of Orillia; Final Report. Prepared by
Ainley and Assoc. for Ontario Ministry of the
Environment, Rexdale, Ontario, September, 1982.
82
-------
CHAPTER S
Sludge Handling
5.1 Introduction
The importance of sludge handling in the design of
new facilities or in the retrofitting of existing plants for
phosphorus removal cannot be over emphasized. The
difficulty in predicting generation rates and
characteristics of sludges derived from phosphorus
removal operations is further complicated by the wide
variability of data and the inconsistent information
reported from operating phosphorus removal facilities.
Retrofitting of existing facilities to achieve phosphorus
removal presents difficult design challenges with
respect to sludge handling. For example, addition of
chemicals to precipitate phosphorus from the liquid
stream is relatively straightforward and can be
implemented with relatively small capital expenditure.
However, the additional sludge generated and the
impact on sludge characteristics can easily result in
overloading of existing thickening, stabilization, and
dewatering equipment, or require significant increases
in operational staff time to operate and maintain this
equipment. Installation of new capital-intensive
equipment may be required to effectively process the
additional sludge. On the other hand, at some
facilities, implementation of a phosphorus removal
system may not significantly impact the existing
sludge handling operation. Such variability of
experience with sludge handling at plants removing
phosphorus makes development of firm design
recommendations difficult.
The objectives of this chapter are to:
1. Present information on the quantities and
characteristics of sludges derived from chemical
and biological phosphorus removal processes.
2. Provide specific recommendations for the design of
sludge handling processes and equipment used in
plants removing phosphorus.
3. Summarize the experience with sludge handling
processes at full-scale phosphorus removal
facilities.
It should be noted that, in some cases, specific
design criteria cannot be presented because of the
wide variability in sludge characteristics, resulting in
variations in performance of sludge handling
processes and equipment. In these instances, full-
scale experience is summarized and a design
approach is outlined.
5.2 Current Practice for Handling
Chemical Sludges
5.2.7 Introduction
In 1976 and 1977, an extensive EPA-sponsored
survey was conducted of 174 municipal wastewater
treatment plants in the United States and Canada
which used chemicals to remove phosphorus (1). The
purpose of this survey was to quantify the effects of
chemical addition on the sludge handling and disposal
operations at full-scale plants.
This section summarizes the results of this survey
with regard to the types of facilities, points of
chemical addition, sludge generation rates, and
prevalence of various sludge treatment and disposal
methods. Information regarding the impacts of
chemical addition on unit sludge handling processes
is provided under those sections of this chapter which
deal with sludges derived from use of specific
chemicals.
5.2.2 Points of Chemical Addition and Methods of
Combining Sludges
Table 5-1 shows a breakdown of the 174 plants by
point(s) of chemical addition and by how the chemical
sludges are combined with organic sludges at
different locations within the plant.
Chemical addition to secondary treatment was the
most common option practiced, accounting for 62
percent of the plants surveyed. Twenty-six percent
of the plants added chemicals to primary treatment, 5
percent of the plants added chemicals to both primary
and secondary treatment, and 6 percent of the plants
added chemicals to tertiary facilities.
Of the plants practicing chemical addition to primary
treatment and generating both primary and secondary
sludges, 94 percent of the plants combined their
83
-------
Table 5-1. Combination of Chemical Sludges with Other Sludges for Processing as Practiced by Plants in EPA Survey (1).
Point at Which Chemical
Sludges and Other
Sludges were Comvined
Number of Plants by Point(s) of Chemical Addition
Primary
Addition
Secondary
Addition
Tertiary
Addition
Primary and
Sec. Addition
Primary and
Tert. Addition
Sec. and Tert.
Addition
Total
Plants w/o primary or
secondary clarification
No primary clarifier
No secondary clarifier
NA1
13
28
NA
1
NA
NA
NA
NA
NA
1
NA
30
13
Plants with primary and
secondary clarification
Sludges not combined
Combined in or before
primary clarifier
Combined in or before
thickener
Combined in or before
digester
Combined in or before
dewatering device
Total
2
12
13
6
0
46
10
26
14
24
6
108
6
0
2
2
0
11
1
3
1
3
0
8
0
0
1
0
0
1
0
0
0
0
0
1
19
41
31
35
6
1752
1 Not applicable.
2 One plant with tertiary addition is counted twice because it has no primary sludge and it has secondary and tertiary sludges which are not
combined.
chemical primary sludges with organic secondary
sludge before processing. Of these plants combining
their sludges, 39 percent accomplished this by
returning secondary sludge to the primary clarifiers.
The remainder combined their sludges in or before
thickening, stabilization, or dewatering processes.
Of the plants practicing chemical addition to
secondary treatment and which generated both
primary and secondary sludges, 88 percent combined
their sludges before processing. Of the plants
combining their sludges, 37 percent practiced this by
returning secondary sludge to the primary clarifier.
5.2.3 Sludge Generation Rates and
Characteristics
The additional sludge generated from use of
chemicals for phosphorus precipitation is a key
design consideration. Although chemical addition to
the liquid stream is easily implemented, the impacts
on thickening, stabilization, dewatering, and disposal
operations are often severe. This is due to both the
additional mass and volume of sludge generated as
well as the effects on thickening and dewatering
characteristics.
Figure 5-1 summarizes volumes of sludge produced
(before thickening) per cubic meter of influent flow as
reported in the EPA survey of 174 plants (1). It is
apparent that lime addition produced significantly
greater volumes of sludge than mineral salt addition.
Chemical addition to secondary processes produced
greater volumes of sludge than chemical addition to
the primary when considering primary and secondary
processes separately. However, on a total plant basis,
the additional sludge volumes resulting from chemical
addition were approximately equivalent for both
primary and secondary chemical addition.
Average increases in sludge volume before thickening
were 25 percent for iron salt addition, 58 percent for
aluminum salt addition, and several hundred percent
for lime addition (1). Bear in mind that these figures
are volumes before thickening, and as such do not
take into account the relative thickening
characteristics of sludge which will affect volumes
requiring dewatering; nor do they consider dewatering
characteristics which will affect volumes requiring final
disposal.
It is important to note that the reported sludge
volumes represent a diverse array of plants with wide
ranges of wastewater characteristics, treatment
process configurations, operating modes, and effluent
quality, all of which significantly impact sludge
generation rates. Therefore, these figures are for
illustrative purposes only, and should not be used for
design.
Figure 5-2 summarizes mass sludge production data
from the EPA survey (1). Values are expressed as
kilograms of dry solids per cubic meter of influent
flow. Again, it should be noted that these figures
represent plants with wide ranges of wastewater and
process characteristics, which can significantly impact
the mass of sludge generated. Use of these figures
for design purposes or chemical selection is not
recommended.
Table 5-2 summarizes data on solids content of
chemical sludges at the 174 plants surveyed in the
EPA-sponsored study. The total and volatile solids
content are for the sludges prior to thickening or other
84
-------
Figure 5-1 Volumetric sludge production from EPA survey - m3 sludge/m3 plant influent (1).
Chemical Added
to Primary-
Primary
Sludge
Only
35,200
4,190
2,710
Chemical Added
to Secondary-
Secondary
Sludge
Only
17,080
8,300
60,350
Chemical Added to Primary -
Total Plant Sludge
5.88C
4,710 4.750
3,000
| I Iron Salt Addition Plants
^^m Iron Salt Addition Plants Prior to P Removal
E%%1 Lime Addition Plants
Chemical Added to Secondary •
Total Plant Sludge
4,650 4,640
5.71C
5.110
Aluminum Salt Addition Plants
Aluminum Salt Addition Plants Prior to P Removal
85
-------
Figure 5-2 Mass sludge production from EPA survey - kg sludge/m3 plant influent (1).
Chemical Added
to Primary-
Primary
Sludge
Only
0.44
0.18
0.12
Chemical Added
to Secondary-
Secondary
Sludge
Only
0.10
0.09
Chemical Added to Primary -
Total Plant Sludge
0.54
0.34
0.27
0.17
0.29
I
I
Chemical Added to Secondary
Total Plant Sludge
0.22
0.21
0.18
0.17
I 1 Iron Salt Addition Plants
Iron Salt Addition Plants Prior to P Removal
Lime Addition Plants
Aluminum Salt Addition Plants
Aluminum Salt Addition Plants Prior to P Removal
86
-------
processing. For chemical addition to primary
treatment, iron sludges had the highest average total
solids content at 5.26 percent compared to alum
sludge at 3.95 percent and lime sludge at 1.1
percent. For chemical addition to secondary
treatment, secondary biological-iron sludge had an
average total solids content of 0.93 percent vs. 1.41
percent for biological-aluminum sludge. When
combined with primary sludges, the sludge solids
contents for iron and aluminum addition to secondary
treatment were similar at 4.13 and 3.82 percent,
respectively. Although these data are useful for
illustrative purposes, the values are averages
obtained from plants with different wastewater and
process characteristics. This may explain some
inconsistencies in the data. For example, for iron
addition to primary treatment, total solids for primary
sludge is 5.26 percent, while total solids for the
combined (total plant) sludge is 5.73 percent. In
practice, this would not occur, since secondary
sludge always has a significantly lower total solids
content than primary sludge.
5.2.4 Prevalence of Various Sludge Treatment and
Disposal Options
Table 5-3 summarizes the responses from plants
contacted in the EPA survey regarding the types of
processes employed for thickening, stabilization,
conditioning, dewatering, drying, reduction, and
disposal of sludge.
Of the thickening options, gravity thickening was by
far the most prevalent, as would be expected. In most
cases, flotation thickening was used for secondary
sludge only. The five plants which used centrifuges
for thickening added lime to primary treatment for
phosphorus removal.
Aerobic and anaerobic digestion were the most
prevalent stabilization techniques. Of the plants using
iron salts, 20 percent used aerobic digestion. Of the
plants using aluminum salts, 46 percent used aerobic
digestion, while 54 percent used anaerobic digestion.
Chemical conditioning was the most prevalent method
of conditioning, followed by thermal conditioning.
Seventy-five percent of the plants using iron salts
and 74 percent of the plants using aluminum salts
employed chemical conditioning.
Sand drying beds and vacuum filters were the most
prevalent dewatering techniques. Of the plants using
iron salts, 49 percent used sand drying beds and 34
percent used vacuum filters. Of the plants using
aluminum salts, 44 percent used sand drying beds,
24 percent used vacuum filters, and 11 percent used
centrifuges. None of the plants using iron employed
centrifuges for dewatering.
Three plants in the survey reported using heat dryers
for further sludge volume reduction, and 22 plants
employed multiple hearth or fluidized bed incineration.
The most prevalent means of final sludge disposal
was application to agricultural land, lawns, or gardens,
accounting for 52 percent of the plants surveyed.
This was followed by sanitary landfill (27 percent),
private or authority-owned dump site (16 percent),
and land reclamation (5 percent).
5.3 Sludge Derived from Addition of
Aluminum Salts
5.3.7 Sludge Characteristics
Aluminum salts may be employed for phosphorus
removal by addition to primary, secondary, or tertiary
treatment processes. Secondary addition is the most
common, accounting for approximately 82 percent of
the plants using aluminum salts as determined by the
EPA survey (1). Addition of aluminum to primary
treatment accounted for only about 14 percent of the
plants using aluminum salts. The few remaining plants
added aluminum salts to tertiary treatment (3 percent)
or to both primary and secondary treatment (1
percent).
Sludge characteristics are dependent on numerous
factors including wastewater composition, chemical
dosage, point of chemical addition (primary,
secondary or tertiary), whether the various sludges
are combined before processing and in what
proportions, detention time in clarifiers or holding
tanks, and other factors. Although it is difficult to
predict the impact of chemical addition on thickening
and dewatering characteristics, two conclusions can
be drawn with some certainty:
1. That the addition of aluminum salts will result in an
increase in sludge volume and mass, and
2. The thickening and dewatering characteristics will
likely be different than if no chemical were added.
Actual experience with processing chemical -
biological sludges has been extremely variable. In
some cases, chemical addition has improved
thickening and dewatering characteristics, while in
other cases chemical addition has had a detrimental
impact. Many problems can be traced to overloading
of unit processes due to increased volumes of sludge
from chemical addition.
Knowledge of general sludge characteristics and
chemical dosages is generally of little value in
predicting the amenability of sludges to thickening,
dewatering, and stabilization processes. This is due to
the interaction and impact of the host of factors which
affect performance of sludge processing operations.
Design of sludge handling systems must therefore be
87
-------
Table 5-2. Combination of Chemical Sludges with Other Sludges for Processing as Practiced by Plants in EPA Survey (1).
Sludge Characteristics (before processing)
Type of Chemical Sludge and Whether
Combined with Other Plant Sludge(s)
Iron addition to primary
Primary sludge
Total plant sludge
Iron addition to secondary
Secondary sludge
Total plant sludge
Iron addition to tertiary
Tertiary sludge
Total plant sludge
Alum addition to primary
Pnmary sludge
Total plant sludge
Alum addition to secondary
Secondary sludge
Total plant sludge
Lime addition to primary
Primary sludge
Total plant sludge
Lime addition to tertiary
Tertiary sludge
Total plant sludge
i oiai ooiio
Range
3.4 - 8.0
2.31 - 10.0
0.2 - 4.0
0.5 - 7.75
4.0
4.64 - 5.0
3.3 - 43.5
3.96 - 5.0
0.4 - 4.4
1.0 -7.0
0.7 - 1.5
0.64 - 0.82
2.5 - 4.0
1.95
s, percent
Average
5.26
5.73
0.93
4.13
4.0
4.82
3.95
4.49
1.41
3.82
1.1
0.73
3.3
1.95
voiatue s>(
Range
45-69
40 -70
50-70
45 - 72
35
62
61 -67
46-70
60-78
52-70
N/A1
N/A
11-30
39
anas, percent ot 1 1>
Average
57
57
62
62
35
62
65
59
67
59
N/A
N/A
21
39
1 Not available.
based, at a minimum, on laboratory tests such as
settlometer tests, specific resistance, filterleaf,
capillary suction time, and other tests. Where
possible, pilot- or full-scale units should be used to
establish design criteria for full-scale sludge
thickening and dewatering devices. Specific tests
used to characterize sludges resulting from chemical
phosphorus precipitation processes are discussed in
subsequent sections of this manual.
5.3.2 Sludge Generation Rates
As previously discussed, addition of chemicals to
remove phosphorus will increase the mass of sludge
generated. This increase can be attributed to three
components (2):
1. Formation and removal of chemical solids such as
metal phosphates and metal hydroxides,
2. Improved removals of organic solids during
clarification, and
3. Removal of dissolved solids.
Several methods are available to estimate sludge
generation rates for design purposes. When an
existing plant is to be upgraded for chemical
phosphorus removal, the preferred approach is to
conduct full-scale trials at the facility and measure
sludge production under expected operating
conditions. This allows optimization of chemical
dosages and determination of the impact of
operational changes such as modification of sludge
age. For new facilities, pilot plant tests are the best
approach to estimate sludge generation rates.
However, such studies are costly and may not be
economically feasible for small plants. In such cases,
jar tests should be performed to provide data to
estimate additional sludge generation resulting from
chemical addition, although such tests may not be
representative of full-scale, dynamic conditions.
Procedures are also available to allow calculation of
the theoretical mass of solids generated from
wastewater treatment schemes employing chemical
phosphorus precipitation (2-4). An approach is
described below for estimating the quantity of
additional sludge resulting from use of alum for
phosphorus removal. A procedure for calculating
baseline sludge production (no chemical addition) has
not been shown, since this is described in detail in
Reference 4 for several types of treatment (e.g.
processes activated sludge, trickling filter, etc.).
