PROCESS DESIGN MANUAL
                FOR
         NITROGEN CONTROL
U.S. ENVIRONMENTAL PROTECTION AGENCY

           Technology Transfer
             October 1975

-------
                             ACKNOWLEDGEMENTS
This design  manual  was prepared  for  the  Office of Technology Transfer of the U.S.
Environmental Protection Agency. Coordination and preparation of the manual was carried
out by the firm  of Brown and Caldwell, Walnut Creek, California, under the direction of
Denny S. Parker  with assistance from Richard W. Stone and Richard J. Stenquist. Chapters
7, 8, and portions of Chapter 9 were  prepared by  Gordon Gulp of Culp/Wesner/Culp.
Clair N. Sawyer and Perry L. McCarty served as consultants to the U.S. EPA for the purpose
of reviewing portions of the text. U.S.  EPA reviewers were Edwin F. Earth and Irwin J.
Kugelman of the U.S. EPA National Environmental Research Center, Cincinnati, Ohio, and
Robert S. Madancy of the Office of Technology Transfer, Washington, D.C.
                                    NOTICE
The mention  of trade names of commercial products in this publication is for illustration
purposes and does not constitute endorsement or recommendation for use by the U.S.
Environmental Protection Agency.
                                        11

-------
                                     ABSTRACT
This manual presents  theoretical and process design  criteria for the implementation of
nitrogen control technology in municipal wastewater treatment facilities. Design concepts
are emphasized as much as possible through examination of data from full-scale and pilot
installations.

Design  data  are  included on biological  nitrification  and  denitrification,  breakpoint
chlorination, ion exchange and air stripping. One chapter presents the concepts involved in
assembling various  unit  processes into rational treatment trains and presents  actual case
examples of specific treatment systems that incorporate nitrogen control processes.
                                        111

-------
                                 CONTENTS


Chapter                                                              Page

          ACKNOWLEDGEMENTS                                      ii

          ABSTRACT                                                 iii

          CONTENTS                                                 v

          LIST OF FIGURES                                           xviii

          LIST OF TABLES                                            xxiii

          FOREWORD                                                xxvii

   1       INTRODUCTION                                            1-1

          1.1  Background and Purpose                                   1-1
          1.2  Scope of the Manual                                      1-2
          1.3  Guide to the User                                         1-2

   2       NITROGENOUS MATERIALS IN THE ENVIRONMENT AND
            THE NEED FOR CONTROL IN WASTEWATER EFFLUENTS      2-1

          2.1  Introduction                                             2-1
          2.2  The Nitrogen Cycle                                       2-1

               2.2.1   The Nitrogen Cycle in Surface Waters and Sediments      2-5
               2.2.2  The Nitrogen Cycle in Soil and Groundwater             2-5

          2.3  Sources of Nitrogen                                       2-8

               2.3.1   Natural Sources                                    2-8
               2.3.2  Man-caused Sources                                 2-9

          2.4  Effects of Nitrogen Discharge                               2-12

               2.4.1   Biostimulation of Surface Waters                      2-12
               2.4.2  Toxicity                                           2-13
               2.4.3  Effect on Disinfection Efficiency                      2-13
               2.4.4  Dissolved Oxygen Depletion in Receiving Waters          2-14
               2.4.5  Public Health                                      2-14
               2.4.6  Water Reuse                                       2-16

-------
                              CONTENTS - Continued


Chapter                                                                    Page

           2.5  Treatment Processes for Nitrogen Removal                     2-16

                2.5.1  Conventional Treatment Processes                       2-16
                2.5.2  Advanced Wastewater Treatment Processes               2-17
                2.5.3  Major Nitrogen Removal Processes                      2-17

                      2.5.3.1  Biological Nitrification-Denitrification           2-18
                      2.5.3.2  Breakpoint Chlorination                        2-18
                      2.5.3.3  Selective Ion Exchange for Ammonium Removal  2-19
                      2.5.3.4  Air Stripping for Ammonia Removal             2-19

                2.5.4  Other Nitrogen Removal Processes                      2-20
                2.5.5  Summary                                            2-20

           2.6  References                                                 2-21

   3       PROCESS CHEMISTRY AND BIOCHEMISTRY OF
            NITRIFICATION AND DENITRIFICATION                      3-1

           3.1  Introduction                                                3-1
           3.2  Nitrification                                                3-1

                3.2.1  Biochemical Pathways                                 3-1
                3.2.2  Energy and Synthesis Relationships                     3-2
                3.2.3  Alkalinity and pH Relationships                         3-4
                3.2.4  Oxygen Requirements                                 3-6
                3.2.5  Kinetics of Nitrification                                3-6

                      3.2.5.1  Effect of Ammonia Concentration on Kinetics    3-6
                      3.2.5.2  Relationship of Growth Rate to Oxidation Rate  3-7
                      3.2.5.3  Relationship of Growth Rate to Solids
                               Retention Time                              3-8
                      3.2.5.4  Kinetic Rate Constants for Temperature and
                               Nitrogen Concentration                       3-8
                      3.2.5.5  Effect of Dissolved Oxygen on Kinetics          3-12
                      3.2.5.6  Effect of pH on Kinetics                       3-13
                      3.2.5.7  Combined Kinetic Expressions                  3-14
                                       VI

-------
                              CONTENTS - Continued


Chapter  »                                                                   Page

                3.2.6  Population Dynamics                                   3-17
                3.2.7  Nitrification Rates in Activated Sludge                   3-21
                3.2.8  Nitrification Rates in Trickling Filters and Other
                        Attached Growth Systems                             3-27
                3.2.9  Effect of Inhibitors on Nitrification                      3-27

            3.3  Denitrification                                               3-29

                3.3.1  Biochemical Pathways                                  3-29
                3.3.2  Energy and Synthesis Relationships                      3-30
                3.3.3  Alkalinity and pH Relationships                         3-34
                3.3.4  Alternative Electron Donors                            3-34
                3.3.5  Kinetics of Denitrification                              3-36

                       3.3.5.1  Effect of Nitrate on Kinetics                    3-36
                       3.3.5.2  Relationship of Growth Rate to Removal Rate    3-36
                       3.3.5.3  Solids Retention Time                          3-37
                       3.3.5.4  Kinetic Constants for Denitrification            3-37
                       3.3.5.5  Effect of Carbon Concentration on Kinetics      3-40
                       3.3.5.6  Effect of pH on Kinetics                       3-41
                       3.3.5.7  Combined Kinetic Expression                   3-42

                3.3.6  Effect of DO on Denitrification Inhibition                3-43

            3.4  References                                                   3-43

   4        BIOLOGICAL NITRIFICATION                                   4-1

            4.1  Introduction                                                 4-1
            4.2  Classification of Nitrification Processes                         4-1
            4.3  Combined Carbon Oxidation-Nitrification in Suspended
                  Growth Reactors                                            4-2

                4.3.1  Activated Sludge Modifications                          4-2

                       4.3.1.1  Complete Mix Plants                           4-2
                       4.3.1.2  Extended Aeration Plants                      4-4
                       4.3.1.3  Conventional or Plug Flow Plants                4-4
                                        Vll

-------
                              CONTENTS - Continued
Chapter                                                                      Page

                       4.3.1.4  Contact Stabilization Plants                    4-4
                       4.3.1.5  Step Aeration and Sludge Reaeration Plants      4-6
                       4.3.1.6  High Rate and Modified Activated Sludge        4-6
                       4.3.1.7  High Purity Oxygen Activated Sludge Plants      4-6

                 4.3.2  Utility of Nitrification Kinetic Theory in Design          4-7
                 4.3.3  Complete Mix Activated Sludge Kinetics                 4-7

                       4.3.3.1  Effect of Temperature and  Safety Factor on
                                Design                                       4-13
                       4.3.3.2  Consideration in the Selection of SF             4-13

                 4.3.4  Extended Aeration Activated Sludge Kinetics             4-20
                 4.3.5  Conventional Activated Sludge (Plug Flow) Kinetics      4-20

                       4.3.5.1  Considerations in the Selection of the Safety
                               Factor                                        4-23
                       4.3.5.2  Kinetic Design Approach                       4-23

                 4.3.6  Contact Stabilization Activated Sludge Kinetics           4-23

                       4.3.6.1  Design Example                               4-24

                 4.3.7  Step Aeration Activated Sludge Kinetics                 4-30
                 4.3.8  Operating Experience with Combined Carbon  Oxidation-
                        Nitrification in Suspended Growth Reactors             4-30

                       4.3.8.1  Step Aeration Activated Sludge In a Moderate
                                Climate                                      4-30
                       4.3.8.2  Step Aeration Activated Sludge in a Rigorous
                                Climate                                      4-32
                       4.3.8.3  Conventional Activated Sludge In a Rigorous
                                Climate                                      4-32

            4.4  Combined Carbon Oxidation-Nitrification In Attached Growth
                  Reactors                                                   4-35

                 4.4.1  Nitrification with Trickling Filters in Combined Carbon
                        Oxidation-Nitrification Applications                    4-35

                                       viii

-------
                              CONTENTS - Continued
Chapter                                                                      Page

                       4.4.1.1  Media Selection                                4-35
                       4.4.1.2  Organic Loading Criteria                        4-37
                       4.4.1.3  Effect of Media Type on Allowable Organic
                                Loading                                     4-39
                       4.4.1.4  Effect of Recirculation on Nitrification           4-40
                       4.4.1.5  Effect of Temperature on Nitrification           4-40
                       4.4.1.6  Effect of Diurnal Loading on Performance        4-41

                4.4.2  Nitrification with the Rotating Biological Disc Process in
                        Combined Carbon Oxidation-Nitrification Applications   4-41

                       4.4.2.1  Loading Criteria for Nitrification                4-43
                       4.4.2.2  Effect of Temperature                          4-43
                       4.4.2.3  Effect of Diurnal Load Variations               4-43

            4.5  Pretreatment for Separate Stage Nitrification                    4-45

                4.5.1  Effects of Pretreatment by Chemical Addition            4-46

                4.5.2  Effects of Degree of Organic Carbon Removal             4-50
                4.5.3  Protection Against Toxicants                            4-51

            4.6  Separate Stage Nitrification with Suspended Growth Processes    4-52

                4.6.1  Application of Nitrification Kinetic Theory to Design      4-52
                4.6.2  Solids Retention Time Approach                         4-53

                       4.6.2.1  Choice of Process Configuration                 4-53
                       4.6.2.2  Choice of the Safety Factor                     4-54

                4.6.3  Nitrification Rate Approach                            4-55
                4.6.4  Effect of the BODs/TKN Ratio on Sludge Inventory
                        Control                                              4-57
                4.6.5  Comparison of the Use of Conventional Aeration to
                        the Use of High Purity Oxygen                         4-58

                       4.6.5.1  High Purity Oxygen Nitrification With and
                                 Without pH Control                          4-59
                                        IX

-------
                              CONTENTS - Continued


Chapter                                                                     Page

                       4.6.5.2  Comparison of Conventional Aeration and
                                High Purity Oxygen at the Same pH            4-59

            4.7  Separate State Nitrification with Attached Growth Processes      4-61

                4.7.1   Nitrification with Trickling Filters                       4-62

                       4.7.1.1  Media Type and Specific Surface                4-62
                       4.7.1.2  Loading Criteria                               4-62
                       4.7.1.3  Effect of Recirculation                         4-66
                       4.7.1.4  Effluent Clarification                          4-66
                       4.7.1.5  Effect of Diurnal Load Variations               4-66
                       4.7.1.6  Design Example                               4-68

                4.7.2  Nitrification with the Rotating Biological Disc Process     4-69

                       4.7.2.1  Kinetics                                      4-70

                4.7.3  Nitrification with Packed-Bed Reactors                  4-72

                       4.7.3.1  Oxygenation Techniques                       4-73
                       4.7.3.2  Media Type, Backwashing and Loading Criteria   4-73

            4.8  Aeration Requirements                                       4-76

                4.8.1   Adaptability of Alternative Aeration Systems to Diurnal
                        Variations in Load                                    4-79
                4.8.2  Oxygen Transfer Requirements                          4-80
                4.8.3  Example Sizing of Aeration Capacity                    4-84

            4.9  pH Control                                                  4-85

                4.9.1   Chemical Addition and Dose Control                    4-86
                4.9.2  Effect of Aeration Method on Chemical Requirements     4-86

            4.10 Solids-Liquid Separation                                      4-89
            4.11 Considerations for Process  Selection                           4-94

                4.11.1 Comparison to Physical-Chemical Alternatives            4-94

-------
                             CONTENTS - Continued

Chapter                                                                     Page

                4.11.2 Choice Among Alternative Nitrification Systems          4-94

           4.12 References                                                  4-99

   5       BIOLOGICAL DENITRIFICATION                                5-1

           5.1  Introduction                                                5-1
           5.2  Denitrification in Suspended Growth Reactors Using
                 Methanol as the Carbon Source                               5-1

                5.2.1   Denitrification Rates                                   5-3
                5.2.2   Complete Mix Denitrification Kinetics                   5-4

                       5.2.2.1  Effect of Safety Factor on Steady-State
                                Effluent Quality                             5-8
                       5.2.2.2  Effect of Diurnal Load Variations on Effluent
                                Quality                                     5-8

                5.2.3   Plug Flow Denitrification Kinetics                       5-10
                5.2.4   Effluent Quality from Suspended Growth
                        Denitrification Processes                              5-12

                       5.2.4.1  Experience at Manassas, Va.                    5-12
                       5.2.4.2  Experience at the CCCSD's Advanced
                                Treatment Test Facility.                       5-13

           5.3  Denitrification in Attached Growth Reactors Using Methanol
                 as the Carbon  Source                                        5-15

                5.3.1   Kinetic Design of Attached Growth Denitrification
                        Systems                                             5-15
                5.3.2   Classification of Column Configurations                  5-17

                       5.3.2.1  Nitrogen Gas Filled Denitrification Columns -
                                Packed Bed                                  5-17
                       5.3.2.2  Submerged High Porosity Media Columns -
                                Packed Bed                                  5-22
                       5.3.2.3  Submerged Low Porosity Fine Media
                                Columns - Packed Bed Configuration           5-23
                                       XI

-------
                             CONTENTS - Continued
Chapter                                                                     Page

                      5.3.2.4 Submerged High Porosity Fine Media Columns -
                                Fluidized Bed                                5-29
                      5.3.2.5 Comparison of Attached Growth
                                Denitrification Systems                        5-32

            5.4  Methanol Handling, Storage, Feed Control, and Excess
                 Methanol Removal                                          5-32

                5.4.1  Properties of Methanol                                 5-33
                5.4.2  Standards for Shipping, Unloading, Storage and
                        Handling                                            5-33
                5.4.3  Methanol  Delivery and Unloading                        5-34
                5.4.4  Methanol  Storage                                      5-36
                5.4.5  Transfer and Feed                                     5-37
                5.4.6  Methanol  Feed Control                                 5-37
                5.4.7  Excess Methanol Removal                              5-38

            5.5  Combined Carbon Oxidation-Nitrification-Denitrification
                 Systems with Wastewater and Endogenous Carbon Sources       5-39

                5.5.1  Systems Using Endogenous Respiration in a Sequential
                        Carbon Oxidation-Nitrification-Denitrification System    5-39
                5.5.2  Systems Using Wastewater Carbon in Alternating
                        Aerobic/Anoxic Modes                                5-42

                      5.5.2.1 Aerobic/Anoxic Sequences in Oxidation Ditches  5-42
                      5.5.2.2 Denitrification in an Alternating Contact Process  5-48
                      5.5.2.3 The Bardenpho Process                         5-49
                      5.5.2.4 Alternating Aerobic/Anoxic System Without
                                Internal Recycle                              5-52
                      5.5.2.5 Kinetic Design of Alternating Aerobic/Anoxic
                                Systems                                     5-55

            5.6  Solids-Liquid Separation                                      5-58
            5.7  Considerations for Process Selection                           5-60

                5.7.1  Comparison to Physical-Chemical Alternatives            5-61
                5.7.2  Choice Among Alternative Denitrification Systems        5-61

            5.8  References                                                  5-64
                                        xii

-------
                             CONTENTS - Continued


Chapter                                                                     Page

   6        BREAKPOINT CHLORINATION                                  6-1

            6.1  Process Chemistry                                            6-1

                6.1.1   Chemical Stoichiometry                                6-1
                6.1.2   The Breakpoint Curve                                  6-5

            6.2  Process Application Considerations                            6-5

                6.2.1   Chlorine Dosage Requirement                           6-6

                       6.2.1.1  Effect of Pretreatment                         6-6
                       6.2.1.2  Effect of pH and Temperature                  6-7
                       6.2.1.3  Initial Mixing of Chlorine                       6-7

                6.2.2   Residual Nitrogenous Materials                          6-9
                6.2.3   Alkalinity Supplementation                            6-12
                6.2.4   Reaction Rates                                        6-12
                6.2.5   Effect on Total Dissolved Solids                         6-13
                6.2.6   Reactions with Organic Nitrogen                        6-14
                6.2.7   Disinfection                                          6-15

            6.3  Process Control Instrumentation                               6-15

                6.3.1   Process Control System                                 6-15

                       6.3.1.1  Chlorine Dosage Control                       6-15
                       6.3.1.2  pH Control                                   6-18

                6.3.2   Process Control Components                            6-18

            6.4  Dechlorination Techniques                                    6-18

                6.4.1   Sulphur Dioxide Dechlorination                         6-19

                       6.4.1.1  Stoichiometry                                 6-19
                       6A.I.2  Reaction Rates                                6-20
                       6.4.1.3  Significance of Sulphur Dioxide Overdose        6-20
                       6.4.1.4  Process Application and Control                6-20

                                        xiii

-------
                              CONTENTS - Continued


Chapter                                                                      Page

                6.4.2  Activated Carbon Dechlorination                        6-21

                       6.4.2.1  Stoichiometry                                  6-21
                       6.4.2.2  Process Application                            6-22

            6.5  Design Example                                              6-22
            6.6  Considerations for Process Selection                            6-24
            6.7  References                                                   6-25

   7        SELECTIVE ION EXCHANGE FOR AMMONIUM REMOVAL        7-1

            7.1  Chemistry and Engineering Principles                           7-1

                7.1.1  Basic Concept                                          7-1
                7.1.2  Ion Exchange Principles                                 7-2
                7.1.3  Properties of Clinoptilolite                              7-3

                       7.1.3.1  Selectivity                                     7-3
                       7.1.3.2  Mineralogical Classification                      7-5
                       7.1.3.3  Total Exchange Capacity                       7-7
                       7.1.3.4  Chemical Stability                              7-7
                       7.1.3.5  Physical Stability                              7-8
                       7.1.3.6  Density                                       7-9

            7.2  Major Service Cycle Variables                                  7-9

                7.2.1  pH                                                   7-9
                7.2.2  Hydraulic Loading Rate                                 7-9
                7.2.3  Clinoptilolite Size                                      7-9
                7.2.4  Pretreatment                                           7-10
                7.2.5  Wastewater Composition                                7-10
                7.2.6  Length of Service Cycle                                 7-10
                7.2.7  Bed Depth                                             7-11
                7.2.8  One Column vs. Series Column Operation                 7-12
                7.2.9  Determination of Ion Exchanger Size                     7-14

            7.3  Regeneration Alternatives                                     7-16

                7.3.1  Basic Concepts                                         7-16

                                        xiv

-------
                              CONTENTS - Continued


Chapter                                                                     Page

                7.3.2  Regeneration Process                                  7-17

                       7.3.2.1   High pH Regeneration                          7-17
                       7.3.2.2  Neutral pH Regeneration                       7-17
                       7.3.2.3  Effects on Effluent TDS                        7-19

                7.3.3  Regenerant Recovery Systems                          7-19

                       7.3.3.1   Air Stripping of High pH Regenerant            7-19
                       7.3.3.2  Air Stripping of Neutral pH Regenerant          7-22
                       7.3.3.3  Steam Stripping                               7-24
                       7.3.3.4  Electrolytic Treatment of Neutral pH
                                Regenerant                                  7-26

            7.4  Considerations in Process Selection                            7-27
            7.5  References                                                  7-28

   8        AIR STRIPPING FOR NITROGEN REMOVAL                     8-1

            8.1  Chemistry and Engineering Principles                          8-1

                8.1.1  Basic Concept                                         8-1

            8.2  Environmental Considerations                                 8-1

                8.2.1  Air Pollution                                          8-2
                8.2.2  Washout of Ammonia from the Atmosphere              8-4
                8.2.3  Noise                                                8-5

            8.3  Stripping Tower System Design Considerations                  8-5

                8.3.1  Type of Stripping Tower                               8-5
                8.3.2  pH                                                   8-5
,                8.3.3  Temperature                                          8-6
                8.3.4  Hydraulic Loading                                     8-7
                8.3.5  Tower Packing                                        8-9

                       8.3.5.1   Packing Depth                                8-9
                       8.3.5.2  Packing Material and Shape                     8-9

                                       xv

-------
                              CONTENTS - Continued


Chapter                                                                      Page

                       8.3.5.3  Packing Spacing and Configuration               8-10

                8.3.6  Air Flow                                              8-11
                8.3.7  Scale Control                                          8-12

            8.4  Ammonia Recovery or Removal From Off-Gases                 8-13

                8.4.1  Acid Systems                                          8-13
                8.4.2  Nitrification-Denitrification                             8-15

            8.5  Stripping Ponds                                              8-17
            8.6  Considerations in Process Selection                             8-19
            8.7  References                                                   8-20

   9        TOTAL SYSTEM DESIGN                                         9-1

            9.1  Introduction                                                 9-1
            9.2  Influence of Effluent Quality Objectives on Total System
                 Design                                                      9-1
            9.3  Other Considerations in Process Selection                       9-4
            9.4  Interrelationships with Phosphorus Removal                     9-4

                9.4.1  Alternative Systems                                    9-5
                9.4.2  Considerations in System Selection                       9-8

                       9.4.2.1  Phosphorus Removals Obtainable                9-8
                       9.4.2.2  Impacts on Sludge Handling                     9-9
                       9.4.2.3  Reliability                                     9-12
                       9.4.2.4  Flexibility of Operation in Multipurpose
                                Treatment Units                              9-12
                       9.4.2.5  Cost                                          9-12

            9.5  Case Examples                                               9-12

                9.5.1  Case Examples of Nitrification for Ammonia Reduction    9-13

                       9.5.1.1  Jackson, Michigan                              9-13
                       9.5.1.2  Valley Community Services District, California    9-16
                                      xvi

-------
                             CONTENTS - Continued
Chapter                                                                    Page

                      9.5.1.3  Livermore, California                          9-22
                      9.5.1.4  San Pablo Sanitary District, California            9-23

                9.5.2  Case Examples of Nitrification-Denitrification for
                       Nitrogen Removal                                    9-29

                      9.5.2.1  Central Contra Costa Sanitary District, California  9-29
                      9.5.2.2  Canberra, Australia                            9-39
                      9.5.2.3  Washington, D.C.                              9-40
                      9.5.2.4  El Lago, Texas                                9-49

                9.5.3  Case Examples of Breakpoint Chlorination for Nitrogen
                       Removal                                            9-55

                      9.5.3.1  Sacramento, California                         9-55
                      9.5.3.2  Montgomery County, Maryland                 9-62

                9.5.4  Case Examples of Selective Ion Exchange for Nitrogen
                       Removal                                            9-70

                      9.5.4.1  Upper Occoquan Sewage Authority, Va.         9-70
                      9.5.4.2  Rosemount, Minnesota                        9-77

                9.5.5  Case Examples of Air Stripping for Nitrogen Removal       9-81

                      9.5.5.1  South Lake Tahoe, California                   9-81
                      9.5.5.2  Orange County Water District, California         9-87

           9.6  References                                                  9-91

APPENDIX A - GLOSSARY OF SYMBOLS                                    A-1

APPENDIX B - METRIC EQUIVALENTS                                     B-l
                                      xvu

-------
                                 LIST OF FIGURES
Figure No.                                                                     Page

  2-1    The Nitrogen Cycle                                                   2-2
  2-2    The Nitrogen Cycle in Surface Water                                    2-6
  2-3    The Nitrogen Cycle in Soil and Groundwater                             2-7
  2-4    Allowable Effluent Discharge into the Thames Estuary                    2-15
  3-1    Temperature Dependence of the Maximum Growth Rates of Nitrifiers      3-9
  3-2    Temperature Dependence of the Half Saturation Constants of Nitrifiers     3-9
  3-3    Comparison of Effect of Temperature on Nitrification in Suspended
          Growth and Attached Growth Systems                                 3-11
  3-4    Effect of Dissolved Oxygen on Nitrification Rate                         3-13
  3-5    Effect of pH on Nitrification Rate                                       3-15
  3-6    Effect of Solids Retention Time on Effluent Ammonia Concentration
          and Nitrification Efficiency                                           3-20
  3-7    Effect of Ammonia Concentration on Nitrification Rate                   3-24
  3-8    Effect of Temperature and Fraction of Nitrifiers on Nitrification Rate      3-25
  3-9    Effect of BODs/TKN Ratio on Nitrification Rate - Experimental
          Attached Growth System                             •                3-26
  3-10   Effect of Temperature on Denitrification Rate                           3-39
  3-11   Effect of pH on Denitrification Rate                                    3-41
  4-1    Modifications of the Activated Sludge Process                            4-5
  4-2    Effect of the Safety Factor on Steady State Effluent Ammonia Levels
          in Suspended Growth Systems                                         4-16
  4-3    Diurnal Variations at the Chapel Hill, N.C. Treatment Plant                4-17
  4-4    Effect of SF on Diurnal Variation in Effluent  Ammonia                   4-19
  4-5    DO and Ammonia Nitrogen  Profile in a Plug-Flow System                 4-21
  4-6    Nitrification Efficiency as a  Function of Process Parameters               4-25
  4-7    Effect of Organic Load on Nitrification Efficiency of Rock Trickling
          Filters                                                              4-38
  4-8    A Typical Rotating Biological Disc Process                              4-42
  4-9    Effect of BOD5 Concentration and Hydraulic Load on Nitrification
          in the RBD Process                                                  4-44
  4-10   Temperature Correction Factor for Nitrification in the RBD Process        4-45
  4-11   Pretreatment Alternatives for Separate Stage Nitrification                 4-46
  4-12   Rancho Cordova Wastewater Treatment Facility Effluent Ammonia
          Characteristics, March 19-20, 1974                                    4-54
  4-13   Observed Nitrification Rates at Various Locations                        4-56
  4-14   Covered High Purity Oxygen Reactor with Three Stages and
          Mechanical Aerators                                                 4-58
  4-15   Surface Area Requirements for Nitrification - Midland Michigan            4-64
                                       XVlll

-------
                         LIST OF FIGURES - Continued
Figure No.                                                                    Page

  4-16   Surface Area Requirements for Nitrification - Lima, Ohio                 4-65
  4-17   Surface Area Requirements for Nitrification - Sunnyvale, California        4-67
  4-18   Nitrification Rates as a Function of Stage Effluent Concentration          4-70
  4-19   Design Relationships for a 4-Stage RED Process Treating Secondary
          Effluent                                                           4-71
  4-20   Schematic Diagram of a Packed-Bed Reactor                            4-72
  4-21   Temperature Dependence of Detention Time for Complete
          Nitrification, (<2 mg/1 NH|-N) at Steady State in the PER               4-74
  4-22   Relation  Between Ammonia Peaking and Hydraulic Peaking Loads for
          Treatment Plants with No In-Process Equalization                      4-77
  4-23   Relationship of Maximum/Minimum Nitrogen Load Ratio to
          Maximum/Average Flows                                            4-78
  4-24   Relationship of Aeration Air Requirements for Oxidation of
          Carbonaceous BOD and Nitrogen                                      4-83
  4-25   Effect of Temperature on Thickening Properties of Oxygen Activated
          Sludge at MLSS = 4000 mg/1                                         4-91
  4-26   Effect of Temperature on Thickening Properties of Oxygen Activated
          Sludge at MLSS = 7000 mg/1                                         4-92
  5-1    Suspended Growth Denitrification Systems Using Methanol               5-2
  5-2    Observed Denitrification Rates for Suspended  Growth Systems
          Using Methanol                                                     5-3
  5-3    Effect of Safety Factor  on Effluent Nitrate Level in Suspended Growth
          System                                                             5-9
  5-4    Effect of Diurnal Variation in Load on Effluent Nitrate Level in
          Complete Mix Suspended Growth System    .                          5-11
  5-5    The Three Sludge System as Tested at  Manassas, Va                      5-12
  5-6    ATTF System for Nitrogen and Phosphorus Removal                     5-14
  5-7    Design Details of Nitrogen Gas Filled Denitrification Column              5-19
  5-8    Typical Process Schematic for Submerged High Porosity Media Columns    5-22
  5-9    Surface Denitrification Rate for Submerged High Porosity Media Columns  5-24
  5-10   Nitrification-Denitrification Flow Sheet Utilizing Low Porosity Fine
          Media in Columns                                                   5-25
  5-11   Column Depth vs. Specific Surface Area                                 5-27
  5-12   Fluidized Bed Denitrification System                                   5-29
  5-13   Volume Denitrification  Rate for Submerged High Porosity Fine Media
          Columns                                                           5-31
  5-14   Feedforward Control of Methanol Based on Flow and Nitrate Nitrogen      5-38
  5-15   Sequential Carbon Oxidation-Nitrification-Denitrification                 5-40
  5-16   Denitrification Rates Using Endogenous Carbon Sources                  5-41

                                      xix

-------
                          LIST OF FIGURES - Continued
Figure No.                                                                    Page

  5-17   Pasveer Ditch or Endless Channel System for Nitrogen Removal            5-43
  5-18   Vienna-Blumenthal Wastewater Treatment Plant                          5-44
  5-19   Alternating Contact Process                                            5-49
  5-20   Operational Sequencing of One of Two.Aeration Tanks in Alternating
          Contact Process                                                     5-50
  5-21   The Bardenpho System - Sequential Utilization of Wastewater Carbon
          and Endogenous Carbon                                              5-51
  5-22   Blue Plains Alternating Anoxic Aerobic System                           5-52
  5-23   Effect of Temperature on Peak Denitrification Rates with Wastewater
          as Carbon Source                                                    5-57
  5-24   Comparison of Denitrification Systems                                 5-63
  6-1    Relative Amounts of HOC1 and OC1~ at Various pH Levels                6-2
  6-2    Effects of pH and Temperature on Distribution of Ammonia and
          Ammonium Ion in Water                                             6-3
  6-3    Theoretical Breakpoint Chlorination Curve                              6-6
  6-4    Effect of C^NH^-N on Nitrogen Residuals in Lime Clarified
          Filtered Secondary Effluent                                          6-11
  6-5    Comparison of Germicidal Efficiency of Hypochlorous Acid,
          Hypochlorite Ion, and Monochloramine for 99 Percent Destruction
          of E. Coli at 2-6 C           „                                        6-16
  6-6    Breakpoint Chlorination Control - Functional Schematic                  6-17
  7-1    Selective Ion Exchange Process                                        7-2
  7-2    Generalized Ion Exchange Isotherms                                    7-4
  7-3    The 23 C Isotherms for the Reaction, (Ca)z + 2(NHf )N = 2(NH|)Z. +
         (Ca)N with Hector Clinoptilolite and IR 120                            7-5
  7-4    Selectivity Coefficients vs. Concentration Ratios of Sodium or
          Potassium and Ammonium in the Equilibrium Solution with Hector
          Clinoptilolite at 23 C for the Reaction (Y)z + (NH^ = (NH^ + (Y)N  7.6
  7-5    Selectivity Coefficients vs. Concentration Ratios of Calcium or Magnesium
          and Ammonium in the Equilibrium Solution with Hector Clinoptilolite
          at 23 C for the Reaction (X)z + 2(NHj)N = 2(NHj)z + (X)N            7-7
  7-6    Isotherms for Exchange of NH^ for K+, Na+, Ca"*"1", and Nig*"1" on
          Clinoptilolite                                                       7-8
  7-7    Minimum Bed Volumes as a Function of Influent  NH^-N Concentration
          to Reach 50 Percent Breakthrough of Ammonium                      7-12
  7-8    Ammonium Breakthrough Curves for a 6 ft Clinoptilolite Bed at
          Various Flow Rates                   ,                              7-13
  7-9    Effect of Bed Depth on Ammonium Breakthrough at 9.7 BV/hr            7-14
                                        xx

-------
                         LIST OF FIGURES - Continued
Figure No.                                                                    Page

  7-10   Variation of Ammonium Exchange Capacity with Competing Cation
          Concentration for a 3 ft Deep Clinoptilolite Bed                        7-15
  7-11   Ammonium Elution with 2 Percent Sodium Chloride Regenerant           7-18
  7-12   Example Ion Exchange - Air Stripping System for High pH Regenerant      7-20
  7-13   Flow Diagram of Neutral pH Regeneration System Using Air Stripping      7-23
  7-14   Typical Elution Curve                                                 7-25
  7-15   Simplified Flow Diagram of Electrolytic Regenerant Treatment
          System                                                             7-26
  8-1    Ammonia Stripping Process                                           8-2
  8-2    Types of Stripping Towers                                             8-6
  8-3    Effect of Temperature on Ammonia Removal Efficiency Observed
          at Blue Plains Pilot Plant                                             8-7,
  8-4    Percent Ammonia Removal vs. Surface Loading Rate for Various
          Depths of Packing                                                   8-8
  8-5    Illustrative  Packing Configuration                                      $-10
  8-6    Effect of Packing Space on Air Requirements and Efficiency of
          Ammonia Stripping   ,                                              8-11
  8-7    Process for Ammonia Removal and Recovery                            8-14
  8-8    Ammonia Elimination System                                         8-16
  8-9    Ammonia Stripping Pond System                                      8-|8
  9-1    Alternate Process Sequencing for Systems Yielding Combined Nitrogen
          and Phosphorus Removal - Systems with Coagulant Addition to
          Primary Sedimentation                                              9-6
  9-2    Alternative Process Sequencing for Systems Yielding Combined
          Nitrogen and Phosphorus  Removal - Systems with Coagulant Addition
          after Primary Treatment                                             9-7
  9-3    Schematic Flow Diagram - South Lake Tahoe, California Plant             9-10
  9-4    Jackson, Michigan Waste water Treatment Plant Flow Diagram             9-15
  9-5    Valley  Community Services District (Calif.) Wastewater Treatment
          Plant  Flow Diagram                                                 9-19
  9-6    Holding Basin at the Valley Community Services District (California)
          Wastewater Treatment Plant                                          9-21
  9-7    City of Livermore Water Reclamation Plant (Calif.) Flow Diagram         9-22
  9-8    Aeration Tank at the Livermore Water Reclamation Plant (California)
          with Roughing Trickling Filters in Background                          9-25
  9-9    San Pablo Sanitary District Treatment Plant (California) Flow Diagram     9-27
  9-10   Liquid Process Flow Sheet - CCCSD Water Reclamation Plant              ft-33
                                       xxi

-------
                          LIST OF FIGURES - Continued
Figure No.                                                                    Page

  9-11   Nitrification-Denitrification System at the CCCSD Water Reclamation
          Plant                                                              9-36
  9-12   Solids Flow Diagram at the CCCSD Water Reclamation Plant              9-38
  9-13   Process Flow Diagram for the Lower Molonglo Water Quality
          Control Centre (Canberra, Australia)                                   9-41
  9-14   Section Through Nitrification Tanks at the LMWQCC, Canberra,
         Australia                                                            9-44
  9-15   Washington, D.C. Blue Plains Treatment Plant Flow Diagram of
          Primary and Secondary  Systems                                       9-46
  9-16   Washington, D.C. Blue Plains Treatment Plant Flow Diagram of
          Nitrification and Denitrification Systems                               9-47
  9-17   Washington, D.C. Blue Plains Treatment Plant Flow Diagram of
          Filtration and Disinfection Systems                                    9-48
  9-18   El Lago, Texas Wastewater Treatment Plant, Flow Diagram                9-51
  9-19   El Lago, Texas Denitrification Columns, Coarse Media Type on Right
          and Fine Media Type on Left                                         9-53
  9-20   Hypochlorite Generation Schematic - Sacramento Regional
          Wastewater Treatment Plant                                          9-59
  9-21   Plan and  Section of the Breakpoint Facility at the Sacramento
          Regional Wastewater Treatment Plant                                  9-60
  9-22   Flow Diagram of the Montgomery County, Maryland Plant                9-63
  9-23   Membrane Cell Used for Hypochlorite Production                        9-64
  9-24   Overall System Using Membrane Cells for Hypochlorite Production         9-65
  9-25   Schematic of Montgomery County, Maryland Breakpoint
          Chlorination Process                                                 9-67
  9-26   Flow Diagram - Upper Occoquan Sewage Authority Plant (Virginia)        9-71
  9-27   Plan and  Section of Ion Exchange Beds at Upper Occoquan Plant          9-73
  9-28   Added Details - Ion Exchange Beds at Upper Occoquan Plant              9-74
  9-29   Plan View of ARRP Module - Upper Occoquan Plant                      9-75
  9-30   Section of ARRP Module - Upper Occoquan Plant                        9-76
  9-31   Schematic of Rosemount, Minnesota Plant                              9-79
  9-32   Orange Co. Ammonia Stripping/Cooling Tower Section                   9-88
  9-33   Overall View of the Orange County Water District                        9-89
  9-34   Stripping Tower Packing  Module at the Orange County Water
          District Plant                                                       9-90
                                       xxii

-------
                                 LIST OF TABLES
Table No.                                                                       Page

  2-1    Estimated Nitrogen Loadings for the San Francisco Bay Basin              2-11
  2-2    Effect of Ammonium Removal on Total Oxygen Demand of
          Wastewater Treatment Plant Effluent                                    2-14
  2-3    Effect of Various Treatment Processes on Nitrogen Compounds             2-21
  3-1    Relationships for Oxidation and Growth in Nitrification Reactions
          in Relationship to the Carbonic Acid System                             3-4
  3-2    Alkalinity Destruction Ratios in Experimental Studies                     3-5
  3-3    Maximum Growth Rates for Nitrifiers in Various Environments             3-10
  3-4    Half-Saturation Constants for Nitrifiers in Various Environments            3-10
  3-5    Relationship between Nitrifier Fraction and the BOD5/TKN Ratio          3-23
  3-6    Compounds Toxic to Nitrifiers                                          3-28
  3-7    Comparison of Energy Yields of Nitrate Dissimilation vs. Oxygen
          Respiration for Glucose                                                3-30
  3-8    Relationships for Nitrate Dissimilation and Growth in Denitrification
          Reactions                                                             3-32
  3-9    Combined Dissimilation-Synthesis Equations for Denitrification             3-33
  3-10   Values of Denitrification Yield and Decay Coefficients for Various
          Investigations Using Methanol                                          3-38
  4-1    Classification of Nitrification Facilities                                    4-3
  4-2    Calculated Design Parameters for a 1 mgd Complete Mix Activated
          Sludge Plant                                                           4-14
  4-3    Design Data Whittier Narrows Water Reclamation Plant                    4-31
  4-4    Nitrification Performance  at the Whittier Narrows Water Reclamation
          Plant                                                                 4-33
  4-5    Average Nitrification Performance at Flint, Michigan for 8 Months          4-34
  4-6    Effect of Temperature and Solids Retention Time on Nitrification
          Efficiency  at Flint, Michigan                                           4-34
  4-7    Nitrification Performance  at the Jackson, Michigan Wastewater
          Treatment Plant                                                       4-36
  4-8    Comparative Physical Properties of Trickling Filter Media                 4-37
  4-9    Organic Nitrogen Reductions in Nitrifying Trickling Filters                 4-39
  4-10   Loading Criteria for Nitrification with Plastic Media at Stockton            4-40
  4-11   Effect of Recirculation on Nitrification in Rock Trickling Filters at
          Salford, England                                                      4-41
  4-12   Effect of Alum Addition to Wastewater on Alkalinity                      4-47
  4-13   Comparison of Process Characteristics for Oxygen Nitrification
          Systems with and without pH Control at Blue Plains, Washington, D.C.     4-60
  4-14   Comparison of Process Characteristics of Conventionally Aerated  and
          High Purity Oxygen Systems with pH Control at Blue Plains,
          Washington, D.C.                                                      4-61
                                        xxiii

-------
                            LIST OF TABLES - Continued
Table No.                                                                      Page

  4-15   Commercial Types of Plastic Media for Separate Stage Nitrification
          Applications                                                         4-63
  4-16   Nitrification in Separate Stage Rock Trickling Filters                     4-66
  4-17   Packed Bed Reactor Performance When Treating Secondary
     ;     Effluents                                                            4-75
  4-18   Peaking Factors Versus Frequency of Occurrence for Primary
          Treatment Plant Effluent                                             4-79
  4-19   Relation of Oxygen Transfer Efficiency to Aerator Power Efficiency       4-83
  4-20   Air Requirements for Nitrification Activated Sludge Plants                4-84
  4-21   Effect of Oxygen Transfer Efficiency and Residual Alkalinity on
          Operating pH                                                        4-88
  4-22   Comparison of Nitrification Alternatives                                 4-95
  5-1    Denitrification Performance: Final Four Months of Operation at
          Manassas, Virginia                                                    5-13
  5-2    ATTF Performance Summary, April 16 to July 15, 1972                  5-16
  5-3    Types of Denitrification Columns and Measured Denitrification Rates      5-18
  5-4    Summary of Operation - Nitrogen Gas Filled Denitrification Column       5-21
  5-5    Neptune-Microfloc Media Designs for Denitrification                     5-26
  5-6    Comparison of Suspended Solids Removal Efficiency for Submerged
          Fine Media Denitrification Columns                                    5-28
  5-7    Properties of Methanol                                                5-33
  5-8    Pilot Tests of Wuhrman's Sequential Carbon Oxidation-Nitrification
          Denitrification System                                               5-40
  5-9    Design Data for the Vienna-Blumenthal Treatment Plant                  5-45
  5-10   Operation and Performance of the Vienna-Blumenthal Plant
          24-Hour Investigations                                               5-46
  5-11   Operation and Performance of Oxidation Ditch Operated for
          Nitrogen Removal in South Africa                                     5-47
  5-12   Performance of the "Bardenpho" Process at Pretoria, South Africa         5-51
  5-13   Summary of Operation and Performance of the Blue Plains Alternating
          Aerobic/Anoxic System                                              5-54
  5-14   Observed Nitrification and Denitrification Rates for Blue Plains
          Alternating Anoxic/Aerobic System                                    5-55
  5-15   Effect of Stabilization Tank on Denitrified Effluent at the Central
          Contra Costa Sanitary District's Advanced Treatment Test Facility        5-59
  5-16   Denitrification Process Parameters at the Central Contra Costa
          Sanitary District's Advanced Treatment Test Facility                    5-59
  5-17   Comparison of Denitrification Alternatives                              5-62
                                       XXIV

-------
                           LIST OF TABLES - Continued
Table No.                                                                      Page

  6-1    Effect of Pretreatment on C^NH^-N Breakpoint Ratio                   6-8
  6-2    Effect of Pretreatment on Formation of Nitrogenous Residuals at
          Breakpoint                                                          6-10
  6-3    Effects of Chemical Addition on Total Dissolved Solids in Breakpoint
          Chlorination                                                        6-13
  6-4    Effect of Breakpoint Chlorination on Soluble Organic Nitrogen            6-14
  7-1    Influent Composition for Selective Ion Exchange Pilot Tests at
          Different Locales                                                    7-11
  9-1    Effluent Nitrogen Concentrations in Treatment Systems Incorporating
          Nitrification-Denitrification                                           9-2
  9-2    Effluent Nitrogen Concentrations in Treatment Systems Incorporating
          Ion Exchange                                                       9-2
  9-3    Effluent Nitrogen Concentrations in Treatment Systems
          Incorporating Breakpoint Chlorination                                 9-3
  9-4    Effluent Phosphorus Concentration from Alternative Systems              9-11
  9-5    Design Data, Jackson, Michigan Wastewater Treatment Plant               9-14
  9-6    VCSD Wastewater Treatment Plant Design Data                          9-17
  9-7    Nitrification Performance at Valley Community Services District
          Wastewater Treatment Plant, California                                9-20
  9-8    Nitrogen Analyses on 24 Hour Composite Effluent Samples at the
          Valley Community Services District Treatment Plant                     9-21
  9-9    Design Data - Livermore Water Reclamation Plant                        9-24
  9-10   Nitrification Performance at the Livermore Water Reclamation Plant       9-26
  9-11   Design Data, San Pablo Sanitary District Treatment Plant                  9-28
  9-12   Nitrification Performance at the San Pablo Sanitary District
          Treatment Plant                                                     9-30
  9-13   Average Process Loading Conditions at the San Pablo Sanitary
          District Treatment Plant During Special Test, May 19th to July 8th,
          1974                                                               9-31
  9-14   Performance Summary for the San Pablo Sanitary District Treatment
          Plant During Special Testing, May 19th to July 8th, 1974                9-31
  9-15   Central Contra Costa Sanitary District Water Reclamation  Plant -
          Design Data                                                         9-34
  9-16   Lower Molonglo Water Quality Control Centre, Design Data                9-42
  9-17   Anticipated Performance Data and Effluent Standards - Blue
          Plains Plant                                                          9-45
  9-18   Design Data, El Lago, Texas Wastewater Treatment Plant                  9-52
  9-19   Initial Performance of Fine Media Denitrification Columns at
          El Lago, Texas - June 4 to July 6, 1973                                 9-54

                                        xxv

-------
                           LIST OF TABLES - Continued
Table No.                                                                    Page

  9-20   Initial Performance of Coarse Media Denitrification Columns - at
          El Lago, Texas - July 8 to August 31,1973                             9-54
  9-21   Subsequent Performance of Fine Media Denitrification Columns at
          El Lago, Texas - October 1 through October 31, 1974                   9-55
  9-22   Design Criteria for Hypochlorite Production Facility Sacramento
          Regional Wastewater Treatment Plant                                 9-57
  9-23   Capital Cost Breakdown for Breakpoint Chlorination at the
          Sacramento Regional Wastewater Treatment Plant                      9-61
  9-24   Total Annual Cost Breakdown for Breakpoint Chlorination at the
          Sacramento Regional Wastewater Treatment Plant                      9-61
  9-25   Design Criteria for Hypochlorite Production Facility at the
          Montgomery County Facility                                         9-66
  9-26   Breakpoint Chlorination Design Criteria for the Montgomery County
          Facility                                                           9-68
  9-27   Estimated Costs of Breakpoint Chlorination at the Montgomery
          County Plant                                                      9-69
  9-28   Design Criteria Selective Ion Exchange Process for Ammonium
          Removal at the Upper Occoquan Plant                                 9-72
  9-29   Regeneration and Regenerant Recovery System Design Criteria at
          the Upper Occoquan Plant                                           9-77
  9-30   Estimated Costs of Selective Ion Exchange at the  Upper Occoquan
          Plant                                                              9-78
  9-31   Rosemount Ion Exchange Design Criteria                               9-80
  9-32   Design Data, Ammonia Stripping Tower at South Lake Tahoe,
          California                                                          9-82
  9-33   Operating Costs for Ammonia Stripping for Continuous Operation of
          Tahoe Air Stripping Tower at 7.5 mgd                                 9-84
  9-34   Design Data and Estimated Nitrogen Removals for All-Weather
          Ammonia  Stripping at South Tahoe, California                         9-85
                                      xxvi

-------
                                    FOREWORD
The formation of the United States Environmental Protection Agency marked a new era of
environmental  awareness in  America. This  agency's  goals  are  national  in scope and
encompass broad  responsibility  in  the area of  air and  water pollution,  solid wastes,
pesticides, and radiation. A vital part of EPA's national water pollution control effort is the
constant development and dissemination of new technology for waste water treatment.