Additional sludge generated as a result of alum
addition is due to formation and removal of chemical
solids, improved removal of suspended solids, and
removal of dissolved solids. Another consideration in
estimating sludge production is that the quantity of
biological sludge which is generated may be affected
by selection of the point of chemical addition. For
example, while alum addition to the primary clarifier
will increase the production of primary sludge,
improved BOD and SS removal in primary treatment
will reduce the organic loading to downstream
88
-------
Table 5-3. Prevalence of Treatment and Disposal Processes for Chemical Sludges Among Plants in EPA Survey (1)*.
Major Chemical Used and Point of Addition
Sludge Treatment
and Disposal Unit
Thickening
Gravity
Flotation
Centrifuge
Stabilization
Composting
Aerobic digestion
Anaerobic digestion
Lime stabilization
Conditioning
Chemical conditioning
Elutriation
Thermal conditioning
Dewatering
Pressure filter
Sand drying bed
Centrifuge
Vacuum filter
Lagoon
Hor. moving screen
concentrator
Cylindrical rotating
gravity filter
Heat Drying
Flash dryer
Multiple hearth dryer
Reduction
Incineration
Final Disposal
Land reclamation
Sanitary landfill
Agricultural fields
lawns, gardens
Private- or authority-
owned dump site
Primary
18
1
0
1
4
27
0
11
1
2
1
17
0
12
4
0
0
0
0
7
0
16
18
8
Iron Salt
Sec.
25
4
0
0
11
37
0
12
2
4
3
20
0
13
4
0
0
0
1
5
1
18
39
10
Aluminum Salt
Tertiary
2
0
0
0
1
1
0
1
0
0
0
0
0
1
1
0
0
0
0
0
0
1
1
0
Primary
4
0
0
1
1
5
1
2
0
1
1
2
1
1
1
0
0
0
0
0
1
1
7
2
Sec.
18
5
0
0
24
25
0
12
0
4
1
18
4
10
3
2
1
1
1
5
6
13
30
8
Tertiary
1
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
1
0
Primary
8
0
5
0
0
1
0
2
0
0
2
1
0
0
0
0
0
0
0
5
1
4
5
2
Lime
Sec.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Tertiary
0
0
0
0
0
0
0
0
0
0
1
0
1
0
2
1
0
0
0
0
0
0
1
1
Total
76
10
5
2
41
97
1
40
3
11
8
58
6
37
16
3
1
1
2
22
9
53
102
31
* One plant may use more than one method of thickening, dewatenng, etc.
biological processes and, hence, reduce the quantity
of biological solids generated. A procedure for
estimating quantities of additional sludge resulting
from alum addition is given below.
a. Chemical sludge
This calculation assumes that the aluminum reacts
with phosphorus compounds first, and that excess
aluminum forms aluminum hydroxide.
Given:
Pin = 8 mg/l
pout = 1 mg/l
AI:Pdose = 2.2:1
Atomic weight of P = 31
Atomic weight of Al = 27
Atomic weight of AIPO4 =122
Atomic weight of AI(OH)3
Al dose = 2.2 x 8 x (27/31) = 15 mg/l
Stoichiometry:
Al + P04 = Al PO4
Al + 3 OH = AI(OH)3
(8 - 1 mg/l P) v 31 = 0.23 mmole/l AIPO4
produced
(15 mg/l Al) * 27 = 0.56 mmole/l Al added
0.56 - 0.23 = 0.33 mmole/l Al in excess to AI(OH)3
AIPO4 sludge: 0.23 mmole/l x 122 = 28.1 mg/l AIPO4
AI(OH)3 sludge: 0.33 mmole/l x 78 = 25.7 mg/l
AI(OH)3
Total chemical sludge: 28.1 mg/l AIPO4
25.7 mg/l AI(OH)3
89
-------
53.8 mg/l Al sludge
Chemical sludge produced = 53.8 mg sludge/liter of
wastewater treated (449 lb/106 gallons).
Because the stoichiometry is only an approximation of
the chemical reactions which occur, and because
some data indicate greater quantities of chemical
sludge than predicted, it has been recommended that
the calculated sludge production value be increased
by 35 percent (2).
Design estimate of chemical sludge production:
1.35 x 53.8 mg/l = 72.6 mg sludge/liter of wastewater
treated (605 !b/106 gallons)
b. Sludge from improved removal of suspended
solids
Additional sludge resulting from improved suspended
solids removal is calculated by assuming a greater
removal efficiency. For example, in primary
clarification without chemical addition, removals of
suspended solids are typically 50 percent. With alum
addition, however, removal efficiencies may be 75
percent or greater. The additional sludge generated
by chemical addition to the primaries can be
calculated as shown below.
Given:
SSjn = 200 mg/l
SS removal efficiency (no chemicals) = 50 percent
SS removal efficiency (with alum) = 75 percent
Additional sludge generated:
(0.75 - 0.50) x 200 mg/l = 50 mg sludge/liter of
wastewater treated
(417 lb/106 gallons)
When aluminum is added to the secondary biological
process such as the activated sludge basin, additional
sludge production from improved SS removal may not
be evident. However, generation of chemical solids
(AIPO4 and AI(OH)3) can be assumed to be
approximately equal regardless of the point of
chemical addition. For a well operated standard rate
activated sludge plant, addition of aluminum salts to
the aeration basin may have little impact on
clarification efficiency. However, for some high rate
activated sludge or trickling filter plants, secondary
clarification efficiency may improve significantly,
resulting in additional sludge production. An
improvement in effluent suspended solids
concentration from 20 to 10 mg/l will result in 10 mg
of additional sludge per liter of wastewater treated (83
Ib per million gallons).
For tertiary applications of aluminum following
secondary biological treatment, additional sludge
resulting from improved removal of SS will generally
be small. Estimates of tertiary sludge production from
SS removal are based on anticipated SS
concentrations in secondary and tertiary effluents.
c. Sludge from removal of dissolved solids
Data exist to show that dissolved solids are removed
as a result of chemical addition (2). Reported
removals of soluble TOG are about 30% percent
using alum. Removal of soluble COD has been
reported to be approximately 40 percent. The sludge
mass resulting from removal of dissolved solids must
be estimated indirectly from soluble TOG, COD, or
BOD loadings. The following relationships can be
used for this purpose. The derivation of these
relationships may be found in Reference 2.
Sludge mass resulting from removal of dissolved
solids (assuming 30 percent removal of soluble
organics)
= STOCjn (mg/l) x 0.30 x 2.5 x 1.18
= SCODjn (mg/l) x 0.30 x 1.1 x 1.18
= SBODjn (mg/l) x 0.30 x 1.6 x 1.18
where:
STOCjn = soluble TOG in influent
SCODjn = soluble COD in influent
SBOD,n = soluble BOD in influent
The latter two equations are only applicable to
influents prior to biological oxidation processes (2).
d. Design example
Estimate the sludge production from a 3,785-m3/d
(1-mgd) conventional activated sludge plant with and
without alum addition to the primary clarifier.
Given:
Pin = 8 mg/l
BODjn = 200 mg/l
SSjn = 220 mg/l
VSS = 0.75 x SS
STOCjn = 50 mg/l
Sludge from primary treatment without alum addition:
Assume 50 percent SS removal, 30 percent BOD
removal;
0.5 x 220 mg/l x 3,785 m3/d x 0.001 =416 kg/d
(917lb/d)
Sludge from primary treatment with alum addition:
Assume 90 percent P removal, 75 percent SS
removal, 50 percent BOD removal, 30 percent STOC
90
-------
removal (from test results); assume AI:P dosage ratio
of 2.0 (from test results)
Al dose = (2.0 x 8 mg/l) x (27/31) = 13.9 mg/l Al
P removed = 0.90 x 8 mg/l = 7.2 mg/l P
AIPO4 sludge = 7.2/31 = 0.23 mmole/l AIPO4
Total Al added = 13.9/27 = 0.51 mmole/l Al
Excess Al = 0.51 - 0.23 = 0.28 mmole/l Al
AIPO4 sludge = 0.23 x 122 = 28.1 mg/l AIP04
AI(OH)3 sludge = 0.28 x 78 = 21.8 mg/l AI(OH)3
Total chemical sludge produced:
28.1 mg/l + 21.8 mg/l = 49.9 mg/l
Design chemical sludge production:
49.9 mg/l x 1.35 = 67.4 mg/l
or
67.4 mg/l x 3,785 m3/d x 0.001 = 255 kg/d (562 Ib/d)
Sludge from SS removal:
0.75 x 220 mg/l x 3,785 m3/d x 0.001 = 625 kg/d
(1,375 Ib/d)
Sludge from removal of dissolved solids (DS):
50 mg/l x 0.30 x 2.5 x 1.18 = 44.2 mg/l
or
44.2 mg/l x 3,785 m3/d x 0.001 = 167 kg/d (368 Ib/d)
Secondary sludge - no alum addition to primary
(from reference 4):
WAST = PX +
nv
Px = Y (Sr) - Kd (M)
= (Y) (Sr) /(1 + Kd SRT)
where:
WASj = waste activated sludge production, kg/d
Px = net growth of biological solids (VSS), kg/d
lnv = inert SS fed to process, kg/d
ET = effluent SS, kg/d
Y = gross yield coefficient, kg/kg
Sr = BOD removed, kg/d
Kd = decay coefficient, day1
M = system inventory of microbial solids
(VSS), kg
SRT = solids retention time, days
Assume:
Y = 0.67 kg/kg from test data
Kd = 0.06 day-1 from test data
BODout = 15 mg/l
SRT = 10 days
BODm - BODout = (1 - 0.3) (200 mg/l) - 15 mg/l
Sr = 125 mg/l x 3,785 m3/d x 0.001
= 473 kg/d
Px = (0.67 kg/kg x 473 kg/d)/[1 + (.06 d-1) (10 d)]
= 198 kg/d
Inert SS feed = (1 - 0.5) (1 - 0.75) (220 mg/l)
= 27.5 mg/l
lnv = 27.5 mg/l x 3,785 m3/d x 0.001
= 104 kg/d
Assume effluent SS = 20 mg/l;
ET = 20 mg/l x 3,785m3/d x 0.001
= 76 kg/d
WAST = Px + lnv - ET
= 198 kg/d +104 kg/d - 76 kg/d
= 226 kg/d (498 Ib/d)
Secondary sludge - alum addition to primary:
Calculate reduced BOD loading due to improved BOD
removal in primaries;
BODjn = (1 - 0.5) (200 mg/l) (3,785 m3/d) (0.001)
= 379 kg/d
BODout = 15 mg/l x 3,785 m3/d x 0.001
= 57 kg/d
(assumes no improvement in secondary effluent
quality)
Sr = 379 kg/d - 57 kg/d = 322 kg/d
Px = (0.67 kg/kg x 322 kg/d)/[1 + (.06 d-1) (10 d)]
= 135 kg/d
Inert SS feed = (1 - 0.75) (1 - 0.75) (220 mg/l)
= 13.8 mg/l
lnv = 13.8 mg/l x 3,785 m3/d x 0.001
= 52 kg/d
Assume effluent SS = 20 mg/l;
ET = 20 mg/l x 3,785m3/d x 0.001
= 76 kg/d
(assumes no improvement in secondary effluent
quality)
WAST = Px + lnv - ET
= 135 kg/d +52 kg/d - 76 kg/d
= 111 kg/d (224 Ib/d)
91
-------
Summary of sludge production calculations:
Table 5-4. Typical Gravity Thickener Design Criteria.
Sludge production, kq/d
w/o alum w/alum
Primary clarifier
SS sludge
DS sludge
Chemical Sludge
Total
Secondary clarifier
WAS
Total sludge
5.3.3 Thickening
416
0
0
416
226
642
625
167
255
1,047
111
1,158
5.3.3.1 Gravity Thickening
Relatively little definitive design and performance
information is available in the literature for gravity
thickening of alum sludges. Thickening characteristics
of sludges generated from plants practicing chemical
phosphorus precipitation will vary depending on
wastewater characteristics, point of chemical addition,
chemical dosage, wastewater treatment processes
employed, whether sludges are combined before
thickening, relative proportions of chemical and
biological sludges, sludge characteristics before
thickening, and other factors. Because of the wide
variability in thickening characteristics, thickening
tests should be conducted whenever the actual
sludge is available, as in the case of retrofitting or
expanding existing facilities. In such cases, full-scale
trials with chemical addition are recommended to
generate representative samples of sludge for
conducting thickening tests. Procedures for
conducting such tests may be found elsewhere (5-
8).
In some cases where new treatment facilities are to
be designed, the design engineer does not have the
benefit of having existing sludges on which to
conduct thickening tests, and must rely on published
design guidelines. Table 5-4 provides design criteria
for thickeners receiving various types and
combinations of alum and non-chemical sludges.
In addition to mass loading criteria shown in Table
5-4, hydraulic loading must also be considered in
the design of gravity thickeners. For primary sludges,
typical maximum overflow rates are 24-30 m3/m2/d
(600-720 gpd/ft2). However, for waste activated
sludge or combinations of primary and waste
activated sludges, hydraulic loadings rates should be
lower than this, 4-8 m3/m2/d (100-200 gpd/ft2) (4).
Experience with thickening of alum sludges in
combination with non-chemical sludges has been
highly variable. Some reports indicate an
improvement in thickening characteristics of sludges
Sludge Type
Primary
Primary w/alum1
Primary + WAS
(Primary w/alum)
+ WAS
Primary + (WAS
w/alum)
WAS
WAS w/alum
Tertiary w/alum2
Influent
Solids
Cone.
percent
2-7
2.1-3.7
0.5-1.5
-
0.2-0.4
0.5-1.5
0.5-1.5
0.5-1.5
Expected
Underflow
Cone.
percent
5-10
2.5-6.7
4-6
-
4.6-6.5
2-3
1.5-2.0
2.5-3.0
Mass
Loading
kg/m2/hr
3.9-5.9
0.4-1.0
1.0-2.9
-
2.4-3.4
0.5-1.5
0.6-0.8
0.6-1.0
Ref.
g
10
8
-
8
9
11
12
1 Data reflect use of anionic polymer to assist in thickening.
2 From water treatment plant sludge containing alum and
powdered activated carbon.
when alum addition was initiated; others show
detrimental impacts on thickener performance (1,3).
Unfortunately, in the latter cases, it is difficult to
determine whether poor performance was due to the
change in sludge characteristics or the additional
solids loading from increased sludge production.
In pilot-studies on alum-primary sludges, it was
found that underflow solids concentration generally
decreased as the amount of chemical solids in the
feed sludge increased (10). Note that in Table 5-4
recommended mass loadings to gravity thickeners
decrease substantially for primary sludge with alum
vs. primary sludge alone.
Several investigators have noted an improvement in
the settlability (decrease in SVI) of mixed liquor solids
where aluminum salts are dosed directly into the
aeration basin (14-16). Little data are available
regarding thickening characteristics of the resulting
sludges, however. In general, where alum is added to
the aeration basin of the activated sludge process,
recommended loadings to gravity thickeners are
similar to those for which no chemicals are added.
No data were found for thickening of tertiary
wastewater sludges resulting from alum addition. In
many cases, alum is added to the secondary effluent
immediately prior to tertiary filtration without
intermediate clarification. Tertiary alum sludges would
be expected to behave similarly to alum sludges from
potable water treatment.
Addition of polymer to the influent sludge of a gravity
thickener has been found to improve solids capture
and increase underflow solids concentrations (1, 10).
5.3.3.2 Flotation Thickening
Flotation thickening is often employed for thickening
of waste activated sludge from extended aeration
92
-------
plants without primary clarification, or for secondary
sludges where primary and secondary sludges are
thickened separately. In the majority of cases,
flotation thickening is not applied to primary sludges
or combined primary-secondary sludges. Where
chemicals are added to the primary clarifier for
phosphorus removal, organic loadings to downstream
biological processes are reduced substantially, thus
reducing the quantity of biological sludge generated.
In such cases, separate thickening of primary and
secondary sludges is not justified and the sludges
would most likely be combined prior to thickening by
gravity.
In the EPA-sponsored survey of plants removing
phosphorus, it was found that 6 percent of the
facilities employed flotation thickening (1). In all but
one of these plants, flotation thickening was used on
iron or aluminum waste activated sludge alone. Half
of these facilities aerobically digested the sludge
before thickening. Of the nine plants surveyed, four
reported significant changes in flotation thickener
performance when processing chemical sludges; four
reported no change at all. The ninth plant did not
comment on the impacts of chemical sludge handling.
Of the four plants reporting significant impacts, three
reported negative impacts, which included:
1. The need to reduce hydraulic loading rate,
2. The need to use both anionic and cationic
polymers, and
3. A decrease in float (thickened) solids
concentration.
The plant reporting a positive impact indicated an
increase in float solids concentration. Some plants
noted improved performance with polymer addition.
Of the nine plants that used flotation thickening, four
added alum to the aeration basins, and two of these
plants aerobically digested the sludge prior to
thickening. Feed solids concentrations were 0.5-1.5
percent, while float solids were typically 3.5-5
percent (1).
In batch laboratory flotation tests on aerobically
digested waste activated sludge from a contact-
stabilization plant, no difference was found between
the thickening properties of the control digester and
alum digester sludge. The sludge was concentrated
from 0.75 percent TSS to 3.5 percent TSS without
polymer addition at an air: solids ratio of 0.03 (17).
Mass loadings of waste activated sludge (no
chemicals) to dissolved air flotation thickeners vary
widely in practice, ranging from 2.8 to 34.0 kg/m2/hr
(0.6-6.9 Ib/ft2/hr) (4). Typical design loadings for
flotation thickeners receiving non-chemical waste
activated sludge are 2.4-3.9 kg/m2/hr (0.5-0.8
Ib/ft2/hr) without polymer and 4.9-7.3 kg/m2/hr (1.0-
1.5 Ib/ft2/hr) with polymer (8) [In Chapter 4 of
Reference 8, Sludge Thickening, WPCF MOP No.
FD-1, all solids loadings shown in metric units are
calculated incorrectly; use only values shown in
English units; kg/m2/hr = 4.9 x Ib/ft2/hr].
For dissolved air flotation thickening of waste
activated sludge with alum, solids loading rates of
10-24 kg/m2/hr (2-5 Ib/ft2/hr) with polymer have
been recommended (11).
Pilot-studies have been conducted on flotation
thickening of chemical primary sludges alone (10).
Unfortunately, flotation thickening is seldom used for
primary sludge. Solids loadings for alum sludges were
6.8-24.9 kg/m2/hr (1.4-5.1 Ib/ft2/hr). Feed solids
(TSS) were 0.8-2.7 percent, and float solids (TSS)
were typically 2.5-5.0 percent. Polymer addition was
necessary to achieve good performance (10).
5.3.4 Stabilization
5.3.4.1 Aerobic Digestion
In the EPA survey of 174 plants removing
phosphorus, 41 plants (24 percent) employed aerobic
sludge digestion (1). Only four plants reported
problems with stabilization of chemical sludge; the
majority of these problems were directly attributed to
increased sludge volumes from chemical addition. In
some cases, generation of a more concentrated
sludge as a result of alum addition necessitated
increasing the air supply to maintain adequate mixing
of the digester contents.
No deterioration of supernatant quality was reported
as a result of increased sludge volumes or changed
sludge characteristics upon chemical addition (1).
Laboratory studies on waste activated sludge with
alum showed little impact of the presence of
aluminum precipitates on the aerobic digestion
process (18,19).
In a full-scale study of alum addition at a 1.7 mgd
contact-stabilization plant, TSS reduction in that
alum sludge during aerobic digestion was only 12
percent vs. 31 percent in the control sludge. This
does not provide an adequate measure of the degree
of organics oxidation which occurred, however, since
the alum sludge would be expected to have a higher
percentage of non-volatile solids. Data on volatile
solids reduction were not provided. The digested
alum sludge thickened significantly better than the
control when subjected to laboratory tests (17).
There has been some concern regarding the possible
release of phosphorus from the solid to the liquid
phase during anoxic storage of aerobically disgested
phosphorus-laden sludges. Laboratory studies were
conducted on untreated primary sludge and mixed
93
-------
primary-chemical sludges to address this issue (20).
It was found that during storage of primary sludge at
different degrees of stability (digestion times),
phosphorus was always released from the solid phase
to the liquid phase. Orthophosphorus concentration
increased from an initial 5-10 mg/l P to 25-50 mg/l
P after 10-12 days of anoxic storage. However, for
mixed primary and alum secondary sludge, no release
of orthophosphate took place during anoxic storage,
regardless of the degree of stability and digester
temperature (20).
5.3.4.2 Anaerobic Digestion
Review of the literature on anaerobic digestion of
alum sludges indicates mixed results regarding the
impact of aluminum addition on the digestion process.
In one laboratory study, it was found that alum-
precipitated phosphorus concentrated in the sludge of
the digester and was not released during anaerobic
digestion. Phosphorus concentrations in the
supernatant from the digester receiving alum sludge
were less than in the control. No toxic effects of the
alum sludge were noted, as evidenced by volatile acid
and gas production characteristics (21). Similar
results were found in another laboratory study (22).
However, in a closely controlled laboratory study, it
was found that chemical coagulation of organic
materials with alum caused a significant decrease in
the anaerobic digestibility of the resulting sludge as
measured by gas production per mass of organic
solids in the feed. This was attributed to "association
of substrate with coagulant floe, rendering a portion of
the organics less accessible and/or reactive to
microorganisms or their extracellular enzymes" (23).
Other laboratory-scale studies showed digester
performance, as measured by gas production,
methane production, volatile solids reduction, and
COD reduction, decreased with increasing alum
dosages. At an alum dosage of 200 mg/l, digester
performance was 92 percent of the control value,
decreasing to 82 percent at an alum dosage of 400
mg/l. Reduced alkalinity in the digesters receiving
alum sludge was noted (24). At the plant from which
the sludges were derived for the above laboratory
tests, gas production decreased to about 50 percent
of normal levels upon addition of 250 mg/l of alum to
the raw wastewater (24).
In the EPA survey of plants removing phosphorus, 56
percent of plants employed anaerobic digestion. Of
these, 22 percent reported that chemical addition was
having a significant impact on their digestion process.
Negative impacts reported included (1):
1. Increased energy requirements for sludge mixing,
pumping, and heating,
2. Difficultly in achieving adequate digester mixing
and heating,
3. Increased labor requirements for sludge pumping,
4. Poor solids-liquid separation, and
5. Reduction in digester efficiency.
At Richardson, Texas, solids stratification and
digester upset occurred when alum was added to the
raw wastewater. When the point of alum addition was
moved to the secondary process, no digestion
problems were observed. At Ashland, Wisconsin,
alum was added to the secondary treatment stage
(step aeration activated sludge), and sludge was
stabilized with two stage digestion. After
commencement of alum addition, solids-liquid
separation no longer occurred. Polymer addition did
not solve the problem. Similar problems in solids-
liquid separation occurred in the secondary digesters
at Three Rivers, Michigan and Gladstone, Michigan.
Several plants reported significant increases in sludge
volume which reduced detention times and exceeded
heat exchanger capacity (1).
Special considerations should be given to the design
of anaerobic digesters receiving chemical sludges.
The most important is the capability to handle
increased sludge volumes resulting from chemical
addition. Procedures for estimating sludge mass have
been provided in this chapter. Sludge volume
estimates require knowledge of sludge concentrations
resulting from clarification and thickening, which can
be best determined from pilot or full-scale trials. In
existing plants, expansion of thickener capacity and/or
addition of polymer to thickeners can reduce the
volume of sludge fed to the digesters. In addition to
sludge volume increases, consideration should be
given to potential performance inhibition from alum
addition, which may justify longer detention times.
Mixing is also very important in order to achieve
optimal volatile solids destruction and gas production.
Mixing maintains contact between the active biomass
and the substrate; creates physical, chemical, and
biological uniformity throughout the digester;
disperses metabolic end products and toxic materials;
and prevents formation of scum layer and deposition
of suspended matter (4). Even in digesters receiving
conventional non-chemical sludges, mixing is often
inadequate (1). Unfortunately, little design guidance is
available to define "adequate" mixing. In general,
strong mixing can be achieved if the power dissipated
in the tank is 5-8 W/m3 (0.2-0.3 hp/1,000 ft3) of
digester volume. Velocity gradients of 50-85
m/sec/m have been recommended (4). Use of values
in the high end of these ranges is prudent when
digesting chemical sludges.
Poor liquid-solids separation in secondary digesters
may occur when handling alum sludges. This results
in a high BOD and SS load in the return supernatant
and potentially a thin sludge for dewatering or
94
-------
disposal. Polymer addition to the feed sludge or in
conjunction with the primary coagulant may be of
some value. Poor liquid-solids separation may be
the result of digester "crowding." Thickening of raw
sludges can reduce the volume for digestion. More
rapid removal of sludge from the digester can also
reduce crowding. However, this requires sufficient
dewatering and disposal capability, and consideration
must be given to reduced detention time and the
potential impacts on sludge stability. In some cases,
changing the primary coagulant or modifying the point
of chemical addition can improve liquid-solids
separation (1).
5.3.5 Conditioning
5.3.5.1 Chemical Conditioning
Chemical conditioning is frequently used to improve
the dewaterability of organic sludges. Addition of alum
to primary or secondary processes will likely change
the dewatering characteristics of the sludges and the
resulting conditioning requirements. Unfortunately, it
is virtually impossible to provide specific design
criteria for chemical conditioning of alum sludges due
to the numerous factors which affect conditioning
requirements and dewatering characteristics.
Polymers have become increasingly popular as
conditioning agents for dewatering combinations of
chemical and organic sludges, although ferric chloride
and ferric chloride plus lime are also used for this
purpose.
The only generalization that can be made regarding
conditioning requirements of chemical-organic
sludges is that conditioner dosages can be expected
to be higher than for conventional non-chemical
sludges. Farrell predicted that costs of conditioning
may be as much as 40 percent higher when alum is
used as a coagulant compared to the baseline cost of
conditioning anaerobically digested primary plus
waste activated sludge without chemical addition to
the wastewater (25). In some cases, conditioning
costs per unit weight of dry solids have more than
tripled (2). In general, sludges should be conditioned
and dewatered when "fresh," as storage has been
shown to greatly increase conditioning requirements
and adversely affect dewatering characteristics (26).
For conventional non-chemical sludges, typical
conditioner dosages of ferric chloride and lime are
20-63 kg/Mg (40-125 Ib/ton) and 75-277 kg/Mg
(150-550 Ib/ton) of dry solids, respectively. For
polymer, typical dosages are 0.3-5 kg/Mg (0.5-10
Ib/ton) of dry solids (4). Novak and O'Brien studied
polymer conditioning of chemical sludges in the
laboratory. It was found that for the near neutral pH
range, anionic polymers with a range of 15-30
percent hydrolysis require the least dose and
significantly reduce specific resistance. For neutral
and slightly acidic sludges, cationic polymers function
effectively, although dosage requirements are greater
than for anionic and nonionic polymers. For
conditioning of sludges prior to vacuum or pressure
filtration, results indicated that the benefit of polymer
conditioning is improvement in the filtering rate and
not by increasing the cake solids concentration (27).
Proper dosages of chemical conditioners can only be
determined through trial and error procedures using
the actual sludge to be dewatered. Where possible,
results of laboratory tests should be confirmed at
pilot- or full-scale.
5.3.5.2 Thermal Conditioning
Thermal conditioning is a process by which sludge is
subjected to temperatures of 177-240°C (350-
465°F) in a reaction vessel at pressures of 1,720-
2,760 kn/m2 (250-400 psig) for a period of 15-40
minutes. One variation of the process involves
injection of a small amount of air into the system.
Thermal conditioning changes the cellular structure of
the sludge, allowing the resulting material to be
readily thickened and dewatered.
There is evidence in the literature to indicate that
sludges resulting from chemical precipitation of
phosphorus have an adverse impact on thermal
sludge conditioning processes (1,28). At Midland,
Michigan, ferric chloride addition to primary treatment
was initiated to reduce effluent phosphorus
concentrations. When no ferric chloride was added,
the thermally conditioned sludge thickened to 13
percent solids. Upon addition of 19 mg/l FeCl3 to the
raw wastewater, thickening to only 9 percent solids
was possible. However, by increasing the
temperature of the thermal conditioning unit from
185°C (365°F) to 202°C (395°F), the sludge could
be thickened to 22.5 percent solids. Full-scale trials
with alum addition were also conducted. When
operated at 202°C (395°F), the thermal conditioning
unit produced a sludge which could only be thickened
to 16 percent solids. Percent solids of the vacuum
filter cake was 41 percent for the thermally
conditioned alum sludge vs. 56 percent for the
thermally conditioned ferric chloride sludge.
5.3.5.3 Freezing
Subjecting alum sludge to freezing conditions is
another conditioning technique that may be applicable
in cold climates. In experiments at Ely, Minnesota, it
was found that alum sludges with solids
concentrations of 0.25-0.32, after subjecting to a
natural freeze thaw cycle, dewatered to 16.8-18
percent solids. Freezing rates of alum sludges were
similar to that of water (29). Reed et al have
proposed design criteria and procedures for sludge
dewatering systems utilizing natural freezing in cold
climate (30).
5.3.6 Dewatering
The following sections address dewatering systems
for sludges resulting from chemical phosphorus
95
-------
removal processes. Much of the research on
dewatering of chemical sludges was conducted during
the 1970s. Since that time, there have been
significant developments in sludge dewatering
equipment, including displacement of traditional
vacuum filters with belt filter presses, and
improvements in centrifuge design. Unfortunately,
little data have been published recently on the
performance of these devices when processing
chemical-laden sludges. For this reason, it is
recommended that the engineer approach such
designs cautiously. Manufacturers should be
contacted for additional data and a list of facilities that
are dewatering chemical sludges prior to process or
equipment selection.
5.3.6.1 Drying Beds
Of the 174 plants practicing phosphorus removal and
which respond to the EPA survey, 58 (33 percent) of
the plants dewatered their sludges on sand drying
beds. Twenty of these plants used alum for
phosphorus precipitation, three of which reported
problems with dewatering. Problems ranged from
longer drying times to complete failure to dewater,
requiring use of alternative dewatering methods (1).
Novak and Montgomery investigated sand bed
dewatering on a variety of chemical sludges from
water treatment plants (31).
Little design information is available for sand bed
dewatering of sludges from plants using alum for
phosphorus removal. For that matter, there is little
published design criteria for any sludge other than
anaerobically digested. For anaerobically digested,
unconditioned sludge, recommended solids loadings
are 134 kg/m2/yr (27.3 Ib/ft2/yr) for "primary", 110
kg/m2/yr (22.4 Ib/ft2/yr) for "primary plus chemicals"
and "primary plus low-rate trickling filter", and 73
kg/m2/yr (14.9 Ib/ft2/yr) for "primary plus waste
activated sludge" (4,32). Other design criteria are in
terms of bed surface area per capita, which is of
limited value and has no rational design basis.
Addition of polymer to sludge prior to application to
drying beds has been shown to improve dewatering
and shorten drying time (1). Conclusions from the
EPA survey suggested the following modifications to
improve drying bed performance (1):
1. Improving performance of upstream facilities (e.g.
thickeners, digesters).
2. Adding chemicals to improve dewatering
characteristics.
3. Optimizing sludge loading rates and bed turnover
rates.
4. Changing the drying bed filter material.
5. Covering open beds where climatic conditions
adversely affect performance.