It  is now clear that only the most effective design and operation of wastewater treatment
facilities, using the latest available techniques, will be adequate to meet the future water
quality objectives and to ensure protection of the nation's waters. It is essential that this
new technology be incorporated into the contemporary design of waste treatment facilities
to achieve maximum benefit of our pollution control expenditures.

The purpose of this manual is to provide the engineering community and related industry
with a new source of information to  be  used in  the planning and  design of present and
future  wastewater treatment facilities. It  is recognized  that there are a number of design
manuals and manuals of standard practice, such as those published by the Water Pollution
Control Federation, available in  the field that  adequately describe and interpret current
engineering practices as  related to traditional plant design. It is the intent of this manual to
supplement  this existing body of knowledge by  describing new treatment methods, and by
discussing  the  application  of new  techniques for more effectively  removing  a broad
spectrum of contaminants from wastewater.

Much of the information presented is based  on the evaluation  and operation of pilot,
demonstration, and  full-scale plants. The design criteria thus generated represent typical
values. These values should  be  used  as a  guide and  should  be tempered with sound
engineering judgment based on a complete analysis of the specific application.

This manual is  one of several available through the EPA Office of Technology Transfer to
describe recent technological advances  and new information. Future editions will be issued
as warranted by  advancing  state-of-the-art  to include new  information as it  becomes
available,  and to revise design criteria as additional full-scale operational information is
generated.
                                        xxvii

-------
                                    CHAPTER 1

                                  INTRODUCTION
1.1 Background and Purpose

Man's influence  on the environment is receiving increasing public and scientific attention.
The quality  of some of  the  nation's water  bodies has been subjected  to  continuing
degradation as a result of man's activities. While there has been considerable  success in
reversing this trend,  one  roadblock to greater progress often  has  been the lack of the
necessary technology to reliably  and  economically remove the  pollutants  which are the
cause of degradation  of receiving waters. While conventional technology is well  developed
for removing organics  from wastewater,  the  processes  for  the control of nitrogen in
wastewater effluents have been developed only recently.

The beginnings of the implementation  of nitrification on a significant scale occurred in the
U.S. as late as the 1960's. The practice of nitrification was widespread in England much
earlier. The first implementation of full nitrogen removal was as late as 1969 at South Lake
Tahoe  in California and  even  this installation encountered many problems. A flurry of
research and development activity on  the various nitrogen control methods occurred  very
recently beginning in the late  1960's  and  continues  to date.  Recent legislation and state
regulatory activities have spurred many localities into nitrogen control projects.

Nitrogen control techniques are divided  into  two broad  categories.  The first group of
nitrogen control processes is involved with the conversion of organic and ammonia nitrogen
to nitrate  nitrogen, a less objectionable  form. These  processes are termed nitrification
processes. The second category involves processes which result in the removal of nitrogen
from the wastewater,  not just merely the conversion of nitrogen from one form to another
form in the wastewater. This latter group includes biological nitrification-denitrification, ion
exchange, ammonia stripping and breakpoint chlorination.

The purpose of this manual is the dissemination  of the available data on the nitrogen control
techniques developed to  date.  Further, this manual is not simply an assembly of data,
rather, data from  a variety of  sources has been  scrutinized and reasonable design criteria
drawn  on the basis of all  available sources. Where design procedures come directly from a
single investigator, appropriate reference is made to the work.

This manual could not have been prepared five years ago because of the state of nitrogen
control technology at that time.  It may well be that continuing research will require an
update of this manual in the future. Nonetheless, the body of knowledge on nitrogen control
techniques is now well developed and municipalities and local agencies have a firm basis upon
which  to  plan  those  wastewater treatment  facilities  which  require  nitrogen  control
techniques.

                                         1-1

-------
1.2 Scope of the Manual

This manual presents theoretical and process design information on a number of nitrogen
control processes. While all of the possible nitrogen removal processes are discussed, details
are presented only on those general methods which are most technically and economically
feasible, as evidenced by their actual or planned full-scale application. One exception to this
is  nitrogen  control in oxidation ponds;  material on nitrogen control in oxidation pond
systems was not included because of the paucity of generally applicable design information.
Another exception is land treatment; nitrogen removal by land treatment systems is beyond
the scope of this manual.

The information in  this  manual was developed  from  the  following sources:  (1)  the
experience of the individuals involved in the preparation of the manual,  (2)  the EPA
research, development and demonstration program,  (3) the literature,  (4) from progress
reports on on-going projects,  (5) from private communication with investigators active in
the field, and (6) from operating personnel at  existing wastewater treatment plants.

1.3 Guide to the User

A  perusal of  the table of contents  will give the reader a  fairly complete picture of the
subject matter contained  in this manual. The following chapter-by-chapter description is
oriented toward providing a general description of the contents of each chapter.

Chapter 2,  Nitrogenous  Materials in the Environment and the Need  for  Control in
Wastewater Effluents, describes the sources of nitrogen compounds entering  water bodies,
the nitrogen transformations which  take place  in the  environment, and the effects of
nitrogen compounds as pollutants. Also given in Chapter 2 is a general introduction into the
various types of nitrogen control methods and their applicability to the individual chemical
forms of nitrogen. Chapter 2 is useful for establishing the rationale for nitrogen removal.

Chapter 3, Process Chemistry and Biochemistry of Biological Nitrification and Denitrifica-
tion, is a presentation of the basic factors affecting the growth of nitrifying and denitrifying
organisms.  With  an understanding of these  factors  on a  fundamental  level, the  design
concepts evolved in Chapters 4 and 5  can be better appreciated. However, should the  reader
decide not to involve himself in basic theory, Chapters 4 and 5 are  designed  to stand by
themselves without requiring reference to Chapter 3 except when detailed explanations of
individual points are required.

Chapter 4, Biological Nitrification, presents design criteria for a wide variety of nitrification
processes. Since it has been anticipated  that the greatest number of manual  users will be
concerned with  ammonia oxidation,  as opposed to nitrogen removal, Chapter 4 presents
more  material than any  other chapter. Both combined carbon oxidation-nitrification  and
separate stage nitrification systems are described with details, whether  given on attached
growth or suspended  growth processes.  The alternative methods for pretreatment for

                                         1-2

-------
organic carbon removal prior to separate stage nitrification systems are presented. Sections
are included on aeration, pH control, and solids-liquid separation.

Chapter 5, Biological Denitrification,  completes the sequence of the three chapters on the
biological approach to  nitrogen removal. Design information is provided for both attached
growth and suspended  growth denitrification systems. For those systems using methanol as
the carbon source  for denitrification,  a  section is  included  describing the methods for
chemical handling. The increasingly popular systems using wastewater carbon sources are
described  in detail. Chapter 5  concludes  with a section on solids-liquid separation and a
qualitative comparison  of the alternative denitrification techniques.

Chapter 6, Breakpoint Chlorination,  is the first  of a  set of three chapters  on  physical-
chemical techniques  for nitrogen removal. Basic process chemistry is presented along with a
host of process design considerations for breakpoint chlorination applications. Because of
the importance of process control, details of methods are given. Information is presented on
the removal of toxic  chlorine residuals.

Chapter 7, Selective  Ion Exchange for Ammonium Removal, is a presentation of the design
concepts  involved in the use  of  clinoptilolite,  a natural zeolite exchange  material,  for
ammonium removal  from  wastewater. Ion exchange fundamentals are discussed along with
clinoptilolite   properties.  Process  loading and regeneration  relationships  are presented.
Alternative methods  of regenerant recovery are described.

Chapter 8, Air Stripping for Ammonia Nitrogen Removal,  describes  the  application of
ammonia  stripping to  wastewater  treatment. The  air pollution aspects  of the method are
discussed  and  general conclusions  drawn. The major factors affecting design and process
performance are  described. The problem of equipment  scaling and its control  is given
detailed consideration. Methods of removing ammonia and controlling the  carbon dioxide
levels in the stripping air are described.

Chapter 9, Total System Design, describes the concepts involved in assembling various unit
processes into  rational  treatment trains that can accomplish not only nitrogen removal, but
organics removal and phosphorus removal  (where it is required). The main thrust of Chapter
9 is to present actual examples of treatment systems that incorporate the nitrogen control
processes described in the previous chapters  of this manual. Design concepts that evolved to
suit local circumstances are given emphasis.
                                         1-3

-------
                                   CHAPTER 2

          NITROGENOUS MATERIALS IN THE ENVIRONMENT AND THE
               NEED FOR CONTROL IN WASTEWATER EFFLUENTS
2.1 Introduction

Various compounds containing the element nitrogen are becoming increasingly important in
wastewater management programs because of the many effects that nitrogenous materials in
wastewater effluent can have on the environment. Nitrogen, in its various forms, can deplete
dissolved oxygen levels in  receiving  waters, stimulate aquatic  growth, exhibit toxicity
toward aquatic life, affect chlorine disinfection efficiency, present a public health hazard,
and affect the suitability of wastewater for reuse. Biological  and chemical processes which
occur in wastewater  treatment plants and  in the natural  environment can change  the
chemical form in which nitrogen exists. Such change may eliminate one deleterious effect of
nitrogen while producing, or leaving unchanged, another effect. For example, by converting
ammonia in raw wastewater  to nitrate, the oxygen-depleting and toxic effects of ammonia
are eliminated, but the biostimulatory effects may not be changed significantly.

It is important, therefore, prior to the detailed discussions  of nitrogen  removal processes
which form the principal content of this manual, to review the chemistry of nitrogen and
the effects that the various  compounds can  have. Several specific aspects are discussed in
this  chapter.  First, the nitrogen  cycle  for both surface water  and soil/groundwater
environments  is described,  with emphasis on the important compounds and reactions
associated  with  each.  Second, sources of nitrogen, both  natural and man-caused,  are
discussed. Important  elements of the latter category include domestic  and industrial
wastewater, urban  and suburban runoff, surface and  subsurface agricultural drainage, and
emissions to the atmosphere which may eventually enter the aquatic environment through
precipitation or dustfall.  Then, the  effects  of nitrogen  discharge  to   surface  water,
groundwater, and land are summarized. And finally, introductory to the following chapters,
a brief discussion is presented on the relationship between the various nitrogen compounds
and process removal efficiency.

2.2. The Nitrogen Cycle

Nitrogen exists in many compounds because of the high number of oxidation states it can
assume. In ammonia or organic  compounds, the form most closely associated with plants
and animals, its oxidation state is minus 3. At the other extreme its oxidation state is plus 5
when in the nitrate  form. In the environment, changes from one oxidation state to another
can be brought about biologically by living organisms. The relationship between the various
compounds and the transformations which can occur are often  presented schematically in a
diagram known as the nitrogen cycle. Figure 2-1 shows a common manner of presentation J
The atmosphere serves as a reservoir of N2 gas from which nitrogen is removed naturally by

                                        2-1

-------
electrical discharge and nitrogen-fixing organisms and artificially by chemical manufactur-
ing. Nitrogen gas is returned to the atmosphere by the action of denitrifying organisms. In
the fixed state, nitrogen can undergo the various reactions shown. A general description of
the nitrogen cycle is presented here, and aspects of particular importance to surface water
and soil/groundwater environments are discussed in the following sections.
                                   FIGURE 2-1
                 THE NITROGEN CYCLE (AFTER REFERENCE 1)
                FECAL
               MATTER
               ORGANIC
                 N
   ANIMAL

   PROTEIN
   ORGANIC
      N
               PLANT
              PROTEIN

              ORGANIC
                 N
                                       2-2

-------
Transformation  reactions of importance include fixation, ammonification, assimilation,
nitrification  and  denitrification.2  These  reactions can  be  carried out  by  particular
microorganisms with either a net gain or loss of energy ; energy considerations often play an
important  role in determining the reaction which  occurs. The principal compounds of
concern in the nitrogen  cycle are  nitrogen gas, ammonium, organic nitrogen, and nitrate.
These compounds and their oxidation states are shown below:

                              -3         0       +3     +5
                           NH, / NH+ - N~  - NO! - NO!
                              j . *    T"     2^       2t      j
                            Organic
                           Derivatives

It is important to note that at neutral pH values  there is very little molecular ammonia
       in wastewater as most is in  the form of the ammonium  ion (NH4). The distribution
of ammonia and ammonium as a function of pH is discussed in Section 6.1.1.

Fixation of nitrogen from N2 gas to  organic nitrogen  is accomplished biologically by
specialized  microorganisms. This reaction requires an investment of energy. Biological
fixation accounts for most of the natural transformation of nitrogen to compounds which
can be used by plant and animal life. Lightning fixation has been estimated to account for
approximately  15 percent of the total which occurs naturally. 3  Industrial fixation was
initially  developed  in  the early  20th  Century for  manufacture  of both fertilizer and
explosives. Presently, nitrogen fixed by industry is about  half the amount that is removed
from the atmosphere by natural means.

Ammonification is the change from organic nitrogen to the ammonium (NH3/NH4) form.
This occurs to dead animal and plant tissue and to animal fecal matter.

                 Protein (organic N) + microorganisms - »- NH-/NH.

Nitrogen in urine exists  principally as urea. Urea is hydrolyzed by the enzyme urease to
ammonium carbonate.

                     H0NCONH0 + 2H00  Enzyme*  (NH,),,CO,
                       ^        i      2.   Urease         °> z   •*

Assimilation is the use of ammonium or nitrate compounds to form plant protein and other
nitrogen-containing compounds.

                 NO^ + CO 2 + green plants + sunlight  - *•  protein


                 NH3/NH4 + CO2 + green plants + sunlight  - »•  protein


                                        2-3

-------
Animals require protein from plants or from other animals. With certain specific exceptions,
they are incapable of converting inorganic nitrogen forms into organic forms.

The  term  "nitrification" is applied to the biological oxidation of ammonium, first to the
nitrite, then to the nitrate, form. The bacteria responsible for these reactions are termed
chemoautotrophic because they use inorganic chemicals as their source of energy. Generally,
the Nitrosomonas genera are involved in conversion of ammonium to nitrite under aerobic
conditions as follows:

                   2NH* f 302  bactena»  2NO~ + 4H+ + 2H2O


The  nitrites are in turn  oxidized to nitrate  generally by Nitrobacter according to  the
following reaction:

                              -1 o_ bacteria.  2No:
                                   ^               J
The overall nitrification reaction is as follows:
To oxidize 1 mg/1 of ammonia-nitrogen requires about 4.6 mg/1 of oxygen when synthesis of
nitrifiers is neglected. The  nitrate thus formed may be used in assimilation as described
above to  promote  plant growth, or it may  be used in denitrification, wherein  through
biological reduction, first nitrite and then nitrogen gas are formed. A fairly broad range of
bacteria can  accomplish  denitrification, including Psuedomonas, Micrococcus, Achromo-
bacter, and Bacillus. In simplified form, the reaction steps are as follows:

                 NO~ + 0.33 CH3OH - •>  NO" + 0.33 CO2 + 0.67 H<0

                      (organic carbon
                          source)

                 NO~ + 0.5 CH3OH  - *• 0.5N2 + 0.5 H2O + OH" + 0.5 CO2

                     (organic carbon
                         source)

Here methanol is used as the example organic carbon source, although many natural and
synthetic organic compounds can serve as the carbon source for denitrification.
                                        2-4

-------
Oxidation  of organic  matter to carbon dioxide and  water furnishes energy for bacteria.
Either oxygen or nitrate may be used for the oxidation, but the use of oxygen results in the
release of more energy.  When both oxygen and nitrate are present, bacteria preferentially
use  oxygen. Therefore,  use of nitrate  for denitrification can only  occur under  anoxic
conditions, an important consideration when attempting to remove nitrate from wastewater.

Nitrite, since it is an intermediate in the nitrification and denitrification processes, can link
the  nitrification and denitrification steps directly without passing through nitrate. First,
nitrite is formed  from  oxidation of ammonium  by Nitrosomonas,  then nitrite can  be
denitrified to nitrogen gas.  By this route less oxygen is required for nitrification and  less
organic matter (energy)  is required  for denitrification. This is  a special case, however, and
not broadly applicable to municipal wastewater treatment.

In discussing the nitrogen cycle, it is useful to differentiate between the surface water and
sediment environment and the soil/groundwater environment. This aids in understanding the
roles that nitrogenous compounds play in each and the problems which can be encountered.

     2.2.1 The Nitrogen Cycle in Surface Waters and Sediments

A modified representation of the nitrogen cycle applicable to the surface water environment
is presented in Figure 2-2.4 Nitrogen can be added by precipitation and dustfall, surface
runoff, subsurface  groundwater  entry, and  direct discharge  of wastewater  effluent.  In
addition, nitrogen from  the atmosphere can be  fixed  by certain photosynthetic blue-green
algae and some bacterial species.

Within the aquatic environment ammonification, nitrification,  assimilation, and denitrifica-
tion can occur as shown in Figure 2-2.  Ammonification of organic matter is carried out by
microorganisms. The ammonium  thus formed, along with nitrate, can be assimilated  by
algae and aquatic plants;  such growths may create water quality problems.

Nitrification of  ammonium can occur with a resulting depletion of the dissolved oxygen
content of the water. To oxidize  1.0 mg/1 of ammonia-nitrogen,  4.6 mg/1 of oxygen is
required.

Denitrification produces  nitrogen  gas which may escape to the atmosphere. Because anoxic
conditions are required,  the oxygen-deficient hypolimnion (or  lower layer) of lakes and the
sediment zone of streams and lakes are important zones of denitrification action.4

     2.2.2 The Nitrogen Cycle in Soil and Groundwater

Figure 2-3 shows the major aspects of the nitrogen cycle associated with the soil/ground-
water environment.^  Nitrogen can enter the  soil  from wastewater or wastewater effluent,
artificial fertilizers,  plant and animal matter, precipitation,  and  dustfall.  In  addition,
                                         2-5

-------
fHECIflTATION
  AND
 DUSTFALL
                                               FIGURE 2-2
                      THE NITROGEN CYCLE IN SURFACE WATER (AFTER REFERENCE 4)
                RUNOFF
WASTEWATER
 EFFLUENT
                                         ATMOSPHERE

-------
                                              FIGURE 2-3
                THE NITROGEN CYCLE IN SOIL AND GROUNDWATER (AFTER REFERENCE 5)
PRECIPITATION
  AND
 DUSTFALL
WASTEWATER
  AND
WAfTEWATER
 EFFLUENT

-------
nitrogen-fixing bacteria convert. nitrogen gas into  forms available to plant life. Man has
increased the amount of nitrogen fixed biologically by cultivation of leguminous crops (e.g.,
peas  and  beans).  It  is  estimated  that  nitrogen  fixed  by  legumes now" accounts for
approximately 25 percent of the total fixed.3

Usually more than 90 percent of the nitrogen present in soil is organic, either in living plants
and animals or in humus originating from decomposition of plant and animal residues. Most
of the remainder is ammonium (NIfy), which is tightly bound to soil particles.

The nitrate content is generally low due to assimilation by plant roots and leaching by water
percolating through the soil. Nitrate  pollution is the principal groundwater quality problem
in many areas.  Denitrification, which is the dominating reaction below the aerobic top layer
of soil, rarely removes all nitrates added to the soil  from fertilizers or wastewater effluents.
Thus, most of the nitrogen  which is not  assimilated by  plant growth eventually enters the
groundwater table in the nitrate form.

2.3 Sources of Nitrogen

Nitrogenous materials may enter the aquatic environment from either natural or man-caused
sources. Further,  the quantities from natural sources are often increased  by man's activity.
For example,  while some nitrogen may  be expected in rainfall, the combustion of fossil
fuels or the application of liquid ammonia agricultural fertilizers with subsequent release to
the air through volatilization can increase rainfall concentrations of nitrogen substantially. It
is useful to have an understanding of the various sources  of nitrogenous materials and to
have an appreciation of the quantities of nitrogen which may be expected from each.

Although  the  source of nitrogen causing a specific pollution  problem is often obvious,
difficulty  may be  encountered in determining which of several possible sources is most
important. As  an example, if a  stream with  excessive aquatic growths due to nitrogen
receives effluent  from  a  sewage  treatment plant,  drainage from fertilized cropland,  and
runoff from pastures or feedlots, the  contribution of nitrogen from the treatment plant may
be a  small fraction of that from the other two  sources. Thus, in analyzing a nitrogen
pollution problem,  care must be taken to ensure that all  possible sources are investigated
and that the amount to be expected from each is accurately estimated. Once an estimate is
made, nitrogen control measures can be oriented toward the more significant sources.

     2.3.1  Natural Sources

Natural sources of nitrogenous substances include precipitation, dustfall, nonurban runoff,
and biological fixation. Amounts  from all may be increased in some way by man. It may be
quite difficult  to determine quantities which might be expected under completely natural
conditions.

In order to find levels of  nitrogenous substances in precipitation  which are as close to
"natural" as possible,  it is  necessary to take samples far from urban or agricultural areas.
                                         2-8

-------
Even these values may be suspect, however. In one review of nutrient levels in precipitation,
total nitrogen in rainfall  in Sweden was  cited as 0.2 mg/1.6 The average concentration of
nitrogen in western snow samples, mainly in the Sierra Nevada Mountains, was 0.15 ppm of
ammonia-nitrogen,  0.01  ppm of nitrite-nitrogen and 0.02 ppm of nitrate-nitrogen. How
representative such values are of "natural" conditions cannot be determined with any
certainty.

The quantities of nitrogen in nonurban runoff from non-fertilized land may be expected to
vary greatly, depending on the erosive characteristics of  the soil. One study found that
runoff from forested land in Washington contained 0.13 mg/1 of nitrate-nitrogen and 0.20
mg/1 of total nitrogen.7

Biological fixation  may add  nitrogen  to both soil and surface water environments. Of
particular interest is the role of fixation in eutrophication  of lakes. Certain photosynthetic
blue-green algae, such as  the  species of Nostoc, Anabaena, Gleotrichia and Calothrix, are
common nitrogen fixers.^

As  much as 14 percent of the total nitrogen entering eutrophic Lake Mendota, Wisconsin,
was added by fixation.^  The role of nitrogen fixation in oligotrophic lakes has not been
established.

     2.3.2 Man-caused Sources

The activities  of  man   may  increase  quantities  of  nitrogen  added  to  the  aquatic
environment  from  three  of the  sources  discussed  above: precipitation, dustfall, and
nonurban  runoff. These sources are increased principally by fertilization of agricultural land
and the combustion of fossil fuels.

Other man-related sources include runoff from urban areas and livestock feedlots, municipal
wastewater effluents, subsurface drainage from agricultural lands and from septic tank leach
fields, and industrial wastewaters.

Nitrogen concentrations in raw municipal wastewaters are well documented.^'"'" Values
generally range  from  15  to 50  mg/1,  of which approximately 60 percent  is ammonia-
nitrogen, 40 percent is organic nitrogen, and a negligible amount (one percent) is nitrite- and
nitrate-nitrogen. Unless wastewater  treatment facilities are  designed  to remove  nitrogen
specifically, most will pass through the treatment works  to the receiving  waters or land
disposal site.  An estimate for the total amount of nitrogen discharged into sewerage systems
in domestic wastewater is 0.84 million metric tons per year in the United States."

Nitrogen discharged into individual septic tank systems can also create pollution problems.
It has  been estimated that up to 25 percent of the national population utilizes individual
systems, 9  contributing up to  0.23  million  metric  tons  of nitrogen  annually.  In  a
well-operating septic tank system, most of the nitrogen leaving the tank will be converted to
nitrate in  the leaching field. This may then percolate downward to a groundwater table.

                                        2-9

-------
Problems from  high nitrate  concentrations  occasionally occur when  septic tank  waste
disposal  is located  near shallow wells used for water supply, particularly on the fringes of
urban areas where the population density may  be fairly high.

The nitrogen content of industrial wastes varies dramatically from one industry to the next.
Among those industries whose wastewater nitrogen contents may be quite high are meat
processing plants,  milk  processing plants,  petroleum  refineries,  ice  plants,  fertilizer
manufacturers, certain synthetic fiber plants, and industries using ammonia for scouring and
cleaning  operations.^

Feedlot runoff constitutes a source of nitrogen which has become significant as  a result of
the  increased  number of concentrated,  centralized  feedlots. Ammonium  is a major
constituent of feedlot waste as a result of urea hydrolysis. Ammonia-nitrogen concentra-
tions may reach 300 mg/l,4>8,10 an(j organjc nitrogen concentrations of up to 600 mg/1
have  been reported.^>10 The total annual nitrogen  load  from livestock in  the U.S. is
estimated to  be  6.0 million metric tons.4 While the majority of the animals are apparently
still raised on small farms, the trend toward  feedlot  operations is  continuing, and unless
steps are taken to prevent  drainage and runoff, serious  localized problems can occur.

Urban runoff can contribute significant quantities of nitrogen to receiving waters during and
after periods  of precipitation. Average concentrations which have been reported are 2.7 mg/1
total nitrogen in Cincinnati, 11 2.1 mg/1 total nitrogen in Washington, D.C.,12 2.5 mg/1 total
nitrogen  in Ann Arbor, Michigan,^ and 0.85  mg/1 organic nitrogen in Tulsa, Oklahoma. 14
Sanitary  or combined sewer overflows can also add to the nitrogen load.

The use  of artificial fertilizers has increased the nitrogen  concentrations which can  be
expected in nonurban runoff. In  rural Ohio, runoff from a 1.45 acre field planted in winter
wheat contained an average of 9  mg/1 total nitrogen. 15 For agricultural land in Washington,
the nitrate-nitrogen concentration was  1.25  mg/lJ On  a 75-acre site  in North Carolina
which consisted of grassed pasture, wooded pasture, corn  field, and orchard,  the mean
nitrogen  concentration in  the runoff was 1.2 mg/1.16

Subsurface irrigation drainage from fertilized  cropland can contain high concentrations of
nitrates.  In agricultural areas of California's San Joaquin Valley, monitoring of subsurface
tile drainage  systems between 1966 and 1968 showed average nitrate-nitrogen concentra-
tions of  19.3  mg/1.17

In the same way that increased nitrogen concentrations in nonurban runoff and subsurface
drainage  have been caused by man's activities, increased nitrogen levels in precipitation and
dustfall have also resulted. For example, high ammonium concentrations in spring rains in
California  are due to the  use of liquid ammonium  fertilizers there."  Most atmospheric
nitrogen  (other  than nitrogen gas), however, is associated with soil picked up by the wind
and can  be returned to earth by  gravitational  settling (dry fallout) or in precipitation, and
several studies have been  conducted to  determine  the  quantities to be expected from such

                                        2-10

-------
sources. The  10-year average of ammonia- plus nitrate-nitrogen concentrations in rainfall at
Geneva, New York, was  1.1  mg/1.6 Snow samples from Ottawa, Canada, over 17 years
contained an average of 0.85  ppm inorganic nitrogen." Rainwater from the same area for
the  same  period  had  concentrations  of 1.8  mg/1  ammonia-nitrogen  and  0.35  mg/1
nitrate-nitrogen. In rainfall measurements at Cincinnati, Ohio, total and inorganic nitrogen
concentrations were 1.27  and 0.69 mg/1, respectively.^ For a rural area near Coshocton,
Ohio, the respective concentrations were 1.17 and 0.80 rrig/1.^

A study made near Hamilton,  Ontario, was cited^ which related dustfall to rainfall. It was
found that the nitrogen fall totaled 5.8 Ib per acre per year. Approximately 61 percent of
the nitrogen came down on rainy days, which constituted 25 percent of the days monitored
during the test.

In a study on dustfall in Seattle^ the fall rate  for soluble nitrate-nitrogen was 0.63 Ib per
acre per year. The concentration of nitrate-nitrogen in the total dustfall was 700 ppm.

As a summary to this discussion of sources of nitrogen,  Table  2-1  shows estimates of
nitrogen  quantities discharged from  various sources  in the San Francisco  Bay  Basin,
California. 1 ^ The bay basin has a population of about 4,500,000 people, a land  area of
4,300 square miles, and a  water surface area of about 450 square miles. Because of the high
population density,  the greatest  amount of nitrogen  discharged  is from  municipal  and
industrial sources. This table is presented only as an example. Care must be taken for each
case to accurately evaluate the  significance of each source.
                                    TABLE 2-1

                     ESTIMATED NITROGEN LOADINGS FOR
                        THE SAN FRANCISCO BAY BASIN

Identified Nitrogen Source

Municipal wastewater, before treatment
Industrial wastewater, before treatment
Vessel wastes , before treatment
Dustfall directly on Bay
Rainfall directly on Bay
Urban runoff
Non-urban runoff
Nitrogen applied to Irrigated agricultural land
Nitrogen from dairies and feedlots
Total
Nitrogen mass emission.
thousand Ib per. year
(thousand kg per year)
55,000 (26,000)
35,000 (16,000)
130 ( 60)
1,300 ( 590)
870 ( 390)
3,100 ( 1,400)
4,100 ( 1,900)
2,000 ( 900)
13,000 ( 6,000)
118,000 (53,000)
Percent
of
total
49
30
0.1
1 .1
0.8
2.7
3.6
1.7
11
100
 From Reference 19
 a
 'A major source not Included is biological fixation
 An estimated 50 percent percolates to giroundwater
                                        2-11

-------
2.4 Effects of Nitrogen Discharge

It was previously noted that nitrogenous compounds discharged from wastewater treatment
facilities can have several deleterious effects. Although biostimulation of receiving waters
has generated the most concern in recent years, other less well publicized impacts can be of
major importance  in  particular situations. These impacts include toxicity  to  fish life,
reduction of chlorine disinfection efficiency, an increase in the dissolved oxygen depletion
in receiving  waters,  adverse public health effects — principally in groundwater, and  a
reduction in the suitability for reuse.

     2.4.1 Biostimulation of Surface Waters

A major problem in the field of water pollution  is eutrophication, excessive plant growth
and/or algae "blooms" resulting from over-fertilization of rivers, lakes, and estuaries. Results
of eutrophication   include deterioration in the appearance of previously clear waters, odor
problems from  decomposing algae, and a lower dissolved oxygen  level which can adversely
affect fish life.

Four basic factors are required  for algal growth: nitrogen, phosphorus, carbon dioxide, and
light energy.  The absence of any one will limit growth. In special cases, trace micronutrients
such  as cobalt,  iron,  molybdenum and manganese may be limiting factors under natural
conditions.

Good generalizations concerning which factor is growth limiting and at what concentration
are difficult to make. Light and carbon dioxide are essentially impossible to control. Both
nitrogen and  phosphorus are present in  waste discharges and hence subject to control. The
questions which must usually be answered when faced with a eutrophication problem are: is
nitrogen or phosphorus (or neither) the  limiting nutrient, and  if either one is, can the
amount entering the receiving water be significantly reduced by removing that nutrient from
the waste stream? In some cases algal assay procedures may allow a conclusion as to which
nutrient is limiting. Under some circumstances,  however, removal of both nitrogen and
phosphorus may be  undertaken to limit algal growth.

Eutrophication   is  of most concern in lakes  because nutrients  which  enter tend to be
recycled within  the lake and build up  over a period of time. 9 A river, by contrast, is a
flowing system.  Nutrients are always entering or leaving any given section. Accumulations
tend to occur only  in sediment or in slack water, and the effects of these accumulations are
normally moderated by periodic flushing by floods.

In estuaries and oceans, nitrogen compounds are  often present in very low concentrations
and  may limit  the  total biomass and the types of species it  contains. 9 Thus, upwelling,
which brings nutrient-rich waters to the surface, may result in periodic blooms of algae or
other aquatic life. While in some estuaries discharges from wastewater treatment plants may
increase nitrogen concentrations to the level where blooms occur, the high dilution provided

                                        2-12

-------
by a direct ocean discharge probably eliminates the danger of algae blooms caused by such
discharges.  In summary, while nitrogen in wastewater treatment plant effluents can in
particular cases cause undesirable aquatic growths, determination of the limiting constituent
and other sources of that constituent (such as feedlot runoff or fixation) should be made
before the decision is made to require nitrogen removal from municipal wastewaters.

     2.4.2 Toxicity

The principal  toxicity  problem is  from ammonia in the molecular form (NH3> which can
adversely  affect fish life  in receiving  waters. A  slight increase in pH may  cause a great
increase in toxicity as the ammonium ion  (NH4) is transformed to ammonia in accordance
with the following equation.

                          NH* + OH~ —- NH3 + H20

Factors which may increase ammonia toxicity at a given pH are: greater concentrations of
dissolved oxygen and carbon dioxide; elevated temperatures; and bicarbonate alkalinity.
Reported levels at which acute toxicity is  detectable have ranged from 0.01  mg/K to over
2.0 mg/l^O of molecular ammonia-nitrogen.

     2.4.3 Effect on Disinfection Efficiency

When chlorine, in the  form of chlorine gas or hypochlorite salt, is  added to wastewater
containing ammonium, chloramines, which are less effective  disinfectants, are formed. The
major reactions are as follows:

                  NH* + HOC1 i=^  NH2C1 (monochloramine) + H2O + H+


                  NH2C1 + HOC1  :^=^ NHC12 (dichloramine) + H2O


                  NHC12 + HOC1  y*" NC13 (nitrogen trichloride) + H2O
Only after the addition of large quantities of chlorine does free available chlorine exist. If
the effluent ammonia-nitrogen concentration were 20 mg/1, about 200 mg/1 of chlorine
would be required to complete the reactions with ammonium and organic compounds. Only
rarely in wastewater treatment is this level of chlorine addition ("breakpoint" chlorination)
used.  Therefore,  as a  practical matter, the  less effective combined chlorine residuals
(monochloramine and  dichloramine) must be relied upon for disinfection. This results in
increased chlorine dose requirements for the same level of disinfection. Further information
on the relative effectiveness of free  chlorine  and combined residuals is presented in Section
6.2.7.
                                        2-13

-------
     2.4.4 Dissolved Oxygen Depletion in Receiving Waters

Ammonium can be biologically oxidized to nitrite and then to nitrate in receiving waters
and  thereby add to the oxygen demand imparted by carbonaceous materials. Table 2-2
shows a typical  example of the removal of total oxygen demand obtainable with varying
degrees  of treatment.  If  either conventional  biological  treatment  or  physical-chemical
treatment is utilized to provide 90 percent BOD5 removal, an effluent will be discharged
which still contains over 100 mg/1 of oxygen demand. This high level of oxygen demand
may cause significant oxygen  depletion in the receiving water if insufficient dilution is
available. Nitrification (or ammonia nitrogen removal) will reduce the total oxygen demand
of the effluent to less than 40 mg/1.

The Potomac Estuary in the United States22 and the Thames Estuary in Great Britain^^ are
examples  of estuaries which are greatly affected by nitrification. Figure 2-4  shows, as a
function of the degree of nitrification provided  by wastewater treatment facilities, the
estimated  discharge  into  the  Thames  Estuary  which will  cause the maximum  oxygen
depletion to be 10 percent of saturation. The calculation assumes an  effluent BOD5 of 20
mg/1, an effluent organic plus ammonia-nitrogen concentration of 19 mg/1, and discharge at
a  point  10 miles above London Bridge.  From the figure,  the allowable discharge for
non-nitrified effluent is about 12 mgd, while for completely nitrified effluent, over 40 mgd
can be discharged.

     2.4.5 Public Health

The  public health hazard from nitrogen is associated with the nitrate form and is limited
principally to groundwater where high concentrations can occur. Nitrate  in drinking water

                                    TABLE 2-2
              EFFECT OF AMMONIUM REMOVAL ON TOTAL OXYGEN
           DEMAND OF WASTEWATER TREATMENT PLANT EFFLUENT


Parameter

Organic matter, mg/1
Organic oxygen demand, mg/1
Organic and ammonia nitrogen, mg/1
Nitrogenous oxygen demand (NOD) mg/1
Total oxygen demand (TOD) mg/1
Percent of TOD due to nitrogen
Percent organic oxygen demand removed
Percent of TOD removed

Raw
wastewater

250
375b
25
115C
490
23.5
-
-
Final effluent
Organic carbon
removal
only
25
37
20
92°
129
71.3
90
73.7
With ammonium
and organic carbon
removal
20
30
1.5
7c
37
18.9
92
92.5
 "After Reference 21
  Taken as 1.5 times organic matter
 cTaken as 4.6 times the nitrogen level
                                       2-14

-------
was first  associated in  1945  with methemoglobinemia,  a sometimes fatal blood disorder
which affects infants less than three months old. When water high in nitrate is used for
preparing infant formulas, nitrate is reduced to nitrite in the stomach after ingestion. The
nitrites react with hemoglobin in the blood to form methemoglobin, which is incapable of
carrying oxygen. The result is suffocation accompanied by a bluish tinge to the skin, which
accounts for the use of the term "blue babies" in conjunction with methemoglobinemia. In
suspect areas water should be analyzed for both nitrite and nitrate since either form will
cause methemoglobinemia.

                                 FIGURE 2-4

       ALLOWABLE EFFLUENT DISCHARGE INTO THE THAMES ESTUARY
                          (AFTER REFERENCE 23)
             50
        m

        TJ
        O»
        E

        LJ
        O
        CL
        <
        X
        o
        UJ
        ID
        U_
        UJ
        UJ
        _l
        CD
        I
        o
40
     BOD5 OF WASTEWATER EFFLUENT: 20 mg/l
     POINT OF DISCHARGE : IOMILES
       ABOVE  LONDON  BRIDGE
     MAXIMUM OXYGEN  DEPLETION:
       10 PERCENT   OF  SATURATION
                                                                   100
                           PERCENT  NITRIFICATION
                         OF  WASTEWATER  EFFLUENT
                                    2-15

-------
Since 1945  about 2,000 cases of methemoglobinemia  have been reported in the U.S. and
Europe, with a mortality rate of seven to eight percent. Because of difficulty in diagnosing
the disease and because no reporting is required, the actual incidence may be many times
higher. !0

The  EPA's interim primary drinking water standard (40  CFR Part 141) for nitrate is 10 mg/1
as nitrogen.  This standard is exceeded most often in shallow wells in rural areas.

     2.4.6 Water Reuse

While direct wastewater reuse for domestic water supply is not yet a reality  because  of
public health considerations, plans for industrial reuse are being carried out in several areas.
When reclaiming wastewater for industrial purposes, ammonia may need to be removed in
order to  prevent  corrosion.  Further, nitrogen compounds can  cause biostimulation  in
cooling towers and distribution structures.

2.5 Treatment Processes for Nitrogen Removal

In the past several years the number  of  processes utilized in  wastewater  treatment has
increased rapidly. Many of these processes have been developed with the specific purpose of
transforming nitrogen compounds or removing nitrogen from the wastewater stream. Others
can remove  several compounds, including significant amounts of nitrogen. Still others may
remove  only a small amount of nitrogen or a  particular form of nitrogen which is a small
fraction of the total.

In determining which method is  most suitable for a particular application, consideration
must be given to six principal aspects: (1) form and concentration of nitrogen compounds in
the process influent,  (2)  required  effluent quality,  (3) other treatment processes to  be
employed, (4) cost,  (5)  reliability,  and (6) flexibility. Great care  must  be taken  in
developing and evaluating alternatives.

Presented  below  are  brief descriptions of the various processes employed  in wastewater
treatment facilities which, to varying degrees,  remove nitrogen  from the  waste stream.
Process  characteristics, compound selectivity, and normal range of efficiency are presented.
It is stressed that this  discussion  is  descriptive and is  intended only  to provide  an
introduction to the following chapters of this manual.

     2.5.1 Conventional Treatment Processes

Nitrogen in raw domestic wastewaters is principally in the form of organic nitrogen, both
soluble and  particulate, and ammonia. The soluble organic nitrogen is mainly in the form of
urea and amino acids. Primary sedimentation acts to  remove a portion of the particulate
organic  matter. This generally will  amount to less than 20 percent of the  total nitrogen
entering the plant.

                                        2-16

-------
Biological treatment will remove more particulate organic nitrogen and transform some to
ammonium and other inorganic forms. A fraction of the ammonium present in the waste
will be assimilated into organic materials of cells  formed by the biological process. Soluble
organic nitrogen is partially transformed to ammonium  by microorganisms, but concentra-
tions  of  1 to  3 mg/1 are usually found in biological treatment effluents.24 Through these
processes, an additional 10 to 20 percent of the total nitrogen is removed when biological
treatment and secondary sedimentation follows primary  sedimentation. Thus, total nitrogen
removal for a conventional primary-secondary facility will generally  be less than about 30
percent.

     2.5.2 Advanced Wastewater Treatment Processes

Advanced treatment processes  designed  to remove  wastewater constituents other  than
nitrogen  often remove some nitrogen  compounds as well. Removal is often restricted to
particulate forms, and overall removal efficiency is rarely high.

Tertiary filtration can remove a significant fraction of the organic nitrogen present. Overall
removal depends on the amount of nitrogen in the suspended organic  form. As noted above,
most  of  the organic nitrogen  in secondary effluent is insoluble,  but ammonium usually
accounts for the majority of the total nitrogen. Carbon adsorption, used to remove residual
organics, will  also remove organic nitrogen. The amount of organic nitrogen remaining at
that point in the treatment scheme will generally be quite small.

Electrodialysis and reverse osmosis are tertiary processes used primarily for reduction of
total dissolved solids. Nitrogen  entering such systems is  mainly in the ammonium or nitrate
form. Electrodialysis can  be expected  to remove about 40 percent of these forms; reverse
osmosis,  80 percent. However, these processes are not currently in use for treatment of
municipal wastewater.

Chemical  coagulation,  often utilized   for phosphate  removal, also aids  in  removal  of
particulate matter, including particulate organic nitrogen. While chemical coagulation does
not remove ammonium directly, lime addition is used prior to ammonia stripping (discussed
in Section 2.5.3.4) in order to raise the  pH and allow the process to proceed.

Land disposal may be used to remove  nitrogen. Removal occurs when the effluent is used
for irrigation  purposes  with  the  nitrogen  assimilated  by  growing crops  which  are
subsequently harvested. However, nitrogen removal by land treatment systems is not within
the scope of this manual.

     2.5.3 Major Nitrogen Removal Processes

The major processes considered in this manual are nitrification-denitrification, breakpoint
chlorination (or superchlorination), selective ion exchange for ammonium removal, and air
                                         2-17

-------
stripping for ammonia removal (ammonia  stripping). These are the processes which  are
technically and economically most viable at the present time.

          2.5.3.1  Biological Nitrification-Denitrification

Biological nitrification does not increase the removal of nitrogen from the waste stream over
that achieved by conventional biological treatment. The principal effect of the nitrification
treatment process is to transform ammonia-nitrogen to nitrate. The nitrified effluent can
then be  denitrified  biologically. Nitrification is also used without subsequent biological
denitrification  when  treatment  requirements  call for  oxidation  of  ammonia-nitrogen.
Oxidation of ammonium  can  be as high  as 98  percent. Overall  transformation to nitrate
depends on the extent to  which organic nitrogen is transformed to ammonia-nitrogen in the
secondary  stage or  is removed by  another  process. Nitrification  can be  carried out in
conjunction with secondary treatment or  in a tertiary stage; in both cases, either suspended
growth reactors (activated sludge) or attached growth reactors (such as trickling filters) can
be used.

Biological denitrification  can also be carried out in either suspended growth or attached
growth reactors. As previously noted, an anoxic environment is required for the reactions to
proceed. Overall removal efficiency in a nitrification-denitrification plant can range from 70
to 95 percent.

         2.5.3.2 Breakpoint Chlorination

Breakpoint chlorination (or superchlorination) is accomplished by the  addition of chlorine
to the waste stream in an amount sufficient to  oxidize ammonia-nitrogen to  nitrogen gas.
After sufficient chlorine is added to oxidize the  organic matter and other readily oxidizable
substances  present, a stepwise reaction of chlorine with ammonium takes place. The overall
theoretical reaction is as follows:

                         3C12 + 2NH* 	*-  N2 + 6HC1 + 2H+


In  practice,  approximately  10  mg/1  of chlorine  is  required  for every 1  mg/1  of
ammonia-nitrogen. In addition, acidity produced by the reaction must be neutralized by  the
addition of caustic soda or lime. These chemicals add greatly to the total dissolved solids and
result  in  a substantial  operating  expense. Often dechlorination  is  utilized  following
breakpoint  chlorination in order to reduce the toxicity of the chlorine residual in  the
effluent.

An important  advantage of this method  is that ammonia-nitrogen  concentrations can be
reduced  to near zero in  the effluent. The  effect of breakpoint chlorination on  organic
nitrogen is  uncertain, with contradictory results  presented in the literature. Nitrite and
nitrate are not removed by this method.

                                        2-18

-------
          2.5.3.3 Selective Ion Exchange for Ammonium Removal

Selective ion exchange for removal of ammonium from wastewater can be accomplished by
passing  the  wastewater through  a column  of clinoptilolite, a naturally occurring zeolite
which has a high selectivity  for ammonium ion. The first extensive study was undertaken in
1969 by Battelle Northwest in a federally sponsored demonstration project. Regeneration of
the clinoptilolite is undertaken when all the exchange sites are utilized and breakthrough
occurs.

Filtration prior to ion exchange is  usually required  to  prevent  fouling  of  the  zeolite.
Ammonium removals  of 90-97  percent can  be expected.  Nitrite,  nitrate, and  organic
nitrogen are not affected by  this process.

          2.5.3.4 Air Stripping for Ammonia Removal

Ammonia in the molecular form is a gas which dissolves in water to an extent controlled by
the partial pressure of the ammonia in the air adjacent to the water.  Reducing the partial
pressure causes ammonia to  leave the water phase and enter the air. Ammonia removal from
wastewater can be effected by bringing small drops of water in contact with a large  amount
of ammonia-free air. This physical process is termed desorption, but the common  name is
"ammonia stripping."

In order to strip ammonia from wastewater, it must  be in the molecular form (NH3) rather
than the  ammonium ion (NHjj)  form.  This is accomplished  by raising the pH of the
wastewater to 10 or 11, usually by the addition of lime. Because lime addition is often used
for phosphate removal, it  can serve a  dual role. Again,  nitrite, nitrate, and organic nitrogen
are not affected.

The principal problems associated with ammonia  stripping are its  inefficiency  in cold
weather, required shutdown during freezing conditions, and formation  of calcium carbonate
scale in  the air stripping tower.