5.3.6.2 Vacuum Filtration
Many existing wastewater treatment plants employ
vacuum filtration for sludge dewatering. In the EPA
survey, 21 percent of the plants removing phosphorus
used vacuum filtration. Most plants reported
significant increases in sludge generation rates upon
alum addition for phosphorus removal, which
necessitated longer operating times for the vacuum
filters (1).
Reports in the literature vary widely regarding the
impact of alum addition to the wastewater on vacuum
filter dewatering. Laboratory Buchner funnel tests
showed that alum addition to secondary treatment
produced a waste activated sludge that was easier to
dewater, measured as specific resistance. Improved
dewaterability was also found for combined raw
primary and alum waste activated sludge. These
sludges also required less ferric chloride to condition
(22). In another laboratory study, specific resistance
and filter leaf tests were conducted on aerobically
digested waste activated sludge with and without
alum addition prior to secondary clarification. Results
showed that the alum digested sludge filtered slightly
better than the control sludge. Filter leaf tests
indicated that the conditioned alum waste activated
sludge could be dewatered from 1.6 percent TSS to a
cake concentration of 16 percent TSS at a filtration
rate of 15 kg/m2/hr (3.0 Ib/ft2/hr) (17).
In laboratory studies investigating dewatering of
primary sludges, alum-primary sludge derived by
addition of 200 mg/l alum to raw wastewater exhibited
lower resistivity and capillary suction time (improved
dewatering) compared to the control; however, filter
leaf cake solids were 24.0 percent for the alum-
primary sludge vs. 32.5 percent for the control.
Required filter area for the alum sludge was projected
to be 2.73 times that required for primary sludge
without chemicals (33). Required filter area was
calculated based on estimated sludge production rate
as well as dewatering characteristics (Buchner funnel
data) (32).
Pilot studies were conducted by Envirotech
Corporation on thickening and dewatering of chemical
primary sludges. Data from vacuum filtration of
alum-primary and primary sludge indicated that alum
addition to the raw wastewater adversely affected
vacuum filtration of the resulting sludge. The required
lime conditioning dose increased and the filtration rate
decreased as the alum dosage increased. Cake solids
were approximately the same for the alum dosages
investigated, although were lower than primary sludge
with no chemicals. Vacuum filtration performance for
the alum-primary sludge is summarized in Table 5-
5 (10).
96
-------
Table 5-5. Summary of Vacuum Filter Performance for
Alum-Primary Sludge.
Percent P Removal with Alum
Criterion
Lime conditioning dose, percent
Filter yield, kg/m2/hri
Cake solids content, percent2
Solids capture, percent
80
25
13.5
25.5
97-99
90
25
8.6
25.6
97-99
1 Excluding chemicals.
2 Including chemicals.
Table 5-6. Design Vacuum Filtration Rates for
Conventional Sludges (35).
Type of Sludge
Raw Primary
Raw (primary + WAS)
Raw (primary + TF)
WAS
Anaerobically dig. (primary + WAS)
Anaerobically dig. (primary + TF)
Filtration
Rate
kg/m2/hr
24-48
10-24
15-29
5-10
15-24
20-29
Cake
Solids
percent
25-30
16-24
20-26
12-18
20-24
20-28
At the 91,000-m3/d (24-mgd) West Windsor
primary plant in Windsor, Ontario, effluent phosphorus
levels of 1 mg/l were achieved with 90 mg/l alum and
0.4 mg/l polymer. Sludge production (dry solids)
increased from 115 to 259 kg/1,000 m.3 (960 to 2,160
lb/106 gal), while the solids content of the primary
sludge dropped from 11.5 to 7.6 percent. Vacuum
filter yield dropped from 55 to 28 kg/m2/hr (11.3 to
5.8 Ib/ft2/hr), and cake solids were reduced from 31.1
to 19.2 percent. Sludge conditioning with ferric
chloride and lime became more difficult, increasing
the costs for conditioning chemicals threefold (34).
At Windsor's 15,000-m3/d (4.0-mgd) Little River
conventional activated sludge plant, 150 mg/l alum
was added to the raw wastewater. Primary sludge,
containing a small amount of waste activated sludge,
showed a reduction in solids concentration from 6.2
to 5.7 percent, while sludge production rose from 189
to 293 kg/1,000 m.3 (1,580 to 2,440 lb/106 gal). Filter
yield dropped from 25 to 22 kg/m.2/hr (5.2 to 4.6
Ib/ft2/hr). Filter cake solids were 15.9 percent when
dewatering alum sludge (34).
At Lakewood, Ohio prior to alum addition to
secondary treatment, average total solids in
anaerobically digested sludge (vacuum filter feed)
was 4.45 percent. Sludge dewatered to 23.8 percent
solids. After alum addition, digested sludge feed
solids concentration increased to 6.5 percent, but the
dewatered cake solids dropped to 21.4 percent.
Although the dry mass of sludge generated increased
from 590 Mg/yr (650 tons/yr) in 1974 to 1,650 Mg/yr
(1,820 tons/yr) in 1976, vacuum filter operation and
maintenance costs per unit mass of dry solids
increased only 14 percent (4).
Design filtration rates and cake solids for various
types of conventional sludges are shown in Table 5-
6 (35). Relative vacuum filtration characteristics for
various components of wastewater sludges are shown
in Table 5-7. Composite characteristics are a
function of the proportionate amount of each
component in the total mixture. This is based on the
assumption that the accumulative effects on vacuum
Table 5-7. Component Vacuum Filtration Characteristics
(31).
Sludge Component
Filtration
Rate*
Cake
Solids
Primary
WAS
TF
AL(OH)3»AIPO4
Fe(OH)3»FeP04
kg/m2/hr
29-48
1-2
7-10
5-7
7-10
percent
28-30
12-18
12-15
14-16
12-15
" Values shown are indicative only of the relative effects of various
components on the dewatering characteristics of a sludge
mixture.
filtration are a function of the sum of the individual
effects (35).
Table 5-8 lists estimated design factors for
conventional and alum sludges, using vacuum
filtration of anaerobically digested primary plus waste
activated sludge as a baseline (2). The factors in the
table can then be used to predict other yields. If no
actual data are available on which to predict yields for
other sludges, a baseline value of 20 kg/m2/hr (4
Ib/ft2/hr) can be used (2).
5.3.6.3 Centrifugation
Several studies have investigated centrifugal
dewatering of sludges from plants using aluminum
salts for phosphorus control (10,17,32,36-38). In
general, such sludges are amenable to centrifugation.
However, decreased cake solids and deterioration in
centrate quality may be expected. Polymer is
effective in increasing solids capture, although often
with a corresponding decrease in cake solids.
Studies by Envirotech using a pilot-scale solid bowl
centrifuge showed that as the fraction of alum solids
in the alum-primary sludge increased, both cake
solids and maximum percent solids recovery
decreased for any given hydraulic loading rate. As
percent solids capture improved with polymer
97
-------
Table 5-8. Design factors for Vacuum Filtration of
Conventional Plus Aluminum Sludges (2).
Sludge Type
Al to primary
Digested primary + WAS
Raw primary + WAS
Raw primary + TF
Digested primary + TF
Digested primary
Raw primary
Al to aeration
Raw primary + WAS
Digested primary + WAS
Relative
Yield
kg/m2/hr
1.2
0.85
1.35
1.0
1.05
1.5
1.2
0.85
Rel. Cost of
Conditonmg*
percent
1.3
1.4
1.2
1.3
1.0
0.9
1.3
1.4
" Yield and cost factors related to yield and costs obtained with
digested primary + WAS when no chemicals are added to
wastewater.
addition, cake solids decreased. For solids captures
greater than 80 percent, cake solids were typically 15
to 18 percent (10).
Baillod et a/, investigated pilot-scale basket
centrifugation of aerobically digested sludge from a
contact stabilization plant with alum addition to the
raw wastewater. Using the manufacturer's scale-up
procedure, results indicated that a full size 1.2-m
(4-ft) diameter basket centrifuge could dewater the
digested alum-biological sludge at 1.6 percent TSS
to a 16 percent IS cake at a rate of 159 kg dry
solids/hr (350 Ib/hr) with no chemical addition. Solids
capture was 96 percent (17).
Canadian researchers investigated use of solid bowl
and basket centrifuges for dewatering of several
different sludges from pilot- and full-scale facilities.
Increasing polymer dosages at the solid bowl
centrifuge resulted in increased solids recovery for all
sludges tested with the basket centrifuge. Recoveries
in excess of 90 percent were possible without
polymer addition for all waste activated sludges
investigated. However, for the anaerobically digested
alum sludge, polymer addition was necessary to
obtain high solids recoveries. For alum waste
activated sludges, an optimum polymer dosage of 1
kg/Mg (2 Ib/ton) was suggested, resulting in cake
solids concentrations of 10-11 percent. Waste
activated sludge with alum dewatered slightly better
than the control sludge. It was also found that particle
size impacted centrifuge performance, with
decreasing particle size resulting in decreased solids
recovery and cake solids (33,36,37).
Mininni et a/, investigated dewatering of aerobically
digested waste activated sludge from a plant with no
primary clarification. When alum was added to the
raw wastewater, centrifuge cake solids dropped from
15.8 to 11.3 percent. Polymer was used as a
conditioning agent in both cases, at dosages of 0.3-
0.5 percent dry weight. Conditioner dosage increased
during alum addition. The most significant machine
variables affecting cake solids content were beach
residence time and liquid ring height (38). It was
estimated that costs for sludge conditioning,
dewatering, and disposal (including amortization of
capital) would increase by 63-74 percent, depending
on plant size, when alum was added to the raw
wastewater for phosphorus removal (39).
There are many variables which affect performance of
centrifuges. Some of the more important process
variables include source and type of sludge, feed
solids content, percentage of chemical solids, feed
rate, and conditioner dosage. Machine variables for a
solid bowl centrifuge include bowl design, bowl
speed, pool volume, conveyor speed, and conveyor
pitch (35).
Because of the variability in sludge characteristics,
pilot testing is recommended where feasible.
Procedures are available from the various
manufacturers for scale-up of pilot-test results.
5.3.6.4 Pressure Filtration
Less than 5 percent of the phosphorus removal plants
contacted in the EPA survey practiced pressure
filtration for sludge dewatering. None of these
facilities used alum as the primary coagulant for
phosphorus removal (1). Plate and frame filter
presses are often used when it is desirable to
produce a cake with a high solids content, as in
preparing sludge for incineration.
Information on pressure filtration of sludges from
plants using alum for phosphorus removal is scant.
Envirotech evaluated pressure filtration during the
investigation of chemical-primary sludge dewatering.
Results for alum-primary and primary sludge are
summarized below (10):
1. Filtration rates for alum-primary sludge were
approximately double those for primary sludge.
Filtration rates for alum-primary sludges typically
are 1.0-4.5 kg/m2/hr (0.2-0.9 Ib/ft2/hr),
depending on conditioner (lime) dosage and cake
thickness.
2. Cake solids were higher for primary sludge, 20-40
percent vs. 15-30 percent for alum-primary
sludge.
3. Conditioning requirements using lime were 24-25
percent by weight for alum-primary sludge vs.
37-65 percent for primary sludges
4. Decreasing the cake thickness significantly
increased filtration rate and cake solids for the
primary sludge. This effect was less pronounced
98
-------
as the percentage of alum solids in the sludge
increased.
Minnini et at. studied pressure filtration of waste
activated sludge (no primary clarification) using
aluminum chlorohydrate as the conditioning agent.
When alum was used for phosphorus removal, cake
solids concentrations averaged 24.5 percent. With no
chemical addition to the wastewater, cake solids
concentrations averaged 32.1 percent. Conditioner
dosages ranged from 1.1 to 1.5 percent by weight as
A^Os- Operating pressure in the filter press averaged
6 kg/cm2 (85 psi), with cycle times of approximately 3
hours (38,39).
Additional information on experiences with pressure
filtration of chemical sludges may be found in
Sections 5.4 and 5.5.
5.3.6.5 Belt Press Filtration
Belt filter presses have become quite popular in
recent years for dewatering sewage sludges, since
they are capable of producing dryer cakes than
vacuum filters or centrifuges, and are less costly than
plate and frame filter presses. Unfortunately, there is
little information on their use at plants removing
phosphorus by chemical addition.
The EPA survey makes reference to the use of a belt
filter press for dewatering alum waste activated
sludge at Westfield, New York (1). Since the plant
was designed for phosphorus removal using alum, it
is not possible to compare belt filter press
performance without alum addition. The waste
activated sludge is reported to be difficult to dewater,
as there is no primary treatment. Cake produced by
the belt filter press has an average total solids
content of 11.5 percent. Polymer conditioning with
polymer was found to be more successful than
conditioning with ferric chloride and lime (1).
Data on performance of belt filter presses for
dewatering of conventional sludges may be found in
the EPA Process Design Manual on Sludge
Treatment and Disposal (4).
5.3.7 Incineration
Thirteen percent of phosphorus removal plants which
responded to the EPA survey incinerated their
chemical sludges. Of these 22 plants, 6 reported
significant impacts on incineration as a result of
chemical addition for phosphorus removal. All plants
incinerated a combination of primary and secondary
sludges. Three of the plants that reported problems
used alum for phosphorus precipitation. These were
Warren, Michigan; Coloma, Michigan; and
Richardson, Texas (1).
Critical variables for incineration of sewage sludge
include:
1. Moisture content of the sludge,
2. Calorific value of the sludge, and
3. Relative proportion of volatile and inert material.
It is evident that the addition of alum for phosphorus
removal can significantly impact these variables due
to deterioration in sludge dewaterability (higher cake
moisture content), and a higher concentration of inert
chemical solids, which reduces the calorific value on
a unit mass basis, thus increasing auxiliary fuel
requirements. Table 5-9 lists the potential problems
associated with chemical sludge incineration, and
recommended solutions (40).
Table 5-9. Potential Problems with Chemical Sludge
Incineration (40).
Problem Solution
Greater sludge volume
Lower caloric value
Increased cake moisture
content
Formation of clinkers
More inert solids
Increase incinerator capacity or run
time.
Increase supplementary fuel
requirements.
Improve dewatering by:
1) modifying dewatering operation
or sludge conditioning;
2) changing primary precipitating
chemical or point of addition;
3) changing dewatering equipment
or
Increase supplemental fuel
requirements.
Decrease incineration temperature
to below flash point; decrease
residence time.
Increase ash disposal capacity.
Specific information regarding the impacts of alum
addition on incineration processes is not available,
although additional information on the impacts of
other chemicals such as ferric chloride and lime may
be found in Sections 5.4 and 5.5.
5.3.8 Disposal
Implementation of phosphorus removal by addition of
alum to the wastewater has several implications
regarding sludge disposal. The most significant
impact is on the volumes of sludge for disposal. As
discussed earlier, addition of alum for phosphorus
removal will likely cause a significant increase in both
the mass and volume of sludge to be transported and
disposed. This will result in increased disposal costs.
Another concern is the increased metal content of
sludges due to addition of precipitant. Addition of
alum will increase the metal content of the sludge.
Considerable research has been conducted in
Canada regarding the potential impacts on crop yield,
99
-------
organic matter degradation, nitrogen availability, and
other soil parameters (41-44).
From review of the available literature on land
application of alum sludges, it appears that addition of
alum for phosphorus removal does not adversely
affect the agricultural value of the resulting sludge
compared to non-chemical sludges. Although some
reduction of crop yield was noted with alum sludge in
the Canadian studies, this was believed to be due to
the high content of petroleum hydrocarbons, which
was 14-30 times higher than the other sludges
investigated (44). Kirkham and Dotson found that the
presence of aluminum and iron phosphate
precipitates did not affect the growth of barley in loam
soil irrigated with wet primary sludges (45).