The effect of cold weather has been well documented at the South Lake Tahoe Public
Utility  District  where  ammonia stripping is used for a 3.75  mgd  tertiary facility. The
stripping tower is designed to remove 90 percent of  the incoming ammonium during warm
weather. During  freezing  conditions, the tower is  shut down. One mechanism of scale
formation is attributed to the carbon dioxide in the air reacting with the alkaline wastewater
and precipitating as calcium carbonate.22 in some instances, removal with a water jet has
been possible; in other applications the scale has been extremely difficult to remove. Some
factors which may affect the nature of the scale are: orientation of air flow, recirculation of
sludge, pH of the wastewater, and chemical makeup of the waste water. 2 2
                                        2-19

-------
     2.5.4 Other Nitrogen Removal Processes

In addition to the processes listed above, there  are other methods for nitrogen removal
which might usefully be discussed. Most are in the experimental stage of development or
occur coincidentally with another process.

Use  of anionic exchange resins  for removal of nitrate  was  developed  principally for
treatment of irrigation return waters.22 TWO major unsolved problems are the lack of resins
which have  a  high selectivity  for nitrate over  chloride and disposal of nitrogen-laden
regenerants.

Oxidation  ponds can  remove nitrogen through microbial denitrification in the anaerobic
bottom layer or by ammonia emission to the atmosphere. The latter effect is essentially
ammonia stripping but is relatively inefficent due  to a low surface-volume ratio and low pH.
In a study of raw wastewater lagoons in  California, removals of 35-85 percent were reported
for well-operated lagoons. ^

Nitrogen in oxidation ponds is  assimilated by algal cultures. If the algal cells are removed
from the pond effluent stream, nitrogen removal  is thereby effected. Methods for removal
of  algae   are  summarized in  the  EPA Technology  Transfer  Publication, Upgrading
Lagoons. 25

It was noted previously that in secondary biological treatment and in nitrification, some
nitrogen is incorporated in bacterial  cells and is  removed from  the  waste stream with the
sludge.  If an organic carbon source such as ethanol or glucose is added to the wastewater,
the  solids  production will be increased and a greater nitrogen removal will be  effected.
Disadvantages are that large quantities of sludge are produced and that difficulties occur in
regulating  the  addition of the  carbon  source, with high effluent  BOD5  values or high
nitrogen levels resulting. 2

     2.5.5  Summary

Table 2-3 summarizes the effect of various treatment processes on nitrogen removal. Shown
is the effect that the process has on  each  of the three major forms: organic  nitrogen,
ammonium, and nitrate. In the last column is shown normal removal percentages which can
be expected  from that process. Overall removal for a particular treatment plant will depend
on the types of unit processes and their relation to each other. For example, while many
processes   developed  for nitrogen removal are ineffective in removing organic  nitrogen,
incorporation of chemical coagulation or multimedia  filtration  into the overall flowsheet
can  result in a low  concentration of  organic nitrogen  in the plant effluent. Thus, the
interrelationship between  processes must be carefully analyzed in  designing for nitrogen
removal. Further discussion of process interrelationships is presented in Chapter 9.
                                         2-20

-------
                                      TABLE 2-3
  EFFECT OF VARIOUS TREATMENT PROCESSES ON NITROGEN COMPOUNDS



Treatment process
Conventional treatment processes
Primary
Secondary

Advanced wastewater treatment processes
Filtration0
Carbon sorption
Electrodialysis

Reverse osmosis

Chemical coagulation0
Land application
Irrigation

Infiltration/percolation
Major nitrogen removal processes
Nitrification
Denitrification
Breakpoint chlorination
Selective ion exchange for ammonium

Ammonia stripping
Other nitrogen removal processes
Selective ion exchange for nitrate
Oxidation ponds


Algae stripping

Bacterial assimilation

Effect on constituent

.Organic N

10-20% removed
15-25% removedb
urea -*. NH3/NHj

30-95% removed
30-50% removed
100% of suspend
organic N removed
100% of suspend
organic N removed
50-70% removed

_». NH3/NH4

-*• NH3/NH4

limited effect
no effect
uncertain
some removal, un-
certain
no effect

nil
partial transformation
to NH3/NHJ

partial transformation
to NH3/NH4
no effect ,

NH3/NH4

no effect
;10% removed


nil
nil
4 J% removed

85% removed

ni..

-»• N03
-»• plant N
-»• NOJ

--NO;
no effect
90-100% removed
90-97% removed

60-95% removed

nil '
partial removal
by stripping

-». cells

40-70% removed

NO3

no effect
nil


nil
nil
40% removed

85% removed

nil

-». plant N

-» N2

no effect .
80-98% removed
no -effect
no effect

no -effect

75-90% removed
partial removal by
nitrification-
df. nitrification
-*. cells

limited effect
Removal of
total nitrogen
entering process,
percent9

5-10
10-20


20-40
10-20
35-45

80-90

20-30

40-90

0-50

5-10
70-95
80-95
80-95

50-90

70-90
20-90


50-80

30-70
 Will depend on the fraction of influent nitrogen for which the process is effective, which n.ay depend on other processes
 in the treatment plant.
 Soluble organic nitrogen, in the form of urea and amino acids, is substantially reduced by secondary treatment.
 GMay be used to remove particulate organic carbon in plants where ammonia or nitrate are removed by other processes.
2.6 References

 1.  Sawyer,  C.N., and P.L. McCarty,  Chemistry for Sanitary Engineers.  New York,
     McGraw-Hill Book Co., 1967.
                                                               I
 2.  Christensen,  M.H., and P.  Harremoes,  Biological  Denitrification  in  Wastewater
     Treatment. Report 2-72, Department of Sanitary Engineering, Technical University of
     Denmark, 1972.

 3.  Delwiche, C.C., The Nitrogen  Cycle.  Scientific  American, .223, No. 3,  pp 137-146
     (1970).
                                          2-21

-------
 4.  Martin,  D.M., and  D.R. Goff,  The Role of Nitrogen in the Aquatic Environment.
    Report  No.  2, Department of Limnology, Academy of Natural  Sciences of Phila-
    delphia, 1972.

 5.  Sepp, E., Nitrogen  Cycle in Groundwater. Bureau of Sanitary Engineering, State of
    California Department of Public Health, 1970.

 6.  McCarty, P.L., et al, Sources of Nitrogen and Phosphorus in Water Supplies. JAWWA,
    59, pp 344 (1967).

 7.  Sylvester,  R.O.,  Nutrient  Content of  Drainage  Water  for  Forested,  Urban, and
    Agricultural Areas. Alfjae and Metropolitan Wastes, Robert A. Taft Sanitary Engineer-
    ing Center, Tech. Rep. W61-3, 1963.

 8.  Reeves,  T.G., Nitrogen Removal: A  Literature  Review.  JWPCF, 44,  No.  10, pp
    1896-1908(1972).

 9.  Nitrogenous  Compounds in the Environment.  Hazardous Materials Advisory Com-
    mittee (to the EPA), EPA-ASB-73-001, December,  1973.

10.  Kaufman, W.J., Chemical Pollution of Ground Waters. JAWWA, 66, No. 3, pp 152-159
    (1974).

11.  Weibel,  S.R., Anderson, R.J., and R.L. Woodward, Urban-Land Runoff as a Factor in
    Stream Pollution. JWPCF, 43, p 2033 (1971).

12.  American  Public Works Association, Water Pollution  Aspects  of Urban Runoff.
    FWPCA Report No. WP-20-15, January, 1969.

13.  Burn, R.J., Krawezyk, D.F., and G.T. Harlow, Chemical and Physical Comparison of
    Combined and Separated Sewer Discharges. JWPCF, 40, pp 112 (1968).

14.  Avco Economic  Systems Corporation,  Storm  Water Pollution  From  Urban Land
    Activity. EPA Report No. 11034 FKL 07/70, July, 1970.

15.  Weibel,  S.R., et al, Pesticides and Other Contaminants  in Rainfall  and Runoff.
    JAWWA, 58, pp 1075 (1966).

16.  Dept. of Biological and Agricultural Engineering, North Carolina State  University at
    Raleigh, Role of Animal Wastes in Agricultural Land Runoff. EPA Report No. 13020
    DGX 08/71, August 1971.

17.  California Departmen't  of  Water  Resources, Nutrients From  Tile Drainage Systems.
    EPA Report No. 13030 ELY 5/71-3, May 1971.

                                      2-22

-------
18.  Johnson, R.E., Rossano, A.T., Jr., and R.O. Sylvester, Dustfall as a Source of Water
     Quality Impairment, ASCE, JSED, 92, No. SA1, pp 145 (1966).

19.  California State Water Resources Control Board, Tentative Water Quality Control Plan,
     San Francisco Bay Basin. November, 1974.

20.  Brown and Caldwell/Dewante and Stowell, Feasibility Study for the Northeast-Central
     Sewerage Service Area.  Prepared  for County of Sacramento, Department of Public
     Works, November 1972.

21.  Ehreth, D.J., and E. Barth, Control of Nitrogen in Wastewater Effluents. Prepared for
     the EPA Technology Transfer Program, March, 1974.

22.  Nitrogen Removal  from  Wastewaters.  EPA  Advanced  Waste Treatment  Research
     Laboratory, ORD-17010, October, 1970.

23.  Effects  of Pollution Discharges on the Thames Estuary. Department of Scientific and
     Industrial Research, Water Pollution Research Technical Paper No. 11, Her Majesty's
     Stationery Office, 1964.

24.  Parkin,  G.F., and P.L. McCarty, The Nature, Ecological Significance, and Removal of
     Soluble Organic Nitrogen in Treated Agricultural Wastewaters. Stanford University,
     prepared  for  the  Bureau  of  Reclamation,   Contract  USDI- 14-06-200-6090-A,
     September, 1973.

25.  Caldwell, D.H., Parker, D.S., and W.R. Uhte, Upgrading Lagoons. Prepared for the EPA
     Technology Transfer Program, August, 1973.
                                       2-23

-------
                                     CHAPTER 3

                  PROCESS CHEMISTRY AND BIOCHEMISTRY OF
                     NITRIFICATION AND DENITRIFICATION
3.1 Introduction

The purpose  of this  chapter is to present a treatment process-oriented review of the
chemistry and biochemistry of nitrification and denitrification. An understanding of this
subject is useful for developing an appreciation of the factors affecting the performance,
design, and operation of nitrification and denitrification processes. Subsequent chapters deal
with design aspects of nitrification (Chapter 4) and denitrification  (Chapter 5). Since these
latter  chapters are laid  out to be used without reference to this  chapter, review of the
theoretical material in this chapter is not mandatory.

Biological processes for control of nitrogenous residuals in effluents can be classified in two
broad  areas.   First, a  process designed to  produce an effluent where  influent nitrogen
(ammonia  and organic  nitrogen) is  substantially  converted to nitrate nitrogen can be
considered.  This  process,  nitrification,  is  carried out  by bacterial  populations  that
sequentially oxidize ammonia  to nitrate with intermediate formation of nitrite. Nitrification
will satisfy effluent or receiving  water standards where reduction of residual nitrogenous
oxygen demand due to ammonia is mandated or where ammonia reduction for other reasons
is  required for the treatment  system.  The second  type of process, denitrification, reduces
nitrate to nitrogen gas and can be used following  nitrification when the total nitrogenous
content of the effluent must be reduced.

3.2 Nitrification

The two principal genera of importance in biological nitrification processes  are Nitro-
somonas and Nitrobacter. Both of these groups are classed as autotrophic organisms. These
organisms are  distinguished from heterotrophic bacteria  in  that  they derive  energy for
growth  from  the  oxidation  of inorganic  nitrogen  compounds,  rather than  from the
oxidation of organic matter. Another  feature of these organisms is that inorganic carbon
(carbon dioxide) is used for synthesis rather than organic  carbon. Each group is limited to
the oxidation of  specific  species of nitrogen compounds. Nitrosomonas can  oxidize
ammonia to  nitrite, but cannot  complete  the  oxidation to nitrate. On the other hand,
Nitrobacter is limited to the oxidation of nitrite to nitrate. Since complete nitrification is a
sequential reaction, treatment processes must  be designed to provide an environment
suitable to the growth of both  groups of nitrifying bacteria.

     3.2.1 Biochemical Pathways

On a biochemical level, the nitrification process is more complex than simply the sequential
oxidation of ammonia  to nitrite by Nitrosomonas and the subsequent oxidation of nitrite to
                                         3-1

-------
nitrate by Nitrobacter. Various reaction intermediates and enzymes  are involvedJ More
important than an  understanding  of these pathways is knowledge of the response  of
nitrification organisms to environmental conditions.

      3.2.2. Energy and Synthesis Relationships

The stoichiometric reaction for oxidation of ammonium to nitrite by Nitrosomonas is:

                 NH*  +  1.5O2  	+•  2H+  +  H2O +  NO~                (3-1)

    &M*
The Jess of free energy by this reaction at physiological concentrations of the reactants has
been  estimated by  various investigators  to  be  between  58  and 84  kcal per mole  of
ammonia. *>2

The reaction for oxidation of nitrite to nitrate by Nitrobacter is:

                       N0~   + 0.5 02  	»» N0~     .                       (3-2)


This  reaction has  been estimated to release between  15.4 to 20.9 kcal per mole of nitrite
under in  vivo conditions. 2 Thus, Nitrosomonas obtains more energy per mole of nitrogen
oxidized than Nitrobacter.  If it assumed that the cell synthesis per unit of energy produced
is equal, there should be  greater mass of Nitrosomonas formed than Nitrobacter per mole of
nitrogen oxidized. As will be seen, this is in fact the case.

The  overall oxidation of ammonium  by both groups is obtained by  adding Equations 3-1
and 3-2:

                 NH*  +  202 —*•  N0~ + 2H+  +  H23  +  C5H?NO2  +  H+     (3-5)
                                                        Nitrobacter
                                        3-2

-------
Equations 3-1, 3-4 and 3-5 have terms showing the production of free acid (H+) and the
consumption of gaseous carbon dioxide (CC>2)- In  actual fact, these reactions take place in
aqueous systems in the context of the carbonic acid system. These reactions usually take
place at pH  levels less than 8.3. Under this circumstance, the production of acid results in
immediate reaction with bicarbonate ion (HCOs) with the production of  carbonic acid
(H2CO3). The consumption of carbon dioxide by the organisms results in some depletion of
the  dissolved  form of carbon  dioxide,  carbonic  acid (H2CO3). Table 3-1 presents  the
modified forms of Equations 3-1 to 3-5 to reflect  the changes in the carbonic acid system.
As  will be later described in  Sections 3.2.3 and  3.2.5.6,  the variations occurring in pH
resulting from changes in  the  carbonic acid system can significantly affect nitrification
process performance.

The equations  for energy yielding reactions (Equations 3-1  and  3-2) can be combined with
the equations  for  organism synthesis (Equations  3-4 and 3-5) to form overall synthesis-
oxidation  relations by knowledge  of the yield coefficients for the nitrifying organisms.
Experimental yield values for Nitrosomonas range  from 0.04 to 0.13 Ib VSS grown per Ib
ammonia nitrogen oxidized. 1 Experimental yields for Nitrobacter are in the range from 0.02
to 0.07 Ib VSS grown per Ib of nitrite nitrogen oxidized.l Values based on thermodynamic
theory  are  0.29  and  0.084  for  Nitrosomonas and  Nitrobacter, respectively.2  The
experimentally based yield may be lower than  theoretical values due to the  diversion of a
portion of the free energy released by oxidation to microorganism maintenance functions.2

Equations for  synthesis-oxidation using representative measurements of yields and oxygen
consumption for Nitrosomonas and Nitrobacter are  as follows: 3,4
             +  76 (X,  +  109HCO"  	»-  C^HJSKX,  + 54NCX, + 57tt,O    (3-6)
                    Z*             3          J  I    £          £        £
                                          Nitrosomonas
             +  104H2C03


    400NO"  +  NH*  +  4H,,CO, + HCO~  + 195 Oo —». C^H-NO0       ,, 7,
           2*        H"        2*   j         J           Zi        j  I    £       \.^~' )
                                                             Nitrobacter
                3H2O  +  400NO~
Using Equations 3-6 and 3-7, the overall synthesis and oxidation reaction is:
      NH^ +  1.8302  +  1.98HC03 —*~ 0.021 C5H?NO2  +  1.041H2O       (3-8)

           + 0.98NO~  +  1.88H2CO3
                                        3-3

-------
In these equations, yields for Nitrosomonas and Nitrobacter are 0.15 mg cells/mg NH^-N
and 0.02 mg cells/mg NO2-N, respectively. On this basis, the removal of twenty- mg/1 of
ammonia nitrogen would yield only 1.8 mg/1  of nitrifying  organisms. This relatively low
yield  has  some  far reaching implications, as will be  seen  in  Section  3.2.7. Oxygen
consumption ratios in  the equations are 3.22 mg O2/mg Nffy-N  oxidized and 1.11  mg
O2/mg NOJ-N oxidized, which is in agreement with measured values.^

     3.2.3. Alkalinity and pH Relationships

Equation 3-3A (Table 3-1) shows that alkalinity is destroyed by the oxidation of ammonia
and carbon dioxide (H2CO3 in the aqueous phase) is produced. When synthesis is neglected,
it can be calculated that 7.14 mg of alkalinity  as CaCOs is destroyed per mg of ammonia
nitrogen oxidized. The effect of synthesis is relatively small; in Equation 3-8, the ratio is
7.07 mg of alkalinity per mg of ammonia nitrogen oxidized.  Experimentally determined
ratios are presented in  Table 3-2; differences between the experimental and theoretical
ratios are due either to errors in alkalinity or nitrogen analyses or the inadequacy of theory


                                   TABLE 3-1
       RELATIONSHIPS FOR OXIDATION AND GROWTH IN NITRIFICATION
        REACTIONS IN RELATIONSHIP TO THE CARBONIC ACID SYSTEM
     Reaction
                   Equation
                                                 Equation
                                                   No.
 Oxidation -
    Nitrosomonas

 Oxidation -
    Nitrobacter

 Oxidation -
    overall

 Synthesis -
    Nitrosomonas
 Synthesis -
    Nitrobacter
NH. +1.50_+  2HCO
   4       £
NO. +0.5O.-*-NO
                        NO, + 2H.CO, + H.O
                           L      Z   O    £t
              2HCO  -^
13NH4 + 23HCO3
3-1A
                                                   3-2
                                                   3-3A
                                                   3-4A
       10NO
                                                   3-5A
                                      3-4

-------
                                    TABLE 3-2
        ALKALINITY DESTRUCTION RATIOS IN EXPERIMENTAL STUDIES
System
Suspended growth
Suspended growth
Suspended growth
Attached growth
Attached growth
Attached growth
mg alkalinity destroyed
' mg NH~4-N oxidized
6.4
6.0
7.1
6.5
6 . 3 to 7 . 4
x .7.3
Reference
5
6
7
8
' ' 9 '
2
   As CaCO  ; the theoretical value is 7.1
           O

to completely explain the phenomenon.  A ratio of 7.14 mg alkalinity destroyed per mg of
ammonia nitrogen oxidized may be used for engineering calculations.

These changes may have a  depressing effect on  pH in the nitrification system, as the
relationship for pH in the system is:
                        pH  =  pKj  -  log
(3_9)
Since nitrification reduces the HCOs level and increases the H2CO3 level, it is obvious that
the pH would tend to be reduced. The effect is mediated by stripping of carbon dioxide
from  the liquid  by the process of aeration and the pH is elevated upwards. If the carbon
dioxide is not stripped from the liquid, such as in enclosed high purity oxygen systems, the
pH can be depressed as low as 6.0.  It has been calculated that  to maintain the pH greater
than 6.0 in an enclosed system, the alkalinity of the wastewater must be  10 times greater
than the amount of ammonia nitrified.2

Even in open  systems  where the carbon  dioxide is continually stripped from the liquid,
severe  pH  depression can occur when the  alkalinity  in  the  wastewater approaches de-
pletion by the acid produced in the nitrification process.  For example,  if in a wastewater
20 mg/1 of ammonia nitrogen is nitrified, 143 mg/1 of alkalinity as CaCOs will be destroyed.
In many wastewaters  there is insufficient alkalinity initially  present to leave a sufficient
residual for buffering the wastewater during the nitrification process. The significance of pH
depression in  the process is that nitrification rates are rapidly depressed as the pH is reduced
                                        3-5

-------
below 7.0 (see Section 3.2.5.6). Procedures for calculating the operating  pH in aeration
systems are presented in Section 4.9.

     3.2.4 Oxygen Requirements

The theoretical oxygen requirement for nitrification, neglecting synthesis, is 4.57 mg O2/mg
NH^-N  (Equation  3-3).  Synthesis has  an effect  on  oxygen  requirements; the oxygen
requirement is calculated  to be, from Equation 3-8, 4.19 mg C>2/mg NH^-N. An oxygen
requirement  sufficiently  accurate  to be  used in engineering calculations  for aeration
requirements is 4.6 mg O2/mg NH^-N.

The oxygen  demand for  nitrification is significant;  for instance if 30  mg/1 of ammonia
nitrogen is oxidized by the nitrification system, about 138 mg/1  of oxygen will be required.
Caution: in virtually  all practical nitrification systems, oxygen  demanding materials other
than ammonia are  present  in  the  wastewater,  raising the total oxygen requirements of
nitrification systems even higher (see Section 4.8).

      3.2.5 Kinetics of Nitrification

A complete review of the kinetics of biological systems is beyond the scope of the manual;
however,  several excellent reviews are available. 10,11,12  Rather,  the basics of biological
kinetics are drawn upon to  usefully portray a mathematical description of the oxidation of
ammonia and nitrite. In the  succeeding portions of this section, the impact of a variety of
environmental factors on  the rates  of growth and nitrification are considered. A  combined
kinetic  expression  is then formulated which  accounts for  the  effects  of  ammonia
concentration, temperature, pH and dissolved oxygen concentration.

At several points, reference  is made  to data developed from various types of nitrification
processes.  Comprehensive descriptions of the various nitrification processes are presented in
Chapter  4 and  will  not  be reproduced herein. One  distinction that needs to be clearly
understood in discussions  in  this chapter  is the difference  between  combined carbon
oxidation-nitrification processes and  separate stage nitrification processes. The  combined
carbon  oxidation-nitrification  processes oxidize  a high  proportion of influent organics
(BOD) relative to the ammonia nitrogen content. This  causes relatively low populations of
nitrifiers to be present in the biomass.  Separate stage nitrification systems, on the  other
hand, have a relatively low BODs  load relative to  the  influent  ammonia load. As a result,
higher proportions  of nitrifiers are  obtained. Separate stage nitrification can be provided in
municipal  treatment  applications when  a high level of organic  carbon removal is provided
prior to the nitrification stage. This level of treatment is generally greater than provided by
primary treatment. Other differences between these classes of processes  can be drawn, but
these are left for detailed discussion in Section 4.2.

          3.2.5.1 Effect of Ammonia  Concentration on  Kinetics

A description of ammonia and nitrite oxidation can be derived  from an examination of the
                                         3-6

-------
growth kinetics of Nitrosomonas and Nitrobacter. Nitrosomonas' growth is limited by the
concentration  of  ammonia  nitrogen,  while  Nitrobacter 's growth is limited by  the
concentration of nitrite.

The kinetic equation proposed by Monodl3 is used to describe the kinetics of biological
growth of either Nitrosomonas or Nitrobacter:
                     M
                        K  + S
                          s
where:     M   =    growth rate of microorganisms, day   ,
           A                                                 _1
           M   =    maximum growth rate of microorganisms, day  ,
           K   =    half velocity constant =  substrate concentration, mg/1,
                     at half the maximum growth rate and
           S   =    growth limiting substrate concentration, mg/1.

Since the maximum  growth rate of Nitrobacter is  considerably larger than the maximum
growth rate of Nitrosomonas, and since Ks values for both organisms are less than 1 mg/l-N
at temperatures  below 20 C, nitrite does not accumulate in large amounts  in biological
treatment systems under steady state conditions. For this reason, the rate of nitrifier growth
can be modeled  with Equation 3-10 using  the rate limiting step, ammonia conversion to
nitrite.  For  cases   where  nitrite  accumulation   does  occur,  other  approaches  are
available. 14, 15, 16

          3.2.5.2 Relationship of Growth Rate to Oxidation Rate

The ammonia oxidation rate can be related to the Nitrosomonas growth rate, as follows:
where:      MXJ  =   Nitrosomonas  growth rate, day  ,
            H~,  =    peak Nitrosomonas growth rate, day  ,
                     MN                                         +
            q^T  =    -y—   =   peak ammonia oxidation rate, Ib NH. - N
                       N        oxidized/lb VSS/day,
            q-^  =    ammonia oxidation rate, Ib NH4 - N oxidized/lb VSS/day

            Y^  =    organism yield coefficient, Ib Nitrosomonas grown (VSS) per
                     Ib NH4 - N removed,

            N   =    NH^ - N concentration, mg/1, and

            KN  =    half-saturation constant, mg/1 NH4  - N, mg/1.

                                        3-7

-------
In Equations 3-10 and 3-11, only the effect of ammonia concentration is considered; in later
sections, the effects of temperature, pH, and dissolved oxygen are considered.

          3.2.5.3. Relationship of Growth Rate to Solids Retention Time

The growth rate of organisms can be related to  the design of activated sludge systems by
noting the inverse relationship between solids retention time and growth rate of nitrifiers:

                                    0   =   4_                               (3-12)
                                     c       M

where:      6   =   solids retention time, days.


The solids retention time can be  calculated from system operating data by dividing the
inventory of microbial mass in the treatment system by the quantity of biological  mass
wasted daily. Equations applicable for this calculation are presented in Section 4.3.3.

          3.2.5.4 Kinetic Rate Constants for Temperature and Nitrogen Concentration

The most widely accepted kinetic constants for the nitrifers are those presented by Downing,
and coworkers. 1 ' > * ° Their results are presented in Figures 3-1 and 3-2. As can be seen,  both
the maximum growth rate, /i, and the half saturation constants, Ks for Nitrosomonas and
Nitrobacter are markedly affected  by temperature. Further, the maximum growth rate for
Nitrosomonas in activated sludge was found to be considerably less than for Nitrosomonas
in pure culture.

Kinetic constants found by other investigators are summarized in Tables 3-3 and 3-4. The
observations   of  maximum  growth  rates  of Nitrosomonas  of  Gujer  and Jenkins %
Wuhrmann^ Loehr, et al.H, Poduska and Andrews^, and Lawrence and  Brown24; are
closer to the pure  culture values of Nitrosomonas rather than the activated sludge values in
Figure 3-1. This suggests that some additional parameter such as dissolved oxygen (DO) may
have  been limiting Downing's activated sludge  measurements. ^ The influence  of DO  is
discussed in the next section. For  illustrative use in this manual the pure culture values of
Downing, et al.  for Nitrosomonas are used for considering the effect of temperature and
ammonia on growth and nitrification rates. The expressions for the half saturation constant
for oxidation of ammonia-nitrogen, KN, is:
                                                   sN                        (3-13)

                   T   =  temperature, C.

The expression for the effect of temperature on the maximum growth rate of Nitrosomonas

                       2N  =  0.47e°-098 day'1                         (3-14)

                                         3-8

-------
                            FIGURE 3-1
   4.0

r  3.0

2  2.0
             TEMPERATURE DEPENDENCE OF THE MAXIMUM
                   GROWTH RATES OF NITRIFIERS
lu
O
(t
   1.0

   °-8
   O.6
   0.4
 $
 S
 X  O.2
 S
< a
   O.I
               I
I
I
                                                  0.098 (T-15)    —
                                       = O.I8e
                                             O.II6(T-I5)
                         I
               12
                         16         20
                         T, TEMPERATURE, C
                                             24
                              28
                                                                32
                             FIGURE 3-2
          TEMPERATURE DEPENDENCE OF THE HALF SATURATION
                      CONSTANTS OF NITRIFIERS
                         16         2O
                         T, TEMPERATURE, C
                                                       28
                                                                32
                                3-9

-------
                                  TABLE 3-3
                  MAXIMUM GROWTH RATES FOR NITRIFIERS
                         IN VARIOUS ENVIRONMENTS

Organism

Nitrosomonas









Nitrobacter

a.. , day- 1 at stated temperature , C
' JN
8









0.25


12
0.40
0.34










IS






0.21



0.28

16


0.57









20





0.71
0.48


0.5


. 21
0.85
0.65








0.34

23




0.37



1.08


1.44
25



0.17


0.55



0.53


prtf
Kei .
4
4
19
20
21
22
11
23
15
24
11,23
15

r i
tiii vironmsnt
Activated sludge, wash out
Activated sludge, math model
Activated sludge
Activated sludge
Activated sludge
Activated sludge
Synthetic river water

Activated sludge
Activated sludge
Synthetic river water
Activated sludge
                                  TABLE 3-4
                     HALF-SATURATION CONSTANTS FOR
                    NITRIFIERS IN VARIOUS ENVIRONMENTS
Organism
Nitrosomonas







Nitrobacter






Ks, mg/l-N at stated temperature, C
15

2.8

0.5 to 1.0





0.7





20

3.6

0.5 to 1.0

1.0

0.5

1.1




0.07
25
0.37
3.4


3.5



0.25
0.7



5

28











5



30


10







6




32












8.4


Ref.
20
11,23
1,25
2
1,26
1,27

28
20
11,23
1,29
1,30
1,31
1,26
28
Environment
Activated sludge
Synthetic river water
Lab culture
Warburg analyses
Lab culture
Lab culture and activated
sludge
Lab culture
Activated sludge
Synthetic river water
Lab culture
Lab culture
Lab culture
Lab culture
Lab culture
Somewhat differing temperature effects  have been found  for attached growth systems.
Huang and Hopson's summary, with some modifications, is shown in Figure 3-3 for attached
growth systems.34 Downing and coworkers' relationship for Nitrosomonas (Equation 3-14)
is also shown for comparison purposes.  Comparing the  suspended growth and attached
nitrification data, one can conclude that attached growth systems have  an  advantage in
withstanding low temperatures (<15 C) without as  severe losses in nitrification rates.
However, measurement of nitrification rates for suspended growth systems are not normally
made on the same basis as attached growth systems. In suspended growth systems, rates are
                                     3-10

-------
expressed on a per unit of biomass basis (MLVSS is used). Precise measurement of biomass
is normally not possible  in attached growth systems so other parameters are used such as
reaction rate  per unit  surface  or volume.  Therefore,  attached  growth  systems  can
compensate  for colder temperature  conditions by the  effective slime growth growing
thicker.  Thus,  if rates  could be expressed on  a unit biomass basis for both system types,
reaction rate variation with temperature might be more similar.

It could be argued that compensation for low temperature in  suspended growth systems
could be provided by an increase in mixed liquor level, much as  an increase in slime growth
occurs in  attached growth systems. However, suspended growth systems are limited by
reactor-sedimentation tank interactions which at cold temperatures might prohibit this due
to reduction of thickening rates of the sludge (cf. Section 4.10).

Other differences in  reaction rates shown in Figure 3-3 may arise from the fact that some
determinations were on  separate  stage nitrification systems while others were made on
combined carbon oxidation-nitrification processes.

                                   FIGURE 3-3
o
o
 fc
 o
 
-------
         3.2.5.5 Effect of Dissolved Oxygen on Kinetics

The concentration of dissolved oxygen (DO) has a significant effect on the rates of nitrifier
growth and nitrification in biological waste treatment systems. The Monod relationship has
been  used  to model  the  effect  of dissolved oxygen, considering oxygen to  be a. growth
limiting substrate, as follows:

                                    D0
                         =  u
                            M
                             N K0   + DO
                                                                              (3-15)
where:      DO  =   dissolved oxygen, mg/1, and
            Kn  =   half-saturation constant for oxygen, mg/1.
             U2

British investigators found  that  the KQ2 value was  about 1.3 mg/1  at an  unspecified
temperature. 3 5 One U.S. investigator  has suggested a relationship that would indicate
half-saturation constants of 0.15 mg/1 at 15 C and 0.42 mg/1 at 25 C, but did not provide
supporting data. 14 Studies conducted by the  Los Angeles County Sanitation Districts at its
Pomona Water Renovation Plant represent one of the most careful attempts to evaluate the
effect  of  DO  on nitrification  rate.3°  This facility  is  a  combined carbon  oxidation-
nitrification plant. Sludge samples were  withdrawn, dosed with ammonia, and aerated at
various DO levels. Nitrification rates determined from the data collected are shown in Figure
3-4.36 Fitted to this data  is  a Monod expression for nitrification rate  as a function of DO.
The KQO  determined from  this data is 2.0 mg/1. Temperature was not specified, but
indicated to be above 20 C.

Several investigations have provided indirect evidence of the importance of the effect of DO
on  nitrification rate. A treatment plant  operated continuously at a DO near  1 mg/1 gave
lower degrees of nitrification than plants held at 4 and 7 mg/1.3 ' When small scale activated
sludge plants were held at 1,2, 4, and 8 mg/1.  British investigators found that the nitrifi-
cation rates at 2.0 mg/1 were about 10 percent lower than at higher levels of DO, although
nitrification was complete. 35  pilot investigations at the Metro Sewer District of Cincinnati,
Ohio, showed that when the  DO was held at 2 mg/1,  only  about 40 percent nitrification
occurred,  but when the DO  was  increased to 4 mg/1, about 80 percent nitrification took
place. 3°   Murphy found that in two parallel  activated sludge plants, that nitrification was
enhanced in the plant maintaining the DO at  8 mg/1 compared to the plant where the DO
was maintained at  1 mg/1. 3 9

The influence of DO on nitrification rates has been somewhat controversial, as examples of
plants can  be found with completely nitrified  effluents with  operating DO levels of 0.5
mg/1. However,  this type of evidence does not indicate that nitrification rate was unaffected,
merely that nitrification  could be completed in the  presence  of  a low DO  level. Low
nitrification rates, depressed by low DO levels, can still be sufficient to cause complete
nitrification if the aeration tank detention time is large enough.

                                        3-12

-------
                                        FIGURE 3-4
                EFFECT OF DISSOLVED OXYGEN ON NITRIFICATION RATE
 Ul
 ^    0.15
 Q:
 S  Q
 9\
 *  V)  0.10
 O  tn
lU ^
Q 2
* \
~ ^ 0.05
    3
        o
                                                   °	)
                                                  .0 + DO /
                                    Data of   Nagel  and Haworth (Ref.  36)

                                  I            I            I            I
                      O.5         1.0         1.5         2.O
                             DO, DISSOLVED  OXYGEN,  mg/l
                                                                     2.5
3.0
While the general effect of DO on kinetics is firmly established, there needs to be further
study to determine the factors affecting the values of KQ2- All of the various estimates are
from systems where combined carbon oxidation-nitrification is practiced and no measure-
ments have been made on separate stage nitrification systems.  Kc>2 values for separate stage
nitrification  systems  may very  well be  different than  those  for  combined  carbon
oxidation-nitrification  systems. Further refinement  of KQ2  values can be expected.  For
illustrative use in this manual, a value of KQ2 °f 1 -3 rng/1 has been assumed. This value  falls
in the middle of the range of KQ2 observations (0.15 to 2.0 mg/l) and is of a magnitude
such that if the operating DO is 2.0 mg/l or less, the nitrification (or nitrifier) growth rate is
60.6 percent (or less)  of the  peak rate. This order of reduction in rate could  account for
most of the  difference in  growth rate observed by  Downing, et al.  (see Section 3.2.5.4)
between river water values at high DO levels and activated sludge operating data at DO levels
of 2.0 mg/l of less.

          3.2.5.6 Effect of pH on Kinetics

The  hydrogen ion concentration (pH) has  been generally found to have a strong effect on
the rate of  nitrification. Figure 3-5 presents typical pH relationships  from a number of
investigations. The results  of other  investigations have been  summarized in  the  litera-
ture. 1>44 There is a wide range in reported pH optima; the almost universal finding is that as
the pH moves to the acid  range, the rate of ammonia  oxidation declines. This has been
found to be true for both unacclimated and acclimated cultures,  although acclimation tends
                                        3-13

-------
to moderate pH effects. The findings for an attached growth reactor (Curve E, Figure 3-5)
are very similar to the findings for an activated sludge (Curve C). In neither case were the
cultures  acclimated  to each pH value prior  to  determining nitrification rates.  When a
three-week acclimation period was provided for the attached growth reactor, it was found
that the rate at pH 6.6 rose to  85 percent of the optimum rate at pH 8.4 to 9.0.34

Various  investigators have  reported  the effects of pH depression on nitrification. For
instance, in an activated sludge with insufficient wastewater alkalinity, pH values of 5 to 5.5
were attained. This high acid  concentration resulted  in a cessation of nitrification; at the
same time sludge bulking occurred. The point at which the rate of nitrification decreased
was  pH  6.3-6.7.1  Parallel  investigations on air  and  high  purity oxygen-activated sludge
systems at the  Blue Plains Treatment Plant, Washington, D.C., have shown that depressed
pH values in the last oxygen activated sludge stage produced slightly lower nitrification rates
than when the  system was operated at higher pH.45  Further information on the EPA
investigations at the Blue Plains Treatment Plant is presented later in Section 4.6.5.

In a study of the  effect of abrupt changes in pH, it was found that an abrupt change in
reactor pH from 7.2 to 6.4 caused no adverse  effects. However, when the pH was abruptly
changed  from  7.2  to 5.8, nitrification performance deteriorated  markedly as  effluent
ammonia levels rose from approximately zero to 11 mg/1 NH^-N. A return to pH 7.2 caused
rapid improvement indicating that the lowered pH was only inhibitory and not toxic.'^
Haug and McCarty showed that nitrifiers could adapt to nitrify at pH levels as low as 5.5 to
6.0.2 However, since the concentration of biomass in  their column was not defined at each
pH level, no conclusions can be drawn from their work as to the  effect of pH on the peak
ammonia oxidation rate, qj^f.

For illustrative use in this manual, the equation of Downing, et alA3, showing the effect of
pH on nitrification is adopted. For pH values < 7.2:

                MN  = £N (1  - 0.833(7.2  - pH))                              (3-16)

For pH  levels  between 7.2 and 8.0,  the  rate is assumed constant. This  expression was
developed  for  combined  carbon oxidation-nitrification systems,  but its application  to
separate stage nitrification systems is probably conservative.

Because  of the  effect of pH on  nitrification  rate, it  is especially important that there be
sufficient alkalinity in the wastewater to balance  the  acid produced by nitrification.
Equation 3-3A (Table 3-1)  indicates  that 7.14 mg of alkalinity  are destroyed per mg of
ammonia nitrogen. Caustic or  lime addition may be required to supplement moderately
alkaline wastewaters. Design considerations for pH control are presented in Section 4.9.

          3.2.5.7 Combined Kinetic Expressions

In previous sections, the effect of ammonia level, temperature, pH and dissolved oxygen on
nitrification rate has been presented.  In all practical systems, these parameters act to affect

                                        3-14

-------
                                   FIGURE 3-5

                     EFFECT OF pH ON NITRIFICATION RATE
                            (AFTER SAWYER, ET AL.)
v, ^ PERCENT OF MAXIMUM
i m RATE OF NITRIFICATION
» -<
O f\> •& O> Q> Q
r o» ° O O O O O
i i i i
/ .
- / /
r /
-/ (vf> '
\y 7 '?
^Js?
* \ i i i

/'
j^ •
^ff •
r^f *
f *
f •
•
»
B:
*
i i /i i
^7 "^
*


i i 11
vS>' ' "
\\v
• X .
• Nk \
• \. ^
* >^
_9 ^
*
:B
i i: i i

O 7.O 8.0 9.O 10.0
ENVIRONMENT REFERENCE
A Nitrosomonas- pure culture Engle and Alexander ( 40)
B Nitrosomonas - pure culture Myerhof (41)
C Activated sludge at 20 C Sawyer, et al. (42)
D Activated sludge Downing, et al. (43)
E Attached growth reactor at 22 C Huang and Hopson (34)
the nitrification rate simultaneously. It has been shown that me combined effect of several
limiting factors on  biological growth  can be introduced  as a product of Monod-type
factors: 46
                                       N
               K
                                     N
                                     Lv
                                          N
                                                                          (3-17)
where:      L   =
            N   =
            P   =
KL,KN,andKp =
concentration of growth limiting substance L,
concentration of growth limiting substance N,
concentration of growth limiting substance P, and
half-saturation constants for substances L, N, and P, respectively.
                                      3-15

-------
This concept has also been applied to the analysis of algal growth kinetic data^7 and to
denitrification kinetics.'*"

Taking this approach for nitrification, the  combined kinetic expression for nitrifier growth
would take the form:
                                                   (1 - 0.833(7.2 - PH))      (3-18)
where:
                =   maximum nitrifier growth rate at temperature, T, and pH.
Downing, et al. 43 used this procedure to describe nitrifier growth rates, excepting that no
term for DO was included. Using specific values for temperature, pH, ammonia and oxygen,
adopted  in this manual in Sections 3.2.5.4, 3.2.5.5., and 3.2.5.6, the  following expression
results for pH < 7.2 for Nitrosomonas valid for temperatures between 8 and 30 C:
                    = 0.47
                                                            2 _pR)
                                  N
                       N +  10
                              0.051T -  1.158
                                                      DO
                                                   DO +  1.3
                                                                 day
-1
                                                                             (3-19)
The first term in brackets accommodates the effect of temperature. The  second term in
brackets considers the effect of pH. For pH > 7.2, the second quantity in brackets is taken
to be unity. The third term in brackets is the Monod expression for the effect of ammonia
nitrogen concentration. Similarly, the fourth term in brackets accounts for the effect of DO
on nitrification rate. Equation 3-19 has been adopted for illustrative use in  this manual. As
more reliable  data  becomes  available,  Equation 3-19 can be modified to suit particular
circumstances.
An example evaluation of Equation 3-19 at T = 20 C, pH = 7.0, N = 2.5 mg/1, and DO equal
to 2.0 mg/1 is as follows:
               MN  =  0.47 [1.63] [.833]  [0.775]  [0.606]  =  0.300  day
                                                                        "1
The ammonia removal rate is defined as done previously (Equation 3-11):
 qN  =
                                              DO  ) 0  - °-833<7-2 -
                                        3-16

-------
where:      Q    =   -~—                                                    (3-21)
In the numerical example above, with a yield coefficient of 0.15  Ib VSS  per Ib NIfy-N
removed,  the  nitrification rate is 2 Ib NH^-N oxidized per Ib VSS per day. This rate is
expressed per  unit of nitrifiers, assuming that there are no other types of bacteria in the
population. Nitrification rates of a comparable magnitude have been found experimentally
by a number of investigators for laboratory enrichment cultures of nitrifiers.2,23,28 AS wjn
be seen in Section 3.2.7, nitrification rates in mixed cultures are much lower.

     3.2.6 Population Dynamics

In previous  sections, the kinetics of the growth of nitrifiers have  been presented. In all
practical  applications  in wastewater  treatment, nitrifier growth  takes place  in waste
treatment processes where other types of biological growth occurs. In no case  are there
opportunities for pure cultures to develop. This fact has significant  implications in process
design for nitrification.

In both combined carbon oxidation-nitrification systems and in separate stage nitrification
systems,  there is sufficient organic matter in the wastewater to  enable  the growth of
heterotrophic bacteria. In this situation, the yield of heterotrophic  bacteria growth is greater
than the yield of the autotrophic nitrifying bacteria. Because of this dominance of the culture,
there is the danger that the growth rate of the heterotrophic organisms will be established at
a value exceeding the maximum possible  growth rate of the nitrifying organisms. When this
occurs, the  slower growing nitrifiers will gradually diminish in  proportion to the total
population and be washed out of the system.43

Thus, for consistent nitrification to occur, the following design condition must be satisfied,
assuming pH and DO do not limit nitrifier growth:
                                   A
                                   MN > Mb                                     (3-22)
where:      JUN  =   maximum growth rate of the nitrifying population,
            ju,    =   growth rate of the heterotrophic population.
Reduced DO or pH can act to depress the peak nitrifier growth rate and cause a washout
condition. An expression for this possible reduced rate of growth is:

                   „   =        - DO - ^  (1 _ 0.833(7.2 - pH))         (3-23)
                   MN     MN  I  K     +  DO
                                        3-17

-------
where:      MXT  =    maximum possible nitrifier growth rate under environmental
                     conditions of T, pH, DO, and N ^K.
As before, the last expression in brackets is  taken  to be unity above a pH of 7.2. The
relationship between actual nitrifier growth rate and maximum possible nitrifier growth rate
can easily be seen from Equations 3-8 and 3-23:

                                •  /    N    \                               ,o ^A-\
                         "N  =  *N ^K^TN j                               (3'24)

A more rigorous condition for prevention of nitrifier washout than Equation 3-22 is:


                              M  ^ M                                        (3-25)
More  typically, Equation 3-25 is expressed in reciprocal form in terms of solids retention
time as: 1 1

                                0d ^ em                                    (3-26)
                                                                             ^
where:      6    =    solids retention time of design, days, and
             L/

            6    =    minimum solids retention time, days, for nitrification at given pH,
                     temperature and DO.
Since n or#c is fixed by the environmental conditions (T, pH and DO), Equations 3-25 or
3-26 is satisfied by modifying/^ or.0c. The various ways of satisfying these relationships can
be established  by examining  the  following  growth relationship for the heterotrophic
population ;10,11,12



                                 c

where:      Y,  =    heterotrophic yield coefficient, Ib VSS grown per Ib of substrate
                     (BOD or COD) removed, and
            q,   =    rate of substrate removal, Ib BOD (or COD) removed/lb VSS/day,
                                           -1
            K,  =    "decay" coefficient, day  , and

            ju«   =    net growth rate for heterotrophic population.


                                       3-18

-------
 The rate of substrate removal is defined as:
                                    so -  si
 where:      SQ   =   influent total BOD (or COD), mg/1,
             Sj   =   effluent soluble BOD (or COD), mg/1,
             HT  =   hydraulic detention time, days, and
             Xj   =   MLVSS, mg/1.

                                                                             d
 Since both  Yfc  and Kd are assumed to be constant,  the only way /^ or  6 c can be
 manipulated is by  altering  q^.   One way the substrate removal rate can  be reduced is
 to place an organic carbon removal step ahead of the nitrificaiian stage, creating a "separate
 stage" nitrification  process.  The result of this procedure is to reduce the food available to
 the heterotrophic bacteria and to lessen their dominance in controlling the solids retention
 time. Separate nitrification stages can have very long solids retention times (  flf. = 15 to 25
 days). Another procedure for reducing the substrate removal rate, without  separating the
 carbon oxidation and nitrification processes, is  to increase the  biological solids  in  the
 system. This can be done by increasing the concentration of biological solids under aeration
 (the MLVSS in  the activated sludge system)  or by increasing the volume of the oxidation
 tank while maintaining the concentration  of biological solids at the same concentration.