Increased metal content in alum sludges may impact
sludge loading rates on agricultural land. Heavy metal
analyses must be conducted on a case-by-case
basis for the sludge to be applied. Application rates,
whether governed by nitrogen or heavy metal
loadings, will be determined by state and federal
regulations.
Alum addition is unlikely to affect the suitability of a
sludge for disposal at a sanitary landfill. However,
many landfills require that organic materials meet a
maximum moisture content criterion. As alum addition
may increase moisture content in cakes resulting
from dewatering processes, this may limit the
suitability for landfill disposal.
5.4 Sludge Derived from Addition of Iron
Salts
5.4.1 Sludge Characteristics
Iron salts may be employed for phosphorus removal
by addition to primary, secondary, or tertiary
treatment processes. Of those plants that used iron
and responded to the EPA survey, 32 percent added
iron to primary treatment, and 57 percent added iron
to secondary treatment. A small number of plants (3
percent) added iron to tertiary processes, while
several plants used iron in combination with other
chemicals such as lime. Four percent of the plants
added iron to both primary and secondary treatment
(1). Iron salts are employed more frequently on a
percentage basis than aluminum salts in primary
treatment.
Knowledge of sludge characteristics, as defined by
conventional characterization parameters, is of little
value in predicting the amenability of iron sludges to
thickening and dewatering operations, since so many
variables interact to affect performance. As discussed
in Section 5.3.1 design of sludge handling systems
must be based on laboratory tests at a minimum, and
preferably on pilot- or full-scale tests.
5.4.2 Sludge Generation Rates
As with aluminum addition, increased solids
production during iron addition to wastewater results
from (2):
1. Formation and removal of chemical solids such as
metal phosphates and metal hydroxides,
2. Improved removals of organic solids during
clarification, and
3. Removal of dissolved solids.
Although procedures are available to estimate the
quantities of additional solids resulting from chemical
addition, these estimates may not be accurate for a
particular wastewater and treatment plant. For
upgrading of existing plants, full-scale trials under
controlled conditions will provide the best data
regarding sludge production. For new facilities, pilot-
plant tests are preferred for accurate prediction of
sludge production.
The procedure for estimating sludge generation is
very similar to that described in Section 5.3.2 for
alum sludge, and consists of determining production
of chemical solids, generation of additional solids
removed during clarification, and removal of dissolved
solids.
a. Chemical Sludge
This calculation assumes that the iron reacts with
phosphorus compounds first, and that excess iron
forms iron hydroxide.
Given:
Pjn = 8 mg/l
Pout = 1 mo/1
Fe: P dose = 2.2:1
Atomic weight of Fe = 56
Atomic weight of P = 31
Atomic weight of FePCU = 151
Atomic weight of Fe(OH)3 = 107
Fe dose = 2.2 x 8 mg/l x (56/31)
= 32 mg/l as Fe
Stoichiometry:
Fe + PO4 = FePO4
Fe + 30H = Fe(OH)3
(8 - 1 mg/l P) T 31 = 0.23 mmole added/I FePC>4
produced
32 mg/l Fe * 56 = 0.57 mmole/l Fe added
0.57 - 0.23 = 0.34 mmole/l Fe in excess to
Fe(OH)3
100
-------
FePO4 sludge: 0.23 mmole/l x 151 = 34.7 mg/l
FePO4
Fe(OH)3 sludge: 0.34 mmole/l x 107 = 36.4 mg/l
Fe(OH)3
Total chemical sludge: 34.7 mg/l FeP04
36.4 mg/l Fe(OH)3
71.1 mg/l Fe sludge
Chemical sludge produced = 71.1 mg sludge/liter of
wastewater treated (592 lb/106 gallons). Adjust
estimate by factor of 1.35 to account for additional
chemical solids not predicted by equation (2):
Design estimate of chemical sludge production:
1.35 x 71.1 mg/l = 96.0 mg sludge/liter of
wastewater (800 lb/106 gallons)
b. Sludge from improved removal of suspended
solids (primary clarifier)
Given:
SSjn = 200 mg/l
SS removal efficiency (no chemicals) = 50 percent
SS removal efficiency (with iron) = 75 percent
Additional Sludge Generated:
(0.75 - 0.50) x 200 mg/l =
50 mg sludge/liter of
wastewater treated (417
lb/106 gallons)
c. Sludge from removal of dissolved solids
Additional sludge mass (assuming 30% removal of
soluble organics using iron salts)
= STOCjr, (mg/l) x 0.30 x 2.5 x 1.18
= SCODjp (mg/l) x 0.30 x 1.1 x 1.18
= SBODjn (mg/l) x 0.30 x 1.6 x 1.18
The latter two equations are only applicable to
influents prior to biological oxidation processes.
Given:
STOCjn = 50 mg/l
Additional sludge generated:
50 mg/l x 0.30 x 2.5 x 1.18 = 44.2 mg sludge/liter of
wastewater (369 lb/106
gallons)
If iron salts are added to primary treatment, improved
BOD removals will result in reduced loadings to
secondary biological processes, and thus reduced
secondary sludge production. This must be
accounted for in calculating estimates of total plant
sludge production. Where iron salts are added to
secondary processes, additional sludge resulting from
improved removal of SS in the secondary clarifier will
probably be small for well operated activated sludge
plants. However, for plants producing effluents with
SS concentrations greater than 25-30 mg/l, an
improved effluent quality from iron salt addition may
result in significant quantities of additional sludge.
A design example is shown in Section 5.3.2 for
calculating sludge production with and without alum
addition. The same approach can be used for iron
salt addition. The major difference is in the
stoichiometry used to predict chemical sludge
production.
5.4.3 Thickening
5.4.3.1 Gravity Thickening
As with alum sludges, reported impacts of primary or
secondary iron addition on sludge thickening
characteristics have been variable. In the survey of
phosphorus removal plants that was conducted for
EPA, respondents indicated both positive and
negative impacts on thickener performance.
Unfortunately, it was impossible to determine from the
information provided in this report whether negative
impacts were due directly to the addition of iron or to
the additional sludge volumes which may have
overloaded the thickener. Because of the many
factors which affect thickening characteristics of
sludges, thickening tests should be conducted on the
actual sludges if available. For retrofit applications,
full-scale trials should be conducted to generated
representative samples of sludge upon which proper
thickening tests can be run. Design criteria can then
be established with confidence.
For design of gravity thickeners for new facilities, the
designer does not have the benefit of having existing
sludges on which to conduct thickening tests, and
must rely on published design guidelines. Table 5-
10 provides design criteria for thickeners receiving
various types and combinations of iron and non-
chemical sludges.
In addition to mass loading criteria shown in Table
5-10, hydraulic loading must also be considered in
the design of gravity thickeners. For primary sludges,
typical maximum overflow rates range from 1,000-
1,200 l/m2/hr (25-30 gal/ft2/hr). However, for waste
activated sludge or combinations of primary and
waste activated sludges, hydraulic loadings rates
should be considerably lower, 160-320 l/m2/hr (4-8
gal/ft2/hr) (4).
In pilot-studies of ferric-primary sludges, it was
found that underflow solids concentration decreased
as the amount of chemical solids in the feed sludge
increased (10). This is shown graphically in Figure
5-3. Ferric-primary sludge was found to exhibit
superior thickening characteristics to alum-primary
101
-------
Table 5-10. Typical Gravity Thickener Design Criteria.
Sludge Type
Primary
Primary w/Fe*
Primary + WAS
Primary + TF
(Primary w/Fe) +
WAS
(Primary w/Fe) +
TF
Primary + (WAS
w/Fe)
WAS
WAS w/Fe
Tertiary w/Fe
Influent
Solids
Cone.
percent
2-7
1.8-5.2
0.5-1.5
2-6
1.8
0.4-0.6
1.5
0.5-1.5
0.5-1.5
0.5-1.5
Expected
Underflow
Cone.
percent
5-10
2.2-6.4
4-6
5-9
3.6
6.5-8.5
3
2-3
1.5-2.0
3-4
Mass
Loading
kg/m2/hr
3.9-5.9
0.3-1.3
1.0-2.9
2.5-4.2
1.3
2.9-4.2
1.3
0.5-1.5
0.6-0.8
0.4-2.1
Ref.
9
10
8
8
8
8
8
9
11
8
Data reflects use of anionic polymer to assist in thickening.
Figure 5-3 Range of thickener operating periods for
ferric-primary sludge (10).
Max. Thickener
Underflow Solids, g/l
90 r-
10 20 30 40 50
Chemical Solids in Sludge, weight percent
60
sludge (see Figure 5-4). Table 5-11 shows a
summary of the results at a loading rate of
approximately 0.83 kg/m.2/hr (0.17 Ib/ft2/hr).
Canadian studies on mineral addition to an extended
aeration plant found that addition of ferric chloride
deteriorated the settleability of the mixed liquor
Figure 5-4 Range of thickener operating periods for
alum-primary sludge (10).
Max. Thickener
Underflow Solids, g/l
90 i-
80
10 20 30 40 50
Chemical Solids in Sludge, weight percent
60
Table 5-11
Effect of Phosphorus Removal on Gravity
Thickening Properties of Alum-Primary and
Ferric-Primary Sludge (10).
Alum-Primary Sludge
T-P
Removal
percent
80
90
95
Chemical
Sludge
Weight
percent
18
23
32
Underflow
TS
percent
4.7
4.1
3.3
Ferric-Primary Sludge
Chemical
Sludge
Weight
percent
22
28
3.8
Underflow
TS
percent
5.5
5.4
5.3
Basis: Raw wastewater with 100 mg/l TSS and TP of 5 mg/l.
Solids loading rate: 0.83 kg/m2/hr (0.17 Ib/ft2/hr).
suspended solids. Secondary clarifier underflow
sludge was 1.1-1.3 percent solids with iron addition
to the aerator vs. 1.7-1.8 percent with alum addition
and 2.0 percent with no chemical addition (32).
In general, limited data are available on thickening
characteristics of combined primary and secondary
sludges with iron addition to the secondary process.
Lower thickened sludge concentrations may be
expected, and solids loading rates are generally lower
than for combined sludges without chemicals.
Polymer addition to the influent sludge may improve
thickener performance.
102
-------
5.4.3.2 Flotation Thickening
Flotation thickening is most often applied to waste
activated sludges. Seldom is it used for thickening
primary or combined primary-secondary sludges. In
the EPA survey of plants removing phosphorus, 6
percent of the facilities employed flotation thickening.
Of the four plants which flotation-thickened iron-
secondary sludge, three reported no impact of iron
addition on thickener performance. The fourth plant,
which aerobically digested the iron-secondary
sludge prior to thickening, had always treated iron
sludge and could not make a comparison with non-
chemical sludge (1). One plant reported use of
flotation thickening for a combination of iron-primary
and aerobically digested secondary sludges.
Table 5-12 summarizes the results reported from
the six plants responding to the EPA survey. For
iron-secondary sludges, the maximum thickened
solids concentration achieved was approximately 5
percent (1).
Table 5-12. Performance of Flotation Thickeners for
Treating Iron Sludges (1)
Type of Sludge
Iron-secondary
Iron-secondary
Iron-secondary
Iron-secondary
(aer. digested)
Iron-primary +
aer. digested
secondary
Feed
Solids
Cone.
percent
0.9
1.0-1.5
1.0
1.5
Thickened
Solids
Cone.
percent
5.1
3.5-4.0
2.5
2.5
5-6
Comments
Polymer added.
Polymer used
when loadings
high.
Polymer did not
improve.
Plant has no
primary
treatment.
Both cationic and
anionic polymer
added.
For flotation thickening of non-chemical activated
sludge, typical design loadings are 2.4-3.S kg/m2/hr
(0.5-0.8 Ib/ft2/hr without polymer and 4.9-7.3
kg//m.2/hr (1.0-1.5 lb/«2/hr) with polymer (8).
Hydraulic loading rates are typically 29-117 m3/m2/s
(0.5-2 gpm/ft2) (8). For iron-activated sludge, solids
loadings of 10-24 kg/m2/hr (2-5 Ib/ft2/hr) with
polymer have been recommended (11).
Recommended hydraulic loading rates are 58-88
m3/m2/s (1.0-1.5 gpm/ft2) (11).
5.4.4 Stabilization
5.4.4.1 Aerobic Digestion
Twenty-four percent of the phosphorus-removal
plants contacted in the EPA survey reported use of
aerobic digestion for chemical sludge stabilization (1).
Few problems were reported, although several plants
indicated problems due to increased volumes of
sludge that exceeded the design capacity of the
digesters.
A Canadian laboratory study investigated the impact
of chemical addition to the aeration basin on the
aerobic digestion of the resulting waste activated
sludge. The major conclusion was that the
performance of the aerobic digestion process did not
differ in any practical degree when iron or aluminum
precipitates were present in the sludge (18). In
addition, the release of soluble organic carbon and
nutrients into the liquid phase was not enhanced by
the presence of iron and aluminum precipitates. Batch
digester operation resulted in greater destruction of
volatile solids and a lower oxygen uptake rate than
semi-continuous operation. However, semi-
continuous operation at a loading rate of 1 kg volatile
solids/m3/d (0.06 Ib/ft3/d) provided digested sludges
with better supernatant quality as well as superior
settling and dewatering characteristics (18).
Relatively little information is available regarding the
impact of iron salt addition to wastewater on the
aerobic digestion of the resulting sludges. More
information is available on alum sludges (1,17-20).
Dick suggests that "Chemical precipitation of
phosphorus would be expected to affect aerobic
stabilization processes (aerobic digestion and
composting) the same way it affects anaerobic
processes. That is, changes in the amount of organic
solids removed, their concentration in sludge, and
their availability in sludge would influence
performance (13)."
5.4.4.2. Anaerobic Digestion
Anaerobic digestion is a common sludge stabilization
technique, particularly for plants larger than 19,000
m3/d (5 mgd). In the EPA survey of plants removing
phosphorus, 5.6 percent of the plants employed
anaerobic digestion. Of these, 22 percent reported
that chemical addition was having a significant impact
on their digestion process. Negative impacts reported
included (1):
1. Increased energy requirements for sludge mixing,
pumping, and heating,
2. Difficulty in achieving adequate digester mixing and
heating,
3. Increased labor requirements for sludge pumping,
4. Poor solids - liquid separation, and
5. Reduction in digester efficiency.
Process-related problems reported with iron sludges
included poor digestion, accompanied by decreased
digester pH, reduced gas production, and/or
decreased gas production; digester upsets
characterized by loss of methane production, low pH,
103
-------
and low volatile solids destruction; and digester
foaming (1). On the other hand, positive effects have
been reported, such as improved digester sludge
settleability and supernatant quality, and increased
volatile solids destruction (1). Such conflicting
information is similar to that reported for alum sludge
(see Section 5.3.4.2).
Malhotra et al. investigated anaerobic digestion of iron
phosphate sludges on a laboratory scale. Conclusions
from this study are as follows (46).
1. For conventional activated sludge plants using
ferrous iron for phosphorus precipitation,
phosphorus removal efficiency will not be
drastically reduced by the return of phosphorus in
anaerobic digester supernatant.
2. The pH, alkalinity of volatile acids and volatile
solids destruction were similar in both the control
digesters and test digesters receiving iron sludge.
3. The ferrous iron present in the feed sludge did not
cause digester upset up to a maximum level of 5.5
percent Fe by weight dry solids.
4. The quantity and quality of the gas was not altered
significantly with iron sludge digestion.
5. With primary sludge containing ferrous-iron
precipitated phosphorus, significant uptake of total
soluble phosphorus was observed during digestion.
With thickened waste activated sludges containing
iron-precipitated phosphorus, significant
phosphorus release was observed during digestion.
This was possibly due to the conversion of ferric
phosphate to ferrous phosphate plus phosphate
ions during anaerobic digestion.
Other laboratory studies by Dentel and Gosset
concluded that chemical coagulation of organic
materials with alum or ferric chloride caused a
decrease in anaerobic digestibility of the resulting
sludge. The effect was distinct from any effects of
increased loading or differences in pH or alkalinity,
and was not attributable to toxicity, increased fixed
solids concentration, or phosphate limitation. Results
suggested that the mechanism responsible for
decreased digestibility was association of substrate
with coagulant floe, rendering a portion of the
organics less accessible and/or less reactive to
microorganisms (23).
In another study, Gossett et al. found that, at ferric
chloride doses of 150 mg/l to wastewater, anaerobic
digester performance was 90 percent of the control;
at 200 mg/l, it was 78 percent. For chemical sludges,
organic nitrogen decomposition was about 50 percent
less than for the control sludge, lowering ammonia
production which in turn reduced alkalinity. During
plant-scale studies, addition of 150 mg/l of ferric
chloride to raw wastewater resulted in a 25 percent
reduction in digester gas production (24).
The most important consideration in digestion of
chemical sludges is the large volumes of sludges
generated during chemical precipitation. Provision of
adequate digester mixing is also critical. Further
discussion of these criteria is found in Section
5.3.4.2.
5.4.5 Conditioning
Chemical conditioning requirements for iron sludges
can be expected to be higher than for non-chemical
sludges (1). However, because of the many variables
which affect conditioning requirements and
dewatering characteristics, specific design criteria
cannot be provided. Laboratory tests should be
conducted using samples of the actual sludge to be
dewatered in order to determine chemical
requirements. Where possible, such results should be
confirmed at pilot or full scale.
Thermal conditioning may also be affected by the
presence of iron or aluminum precipitates (1,28).
Because of the similarities in the results of studies on
conditioning of aluminum and iron sludges, reference
is made to Section 5.3.5 for further discussion of
conditioning of chemical sludges.
5.4.6 Dewatering
As described in Section 5.3.6, significant advances
have occurred in dewatering wastewater sludge,
including introduction of the belt filter press and
improvements in centrifuge design. Much of the
performance data and design criteria decribed below
resulted from work conducted in the 1970s. Little data
are available on dewatering chemical-laden sludges
using state-of-the-art (1987) technology. For this
reason, the engineer should proceed cautiously with
regard to selection and sizing of dewatering
equipment. Manufacturers should be contacted
regarding design criteria for dewatering chemical-
laden sludges.
5.4.6.1 Drying Beds
Sand drying beds are a popular and economical
technique for sludge dewatering at small wastewater
treatment plants. A third of the phosphorus removal
plants responding to the EPA survey used drying
beds for dewatering sludge (1). Only 15 percent of
the plants using drying beds reported problems with
dewatering chemical sludges. Many of the problems
were directly related to handling the increased volume
of sludge generated by the addition of chemicals for
phosphorus control. However, several plants reported
that the sludge was more difficult to dewater (1).
Novak and Montgomery studied the use of sand
drying beds for dewatering chemical sludges from
water treatment plants (31). Conclusions from this
work are discussed in Section 5.3.6.1.
104
-------
Rational design criteria for other than anaerobically
digested sludge are virtually non existent (4). As
discussed in Section 5.3.6.1, recommended solids
loadings for anaerobically digested sludges are 73-
134 kg/m2/yr (15-27 Ib/ft2/yr).
Conclusions from the EPA survey suggested the
following modifications to improve the performance of
sand drying beds (1):
1. Improving performance of upstream facilities (e.g.
thickeners, digesters).
2. Adding chemicals such as polymer to improve
dewatering characteristics.
3. Optimizing sludge loading rates and bed turnover
rates.
4. Changing the drying bed filter material.
5. Covering open beds where climatic conditions
affect performance.
5.4.6.2 Vacuum Filtration
Twenty-one percent of the phosphorus removal
plants responding to the EPA survey reported use of
vacuum filters for sludge dewatering. Most plants
indicated significant increases in sludge generation
rates which required longer operating times for the
vacuum filters (1).
As with alum sludges (Section 5.3.6.2), reports in the
literature vary widely regarding the impact of iron salt
addition of wastewater on vacuum filter dewatering of
the resulting sludge. In the EPA survey, some plants
reported significant increases in filter yield and cake
solids content, while others reported decreased filter
yields when dewatering iron sludges (1). Campbell
and LeClair reported deteriorated dewaterability when
ferric chloride was used to remove phosphorus, but
improved dewaterability with alum (36). However,
Mininni ef a/, found that both ferrous iron or aluminum
deteriorated dewaterability, with aluminum having a
more deleterious effect (38).
Envirotech Corporation conducted comprehensive
pilot studies on dewatering of chemical-primary
sludge (10). Vacuum filter performance, as measured
by conditioner dose and filtration rate, was adversely
affected by the presence of iron chemical solids in
the feed sludge. Conditioner dose increased and
filtration rate decreased as the proportion of iron
chemical solids in the feed sludge increased. Cake
solids content in dewatered ferric-primary sludge
was insensitive to the proportion of chemical solids in
the feed sludge; however, vacuum filter dewatering of
ferric-primary sludge produced sludge cakes of
higher solids than with non-chemical primary sludge.
Performance of vacuum filters for dewatering iron-
primary sludge was slightly better than for dewatering
alum-primary sludge. Vacuum filter performance for
iron-primary sludge is summarized in Table 5-13
(10).
Table 5-13. Summary of Vacuum Filter Performance for
Iron-Primary Sludge (10).
Percent P Removal with FeCI3
Criterion
Lime conditioning dose, percent
Filter yield, kg/m2/hr1
Cake solids content, percent2
Solids capture, percent
80
30
10.2
34.5
97-99
90
30
8.6
34.5
97-99
1 Excluding chemicals.
2 Including chemicals.
*
At Sheboygan, Wisconsin, the secondary trickling
filter plant began adding ferric chloride in 1972 to the
effluent of the trickling filters in order to accomplish
phosphorus removal. Degritted primary sludge is
blended with secondary sludge, gravity thickened,
and dewatered by vacuum filtration. Comparison of
vacuum filter data for 1970 and 1976 showed a
decrease in filter yield from 20.3 to 12.8 kg/m2/hr (4.2
to 2.6 Ib/ft2/yr).
Filter feed solids dropped from 8.6 to 7.0 percent, and
cake solids were reduced from 25.5 to 21.5 percent.
Filtrate quality improved from 658 ppm to 442 ppm
SS. While only polymer was used for conditioning 80
percent of the time in 1970, both polymer and ferric
chloride were required for conditioning after chemical
phosphorus removal was implemented (1).
The City of Midland, Michigan, operates a high rate
trickling filter plant with vacuum filtration of primary
and secondary sludge. Prior to installation of thermal
conditioning units, raw sludge was conditioned with
polymers. Vacuum filter yields were 24 kg/m2/hr (5
Ib/ft2/hr). When 19 mg/l FeCIa was added to primary
treatment for phosphorus removal yields were
reduced to 15 kg/m.2/hr (3 Ib/ft2/hr), although cake
solids increased from 25.5 to 39.3 percent TS. During
a period of time when thermal conditioning was on-
line but when chemical phosphorus removal was not
being practiced, filter yields were 39 kg/m2/hr (8
Ib/ft2/hr), and cake solids were 48 percent. Upon
ferric chloride addition to the primaries, filter yield
dropped to 20 kg/m2/hr (4 Ib/ft2/hr), and cake solids
were reduced to 44 percent. Raising the operating
temperature of the thermal conditioning units from
185°C (365 °F) to 202 °C (395 °F) resulted in filter
yields of 78 kg/m.2/hr (16 Ib/ft2/hr), and cake solids of
54 percent (1).
Design vacuum filter loading rates for various types of
conventional sludges are shown in Table 5-6 (35).
105
-------
Relative vacuum filtration characteristics for various
components of wastewater sludges are shown in
Table 5-7. Composite characteristics are a function
of the proportionate amount of each component in the
total mixture. This is based on the assumption that
the accumulative effects on vacuum filtration are a
function of the sum of the individual effects (35).
Table 5-14 lists design factors for conventional and
iron sludges using vacuum filtration of anaerobically
digested primary plus waste activated sludge as a
baseline (2). The factors in the table can then be
used to predict other yields. If no actual data are
available on which to predict yields for other sludges,
a baseline value of 20 kg/m^/hr (4 Ib/ft2/hr) can be
used (2).
Table 5-14. Design factors for Vacuum Filtration of
Conventional Plus Iron Sludges (2).
Sludge Type
Relative
Yield
Rel. Cost of
Cond toning*
kg/m2/hr percent
Fe to primary
Raw primary + WAS
Raw primary + TF
Digested primary + WAS
Digested pnmary + TF
Digested primary
Raw primary
Fe to aeration
Raw primary + WAS
Digested primary + WAS
.3
.5
3.95
.1
.2
.6
1.3
3.95
1.1
1.0
1.2
1.1
0.8
0.7
1.1
1.2
* Factors related to yield and costs obtained with digested primary
+ WAS when no chemicals are added to wastewater.
Another important design consideration is corrosion of
metal components due to the corrosive nature of
ferric chloride. Use of stainless steel components
may be justified if ferric chloride is to be employed for
phosphorus removal.
5.4.6.3 Centrifugation
Data from the literature indicate that sludges derived
from processes employing iron salts for phosphorus
removal are amenable to dewatering by
centrifugation. However, decreased cake solids and
poorer centrate quality compared to conventional
non-chemical sludges may be expected in some
cases.
Envirotech conducted extensive pilot studies on
dewatering of chemical primary sludges. For alum-
primary sludges centrifuge performance relative to
polymer requirements and cake solids concentrations
was adversely affected as the quantity of aluminum
solids in the feed sludge increased. However, for
ferric-primary sludge, cake solids concentrations
increased as the amount of iron chemical solids
present in the feed sludge increased, and were
significantly higher than for alum primary sludges.
Polymer requirements and machine capacity to
achieve a given level of solids capture were not
affected as the quantity of iron chemical solids in the
feed sludge increased. At total solids capture of 95
percent, centrifugal dewatering of ferric-primary.
sludge produced cakes of 22-25 percent solids vs.
20-21 percent solids for the control primary sludge
with no iron (10).
Campbell and LeClair reported on pilot scale
centrifugal dewatering of waste sludge generated by
an extended aeration pilot plant (36). Using a basket
centrifuge, it was found that the ferric chloride sludge
was more difficult to dewater than either the alum
sludge or the control sludge with no chemical. Cake
solids for the iron sludge were 8.4-12.0 percent TS.
With a solid bowl centrifuge, cake solids achieved
with the iron sludge were essentially the same as the
control sludge, but consistently lower than the alum
sludge. Cake solids with iron sludge were 3.4-8.4
percent TS. Polymer addition was required to obtain
solids recoveries in excess of 9.5 percent (36).
Centrifugation of anaerobically digested primary and
iron- secondary from the North Toronto plant was
also investigated. A feed sludge of 6.6 percent TS
was dewatered to 17.3-24.0 percent TS with
polymer dosages of 4.4-6.5 kg/Mg (8.8-13.0
Ib/ton). Solids recoveries were 98-99 percent (33).
Mininni et a/, investigated centrifuge dewatering of
aerobically digested waste activated sludge from a
plant with no primary clarification. When ferrous
sulfate was added to the raw wastewater, sludge cake
concentrations dropped from 15.8 percent to 13.0
percent. Cationic polymer was used as a conditioning
agent at similar dosages in both cases. The most
significant machine variables were beach residence
time and liquid ring height (38). Estimated sludge
handling costs were slightly lower for iron addition
than for aluminum addition (39).
As discussed in Section 5.3.6.3, process variables
affecting centrifuge performance include source and
type of sludge, feed solids content, percentage of
chemical solids, feed rate, and conditioner dosage.
Machine variables for a solid bowl centrifuge include
bowl design and speed, pool volume, conveyor
speed, and conveyor pitch (35).
Pilot testing is strongly recommended where possible,
as the variability in sludge characteristics makes
performance impossible to predict. Procedures are
available from the various manufacturers for scale-
up of pilot-test results.
5.4.6.4 Pressure Filtration
Limited information is available on pressure filtration
of iron sludges. In the EPA survey, plants reporting
106
-------
use of plate and frame filter presses on iron sludges
or combinations of iron and conventional sludges
included Saline, Michigan; Kenosha, Wisconsin; and
Brookfield, Wisconsin (1). At Saline, anaerobically
digested iron-primary plus secondary sludge at 9
percent TS was dewatered to 41 percent TS; at
Kenosha, anaerobically digested primary plus iron-
secondary sludge at 4.8 percent TS was dewatered
to 40 percent TS; and at Brookfield aerobically
digested primary plus iron-secondary sludge at 7 to
8 percent TS was dewatered to 43 percent TS (1).
At Brookfield, Wisconsin, filter press performance
improved substantially upon addition of pickle liquor
(ferrous sulfate) to the aeration basins for phosphorus
removal. The sludge mass feed rate (not including
admixtures) increased 70 percent from 227 to 384 kg
TS/hr (500 to 845 Ib/hr); average run length
decreased by 40 percent from 2.8 to 1.7 hr/run; and
the percentage of solids fed to the filter which were
recovered in the cake increased from 75 to 90
percent. The cake solids content remained essentially
constant at 43 percent TS (1).
Mininni et al. investigated pressure filtration of waste
activated sludge from a plant with no primary
clarification. Upon addition of ferrous chloride to raw
wastewater for phosphorus control, cake solids
decreased from 32.1 to 30.0 percent TS. Aluminum
chlorohydrate was used as a conditioner at dosages
of 12.7 and 12.1 g AI2C>3/kg (25.4 and 24.2 Ib/ton) of
dry solids during the periods of no chemical addition
and during ferrous sulfate addition, respectively
(38,39).
For further general information on pressure filtration of
sludge the reader is referred to the EPA Process
Design Manual on Sludge Treatment and Disposal
(4).
5.4.6.5 Belt Pressure Filtration
Many plants are now using belt filter presses for
sludge dewatering, since, in general, dryer sludge
cakes are possible compared to vacuum filters and
centrifuges, and the devices are economically
competitive with other dewatering equipment.
Unfortunately, virtually no information is available in
the literature regarding their performance on sludges
derived from the addition of iron salts for phosphorus
removal.
Performance data and design criteria for belt filter
presses used for dewatering of non-chemical
sludges may be found in Reference 4. Pilot-testing
is recommended if belt filter presses are under
consideration for dewatering of sludges from chemical
phosphorus removal systems.
5.4.7 Incineration
Of 22 plants contacted in the EPA survey of
phosphorus removal facilities which employed
incineration, six reported significant impacts on
incineration as a result of chemical addition for
phosphorus removal. Two of these plants used iron
salts for phosphorus removal. These were Wyandotte,
Michigan and Sheboygan, Wisconsin.
Wyandotte reported major problems with clinker
buildup in the multiple hearth incinerator. This
increased the time required for cleaning drop holes in
the hearth, and increased wear on the rabble arms.
Sheboygan operates a fluidized bed incinerator, which
receives sludge cake from a vacuum filter. Ferric
chloride is added to the effluent of the trickling filters
for phosphorus removal. Phosphorus removal has
adversely affected incinerator capacity due to the
increased moisture content of the sludge cake from
the vacuum filter. Cake solids dropped from 25.5 to
21.5 percent after phosphorus removal was initiated.
The volatile fraction was reduced from 73-74
percent to 65 percent. The overall impact was a
reduction in incinerator feed rate from 755 to 475 kg
dry solids/hr (1,660 to 1,050 Ib/hr), and an increase in
fuel consumption from 246 to 517 l/Mg (59 to 124
gal/ton) (1).
Other problems associated with incineration of
chemical sludge have been reported at Sheboygan.