 The  level  set for biological solids retention time,  6 c  , establishes  the biological solids
 retention time  (or growth rate) of the nitrifiers, since selective wasting of the  heterotrophic
 population is not practical. Therefore, the design solids retention time can be related to the
 effluent ammonia level through the Monod relationship and the inverse relationship between
 nitrifier  growth  rate and solids retention time (Equations 3-12 and 3-24).  With specific
 values of  0  £  >  T,  pH  and  DO, Equations 3-12  and 3-24 can be solved for the effluent
 ammonia level. Figure 3-6 was developed by such a  procedure, with the assumption of T =
 15 C, DO  =  2 mg/1, and pH  >7.2 <8. Also plotted is the nitrification efficiency, assuming
 an influent Total Kjeldahl Nitrogen  (TKN) concentration of 25 mg/1 and that  all of the
 influent nitrogen is available for nitrification. As can be seen, significant breakthrough of
 ammonia from the system does not occur  unless the solids retention time is reduced below 5
 days.  At  that point, ammonia nitrogen breakthrough is  very abrupt,  rising from  1  mg/1 at
 6 c  = 4.9 days to 15 mg/1 at  6 c  = 3.6  days. The principal cause of the sharpness of the
 ammonia breakthrough is due to the low value of KN (0.40 mg/1 NHlj. — N in this case). A
 number  of investigators have experimentally  determined very similar relationships to that
 shown in Figure 3-6 J 9, 20, 21

 Lawrence  and  McCarty^l have introduced the concept of a  safety factor (SF) in the
• application of biological treatment process kinetics to design. The safety factor was defined

                                          3-19

-------
as the ratio of the design solids retention time to the minimum solids retention time; the
safety factor can also be related to the nitrifier growth rates through Equation 3-12. The
expression for the SF is:
                              SF  =
           (3-29)
A conservative safety factor was recommended to minimize process variations caused by pH
extremes, low DO concentrations and toxicants. 11 Also, the SF can be used to ensure that
ammonia breakthrough does not occur during diurnal peaks in load (see Section 4.3.3).

Interestingly, Equation 3-24 can  be manipulated to show that the specification of a SF of 2
will establish an  ammonia  level equal  to KN in the effluent of a complete mix activated
sludge plant, if there is a high DO level (DO not limiting). This is because the process will be
operating at one-half its maximum growth rate. Recall that the half-saturation constant, KN,
in this Monod expression is  defined as the level of substrate which will cause the organism to
                                   FIGURE 3-6

          EFFECT OF SOLIDS RETENTION TIME ON EFFLUENT AMMONIA
               CONCENTRATION AND NITRIFICATION EFFICIENCY
25



20





\ IS
o>
6

-
z
i
2 '°
2
Ul
-J
I1- 5
u.
UJ


D


—

—



i-

—



—


—


_


-


1 x*» "" " 1 | 1 I 1
s
^.
	 PERCENT REMOVAL
— *~ —



—

—


•
—







/-EFFLUENT AMMONIA
\ / ~~
v/
r* — . — ' 	 * ' ' i
IUU

90
2
80 £
tt
iu
0.
70
>?
60 2
0

u.
SO u.
Uj

40 o
30 i
U.
20 t

*
IO


                         IO        15        2O         25
                     DESIGN SOLIDS  RETENTION  TIME,  DAYS
3O
35
                                       3-20

-------
 grow at half its maximum rate. Many nitrifying activated sludge plants have been observed
 to have 1 to 2 mg/1 of ammonia in their effluents, values close to the theoretical value of
 KN-

 Criteria for establishing the safety factor are presented in Chapter 4. Furthermore, specific
 examples of the use of the kinetic expressions developed in this section are presented in
 Sections 4.3.3. and 4.3.5.

      3.2.7 Nitrification Rates in Activated Sludge

 The basic design approach for separate stage nitrification systems (activated sludge type) has
 been on a different basis than for combined carbon oxidation-nitrification systems. Rather
 than use the sludge growth rate or solids retention time approach described in Section 3.2.6,
 the practice frequently has been to base reactor sizing for separate stage systems on the basis
 of nitrification rates in terms of Ib NH^-N oxidized/lb MLVSS/day.5>42,49 However, it will
 be shown that this parameter is fundamentally related to the nitrification kinetics previously
 presented in this chapter.

 The nitrification rate can  be   calculated  from  the ammonia oxidation  rate,  q^,  by
 recognizing that the nitrifiers are  only a  fraction of the total mass of biological solids in a
 nitrification system. The  other biological solids in  the system result from the growth of the
 hetero trophic population. On this basis, the nitrification rate, r^, is as follows:
                                                                                 (3-30)
                                                                                 (3-3D
 where:      f    =    nitrifier fraction of the mixed liquor solids, and

             rN  =    nitrification rate, Ib NH* - N oxidized/lb MLVSS/day.


 A similar expression for the peak nitrification rates (in activated sludge) is:

                               /\      A     „
where:      rN  =    peak nitrification rate, Ib NH. - N oxidized/lb MLVSS/day.
This latter rate is normally determined experimentally in activated sludge systems, as will be
described in Section 4.6.3.

Specific analytical techniques for determination of the nitrifier fraction have not as yet been
developed. However, f can be estimated from knowledge of the biological yields of the
auto trophic and heterotrophic populations, as follows:

                                         3-21

-------
                                     M
                             f  =
                                       N
                                                                            (3-32)
where:     MN =   nitrifiers grown through oxidation of ammonia, and
           M   =   heterotrophs grown through oxidation of organic carbon;
             c
MXT and M  can be estimated as follows:
  N      c

                             MN  =
                                                                            (3-33)
where:     N~  =
                    TKN in the influent, mg/1, and
                =   NH   - N in the effluent, mg/1.
                               Mc  =  Vso-V
                                                                            (3-34)
where:     Sn  =

           Sl   =
                    carbon (BOD5 or COD) in the influent, mg/1
                    carbon (BOD5 or COD) in the effluent, mg/1, and

                    net yield of VSS of heterotrophs per unit of carbon (BOD, or
                    COD) removed.
This procedure neglects the ammonia assimilated by heterotrophic growth and therefore is
approximate. A further approximation is that the net yield of the heterotrophs  has been
assumed constant, whereas it is known that it varies with solids retention time.

Several examples  can  be drawn  for separate  stage nitrification  systems  as compared to
combined carbon oxidation-nitrification systems. In a separate  stage system, illustrative
values of BOD5 removed and TKN oxidized would be 50 mg/1 and 25 mg/1, respectively. A
reasonable estimate  for YN is 0.15 Ib/lb NH^-N rem. (Section 3.2.2) and for Yb (BOD),
0.55 Ib/lb BOD5 removed. Thus,  for this separate stage example,  f can be calculated to be:
                           f  =
                                     0.15(25)
                                 0.55(50 + 0.15(25)
                                                    =  0.12
A similar example  can be drawn  for  combined carbon  oxidation-nitrification systems.
Assuming 200 mg/1 of BOD5 removed and 25 mg/1 of TKN oxidized, f can be calculated to
be:
                           f  =
                                    0.15(25)
                                0.55(200) + 0.15(25)
                                                    = 0.033
                                       3-22

-------
Thus,  it  can  be  seen  that the  fraction of  nitrifiers  is  lower in  combined  carbon
oxidation-nitrification systems that in separate stage nitrification systems.

Equation  3-32  can be reexpressed in terms of the BOD5/TKN ratio  in the influent, by
assuming effluent BOD and ammonia are negligibly small:
                              f =
                                           1
                                     !o_iL
                                     N0   YN
(3-35)
Table 3-5 presents numerical values for the fraction of nitrifiers, using Equation 3-35, and a
condition where YN = 0.15 and Y^ (BOD5) = 0.55. For most separate stage nitrification
systems, the BOD5/TKN ratio is greater than 1.0 and less than 3.0 (Section 4-2). Thus even
in separate  stage systems, the fraction of nitrifiers is relatively low.  For the assumed yield
values, the fraction is less than 20 percent and greater than 8 percent. It must be emphasized
that the values of nitrifier fraction given in Table 3-5 are estimates only, and not supported
by actual measurements of nitrifier fractions. Table 3-5 does have value in that it shows, at
least qualitatively, the impact of influent BOD5/TKN ratio on the fraction of nitrifiers in a
nitrification system.  The  influence of this ratio on nitrifier fraction and nitrification  rates
was recognized as early as 1940, when Sawyer showed that the BOD5/NH3 ratio correlated
with the nitrifying ability of various activated sludges.^O

The  effect  of ammonia concentration on the nitrification rate is portrayed in Figure 3-7.
The  assumptions made were: f = 0.1, pH > 7.2 <8, and the DO = 2.0 mg/1. Equations 3-20,
3-21, and 3-27 were employed to construct the figure. As can be seen, the nitrification rates

                                   TABLE 3-5
                RELATIONSHIP BETWEEN NITRIFIER FRACTION
                          AND THE BOD5/TKN RATIO
BOD /TKN ratio
o
0.5
1
2
3
4
Nitrifier
•a
fraction
0.35
0.21
0.12
0.083
0.064
BOD /TKN ratio
0
5
6
7
8
9
Nitrifier
fraction
0.054
0.043
0.037
0..033
0.029
    Using Equation 3-35 and YN=0.15, Y  = 0.55.
                                        3-23

-------
are  relatively unaffected by ammonia concentration above 2.5 mg/1 ammonia N.  As the
nitrification rates approach their plateau values, nitrification approaches a zero order rate,
uninfluenced by ammonia level. It has been shown that for suspended growth processes, the
rate of removal approximates a zero order reaction.42,45,51  However, in none of these
cases were nitrification rates determined at ammonia concentrations at or below the value of
    where non-zero order rates effects would be evident.
                                  FIGURE 3-7
        EFFECT OF AMMONIA CONCENTRATION ON NITRIFICATION RATE
     O.3
                                                        f = 0.1
                                                      D0= 2.0 mg/l
                                                      pH  >7.2 <8
                               10                      2O
                                    NH4-N,  mg/l
30
                                     3-24

-------
The  fraction of nitrifiers  has a marked effect on the nitrification rates.  Figure 3-8,
demonstrating this effect, was developed similarly to Figure 3-7, excepting that the effluent
ammonia nitrogen  concentration  was assumed to be 2.5  mg/1. A principal means of
increasing the nitrification rate is to increase the fraction of nitrifiers. From Table 3-5, it can
be seen that this can be accomplished by lowering the BOD5/TKN ratio. In terms of plant
                                   FIGURE 3-8

                 EFFECT OF TEMPERATURE AND FRACTION OF
                       NITRIFIERS ON NITRIFICATION RATE
              1.2
         CO
         00
         •J
N
Q
         I
        •s
         QQ
         -J
UJ

ft:

o

o
it
ftl
              1.O
             0.8
             0.6
             O.4
             0.2
             DO  =  2.0  mg/l
             NO  =2.5  mg/l  NH4-N
             pH  >7.2  <8
                10
                                          I
                                 20
                       TEMPERATURE,  C
30
                                     3-25

-------
 design, the BOD5/TKN ratio can be altered by increasing the organic carbon removal ahead
 of the nitrification unit.

 It must  be emphasized  that the nitrification  rates developed  in  this section are only
 estimated relationships based on theoretical  considerations.  Actual  measured values are
 presented in Section 4.6.3.

 As a practical example of the effect of the BOD5/TKN ratio on nitrification rates, rate data
 for an attached growth system^ are plotted against the BOD5/TKN ratio in Figure 3-9. The
 effect of BOD5 in the synthetic waste was to displace nitrifiers with heterotrophic bacteria in
 the  bacterial film,  thereby reducing the nitrifier fraction. and the nitrification  rate.
 Interestingly, a small amount of BOD5 ^10 mg/1)  was  found to enhance the nitrification ,
 rate.


                                   FIGURE 3-9

                EFFECT OF BOD5/TKN RATIO ON NITRIFICATION
             RATE - EXPERIMENTAL ATTACHED GROWTH SYSTEM
   100
o
II
 10
Q
O
03
Uj
u.
o
Uj
o

-------
     3.2.8 Nitrification  Rates in  Trickling Filters and  Other Attached  Growth Systems


Discussion of  kinetic  rates  in  the previous sections  has been  primarily  oriented  to
nitrification in activated sludge type (suspended growth) nitrification systems, although
some comparisons have been  drawn  with attached growth system  measurements for
comparative purposes.  The growth and oxidation rate relationships presented in Sections
3.2.5.7 and 3.2.6 are  directly  applicable'to design of suspended growth systems, as will be
shown in  Chapter 4. While these relationships are operative in attached growth systems, their
application is  complicated by the fact  that oxygen mass transfer limitations through the
bacterial slimes may  limit reaction rates in some situations.  A biofilm model has been
developed which yields insight into which factors are controlling^,53 bu{ jt ^as no{ yet
been extended to the point where it can be applied directly to design applications. As a
consequence, the design  relationships presented for attached growth nitrification in Chapter
4 are empirically  based  and therefore,  less  theoretically  precise  than those developed for
suspended growth sytems. Though the design relationships  presented are empirical,  where
possible the loading relationships are presented on a basis that is at least consistent with the
biofilm model. For instance, ammonia  nitrogen oxidation rates are expressed  on a surface
area basis when describing separate stage nitrification in trickling  filters and the rotating
biological disc process (Sections 4.7.1 and 4.7.2).

Some of  the  conclusions that  can  be drawn  from the  biofilm model  are of interest in
considering surface ammonia removal rates in attached growth  systems. The biofilm model
shows that the ammonia oxidation rate in attached growth systems should not  be decreased
as drastically under adverse environmental conditions as in suspended growth systems.52
This finding is consistent with the observation made in  Section 3.2.5.4,  namely that
attached growth systems have an advantage  over suspended growth systems in withstanding
lower temperatures without as severe losses in nitrification rates.

The biofilm model also  shows that  the dissolved oxygen  concentration must  be 2.7 times
the ammonia nitrogen concentration to prevent oxygen transfer from limiting nitrification
rates in attached growth systems.52 Two operational procedures have been suggested to
overcome this limitation: (1)  dilution of the ammonia nitrogen through recirculation and
(2) increasing the  oxygen transfer  through the  use of high purity oxygen.5 2  The first
recommendation has been made in this manual in Sections 4.4.1.4 and 4.7.1.3.

     3.2.9 Effect of Inhibitors on Nitrification

Certain heavy metals  and organic compounds are toxic to nitrifiers. To date,  these effects
have not  been  quantitatively  incorporated into the kinetic description of nitrifier growth,
although such approaches have been used to describe toxicity in other biological systems. A
listing  of  substances toxic to  unacclimated  nitrifying organisms is presented in Table 3-6,
which is drawn primarily from  the review by Painter. 1
                                        3-27

-------
Sawyer, on reviewing the English literature, suggests that 1 0 to 20 mg/1 of heavy metals can
be tolerated due to the low ionic concentrations at high pH values of 7.5 to 8.0.^6 it has
also been pointed out that precipitated metals (such as hydroxides) that concentrate in the
sludge  can  be  disastrous  to the sludge if the pH falls and the precipitate dissolves. Such
conditions  may  occur in the sludge collection zone of the secondary clarifier  where
continuing  organism activity may cause  low pH values. Alternatively, low pH values may
occur when pH control systems fail.
        found that silver (Ag) was extremely toxic to nitrification of secondary effluent on
a plastic media trickling filter. Levels of 2 ppb in the influent to the filter were concentrated
to 5 ppm in the biomass on  the media. This inhibitory effect was found to severely reduce
allowable loading rates and result in only partial nitrification.
                                    TABLE 3-6

                       COMPOUNDS TOXIC TO NITRIFIERS
                              (AFTER PAINTER (1))
                             Organics
                      Thiourea
                      Allyl-thiourea
                      8-hydroxyquinoline
                      Salicyladoxine
                      Histidine
                      Amino acids
                      Mercaptobenzthiazole
                      Perchloroethylene'3
                      Trichloroethylenec
                      Abietec acid0
Inorganics
  Zn
  OCN
-1
-1
  Cu
  Hg
  Cr
  Ni
  Ag
                         Also reference 54
                         Reference 5
                         Reference 55
The rate and change of magnitude of environmental conditions are nearly as critical to the
biomass as the conditions themselves. It has been shown that nitrifiers can adapt to toxic
substances when they  are consistently present at concentrations higher than cause toxic
effects in slug discharges. 1 >$& Unfortunately, slug discharges are often present in municipal
systems and can result from industrial dumps or from urban stormwater inflow.

Under the unusual conditions of discharge of highly  concentrated industrial wastes into
municipal systems  that contain either nitrite or ammonia, the resulting high concentrations
of ammonia or nitrite  in the municipal waste can be temporarily toxic to the  nitrifying
                                       3-28

-------
population. 5 9 When these conditions are suspected, the reader is referred to reference 59
which contains  charts which allow  the identification of the  regions of toxicity.  Under
normal municipal conditions of pH and concentrations complete nitrification will occur.
When concentrated industrial wastes are present, slug discharges should be avoided; rather,
storage facilities should  be provided so that  wastes  can be metered into  the collection
system at a rate sufficient to ensure dilution to safe loads.

In sum, the possibility of toxic inhibition must be recognized in the design of nitrification
systems. Either implementation of  source  control  programs  or inclusion of upstream
toxicity  removal processes may be required, particularly in those cases  where significant
industrial dischargers are  tributary to the collection system.

3.3 Denitrification

The biological process of denitrification involves the  conversion of nitrate nitrogen to a
gaseous nitrogen species. The gaseous product  is primarily nitrogen gas  but also may be
nitrous oxide or nitric  oxide.  Gaseous nitrogen  is  relatively unavailable  for biological
growth, thus denitrification converts nitrogen which may be in an objectionable form to one
which has no significant effect on environmental quality.

As  opposed  to  nitrification,  a  relatively  broad  range  of  bacteria  can  accomplish
denitrification, including Psuedomonas, Micrococcus,  Archromobacter and Bacillus. These
groups accomplish nitrate reduction by what is  known as a process of nitrate dissimilation
whereby nitrate or nitrite replaces oxygen in the respiratory processes of the organism under
anoxic conditions. Because of the  ability of these organisms to use either  nitrate or oxygen
as the terminal electron acceptors while  oxidizing organic matter,  these  organisms  are
termed facultative heterotrophic bacteria.

Confusion has arisen in the literature  in terminology; the process has been  termed anaerobic
denitrification. However, the principal biochemical  pathways are not anaerobic, but merely
minor modifications of  aerobic biochemical pathways. The term anoxic denitrification is
preferred, since it describes the environmental condition of the absence of oxygen, without
implying the nature of the biochemical pathways.

     3.3.1 Biochemical Pathways

Specific  information on the  specific  biochemical   reaction   intermediates  involved  in
denitrification are available in the literature^  and  only certain concepts  are of interest in
process design applications. Denitrification is a two-step process in which the first step is a
conversion of nitrate to  nitrite. The second step carries nitrite through two intermediates to
nitrogen gas. This two-step process is  normally termed  "dissimilation."

Denitrifiers  are also capable of an assimilation  process whereby nitrate (through nitrite) is
converted to ammonia. Ammonia  is then used for the bacterial cell's nitrogen requirements.

                                         3-29

-------
If ammonia  is already present,  assimilation of  nitrate  need  not occur to satisfy cell
requirements.

As  will be shown in Section 3.3.2, electrons pass from  the carbon  source (the electron
donor) to nitrate or nitrite (the electron acceptor) to promote the conversion to nitrogen
gas. This involves the nitrifiers "electron transport system" and is involved with the release
of energy from the carbon source for use in organism growth. It happens that this electron
transport  system  is identical to that used for respiration by organisms oxidizing organic
matter aerobically,  except  for one  enzyme. Because of  this close relationship, many
facultative bacteria  can shift between using oxygen  or  nitrate (or  nitrite) rapidly and
without difficulty.

     3.3.2 Energy and Synthesis Relationships

The use of oxygen as the final electron acceptor is more energetically favored than the use
of nitrate. Table 3-7 compares the energy yields per mole of glucose when oxygen and
nitrate are used as electron acceptors.61 The greater free energy released for oxygen favors
its use whenever  it is available. Therefore, denitrification  must be conducted in  an anoxic
environment to ensure that nitrate, rather than oxygen, serves as the final electron acceptor.

                                  TABLE 3-7

       COMPARISON OF ENERGY YIELDS OF NITRATE DISSIMILATION
                  VS OXYGEN  RESPIRATION FOR GLUCOSE
                          Reaction
                                                                  Energy yield
                                                                per mole glucose,
                                                                   kilocalories
Nitrate dissimilation
   5C-H,,O_ + 24KNO,-*- 30 CO0 + 18H0O + 24KOH + 12N
      D-l^D         *3          /       £
                                                                       570
Oxygen respiration
                                                                       686
   C.H.0OC + 600 -^6CO0 + 6H.O
     o  1 Z  b      £i        L      &
   Reference 61

Methanol, rather than  glucose or any other organic, has seen  widest use as the electron
donor in the U.S. (see Section 3.3.4). Using methanol as an electron donor and neglecting
synthesis for the moment, denitrification can be represented as a two-step process as shown
in Equations 3-36 and 3-37:
                                        3-30

-------
     First Step

          NO~ +  0.33CH3OH  =  NO~  +  0.67 H2O                          (3-36)

     Second Step

          NO~ +  0.5CH3OH  =  0.5 N2  +  0.5 CO2  +  0.5 H2O  +  OH~     (3-37)


The  overall transformation is obtained by addition of Equations 3-36 and 3-37 to yield
Equation 3-38:
                 0.833 CH3OH  = 0.5 N2 + 0.833 CO2 +  1.167H2O + OH~    (3-38)
In this equation, methanol serves as the electron donor and nitrate as the electron acceptor.
This can be shown by splitting up Equation 3-38 into the following oxidation-reduction half
reactions:

           NO"  +  6H+ + 5e~  - - 0.5 N2  +  3 H2O                     (3-39)
     (electron acceptor)
            0.833 CH3OH  +  0.833 H2O 	•-  0.833 CO2  +  5H+  +  5e      (3"40)

           (electron donor)
Equation 3-38  can  be obtained by adding Equations 3-39  and 3-40 and the following
equation for water:

                            H2O  =  H+ +  OH"
From Equations 3-39 and 3-40 the meaning of the terms of electron donor and acceptor are
clear. Nitrate gains electrons and is reduced to nitrogen gas, hence it is the electron acceptor.
The  carbon source, methanol, loses electrons and is oxidized to carbon dioxide, hence it is
the electron donor.

As mentioned in Section 3.2.2, these reactions take place in the context of the carbonic acid
system.  Equations 3-36 and 3-38 have been modified in Table 3-8 to reflect the fact that
hydroxide (OH~)  produced reacts  with carbonic acid  (carbon  dioxide) to  produce
bicarbonate  alkalinity.  Also  shown in Table 3-8  is the  equation  of synthesis for  those
organisms deriving energy through nitrate respiration.^^

                                       3-31

-------
The equations for energy yielding reactions (Equations 3-36 and 3-37) have been combined.
with the equation for oganism synthesis (Equation 3-41, Table 3-8) through knowledge of
organism yields and are summarized in Table 3-9. Also, shown for completeness is the
combined expression for oxygen respiration (Equation 3-44) since, if any oxygen is present,
it will be used preferentially. Similar expressions can be developed for other organic sources
serving as electron donors if organism yields are known.63

                                   TABLE 3-8

        RELATIONSHIPS FOR NITRATE DISSIMILATION AND GROWTH IN
                         DENITRIFICATION REACTIONS
           Reaction
            Equation
                                      Equation
                                        No.
  Nitrogen dissimilation
     Nitrate to nitrite
     Nitrite to nitrogen gas
     Nitrate to nitrogen gas
  Synthesis - denitrifiers
NO, +0.33CH,OH =
   «5          o

  NO" +0.33H.O +0.33H  CO,
     Z        Z          Z -   o
NO, +0.5 CH,OH + 0.5HCO, =
   Z          o            Z   O

   0.5N  + HCO~ + HO
       Z        o    Z
NO3 + 0.833
                     + 0.167H2CO3
                                = 0.5 N2 4  1 .33 H2O + HCO~
14CH3OH + 3N03
                                  3C H O  N + 20 HO + 3HCO~
                                     O  / Z        Z           O
                                       3-36A
                                      •3-37A
3-38A
                                       3-41
The theoretical methanol requirement for nitrate reduction, neglecting synthesis, is 1.9 mg
methanol per mg nitrate N (Equation 3-36, Table 3-9). Including synthesis (Equation 3-42),
the requirement is increased to 2.47. Similarly, Equations 3-43 and 3-44 in Table 3-9 allow
calculation of methanol requirements for nitrite reduction and deoxygenation to allow a
combined expression to be formulated for the methanol requirement:62
            Cm  = 2.47NO3-N +  1.53NO~-N  +  0.87 DO
                                          (3-45)
                                      3-32

-------
 where:
             m
            NO" - N =

            NO" - N =

            DO
 required methanol concentration, mg/1,

 nitrate concentration removed, mg/1,

 nitrite concentration removed, mg/1, and

 dissolved oxygen removed, mg/1.
Biomass production can be calculated similarly:
                C.   = 0.53 NO" -N  +  0.32 NO" -N  +  0.19 DO
where:      C.   =    biomass production, mg/1.
                                                     (3-46)
                                    TABLE 3-9

               COMBINED DISSIMILATION-SYNTHESIS EQUATIONS
             FOR DENITRIFICATION (AFTER MC CARTY, ET AL. (62))
    Transformation
                   Equation
Equation
  No.
 Overall nitrate removal
 Overall nitrite removal
 Overall deoxygenatlon
     1.08CH3OH
                          0.056C5H7N02 + 0.47N2 + 1.68H2O + HCO~
     0.53H2CO3 + 0.67CH3OH =
                                       1.23H2O + 0.48N2 + HCO~
O, + 0.93CH.OH + 0.056NO, =
                          0.056C..H NO  + 1.04H O + 0.59H CO, + 0.056HCO,
                                O /  £,       £,        Z  i3          O
                                                                            3-42
                                                                            3-43
                                                                            3-44
Most  experimental data is expressed in terms of the "M/N ratio," which is the mg of
methanol  per  mg  of initial nitrate  nitrogen concentration.  The  ratio  includes the
requirements for nitrite  and  oxygen,  which  are usually small relative  to the nitrate
requirement. For instance, for a NO§" value of 25 mg/1 of nitrate -N, 0.5 mg/1 nitrite -N and
3.0 mg/1 dissolved oxygen, the methanol requirement can be calculated to be 64.1 mg/1
from Equation 3-45.  The M/N ratio is therefore 2.57 (64.4/25), which is  only 4 percent
greater than the requirement for nitrate alone (2.47).

Values of the "M/N"  ratio required  for complete denitrification range  from the  levels
estimated from Equation 3-45 at 2.5 Ib methanol per Ib of nitrate nitrogen removed up to
3.0 Ib  methanol per  Ib  of nitrate -N  removed.5>6>64,65,66 Departures  of  methanol
                                       3-33

-------
requirements from Equation 3-45 are most likely due to variations in sludge yields among
experimental systems. It has been suggested that column denitrification systems require a
lower M/N ratio than suspended growth systems due to the higher concentration of biomass
maintained in the  column systems.67 Higher biomass levels produce longer solids retention
times and reduce  organism  yields due to increased endogenous metabolism.  In  turn, this
lower yield would result in less methanol required for synthesis and reduce the  "M/N"
ratio. 67

In general, an M/N ratio of 3.0 will enable "complete" denitrification (95 percent removal of
nitrate) and this value may be used for design purposes when methanol is employed as the
carbon source for denitrification.

      3.3.3 Alkalinity and pH Relationships

Equations 3-42 and 3-43 (Table 3-9) show that bicarbonate is produced and carbonic acid
concentration is reduced whenever nitrate or nitrite is  denitrified  to nitrogen gas. The
stoichiometric quantity of alkalinity produced is 3.57 mg alkalinity as CaCC>3 produced per
mg of nitrate or nitrite -N reduced to nitrogen gas.

Since both the alkalinity concentration is increased and the carbonic acid concentration is
reduced, the  tendency of  denitrification is to at least partially reverse  the effects of
nitrification and raise the pH of the biological reaction (Equation 3-9). Denitrification only
partially offsets the alkalinity loss caused by nitrification, since the alkalinity gain per mg of
nitrogen is only one-half the loss caused by nitrification (see Section 3.2.3).

Measured alkalinity  production has been reported to be somewhat  lower  than  indicated
theoretically. Experiments with  an attached growth process showed that the  alkalinity
produced averaged 2.95 mg as CaCC>3 per mg of nitrogen reduced. 65 Similarly, the ratio for
a suspended growth system  was 2.89.6 Departures from theory may be due to the fact that
Equations  3-42 and  3-43  (Table  3-9)  represent over-simplications of  the biological
transformations taking place and do not include all factors affecting alkalinity production.

A value for alkalinity production suitable for engineering  calculations would be  3.0 mg
alkalinity as CaCO3 produced per mg nitrogen reduced.

     3.3.4 Alternative Electron Donors

Although methanol has found a predominance  in  U.S. practice as the electron  donor of
choice,  the  significance of the cost of the organic chosen  for the process has led to the
consideration of alternative electron donors available, particularly those from waste sources.
Considering alternate  commercial sources, methanol seems to continue to  be  the  most
economic choice,  because price  increases in alternate  sources have  paralleled  those for
methanol.
                                        3-34

-------
 A variety  of compounds that can  substitute  for  methanol  have been experimentally
 evaluated,    but design data are available only for municipal wastewater organics, volatile
 acids, brewery wastes, and molasses.  The use of wastewater organics for denitrification is
 discussed  extensively in  Section 5.5.2. Denitrification rates with wastewater  oganics are
 approximately one-third of those when methanol is  employed. Therefore, denitrification
 reactors must be proportionately larger. Since using wastewater organics adds ammonia and
 organic nitrogen to the wastewater, the sequence of nitrification-denitrification steps must
 be  modified to ensure that  these  compounds  do  not escape from  the system. Thus,
 wastewater  organics are not completely interchangeable  with  methanol; their attraction,
 however, is  the possible reduction in operating costs with the  elimination of the need for
methanol in the treatment  plant.

In studies conducted for the development of the City of Tampa, Florida's treatment plant,
it  was shown that  brewery wastes  could  substitute  for methanol when used in both
suspended growth and column  denitrification systems.68 Bench scale  studies exhibited
denitrification rates of 0.25 to 0.22 Ib NC>3 -N rem./lb MLVSS/day with brewery wastes
compared to 0.18  Ib NC>3 -N rem./lb MLVSS/day with methanol at a temperature in the
range of 19 to 24  C. Solids production was  found to be greater with brewery wastes than
methanol, but values were not given. Removal efficiencies were similar in a parallel test of
brewery wastes and methanol using columnar denitrification. 68

Volatile acids  have also  been used as a  carbon source  for denitrification.  In studies of
nitrate reduction in wastewaters generated in  the manufacture of nylon intermediates, it was
found that a mixture of C\ to C§ volatile acids  was very effective as a  carbon source for
denitrification.69 Denitrification rates with  this mixture were 0.36 Ib. NO§  -N rem./lb
MLVSS/day at 20  C and 0.10 Ib NO5 -N rem./lb MLVSS/day at 10 C. These rates compare
favorably with those measured for use with methanol (see Section 5.2.1). Volatile acids can
be produced from wastewater organics by anaerobic fermentation or by low temperature wet
oxidation. In either  case,  the product will contain varying amounts of ammonia nitrogen
which  may have to be removed in the process (as described in Section 5.5.2) or removed
prior to use by ammonia stripping.

Molasses  was tested  at the Central Contra Costa Sanitary District's Advanced Treatment
Test facility as a substitute for methanol. 7® Peak denitrification  rates  at 16 C in a suspended
growth reaction were found to be only 0.036 Ib NO3 -N rem./lb MLVSS/day. In  addition to
having a  slower reaction rate with molasses, the sludge tended to bulk to a greater degree
than with methanol, rising  from a sludge volume index (SVI) averaging 164 ml/gram to one
having a SVI averaging 257 ml/gram. This caused a decrease  in the settling rates of the
sludge when molasses was employed.

Some of the alternatives cause greater sludge production than  others. For instance, about
twice as much sludge is produced per mg of nitrogen reduced when saccharose is used than
when methanol is  employed. On the other  hand, acetone, acetate  and  ethanol produced
similar quantities of sludge to that produced when methanol is employed.62

                                        3-35

-------
Methanol has certain advantages over wastewater carbon sources. It is free of contaminants,
such as nitrogen, and therefore can be used directly in the process without taking the special
precautions that must be made for use of with a waste carbon source. Second, the product is
of consistent quality while wastewater sources may vary in strength and composition either
daily  or seasonally, complicating process  control and optimization. Use of wastewater
sources will require regular assaying of the source to check its purity, strength and biological
availability. Methanol also has the advantage of being nationally distributed while  suitable
waste carbon sources  may not be geographically close to the point of use. Nonetheless, the
significant  disadvantage of methanol is  its cost and  this alone mandates  the necessity of
economic comparisons of alternate carbon sources.

     3.3.5  Kinetics of Denitrification

Just as in the case of nitrification (Section  3.2.5),  environmental factors have a significant
effect on the kinetic rates of denitrifier  growth and nitrate removal. Factors considered in
subsequent sections are temperature,  pH, carbon concentration and nitrate concentration. A
combined kinetic expression incorporating all these  factors is presented.

         3.3.5.1 Effect of Nitrate on Kinetics

The absence of significant quantities  of nitrite in denitrification systems^.?! has led to the
description of the  kinetics of denitrification as a one-step  process from nitrate to nitrogen
gas. The Monod expression is employed to  describe the influence of nitrate on growth rate:
                        D      D   Kp     D


where:      /UD  =   growth rate, day"
            A                                           _1
            Mp  =   maximum denitrifier growth rate, day

             D  =   concentration of nitrate nitrogen, mg/1, and

            Kp  =   half saturation constant, mg/1 NO-  - N.


         3.3.5.2 Relationship of Growth Rate to Removal Rate

Denitrification rates  can be  related  to  the  oganism  growth  rates  by the  following
relationship:
                               qD  =  "D/YD                                 (3-48)

where:      q^  =   nitrate removal rate, Ib NOl - N rem./lb VSS/day, and

            Y~  =   denitrifier gross yield, Ib VSS grown/lb NO~ - N removed.

                                        3-36

-------
 Similarly,  peak denitrification rates  are  related to maximum denitrifier growth rates as
 follows:

                                   Sn  = *VYD                              (3-49)
          3.3.5.3 Solids Retention Time                                              .

 Consideration  of solids  production and  solids retention time is an  important design
 consideration.  A mass balance of the biomass in  a completely mixed  reactor yields the
 relationship. 10,11


                                   = YDqD  '  Kd                            <
where:      0£  =    solids retention time, days, and

            K   =    decay coefficient, day  .
          3.3.5.4 Kinetic Constants for Denitrification

The value of the half saturation constant, Kj>, is very low. Investigators at the University of
California at Davis, found  Kj) for suspended growth systems to be 0.08 mg/1 NO§  -N
without solids recycle and 0.16 mg/1 NOJ- N with solids recycle at 20 C.71>48 por attached
growth systems the value of Kj) was found  to be 0.06 mg/1 NO;} -N at 25 C.^2,73 n can j,e
seen from examination of Equation 3-47 using these values of Kj) that nitrate has almost no
effect  on denitrification above  1-2 mg/1 nitrate -N and approaches a zero order rate. The
observations of several investigators suport these low values of Kr_>, as they have reported
zero order rates above 1-2 mg/1 nitrate -N.60,74,75,76

Yield and decay coefficients from data of  a number of investigations are shown in Table
3-10. In most cases only net yields are reported or can be calculated from the data reported.
The relationship between the gross yield in Equation 3-50 and the net yield is:
                               Y a      K             Y
                               YDqD    Kd            YD
                        D
 where:     YD  =    denitrifier net yield, Ib VSS/lb NO3 - N rem.


The data  of Stensel, et al. 74 for KJ has been used in some cases to derive calculated Y
values for those cases  where none was reported (Table 3-10). Data from another study4°

                                         3-37

-------
allow calculation of both YD and Kj at 20 C. The value of Kj of 0.04 day"1 is consistent
with Stensel's findings at 20 C. Values suitable for use in engineering calculations are YD =
0.6 to 1.20 and Kd = 0.04 day1.

It is notable that when an aerobic stabilization step was incorporated into the process after
anoxic denitrification, net yields reduced by almost an order of magnitude.78,80'T'}1js effect
was attributed  to enhanced endogenous metabolism  where oxygen is provided as an
electron-acceptor.80 Table 3-10 shows a calculated value for K^ under these conditions of
0.19  days"1, almost  five times the rate when  nitrate serves as  the electron  acceptor for
endogenous metabolism.  This concept is supported  by the results of other investigators
whose data show that the endogenous respiration rates when expressed on an equivalent
basis are significantly greater when oxygen serves as the electron acceptor than when nitrate
      1'** 2
Both  organism  growth  rate and nitrate  removal  rate are significantly  affected  by
temperature.  Only one investigator has reported growth rates, 74 all others have reported
removal rates. To  show the effect of temperature on growth and denitrification rates, the
available data have been  summarized in Figure 3-10 on a basis that is normalized with
respect to the rate at 20 C. It can be seen that denitrification proceeds at a reduced rate as

                                    TABLE 3- 10

        VALUES OF DENITRIFICATION YIELD AND DECAY COEFFICIENTS
                FOR VARIOUS INVESTIGATIONS USING METHANOL


Process description

Suspended growth, no solids
recycle, continuous



Suspended growth, batch

Suspended growth, solids
recycle, continuous


Suspended growth, solids
recycle, continuous.
aerated stabilization


Ref.

74
74
74
71
77
7Sa
62
5
48
78
79
78


q
D'
, -1
days
Variable
Variable
Variable
0.12 to 0.32
0.16 to 0.9
0.24 to 3.8
Variable
b
0.131 to 0.347
0.25
e
0.30


Y
D'
Ib VSS
Ib NC>3-N rem.
Variable
Variable
Variable
0.55 to 1.4
0.57 to 0.73
0.45 to 1.43
0.53
0.58
0.542 to 0.703
0.49
0 . 7 to 1 . 4
0.061


Y ,
D'
Ib VSS
Ib NOjj-Nrem
0.57
0.63
0.67
-
-
-
-
0.77C
0.84
0.65C
0.83 to 1.67
0.65d


K,,
d
_1
days
0.05
0.04
0.02
-
-
-
-
0.04d
0.04
0.04°
0.04d
0.19C




Temp, C

10
20
30
20
20
5 to 27
20
10 to 20
20
16 to 18

19 to 20


  3 Substrate was sodium citrate
    q not given, but Qc = 8.0
  c Calculated
    Assumed
  6 6  = 3 to 6 days
                                        3-38

-------
o
o
CM
t-
«*
UJ
ft
a:
h-
o
or
tt
o
UJ
s
UJ
Q
U.
O
iu
                                   FIGURE 3-10
              EFFECT OF TEMPERATURE ON DENITRIFICATION RATE
    /SO
           Key
              Ref.
                          qD   ^ID  Electron
                         days-' days*1 donor
                           at 20 C
                 System
    160
         X X X  X X
              75,83  -
               60
                5
             •  84
                         1,07
                         0.24
    /4O
          a a a a a a
               85
               74
               86
               86
               87
                         0.33
                         1.7
1.04
Sodium Citrate
Endogenous
Methanol
Methanol
Methanol
Methanol
Methanol
Methanol
Methanol
    I2O
     110
80
     60
     40
       S.G. = Suspended  Growth
       A.G. = Attached Growth
     20
                                         _L
                              10          15         20
                                TEMPERATURE, C
                                                               25
                                                                    30
                                    3-39

-------
low as 5 C. Above 20 C, four out of seven sets of data indicate that the denitrification rates
find plateau values at some temperature and do not keep climbing. The parallel systems
study of Murphy et a/. 8 6 js interesting in that it shows attached growth systems to be less
affected by  cold  temperatures than  suspended growth  systems.  Differences between
attached growth systems and suspended growth  systems  may  reflect  differences in the
method of measurement rather than, differences in organism reaction rate. Attached growth
system removal rates were expressed on a unit surface basis while suspended growth systems
were  expressed per unit of biomass (MLVSS) in Murphy's study. 86 it is probable that
surface  slimes in attached  growth systems expand  in cold  weather and compensate for
reduced reaction rates. If the biomass level could be measured the rate per unit of biomass
may very well be similar. For instance, in one study parallel  tests of suspended and attached
growth systems were at 30  C. Biomass measurements were  made in both systems and peak
denitrification rates were found to be comparable, 0.38 Ib NO^-N rem./lb MLVSS/day for
the suspended growth system  and 0.45 Ib NC>3~ -N rem./lb MLVSS/day for the attached
growth system. ^8

Further information on the effect of temperature on denitrification rates is presented in
Chapter 5.

         3.3.5.5 Effect of Carbon Concentration on Kinetics

The effect of carbon concentration on the rate of denitrification has been modeled in terms
of a Monod type of expression. When methanol serves as the carbon source, the expression
is: 74,89
                             "   =
                                                                             (3'52)
                              D    ^D  KM  + M


where:      M   =    methanol concentration, mg/1
            KM =    half saturation constant for methanol, mg/1
                     of methanol.


The earliest investigators used a nonspecific test for methanol, COD. 74 AS a result the initial
evaluation  of KM was somewhat obscured. A later more definitive investigation evaluating
KM used a specific test  for methanol.^9 A chemostat system operating at a solids retention
time of five days and a temperature of 20 C was operated in a manner whereby  reaction
rates were limited by methanol and not nitrate. The value of KM was found to be very low,
0.1 mg/1 as  methanol. The practical implication of this finding is that to achieve 90 percent
of the maximum denitrification rate in a suspended growth reactor, only about 1 mg/1 of
methanol need  be  in  the effluent. In other words, great excesses of methanol  above
stoichiometric  requirements  need  not  be  in  the   effluent  from  a  suspended growth
denitrification process to achieve nearly the maximum denitrification rates.
                                        3-40

-------
         3.3.5.6 Effect of pH on Kinetics

Representative observations of the effect of pH on denitrification rates are shown on Figure
3-11. While there are some anomalies, it is apparent that denitrification rates are depressed
below pH 6.0 and above pH 8.0. There is some disagreement about the pH of the optima,
but the data show the  highest rates of denitrification are at least within the range of pH 7.0
to 7.5.
                                     FIGURE 3-11
                    EFFECT OF pH ON DENITRIFICATION RATE
     100
 Uj
 l-
 2   80

 O
o
£
ct

Ul
Q

1
 u.
 o
 UJ
 O
 Ct
 UJ
 Q.
     6O
     40
     20
       6.0
                                     Key
                                            Reference
                                                5
                                               83
                                              60,90
                                               81
                                          I     I
                                                                 b
                               7.0
                                          pH
8.0
9.0

-------
         3.3.5.7 Combined Kinetic Expression

The same approach as employed for nitrification can be used for denitrification to establish
the effects of environmental  conditions on  the rates of  denitrifier growth (and nitrate
removal):

                      «  •
            A
where:      MT)  =   peak rate of denitrifier growth at given temperature, T, and
                     pH, and
            MT^  =   actual rate of denitrifier growth affected by nitrate, methanol,
                     T, and pH.

Relationships for temperature,  pH, nitrate and methanol established in  Sections 3.3.5.3,
3.3.5.4, 3.3.5.5, and 3.3.5.6 can be employed when using this equation to  predict growth
rates or removal rates. Ordinarily, the term for methanol can be neglected (Section 3.3.5.5).
Removal rates can be related to growth rates through Equation .3-48.

The safety factor concept presented in Section 3.2.6 can be applied to denitrification as well
as to nitrification, as the concept has general validity  for biological  systems. Restating the
concept for denitrification:

                                    SF   =  —£-                               (3-29)
                                            0m
                                             c

In the case of denitrification,  the safety  factor can be  related to nitrate removal rates
through Equation 3-50 and the following similar equation for the minimum solids retention
time:

                                —L    =  Yqn  -  k,                          (3-54)
                                  Qm          D     d
                                   c
The use of these equations for design of suspended growth systems is given in Section 5.2.2
in terms of illustrative examples.

The above equations cannot be directly applied to attached growth denitrification because
the reactions take place in a more complex environment than is present in suspended growth
systems. Rates of nitrate removal in the bacterial films developed in denitrification systems
may be affected by the mass transfer of nitrate or methanol  through the bacterial film. A
biofilm model has been developed^2,53  faat may be use(j  to describe  denitrification in

-------
bacterial slimes,, but its use has not yet been extended to the point where it can be used in
system design. However,  the model indicates that removal rates are most usefully expressed
on a unit surface area basis and this is the procedure adopted in Section 5.3.1  to describe
denitrification in the various attached growth systems.

The biofilm model usefully predicts certain properties of attached growth denitrification
that are  significant in design. The model shows  that the nitrate removal rate  in attached
growth  sytems  should not be drastically affected  by adverse environmental conditions
compared to effects in suspended growth systems.52 In Section 3.3.5.4, for instance, it was
shown that attached growth systems are less affected by cold temperatures than suspended
growth systems. Another interesting prediction of the biofilm model is that methanol will
normally be  film transfer limiting rather than nitrate,  unless the methanol is supplied in
concentrations five times as large as  the nitrate concentration^ (an impractical situation).

      3.3.6  Effect of DO on Denitrification Inhibition

The role of dissolved  oxygen in denitrification is  generally to suppress denitrification. This
has been explained  on  the  basis  that the rate  of  dissimilatory  nitrate reduction is
considerably slower than the  rate of aerobic respiration.71 While it has been observed that
denitrification can occur in  the presence  of low levels of DO,°>"° the mechanism  of
denitrification is attributed to an oxygen gradient in the system whereby some cells are at
zero dissolved oxygen and thus able to reduce nitrate. 1 >6>91

3.4 References

  1. Painter,  H.A., A  Review  of Literature  on  Inorganic Nitrogen  Metabolism  in
    Microorganisms. Water Research, 4, No. 6, pp 393-450 (1970).

  2. Haug, R. T.,  and P.  L. McCarty,  Nitrification with the Submerged Filter. Report
    prepared by  the  Department of Civil  Engineering, Stanford  University for the
    Environmental Protection Agency, Research Grant No. 17010 EPM, August, 1971.

  3. Notes on Water  Pollution No.  52,  Water Pollution Research Laboratory, Stevenage,
    England, 1971.

  4. Gujer, W. and  D. Jenkins, The Contact Stabilization Process-Oxygen and Nitrogen  Mass
    Balances. University of  California, Sanitary  Engineering Research Lab, SERL Report
    74-2, February, 1974.

  5. Mulbarger,  M. C., The Three Sludge System for Nitrogen and Phosphorus Removal.
    Presented at the 44th Annual Conference of the Water Pollution Control Federation,
    San Francisco, California, October, 1971.
                                        3-43

-------
 6. Horstkotte, G. A., Niles, D.G., Parker, D.S., and D.H. Caldwell, Full-Scale Testing of A
    Water Reclamation System. JWPCF, 46, No. 1, pp 181-197 (1974).

 7. Newton, D., and T.E. Wilson, Oxygen Nitrification Process at Tampa. In Applications
    of Commercial Oxygen to Water and Waste-water Systems, Ed. by R.E. Speece and J.F.
    Malena, Jr., Austin, Texas: The Center for Research in Water Resources, 1973.