Slag formation has caused plugging of tuyeres and
clogging of exhaust lines, increasing the frequency of
inspections and maintenance. In addition, the plant
manager believes that the ferric chloride in the sludge
is responsible for a high rate of corrosion of metal
ductwork (1).
It is apparent from the Sheboygan experience that
ferric chloride addition for phosphorus removal can
impact critical incineration parameters such as: 1)
cake moisture content, 2) calorific value of the
sludge, and 3) relative proportion of volatile and inert
material. Table 5-9 lists the potential problems
associated with chemical sludge incineration, and
recommended solutions (40).
5.4.8 Disposal
As discussed in Section 5.3.8 for disposal of alum
sludges, the most significant impact of chemical
addition on sludge disposal considerations is the
increased sludge mass and volume for disposal,
resulting in increased disposal costs.
Information in the literature indicates that the
presence of iron precipitates does not adversely
affect the agricultural value of the sludge with respect
to crop yield, organic matter degradation, nitrogen
availability, and other soil parameters (41-45).
However, an increase in metal content of the sludge
may be expected as a result of precipitation with iron.
This may impact allowable loading rates to agricultural
land. Heavy metal analyses must be conducted on a
case-by-case basis for the sludge to be applied.
107
-------
Application rates, whether governed by nitrogen or
heavy metal loadings, will be determined by state
regulations.
Iron addition is unlikely to affect the suitability of a
sludge for disposal at a sanitary landfill. However,
mass landfills require that organic materials comply
with a maximum moisture content criterion. As iron
addition may increase moisture content in sludge
cakes from dewatering processes, this may limit the
suitability for landfill disposal.
5.5 Sludges Derived from Biological
Phosphorus Removal Processes
5.5.7 Characteristics
Sludges derived from biological phosphorus removal
systems exhibit properties similar to conventional
biological sludges. The only possible exception is
sludge derived from the Phostrip process in which a
portion of the total sludge results from lime addition to
the anaerobic stripper vessel. Even with the Phostrip
process, the volume of lime sludge is relatively small
compared to the combined volume of primary and
waste activated sludge.
Because of the mechanism of excess phosphorus
uptake in biological phosphorus removal systems,
resulting waste activated sludges tend to have higher
phosphorus concentrations than conventional
sludges. Typical phosphorus concentrations in waste
activated sludge from the Bardenpho and A/0
processes are 4-6 percent by weight vs. 2-3
percent for conventional waste activated sludges.
Resolubilization of phosphorus during anaerobic
storage of lime- precipitated sludge in the Phostrip
process is unlikely to occur, since the phosphorus is
bound to the calcium ion. However, with A/O,
Bardenpho, and other "pure" biological phosphorus
removal systems, it is recommended that waste
activated sludge be kept aerobic in order to prevent
phosphorus solubilization.
5.5.2 Sludge Generation Rates
Sludge generation rates from biological phosphorus
removal systems are not expected to be significantly
different than for conventional activated sludge
systems, and solids production will vary with
wastewater characteristics and operational
parameters such as SRT.
The Phostrip process will likely generate somewhat
greater masses of sludge because of lime addition to
the anaerobic stripper supernatant. Knowing the
characteristics of the supernatant and the necessary
pH to remove the desired quantity of phosphorus, the
quantity of lime sludge produced can be estimated
using the procedure described in Section 5.6.
Theoretically, some increase in sludge production
would be expected for biological phosphorus removal
systems due to the increased mass of phosphorus
taken up by the organisms. This will be dependent on
the phosphorus content of the waste activated sludge
(WAS). As shown in Section 3.5.2.1, the theoretical
waste activated sludge yield would increase by 8.5
percent if the phosphorus content of the WAS
increased from 2 to 4 percent by weight. If the
phosphorus content increased to 5 percent by weight,
the theoretical mass of WAS production would
increase by 13 percent. It should be noted that on a
volumetric basis, the increased sludge mass may be
counteracted by an improvement in settling
characteristics, as SVI values of less than 80 ml/g
have been reported for the Modified Bardenpho and
A/O processes.
5.5.3 Thickening
Because of the potential release of phosphorus during
gravity thickening of waste activated sludge from
biological phosphorus removal systems, use of
dissolved air flotation thickening is recommended.
This would apply to the purely biological systems, and
would not be a concern with lime sludges derived
from treatment of the anaerobic stripper supernatant
in the Phostrip process. Lime sludges can be
combined with other sludges and handled by
conventional sludge handling processes without
special consideration for phosphorus release. With
"pure" biological systems, however, choice of sludge
handling processes must account for potential
phosphorus resolubilization if thickened, stored, or
stabilized in the absence of oxygen.
5.5.4 Stabilization
Little data are available regarding the fate of
phosphorus during aerobic or anaerobic stabilization
of sludges from biological phosphorus removal
systems. Phosphorus resolubilization would be
anticipated during anaerobic digestion. However, at
Pontiac, Michigan, significant levels of phosphorus in
anaerobic digester supernatant were not observed,
possibly due to formation of an ammonium-
magnesium-phosphate precipitate in the digester.
Phosphorus release may also be possible during
aerobic digestion due to destruction and lysing of
biological solids.
Because of the lack of information on this subject,
consideration may have to be given to chemical
treatment of digester supernatants for phosphorus
removal in order to minimize return of phosphorus to
the head of the plant. Further studies are needed to
assess the magnitude of phosphorus release during
stabilization of biological phosphorus removal sludges.
In some biological phosphorus removal systems
employing long solids retention times, phosphorus-
laden sludges are subjected to dewatering without
108
-------
separate stabilization. The acceptability of this
practice is dependent on regulations as to whether
such sludges are considered stabilized or whether
separate stabilization is required prior to land
disposal.
5.5.5 Conditioning
Sludges from biological phosphorus removal systems
are expected to have similar conditioning
requirements to conventional non-phosphorus
sludges. Blending of lime sludge with other sludges in
the Phostrip process may reduce overall conditioning
requirements.
5.5.6 Dewatering
Sludges from biological phosphorus removal systems
are expected to have dewatering characteristics
similar to those from conventional activated sludge
systems. The lime sludge from the Phostrip process
is not be expected to adversely affect dewatering,
and, based on experience with dewatering of lime
sludges alone, may improve dewatering
characteristics when blended with other sludges. No
specific design information regarding loading rates to
dewatering equipment is available for biological
phosphorus removal sludges. Design criteria for
conventional primary and waste activated sludges
should be used to size dewatering equipment for
sludges from biological phosphorus removal
processes if pilot- or full-scale performance data
are not available.
5.5.7 Incineration
No unique problems are associated with incineration
of sludges from biological phosphorus removal
systems. Sludges from "pure" biological phosphorus
removal prcesses will have volatile solids contents
and Btu values similar to those of conventional
biological wastewater treatment sludges. Phostrip
sludges, if lime sludge from treatment of the stripper
supernatant is included, may have slightly lower
volatile solids contents and Btu values due to the
addition of inert solids from the lime addition step.
However, the overall impact is expected to be small.
5.5.8 Disposal
Sludges from biological phosphorus removal systems
can be disposed of in the same manner as sludges
from conventional biological systems. Higher
phosphorus contents may make sludges from
biological phosphorus removal systems particularly
attractive for agricultural utilization.
5.6 Sludges Derived from Addition of
Lime
Lime may be employed for phosphorus removal by
addition to the primary clarifier or tertiary treatment
process. Lime is also used to remove phosphorus
from the effluent of the phosphorus stripper tanks in
the Phostrip biological phosphorus removal systems
(47). The characteristics of the sludge produced are
dependent on where the lime is added, whether the
low lime or high lime process is used, the alkalinity of
the water, and whether the various wastewater plant
sludges are combined with the lime sludge before
processing and in what proportions they are
combined.
As discussed in Section 4.1, very few plants in the
United States use lime for phosphorus removal. Many
plants originally designed to use lime have abandoned
the lime systems in favor of aluminum or iron salts.
The major disadvantages of lime compared to metal
salts are significant increase in sludge mass, and
greater operation and maintenance requirements for
cleaning and maintaining lime handling equipment.
The following is a brief discussion of lime sludges
with references provided if further detail is desired.
Two general statements about lime sludges can be
made. One is that use of lime for phosphorus removal
results in larger volumes of sludge before thickening
than does the use of metal salts for phosphorus
removal (1,48). Secondly, lime sludges generally
improve the thickening and dewatering properties of
wastewater treatment sludges when lime sludge and
non-lime wastewater sludges are mixed (13).
Due to the fact that the nature of the sludge produced
will vary with the specific wastewater being treated
and the composition of the sludge, the design of the
sludge handling system must be based at a minimum
on laboratory tests such as settleometer tests,
specific resistance, filter leaf, capillary suction time
and other tests. Where possible, pilot- or full-scale
sludge thickening and dewatering devices should be
used to establish design criteria for full-scale sludge
thickening and dewatering devices.
Specific differences expected to be found in the
different lime treatment systems are listed below.
A. Low Lime Treatment - Primary Addition.
1. The sludge contains a smaller percentage of
organics than does high lime treatment (49).
2. The sludge contains no magnesium hydroxide,
which is a gelatinous precipitate that is difficult
to settle (13,49).
B. High Lime Treatment - Primary Addition
1. Sludge removed from high lime treatment has a
higher solids content than that from low lime
treatment (50).
2. The sludge contains magnesium hydroxide, a
gelatinous precipitate.
109
-------
C. Tertiary Lime Process (11).
1. Tertiary lime sludges are similar to primary lime
sludges but do not contain nearly the quantity of
organic materials as primary sludge.
2. The high lime tertiary process will produce a
sludge that behaves much like sludge from a
water softening process. The thickening and
dewaterability of this sludge decreases with
increasing magnesium concentration in the
sludge.
3. High alkalinity waters, (greater than 200 mg/l
CaCOa) require special consideration for
thickening of the sludge as the thickening
operations can become unmanageable.
4. Recycle of sludge from the settling unit
improves thickening of the sludge.
Specific data on the thickening, dewatering and
disposal of lime sludges are contained in the following
sections. Recommended reading for primary lime
sludges are articles by Minton and Carlson (48), who
did a number of studies on these sludges in the early
1970's, and Parker (50,51).
Theoretical calculations of sludge generation rates
appear to give good correlation with observed sludge
volumes in plant studies for tertiary lime addition but
underestimate the quantity for primary lime addition
(48). The quantity of sludge produced by lime addition
depends not only on the degree of phosphate
removal desired, but also on the magnesium
concentration, alkalinity and other characteristics of
the wastewater (52). In fact, both the amount of
sludge produced and the degree of phosphorus
removal achieved by raising the pH to a given level
depends on the nature of the wastewater.
Increases in the mass of sludge solids produced by
lime addition systems have been reported to be
double the mass produced by conventional primary
and secondary treatment systems (53). Minton and
Carlson have reported the mass of lime sludges to be
as much as two to three times the mass of
conventional primary and secondary treatment
systems when lime is added to raw wastewater.
An important factor that must be kept in mind is that
while the sludge mass increases, lime sludges
generally thicken and dewater to high solids content
and the volume of dewatered sludge that needs to be
processed is not necessarily two to three times the
volume of sludge from a wastewater plant without
chemical phosphorus removal.
Procedures have been developed for calculating the
sludge quantity produced from lime sludges. These
procedures were developed for tertiary lime
applications but can be used for raw sewage or other
applications. The procedure is described in detail in
the 1976 EPA Phosphorus Design Manual (4).
5.7 Case Histories
5.7.7 Baltimore, Maryland (54)
The Back River Wastewater Treatment Plant provides
secondary treatment of municipal-industrial
wastewater from Baltimore City and Baltimore County,
Maryland. Liquid stream processes include
preliminary treatment, primary clarification, biological
treatment by parallel trickling filters and activated
sludge process, secondary clarification, and
chlorination. Primary and waste activated sludges are
thickened by gravity, anaerobically digested,
chemically conditioned with polymer, and dewatered
using vacuum filters. Phosphorus is removed by
addition of iron as waste pickle liquor to the activated
sludge aeration basins at a dosage of approximately 5
mg/l as Fe. Full-time addition of waste pickle liquor
began in June, 1981.
Data on sludge handling operations were compared
for periods with and without waste pickle liquor
addition. Prior to waste pickle liquor addition, average
daily sludge production was 62 Mg (69 tons)/d dry
solids, and 353 Mg (390 tons)/d wet. After pickle
liquor addition, average daily sludge production was
80 Mg (88 tons)/d dry solids, and 414 Mg (456
tons)/d wet. This amounted to an increase of 29
percent on a dry solids basis, and 17 percent on a
wet solids basis following initiation of phosphorus
removal. The smaller percentage increase in wet
sludge production was due to generation of a drier
vacuum filter cake as a result of waste pickle liquor
addition. Cake solids increased from 17.6 percent to
19.1 percent, an improvement of 9 percent.
Coupled with an increase in vacuum filter cake solids
was reduction in polymer required for sludge
conditioning. Prior to iron addition, polymer dose
based on dry solids, was 34 kg/Mg (68 Ib/ton); after
iron addition, polymer dose was 30 kg/Mg (60 Ib/ton),
a decrease of approximately 11 percent.
Thickening characteristics of combined raw primary
and secondary sludges did not appear to change
significantly as a result of pickle liquor addition.
Changes were attributed to operational modifications.
Figure 5-5 summarizes the costs of sludge
conditioning and hauling before and after phosphorus
removal. It should be noted that these costs are only
for chemical conditioning and dewatered sludge
hauling, and do not account for additional operation
and maintenance labor or energy requirements.
5.7.2 Lorton, Virginia (55)
The Lower Potomac Water Pollution Control Plant is a
136,000-m3/d (36-mgd) activated sludge facility
110
-------
Figure 5-5 Comparison of sludge quantities, conditioning costs, and hauling costs before and after phosphorus removal;
Baltimore, MD.
62 dry t/d
2,110 hg/d polymer
@$0.15/kg
= $320/d
Before Phosphorus Removal
353 wet t/d hauled
@ $38.31/t
= $l3,520/d
Total Cost
= $l3,840/d
80 dry t/d
2,400 hg/d polymer
@$0.15/kg
= $360/d
After Phosphorus Removal
414 wet t/d hauled
@ $38.31/1
= $15,860/d
Total Cost
= $16,220/d
with advanced wastewater treatment for phosphorus
removal. Liquid stream processes consist of bar
screens, primary clarifier, activated sludge basins,
secondary clarifier, flow equalization basins, tertiary
chemical clarifiers for phosphorus precipitation, and
tertiary filters. Primary sludge is degritted and gravity
thickened. Waste activated sludge is thickened by
dissolved air flotation. Primary and waste activated
sludges are blended in a storage tank, chemically
conditioned with lime and ferric chloride, and
dewatered by vacuum filtration. Dewatered sludge
cake, at 16-18 percent solids, is incinerated in
multiple-hearth furnaces. The tertiary chemical
sludge is conditioned with lime, gravity thickened to
2-3 percent solids, conditioned with anionic polymer,
and dewatered in solid bowl centrifuges. Dewatered
sludge cake, at a solids content of 15-16 percent, is
mixed with incinerated sludge ash and disposed of in
a sanitary landfill.
Originally the plant was designed for phosphorus
removal in a tertiary process using a two-stage lime
and recarbonation system. Because of numerous
design, operation, and maintenance problems and
high operating costs, the plant was modified to
remove phosphorus using ferric chloride. Ferric
chloride is normally added to the plant influent,
activated sludge process, and influent to tertiary
clarification. Ferrous sulfate addition to the return
activated sludge was investigated during the period
April through July of 1983. During this time, ferric
chloride addition to the plant influent and influent to
tertiary clarification was maintained. From June
through October, 1984, ferrous sulfate was added to
the plant influent while ferric chloride addition to the
activated sludge process and tertiary clarifiers was
maintained. Results of these investigations with
respect to impact on solids handling are discussed
below.
Ferrous sulfate addition to return activated sludge
Total suspended solids removal in the secondary
clarifier improved significantly during ferrous sulfate
addition to the return activated sludge. TSS removal
efficiencies were 87 percent during ferrous sulfate
addition and 80 percent during ferric chloride addition.
Performance of dissolved air flotation thickening of
waste activated sludge was slightly improved during
ferrous sulfate addition, with thickened solids
concentrations increasing from 4.4 to 4.9 percent by
weight. However, the improved performance was not
solely attributed to ferrous sulfate addition.
Ferrous sulfate addition to plant influent
Gravity thickening of primary sludge improved with
addition of ferrous sulfate to the plant influent, with
average thickened solids concentrations increasing
from 6.6 to 8.7 percent.
111
-------
Performance of vacuum filters improved during the
period of ferrous sulfate addition to the plant influent.
Average total solids concentration in the vacuum filter
sludge cake increased from 16.4 percent during ferric
chloride addition to 18.7 percent during ferrous sulfate
addition. There was no change in chemical
requirements for conditioning.
Another observation was that the chemical "ferric"
sludge from the tertiary clarifier dewatered to a higher
cake solids content when ferrous sulfate was added
to the influent, increasing from 15-16 percent to
approximately 19 percent. Although no explanation
was provided as to the reason for the improvement,
the same phenomenon was observed when ferrous
sulfate addition was temporarily suspended and then
restarted.
The major conclusion from the study was that,
although ferrous sulfate was more costly than ferric
chloride per unit of phosphorus removed, the savings
resulting from improved thickening and dewatering
characteristics may make its use cost-effective for
this application.
5.8 Costs
No estimates have been provided for the costs
incurred in handling additional sludge associated with
phosphorus removal processes. This is partly due to
the wide variability in sludge generation rates as a
result of variation in process types, operating
strategies, chemical used for phosphorus removal (if
any), effluent discharge limitations, wastewater
characteristics, and other factors. Costs are
dependent not only on the additional quantity of
sludge to be processed, but also on changes in
thickening and dewatering characteristics of the
sludge, types of sludge handling processes
employed, existing sludge handling capacity, labor
and energy costs, and available sludge disposal
options.
Estimates of the additional costs for sludge handling
upon implementation of phosphorus removal can only
be made on a site-specific basis. Sludge generation
rates can be estimated using the procedures outlined
in this chapter. Laboratory tests can be used to
estimate the impact of phosphorus removal on
thickening and dewatering characteristics. A capacity
analysis of sludge handling equipment must be
conducted in order to determine if more tankage or
equipment will be required to process the additional
sludge. A similar analysis of manpower schedules and
requirements must also be made. Finally, the overall
impact on the volume of sludge to be disposed of
must be assessed in order to estimate additional
disposal costs. For chemical precipitation of
phosphorus, cost for sludge handling and disposal is
likely to account for a major portion of the additional
costs associated with phosphorus removal,
cannot be ignored.
and
5.9 References
When an NTIS number is cited in a reference, that
reference is available from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
(703) 487-4650
1. Schmidt, C.J., Hammer, L.E. and M.D. Swayne.
Review of Techniques for Treatment and Disposal
of Phosphorus - Laden Chemical Sludges.
EPA-600/2-79-083, NTIS No. PB80-117542,
U.S. Environmental Protection Agency, Municipal
Environmental Research Laboratory, Cincinnati,
OH, 1979.
2. Process Design Manual for Phosphorus
Removal. EPA-625/1-76-001a, NTIS No.
PB-259150, U.S. Environmental Protection
Agency, Center for Environmental Research
Information, Cincinnati, OH, 1976.
3. Knight, C.H., Mondoux, R.G. and B. Hambley.
Thickening and Dewatering Sludges Produced in
Phosphate Removal. In Phosphorus Removal
Design Seminar Conference Proceedings No. 1,
Canada- Ontario Agreement, Toronto, 1973.
4. Process Design Manual: Sludge Treatment and
Disposal. EPA 625/1-79-011, U.S.
Environmental Protection Agency, Center for
Environmental Research Information, Cincinnati,
OH, 1979.
5. Wastewater Engineering. Metcalf and Eddy, Inc.,
McGraw-Hill Book Co., New York, 1972.
6. Keinath, T.M., Ryckman, M.D., Dana, C.H. and
D.A. Hofer. Design and Operational Criteria for
Thickening of Biological Sludges, Parts I, II, III,
IV. Water Resources Research Institute, Clemson
University. September, 1976.
7. Gravity Thickening. In: Process Design
Techniques for Industrial Waste Treatment. Edited
by C.E. A
-------
PB-299593, U.S. Environmental Protection
Agency, Center for Environmental Research
Information, Cincinnati, OH, 1978.
10. DiQregorio, D., Ainsworth, J.B. and K.J,
Mounteer. Chemical Primary Sludge Thickening
and Dewatering. EPA-600/2-79-055, U.S.
Environmental Protection Agency, Municipal
Environmental Research Laboratory, Cincinnati,
OH, 1979.
11. Nutrient Control. Manual of Practice FD-7. Water
Pollution Control Federation, Washington, D.C.,
1983.
12. Kos, P. Gravity Thickening of Water Treatment
Plant Sludge. JAWWA, Research Edition, 272-
279, May, 1977.
13. Dick, R.I. Management of Phosphorus-Laden
Sludges. In: Proceedings of the International
Conference: Management Strategies for
Phosphorus in the Environment. Edited by J.N.
Lester and P.W. Kirk, Lisbon, Portugal, July,
1985.
14. Long, D.A. and J.B. Nesbitt. Removal of Soluble
Phosphorus in an Activated Sludge Plant. JWPCF
47(1 ):170-184, 1975.
15. Martel, C.J., DiGiano, F.A. and R.E. Pariseau.
Phosphorus Removal in Extended Aeration
Systems by Chemical Clarification. JWPCF
51(1):140-149, 1979.
16. Lin, S.S. and D.A. Carlson. Phosphorus Removal
by the Addition of Aluminum to the Activated
Sludge Process. JWPCF 47(7): 1978-1986,
1975.
17. Baillod, C.R., Cressey, G.M. and R.T. Beaupre.
Influence of Phosphorus Removal on Solids
Budget. JWPCF 49(1):131-145, 1977.
18. Ganczarczyk, J. and M.F.D. Hamoda. Aerobic
Digestion of Organic Sludge Containing Inorganic
Phosphorus Precipitates, Phase 1. Canada -
Ontario Agreement on Great Lakes Water Quality,
Research Report No. 3, Environment Canada,
Ottawa, June, 1973.
19. Mitchell, G.F., Seyfarth, R.H. and F.R. Wilson.
Effect of Alum Addition on Aerobic Digestion of
Activated Sludge. Water and Sewage Works
124(7):58-62, 1977.
20. Eikum, A.S., Carton, D.A. and A. Lundar.
Phosphorus Release During Storage of
Aerobically Digested Sludge. JWPCF,
47(2):330-337, 1975.
21. Grigoropoulos, S.G., Vedder, R.C. and D.W. Max.
Fate of Aluminum - Precipitated Phosphorus in
Activated Sludge and Anaerobic Digestion.
JWPCF 43(12):2366-2382, 1971.
22. O'Shaughnessy, J.C., Nesbitt, J.B., Long, D.A.
and R.R. Kountz. Digestion and Dewatering of
Phosphorus-Enriched Sludges. JWPCF
46(8):1941-1962, 1974.
23. Dentel, S.K. and J.M. Gossett. Effect of Chemical
Coagulation on Anaerobic Digestibility of Organic
Materials. Water Research 16:707-718, 1982.
24. Gossett, J.M., McCarthy, P.L., Wilson, J.C. and
D.S. Evans. Anaerobic Digestion of Sludge from
Chemical Treatment. JWPCF 50(3):533-542,
1978.
25. Farrell, J.B. Design Information on Dewatering
Properties of Wastewater Sludges. Sludge
Handling and Disposal Seminar Conference
Proceedings No. 2, Environment Canada,
Toronto, Ontario, 1974.
26. Williamson, D.J. and G. Wheale. The Influence of
Sludge Storage on Chemical Conditioning Costs.
Water Pollution Control, 80:529-536, 1981.
27. Novak, J.T. and J.H. O'Brien. Polymer
Conditioning of Chemical Sludges. JWPCF
47(10):2397-2410, 1975.
28. Hudgins, R.R. and P.L. Silveston. Wet Air
Oxidation of Chemical Sludges. Research Report
No. 12, Canada-Ontario Agreement on Great
Lakes Water Quality, Environment Canada,
Ottawa, Ontario, 1973.
29. Farrell, J.B., Smith, J.E., Dean, R.B., Grossman,
E. and O.L. Grant. Natural Freezing for
Dewatering of Aluminum Hydroxide Sludges.
JAWWA 62(12):787-791, 1970.
30. Reed, S., Bouzoun, J. and W. Medding. A
Rational Method for Sludge Dewatering Via
Freezing. JWPCF 50(9):911-916, 1986.
31.Novak, J.T. and G.E. Montgomery. Chemical
Sludge Dewatering on Sand Beds. Journal
ASCE-EED 101(2):1-14, 1975.
32. Sludge Dewatering. Manual of Practice No. 20.
Water Pollution Control Federation, Washington,
D.C., 1983.
33. Stickney, R. and B.P. LeClair. The Use of
Physicochemical Sludge Characteristics and
Bench Dewatering Tests in Predicting the
Efficiency of Thickening and Dewatering
Processes. Sludge Handling and Disposal
113
-------
Seminar Conference Proceedings No. 2,
Environment Canada, Ottawa, 1974.
34. Van Fleet, G.L., Barr, J.R. and AJ. Harris.
Treatment and Disposal of Phosphate Sludges in
Ontario. Presented at the Annual Conference of
WPCF, Atlanta, GA, 1972.
35. Wastewater Treatment Plant Design. WPCF
Manual of Practice No. 8, Water Pollution Control
Federation, Washington, D.C., 1977.
36. Campbell, H.W. and B.P. LeClair. Sludge
Dewatering Alternatives for Waste Activated
Sludges from Phosphorus Removal Facilities.
JWPCF 51(5):991-998, 1979.
37. Campbell, H. and B.P. LeClair. Effects of Control
Variables and Sludge Characteristics on the
Performance of Dewatering and Thickening
Devices. In: Sludge Handling and Disposal
Seminar Conference Proceedings No. 2.
Environment Canada, Ottawa, 1974.
38. Mininni, G., Passino, Ft., Santori, M. and L.
Spinosa. Sludge Dewatering in A Conventional
Plant with Phosphorus Removal-ll: Study on
Centrifuge and Filter Press Performance. Water
Research 19(2): 151-156, 1985.
39. Mininni, G., Passino, R., Santori, M. and L.
Spinosa. Sludge Dewatering in a Conventional
Plant with Phosphorus Removal-l: Analysis of
Additional Costs. Water Research 19(2):143-
149, 1985.
40. Shannon, E.E., Plummer, D. and P.J.A. Fowlie.
Aspects of Incinerating Chemical Sludges. In:
Sludge Handling and Disposal Seminar,
Conference Proceedings No. 2. Environment
Canada, Ottawa, 1974.
41. Gaynor, J.D. Soil Degradation of Wastewater
Sludges Containing Chemical Precipitants.
Environmental Pollution:57-64, 1979.
42. Soon, Y.K., Bates, I.E., Beauchans, E.G. and
J.R. Moyer. Land Application of Chemically
Treated Sludge: I. Effects on Crop Yield and
Nitrogen Availability. J. Environmental Quality
7(2):264-269, 1978.
43. Soon, Y.K., Bates, I.E. and J.R. Moyer. Land
Application of Chemically Treated Sewage
Sludge: II. Effects on Plant and Soil Phosphorus,
Potassium, Calcium, and Magnesium and Soil
pH. J. Environmental Quality 7(2):269-273,
1978.
44. Chawla, V.K., Bryant, D.N. and D. Liu. Disposal
of Chemical Sewage Sludges on Land and Their
Effects on Plants, Leachate, and Soil Systems. In:
Sludge Handling and Disposal Seminar,
Conference Proceedings No. 2. Environment
Canada, Ottawa, 1974.
45. Kirkham, M.B. and G.K. Dotson. Growth of Barley
Irrigated with Wastewater Sludge Containing
Phosphorus Precipitants. In: Proceedings of the
National Conference on Municipal Sludge
Management, Pittsburgh, PA, June 11-13, 1974.
46. Malhotra, S.K., Parrillo, T.P. and A.G.
Hartenstein. Anaerobic Digestion of Sludges
Containing Iron Phosphates. Journal ASCE,
Sanitary Engineering Division 97(SA5):629-645,
1971.
47. Heim, C.J., The Phostrip Process for Phosphorus
Removal. Proceedings of the International
Seminar on Control of Nutrients in Municipal
Wastewater Effluents, Volume 1:Phosphorus. San
Diego, CA, 1980.
48. Minton, G.R. and D.A. Carlson. Effects of Lime
Addition on Treatment Plant Operation. JWPCF
48(7):1697-1727, 1976.
49. Mulbarger, M.C., Grossman, E., Dean, R.B. and
O.L. Grant. Lime Clarification Recovery, Reuse,
and Sludge Dewatering Characteristics. JWPCF
41(12):2070-2085, 1969.
50. Parker, D.S., Zadich, F.J. and K.E. Train. Sludge
Processing for Combined Physical-Chemical-
Biological Sludges. EPA-R2-73-250, NTIS
No. PB-223334, U.S. Environmental Protection
Agency, 1973.
51. Parker, D.S., de la Fuente, E., Britt, L.O.,
Spealman, M.L., Stenquist, R.J. and F.J. Zadich.
Lime Use in Wastewater Treatment: Design and
Cost Data. EPA-600/2-75-038, NTIS No.
PB-248181, U.S. Environmental Protection
Agency, Municipal Environmental Research
Laboratory, Cincinnati, OH, 1975.
52. Farrell, J.B. Interim Report of Task Force on
Phosphate Removal Sludges. NTIS No. PB-
238317, U.S. Environmental Protection Agency,
National Environmental Research Center,
Cincinnati, OH, 1975.
53. Minton, G.R. and D.A. Carlson. Primary Sludges
Produced by the Addition of Lime to Raw Waste
Water. Water Research 7:1821- 1847, 1973.
54. The Effect of a Ban on Phosphates in Household
Detergents on the Treatment of Municipal
Wastewater. A report prepared for the Soap and
114
-------
Detergent Association by Gannett Fleming
Environmental Engineers, Inc., February, 1985.
55. Hogge, A.H., Jenkins, J.D. and I.A. Khan. Full
Scale Evaluation of Ferrous Sulfate for
Phosphorus Removal and Its Impact on Solids
Handling. Paper presented at the 58th Annual
Conference of the WPCF, Kansas City, Missouri,
October, 1985.
115
-------
Terms
BOD biochemical oxygen demand
BODs 5-day BOD
COD chemical oxygen demand
DO dissolved oxygen
F/M food-to-microorganism loading
gpd gallons per day
gpm gallons per minute
HOT hydraulic detention time
HRT hydraulic retention time
mgd million gallons per day
mg/l milligrams per liter
MLSS mixed liquor suspended solids
MLVSS mixed liquor volatile suspended solids
NH4-N ammonium nitrogen
NOa-N nitrate nitrogen
SBOD soluble 5-day BOD
SDT stripper detention time
SP soluble phosphorus
SRT solids retention time
SS suspended solids
TBOD total 5-day BOD
TKN total Kjeldahl nitrogen
TN TKN plus oxidized nitrogen
TP total phosphorus
TSS total suspended solids
VSS volatile suspended solids
116
U.S. GOVERNMENT PRINTING OFFICE: 1987- 748-121/67006
------- |