 8. Gasser, J.A., Chen,  C.L.,  and R.P. Miele,  Fixed-Film Nitrification of Secondary
    Effluent. Presented at the  EED-ASCE  Specialty  Conference, Penn. State University,
    Pa., July, 1974.

 9. Osborn, D.E., Operating Experiences with Double Filtration in Johannesburg. J. Inst.
    of Sew. Purif., Part 3, pp 272-281 (1965).

10. Pearson,  E.A., Kinetics of Biological Treatment.  In Advances  in Water  Quality
    Improvement. Ed. by E.F. Gloyna and W.W. Eckenfielder, Austin, Texas: University of
    Texas Press, 1968.

11. Lawrence, A.W., and P.L. McCarty, Unified Basis for Biological Treatment Design and
    Operation.  JSED, Proc. ASCE, 96, No. SA3, pp 757-778 (1970).

12. Jenkins,  D., and W.E. Garrison, Control of Activated Sludge by Mean Cell Residence
    Time. JWPCF, 40, No. 11, pp 1904-1919 (1968).

13. Monod, J., Researches sur la croissance des cultures. Hermann and Cie, Paris, 1942.

14. Stankewich, M. J., Biological Nitrification with the High Purity Oxygenation Process.
    Presented at 27th Annual Purdue Industrial Waste Conference, Lafayette, Indiana,
    May, 1972.

15. Poduska, R.A. and J.F. Andrews, Dynamics of Nitrification in the Activated Sludge
    Process.  Presented  at the 29th Industrial  Waste Conference,  Purdue  University,
    Lafayette, Indiana, May 7-9, 1974.

16. Christensen, D.R. and P.L. McCarty, Bio-treat: A Multi-Process Biological Treatment
    Model.  Presented  at  the 47th Annual Conference of the Water Pollution Control
    Federation, Denver, Colorado, October,  1974.

17. Knowles, G., Downing, A.L., and M.J. Barrett, Determination of Kinetic Constants for
    Nitrifying Bacteria in Mixed Culture, with the Aid of An Electronic Computer. J. Gen.
    Microbiology, 38, p 263 (1965).

18. Downing, A.L., and A.P. Hopwood, Some Observations on the Kinetics of Nitrifying
    Activated Sludge Plants. Schweiz. Zeitsch f Hydrol., 26, No. 2, p 271 (1964).

                                       3-44

-------
19.  Wuhrmann, K., Research Developments'in Regard to Concept and Base Values of the
     Activated Sludge System. Ed. by E.F. Gloyna and W.W. Eckenfielder, Austin, Texas:
     University of Texas Press, 1968.

20.  Melamed, A., Saliternik, C., and A.M. Wachs, BOD Removal and Nitrification of
     Anaerobic Effluent by Activated Sludge. Presented at the 5th IAWPR Conference, San
     Francisco, California, July-August, 1970.

21.  Balakrishnan, S., and W.W. Eckenfielder,  Nitrogen Relationships in Biological Waste
     Treatment Processes — I, Nitrification in the Activated Sludge Process. Water Research,
     3, pp 73-81  (1969).

22.  Loehr,  R.C.,  Prakasam,  T.B.S.,  Srinath, E.G.,  and  Y.D. Joo,  Development  and
     Demonstration  of Nutrient Removal from Animal Wastes.  Report prepared  for the
     Environmental Protection Agency, EPA-R2-73-095, January,  1973.

23.  Stratton,  F.E., and P.L. McCarty, Prediction  of Nitrification Effects on the Dissolved
     Oxygen Balance of Streams. Environmental Science and Technology, 1, No. 5, pp
     405-410(1967).

24.  Lawrence, A.M., and C.G. Brown, Bio kinetic  Approach to Optimal Design and Control
     of Nitrifying Activated Sludge Systems. Presented at the Annual Meeting of the New
     York Water Pollution Control Association, New York City, January 23, 1973.

25.  Hoffman, T., and H.  Lees,  The  Biochemistry of Nitrifying  Organisms   4,  The
     Respiration  and  Intermediate Metabolism of Nitrosomonas. Biochemistry Journal, 54,
     575-583(1953).

26.  Ulken, A., Die Herkunft des Nitrits in der Elbe.  Arch. Hydrobiol., 59, pp 486-501,
     1963.

27.  Loveless.  J.E., and J.E. Painter, The Influent of Metal Ion Concentration and pH Value
     on the  Growth  of a Nitrosomonas Strain Isolated From Activated Sludge.  J. Gen.
     Microbiology, 52, pp 1-14(1968).

28.  Williamson, K.J. and  P.L.  McCarty, Rapid Measurement  of Monod Half-Velocity
     Coefficients for  Bacterial  Kinetics.  Unpublished paper, Stanford University,  May,
     1974.

29.  Lees, H.,  and J.R. Simpson, The Biochemistry of the Nitrifying Organisms. 5, Nitrate
     Oxidation by Nitrobacter. Biochemistry Journal, 65, pp 297-305 (1957).

30.  Gould, G.W., and H. Lees, The Isolation and  Culture of the Nitrifying Organisms.  Part
     I. Nitrobacter. Can. J. Microbiology, 6,  pp 299-307 (1960).

                                       3-45

-------
31.  Landelot,  H.,  and L. Van Tichilen, Kinetics  of Nitrite Oxidation by Nitrobacter
     Winogradskyi, J. Bact., 79, pp 39-42 (1960).

32.  Buswell, A.H.,  Shiota, T., Lawrence, N., and I. Van  Meter, Laboratory Studies on the
     Kinetics of the  Growth of Nitrosomonas with Relation to the Nitrification Phase of the
     BOD Test. Applied Microbiology, 2, pp 21-25 (1954).

33.  Balakrishnan, S., and W.W.  Eckenfielder, Nitrogen Relationships in Biological Waste
     Treatment Processes  — II, Nitrification in  Trickling  Filters, Water Research, 3, pp
     167-174(1969).

34.  Huang, C.S.,  and N.E. Hopson,  Temperature and  pH Effect on  the  Biological
     Nitrification  Process. Presented at  the Annual Winter Meeting,  New York  Water
     Pollution Control Association, New York City, January, 1974.

35.  Water Pollution Research, 1964. London: HMSO, 1965.

36.  Nagel, C.A.  and Haworth,  J.G.,  Operational Factors Affecting Nitrification in the
     Activated Sludge  Process. Presented  at the 42nd Annual  Conference of the Water
     Polluation Control Federation, Dallas, Texas, October (1969), (available as a reprint
     from the County Sanitation Districts of Los Angeles County).

37.  Wuhrman, K.,  Effects of  Oxygen  Tension  on Biochemical Reactions in Sewage
     Purification.  In Advances in Biological Treatment, Proc. 3rd Conf. Biological Waste
     Treatment, Ed. by Eckenfielder, W.W., and J. McCabe. (New York, 1963), Pergamon
     Press.

38.  Schwer, A.D., Letter communication to D.S. Parker — Metropolitan Sewer District of
     Greater Cincinnati, March 9, 1971.

39.  Murphy, K.L.,  Personal communication to D.S. Parker, McMasters University, Ontario,
     July, 1974.

40.  Engel, M.S.,  and M. Alexander, Growth and Metabolism of Nitrosomonas europaea.
     Journal Bacteriology, 76, pp 217-222 (1958).

41.  Meyerhof, O.,  Untersuchungen uber den Atmungsvorgany Nitrifizierenden Bakterien.
     Pflugers Archges Physiol., 166, pp 240-280, 1917.

42.  Sawyer,  C.N.,  Wild,  H.E., Jr., and T.C.  McMahon, Nitrification and Denitrification
     Facilities, Wastewater Treatment. Prepared for the EPA Technology Transfer Program,
     August, 1973.
                                        3-46

-------
43.  Downing, A.L., and G. Knowles, Population Dynamics in Biological Treatment Plants.
     Presented at the 3rd Conference of the IAWPR, Munich, 1966.

44.  Huang, C.S., Kinetics and Process Factors of Nitrification  on a Biological Film Reactor.
     Thesis submitted in partial satisfaction of the requirements for the degree of Doctor of
     Philosophy, University of New York at Buffalo, 1973.

45.  Heidman, J.A., An Experimental  Evaluation of Oxygen and Air Activated  Sludge
     Nitrification Systems With and  Without pH Control. EPA  report fqr Contract No.
     68-03-0349, 1975.

46.  Chen, C.W., Concepts and Utilities of Ecological Model. JSED, Proc. ASCE, 96, No.
     SA5,pp  1085-1097(1970).

47.  Middlebrooks,  E.J., and D.B.  Porcella, Rational Multivariate Algal Growth Kinetics.
     JSED, Proc. ASCE, 97, SA1, pp 135-140 (1971).

48.  Engberg, D.J.,  and E.D. Schroeder, Kinetics and Stoichiometry of Bacterial Denitrifi-
     cation as a Function of Cell Residence  Time.  University  of California at Davis,
     unpublished paper, 1974.

49.  Stamberg,  J.B., Hais, A.B., Bishop, D.F.,  and J. Heidman, Nitrification in  Oxygen
     Activated Sludge. Unpublished paper, Environmental Protection Agency, 1974.

50.  Sawyer,  C.N.,  Activated Sludge  Oxidations,  V.   The Influence of Nutrition  in
     Determining Activated Sludge Characteristics.  Sewage Works Journal, 12, No. 1, pp
     3-17(1940).

51.  Kiff, R.J.,  The Ecology  of Nitrification IDenitrification in Activated Sludge. Water
     Pollution Control, 71, p 475 (1972).

52.  Williamson, K.L.  and P.L. MeCarty, A Model of Substrate Utilization by Bacterial
     Films. Presented  at  the  46th Annual  Conference of the Water Pollution  Control
     Federation, Cincinnati, Ohio, October, 1973.

53.  Williamson, K.L.  and P.L. MeCarty, Verification Studies of the  Biofilm Model for
     Bacterial Substrate Utilization. Presented at the 46th Annual Conference of the Water
     Pollution Control Federation, Cincinnati, Ohio, October,  1973.

54.  Interaction  of Heavy Metals and Biological Sewage Treatment Processes.  Division of
     Water Supply and Pollution Control, USPHS, May, 1965.

55.  Environmental Quality Analysts, Inc., Letter Report  to Valley Community  Services
     District, March, 1974.

                                       341

-------
56.  Sawyer, C.N., Letter communication to D.S. Parker, January 24, 1975.

57.  Davis, G., Personal  communication  with  D.S.  Parker, B.F. Goodrich, Co., August,
     1974.

58.  Tomlinson, T.G., Boon, A.G., and C.N.A.  Trotman, Inhibition of Nitrification in the
     Activated Sludge Process of Sewage Disposal. J. Appl. Bact., 29, pp 266-291 (1966).

59.  Anthonisen, A.C., Loehr,  R.C., Prakasam, T.B.S.,  and E.G. Srinath, Inhibition of
     Nitrification by Un-ionized Ammonia and Un-ionized Nitrous Acid. Presented at the
     47th Annual Conference,  Water  Pollution  Control Federation, Denver,  Colorado,
     October, 1974.

60.  Christensen, M.H.,  and  P.  Harremoes,  Biological Denitrification in Wastewater
     Treatment.  Report 2-72, Department of Sanitary Engineering, Technical University of
     Denmark, 1972.

61.  Delwiche,  C.C., The Nitrogen  Cycle. Scientific American, 223, No. 3, pp  137-146
     (1970).

62.  McCarty, P.L., Beck, L., and P. St. Amant, Biological Denitrification of Waste-waters by
     Addition of Organic Materials. In Proc. of the 24th Industrial Waste Conference, May
     6, 7, and 8, 1969. Lafayette, Indiana: Purdue University, 1969.

63.  McCarty,  P.L.,  Stoichiometry  of Biological  Reactions.  Presented  at  the Summer
     Institute in Water Pollution Control, Biological  Waste Treatment, Manhattan College,
     •New York, N.Y., May, 1974.

64.  Dholakia,  S.G.,  Stone,  J.H.,   and  H.P. Burchfield, Methanol Requirement  and
     Temperature Effects in  Waste-water Denitrification.  U.S. Environmental Protection
     Agency, Washington, D.C., WPCRS 17010 DHT 09/70, August, 1970.

65.  Jeris, John S.5 and R.W. Owens, Pilot Scale  High Rate Biological Denitrification at
     Nassau County, N. Y. Presented at the Winter Meeting of the New York Water Pollution
     Control Association, January, 1974.

66.  Smith, J.M., Masse, A.N., Feige, W.A.,  and L.J. Kamphake, Nitrogen Removal from
     Municipal   Wastewater  by  Columnar  Denitrification.  Environmental  Science  and
     Technology, 6, p 260 (1972).

67.  Michael, R.P., and W.J. Jewell,  Optimization of the Denitrification Process. Journal of
     the Environmental Engineering Division, Proc,  ASCE, in press.
                                       3-48

-------
68.  Wilson, T.E., and D. Newton, Brewery Wastes as a Carbon Source for Denitrification at
     Tampa, Florida. Presented at the  28th Annual Purdue  Industrial Waste Conference,
     May, 1973.

69.  Climenhage, D.C., Biological Denitrification  of Nylon Intermediates  Waste Water.
     Presented  at the 2nd Canadian Chemical Engineering Conference, September, 1972.

70.  Central Contra Costa Sanitary District,  Operating Report, Advanced Treatment Test
     Facility. January, 1974.

71.  Moore, S.F., and E.D. Schroeder, The Effect of Nitrate Feed Rate on Denitrification.
     Water Research, 5, pp 445-452 (1971).

72.  Requa, D.A.,  Kinetics  of Packed Bed Denitrification.  Thesis  submitted  in  partial
     satisfaction of the requirements for  the degree of Master of Science in Engineering,
     University  of California at Davis,  1970.

73.  Requa, D.A., and E.D. Schroeder, Kinetics of Packed Bed Denitrification. JWPCF, 45,
     No. 8,pp 1696-1707(1973).

74.  Stensel, H.D., Loehr, R.C., and A.W. Lawrence, Biological Kinetics of the Suspended
     Growth Denitrification. JWPCF, 45, No.  2, pp 249-261 (1973).

75.  Murphy,  R.L.,  and  R.N. Dawson, The  Temperature Dependency  of Biological
     Denitrification. Water Research, 6, pp 71-83 (1972).

76.  Ericsson, et al., Nutrient Reduction at Sewage Treatment Plants. KTH Publication
     67:5, Stockholm, 1967.

77.  Moore, S.,  and E.D.  Schroeder, An Investigation of the Effects of Residence Time on
     Anaerobic Bacterial Denitrification. Water Research, 4, pp 685-694 (1970).

78.  Parker, D.S.,   Zadick,  F.J.,  and K.E. Train, Sludge Processing for  Combined
     Physical-Chemical-Biological Sludges.  Prepared for  the EPA, Report No. R2-73-250,
     July, 1973.

79.  Sutton,  P.M.,  Murphy,  K.L.,  and R.N.  Dawson,  Low  Temperature Biological
     Denitrification of Wastewater. JWPCF, 47, No. 1, pp 122-134 (1975).

80.  Parker, D.S., Aberley, R.C., and D.H. Caldwell, Development and Implementation of
     Biological  Denitrification for Two Large  Plants.  Presented at the Conference  on
     Nitrogen as a  Water Pollutant, sponsored by  the IAWPR, Copenhagen,  Denmark,
     August, 1975.
                                       3-49

-------
81.  Clayfield, G.W., Respiration  and Denitrification Studies on Laboratory and Works
     Activated Sludges. Water Pollution Control, London, 73, No. 1, pp 51-76 (1974).

82.  Balakrishnan, S., and W.W. Eckenfielder, Nitrogen Relationships in Biological Waste
     Treatment Processes — HI, Denitrification in the Modified Activated Sludge Process.
     Water Research, 3, pp 177-188 (1969).

83.  Dawson, R.N., and K.L. Murphy, Factors Affecting Biological  Denitrification in
     Waste-water.  In Advances in  Water Pollution Research, S.H. Jenkins, Ed., Oxford,
     England: Pergamon Press, 1973.

84.  Parker, D.S., Case Histories of Nitrification and Denitrification Facilities.  Prepared for
     the EPA Technology Transfer Program, May, 1974.

85. 'Bishop,  D.F.,  Personal  communication to D.S. Parker. Environmental Protection
     Agency, Washington, B.C., April,  1974.

86.  Murphy, K.L.,  and P.M.  Sutton, Pilot Scale Studies on Biological Denitrification.
     Presented at the 7th International Conference  on  Water Pollution Research, Paris,
     September, 1974.

87.  Ecotrol, Inc.,  Biological Denitrification  Using  Fluidized Bed Technology.  August,
     1974.

88.  Jewell,  W.J.,  and   R.J.  Cummings, Denitrification of  Concentrated  Wastewaters.
     Presented  at  the Water Pollution  Control  Federation,  Cleveland, October,  1973.

89.  Kaufman, G., and E. Schroeder, Personal communication to D.S. Parker. University of
     California at Davis, July, 1974.

90.  Renner, Production  of  Nitric Oxide and Nitrous  Oxide During Denitrification by
     Comybacterium nephridii. J. Bacterial., 101, pp 821-826 (1970).

91.  Skerman, V.B.D., and I.C. MacRae, The  Influence  of Oxygen  Availability  on the
     Degree of Nitrate Reduction by Pseudomonas Denitrificans. Can. J. Microbiology, 3, pp
     505-530(1957).
                                        3-50

-------
                                     CHAPTER 4

                           BIOLOGICAL NITRIFICATION
4.1 Introduction

The application of biological nitrification in municipal wastewater treatment is particularly
applicable to those cases where an ammonia removal requirement exists, without need for
complete nitrogen removal. Biological nitrification is also the first step of the biological
nitrification-denitrification approach to nitrogen removal.

4.2 Classification of Nitrification Processes

The first means of categorizing nitrification systems concerns the degree of separation of the
carbon  removal and nitrification processes.  The  first  nitrification  processes  developed
combined the functions of carbon oxidation and nitrification in one process. The extended
aeration modification of the activated sludge process is an example of a combined carbon
oxidation-nitrification  process. Combined carbon oxidation-nitrification processes generally
have low populations of nitrifiers due to a high  ratio of BOD5 to Total Kjeldahl Nitrogen
(TKN) in  the influent (see  Section 3.2.7 for a  discussion of  this effect). The bulk of the
oxygen requirement  for this process comes from the oxidation of organics.

Separate stage nitrification is the other category of nitrification processes. In this  process,
there  is a lower BOD5 load relative  to the influent ammonia load. As  a result,  a higher
proportion of nitrifiers is  obtained, resulting in higher rates of nitrification. The bulk of the
oxygen requirements in the  nitrification stage derive from ammonia oxidation. To obtain
separate  stage nitrification,  pretreatment  is required  to  lower the  organic  load  or
BOD5/TKN ratio in the influent to the nitrification stage.

Both the combined carbon oxidation-nitrification and separate stage nitrification processes
can be further subdivided into suspended growth and attached growth processes. Suspended
growth processes are those which suspend  the biological solids in a mixed liquor by some
mixing mechanism. A subsequent clarification stage is required for returning these solids to
the nitrification stage. Attached  growth processes,  on the other hand, retain the bulk of the
biomass on the media and therefore do not require a solids separation step for returning the
solids to the  nitrification reactor. In separate  stage processes operated  in the attached
growth mode, a clarification  step may not be  required since solids synthesis is low and the
sloughed solids are often low in concentration.

There are  many  different  configurations of suspended and attached growth reactors; these
are described in subsequent sections of this manual. For suspended growth reactors refer to
Sections 4.3 and 4.6; for attached growth reactors refer to Sections 4.4 and 4.7.
                                         4-1

-------
Using the classification described  above,  representative nitrification processes have  been
classified in Table 4-1 according to  the degree of separation of the carbon removal and
nitrification processes. Many of these facilities are described in further detail in either this
chapter or Chapter 9. Each facility listed has been categorized according to the BOD5/TKN
ratio  of the wastewater influent to  the nitrification process. Interestingly,  the processes
listed can all be categorized according to whether the BOD5/TKN ratio is less than 3.0 or
greater than 5.0. If the BOD5/TKN ratio is less than 3.0, the system can  be classified as a
separate stage nitrification process.  If the BOD5/TKN is greater than 5.0, the process can be
classed as a combined carbon  oxidation-nitrification process. Also shown in Table 4-1 is the
distribution  of total oxygen demand  in the process between carbonaceous sources (8005)
and  nitrogenous sources  (NOD). It can  be seen that  in  separate stage  processes, the
proportion of nitrogenous oxygen demand is at least 60 percent of the total. In combined
carbon  oxidation-nitrification processes, the proportion of nitrogenous oxygen demand is
lower than 50 percent.

There exists a range of BOD5/TKN ratios between 3.0 and 5.0 where no practical examples
currently exist. Facilities in this range could be considered to provide an intermediate degree
of separation of carbon removal and nitrification.

4.3 Combined Carbon Oxidation-Nitrification in Suspended Growth Reactors

The  conventional activated sludge process has seen relatively  wide application in Great
Britain  for use in obtaining effluents low in ammonia nitrogen.  Much of the U.S.  practice
derives  from  that  experience.  Recent U.S. design  practice  has  provided  amplifying
information.

General design  concepts for  the activated sludge process are covered in the Technology
Transfer publication, Process Design Manual for Upgrading Existing Wastewater Treatment
Plants. 25 The following sections provide an extension of these concepts to combined carbon
oxidation-nitrification applications.

     4.3.1 Activated Sludge Modifications

Not  all of the various  modifications of the activated  sludge process are appropriate for
nitrification applications,  although some see use  where only partial ammonia removal is
required. Figure 4-1 gives pictorial representations of four common modifications.

         4.3.1.1 Complete Mix Plants

Many plants are  designed to operate  on the complete mix principle. Shown on Figure 4-1 is
an example of the feed and withdrawal arrangement for a complete mix plant. The complete
mix  design provides  uniformity  of load to all points within the aeration tank,  easing the
problems of oxygen transfer presented in the head end of the conventional  plants. Complete
mix  plants  can  be  designed for  complete nitrification at loading  rates comparable  to

                                        4-2

-------
                                                        TABLE 4-1
                                    CLASSIFICATION OF NITRIFICATION FACILITIES



Type and Location

Suspended Growth
Manassas, Va.
Hyperion, Los Angeles, Ca.
Central Contra Costa Sanitary
District, Ca.
Llvermore, Ca.
Flint, Michigan
Valley Community Services District, Ca.
Blue Plains, D.C.

Whittier Narrows, LACSD, Ca.
Jackson, Michigan
Tampa, Florida

South Bend, Indiana
New Market, Ontario, Canada
Cincinnati, Ohio

Fitchburg, Mass.
Marlboro, Mass.
Amherst, N. Y.
Denver, Col.
Attached Growth
Stockton, Ca.

Midland, Mich.
Union City, Ca.
Allentown, Pa.
Lima, Ohio


Scale,
mgda


0.2
46
1.0, design 30
pilot
3.3
34
3.8
pilot, design
309
12
13.5
pilot
design 60
pilot
2.4
pilot
pilot
pilot f
pilot
pilot
pilot

pilot
design 58
pilot
pilot
40
pilot


BOD5/TKN
Ratio


1.2
7.3b
2.4
1.0
2.8
5.5
10. 8b
1.3 to 3.0

6.6
9
3.0

1.8
2.6
7.2
1.0°
i.od
3.6d
* 0.8 to 2.0
2.7

5.3

1.1
1.7
1.9
0.79
Oxygen Demand
Distribution in
percentage


BOD5

20
61
34
18
38
65
70
22 to 39

61
66
40

28
36
61
18°
18
40
22
37

54

19
27
30
15

NOD

80
39b
66
82
62
35b
30b
61 to 78

39
34
60

72
64
39
82°
82
60
78
63

46

81
73
70
85



Bef.


1
2
3
4
5
6
3
7, 8

9
10
11

12
13
14
15
16
17
18
19, 20

21

22
23, 24
25
26, 27
Classification -
Degree of separation
Combined
oxidation -
nitrification


X



X
X


X
X




X






X






Separate
stage

X

X
X
X


X



X

X
X

X
X
X
X
X



X
X
X
X



Pre treatment


Activated sludge
Primary treatment
Lime primary treatment
Activated sludge
Boughing filter
Primary treatment
Primary treatment
Activated sludge

Primary treatment
Primary treatment
Activated sludge

Activated sludge
Lime primary treatment
Primary treatment
Activated sludge
Activated sludge
Trickling filter
Activated sludge
Activated sludge

Primary treatment

Trickling filter
Activated sludge
Trickling filter
Activated sludge
a 1 mgd = 0.044 mVsec
  Calculated from effluent
  Approximate, calculated from COD data

  BOD/NH^"- N ratio; BOD/TKN would be about 3. 0

-------
conventional plants. As will  be  shown in Section 4.3.3.2, complete mix plants may have
slightly higher effluent ammonia contents than conventional plants due to increased short
circuiting of the influent to the  effluent. Design procedures for nitrification with complete
mix plants are presented in Section 4.3.3.

          4.3.1.2 Extended Aeration Plants

Extended aeration plants are similar to complete mix  plants excepting that  hydraulic
retention times range from 24 to 48 hr instead of the 2 to 8 hr used in complete mix plants.
Extended aeration  plants are operated to maximize endogenous respiration, consequently
solid  retention times of 25 to 35 days are not uncommon. Because of their long aeration
periods, they suffer from unusual heat losses and low temperatures. Extended aeration
plants, because of their low net growth rate, can be expected to nitrify except at the coldest
of temperatures ( < 10 C). Unless the sludge inventory is kept under control via intentional
sludge wasting, solids are  periodically lost in the effluent and nitrification efficiency wanes.
Section 4.3.4 includes a discussion of design procedures for extended aeration plants.

          4.3.1.3 Conventional or Plug Flow Plants

Conventional plants consist of a series of rectangular tanks or passes with the total tank
length to width ratio of 5 to 50.^5 The hydraulics of the system have been loosely termed a
plug  flow configuration,  so  called because the influent wastewater and return activated
sludge are returned to the head end of the process and the combined  flow must pass along a
long narrow  aeration tank prior to exiting from the system. The degree  to which the process
actually approaches plug flow is  dependent on the amount of longitudinal mixing in the
process.  Conventional plants can  be designed  to  dependably  nitrify using the design
approach presented in Section 4.3.5.

          4.3.1.4 Contact Stabilization Plants

The contact stabilization modification of the activated  sludge process derives from the
alteration of the feed pattern to  the process. Instead of mixing the influent wastewater with
the return sludge, the return activated sludge is separately aerated in a sludge reaeration tank
prior to mixing with the influent wastewater. Backmixing between  the contact tank and the
sludge reaeration  tank  is prevented  by  providing  overflow weirs or  pumps  between the
tanks. BOD5 removal can take  place in the contact  tank which  has a relatively short
detention time,  0.5 to 1 hr based on average dry weather flow (ADWF). BOD5  removals can
be  fairly  high because the bulk  of the organics in domestic wastewater are particulate or
colloidal  and can be adsorbed  to  the biological solids for later  oxidation  in  the sludge
reaeration (or stabilization) tank.

The process is not well suited for complete nitrification, even though  relatively  high solids
retention times  can be maintained in the process because of the inventory of solids in the
sludge reaeration tank.  Nonetheless, insufficient  biological  mass is present in the contact

                                         4-4

-------
                            FIGURE 4-1
        MODIFICATIONS OF THE ACTIVATED SLUDGE PROCESS
   RAW
   WASTEWATER
                          FINAL  \EFFLUENT
                        CLARIFIER I      *
                                                  ' ' EXCESS SLUDGE
                                                           ^
                 CONVENTIONAL ACTIVATED SLUDGE PLANT
   RAW
   WASTEWATER
    OR —	
   PRIMARY
   EFFLUENT
                       tttt ttt
AERATION TANK
    4 * I
                         FINAL  \ EFFLUENT
                       I CLARIF IER I	»•
                           RETURN SLUDGE
                                                     EXCESS SLUDGE
                       COMPLETE MIX PLANT
RAW
WASTEWATER
           SLUDGE



r~^^

i r
SLUDGE
REAERATION
TANK

CONTACT
TANK
                           RETURN  SLUDGE
                                                      EXCESS SLUDGE
                   CONTACT STABILIZATION PLANT
RAW

PRIMARY
^ CLARIFIER ^
•EWATER
JsLUDGE





* *
r
AER
>
ATION
V
r
T A
J
rr t
RETURN
~>
NK

SLUDGE


FIN
CLAR
s
\
AL ^ EFf
IFIER I1
1 EXCESS S
                      STEP  AERATION  PLANT

                                4-5

-------
tank  to  completely nitrify  the ammonia  and since ammonia is  not adsorbed on the
biological floe, ammonia  will  bleed through to the effluent.  Partial  nitrification can be
obtained at levels which can be predicted by methods presented in Section 4.3.6.

         4.3.1.5 Step Aeration and Sludge Reaeration Plants

A typical step aeration plant is illustrated on Figure 4-1.  Like  the conventional plant, the
return sludge is introduced at the head end of the aeration tank. However, the step aeration
plant differs from the conventional plant in that influent wastewater is introduced at several
points along the aeration tank.  This distribution of influent flow reduces the initial oxygen
demand usually experienced in the conventional plant. 25

A variation on  the  step  aeration plant that has been popular on the West Coast  is to
introduce no feed into the first  pass while  directing the flow into the remaining downstream
passes. A sludge reaeration zone is established in  the first pass and this variation has become
known as a  "sludge reaeration plant." Normally,  no provision is made to prevent  back
mixing between the sludge reaeration pass and the downstream passes.

The ammonia bleedthrough characterizing contact stabilization plants is avoided in a step
aeration  plant because of the greater contact times employed  and backmixing of influent
occurs. Nonetheless, some bleedthrough of ammonia as  well as organic nitrogen can occur.
This breakthrough results from short circuiting of influent to the effluent and insufficient
contact time for complete organic nitrogen hydrolysis (ammonification) and oxidation of
ammonia.

         4.3.1.6 High  Rate and Modified Activated Sludge

High rate activated sludge processes (high MLSS) and modified activated sludge (low MLSS)
processes are characterized by low solids retention times (0.5 days). Under these conditions,
a nitrifying activated sludge cannot be developed. The high rate and modified activated
sludge processes are acceptable pretreatment techniques  for  separate stage  nitrification
processes (Section 4.5).

         4.3.1.7 High  Purity Oxygen Activated Sludge Plants

Both covered and uncovered reactors have been used with pure oxygen activated sludge, but
only  the former technique  has seen actual implementation  in full-scale  plants. 28 The
covered reactor approach involves the recirculatioh of reactor off-gases to achieve efficient
oxygen utilization. As  a consequence, the carbon dioxide  which is present in the off-gas is
returned to the liquid. The end result is that high carbon dioxide concentrations build up in
the mixed liquor and recycle gases,  depressing the mixed liquor pH. pH levels as low as 6.0
are not  uncommon. This  effect can have  a depressing  effect on  nitrification rates (cf.
Sections  3.2.5.6 and  4.6.3), resulting in  the  requirement for somewhat longer  solids
retention times for nitrification  than would otherwise be the case.

                                         4-6

-------
Virtually all applications of the high purity oxygen activated sludge process to nitrification
have  been for  separate stage nitrification  applications  (Section  4.6),  rather than  for
combined carbon oxidation-nitrification applications.

     4.3.2 Utility of Nitrification Kinetic Theory in Design

The nitrification kinetic theory presented in Chapter 3 may be directly applied to the design
of those activated sludge modifications compatible with nitrification. The equations must be
adapted to the hydraulic configuration under consideration, but in all cases this adaptation
is relatively straightforward.

Nitrification kinetic theory can be very usefully applied to define the following parameters:

     1.   The safety  factor  required to handle  diurnal  transients in loading to  prevent
         significant ammonia bleedthrough under peak load conditions.

     2.   The design solids retention time under the most adverse conditions of pH, DO and
         temperature.

     3.   The allowable  organic  loading on the combined carbon oxidation-nitrification
         stage.

     4.   The required hydraulic detention  time in the aeration tank at ADWF.

     5.   The excess sludge wasting schedule.

The  following sections present  the design procedures in terms of a number of  specific
examples. The procedure  developed for  each  case has often been  termed the "solids
retention time" design approach.

     4.3.3 Complete Mix Activated Sludge Kinetics

As  a design  example,  consider  a  1 mgd treatment plant that must achieve complete
nitrification at 15 C.  The  plant incorporates primary treatment. Primary effluent BOD5 is
150 mg/1, including solids handling return streams to the primary. Total Kjeldahl Nitrogen
(TKN) is 25 mg/1 as N. As a simplifying assumption, neglect that portion of the TKN that is
assimilated into  biomass or  associated  with refractory  organics. The wastewater has an
alkalinity of 280 mg/1 as CaCO3- The procedure is as follows:

      1.  Establish the safety factor, SF. The SF is affected by the desired effluent quality.
         Assume  a minimum SF of 2.5 is required  due to transient loading conditions at
         this particular plant (see Section 4.3.3.2).

     2.   Establish the minimum  mixed  liquor dissolved  oxygen (DO)  concentration.

                                         4-7

-------
   Consideration of aeration  efficiency  at  the  peak hourly load is required (see
   Section 4.8). Assume a minimum DO of 2.0 mg/1 is selected as a compromise
   between power requirements and a consideration  of the depressing effects of low
   DO levels on the rate of nitrification as discussed in Section 3.2.5.5.

3. Estimate the process operating pH (see Section 4.9.2). Approximately 7.14 mg/1
   of alkalinity as CaCO3 is destroyed per mg/1  of NH4 -N oxidized. Neglecting the
   incorporation of nitrogen into biomass, the alkalinity remaining after nitrification
   will be at least:

                    280- [7.14(25)]= 102 mg/1

   If a coarse bubble aeration system is  chosen,  the  pH should remain above pH 7.2
   and chemical addition is not required for pH control (see Section 4.9.2).

4. Calculate the maximum growth rate of nitrifiers at  15 C, DO =  2 mg/1,  and
   pH >7.2. The appropriate equation to be used was presented in Section 3.2.6 and
   is as follows:
where:
          ~*
          xr
         K
                                                                 _J
                         maximum possible nitrifier growth rate, day  ,
                         environmental conditions of pH, temperature, and DO,
                                                         _1
                         maximum nitrifier growth rate, day  , and

                         half-saturation constant for oxygen, mg/1.
    The last bracketed term is taken as  unity at a pH above 7.2. Using the specific
    values adopted in Section 3.2.5  for JLI^ and KQ2 leads to  the following expres-
    sion:
0.098(T-15)
                                   DO
                                 DO + 1.3
                         - 0.833(7. 2 -pH)
                                                        )J
                                                                          (4-1)
          Using the numbers given above:
                   M  = (0.47)(0.61) = 0.285 day
                                               "1
5.  Calculate the minimum solids retention  time  for nitrification. From Equation
    3-15, the correct expression is:
                                   4-8

-------
                          0m=  -4-                             .        (4-2)

                           c   ,  "N



  where:    d      =   minimum solids retention time, days, for nitrification at pH,

                       temperature and DO.



  For this example:
6.  Calculate the design solids retention time.  From  Equation 3-29,  the  correct

    expression is:



                         0J? = SF.0m                                     (4-3)





    where:    d       =   solids retention time of design, days.
              C/



    For this example:

                                  0  =2.5(3.51) = 8.78 days.
                                   C




7.  Calculate the design nitrifier  growth rate.  From  Equation 3-12,  the  correct

    expression is:
    where:    ju      =   nitrifier growth rate Nitroso mo nas  , day"  .
    For this example:
                                  =  -8778   =0-114 day'1
8.  Calculate the half-saturation constant for ammonia oxidation at 15 C. The proper

   expression is:


                       KN=100.051T-1.158                           (3-13)




   where:   KN    =   half-saturation constant for NH, - N, mg/1, and


            T     =   Temperature, C



   For this example:

                               KN=1(T0-3   = 0.405 mg/1




                                  4-9

-------
 9.  Calculate the steady state ammonia content of the effluent. Equation  3-24  is

    directly applicable to complete mix activated sludge  systems, where Nj is the

    effluent ammonia-nitrogen content:



                                 N.

                     M  =MN  - 1 -                                 (3-24)
                      N    IN   v-
    where:   Nj     =   effluent NH* - N, mg/1



    For this case:



                                                      Nl
                               M  = 0.1 14 = 0.285
                                                   Nj +0.405
                                  =0.27mg/l
    Transient loading effects' on effluent quality are presented in Section 4.3.3.2.



10.  Calculate -the organic removal rate. The design solids retention time 0  applies to
                                                                   C
    both the nitrifier population and the heterotrophic population. Equation 3-27 can

    be applied to determine substrate removal rates:





                       =-=-K                                (3'27)
   where:   Yb     =   heterotrophic yield coefficient, Ib VSS grown per Ib BOD,

                       removed,



            qb     =   rate of substrate removal, Ib BOD5 removed/lb VSS/day, and


            K,     =   "decay" coefficient, day  .



   Assume representative values for Y, and K , :^
                                  o     a




            Yb     =   0.65 Ib VSS/lb BOD rem.


            Kd     =   0.05 day'1
  Therefore:
            0.114 =   0.65qb-0.05



            qb    =   0.252 Ib BOD rem./lb MLVSS/day
                                   4-10

-------
    In the above calculation of qj,, it is assumed that the fraction of nitrifiers is low
    and can be neglected (see Section 4.6.1 for a discussion of this point).

11.  Determine the hydraulic detention time at ADWF.  In this analysis, the MLVSS
    content and effluent soluble BOD must be known. The effluent soluble BOD5
    can be assumed to be very low (say 2 mg/1). The MLVSS content is dependent on
    the mixed liquor total suspended  solids, which is in turn dependent on the
    operation of the nitrification-sedimentation tank (Section 4.10).  Assume for the
    purposes  of this example  that  the  design mixed liquor  content at  15 C is
    2500  mg/1.  At a volatile  content of 75 percent, the MLVSS is 0.75 (2500) =
    1875  mg/1.  From Equation 3-28, the expression for hydraulic detention time is:
                                S - S
                                                                        (4_5)
                                Xlqb

    where:   HT    =    hydraulic detention time, days,

             X.    =    mixed liquor volatile suspended solids, MLVSS, mg/1,

             SQ    =    influent total BOD5, mg/1, and

             Sj     =    effluent soluble BOD5, mg/1.

    For this example, the hydraulic detention time at ADWF is:



                          HT°(1875)(0.252)°°-3'3dayS

                              = 7.5 hours


12. Determine  the organic loading  per unit volume.  The volume required in the
    aeration basin  for 1 mgd flow is:

                    Volume = Q • HT = 1(0.313) = 0.313 mil gal = 41,844 cu ft

    where:   Q     =    influent flow rate, mgd

    The BOD- loading is:

                             (1)(8.33)(150) = 1249 Ib/day
                                  4-11

-------
     The BOD5 load per 1000 cu ft is:
                            1249
                               j  = 29.9 lbBOD5/l000 cu ft/day
13.  Determine the  sludge wasting schedule. Sludge is wasted from the system from
    two sources: (1) solids contained in the effluent from the secondary sedimenta-
    tion tank, and (2) intentional sludge  wasting  from the return sludge or mixed
    liquor. The sludge to be wasted under steady state conditions can be calculated
    from the solids retention time. The total sludge  wasted per day is:

                                 X  + W-X)                             (4-6)
     where:   S      =    total sludge wasted in Ib/day,

              W     =    waste sludge flow rate, mgd

              X~     =    effluent volatile suspended solids, mg/1, and

              X     =    waste sludge volatile suspended solids, mg/1

     The inventory of sludge in the system is:

                          I =  8.33(Xj • V)                                 (4-7)

     where:   I      =    inventory of VSS under aeration, Ib, and

              V     =    volume of aeration  tank, mil gal
     The solids retention time is defined as:
                                                                          (4-8)

     In this case, application of Equation 4-7 yields:

                               I = 8.33(1875X0.313) = 4889 Ib VSS

     Using Equation 4-8 and a design 6  of 8.78 days, the sludge wasted from the
     system is:

                                  S = 4889/8.78 = 557 Ib/VSS day
                                   4-12

-------
     The sludge contained in the effluent at 1 mgd can be calculated assuming that the efflu-
     ent volatile suspended solids is equal to 12 mg/1:

                             8.33(1) (12)= lOOlbVSS/day

     By difference, the Ib of MLVSS to be wasted from the mixed liquor or return sludge is:

                             557 - 100 = 457 Ib VSS/day


           4.3.3.1  Effect of Temperature and Safety Factor on Design

The design example presented in the previous section provided one solution  to a set of
stated conditions.  Alteration  of  the lowest temperature  at  which nitrification will be
supported, or the design safety factor,  or the wastewater strength, or the assumption of
different kinetic constants can materially alter the design.

To give one illustration, Table 4-2 has been prepared using  differing safety  factors (2.0 to
3.0) and differing minimum wastewater  temperatures with design calculations to derive the
computed quantities shown. Assumptions have been made for illustrative purposes as to the
allowable MLSS. Allowable  mixed liquor levels  are  a  function of sedimentation  tank
operation.  The  mixed liquor level  that can  be  maintained will  be affected  by reduced
sedimentation efficiency at lower temperatures.  Consideration of aeration tank-secondary
sedimentation tank  interactions is presented in Section 4.10.

As can be seen  from Table 4-2, low temperature applications (10 C) of combined  carbon
oxidation-nitrification in complete mix activated sludge systems require very long hydraulic
residence times to achieve favorable  conditions for  nitrification. This factor was one of the
reasons for the  development  of separate stage nitrification systems. As temperatures  rise,
required residence times are materially reduced. At  20 C, less than five hours is required for
virtually complete nitrification in  the specific case examined. While it is possible to design
for nitrification  using the relatively low  detention times given in Table 4-2 for 20 C, special
attention must be given to oxygen transfer as a very high oxygen demand is expressed per
unit volume. Considerations for oxygen transfer are given in Section 4.8.

           4.3.3.2 Consideration in the Selection  of SF

In introducing the safety factor  concept to the design  of biological treatment systems,
Lawrence and McCarty29 noted that the SF was necessary to achieve high efficiency of
treatment, to insure process stability and to provide resistance to toxic upsets. Excessively
high safety factors resulted in higher operating and capital costs. It was noted that the safety
factor concept had  been implicitly incorporated into treatment plant design practice by the
selection of solids retention times in excess of 0m
                                            c
                                        4-13

-------
                                            TABLE 4-2
                            CALCULATED DESIGN PARAMETERS FOR A 1 MGD
                              COMPLETE MIX ACTIVATED SLUDGE PLANT
Minimum
temp, for
nitrification,
C

10


15


20

Maximum
possible
nitrlfier.
growth rate,
>»N . day-1

0.175


0.285


0.465

Assumed
allowable
MLSS/MLVSS
mg/I
2,000 s^
//
s' 1, 500
2,500 s'
^<^
^^ 1, 875
3,000 ./
//
s/ 2, 250
Safety
Factor,
SF
2.0
2.5
3.0
2.0
2.5
3.0
2.0
2.5
3.0
Design
solids
retention
time, days
e?
11.5
14.3
17.2
7.0
8.8
10.5
4.3
5.4
6.4
Steady
state
effluent
NH+-N,
mg/1
0.23
0.15
0.11
0.40
0.27
0.20
0.73
0.49
0.36
Organic
removal
rate,
Ib BODrem/
Ib MLVSS-day
0.21
0.19
0.17
0.29
0.25
0.22
0.44
0.36
0.32
Hydraulic
retention
time, a
hours
11.0
12.8
14.0
6.4
7.5
8.5
4.4
5.2
6.0
BOD5
loading
(volumetric) .
lb/1000/cf/day°
20.5
17.5
15.8
34.9
29.9
26.5
51.5
43.0
37.3
* AtADWF

b 62.4 lb/1000 cf/day = kg/m3/day

-------
Because the SF concept  is relatively  new, there is no plant scale experience with its
application  accumulated as yet on which to base broad recommendations. Rather, kinetic
theory  itself is used in this section to establish minimum factors of safety considering the
desired degree of nitrification under steady state and transient  load conditions. It must be
emphasized that these are minimum values and individual designs may exceed these values
for a variety of reasons. For instance, the presence of industrial wastes may adversely affect
nitrification rates, requiring conservatism in the selection of the SF.

Figure 4-2 provides a wider array of safety factors for the design example presented in Table
4-2. As may be seen, the selection  of the SF has a marked effect on the calculated steady
state values of ammonia in the effluent. If relatively complete nitrification is to be obtained
(at steady-state) resulting in 0.5-2 mg/1  of ammonia nitrogen in  the effluent, a minimum SF
of 1.5 is appropriate for application to complete mix  activated sludge  systems. Further,
effluent values for a comparable plug flow system  are also shown in Figure 4-2 (see Section
4.3.5 for plug flow  data). As  may be seen, complete mix systems  have higher effluent
ammonia levels than plug flow systems at the same SF.

In  all practical applications,  waste treatment plants do not  operate at  "steady state."
Significant diurnal  variation in the nitrogen loading on such  systems occurs. Figure 4-3
shows the diurnal variations in influent flow and TKN loading experienced at the Chapel
Hill, N.C. treatment plant. The ratio of the maximum TKN loading to the average was 2.17,
while the ratio of the maximum to minimum was 6.72. The Chapel Hill system is a relatively
small system (1.8 mgd) with high peak to average ratios for all constituents.30 The variation
in  load for each community  will be a function  of the unique  characteristics  of that
community (see Section 4.8), and data must be individually developed for each situation.

TKN load  variations  have a  significant impact  on nitrification kinetics, and ammonia
bleedthrough can occur under peak load situations.31>32 Kinetic theory  can  be applied to
these situations,  however, and the safety  factor  established at levels which will  prevent
ammonia bleedthrough  from causing significant deterioration of effluent quality.

A mass  balance on nitrogen in the organic and ammonia form can be made at any  time
during  a diurnal cycle which states that the influent TKN load is equal to the effluent
ammonia load plus that nitrified in the complete-mix  reactor during any time,  At:

                       NQQAt = qNfX 1 VAt + N {Q At                              (4-9)


where:    N     =    influent TKN concentration, mg/1,

          N.    =    effluent ammonia nitrogen concentration, mg/1,

          Q     =    influent or effluent flow rate, mgd,

          At    =    time increment,

                                        4-15

-------
   V     =   volume of aeration basin, mil gal,
   f     =   nitrifier fraction of the mixed liquor solids
                       FIGURE 4-2
  EFFECT OF THE SAFETY FACTOR ON STEADY STATE EFFLUENT
      AMMONIA LEVELS IN SUSPENDED GROWTH SYSTEMS
2:  3
   0
                                         I
                                              I
                                       A.  at  20C
                      COMPLETE  MIX
    1.0   1.2   1.4   1.6   1.8  -2.0  2.2  2.4   2.6   2.8   3.O
                     SAFETY  FACTOR, SF
  2.5
 I
t-
3:
  1.5

  1.0

  0.5

  0
                                         I     I     r
                                        B.  at  10 C
                    COMPLETE  MIX
    1.0   1.2   1.4
                    1.6   1,8   2.0   2.2  2.4  2.6  2.8   3.0
                     SAFETY  FACTOR, SF
                            4-16

-------
                          FIGURE 4-3

         DIURNAL VARIATIONS AT THE CHAPEL HILL, N.C.

         TREATMENT PLANT (AFTER HANSON, ET AL. (30))
  zoo



  180



  I6O
Ul
v> 120

S
Uj
i. IOO
   8O
                       I
          T
Ul
o
(t
UJ
a.
 6O
   40
                 I
                                I
                                         I
                                                  I
                                  I
       2400
             04OO
0600
                                                20OO
                              1200     1600

                              TIME

        DIURNAL  VARIATION IN WASTEWATER  FLOW
                                                        2400
22O



ZOO



180




160



I4O




IZO




100
Ul
o  60
Ul
   40
   20
                T
                   INFLUENT

                     LOAD
                                     INFLUENT

                                     CONCENTRATION
         I
               _L
  I
          I
                                                        I
       2400
             0400
0800
 I20O

TIME
1600
ZOOO
2400
 DIURNAL VARIATION  IN  NITROGEN  LOAD  AND CONCENTRATION

                             4-17

-------
This equation neglects synthesis terms, assumes all influent organic N is hydrolyzed, and
neglects  terms relating  to the rate of change  of ammonia concentration in the reactor.
Numerical solution techniques are available to handle transient load effects more exactly.32
Equation 4-9, however, is useful for approximating the effects of transient loads.

Equation 4-9 may be solved for Nj, by substitution for the terms for nitrification rate, q-^
and the  term fXjV, representing the inventory of nitrifying organisms. The inventory of
nitrifying organisms can  be  related  to the  solids retention  time through  the  following
equation:
                                A        fX, V
                                d           i - _                          (4.10)
where:    N      =   24 hr-average influent TKN, mg/1

          N,     =   24 hr-average effluent NH.  - N, mg/1

          Q      =   mean flow rate (ADWF), mgd, and

          YN    =   nitrifier yield coefficient, Ib VSS/lb NH* -N removed.


The term C>YN(NO-NI) represents the quantity of nitrifiers grown per day, which must be
wasted each day to establish a steady-stage solids retentiofi time, Q £. The average terms, No
and Nj,  are  flow weighted averages  of nitrogen concentration of an entire  day (the
equivalent of composite samples). Q represents the average dry weather flow (ADWF).
The nitrification rate from Equations 3-20 and 3-24 is:
                                             N

Substitution of Equations 4-10, 4-1 1 and Equation 3-29 into Equation 4-9 yields:
Equation 4-12 can  be used to solve for Nj over a 24-hr cycle since all other quantities in
Equation 4-12 are known or can be estimated.  Initially, Nj can be estimated to be the
calculated steady-state value. Once Equation 4-12 has been applied to generate a 24-hr cycle
of NI  values, a  new value of Nj may  be  calculated. If N\ differs significantly from the
initial assumption, the calculation process can be repeated.
                                        4-18

-------
 Equation 4-12 has been applied to the variations in load observed at Chapel Hill, and using
 the design information used to generate Table 4-2 at a temperature of 15 C. The results of
 this analysis are plotted in Figure 4-4 for three different assumed safety factors, 1.5, 2.0,
 and 2.5. As may be seen, the assumption of the safety factor has a marked effect on the
 average effluent ammonia content, Nj. For this particular case, the ratio of peak to average
 TKN loading was 2.2; the SF had to exceed this ratio (2.5) to produce an effluent that had,
 on the average, less than 1 mg/1 of ammonia-N.

 The application of Equation 4-12 to several other such cases*, showed the  same effect;
 namely, the minimum safety factor should equal or exceed the ratio of peak ammonia load
 to average load to prevent high ammonia bleedthrough at peak loads. This statement may be
 used as "a rule of thumb" for designing suspended growth nitrification systems operated in
 the complete mix mode.

 A flow equalization  procedure applicable  to reducing  diurnal peaking on nitrification
 systems is  presented  in Chapter 3 of the Process Design Manual for Upgrading Existing
 Waste-water Treatment Plants.^ By incorporating flow equalization into  treatment plants,
                                     FIGURE 4-4
          EFFECT OF SF ON DIURNAL VARIATION IN EFFLUENT AMMONIA
 o
 Ul
 o
 o
 o

I
 It
 lu
     20 r
                          24 hr  average
                          composite  IMH^-N
                          concentration
                 04 OO
08OO
     I20O
TIME,  HR
I6OO
2000
                       2400
                                       4-19

-------
the safety factor used in kinetic design of the nitrification tanks may be reduced. Case
examples  for  treatment  plants incorporating flow equalization are presented  in Sections
9.5.1.1, 9.5.1.2 and 9.5.2.1.

     4.3.4 Extended Aeration Activated Sludge Kinetics

The  procedure presented in Section 4.3.3 for complete mix  activated sludge kinetics is
directly applicable  to extended aeration activated  sludge. Extended aeration systems are
usually  operated at such long solids retention times that except during cold temperatures
(5-10 C) nitrification is usually obtained in properly  operated systems.

     4.3.5 Conventional Activated Sludge (Plug Flow) Kinetics

The  approach for  conventional activated sludge plants is similar to that for complete mix
plants with  the exception of the  equations used to  predict effluent quality. The plug flow
model may  be applied to approximate the hydraulic regime in  these plants. The Monod
expression for substrate removal rate (Equation 3-24) must be integrated over the period of
time an element of liquid remains in the nitrification tank. The following is a solution for
plug flow kinetics that can be adapted to this problem as shown :^9


                1           ^N^o'Nl-1
               —- =  	     for r < 1
               0                         No
                c    (N  -N,) + KXTln  —-
where:   r      =   recycle ratio (or return sludge ratio).
                               0
                -   = - ^— - —     forr
-------
A  typical DO and  nitrification  pattern for plug  flow  tanks where  aeration capability is
limited in the front end of the  tank  is presented in Figure 4-5, where an aeration tank
profile for DO and ammonia nitrogen is plotted. As may be seen, nitrification is inhibited in
the first portion of  the tank, because of the DO suppression due to carbon oxidation. Once
the DO rises, the ammonia level falls at a reaction rate that approximates zero order, a
reactor order predicted  by kinetic theory (Section 3.2.7). It is notable that  if sufficient
aeration  capability  had  been available in the head end of the tank, virtually complete
nitrification probably would have been obtained.

Thus, the first portion of plug flow tanks may be  ineffective for nitrification, reducing the
effective  contact time for nitrification. If oxygen supply limitations are present in the head
end of the tank, the plug flow  type reactor's advantage over the complete mix reactor is
reduced.

The degree to  which full-scale  nitrification tanks approach  plug-flow operation can be
examined through reactor diffusion theory.34'35 Reactors can be characterized by an axial
disperson number, D/uL, where D is the axial disperson coefficient in square ft per hr, u is
the mean displacement  velocity  along the tank length, in feet per hr,  and L is the tank
length, ft. In the calculation of the axial disperson number, u and L are known for any


                                    FIGURE 4-5

       DO AND AMMONIA NITROGEN PROFILE IN A PLUG-FLOW SYSTEM
                      (AFTER NAGEL AND HAWORTH (33))
                            L
AERATION
   TANK
                                      -BAFFLES-
£3
20
"\
p l5

~
l 10
_j_~
3;
^ 5
n
b 	 ^ i i i i i
*~~" » """
^^QL
% ^/-^^M* ^^^ ^^^ "~ "
— *^*^ ~~
P\Q 	 C^^ *^^
^•^-^ ^d£.
— T' ^^, —
^r ^^^W ^
s + r *^o,^
_ ^**s N"4 *^-^-
°"* 1 I.I 1 1
f.3
2.0 I
Z
1.5 uj
X
X
1.0 Q
Q
kl
°'5 i
n Q
                                              ISO          200          250
                      DISTANCE ALONG  TANK,  FEET
                                        4-21

-------
particular plant  design and D must be measured.  A valid empirical relationship for D for
both fine and coarse bubble diffused air plants is as  follows:-^

                                D = 3.118 W2(A)°'346                          (4-15)

where:      W   =   tank width, ft and

            A   =   air flow per unit tank volume, in
                     standard cubic feet per minute
                     per 1,000 cu ft.

The axial  disperson coefficient, D/uL is zero for true plug flow plants and infinite (<*=) for
true complete mix plants. Plants with D/uL between 0 to 0.2 are usually classed as plug flow
reactors, while for complete mix systems, D/uL is usually in the range from 4.0 to oo .36
As  an example  calculation, the  Central Contra  Costa Sanitary District (CCCSD)  plant's
nitrification tanks (Section 9.5.2.1) have the following characteristics:

                        Air How (av.) = 51.1 SCFM/1000 CF
                        Width = 35 ft
                        Area of tank = 525 sf
                        Length (all 4 passes) = 1080 ft
                        Flow each tank (4 passes) @ 50% recycle = 22.5 mgd

From the above data, the mean displacement  velocity can be calculated to be 239 ft/hr.
From Equation 4-15, the diffusion coefficient is:

                        D = 3.1 18 (35)2 (51.1)0.346 = 14,939 ft2/hr
          Therefore:
                        D/uL = 14,898/239 (1080) = 0.058

Therefore, the CCCSD nitrification tanks closely approach a plug flow reactor.

Equation  4-15 can be  utilized to evaluate mixing  in actual  plant designs to determine
whether they approach  plug flow closely enough  to allow use of Equation 4-15 to describe
nitrification. It is probable that most plants operated in the conventional mode do approach
plug flow. For those plants with intermediate values of D/uL, complete mix kinetics can be
employed which yield conservative answers.

The hydraulic configuration of nitrification tanks can also be designed to discourage back
mixing. A series of complete  mix tanks can approximate a plug flow reactor. In the case
example for Canberra, Australia (Section 9.5.2.2) complete mix reactors are used in series
for nitrification.  Absolute prevention of back  mixing is provided by virtue of the mixed
liquor overflowing weirs  between  reactors.   Available head at the  site  was utilized,

                                         4-22

-------
eliminating the need for mixed liquor pumping.

          4.3.5. 1 Considerations in the Selection of the Safety Factor

The factors affecting the choice of the SF for plug flow activated sludge are similar to those
for complete mix applications. Diurnal peaking in load has an important influence  on the
choice of the SF, although kinetic models have not been extended to handle diurnal loads in
plug flow systems at the present time. It can be expected that the effects of diurna} loads on
plug flow  systems will be similar to those for complete mix systems as when the effluent
ammonia nitrogen level rises to 2 to 3 mg/1 or above, the rate of removal becomes a zero
order reaction (unaffected by ammonia nitrogen concentration). In zero order reaction
situations, differences between  plug  flow  and  complete  mix  kinetics are negligible.
Therefore, the  adoption  of the criteria advanced for complete  mix  systems (that the
minimum SF equal or exceed the ratio of peak ammonia load to average  daily load),  should
prevent  high ammonia bleedthrough during diurnal peak loads. The problem of low
dissolved oxygen due to carbonaceous load in the head end of plug flow systems should be
considered in  aeration  design for combined carbon oxidation-nitrification applications;
indeed, this factor alone would justify a conservative safety factor.

         4.3.5.2 Kinetic Design Approach

The kinetic  design  approach  for  plug   flow (conventional) plants is  identical to  that
presented in  Section 4.3.3, excepting in  Step 9,  where Equation  4-13  is used instead of
Equation 3-24.  If a  portion  of the nitrification  tank is  rendered ineffective by DO
suppression at its head end, then only the sludge inventory maintained under adequate DO
conditions should be used in the calculation of 0 r or the SF.
                                             L*

     4.3.6  Contact Stabilization Activated Sludge Kinetics

Gujer and  Jenkins37,38 have developed  the  kinetic design procedure for nitrification in
contact-stabilization activated sludge plants. The procedure described herein is a summary
of their approach, and the jeader is referred to their publications for theoretical bases.

The overall nitrifier growth rate in the contact stabilization process  is the weighted mean of
their growth rate in the contact tank and in the stablization (sludge reaeration) tank:
where:    JUN> ju , p.  =  growth rate of the nitrifiers in the overall process,
                         in the contact and stabilization tanks respectively
                         (day'1).

              C, B    =  the fractions of total sludge in the contact and
                         stabilization basins respectively.
                                         4-23

-------
Gujer and Jenkins used Equation 4-16, and Monod type expressions for complete mix tanks
to develop a graphical  solution for nitrification in contact stabilization (Figure 4-6). In
Figure 4-6, the efficiency of nitrification, 7?m-t, is defined  as a fraction by:


                                 (N03}c
                          n . =  	—                                     (4-17)
                           ""   OK.?.


where:   (NO- )    =  NO- - N level in the contact tank, mg/1, and
             j C        J
         (NO-)    =  NOZ - N level in the stabilization tank, mg/1.


         4.3.6.1 Design Example

As a design example, consider a 1 mgd contact stabilization plant operated at a minimum
temperature of 15 C. Influent BOD5 is 150 mg/1, including solids handling returns to the
primary. Total Kjeldahl Nitrogen is 30 mg/1, of which 21 mg/1 is ammonia -N and 9 mg/1 is
organic -N. The wastewater has an alkalinity of 210 mg/1. The effluent requirement is not
more  than  10 mg/1 reduced soluble nitrogen  (organic and ammonia). The procedure is as
follows:

     1.   Establish a reasonable safety factor for nitrification, say 2.5 as in Section 4.3.3.

     2.   Establish the minimum mixed liquor DO; assume 2.0 mg/1 as in Section 4.3.3.

     3.   Establish  the maximum growth rate of nitrifiers,  assuming for the moment that
         there is sufficient alkalinity in the  wastewater to buffer the nitrification pH to
         greater than 7.2 (see step 14). Therefore,

                  AN = 0.285 day'1  as in Section 4.3.3, step 4.

     4.   Calculate  the minimum solids retention time, the design solids retention  time and
         the actual nitrifier growth rate (as in  Section 4.3.3, Steps 5, 6, 7):

                                0m = 3.51 days


                                0d = 8.78 days
                                 C
                                    = 0.114day"1
                                        4-24

-------
                            FIGURE 4-6
              NITRIFICATION EFFICIENCY AS A FUNCTION OF
          PROCESS PARAMETERS (AFTER GUJER AND JENKINS (37))
Key:
         1
       /             2             3
    SLUDGE RECYCLE  RATIO  R/Q

DESIGN EXAMPLE

               4-25

-------
5.   Calculate the organic removal rate, as in Step 10, Section 4.3.3:

              qb = 0.252 Ib BOD rem/lb MLVSS/day

6.   Calculate the VSS produced per unit of wastewater treated:

               (SQ - Sj) MN/qb = (150 - 2)(0.114)/0.252 = 67.0 mg/1

7.   Compute  the  nitrogen  incorporated  into VSS, assuming  12  percent nitrogen
     incorporated into VSS:

              0.12 (67.0) = 8.0 mg/1 N

8.   Compute the  soluble N content of the effluent, assuming no denitrification. The
     effluent soluble N = the total influent N minus N incorporated into VSS:

              30 - 8.0 = 22.0 mg/1 soluble N

9.   Calculate the soluble organic N in the effluent.  Gujer  and Jenkins found that 40
     percent of the influent organic N appeared in soluble form in the effluent:

              0.4 (9) = 3.6 mg/1 organic N

                    ^ 4.0 mg/1 organic N

10.  Calculate ammonia nitrogen in the effluent under steady state conditions:

              Total reduced N - organic N = 10-4 = 6 mg/1 NH^ -N

11.  Calculate nitrate nitrogen in the effluent and in  the  contact tank under steady
     state conditions:

              Total soluble N - total reduced N = (NCQ
                                                    j C

              (N0~)c = 22-10=12mg/l

12.  Calculate the required nitrification efficiency from Equation 4-17:

               7?nit=l 2/(12 + 6) = 0.667

     In this calculation, it is assumed  that the concentration of nitrate nitrogen in the
     stabilization tank (NO-) totals 18 mg/1, since the contact tank concentration is
     12  mg/1 and  with the assumption  that the  6  mg/1 of ammonia  nitrogen is
     completely nitrified in the stablization tank.

                                   4-26

-------
13.  Calculate the required sludge recycle ratio. Assume the fraction of biomass in the
    contact tank is 15 percent (C= 0.15). The dimensionless number A, is used in the
    calculation; A is defined as follows:
                           A =
     For this example:
                           A =
                                 SF
                                     -C
                                 0.15
                                    + 1
(4-18)
                               _L  -0.15
                               2.5
                                     + 1 = 1.60
    The required sludge recycle  ratio,  R/Q,   depends on  the value  of A  and the
    required nitrification efficiency as follows:
                            R/Q =
                                                                       (4-19)
     where:      R    =   recycle flow rate, mgd
                 Q    =   influent flow rate, mgd
 14.
     For this example:
                          R/Q =
                                      0.667
                                   1.60(1-0.667)
                                             •= 1.25
Since Q  has  been assumed to be 1 mgd, the return activated sludge rate is 1.25
mgd.

This example is also worked graphically in Figure 4-6. The top part of the figure is
used first by entering the abscissa with the value of the SF and rising vertically to
the chosen value  of C and  then reading  the value of A on the ordinate. The
bottom  part  of Figure  4-6  is  then used; the  nitrification efficiency, nnit> *s
entered  on the  ordinate  and  traveling  horizontally to the value  of A just
determined and then finding the required recycle ratio on the abscissa. Figure 4-6
also demonstrates  a general result; in order to obtain high nitrification efficiency,
a higher than normal sludge recycle ratio must be employed.

Check the buffering of the wastewater. The quantity  of ammonia nitrified is
reflected by the level of nitrate  in the process effluent. Approximately 7.14 mg/1
of alkalinity  as  CaCC>3 is destroyed per mg/1 of NH^-N  oxidized. The alkalinity
remaining after nitrification would be at least:
                                    4-27

-------
              210-7.14(12)= 124mg/l as CaCO3

    This should  be sufficient residual alkalinity to maintain the pH above  7.2  for
    coarse bubble aeration  systems. If a fine bubble aeration system were  chosen,
    chemical  addition  would  be   required and  the  dose  estimated from  the
    procedures discussed in  Section 4.9.2. Alternatively, a lower operating pH could
    be used with  a longer aeration period.

15.  Calculate  required reactor volumes. As in Section 4.3.3, assume the mixed liquor
    content in the contact tank is 2500 mg/1 at a volatile content of 75 percent. The
    mixed liquor volatile suspended solids in stabilization can be obtained from the
    balance:

                       (Q + R) X,, = RXc                                  (4-20)
                                c     s

    where:      X    =   contact MLVSS, mg/1, and
                 C
                X    =   stabilization MLVSS, mg/1.
     Therefore: X =1875   l + L25   = 3375 mg/1
                 s           1.25
    The total sludge inventory can be calculated from the following equation for
    substrate removal rate:
                                  Q(S  -Sj)
                           q  =	2	L.                              (4.21)
                            b      2XV
     where:      SXV   =    total inventory of MLVSS in the contact and
                             stabilization tanks, Ib
     therefore:   2XV   =    -             = 4889 Ib
                                 (.252)
     Of this inventory, 15 percent is in the contact tank:

                                 0.15(4889) = 733 Ib MLVSS

     The remainder is in the stabilization tank:

                                 4889 - 733 = 4156 Ib MLVSS
                                  4-28

-------
         The volume in contact is:

                            V  =733/(8.33)(l875) = 0.047 mil gal
                             C

         The volume in stabilization is:

                              Vg = 4156/(8.33)(3375) = 0.148 mil gal


         The total volume is 0.148 + 0.047 = 0.195 mil gal

     16. Calculate the residence time in contact.

                              0  =V  /Q = 0.047 days = 1.1 hr
                               c    c

     17. Calculate the sludge wasting schedule. See Section 4.3.3, Step 13.

As can be seen from Figure 4-6, the design of contact stabilization for nitrification is highly
sensitive to the safety factor chosen and the sludge recirculation ratio.  A wide  range of
alternate designs  can be  derived  from variation in these parameters. Required reactor
volumes are also sensitive to the assumed growth rate of nitrifiers, creating a need for kinetic
data of high accuracy when designing for contact stabilization.

The  design procedure described is based on the  assumption  of steady  state operation.
Diurnal variations  in load will cause  average  effluent ammonia levels  to exceed those
calculated  above. To compensate for this, it would be necessary to  use an even higher SF
than  assumed  in  the above example. Regardless  of the safety factor  chosen, contact
stabilization  plants cannot be expected to completely nitrify except under the normally
impractical condition of very high recycle rates (R/Q = 4.0 and above). However, at high
recycle  ratios the major advantage of contact stabilization is lost because the sludge in the
stabilization basin becomes more dilute and the overall basin volume requirements approach
those of the conventional process. This limits the application  of contact stabilization to
situations where only partial nitrification is required.

A further  limitation  on nitrification  in contact  stabilization plants is the incomplete
hydrolysis  of organic nitrogen occurring in the short detention time contact tank. As noted
under step  9 above, about 40 percent of the influent organic nitrogen appears in the process
effluent. Conventional or complete mix  activated sludge plants, on the  other hand, have
sufficient contact time to hydrolyze  the bulk  of  the organic  nitrogen to ammonia thus
making the nitrogen  available to nitrifiers and  leaving very little organic nitrogen  in the
effluent.

The short contact time of the contact tank can create problems in the sedimentation tank.
The mixed liquor solids are not well stabilized in the contact tank prior to sedimentation.

                                         4-29

-------
Denitrification  activity in the sedimentation tank is therefore greater than in conventional
or complete mix plants and floating sludge may be the result. Procedures for circumventing
the floating sludge problem are discussed in Section 4. 1 0.

     4.3.7 Step Aeration Activated Sludge Kinetics

Because of backmixing, the step feed pattern of step aeration plants causes the kinetics of
such plants to more closely approach complete mix than plug flow. As a result, the design
approach  developed for complete mix (Section 4.3.3) can usually be  employed for step
aeration plants as a reasonable approximation. In those step  aeration plants where influent is
fed to the last pass (as in  Figure 4-1), there is the danger that there will be insufficient time
for the organic nitrogen to be hydrolyzed prior to discharge, resulting in elevated quantities
of organic nitrogen in the effluent. Further discussion of this effect is presented in Section
4.3.8.2 which is a description of an operating step aeration plant.

     4.3.8 Operating Experience with Combined Carbon Oxidation-Nitrification in Suspended
          Growth Reactors

While activated sludge-type systems are  commonly used in England to obtain  dependable
nitrification,  their use in the U.S. has not been widespread. Early U.S. activated sludge
plants of the conventional design nitrified in the warmer months of the year or if they were
underloaded. But nitrification became unpopular because of the additional aeration power
cost and the propensity of some sludges  to float in the sedimentation tank when nitrifying,
and it was questioned whether the added expense was worth it in many cases.39,40 AS a
consequence, ways and means were sought to prevent nitrification rather than to encourage
it  through  increasing  organic  loading or  through  tapered aeration  or  by  picking
modifications  of  the  process which were  less  favorable  for nitrification.  This early
experience with the process may have led to uncertainty about its reliability.

Nonetheless, there have  been several  plant-scale  operations  in  the U.S.  which  have
demonstrated the viability of the process. The purpose of this section is to review some of
these cases. Other case examples are presented in Sections 9.5.1 and 9.5.2.

          4.3.8.1 Step Aeration Activated Sludge In a Moderate Climate

The Whittier Narrows Water  Reclamation Plant is a 1 2 mgd activated sludge plant designed
and operated by the Los Angeles County  Sanitation Districts. The basic purpose of the plant
is to reclaim water for groundwater recharge; the entire effluent of the plant is discharged to
spreading basins for recharging groundwater aquifers.
Design data for the plant  are summarized in Table 4-3.41  j^g  piant operates at either a
constant flow rate or a constant oxygen demand load by pumping wastewater from a trunk
sewer and returning grit, skimmings, primary and waste activated sludge back to the trunk.
No solids handling facilities are provided as the solids returned to the trunk are processed at

                                         4-30

-------
a downstream  primary treatment plant. The plant was constructed in 1961 at a cost of
$1,700,000; this cost includes influent pumping, foam fractionation and effluent pipelines
in addition to those items shown in Table 4-3.

Recently,  the  plant has been  operated in a manner promoting nitrification. The three
aeration tanks are operated  in a  3 pass  series configuration; two-thirds of the primary
effluent is added along the first pass, with the head end of the first pass operating as sludge
reaeration.  One-third of the primary effluent is  added to the second pass. The plant
                                    TABLE 4-3

                                  DESIGN DATA
               WHITTIER NARROWS WATER RECLAMATION PLANT
 Plant Flow

 Raw Wastewater Loadings

   Biochemical Oxygen Demand (BOD)
   Suspended Solids '(SS)

 Primary Sedimentation Tanks
   Number
   Overflow Eate
   Detention Time
   BOD Eemoval
   SS Removal

 Air Blowers
   Number
   Discharge Pressure
   Capacity - Total

 Aeration Tanks
   Number
   Detention Time  (@ 12 mgd)
   BODs loading
 Final Sedimentation Tanks
   Number
   Overflow Rate (@ 12 mgd)
   Detention Time (@ 12 mgd + 33% return)
   Weir Rate (@ 12 mgd)

 Chlorine Contact Chambers
   Number
   Detention Time (@ 12 mgd) including time in Foam
     Fractionation Tank & Effluent Pipe
   Chlorine
    12 mgd (0.53 m3/sec)
   270 mg/1
   280 mg/1
     2 (1 stand-by)
  2000 gpd/ft2 (82 m3/m2/day)
   1.1 hr
    35%
    60%
   6.5 psig (0.46 kgf/cm2)
29,500 cfm (840 m3/min)
     3
   6.0 hrs
    45 lb/1000  cf/day)
      (0.18 kg/m3/day)
   790 gpd/ft2 (32.2 m3/m2/day)
   1.7 hrs
12,000 gpd/ft (150 m3/m/day)
    43 min
   600 Ib/day (272 .kg/day)
                                      4-31

-------
performance  reflects its very careful control and operation; operating data for a one-year
period are summarized in Table 4-4." While organic nitrogen data are not available, the data
indicate that year-round complete nitrification has been obtained. Climatic conditions for
this  California treatment plant are very favorable  for nitrification  as  average monthly
wastewater temperatures did not fall, below 21 C for the year examined.

         4.3.8.2 Step Aeration Activated Sludge in a  Rigorous Climate

The Flint, Michigan sewage treatment plant is being upgraded to comply with requirements
of the Michigan Water  Resources Commission which  mandate nitrification for the purpose
of preventing DO  depletion in the Flint River. In connection  with this upgrading, a large
scale  test of combined carbon oxidation-nitrification was  conducted  with  the existing
activated sludge plant over a ten-month period to determine design conditions for the plant
upgrading. The minimum  wastewater temperature tested was 7 C.6 During the test, ferric
chloride and  polymer were added to the primary  treatment stage for phosphorus removal.
This also had the effect of reducing the organic loading to the  aeration tank.

The existing plant  had  three aeration tanks, each with four passes providing a 750,000 cu ft
(21,240 cu m) capacity. With an average design  BOD5  loading of 24,500 Ib/day (11,110
kg/day) to the aeration tanks at a 20 mgd (75,700 cu m/day) flow, the aeration tank load
was 32.7 lb/1000  cu ft/day (523 kg/1000 cu m/day). Flows  were varied to the facility,
however, to provide variation in loads. Three secondary sedimentation tanks were provided
having a design overflow rate of 678 gal/sq  ft/day (27.6m^/m2/day) at ADWF. The plant
was usually operated in a step aeration mode, with one-half the influent directed to the head
ends of the second and third passes.

Average effluent qualities for eight  months of the test are shown in Table 4-5. While nitrate
and nitrite are not shown, it is reported that a relatively  good balance between ammonia
removal and  nitrate production was obtained.6 Nitrite nitrogen  was  always less than 0.1
mg/1. The appearance of high concentrations of organic nitrogen was attributed to the low
rate  of hydrolysis of organic nitrogen  compounds.^ It is probable that the provision  of
feeding wastewater to the last pass  exacerbated the problem by causing insufficient contact
time  for that portion of the wastewater to complete the hydrolysis of organic nitrogen  to
ammonia.

The effect of temperature and solids retention are considered in Table 4-6. Effluent qualities
deteriorated  somewhat  with  colder temperatures, with only 75 percent ammonia removal
being obtained at  IOC. This ammonia bleedthrough  may have been due to diurnal peaking
in ammonia at the  relatively low solids retention time  employed (c.f. Section 4.3.3.2).

         4.3.8.3 Conventional Activated Sludge In a  Rigorous Climate

The Jackson, Michigan wastewater treatment plant is  a 17 mgd conventional activated sludge
plant designed for year-round complete nitrification.^ The existing plant was upgraded in

                                        4-32

-------
                                                 TABLE 4-4

                            NITRIFICATION PERFORMANCE AT THE WHITTIER NARROWS

                                  WATER RECLAMATION PLANT (REFERENCE 9)
Calen-
dar
Month
4

5

6

7

8

9

10

11.

12

1

2

3

Year
1973

1973

1973

1973

1973

1973

1973

1973

1973

1974

1974

1974

Flow,
med
10.4
(0.46)
11.7
(0. 51)
11.9
(0.52)
11.9
(0.52)
12.8
(0. 56)
13.4
(0.59)
13.5
(0. 59)
13.2
(0.58)
11.1
(0.49)
9.9
(0.43)
12.1
(0.53)
12.2
(0.54)
Recycle
ratio
0.39

0.40

0.45

0.46

0.45

0.43

0.41

0.45

0.52

0.60

0.49

0.48

Temp. ,
C
22

24

26

26

27

25

25

24

22

21

21

22

MLVSS,
mg/l
1st pass
2177

2337

2390

2603

2889

2850

2958

2791

2724

2675

2857

2888

3rd pass
1474

1778

2319

8092

2005

1938

2094

2000

1986

1971

2097

1982

SVI,
ml/g
78

64

66

78

77

64

77

62

55

114

88

75

days
13.1

15.4

17.2

37.5

33.4

40.2

20.5

9.4

10.7

16.1

9.8

9.4

HT?
hours
7.0

6.2

6.1

6.1

5.7

5.4

5.4

5.5

6.6

7.4

6.0

6.0

Air
Use,
MCF/day
29.3
(9600)
27.9
(9145)
29.3
(9600)
28.1
(9400)
27.5
(9010)
27.2
(8920)
27.6
(9050)
27.3
(8950)
25.5
(8360)
25.6
(8400)
28.9
(9470)
28.8
(9440)
COD,
mg/l
Primary
effluent
241

243

239

227

223

216

228

235

233

227

242

229

Secondary
effluent
62

47

39

32

30

34

34

33

35

27

31

41

Ammonia-N,
mg/l
Primary
effluent
21.6

23.5

20.8

20.1

18.8

19.2

21.5

21.9

21.8

21.4

22.1

21.2

Secondary
effluent
2.0

3.2

0.8

0.6

0.6

1.4

1.5

2.8b

1.9

0.4

0.9

1.0

Percent
Bemoval
91

86

96

97

97

93

93

87

91

98

96

95

u>
UJ
     Based on influent flow and entire aeration tank

     Blower trouble this month

-------
1973 in response to an order to improve treatment to a point where a minimum dissolved
oxygen  of 4.0 mg/1 could be maintained in the Grand River. An analysis of the assimilative
capacity of the reach indicated that this could only be done if the effluent were completely
nitrified to prevent discharge of NOD to the river.
                                TABLE 4-5

                AVERAGE NITRIFICATION PERFORMANCE AT
              FLINT, MICHIGAN FOR 8 MONTHS (REFERENCE 6)
Parameter
(all values in mg/1
except Temp. )
BOD5
SS
Total Kjeldahl nitrogen
Organic nitrogen
Ammonia nitrogen
Phosphorus
Temp. , C
Raw
wastewater

250
300
27.6
13. 3
14.3
15.4
7.2 to 18.3
Settled
wastewater

131
140
23.3
9.9
13.4
2.7

Secondary
effluent

13.6
24.1
7.8
6.1
1.7
2.3

                               TABLE 4-6

         EFFECT OF TEMPERATURE AND SOLIDS RETENTION TIME
    ON NITRIFICATION EFFICIENCY AT FLINT, MICHIGAN (REFERENCE 6)
Temperature,
C
18 and greater
13
10
7
Solids retention
time, days
4
4-5
6
10 - 12
NH. removal,
percent
95
87
75
50 (Lab) E
  Based on bench scale test results
                                  4-34

-------
Pilot studies 10 indicated that a combined carbon oxidation-nitrification system was more
economical than a two-stage activated sludge system. Design data for this plant are contained
in Section 9.5.1.1. The plant is operated in the conventional (or plug flow) mode. Table 4-7
summarizes the  plant  operating  data; since start-up  the  plant  has obtained complete
nitrification at daily temperatures as low as 8 C.42 A characteristic of this wastewater is that
the  primary effluent  is weak, both in terms of BOD5 and ammonia -N. The mixed liquor
concentrates very well due to the high inert concentration of the raw wastewater, allowing
high mixed liquor levels under  aeration.  Both the  weak wastewater  and the ability to
maintain  the  mixed  liquor at a  high  concentration allow nitrification to be obtained in
hydraulic retention times of less than eight hours even at temperatures of 8-10 C. Nitrate
and organic nitrogen values are not available.

This case history clearly demonstrates that combined carbon oxidation-nitrification can be
dependably accomplished at temperatures as low as  10 C.

4.4  Combined Carbon Oxidation-Nitrification In Attached Growth Reactors

The  two  attached   growth reactor systems seeing application   for  combined carbon
oxidation-nitrification in the U.S. are the trickling filter process and the rotating biological
disc process. Procedures for designing nitrification with these two systems  are described in
this section.

     4.4.1 Nitrification  with Trickling Filters in Combined Carbon Oxidation-Nitrification
          Applications

Trickling  filter  design  concepts  are discussed  extensively in  the  Technology  Transfer
publication, Process Design Manual for Upgrading Existing Wastewater Treatment Plants.^
Therefore, the following discussion is limited to the loading ranges that are applicable for
nitrification in trickling filters used in combined carbon oxidation-nitrification applications.

As is the case  for the  activated sludge system, the development and maintenace of nitrifying
organisms in a trickling filter is dependent  on a variety of factors including organic loading,
temperature, pH, dissolved oxygen and the presence of toxicants. However, in the case of
the  trickling  filter,  there has been  no comparable development  of kinetic  theory  for
combined carbon oxidation-nitrification that can be directly applied with any degree  of
confidence. The approach applied to date  has largely been empirical and relied mostly on
specification of an organic loading rate suitable for application to each media type.21

          4.4.1.1 Media Selection

The types of media currently available are  summarized in  Table 4-8. Rock  applications are
generally limited to four to ten feet in depth; the plastic and redwood media may be built in
towers of  15  to 25  ft in height  due to  their lighter  weight and  greater void space for
ventilation, affording  considerable space saving economies. Loading capabilities of trickling

                                        4-35

-------
                                             TABLE 4-7

                        NITRIFICATION PERFORMANCE AT THE JACKSON, MICHIGAN
                            WASTEWATER TREATMENT PLANT (REFERENCE 42)
Month
8

9

10

11

12

1

2

3

Year
1973

1973

1973

1973

1973

1974

1974

1974

Flow,
mgd
(m3/sec)
14.5
(0.63)
12.4
(0. 54)
13.2
(0.58)
12.2
(0. 53)
11.4
(0. 50)
14.0
(0.61)
14.2
(0.62)
18.4
(0.81)
Recycle
ratio
.38

.42

.38

.40

.42

.43

.49

.43

Temp. ,
G
21.7

20.0

17.2

15.6

12.2

11.1

10.6

11.1

MLSS,
mg/1
4320

4110

4390

4480

4560

4630

4800

4930

SVI,
ml/g
42

45

47

47

45

43

42

38

Sc,
days
15.6

16.4

16.7

18.6

16.4

10.3

11.0

11.1

HT*
hours
7.5

8.8

8.3

9.0

9.9

7.9

8.2

6.3

Air
use,
MCF/dayb
14

14

14

14

14

14

14

14

BOD5,
mg/1
Primary
effluent
75

82

84

94

85

97

104

90

Secondary
effluent
2.5

2.6

3

4

3

4

5

4

Ammonia-N,
mg/1
Primary
effluent
8.4

9.9

11.6

11.0

11.5

9.2

9.3

7.1

Secondary
effluent
0.6

0.7

1.2

0.8

0.6

0.7

0.6

0.5

Percent
Removal
93

93

90

93

95

92

94

93

o\
   Based on influent flow
   14 MCF = 4590 I/sec

-------
filter media are known to be related to the available surface area for biological slime growth.
Specific  surface, or the amount of media surface contained in a unit volume, is  a gross
measure  of the available surface for growth of organisms. Plastic media is available in higher
specific surfaces than  that shown in Table 4-8. Design practice  has been to avoid the use of
media  with  higher specific surface and lower voids due  to  the danger of clogging in
combined  carbon  oxidation-nitrification applications.  However, there has  been  recent
experience which indicates that medias with specific surfaces exceeding 35 sq ft/cu ft (115
m2/m3) have t,een use(} to treat domestic wastewaters without media clogging.43

                                     TABLE 4-8

     COMPARATIVE PHYSICAL PROPERTIES OF TRICKLING FILTER MEDIA


a
Media

Plasticb

Redwood*^

Granite

Granite

Blast Furnace Slag



Nominal
Size
(cm)
24 x 24 x 48
(61 x 61 x 122)
47-1/2 x 47-1/2 x 35-3/4
(121 x 121 x 51)
1-3
(2.5 x 6.5)
4
(10.2)
2-3
(5.1 x 7.6)

Unit
Weight
Ib/cu ft
(kg/mj)
2-6
(32-96)
10.3
(165)
90
(1440)
_

68
(1090)
Specific
Surface
Area
sq ft/cu ft
m2/m3
25-35°
(82-115)
14
(46)
19
(62)
13
(47)
20
(67)



Void Space
percent
94-97

76

46

60

49

        Reference 25
         Currently manufactured in the U.S. by: the Envirotech Corp., Brisbane, Ca.;
         the Munters Corp., Fort Meyers, Fla., and the B.F. Goodrich Co.,  Marietta,
         Ohio
        .Denser media may be used for separate stage applications, see Section 4.7.1.1
         Currently manufactured in the U.S. by Neptune-Microfloc, Corvalis, Or.

          4.4.1.2 Organic Loading Criteria

Observations of the effect of organic loading on nitrification efficiency in rock media and
trickling filters are summarized in Figure 4-7. The data are from the following full-scale and
pilot-scale plants: Lakefield, Minn.,25  Allentown, Pa.,25  Gainesville, Fla.,44,45 Corvallis,
Or.,4^ Fitchburg, Mass.,25 Ft. Benjamin Harrison,  Ind.,25 Johannesburg, South Africa,4^
and  Salford,  England.4**  Several  interesting  factors  affecting design  are evident.  First,
organic loading significantly affects nitrification efficiency. This is principally caused by the
fact  that the bacterial film in the rock becomes  dominated by  heterotrophic bacteria. The
relative high bacterial yield when BOD is removed causes displacement of the nitrifiers from
the film by heterotrophic organisms at high organic loadings.

As  opposed to  nitrification with activated sludge, the  breakthrough  of ammonia in  a
trickling filter is not abruptly affected by loading rate. For rock media, attainment of 75
                                         4-37

-------
percent  nitrification or better requires  the  organic loading  to be  limited to  10-12 Ib
BOD5/1000 cu ft/day (0.16 to 0.19 kg/m^/day). At higher organic loading rates the degree
of nitrification diminishes, such that above 30 to 40 Ib BODs/lOOO cu ft/day (0.48 to 0.64
kg/m^/day) very little nitrification occurs. These findings  are consistent with those of the
National Research Council whose evaluation of World War II military  installations indicates
that the organic loading should not exceed 12 Ib BODs/lOOO cu ft/day (0.19 kg/m3/day)
for rock media. 49

The partial nitrification occurring at intermediate loading rates can cause confusion when
attempting to analyze organic carbon removals across trickling filters with the BOD5 test.
Samples from  the effluent of these  partially  nitrifying trickling  filters  will contain a
                                     FIGURE 4-7
             EFFECT OF ORGANIC LOAD ON NITRIFICATION EFFICIENCY
                           OF ROCK TRICKLING FILTERS
    100
 s:
 UJ
UJ
a.
u.
i
     80
 iy   eo
 o
 u.
 u_
 Ul

 O   4O
     20
                                           i      '      i      '      I
                   m
                                                 0  NO  RECIRCULATION
                                                 B  RECIRCULATION

                                     Kg/m3/day = 62,4 Ib  BOD5/IOOO cu ft/day_
                    0
             O
                                 m

                                  m
                                                                              J)
       0
                  10         20
                      BODe;  LOAD -
                                         30          40
                                     LB/IOOO  CU FT/DAY
50
60
                                      4-38

-------
significant quantity of nitrifiers that could act as seed for promoting nitrification within the
5-day incubation period  of the BOD5 test.21>50 About 1.5 mg/1 of ammonia nitrogen is
added to the dilution water in the BOD5  test and will also be nitrified. This will result in
unexpectedly high oxygen demands.  If the BOD5 test is to measure organics in effluents
from trickling filters,  then nitrification must be suppressed. The same is true for activated
sludge, but to a lesser degree as partial nitrification is less prevalent.

As opposed to the relatively efficient removal of ammonia in trickling filters, it appears that
reductions in organic nitrogen are variable and range between 20 and 80 percent. (Table
4-9). Organic nitrogen reduction can be obtained through employing effluent filtration to
remove particulate organic nitrogen. However, all treatment systems are limited to about 1
to 2 mg/1 of soluble organic nitrogen, contained in refractory organics, and therefore there
are limits to the improvements that can be obtained with effluent filtration.

         4.4.1.3 Effect of Media Type on Allowable Organic Loading

The  specific  surface  of  media selected has a substantial effect  on  the allowable organic
loading  rate  for trickling  filters.  Greater specific   surface in the media  allows greater
biological film development and therefore a greater concentration of organisms within a unit
volume. Therefore,  the organic loading may be higher in cases where the specific surface of

                                     TABLE 4-9

     ORGANIC NITROGEN REDUCTIONS IN NITRIFYING TRICKLING FILTERS

Facility
Location

Gainesville, Fla.a
(pilot)
Johannesburg, S.A.
(full-.scale)








Stockton, Ca.c
(pilot)


Organic
load.
Ib BOD /1 000 cu ft/day
(kg BOI§5/ m3/day)
31.5
(0.50)
31.2
(0.50
19.6
(0.31)
13.1
(0.21)
10.5
(0.17)
6.8
(0.11)
14
(0.22)
22
(0.35)

Depth,
ft
(m)
6
(1.8)
12
(3 . 7)
6
(1.8)
12
(3.7)




21.5
(6.6)




Media

1-1/2 - 2-1/2 in.
(3.8r-c|5.4cm)
2 in.
(5roc$
1-1/2 in.
(3 . 8 cm)
rock
2-3 in.
(5.1ro-c£.6cm)




plastic
27 sf/cu ft
(86 m /in3)



Influent
organic-N,
mg/1
16.6

. 9.8

6.3

8.2

9.9

13.9

11.3

11.4


Effluent
organic-N,
mg/1
7.3

4.7

2.5

3.6

2.2

2.1

8.9

9.0


Organic-N
removal ,
percent
56

52

60

56

78

85

21

21

 a
  References 44, 45
  Reference 47
  References 21/51
                                        4-39

-------
the media is increased over that of rock media. An example is the work of Stenquist, et
al. 21  who showed that plastic media (27 sq ft/cu ft) could be loaded at about 25 lb/1000 cu
ft/day and still  achieve  good  nitrification  (Table 4-10).  The higher allowable loading
attributable to plastic trickling filters was attributed to be at least partly due to the greater
specific surface of plastic media when compared  to rock media.  Another factor favoring
greater capacity of the plastic media filters may be oxygen supply. Rock filters often have
poor ventilation, particularly when water and air temperatures are close or the same.

          4.4.1.4 Effect of Recirculation on Nitrification

The beneficial effects of recirculation on enhancing nitrification in trickling filters is evident
in the data for Salford,  England in Table 4-11. The  imposition of a 1:1 recycle  ratio
consistently improved ammonia removals compared to when no recirculation was the rule.
Trickling  filter plants designed  for  nitrification  should  incorporate provision for recir-
culation.

The minimum hydraulic application rate  for plastic media trickling filters is in the range of
0.5  to 1.0 gpm/sf (0.020 to 0.041 m3/m2/min.).  This minimum rate must be supplied to
ensure uniform wetting of the media. Without recirculation, nitrification design loadings
may result in applied hydraulic loads lower than the minimum hydraulic application  rate.
Recirculation  provides the means for preventing drying out of portions of the media by
ensuring that at least the minimum hydraulic application rate is applied at all times.

          4.4.1.5 Effect of Temperature on Nitrification

Available  data for nitrification are largely  for warm liquid temperatures, and the practical
effects of reduced temperatures  (< 20C) on allowable organic loads for combined carbon
oxidation-nitrification  applications are not known at this time. However, the kinetic rate
data given in  Section 3.2.5.4  would indicate that organic loads would have to be reduced
below those shown in Figure 4-7 for cold weather operation. This reduction in organic load

                                    TABLE 4-10

  LOADING CRITERIA FOR NITRIFICATION WITH PLASTIC MEDIA AT STOCKTON
BOD
load.
lb/100£ cu ft/day
(kg/m /day)
14
(0.22)
22
(0.35)

Temp,
C

26

24

Influent
BOD ,
mg/r

155

131


Depth,.
ft
(m)
21.5
(6.6)
21.5
(6.6)


Media

plastic
(Surfpac)3
Same


Recycle
ratiob

5.5

2.25

Influent
NH4-N,
mg/1

16.5

• 17.5

Effluent
NH^-N,
mg/1

1.0

2.0

Percent
nitrification
(or ammonia
removal)
94

89


Reference


21,51

21,51

 327 sq ft/cu ft (86 m2/m3)
  Recycle ratio is the ratio of recycled effluent to influent. Effluent was recycled prior to sedimentation.
                                         4-40

-------
would reduce the loadings to such low values as to  cause capital costs to be higher than
other available  ammonia  removal techniques,  such  as separate  stage nitrification or a
physical chemical technique.

         4.4.1.6 Effect of Diurnal Loading on Performance

While it is known that  diurnal variations in nitrogen loading will cause variations in effluent
quality,  no  information  is  available  which  would  allow quantitative  guidelines  to  be
formulated.  In  cases  where large  peak to  average  flow  ratios are experienced, flow
equalization before the  nitrification step may be appropriate.

     4.4.2 Nitrification with the  Rotating  Biological Disc Process in Combined Carbon
          Oxidation-Nitrification  Applications

The rotating biological disc (RBD) process is beginning to see use in the  U.S. in combined
carbon  oxidation-nitrification  applications. The following discussion  is abstracted from
Aritonie.52

The RBD process consists of a series of.large-diameter plastic discs, which  are mounted on a
horizontal  shaft  and  placed  in  a concrete  tank. The  discs are slowly  rotated  while
approximately 40 percent of the surface area  is  immersed in the wastewater to be treated.


                                    TABLE 4-11

 EFFECT OF RECIRCULATION ON NITRIFICATION IN ROCK TRICKLING FILTERS
                  AT  SALFORD, ENGLAND (REFERENCE 48, 25)a
BOD5
load
lb/1000 cu ft/day
{kg/m2/day)

22.6
(0.36)
16.3
(0.26)
11.8
(0.19)
9.2
(0.15)
7.7
(0.12)
5.9
(0.095)
4.6
(0.074)
3.2
(0.051)
Influent
BOD5,
mg/1

266

235

191

239

165

192

199

206

Influent
NH4-N
mg/1

33.9

31.3

32.0

43.9

40.5

40.7

38.3

36.6

Effluent
NH^-N,
mg/1
"without
recirculation
19.7

16.9

9.7

125

11.4

5.7

2.8

0.7

with
recirculation
13.6

11.8

4.8

2.2

4.9

2.8

0.9

0.4

Percent
nitrification
without
recirculation
?2

46

70

72

72

86

93

93

with
recirculation
60

62

85

95

88

93

98

99

Media was blast slag, 8 ft (2.4 m) deep. With recirculation a 1:1 ratio was employed.

-------
Shortly after start-up , organisms present in the wastewater begin to adhere to the rotating
surfaces and grow until in about one week, the entire surface area is covered with a layer of
aerobic biomass. In rotation, the discs pick up a thin film of wastewater, which flows down
the surface  of the discs and absorbs oxygen from the air. Microorganisms remove both
dissolved oxygen and organic materials from this thin film of wastewater. Shearing forces
exerted on the biomass as it passes through the  wastewater strip  excess growth from  the
discs into the mixed liquor. The mixing action of the rotating discs keeps the sloughed solids
in suspension,  and  the  wastewater flow  carries them  out of the disc sections  into a
secondary clarifier for separation and disposal. The discs also serve to mix the contents of
each treatment stage. Treated wastewater and sloughed solids flow to a secondary clarifier
where the solids settle out and the effluent passes on for further treatment or disinfection.
The  settled  solids,  which  can thicken up to 4  percent solids  content  in the secondary
clarifier, are removed for treatment and disposal. A flow diagram for a typical application of
the RBD process is shown in Figure 4-8.

                                    FIGURE 4-8

                A TYPICAL ROTATING BIOLOGICAL DISC PROCESS
                     (COURTESY OF THE AUTOTROL CORP.)
            Primary Treatment
Secondary Clarifier
 Raw^A
Waste^
                  cj> Effluent
                                    Solids Disposal

One current disc  design  consists  of  vacuum  formed  polyethylene sheets formed into
concentric corrugations which provide  a high density of surface area. The corrugated sheets
are then  welded together to form a stack of discs with approximately  \1A in. (3.2 cm)
center-to-center spacing.  This  type  of  construction  has  a surface  area density  of
approximately  37  ft 2/ft 3  (121 m^/m^). A key feature of this disc design is the provision of
radial passages extending from the shaft to the outer perimeter of the discs. This assures that
wastewater, air, and stripped biomass can pass freely into and out of the disc assembly. In
the twelve-ft (3.7 m) diameter size, radial passages are provided every thirty degrees.

The disc  units  are normally housed to avoid temperature  drops across  the process, to
prevent algae growth on the disc surface, and to protect the surface from hail or rain which
can  wash the  slimes off. Information on disc  types and on the general  design of these
facilities can be obtained from the various disc manufacturers.

-------
         4.4.2.1 Loading Criteria for Nitrification

As the rotating discs operate in series, organic matter is removed in the first disc stages and
subsequent disc stages are used for nitrification. This separation of function occurs without
the need for intermediate clarification. Nitrification  does not commence until the bulk of
the BOD5 is oxidized. When low levels of BOD5 are reached, the disc stage is no  longer
dominated by heterotrophs, and nitrification can proceed.

Antonie" has summarized the effects of process operating conditions on nitrification. Test
data from  a number  of locations are shown in Figure 4-9B. The degree  of  ammonia
oxidation is related to the hydraulic loa'ding on the rotating media as gallons per day per
unit surface of available surface area. Also important, as it affects population dynamics, is
the influent BOD strength. The data in Figure 4-9B was used to arrive at the design criteria
shown in Figure 4-9A. Two changes have been  made to allow use of the relationships in
design. One  is that there is a maximum ammonia nitrogen concentration for which the data
is  considered valid. This  concentration  is  generally 1/5  to 1/10  the influent  BOD5
concentration.  When  the  ammonia  concentration   exceeds  these  maximum  ammonia
nitrogen  concentrations on the appropriate BOD5 curve, it has been recommended that the
design curve be used which is rated for that ammonia concentration.53 The second change
made in  Figure 4-9A is the identification of a region of unstable nitrification;  that is, a
region of hydraulic loading where either a slight change in the hydraulic loading or influent
BOD strength could result in a displacement of the  nitrifying population. It is considered
advisable to  stay out of that region even, during daily peak flow conditions.53

         4.4.2.2 Effect of Temperature

Investigators at Rutgers University^ found that the nitrifying capability of the discs was
relatively constant in a temperature range from 15 to 26 C. Simliar results were obtained by
Antonie53 who has found no effect of temperature above  13 C. Temperature correction
factors derived from pilot data are shown in Figure 4-10.53 These correction factors should
be  used  to  reduce  the  design  hydraulic  loading  determined  in Figure  4-9A  for  any
waste water temperature lower than 13 C.

         4.4.2.3 Effect of Diurnal Load Variations

Precise design criteria for handling load variations in  RBD units have not been formulated.
The concern has been that high BOD5  concentrations would break through to the last disc
stage during peak loads and  would cause displacement of nitrifiers from that stage.53 While
firm design recommendations have not  been made, there are two  general approaches
available. One is to arbitrarily derate the surface hydraulic loading to the disc to ensure low
BOD loadings at all times. This will result in a larger amount of rotating surface  area. The
other is to install flow equalization to reduce the peak  to average flow ratio. 5 3
                                        4-43

-------
                            FIGURE 4-9
EFFECT OF BOD5 CONCENTRATION AND HYDRAULIC LOAD ON
  NITRIFICATION IN THE RBD PROCESS (AFTER ANTONIE (53))
  100
  95
  90
2
o
Ul
It
  85
o
Q:
5
2
o
5
   80
   75
   70
  100
           350 150
                      120 100  80
                                   Inlet BODg Concentration', mq/t
                                                  Maximum
                                                  Ammonia Nitrogen
                                                  Concentration, mg/l
                                                         Temperature,
                                                           >t3C
      A. Design  Relationship
        NITRIFICATION OF
        PRIMARY  EFFLUENT
         (l gpd/sq ft = 41 l/mz/day)
            0.5      1.0      1.5      2.0      2.5
                    HYDRAULIC LOADING, gpd/Sq.ft
                                                    3.0
                                                            3.5
                                                                    4.0
   95 —
   9O
O
Ui
   85
o
ct
^  80
1
5

I  75
   70
     B. Experimental Data

     _     Inlet BOD5,mg/l 250  \50

          NITRIFICATION OF
          PRIMARY EFFLUENT
        Inlet BOD$ = 75 to 120 mg/l
         O Tallahassee, Flo. 95-120 mg/l
         £ Pewaukee, Wis. 75-87 mg/l
         ffi Spring City, Pa.  126 mg/l
         EB  Bloom, Illinois
         B  Mansfield , Ohio
         -- Washington, Pa.  120 mg/l
        Inlet BOD5= \50 to 350 mg/l
         • Tallahassee , Fla. ISO  to 180 mg/l
         A  Pewaukie.Wis. 220 to 350 mg/l
         X  Luverne, Minn. 175 mg/l
         •  Spencer, Iowa 210 to 350 mg/l
             I	I	I	I
                                             I        I        I
                                              (1 gpd/sq ft = 41 l/mz/day)
                                                         Temperature,
                                                           >I3C
                                                     I  0
            0.5      1.0      1.5      2.0      2.5
                    HYDRAULIC  LOADING,  gpd/sq ft
                                                    3.0
                                                            3.5
                                                                    4.0
                                 444

-------
4.5 Pretreatment for Separate Stage Nitrification

Nitrification facilities have been previously classified in Table 4-1. As can be seen from that
table, in order to obtain a separate stage nitrification process, the influent to that process
must be pretreated to remove organic carbon.  Further, the pretreatment must achieve a
degree of carbon removal greater than is obtained by primary treatment alone, in  order to
reduce the BOD5/TKN ratio  to a sufficiently low level to ensure a significant fraction of
nitrifiers in the biomass. Alternatives listed in Table 4-1  include chemical treatment  in the
primary, activated sludge,  roughing filters, and trickling filters. Not listed, but possible, is
activated  carbon treatment  in  conjunction  with  primary  chemical addition.  These
alternatives are summarized in Figure  4-11. This set of pretreatment alternatives  is not
meant to be exhaustive, but merely  illustrative; other flowsheets are possible.  Detailed
design  of pretreatment steps is beyond the scope of this manual; however, design procedures
may be found in the publications  referenced in Figure 4-11  and the sources cited on Table
4-1.
The  pretreatment  alternative adopted  may have significant effects on the downstream
nitrification stage. In this section, some of these possible effects are considered.

                                  FIGURE 4-10

         TEMPERATURE CORRECTION FACTOR FOR NITRIFICATION
                 IN THE RBD PROCESS (AFTER ANTONIE (53))
 it
 o
o
 tt
 Ul
 0.
 %
 Uj
 K
    3.0
   2.5
    2.0
    1.5
1.0
            \
               \
        I
     85N AMMONIA NITROGEN  REMOVAL,  PERCENT
   _90t
     95,
     99^
                           I
                              I
I
I
                           6      7      8       9      IO
                           WASTEWATER  TEMPERATURE , C

                                       4-45
                                                                   12
                             13

-------
    4.5.1 Effects of Pretreatment by Chemical Addition

The chemical treatment step may cause significant changes in alkalinity and pH in the
downstream  nitrification stage. Several alternate chemicals are available  and  their results
differ.

                                  FIGURE 4-11

    PRETREATMENT ALTERNATIVES FOR SEPARATE STAGE NITRIFICATION
ALTERNATE
CHE
r
RAW *
WASTEWATER
AC
RAW
WASTEWATER
RO
RAW
WASTEWATER
TR
RAW
WASTEWATER
PHI
r
RAW \^
WASTEWATER
:MICAL ADC
EMICAL
PRIMARY
CLARIFIER

PRETREATMENT TECHNIQUES
HTION IN PRIMARY
SEPARATE
NITRIFICATION
(SG or AG)
TIVATEO SLUDGE .. 	 ..
PRIMARY
(OPTIONAL)

ACTIVATED ^/INTERMEDIATE \ SEPARATE
*" SLUDGE — H CLAHIFIER — S 1 AOt 	 ••
y c"-ARIFIERy NITRIFICATION
V^^^/ (SG or AG)
JGHING FILTER
.x— X. /^N.
PRIMARY
CLARIFIER

/ROUGHING \ AERATION AND /WITH IFICATIOW\
*l TniCICLINC V-» NITRIFICATION ...«. IFICATION
VF!LTER J TANK 1 CLARIF.ER 1 »
^ 	 S (SG) \ 	 /
(or RBD Process
CKLING FILTER
PRIMARY


J TRICKLING \ /INTERMEDIATED SE!*"*J'E fc
~^ F.LTER J~*\^ CLAR.FIER^ " N|TRrFICAT,ON
\ 	 / \ 	 / (S3 or AG)
(or RBO Process)
rSICAL-CHEMICAL TREATMENT
EMICAL
PRIMARY
CLARIFIER

MULTIMEDIA ACTIVATED SEPARATE
FILTRATION ADSORPTION NITRIFICATION
(SG or AG)
BOD5 Removal
Prior to
Nitrification
50 to 75
60 to 95
45 to 60
60 to 90
80 to 90
References
25,55
25,36
25,36
25,36
25,55,
56
      KEY
     SG = Suspended growth
     AG = Attached growth
                                     4-46

-------
When alum or ferric chloride is used in the primary treatment stage, both the carbonate and
phosphate components of the alkalinity in wastewater are changed. Table 4-12 summarizes
the changes occurring when  alum is  added  to  wastewater. Effects of iron addition are
similar. Since orthophosphate is present in wastewater as HPO4 and H2PO4 between pH 4.5
and 9.3,57 fable 4-12 shows reactions with aluminum for both forms. The compound
HPO4 is measured  as  part of the alkalinity in wastewaters because of the following
equilibria:
                      H+ + HPO4 — H2PC>4

This reaction is  shifted completely to the right at pH 4.5 which is  approximately the
endpoint titration pH for the conversion of bicarbonate (HCC>3) to carbonic acid (H2CO3)
in the standard alkalinity determination. Table 4-12 shows that the same amount of total
alkalinity is lost regardless of the form of the inorganic phosphorus (HPO^ or
                                TABLE 4-12

        EFFECT OF ALUM ADDITION TO WASTEWATER ON ALKALINITY


  1.  Hydrolysis

        AU (SO.) „  +6 HCOl    + 6 H0O -*• 2 Al (OH), + 3 SO"? + 6 H-CO,
          £   *t  O          O        ^              O      **     6   O
        (54 as Al+3) (300 alkalinity
                    as CaCo3)                      +g
        or 5.6 mg of alkalinity as  CaCOg lost per mg Al   added

  2.  Precipitation of inorganic phosphorus

           form
         A12 (SO4) 3    +2 HCO~    + 2 HPO4 — >- 2
                                  (100 alkalinity
         (54 as Al+3)  (100 alkalinity  as CaCO3)
                      as CaCOg)    (62 as P)
                                                    +3
         or 3:7 mg of alkalinity as CaCO, lost per mg Al  added
                                     O
                                     O
         and . 87 mg Al    required per mg P
      H2PO4
         A12 (S04) 3 +   4 ECO'   + 2 H2PO4  — *- 2 AlPO4+3 80= + 4

         (54 as Al+3) (200 alkalinity (0 alkalinity
                     as CaCO )    as CaCO3)
                    (62 as P)     (62 as P)
        or 3. 7 mg of alkalinity as CaCOg per mg Al+a added
        and 0. 87 mg of Al    required per mg P
                                      4-47

-------
As an illustrative example of the results of alum addition, consider a case where 10 mg/1 of
inorganic phosphorus is precipitated in the primary stage of treatment. Using a value of 0.64
mg P removed per mg of aluminum added results in the following aluminum use: 1
                       10mg/1  „     =   15.5 mg/1 Al' " required (total)
                      0.64   mgj
                             mgAT

The aluminum used in phosphorus precipitation is (Table 4-12):
                10 mg/1 P(0.87  mgA1	) = 8.7 mg/1 AT" for precipitation
                               mgP

By difference the aluminum used in hydrolysis is:

                                          i I I
                     15.5 - 8.7 = 6.8 mg/1 Al   .for hydrolysis

The alkalinity loss due to hydrolysis is (Table 4-12):

                          (5.6mgCaCO~)
                     6.8 	:rrr^- = 37.4 as CaCO-
                                . ,+TT                  J
                            mg Al

The alkalinity loss due to precipitation is (Table 4-12):

                           (3.7 mg CaCO,)
                      8.7  	— = 32.2 as CaCO-
                                  +++                   3
                             mgAl

Thus, a total alkalinity loss of 70 mg/1  will occur when 10 mg/1 of P is removed. This loss,
depending on the initial wastewater alkalinity, may have to be made up with downstream
chemical addition to prevent adverse  effects on  the operating pH of nitrification (see
Section 4.9.2.).

On the other hand, lime addition to the primary may have quite the opposite effect of alum.
The  changes in  alkalinity occurring with lime addition are dependent on lime dose (or pH)
and  the quality of the raw wastewater. Cases of both increases and decreases in alkalinity
with lime treatment may be found.^5

Lime addition  may also cause elevation of the operating pH of the nitrification stage.3
Primary effluent  after lime treatment will typically  have  a pH between 9.5 and 11.0,
depending on lime dose and treatment requirements. 55 This pH is higher than can normally
be discharged or  introduced into downstream treatment units. To reduce the pH, normal

                                        4-48

-------
practice is to "recarbonate" the high pH primary effluent. Conventionally, this involves the
introduction  of gaseous  carbon  dioxide  (CC>2)  into the high pH primary effluent in a
reaction basin of at least 20 minutes detention time. Typically the carbon dioxide is either
drawn from refrigerated storage or furnace stack gases containing carbon dioxide are used.
The recarbonation step can be thought of as the conversion of alkalinity in the hydroxide
form (OH~) to that in the bicarbonate form (HCOJ) as follows: 55

                   Ca(OH)2 + 2CO2  	+•  Ca(HCO3)2

In many cases  there is sufficient carbon  dioxide produced from the oxidation of organic
carbon and from nitrification  to completely satisfy recarbonation  requirements. In  a lime
precipitation-nitrification sequence in California^ it was found that external carbon dioxide
requirements  were minimal  and  only occasionally required. In this case, it was calculated
that   approximately two-thirds  of  the  carbon dioxide  produced  was derived from
nitrification,  while the remaining one-third was derived from the oxidation of the organic
carbon remaining in the primary effluent. When the same process was tried at the  EPA-DC
pilot plant at Blue Plains, it was  found that supplemental carbon dioxide was continuously
required to maintain a neutral pH. This  is because  wastewater in the Washington  area is
weaker than in the California case. There is less production of carbon dioxide because there
are lower  concentrations of oxidizable substances entering the nitrification stage at Blue
Plains.

The tendency is for the high pH effluent  of the lime  primary process to elevate the pH in
the nitrification reactor. Often this effect enhances nitrification rates (see Section 3.2.5.6.).

When lime primary treatment is employed,  care must be taken not to mix the  primary
effluent with the return activated sludge prior to entry into the nitrification tank. The high
pH of the primary effluent return activated sludge  mixture would  be toxic to both the
nitrifiers and heterotrophic bacteria in the return sludge.

A concern often expressed is that the phosphorus removal obtained in  a chemical primary
treatment  step  will be  so great  as to  starve  the  downstream  nitrifying  biomass for
phosphorus  as  a nutrient  for growth. Actually, the requirements for phosphorus in a
separate stage nitrification system are very low. Typically, organism biomass contains about
2.6 percent phosphorus. This number can be used to calculate phosphorus requirements, as
in the following example. Assume a case where 36 mg/1 of TKN are nitrified and 60 mg/1 of
BOD5 are removed in a separate nitrification stage. The quantity of biomass grown  can be
conservatively estimated as follows: (see Section 3.2.7)

                   Nitrifiers:        0.15(36)   =     5.4 mg/1 VSS
                   Heterotrophs:    0.55(60)   =    33.0 mg/1 VSS
                   Total biomass:                   38.4 mg/1 VSS
                                        4-49

-------
The phosphorus required is 2.6 percent of the VSS grown:

                          0.026(38.4)= 1.0 mg/1 as P

Thus, the phosphorus requirement in this example is only 1.0 mg/1 P. In most cases,  the
BOD5 and TKN levels will be less than given in the example, so that the 1 mg/1 requirement
can be considered a maximum requirement for separate stage nitrification.

It is  a  relatively  simple  matter to manipulate chemical  doses in the  primary  so that
phosphorus residuals are sufficient to support biological growth. 3 It may be that low levels
of phosphorus or some trace micronutrient removed by the primary may cause "bulking" in
a suspended growth type nitrification stage. 58 Very high SVI measurements were observed
in the nitrification stage  following lime treatment at  the Central Contra Costa Sanitary
District's Advanced Treatment Test Facility.^ However, it has been found that this bulking
could be very effectively  controlled by a continuous low dose of chlorine to the return
sludge, without impairment of nitrification efficiency.  A dose of 2 to 3  Ib. per 1,000 Ib.
MLVSS per day  (2 to 3 g/kg/day) in the inventory reduced the SVI from 160-200 to 45-80
ml/gram. Overdosing at 5 to 7 Ib./1,000 Ib./day (5  to 7 g/kg/day) caused impairment of
nitrification and higher effluent solids.59
     4.5.2 Effects of Degree of Organic Carbon Removal

The  pretreatment  alternatives arrayed  in Figure 4-11 provide varying degrees of organic
carbon removal  ahead of the nitrification step. Not only is there  variation  among  the
alternatives, but  there  is possible  range  of intermediate effluent  qualities within each
alternative. It was shown in Chapter 3 that high degrees of organic carbon removals led to
the  highest nitrification  rates.  This  implies  that reactor requirements diminish with
increasing degrees of carbon removal in the pretreatment stage.

Very low  levels  of organic carbon in the nitrification  process influent has  differing
implications for suspended growth and  attached growth nitrification reactors. The effluent
solids from the  sedimentation step in  a suspended growth system can exceed the solids
synthesized in the  process when the level of organics  is low. This leads to  a requirement of
continuously wasting solids  from  other  suspended  growth  stages  (carbon  removal or
denitrification) to the nitrification stage or to increase the BOD of the influent in order to
maintain the inventory of biological solids  in the system. This point is discussed further in
Section 4.6.4.

Low levels of organics in the influent to attached growth reactors can be advantageous. The
synthesis of solids  occurring with  low levels of influent organics results in very low levels of
solids in the effluent from an attached  growth reactor. In some cases this can eliminate the
need for a clarification step, especially if multimedia filtration follows or if there is some
other downstream treatment unit such as denitrification.

                                        4-50

-------
     4.5.3 Protection Against Toxicants

All of the pretreatment alternatives portrayed in Figure 4-11 provide a degree of removal of
the toxicants present in raw wastewater. However, the types of toxicants removed by each
pretreatment stage vary among the alternatives. Chemical  primary treatment can be used
where toxicity from heavy metals is the major problem. Lime primary treatment is one of
the most effective processes for removal of a wide range of metals. 5 5 Chemical treatment is
usually not  effective for removal of organic toxicants, unless it is coupled with a carbon
adsorption  step  such  as  in  the  physical-chemical treatment sequence. The biological
pretreatment alternatives (activated sludge, trickling filters, and roughing filters) provide a
degree of protection against both organic and heavy metal toxicants. An exception would be
organics  resistant to  biological oxidation, such as  the  solvents perchloroethylene  and
trichloroethylene which have been identified as toxicants which can upset nitrification. 1 >60

When materials toxic to nitrifiers  are present in the influent raw wastewater on a regular
basis, the pretreatment technique most suitable for their removal can be used in the plant
design to safeguard the nitrifying population. The determination of the most suitable system
configuration need not involve an extensive sampling or elaborate pilot program. A recently
developed bench top analysis can be used to screen alternates. 61 The test procedure involves
batch oxygen uptake tests using a respirometer to measure oxygen utilization. Composite
wastewater samples are subjected to various pretreatments, e.g., alum or powdered activated
carbon via a jar test procedure or to biological oxidation by batch aeration. Each treated
sample is then split and placed into  two  respirometers.  One respirometer  is used as a
non-nitrifying control  by  treatment with  a nitrification  inhibitor  such as Allythiourea
(ATU).61 The other respirometer is inoculated with a small amount of mixed liquor from a
nitrifying activated  sludge plant. Differences between the oxygen used in the control and in
the seeded sample can be used to establish batch nitrification rates. At the end of the test,
the respirometer contents  are sampled and analyzed  for the nitrogen species to confirm
whether  nitrification took  place in the inoculated samples as well as  to check the control.
The adequacy  of the seed  used  can also be checked  by running an inoculated, but
uninhibited sample, known to contain ammonia and organics, but no toxicants.

The batch nitrification rates, determined  by the batch procedure,  can be examined to
determine  the pretreatment  technique   apparently  most  suitable   among  the  options
examined. Often, some of the pretreatment techniques will result in little or no nitrification
in the inoculated sample, indicating inadequate removal of the toxicant(s). In other cases,
the pretreatment techniques will allow vigorous nitrification in the sample indicating good
removal of the toxicant(s).61 The particular pretreatment  technique  that is effective may
also indicate the type of toxicant that is interfering with nitrification and may permit
identification and elimination of the source tributary to the system.  For instance, if lime
treatment is effective,  the problem may be a heavy metal  that can be precipitated by  the
lime.  Alternatively, if biological oxidation is  ineffective  but activated carbon treatment
allows nitrification  to proceed, then a nonbiodegradable  organic is suspect.  Subsequent
specific analyses can then be run in the identified category of compounds. If the toxicants

                                         4-51

-------
cannot  be eliminated by  a  source control  program, often a  pilot study of the process
identified by the bench scale procedure  can be justified to confirm the process selection.
Pilot  studies also  have value in determining the ability of the nitrifiers  to  adapt to the
toxicants, something the batch test is not capable of doing.

4.6 Separate Stage Nitrification with Suspended Growth Processes

There are many examples of separate stage nitrification processes in the U.S. (Table 4-1).
The initial development of the suspended growth process in separate stage application was
oriented to the isolation  of the operation of the carbonaceous removal and nitrification
processes so that each could be separately controlled and optimized. 15 By  placing a carbon
removal  system (originally conceived as a high rate activated sludge system)  ahead of the
separate  nitrification  stage, the sludge would be enriched with nitrifiers as opposed to the
marginal population  present in combined carbon oxidation-nitrification systems. By this
enrichment process, nitrification would be expected to be less temperature sensitive than in
a combined carbon oxidation-nitrification system.

First applications of the process were in the  northern portion of the U.S. where low ( <  10
C) liquid temperatures were obtained in the  wintertime. Subsequently, the system has been
applied  in moderate  climates  such as Florida and  California  where some  of  the other
advantages of the process have made its application desirable.

     4.6.1 Application of Nitrification Kinetic Theory to Design

The kinetic approach for  design of separate stage nitrification in suspended growth systems
is fundamentally  related  to  the  kinetic  design approach used  for  combined carbon
oxidation-nitrification,  as was  shown in Section  3.2.7. To  date, however,  the practice has
been  to  adopt  the  "solids  retention  time"  design  approach  for combined carbon
oxidation-nitrification applications  and to use the nitrification rate approach for separate
stage  design. The  nitrification  rate approach has  been based on experimentally measured
rates, ^ >62,63 rather than attempting to relate the  rates to fundamental kinetic theory (see
Section  3.2.7). The  theoretically  determined  nitrification rates are limited  in  their
applicability chiefly because the nitrifier fraction of the mixed  liquor cannot  be accurately
assessed. Nonetheless, the concepts developed from kinetic theory are applicable even in the
absence of information about the nitrifier fraction.

The solids  retention time  design approach is  also directly applicable to many separate stage
nitrification design problems. The limit of its applicability is related to the difficulties  with
the nitrification rate  approach, that is,the yield of nitrifiers grown through nitrification is
not accurately known. However, for designs  where  the BOD5 level in  the nitrification
influent  is 30  to  60 mg/1 (or BODs/TKN  ratio is 2 to 3) the growth  of  nitrifiers will
generally be a small fraction relative to the heterotrophic population. In this case, the errors
in assumption of the  yield of nitrifiers will be masked by the growth of heterotrophs. For
these cases, the contribution of the nitrifiers  to  the overall process  growth rate may  be

                                         4-52

-------
neglected with the assurance that their contribution is less than 12 percent (Section 3.2.7).
A model  which explicitly considers the nitrifier contribution to the system growth rate is
available which can be used when accurate yields for nitrification become known.™
In situations where the organic carbon is low in the influent (BOD5/TKN ratio 0.5 to 2.0)
to the nitrification stage, both the assumptions for heterotrophic and nitrifier yields become
uncertain. These  low BOD5/TKN  ratio situations are usually cases where a  well stabilized
secondary effluent is being nitrified.  Residual  BOD5 in these effluents is often biological
solids rather  than  residual raw wastewater organic  matter. The  biomass yield  in the
nitrification stage for these cases is less well defined. This leads to uncertainty in estimating
design sludge inventories and wasting  schedules when  using the solids retention time design
approach.

In this section, both the solids retention time approach and the nitrification rate approach
are presented. The direct theoretical  interrelationships of the approaches, as developed in
Section 3.2.7,  should not be overlooked.

     4.6.2 Solids Retention Time Approach

The design procedures developed in Section 4.3  are  directly applicable to  separate stage
nitrification design when the BOD5/TKN ratio  equals or exceeds  2.0. A summary of these
design concepts follows.

          4.6.2.1  Choice of Process Configuration

In general, the favored system for separate stage nitrification is the plug flow system. It was
already  shown in Section 4.3.5  that the plug flow process results in lower effluent ammonia
than a complete  mix process at the same SF, or alternately, the same ammonia level at a
lower SF. The only disadvantage of the process in combined carbon  oxidation-nitrification
applications is the difficulty in supplying  adequate  DO in the  head end of the system,
rendering that zone ineffective for nitrification in cases of low DO. With a carbon removal
step ahead of the separate nitrification stage, the high oxygen demand at the head  end of
the system  is minimized, and  less difficulty is found in designing  aeration systems that
ensure adequate  DO levels throughout  the nitrification  tanks.  With adequate DO levels
throughout the tanks, the full advantage of plug flow kinetics over the kinetics of any other
configuration is obtained.

In cases where a lime precipitation step precedes the nitrification step, it is often desirable
to use the carbon dioxide (CO2) produced by the nitrification process for recarbonation
(c.f. Section 4.5.1).  Since the process of nitrification will be spread throughout the plug
flow reactor,  and recarbonation is desired at  its head end to avoid pH  toxicity,  only a
portion of the carbon dioxide produced in the  process is available for recarbonation at the
head end of the process. A solution to this problem is  to increase the  use of external carbon
dioxide. On the other hand, external carbon dioxide usage may be minimized by adopting
                                         4-53

-------
one of the other process configurations, such as complete mix, that more evenly spread out
the load throughout the aeration  tank, thereby taking full advantage of the in-process
carbon dioxide generation for recarbonation.

         4.6.2.2 Choice of the Safety Factor

The  same considerations discussed in Section 4.3.3.2  are applicable to separate sludge
applications. Diurnal variations in  nitrogen load will effect effluent quality in the same
manner as  for  combined carbon oxidation-nitrification applications.  However, upstream
treatment steps may moderate the fluctuations in nitrogen  load experienced by separate
stage nitrification processes. While primary treatment  and roughing filters have so little
liquid holdup that little or no nitrogen load equalization is provided, this is not true of
either  the  activated sludge  or trickling  filter  pretreatment alternatives. Both  of these
alternatives provide a degree of nitrogen load equalization.

A  typical ammonia load curve  for the secondary  effluent from the Rancho Cordova,
California treatment plant is shown in Figure 4-12. The plant provides activated sludge
treatment. On the date monitored, the average flow was 1.9 mgd with a peak hourly flow of


                                   FIGURE 4- 12
            RANCHO CORDOVA WASTEWATER TREATMENT FACILITY
          EFFLUENT AMMONIA CHARACTERISTICS, MARCH 19-20, 1974
  50
     —   5OO
5>
o
   30
Uj
o
o
t>
  20
   to
       \
       CD
     —   300
     —  2OO
          too
     I—    O
                                     i   i  r
                   i   i  i
                                   NH4-N MASS  FLUX
                               PLANT FLOW
                                 NH^-N  CONCENTRATION
                   = .454 Kg
I   I	|   I   I
                                                I   i
           oeoo
                      1200
                                 I6OO       20OO
                                       TIME  OF DAY
                                                      Z4OO
III     III
                                                                 04OO
                                                                             oaoo
                                        4-54

-------
3.0 mgd, for a flow peaking factor of 1.6. The ammonia loading behaved similarly with an
average load of 252  Ib /day and a peak hourly load of 379  Ib /day for a nitrogen load
peaking factor of 1.6. Using the concept developed in Section 4.3.3.2, the minimum safety
factor adopted for a separate stage nitrification process serving this plant should be 1.6, to
prevent significant ammonia leakage during the peak hour.

     4.6.3 Nitrification Rate Approach

The kinetic  design approach using nitrification  rates places  reliance on  experimentally
determined rates obtained from pilot studies. Available data are summarized in Figure 4-13,
plotted against temperature.  Also shown are two  other principal variables that can effect
nitrification rates, the BODs/TKN ratio and the mixed liquor pH.

In general, the nitrification rates follow the predictions of the theory (Section 3.2.7). As the
temperature rises, nitrification rates increase. The BOD5/TKN  ratio strongly influences the
nitrification rates. Comparing the Manassas,  Blue  Plains and Marlborough  data, it can be
seen that the lower the BOD5/TKN ratio (and the higher  the  nitrifier fraction) the higher
the nitrification rates. Also the effects of pH depression on nitrification rates is apparent.
Particularly  interesting is the data from Blue Plains for the air and oxygen system run at
approximately the same pH, but at differing BOD5/TKN ratios. The air nitrification system,
running at a lower influent BOD5/TKN ratio exhibited higher nitrification rates than the
oxygen system running  at a relatively higher BOD5/TKN ratio. It is notable  that when
oxygen and  air nitrification systems were  run  in parallel  at the same pH and influent
BOD5/TKN ratios,  the same nitrification rates were obtained.^ A further comparison of
oxygen and air nitrification is presented in Section 4.6.5.

In the absence of pilot "data specific to a particular situation, Figure 4-13 can  be used to
approximate design nitrification rates. What must be known or estimated is the BODs/TKN
ratio in the influent, the minimum temperature for  nitrification and the mixed liquor pH. In
essence, Fig. 4-13 is a plot of experimentally determined values of the peak nitrification
rate, as defined in Section 3.2.7 as follows:

                          *  =     f                                         (3-3D
where:      TN   =   peak nitrification rate, Ib NH4 - N oxidized/lb MLVSS/day,

            f    =   nitrifier fraction, and

            q^  =   peak ammonia oxidation rate, Ib NH.  - N rem/lb VSS/day.


In other words, these rates are determined at values where the DO and the ammonia level
are not limiting the rate of nitrification. However, to be useful for design purposes, the
effects of ammonia content and DO should be considered. The  effect of operating DO can


                                        4-55

-------
ON
           0.7
            0.6
         X
         Q
        Q

        \  0.5
        <0
        t/i
        ^
        -J
        §

        -Q  0.4
^
5.5
  TS
IS
S=>
-V
I!
->e -~
o
\;f>-y; >^
  ^^^K^<:;:;^
  V ^, '^\^*^i^ ^, ^< /'  - ^ *\-J>
"-'- -fvi ->^.*>>x;-;-?
" ' *' ^*C>^V"'-""'"'''*-'''"''""°v
                                Marlborough, Mass.
                             (ref. 62) BOD5/TKN = 3.
                                                       +
                                                             /
                                                               Key to Individual Data Points
                                                               •  South Bend, Indiana
                                                                  (ref. 12), pH 7.5,  BOD5/TKN=t.8
                                                               •  Denver, Colorado
                                                                  (ref. 19,20),  BOD5/TKN = 2.7
                         -Blue Plains, pH 6.8 to 7.2 (Air System)
                          (ref.  7  ) BOD5/TKN = 1.3
              10   II    12   13   14   15    16  • 17   18   19   20  21

                                              TEMPERATURE,  C
                                                              Blue Plains, pH 7.0 (Oxygen System)
                                                              ( ref. 8) BOD5/TKN = 3.0
                                                                         +
                                                               22   23  24   25   26  27    28   29   30

-------
be incorporated through the Monod expression for DO. The  effect of desired effluent
ammonia  content can  be  considered through the  safety factor concept.  The  use of
Equations 3-31, 3-20, and 3-29 yields the following expression for either a complete mix, or
plug flow system:                ^    /     nn    \
                          rxi  =—    	—	1                          (4-22)
                          N    SF    I KQ2 + DO j

Effluent ammonia nitrogen  content can be estimated for a complete mix system at steady
state from the equation:
                          i             Ni
                        IF  =   TV^T                                 (4-23)

For plug flow reactors, the effluent ammonia content can be estimated from Equation 4-14.
Criteria for establishing  the  safety factor are discussed in Sections 4.3.3.1, 4.3.3.2,  and
4.6.2.2.

Once the design value of the nitrification rate  is  established, the  design can proceed  in a
manner similar to the F/M  design approach adopted for activated  sludge design. The total
nitrogen load per day and the nitrification rate are used to establish the mass of solids  that
must be maintained in the nitrification reactor. The  volume of reactor is determined from
the allowable mixed liquor  solids level and the inventory of solids  required. The allowable
mixed liquor  level  is  influenced primarily  by the  efficiency of  solids-liquid separation
(Section 4.10).

     4.6.4 Effect of the BOD5/TKN Ratio on Sludge Inventory Control

At  Manassas,  Virginia, Jackson, Michigan, and Contra  Costa,  California, difficulty  was
experienced in maintaining a nitrifying sludge inventory when the influent contained  low
amounts of organics (low BODs/TKN ratios).64,10,4 Typically,effluent solids fluctuated
between 10 and 50 mg/1 and the effluents contained a good deal of dispersed solids  that
were not captured in the secondary clarifier. It has been suggested that a high fraction of the
mixed liquor must be heterotrophic to maintain good bioflocculation in a separate stage
nitrification system. * *• Since  the synthesis of solids in these systems is often less than the
solids appearing in the effluent from the  system, an unstable condition can result. Several
remedies are available. At the three locations mentioned, solids from the upstream activated
sludge process were periodically transferred  to  the separate  stage  nitrification process to
maintain the solids inventory.

In other cases, pretreatment steps have been purposely chosen which do not provide as  high
a degree of carbon removal  as activated sludge and therefore  cause greater synthesis of
heterotrophic  biomass in the nitrification  stage.  Examples  are lime precipitation in the
primary^ and  modified aeration activated  sludge with alum addition. 65 In still another case,
10 percent  of the primary effluent was bypassed around the activated sludge carbon removal
step to a separate stage nitrification step.  11 The amount of primary effluent bypassed can


                                        4-57

-------
be varied to precisely control  the  solids  retention time  and details of a recommended
procedure for accomplishing this control can be found in reference 66.

     4.6.5  Comparison  of the  Use of Conventional Aeration to the Use of High Purity
           Oxygen

A  comprehensive study of the  effect of pH on  covered high purity oxygen  nitrification
systems with comparisons to conventionally aerated systems has recently been completed.^
High  purity oxygen systems  typically  operate  at somewhat lower  pH  levels than
conventionally  aerated systems, and it  has been intimated that this  will result  in lower
nitrification rates and efficiency than conventionally aerated  systems. The reason for the
lower pH is because of the method of oxygenation in covered high purity oxygen systems.
As shown  in Figure 4-14, the  system uses a covered and staged oxygenation basin for
contact of gases and mixed liquor. High  purity oxygen (90+ percent purity) enters the first
stage  and  flows concurrently with the wastewater being treated. The gas is reused in
successive stages, resulting in the buildup of carbon dioxide released by biological activity in
the gas and in the  liquid. This results in  a depression of pH. While the pH is also depressed
by  carbon  dioxide  release in  conventionally aerated systems  (c.f.  Sec. 4.9),  the pH
depression  is less than occurs in  the high purity oxygen system because evolved carbon
dioxide is continually stripped from the system by the aeration air.

The work at the EPA-DC Blue Plains treatment plant consisted of two carefully controlled
pilot investigations as follows: (1) separate stage nitrification with high purity oxygen with
and without pH control and (2)  separate  stage nitrification by conventional aeration and

                                   FIGURE 4-14
                COVERED HIGH PURITY OXYGEN REACTOR WITH
                 THREE STAGES AND MECHANICAL AERATORS
 OXYGEN
 FEED GAS
   WASTE
   LIQUOR
   FEED
   RECYCLE
   SLUDGE'
 EXHAUST
"GAS
                                                                            MIXED LIQUOR
                                                                            EFFLUENT TO
                                                                            CLARIFIER
                                        4-58

-------
high purity oxygenation controlled at the same pH level.  The purpose of this section is to
review the salient features of this work as it affects nitrification design with high purity
oxygen. No attempt will be made to present general high purity oxygen design concepts as
these are covered in the EPA Office of Technology Transfer's publication, Oxygen Activated
Sludge Wastewater Treatment Systems Design Criteria and Operating Experience. 67

          4.6.5.1 High Purity Oxygen Nitrification With and Without pH Control

Operational results are shown in Table 4-13 for a high purity oxygen pilot plant without pH
control and a plant with the last of four stages held at pH 7.0 by lime addition to the first
stage. As may be seen from the table, the system with pH control provided an effluent with
somewhat better quality. Lime addition to the pH controlled reactor caused  buildup of
inerts which resulted in  better thickening sludge and better clarification efficiency at the
expense of greater sludge production.^ The effluent ammonia level was somewhat lower
with pH control, but the difference is not significant when it is considered that a chlorine
dose as little as 10 mg/1 would be sufficient to remove all traces of ammonia in both
wastewaters (c.f. Chapter 6). It should be noted  that both systems were operated at high
solids retention times and therefore had high SF values. Therefore,  differences in effluent
ammonia levels would be expected to be small.

Comparisons of nitrification rates did not present a clear picture of differences between the
two  systems.   However, it was shown  that when the pH drops  below 6.0 nitrification
rates did decline. However, the pH generally did not drop below 6.0 in the system without
pH control until the last reactor stage, but there was so little  ammonia remaining to be
oxidized there that essentially no effect of  pH on  nitrification  performance  could be
discerned.

In sum, the work at Blue Plains demonstrates that the pH of the nitrification reactor can
drop as low as  6.0 and allow acclimation of the nitrification organisms  with attainment of
complete nitrification. The pH should not be allowed to drop below 6.0, except perhaps in
the last reactor stage, and in those cases where the carbon dioxide evolution is sufficient to
cause the pH to drop below 6.0, pH control should  be implemented. From an organics
standpoint, effluent qualities are superior in the pH  controlled system because of lime
addition,  and  this  should  be kept  in mind  when  designing  for  stringent  effluent
requirements.

         4.6.5.2 Comparison of Conventional Aeration and
                High Purity Oxygen at the Same pH

Table 4-14 shows the results of a parallel study of a conventional aerated nitrification
system with a high purity oxygen system; both systems were held at pH 7.0 in the last stage
of a four-stage  system.^ As may be seen from  the table, the concentration of organics and
nitrogen species were virtually identical in the two systems. Greater lime  was required in the
high purity oxygen system than in the  conventional system to maintain the same pH level.

                                        4-59

-------
This greater lime dose resulted in greater sludge production in the oxygen system but also
improved the sludge thickening properties, allowing the same MLVSS level to be maintained
in both systems at a higher MLSS level in the oxygen system. Nitrification kinetic rates were
found  to be the same  in both  systems. Choice between the two pH controlled systems
should be based on economic considerations as the systems are equivalent in other respects.


                                     TABLE 4-13

                COMPARISON  OF PROCESS CHARACTERISTICS FOR
             OXYGEN NITRIFICATION SYSTEMS WITH AND WITHOUT
                pH CONTROL AT BLUE PLAINS, WASHINGTON, D.C.
                                       With pH
                                       control,
                                     days 51-150a
Without pH
  control,
days 26-8Sa
                               Operating Parameters'3
Oxygenation time, hrs°
F/M ratiod
Solids retention time, days
MLSS, mg/1
MLVSS, mg/1
Return sludge SS, mg/1
Sedimentation tank overflow rate ,
gpd/sq ft
m /m2/day
Sludge production, mg/1
Lime dose (CaO) , mg/1
4.0
0.15
13
5,660
2,620
38,650
620
25.3
69
126
3.9
0.16
17
3,520
2,780
13,640
645
26.3
35.4
0
                                Effluent Qualities13

Concentration, mg/1
BOD 5
COD
SS
VSS
TKN
NHj-N
NO^NOj-N

mean
67
152
77
58
21.7
14.9
-
Effluent
mean
6.3e
26.4
10
6.0
1.3
0.15
13.4
standard
deviation
3.1e
6.2
4.8
4.2
0.6
0.11
1.1

mean
70
152
75
58
22.7
150
-
Effluent
mean
8.6e
42.5
24.0
17.1
2.4
0.64
12.8
standard
deviation
3.06
8.2
10.2
8.6
0.7
0.64
12
  Day 1 = January 1, 1974
  Data from reference 8 ; while the data are not from the same time period, data from a common time
  period (days 56-85) showed the same trends.
 GBased on influent flow

  F/M ratio is the ratio of the Ib BODj in the influent to the activated sludge process and the Ib of
  MLVSS inventory under aeration or oxygenation.
 eNitriftcation inhibited
                                         4-60

-------
4.7 Separate Stage Nitrification with Attached Growth Processes

Three  types  of attached  growth  processes have been  employed  for  separate  stage
nitrification. The differences lie in the type of medium provided for biological growth. The
three types of processes are the trickling filter, the rotating biological disc and the packed
bed reactor.
                                   TABLE 4-14

     COMPARISON OF PROCESS CHARACTERISTICS OF CONVENTIONALLY
           AERATED AND HIGH PURITY OXYGEN SYSTEMS WITH pH
                CONTROL AT BLUE PLAINS, WASHINGTON, D.C.
                                      High purity
                                       oxygen,
                                     days 186-2653
 Conventional
diffused aeration,
 days 186-265a
                                Operating Parameters
Oxygenatlon time, hr
F/M ratiod
Solids retention time, days
MLSS, mg/1
MLVSS, mg/1
Return sludge SS, mg/1
Sedimentation tank overflow rate,
gpd/sq ft
3 2
m /m /day
Sludge production, mg/1
Lime dose (CaO) , mg/1
3.8
0.16
10.0
6,355
2,260
40,850
656
26.7
97.6
128
3.7
0.16
9.9
3,890
2,240
16,460
678
27.6
59.2
47
                                  Effluent Qualities

Concentration, mg/1
BOD5
COD
SS
VSS
TKN
NH*-N
NO"+ NO"-N

Influent
mean
56
130
65
49
19.5
15.2
-
Effluent
mean
3. .5°
20.6
8.1
4.8
0.94
0.19
13.6
standard
deviation
1 . 1 '
2.6
2.6
2.1
0.23
0.20
1.2

Influent
mean
56
130
65
49
19.5
15.2
-
Effluent
mean
3.4
20.6
6.2
4.2
1.0
0.19
12.5
standard
deviation
1.3e
3.4
3.4
2.5
0.3
0.16
1.2
 Day 1 = January 1, 1974
b
 Data from reference 8
 Based on influent flow
 F/M ratio is the ratio of the Ib BOD  in the Influent to the activated sludge process
                         o
 and the Ib of MLVSS Inventory under aeration or oxygenation.
eNitrlflcation inhibited
                                        4-61

-------
     4.7.1 Nitrification with Trickling Filters

The development  of two-stage trickling filtration or  double  filtration preceded the
development of two-stage suspended growth systems for nitrification. In fact,  two-stage
filtration was in operation at several military installations during World War 11.4° initially,
the two-stage trickling filtration process was developed to increase the removal of organics
in the effluents from  the high rate trickling filters. Later,  it was observed that under some
operating conditions,  the second stage produced a well nitrified effluent.68

In separate  stage nitrification application, trickling  filters can follow a high rate trickling
filter plant with intermediate clarification, or an activated sludge process or any of the other
alternatives listed in Fig. 4-11.

         4.7.1.1 Media Type and Specific Surface

It has been shown that in separate stage nitrification applications, the rate of nitrification is
proportional to the surface area exposed to the liquid being nitrified.69,70 In other words,
when all other  factors are held constant, the  allowable loading rates can be expected to be
related to the media surface area, rather than to the media volume.

Very little  biological film development  has been  observed in separate stage applica-
tions.71>72 AS  a consequence, pluggage of voids in the media and ponding becomes of less
concern  than   in combined  carbon oxidation-nitrification  applications.  Media of  higher
specific surface than  normally employed may be used. Plastic media is  characterized  by
having very high specific  surface available while maintaining a high void ratio (  > 90
percent).  The high specific surface area of plastic media allows the trickling filter volume to
be reduced, significantly reducing the cost of the distributor arms and the structure.

Available types of plastic media are summarized in Table 4-15. Most experience in the U.S.
has been with the corrugated sheet module type, rather than with the dumped media which
has just  become  available. Media applicable  to nitrification applications is commercially
available in specific surfaces ranging from 27 to 68 sq ft/cu ft (89 to 223 m^/m^).

         4.7.1.2 Loading Criteria

As previously stated,  nitrification rates in trickling filters are related to the wetted surface
area of the media. Therefore, the most rational criterion would be in terms of surface area.
Unfortunately,  information  on specific surface  is  not always available, and volumetric
loading criteria must occasionally be resorted to.

The pilot study at  the Midland, Michigan wastewater treatment plant provides  the most
comprehensive  set of  data currently available on nitrification with trickling filters.22,71 jj-ie
influent to  the pilot plant was well treated  trickling filter effluent with BOD5, SS and
ammonia-N values ranging  from 15-20, 15-20  and 8-18 respectively.  The BODs/TKN ratio

                                         4-62

-------
                                   TABLE 4-15

                  COMMERCIAL TYPES OF PLASTIC MEDIA FOR
                SEPARATE STAGE NITRIFICATION APPLICATIONS
Manufacturer
Envlrotech Corp . , Brisbane ,
Ca.*
B.F. Goodrich, Marietta, Ohio

Enviro Development Co. , Inc.
Palo Alto, Ca.b
Mass Transfer, Ltd. , Houston,
Texas
Norton Co. , Akron, Ohio

Munters Corp., Ft. Meyers, Fla.

Trade
Name
Surfpac

Vinyl Core

Flocor

Filterpack

Actifil

PLASdek

Type
Corrugated shcat
.modules
Corrugated sheet
modules
Corrugated sheet
modules
Dumped rings

Dumped rings

Corrugated sheet
module s
Specific surface available
sf/cu ft (m / m )
27

30.5
45
27
40
36
57
27
42
42
68
(89)

(100)
(148)
(39)
(131)
(118)
(189)
(89)
(138)
(138)
(223)
  Formerly available from the Dow Chemical Co.,  Midland, Mich.
  Under license from ICI, Great Britain; formerly available from the Ethyl Corp.,
  Baton Rouge, La.

was 1.1, indicating a high degree of BOD removal in the pretreatment stage. The pilot unit
was a 21.5 ft (6.55 m) unit filled with Surfpac media. During the 18 month project period, a
variety of climatic conditions were  experienced with wastewater temperatures in the pilot
unit as low as 7 C and as high as 19 C encountered.

The data from the various operating periods for the project have been reexpressed in Figure
4-15 in terms of the surface area required for nitrification and the desired effluent ammonia
nitrogen content. As may be seen, greater surface area is required at low temperature (7 to
11 C) than high temperatures (13 to 1-9 C). Further, to obtain ammonia-N contents below
2.5 to 3.0 mg/1, greater surface area is required than for effluent ammonia contents above
2.5 to 3.0 mg/1 ammonia-N.

When effluent  ammonia-N levels  less than 2.5 mg/1 are desired, consideration should  be
given  to using breakpoint chlorination (Chapter 6) for removing ammonia residuals rather
than increasing the surface area of the filter, as the cost of removing the last 1-3 mg/1 of
ammonia becomes very high because of the very much larger trickling filters required.

Figure 4-15 was developed  solely  from the Midland, Michigan data. Data are available from
two other locations which allow calculation of surface requirements for nitrification.2',73
Figure 4-16 shows surface  reaction rates for Lima,  Ohio  data^7 compared with the trend
lines developed from the Midland, Michigan data. Lesser surface area is required at the Lima
                                        4-63

-------
location for the same degree of nitrification. This lower surface requirement is chiefly due
to the higher wastewater temperature, but the fact that the influent 6005 levels were lower
may also have caused a higher proportion of nitrifiers to be present in the trickling filter's
surface film. In the case of Lima, Ohio, the influent to the nitrification stage is produced by
a step aeration activated sludge plant.

Surface  requirements for nitrification of oxidation pond effluent at Sunnyvale, California
are shown in  Figure 4-17.'^ In  this case,  large quantities of algae were present in the
trickling filter influent. While the bulk of the algae passed through the unit unaffected,  at

                                   FIGURE 4-15
e
5
3
o
Ui
Uj
tt
UJ
              SURFACE AREA REQUIREMENTS FOR NITRIFICATION
                               MIDLAND MICHIGAN
    12,000
 X


 a" 10,000
 kl
 N
 i    8,000
f
 QQ
    6,000
    4,000
    2,000
                                               ©
              1 SF/lb/day = 0.2 m2/ kg/day
                                                   Influent Data (mean)
                                                    BOD5    !5-20mg/l
                                                     SS     15-20 mg/l
                                                    Organic N  1-4 mg/l
                                                    IMH^-N   8-18 mg/l
                                                    BOD5/TKN ~ 1,1
                                                          7 to  II C
                                                                             •f
                                                         \?> \o 19 C
                                                                           ©	—
                                                       ©
                                                         Key :
                                                         H  T = 7 to  11C
                                                         ©  T = 13 to I9C
                   I.O         2.0         3.0        4.0

                           EFFLUENT  AMMONIA-N, mg/l
                                                                 5.0
6.0
                                      4-64

-------
least 20 to 40 percent were trapped and eventually oxidized. This would have affected the
proportion of heterotrophic bacteria in the bacterial film, causing higher surface requirements
for nitrification at Sunnyvale, California than at Midland, Michigan.

Available data with rock media is more sparse and is summarized in Table 4-16 in terms of
ammonia oxidized per unit volume. Rock media is capable of ammonia oxidation at only 15
to 50 percent of the  plastic media rate, on a volumetric load basis. The principal reason for
this is undoubtedly the rock media's lower specific surface, although the lower depth of the
typical rock filter may also have a role to play.

                                  FIGURE 4-16
             SURFACE AREA REQUIREMENTS FOR NITRIFICATION -
                                  LIMA, OHIO
   12,000
   10,000
X
o
S:

-------
                                  TABLE 4-16
      NITRIFICATION IN SEPARATE STAGE ROCK TRICKLING FILTERS


Facility location
Johannesburg, S.A.
(full-scale)







North Hampton,
England
(pilot-scale)


Ref.
47








72



Depth,
ft
(m)
12
(3.7)

12
(3.7)

9
(2.7)

6
(1.8)



Media
2-3 in.
(5.1 to 7.6 cm)
rock
1.5 in.
(3 . 8 cm)
rock
1 in-
(2 . 5 cm)
rock
1.5 in.
(3 . 8 cm)
rock
Influent

BOD
mg/1
28


32


23


80


+.
NH4-N,
mg/1
23.9


25.2


22


33


Effluent

BOD ,
mg/1
14


13


10


10


NHT-N
4
mg/1
8.3


4.4


9.1


11.2


Percent
removed
65


83


59


66


Ammonia - N
oxidized
Ib/lOM cu ft/day
(kg/mVday)
3.5
(0.055)

2.2
(0.035)

2.4
(0.038)

1.0
(0.016)

         4.7.1.3 Effect of Recirculation

An analysis of the Midland, Michigan data and Lima, Ohio, data has led to the conclusion
that  while  recirculation improved nitrification efficiency only marginally  on an average
basis, the periods with recirculation demonstrated greater consistency (less fluctuations)
than  when  no  recirculation  was employed.  '2'  This  conclusion,  together with  the
improvements seen  with recirculation in combined carbon oxidation-nitrification applica-
tions (Section 4.4.1.4), leads to a general recommendation for the provision of recirculation.
A  1:1 recirculation ratio  is considered adequate  at average dry weather  flow for most
applications.

         4.7.1.4 Effluent Clarification

Since the  organisms are  attached to  the  media  in attached growth systems,  effluent
clarification steps are not required in all cases. In the case of Midland, Michigan it was found
that the effluent solids were approximately equal to the influent solids at 9 to 28 mg/1.22
This is because influent BODs levels were low (15 to 20 mg/1). When influent BODs loads
were increased above previous low levels, trickling filter effluent solids rose to 58 mg/1. The
insertion of clarifier allowed this to be reduced to 19 mg/1. Subsequent multimedia filtration
allowed further reduction to about 4 mg/1.

         4.7.1.5 Effect of Diurnal Load Variations

Trickling filters used for nitrification, like any other nitrification process, are affected by
diurnal  variations in nitrogen load. The rule of thumb developed in Section 4.3.3.2  can
likely be applied to trickling filters to prevent high ammonia bleed through during diurnal
peaks in load. Thus, the amount of surface area determined from Figures 4-15, 4-16 or 4-17
                                        4-66

-------
under average daily loading conditions should be multiplied by the ratio of peak ammonia
load to average load to establish design surface area. An alternative would be to provide flow
equalization.
                                 FIGURE 4-1 7
 Q /2,000

\
 Q
 UJ
 N
   10,000
 I
m
 CD
 -J


 U.
 CO
 UJ
 tt
 UJ


 tt

 CO
    8,000
  .  6,000
   4,000
    2,000
              SURFACE AREA REQUIREMENTS FOR NITRIFICATION -
                          SUNNYVALE, CALIFORNIA
                    1
            18, 20O
                                         1
                                                   1
                                                   1
                                              Influent Data (mean)
                                              COD   =210 mg/l
                                               SS   =104 mg/l
                                              Organic N = 10.0 mg/l
                                              NHj-N = 16.7  mg/l
                                                              0
                                                          ©
                                           Sunnyvale, California  Data
                                             13 C to 19 C
                                                         0
                                                                      O-"f
                                                                       18
                                          Midland, Michigan Data
                                             13 C to  19 C


                                           Key:
                                            © Test result  from
                                              Sunnyvale  work

                                                   1          1
                             468

                       EFFLUENT  AMMONIA-N, mg/l
                                                              10
12
                                   4-67

-------
         4.7.1.6 Design Example

As an example consider a  10 mgd conventional activated sludge plant that must be upgraded
to meet effluent requirements of 4 mg/1 ammonia nitrogen and 10 mg/1 suspended solids on
an average  basis. The plant is located in  a temperate zone, and the minimum wastewater
temperature is 15  C.  Present average effluent qualities are 1  mg/1 organic nitrogen, 20 mg/1
ammonia nitrogen,  15 mg/1 of suspended solids, and a BOD5 of 25  mg/1. The  peak to
average nitrogen load ratio is 1.9. Consider as one alternative a plastic media trickling filter.

     1.   Calculate the BODs/TKN ratio:

                  BOD5/TKN = 25/21 = 1.19

     2.   The  closest set  of data  based on  BODs/TKN ratio and  temperature is that for
         Midland, Michigan (Figure 4-15). For an effluent ammonia nitrogen concentration
         of 4 mg/1 at 15 C, the unit surface area requirement is about 3,800 sf/lb Nlfy-N
         oxidized/day.

     3.   Calculate  the ammonia  nitrogen  oxidized daily. The  following equation is
         appropriate:

                    NT = 8.33 • Q(NQ - Nj)                                    (4-24)


          where:      NT  =   ammonia nitrogen oxidized, Ib per day
                     Q    =   average daily flow, mgd
                     NQ  =   influent NH* - N, mg/1
                     Nj  =   effluent NH+ - N, mg/1

          For this example, assuming no  change in the organic nitrogen level, the result is:

                   NT = 8.33 (10) (20-4) = 1,332 Ib/day

     4.    Find the total surface area requirement under average load conditions. Multiply-
          ing  the nitrogen  oxidized per day (step 3) by the unit surface area requirement
          (step 2) results in:

                   (1,332) (3,800) = 5,061,000 sf of media

     5.    Consider   diurnal  peak  loading.  One  approach  would be to provide flow
          equalization. Assume that in this case  site restrictions  prevent this. Therefore,
          increase the surface requirement by the peak  to average nitrogen  load ratio as
          follows (Section 4.7.1.5):

                                       4-68

-------
                   1.5 (5,061,000) = 7,592,000 sf

     6.   Choose a media type and establish media volume requirements. Effluent BOD5
         and SS are low enough so that fairly high density media can be employed. In this
         instance, a corrugated sheet module media having a specific surface of 42 sf/cu ft
         is chosen. Media volume requirements are determined by dividing the total surface
         requirement by the specific surface as  follows:

                   7,592,000/42 = 180,750 cu ft

         This volume could be provided by a variety of configurations; for instance two 75
         ft diameter trickling filters with a media height of 21 ft would have the necessary
         volume of media. Whatever configuration is chosen, the filter shouldn't be less
         than about 12 to  15 ft in height because of the danger of short circuiting. Usual
         practice is to  consult with the media manufacturers)  prior to final selection of
         media configuration.

     7.   Establish recirculation rate. At 10 mgd ADWF, a 1:1 recycle is adequate (Section
         4.7.1.3);  therefore 10 mgd of recirculation capacity is recommended.

     8.   Establish clarification requirements.  Effluent  solids in the nitrification process
         effluent will  be approximately at the influent SS level, 15 mg/1. Therefore, to
         meet a 10 mg/1 requirement, some form of effluent clarification is required such
         as dual or multimedia filtration.

     4.7.2 Nitrification with the Rotating Biological Disc Process

The rotating biological disc  (RED) process, discussed in  Section 4.4.2 for combined carbon
oxidation-nitrification applications  may also be applied to nitrifying secondary effluents.
The process is constructed as shown in Fig.  4-8, excepting that it may be possible to
eliminate the secondary clarifier when the secondary effluent being treated has a BOD5 and
suspended  solids less than  about  20 mg/1.74  Under this circumstance  the very low net
growth occurring in the nitrification stage causes the RBD process effluent suspended solids
to approximately equal  the influent solids level. If lower levels of suspended solids  are
required, the RBD process could be followed directly by  tertiary filtration without the need
for intermediate clarification.7'*

One manufacturer  has  announced  the  availability  of media  especially adapted  to
nitrification.  The  minimal  biomass film  development  in separate stage nitrification
applications has allowed a 50 percent increase in surface area of the corrugated polyethylene
media. Standard  shafts were 100,000 sq ft (9300 m^)  of available  surface area; the new
media is available  at 150,000 sq ft (13,900  m^) of surface per shaft. This results in a
reduction of one-third in the number of shaft assemblies required for nitrification with the
RBD process.74

                                         4-69

-------
         4.7.2.1 Kinetics

The reaction rates occurring in each stage of the RED process treating secondary effluents
have been  analyzed; the correlation between  surface reaction rates and stage  (effluent)
concentration is shown in Figure 4-18.74 jne trend line  does not reach a plateau value but
keeps gradually rising  because  the  biomass developed per unit surface is not constant.
Antonie found  that the amount of culture developed on the rotating surface increased with
increasing ammonia nitrogen concentration. 74

A  stage-by-stage application of Fig. 4-18 allowed the construction of Fig.  4-19 to be used
for design of 4  stage nitrification systems,  the most commonly employed configuration.74
It  can also  be employed for other numbers of stages using the relative capacities shown in
the figure. The  relative capacity factor should be applied to the hydraulic loading to obtain
design values for situations where other than 4 stages are employed.

Very little test  data is available for temperatures below 13 C. For applications below 13 C,
the provisional  recommendation has been made that the temperature correction factors
                                   FIGURE 4-18
         NITRIFICATION RATES AS A FUNCTION OF STAGE EFFLUENT
                   CONCENTRATION (AFTER ANTONIE (74))
 1-
 u.
 o
 CO
Q
\
    1.0
    0.8
 Q
 iu
 lu
 O
I
    O.6
    0.4
    "
              T
                              T
I
         © Phoenix, Arizona,  22.8 to  27.8 C
         A Madison, Wisconsin , 14.5 to 17.3 C
         Q Broward County, Fla., 21 C
         <•> Mansfield, Ohio
         • Tiffin, Ohio,  13.9 C
                                               Conditions:  BOD5<20mg/l
                                                           Lb/day/iOOO sq. ft =
                                                           4.85 kg/1000 sq. m/day
                                                 I	I	1	I
              2        4        6        8       IO      12       14
                   EFFLUENT AMMONIA NITROGEN CONCENTRATION, tng/l
                                                                         16
                                                                                 IB
                                        4-70

-------
developed for combined carbon oxidation-nitrification (Section 4.4.2.2) be applied for
separate stage nitrification. '4

Figure 4-19 may also be used for hour-by-hour analysis of the effects of diurnal variations in
flow on effluent quality.74 This may tend to overestimate effluent quality during peaking
periods, however. To ensure that severe ammonia bleedthrough does not occur during peak
load periods, it would appear prudent to adopt the rule formulated in Section 4.7.1.5 for
trickling filters. Namely, the surface area determined from Figure 4-19 should be multiplied
by the ammonia nitrogen peaking ratio to establish the design surface area.
                                   FIGURE 4-19
             DESIGN RELATIONSHIPS FOR A 4-STAGE RED PROCESS
            TREATING SECONDARY EFFLUENT (AFTER ANTONIE (74))
   too
 l-
 e
 LU
 o
 I3C
                                               BOD5 <20 mg/l
Inlet Ammonia Nitrogen
   Concentration, mg/l
                                 40
                                      30
          1  gpd/sq ft = 41 Jt/mz/day
                                  I
                        2345
                           HYDRAULIC  LOADING, GPD/SQ FT
                                      4-71

-------
    4.7.3 Nitrification with Packed-Bed Reactors

Packed bed reactors (PER) for nitrification are a comparatively recent development, having
progressed  from the laboratory stage to pilot-scale and commercial availability in a period of
only 5 years.23,24,75,76,77,78,79

Figure 4-20 shows one design.77,80 A pgR consists of a bed of media upon which biological
growth occurs overlaying an inlet chamber, much  as in an upflow carbon column or filter.
Wastewater is distributed evenly across the floor of the PBR by baffles, nozzles or strainers,
similar to  the  way backwash water is distributed in down flow rapid sand filters. The
wastewater flow is upward, and a nitrifying biological mass is developed on the large surface
area of the media.
                                FIGURE 4-20

                 SCHEMATIC DIAGRAM OF A PACKED-BED
             REACTOR (PBR).  (AFTER YOUNG, ET AL., REF 77)

                                                              EFFLUENT
                                                     SUPPORT MEDIA
                                                     INLET CHAMBER
         WASTEWATER INFLUENT
                                     4-72

-------
         4.7.3.1 Oxygenation Techniques

 Several means have been employed for supplying the necessary oxygen for nitrification. The
earliest work used injection of air into the feed line entering the chamber.79 A subsequent
pilot-scale investigation used  a similar procedure, excepting that the air was distributed
across the PBR floor, as shown in Figure 4-20.77 High purity oxygen has been used in two
alternative procedures.23,75,76 in one the oxygen was bubbled directly into the PBR.  In
the second procedure, the liquid was preoxygenated in a reaction chamber prior to entry
into the PBR. Since preoxygenation is limited to satisfying the oxidation of about 10 mg/1
NH^-N due to the solubility of oxygen in water, effluent was recycled at a 2 to 3:1 ratio to
provide sufficient oxygen for nitrification.

          4.7.3.2  Media Type, Backwashing and Loading Criteria

Several types  of media have successfully performed in the PBR including 1-1.5 in. (2.5  to
3.8 cm) stones, 0.5  cm gravel, 1.8  mm (effective size) anthracite  and 9 cm  "Maspac," a
plastic dumped media manufactured .by  the Dow Chemical  Company, Midland, Michi-
gan.76,75,23,81,77

In the studies using the relatively light density anthracite and Maspac where ah- was injected
directly into the PBR, no backwashing was found  to  be necessary due to the turbulence
developed in the bed.77 Despite this, the General Filter Company recommends that when
anthracite is used, provision be made for increasing the hydraulic loading in surges for  1 to 2
hours to about four times the average ra'te, with air at a rate of 0.5 to  1.0 scfm/sq ft (2.54 to
5.08  l/s/m^). The frequency  of the surging  will vary, depending on influent quality and
flow rate. With the plastic media, the frequency of the surging can be reduced considerably
because of the high void volume, and in most cases excess solids can be withdrawn simply
by draining or backflushing the unit on a monthly or less frequent schedule.^

With  the gravel media, standard practice was to backwash the reactor at 25 gpm/ft^ (127
 1/s/m^) at least three times per week and in some cases daily.** 1 In the studies with the
stone  media,  backwashing was required with both direct and pre-oxygenation.  Gravity
draining at 6 to 20 gpm/sq ft  (30 to 102 l/s/m^) once or twice per week was sufficient to
prevent clogging. •*>^4

Data  available for formulation of design criteria for PBR units are summarized in  Table
4-17. Oxidation rates fall in the range of 4 to 27 Ib NH^-N oxidized per 1000 cu ft/day
(0.06  to 0.43 kg/m^/day).  Factors  affecting the oxidation rate are the influent quality
(BODj, TKN  and NH^-N), temperature, and the type of media selected as a biological
growth surface. Oxidation rates at Pomona, Ca. were much greater than those at Ames, Iowa
at the same  temperature, which is  very  probably due to the higher BOD5 and  lower
ammonia content of the Ames secondary effluent. Very likely, there was a higher fraction
of nitrifiers in the Pomona biofilm. Interestingly, chemically clarified raw sewage (BOD = 93
mg/1) was compared to secondary effluent (COD = 46) at Pomona and only 60 percent

                                        4-73

-------
nitrification was achieved with the chemically clarified feed at the detention time sufficient
to produce virtually complete nitrification of the secondary effluent.** 1  It is very probable
that  this reduced efficiency was caused by a higher fraction of heterotrophs being present
when chemically clarified wastewater was being treated.

Temperature has a strong effect on the PER process. Figure 4-21 shows  the detention time
required for relatively complete nitrification (< 2 mg/1  NH^-N) at steady state as a function
of temperature. If Figure 4-21  is used for sizing a PBR,  attention must also be given to
diurnal variation in nitrogen loads. It would be prudent  to multiply the detention time
determined from Figure 4-21 by the peak to average nitrogen load ratio to establish the
design  detention time. This should present extremes in ammonia bleedthrough during
diurnal peak conditions.

The  media type chosen affects the amount of surface available for nitrifier growth. For
instance, Table 4-17 shows that anthracite was superior to Maspac media in terms  of
oxidation rate at Ames, Iowa; this may be due to the higher surface area of anthracite media
compared to Maspac.
                                   FIGURE 4-21
       TEMPERATURE DEPENDENCE OF DETENTION TIME FOR COMPLETE
         NITRIFICATION, (<2 mg/1 NH*-N) AT STEADY STATE IN THE PBR
   30
   25
 Uj
 tt
 ft:
 LU
 Q.
 §
 LU
   20
    15
   O
L-  O
   A
 KEY:
   Ref.
S  81
   81
   81
   23
   77
   77
   77
                        Tiiiir~
                              •TREND  LINE  FOR  POMONA DATA  (Ref. 81)
            Location
Media
                      Comment
        One Column, 02
        Two Columns, 02
        One Column, Air
Pomona, Ca.   Gravel
Pomona, Co.   Gravel
Pomona, Co.   Gravel
Union City, Ca. Stone
Ames, Iowa   Anthracite
Ames, Iowa   Maspac
Ames, Iowa   Series Maspac-Anthracite
                                                    I
     40
                  60           SO        IOO     I2O
                  EMPTY  BED DETENTION TIME,  MIN
                                 I4O
                                                    160   180  2OO
                                      4-74

-------
                                                         TABLE 4-17
                     PACKED BED REACTOR PERFORMANCE WHEN TREATING SECONDARY EFFLUENTS

Location
and type
of aeration
Union City, Ca.
Preoxygenatlon
(oxygen)




Bubble
(oxygen)


Ames, Iowa
Bubble
(air)

fc
J
n





Pomona, Co.
Bubble
(oxygen or
air)











Ref.

23,
24




23,
24


77,
83










81














Media
depth ,
ft
(m)

3.0

(.91)



3.0
(.91)


5.0
(1.5)


e.o
(2.4)


13
(4.0)


5.5
(1.7)








11.0
(3.4)



Media
type

1 to 1.5 In

(2 . 5 to 3 . 8 cm)
stone


1 to 1.5 in
(2.5 to 3.8cm)
stone

1 .8 mm (Dio)
U.C. = 1.7
anthracite

Maspac



Series
anthracite
Maspacb

5 mm
gravel








5 mm.
gravel ,
two columns
in series

Surface
loading
gpm/sf
(m3/m2/day)

.15
(8.8)
.21
(12)
.29
(17)
.15
(8.8)
.29
(17)
. 1.0
(59)
0.4
(23)
1.0
(59)
0.4
(23)
0.5
(29) b
0.75
(44)
0.75
(44)
0.59
(35)
0.41
(24)
0.46
(27)
0.39
(23)
1.49
(87)
0.75
(44)
Empty
bed
hydraulic
detention
a
time , min

154

103

77

1S4

77

" 37

94

60

ISO

195

130

55

70

100

90

105

55

110

Ammonia-N
oxidation
rate
lb/1000 cu ft/day
(kg/m3/day)

7.7
(.12)
12.1
(.19)
9.8
(.15)
9.7
(.15)
-

9.4
(.15)
5.9
(0.09)
5.1
(0.08)
3.5
(0.06)
6.1
(0.10)
4.6
(0.07)
26.5
(0.42)
20.1
(0.32)
13.3
(0.21)
14.7
(0.24)
16.4
(0.26)
26.7
(0.43)
13.8
(0.22)

Temp. ,
C

21 to 27

21 to 27

21 to 27

16 to 30

IE to 30

21 to 23



21 to 23



9 to 14

12

27 to 28

25 to 26

19 to 22

20 to 25

20 to 22

26 to 28

22 to 25


Influent Quality, mg/1
BOD5

35

38

37

37

43

39 ;

20

39

20

26

37

7C













SS

27

38

25

30

48

.13

26

43

26

47

63

9c













Organic-N

3.6

-

5.7

5

-

'_

-

-

_

_

-

_

-

-

-

-

_

-

Ammonla-N

14.3

19.6

15.2

18.3

-

8.4

6.8

8.4

6.8

14.4

11.2

18.1

17.6

16.8

16.4

20.7

17.6

18.9


Effluent Quality, mg/1
BOD5

5

3

10

10

25

19'

5

19

8

8

16

_

-

-

-

-

_

-

SS

4

7

7

16

51

;

_

_

_

_

_

.

-

-

-

-

_

-

Organic-N

1.5

-

2.5

2.2 - 4.7

-

;

•

_

_

_

-

_

-

-

-

-

_

-

Ammonia-N

1.0

5.6

6.9

1.8

-

5

1

5

1

1

5

1.9

1.4

2.0

1.7

1.5

1.3

1.9

Nltrite-N

0.6

4.1

1.7

0.4

-

• .. •

_

-

_

.

_

0.6

0.6

0.9

0.4

0.5

0.4

0.2

Nitrate -N

15.9

6.9

6.0

18.3

-

.

_

_

_

_

_

16.6

16.3

16.9

15.6

20.7

17.1

18.2


Removals,
percent
BOD

86

91

74

74

41

51

77

51

59

68

56






~



-

-


SS

87

83

74

48

-6

20

32

42

40-

49

39



_



_







Organic -N

58

-

56

_

-

_

-

-

-

-

_



_



_







NH+-N

93

71

55

91

-

46

90

40

87

92

59

90

92

88

90


93

93

90
 Basis: Influent flow
 A product of the Dow Chemical Co., Midland, Mich.
cAverage for test series treating" activated sludge effluent

-------
Effluent BOD  and  SS levels  are  affected by  the type  of aeration  (see  Table 4-17).
Preoxygenation  allows the  PBR to  produce  effluents  of  similar quality to  tertiary
multimedia, filtration.  Bubble  aeration,  however,  causes  continuous  shearing  of  the
biological film from  the media, resulting in lower reductions of BOD and suspended solids.
When very low levels  of effluent solids are required, effluent filtration  may be required
when bubble aeration is used in the PBR.

4.8 Aeration Requirements

Care  must  be  exercised  in  designing  aeration  systems  for  nitrification.  Unlike BOD,
ammonia is not adsorbed to the biological floe for later oxidation, and therefore the ammonia
must be  oxidized during  the relatively  short  period  it is in the nitrification  reactor.
Therefore, sufficient oxygen must  be provided  to handle  the  load impressed on  the
nitrification process  at all times. This problem is particularly critical when either activated
sludge or a packed bed reactor system is used for nitrification; in the other attached growth
systems  the  aeration  is provided  as the liquid  spills over  the  media and  the design
considerations relate  to proper ventilation rather than oxygen transfer.

Very significant diurnal changes in nitrogen load have been observed. Load variations at the
Chapel  Hill  treatment plant  are shown in  Fig.  4-3  and the load pattern  would be
representative of systems provided with no significant in-process flow equalization. In  this
case, peak  to average nitrogen load rose to  nearly 2.2,  considerably above the  peak to
average flow ratio of 1.44. As an example  of a plant with some in-process  flow equalization,
Fig. 4-12 shows the  load variations observed in the activated sludge effluent of the Rancho
Cordova Wastewater Treatment Plant. In this case the nitrogen peaking is  moderated by the
equalization in  quality provided by the activated  sludge aeration tanks  and  secondary
clarifiers. The ammonia peak to average ratio at 1.63 approximates the flow peak to average
ratio of 1.57.

In  addition to being  affected by  in-process  flow equalization, the diurnal variation in
nitrogen  loading is  also  very significantly  affected by  equalization in  the  wastewater
collection system. Large collection systems serving spreadout urban areas have high built-in
storage providing unintentional flow and quality equalization.  This relationship is indicated
in Fig.  4-22 where the nitrogen load peaking (expressed as the ratio of the maximum hourly
load  to average  load)  is plotted for eight treatment plants having  no significant in-process
equalization. There  is an interesting relationship between  flow peaking and ammonia load
peaking shown in Figure 4-22. In large  plants  such as Blue Plains plant at Washington, D.C.
and Sacramento, California a spread out collection system causes moderation of both flow
and  nitrogen load   peaking.  In  the  smaller  systems,  however, without such "flow
equalization," ammonia load peaking can  be substantial; for example at the Central Contra
Costa Sanitary District's (CCCSD) plant an hourly peaking factor of 2.4 has been measured.
The aeration system must accommodate these changes in loads to avoid  ammonia  bleeding
through during the peak load period. The  diurnal variations in load can be quite extreme; in
Figure 4-23 the  peak to minimum hourly loads are plotted against the flow peaking factor.
Ratios as high as 10:1 have been observed.
                                         4-76

-------
                                 FIGURE 4-22

      RELATION BETWEEN AMMONIA PEAKING AND HYDRAULIC PEAKING
     LOADS FOR TREATMENT PLANTS WITH NO IN-PROCESS EQUALIZATION
     2.5
 8,
 o
 o
 o
 -J
 c
 o
 E
 5
I
 o
o
o
-J
   c
   o
      I.O
                                                   T
                                               Y = 1.457 X -0.217
                                               r =0.823
                                               1 rr\Z/Sec = 43.8 mgd
                                      KEY
                                  Plant
 Sample
                              Lebanon, Ohio
                              Livermore, Ca
                              CCCSD, Co
                              Sacramento, Ca
                              Blue Plains, DC
                              Chapel  Hill, NC
                            Canberra, Australia-
                              Weston  Creek    Raw
                              Belconnen        Raw
                            	I
Primary
Roughing Filter
Primary
Primary
Primary
 Raw
ADWF,
 mgd
  \.2
  3.4
  2!
  45
 274
  1.8
                                                                    Ret.
84
85
63
86
87
30
                 1.0   88
                 10   88
                   (X)
        I.O                    1.5                   2.0
            Maximum Hourly  Flow, mgd/Average Daily Flow, mgd
                                                                     2.5
An early decision must be made during the design process as to what level of peaking of
oxygen demanding substances will be designed for. In addition to peaking of ammonia or
organic nitrogen, a concurrent peak may also occur in the loading of organic substances. If
very low levels of ammonia nitrogen are required at all times care must be used to develop a
statistical base whereby the frequency of peak oxygen loads can be identified. Not only
should daily peaks be identified, but possibly those occurring on weekly or monthly bases.
Table 4-18  presents an example of such an analysis for the primary effluent from two plants
in St. Louis, Missouri using COD as a measure of oxygen demanding substances since in that
                                    4-77

-------
particular instance nitrification was not required. As may be seen, significant departures
from average conditions occur on a fairly frequent basis. Similar analyses may be justified
when designing for nitrification.

The extra aeration capacity required for handling diurnal variations in nitrogen load, coupled
with the extra tankage and equipment required, may dictate in-plant flow equalization in
many instances. The reductions in capital and operating cost of aeration tankage  and
aeration facilities  must be compared with the cost  of flow equalization to determine
applicability to specific cases.  Design procedures for flow equalization are contained in
Chapter 3  of the Process Design Manual for Upgrading Existing Wastewater Treatment
Plants.25
                                    FIGURE 4-23
 ^               RELATIONSHIP OF MAXIMUM/MINIMUM NITROGEN
 £                  LOAD RATIO TO MAXIMUM/AVERAGE FLOWS
 *"  13
  o
 -j
•J
.o
c
o
6
6
 I
                                                   Y = 9.461 X- 8.28
                                                    r = 0.871
                                                 See  Fig.  4-22
                                                 for  Symbols
       I.O                     1.5                     2.0
             Maximum  Hourly Flow, mgd/Average Daily Flow.mgd
                                      4-78

-------
                                    TABLE 4-18

            PEAKING FACTORS VERSUS FREQUENCY OF OCCURRENCE
                  FOR PRIMARY TREATMENT PLANT EFFLUENT
•,.
Frequency of Occurrence
4 hours /day
4 hours /week.
4 hours/month-*
4 hours/3 months
4 hours/6 months
Bissel Point
Treatment Plant
COD load
peaking
factor b
1.30
1.72
2.02
2.25
2.40
Flow
peaking^
factor15
1.18
1.40
1.60
1.72
1.80
Lemay
Treatment Planta
COD load
peaking,
factor
1.40
1.85
2.35
2.62
2.80
Flow
peaking
factor D
1. 20 '
1.44
1.7.0
1.88
1.96
   Data is from reference  89; both plants serve the St. Louis area in Missouri and
   process about 100 mgd each.
   Peaking factor is defined in each case as the ratio of the 4 hour load listed to the
   average daily^load.
     4.8.1 Adaptability of Alternative Aeration Systems to Diurnal Variations in Load

Careful consideration should be given to maximizing oxygen utilization per unit power
input.  In the face of significant load variation, the aeration system should be designed to
match the load variation while economizing on power input. Obviously, designing the
aeration  system ft) provide for the maximum hourly demand 24 hours a day would provide
over aeration the majority of the time with wasteful losses of power.

The  available means  for aeration are summarized in  Section 5.3.4  of the Process Design
Manual for Upgrading  Existing Wastewater  Treatment Plants,  a Technology Transfer
publication.   Of the available aeration devices, the mechanical surface aerator is least well
suited to nitrification applications. This is because they are normally designed to operate at
fixed speed and^therefore must overaerate the majority  of the day to satisfy the peak
oxygen demands. Even when equipped with variable submergence of the blade, the units are
limited  to  matching  less than  a 2:1  variation in load, at best. Therefore, unless  flow
equalization is  provided  somewhere in the system,  mechanical surface aerators are not
capable of matching variations in  nitrogen loads without overaerating the mixed liquor
during a significant portion of the day.
                                        4-79

-------
Using diffused air aeration, air rates can be easily modulated to closely match the load,
merely by turning down or shutting off individual blowers. Thus, the diurnal load variations
can be matched without the necessity of over aerating the mixed liquor and wasting power.
Fine  bubble  diffusers  can  be  arranged across the  tank floor,90  allowing  fairly even
distribution of energy  input.  Gentler  mixing is provided  than with mechanical aeration
plants, providing less tendency for floe breakup.

Submerged turbine aeration systems are intermediate in terms of their responsiveness to the
problem of aeration in nitrification systems. Because of their capability to vary the air rate
to the sparger, they may be designed  to match the load variation in oxygen demand.  A
drawback, however, is that the impeller normally operates at fixed speed, imparting no turn
down capability for a significant part of the power draw. In fact,  in some impeller designs,
the power draw of the ungassed impeller is actually greater than when gas is fed to the unit.

     4.8.2 Oxygen Transfer Requirements

Oxygen  requirements for nitrification alone were  discussed in  Section  3.2.4. Oxygen
requirements in all practical cases are compounded by the oxygen  required for stabilization
of organics.

Reasonably exact  expressions for oxygen requirements for heterotrophic organisms and
nitrifiers  have been  developed.38 The approach, however,  requires pilot plant data to
provide COD balances and sludge yields. In general, this information is not available and a
simpler approach may be adopted.

In normal activated sludge treatment when nitrification is not required, the amount of
oxygen needed to oxidize the BOD5 can be calculated by the following equation:

                           B = X(BOD5)                                        (4-25)

where:      B   =   oxygen required for carbonaceous oxidation, mg/1
            X   =   a coefficient

The coefficient X relates to the amount of endogenous respiration taking place and  to the
type of waste being treated. For normal municipal wastewater, the X value would range
from  .5-.V for high  rate activated sludge  systems  to  1.5  for  extended aeration. For
conventional activated sludge systems X can be taken as 1.0.

In the case of nitrification, the oxygen  requirement for oxidizing ammonia must be  added
to the requirement for  BOD removal. The  coefficient for nitrogen to be oxidized can be
conservatively taken as 4.6 times the TKN content of the influent (Section 3.2.4) to obtain
the nitrogen oxygen demand (NOD) and the value of X in Equation 4-25 can be assumed to
be approximately 1.0. In actual fact, some of the influent nitrogen will be assimilated into

                                        4-80

-------
the biomass or is associated with  refractory organics and  will not be oxidized.  These
assumptions lead to the following oxygen requirement:


                                W = BOD5 + NOD                             (4-26)

where:      W   =   the total oxygen demand mg/1, and
         NOD   =   oxygen required to oxidize a unit of TKN taken as
                     4.6 times the TKN

Since  aeration devices  are  rated  using  tap water at standard  conditions,  the rated
performance of  the  aerator must be converted  to  actual  process  conditions by  the
application  of temperature corrections and  by factors designated  c*and |3 which relate
waste characteristics to tap water characteristics.

Temperature corrections are made by the relationship:

                      (T-20)
                 1 .024      where T = process temperature in degrees C.


The PC factor is the ratio of oxygen  transfer in  wastewater to  that  in tap water and is
represented by the following:

                        KT a (process conditions)
                  CX = -L -                              (4-27)
                        KT a (standard conditions)

Values  of * can  vary  widely in  industrial  waste  treatment  applications, but  for most
municipal plants,  it will range from 0.40 to 0.90.

The /3 factor is the ratio  of oxygen saturation in  waste to that in tap water at  the same
temperature. A value  of 0.95  is  commonly  used. Thus, the actual amount of oxygen
required to  be transferred (W)  can  be determined from the amount transferred under  test
conditions (WQ) by the equation:
                     W0                 -_                               (4_28)
where:      W   =   oxygen transferred at process conditions, Ib/day
            WQ  =   oxygen transferred at standard conditions
                     (T = 20 C, DO = .01 mg/1, tap water), Ib/day
            T    -   process temperature, C.
            C    =   oxygen saturation in water at temperature T, mg/1
             o
            C.   =   process dissolved oxygen level, mg/1
                                        4-81

-------
The  process dissolved oxygen level, Cj, must be set high enough to prevent inhibition of
nitrification rates (Section 3.2.5.5).  For this purpose,  a  minimum value of 2.0 mg/1 is
recommended. This value is also applicable under peak diurnal  load conditions, and  the
practice of allowing the DO to drop below 2.0 mg/1 under peak load is not recommended. If
the DO were to be allowed to drop below 2.0 mg/1 during peak load conditions, excessive
bleedthrough of ammonia could be expected.

Using example values for domestic sewage (
-------
                                FIGURE 4-24

            RELATIONSHIP OF AERATION AIR REQUIREMENTS FOR
             OXIDATION OF CARBONACEOUS BOD AND NITROGEN
DIFFUSER EFFICIENCY, PERCENT
RATED IN TAP WATER -STD CONDITIONS
^coSivlSoJcoc

v
N



Terr



\


iperat




X

ure 2




x

0 C





kg/m3/day = 62.4 Ib/



\




**X_
^^^




^^




	




1 	

1000




— —


C F/day










      500
               IOOO                       I5OO
AIR REQUIRED, CU FT/LB  BOD5 + NOD
I TOO
theoretical considerations for coarse bubble aeration systems. It should be noted that none
of the plants listed in Table 4-20 have automated blower control systems linked to DO
probes. When the Livermore, California plant staff manually adjusted the blower output
according to the DO probe reading, they were able to reduce the air requirement to an
average for the day of 680 CF/lb BODs + Ib NOD (42 m3/kg) based on hourly variations in
                              TABLE4-19
             RELATION OF OXYGEN TRANSFER EFFICIENCY TO
             AERATOR POWER EFFICIENCY (AFTER NOGAJ (92))
                 Diffuser oxygen
               transfer efficiency,
                    percent
            (at standard conditions)
                      Aerator power
                        efficiency
                      Ib 0_/bhp/hr
                          Lt
4
6
8
10
12
1.23
1.85
2.46
3.08
3.70
                                   4-83

-------
        Thus,  it is possible  that automated blower  operation  may reduce  aeration
requirements.

In addition to determining the total air requirement, attention should also be given to air
distribution within the aeration tanks. If the conventional (or plug flow) mode of operation
is established as the normal operating procedure, the air requirements will be greatest at the
head end of the aeration tanks (see Section 4.3.5) .

                                   TABLE 4-20

    AIR REQUIREMENTS FOR NITRIFICATION ACTIVATED SLUDGE PLANTS
Treatment
plant
Medford, Oregon
Flint, Michigan
Livermore, Calif.
Central Contra Costa
Sanitary District, Calif.
Jackson, Mich.
Hyperion Treatment Plant
West Battery,
Los Angeles, Ca.
Whittier Narrows ,
County Sanitation
Districts of Los Angeles
County, Ca.
San Pablo Sanitary District
Treatment Plant, Ca.
Configuration
Plug flow or
step aeration
Plug flow or
step aeration
Separate stage:
roughing bio-
filter followed
by nitrification
Separate stage:
lime followed
by nitrification
Plug flow
Plug flow
Step aeration
Plug flow;
roughing filter
followed by
nitrification
Diffuser type
Coarse
bubble
—
Coarse
bubble
Coarse
bubble
Coarse
bubble
Coarse
bubble
Coarse
bubble
Coarse
bubble
Oxygen demand
distribution, %
BOD
b
73
65
50
21
66
61
59b
59
NOD
27
35
50
79
34
39
41
41
cu ft/lb of BOD
and NOD3
1390
1280
1250
1700
910
1160
1180b
1410C
Data reference
93
6
5
3
42
2
9
94
  al cu ft/lb = .062 m3/kg

  Assuming primary effluent BOD5/COD ratio of 0.6

  CDuring June 1974 when nitrification run at design loads
     4.8.3 Example Sizing of Aeration Capacity

 As a  design example,  consider a  10 mgd  plant with lime  clarification in the primary
 sedimentation stage:
                                        4-84

-------
    Given the following primary effluent properties:

                  Design flow = 10 mgd
                  Average BOD 5 load = 4,170 Ib/day
                  Average TKN load = 2,100 Ib/day
                  Peak/average TKN load ratio = 1.8
                  Peak/average BODs load ratio = 1.5 (coincident with peak TKN load)

    Average load condition:

                  BOD5= 4,170 Ib/day
                  NOD = 4.6 x 2,100 = 9,660 Ib/day
                  BOD5 + NOD = 13,830 Ib/day

     Peak hourly load condition:

                  BOD5 = 4,170 x 1.5 = 6,260 Ib/day
                  NOD = 9,660 x 1.8 = 17,390 Ib/day
                  BOD5 + NOD = 23,550 Ib/day

     Ratio peak hour to average load: 1.7

     For the  purposes  of this example assume a fine bubble  diffused air aeration with a
     design figure of 725 cf/lb BOD5 + NOD.

           Average aeration requirement:

                  725 x 13,830/1,440 = 6,963 CFM

           Peak aeration requirement:

                  725 x 23,55O/1,440 = 11,860 CFM

In sizing plant air  requirements, separate provision must be made for preaeration, air in
mixed liquor and return sludge channels, and air requirements for aerated stabilization in
downstream denitrification processes.

4.9 pH Control

The implications of adverse operating pH and its causes have been discussed previously in
Sections 3.2.3 and  3.2.5.6. In cases where the alkalinity of the wastewater will be depleted
by the acid produced by nitrification, the natural alkalinity of the wastewater will have to
be  supplemented by chemical  addition. As discussed in Section  4.5.1, the effects  of
operation of  upstream  processes on alkalinity  and pH must be  considered. Alum or iron

                                       4-85

-------
addition tend to deplete wastewater alkalinity, whereas in some instances lime addition
increases the alkalinity.

     4.9.1 Chemical Addition and Dose Control

Two alternative chemicals, caustic (NaOH) and lime (CaO or Ca(OH)2) are in predominate
use for pH control. As lime is less expensive than caustic for the same change in alkalinity,
lime will generally be favored,  except in smaller plants. Procedures  for the  feeding and
storage of these  chemicals are  described in two EPA Technology Transfer publications,
Process  Design  Manual for  Phosphorus Removal^ and Process  Design Manual for
Suspended Solids Removal. 96

The need for pH adjustment may vary diurnally. In one case it was found that an alkalinity
deficit occurred  daily only for several hours at the peak nitrogen load condition. At other
times,  sufficient alkalinity was available. This condition caused a cyclic variation in pH. In
situations where diurnal  variation in the pH depression may be encountered,  continuous
on-line monitoring of pH for the purpose of controlling chemical addition seems justified. In
the case of suspended  growth systems operated in the plug flow mode, probes may  be
positioned at several points in the aeration basin, with provision  for addition of chemical at
several points.62 jn the case of attached growth sytems and suspended growth systems
operated in the complete mix  mode, effluent monitoring of pH  would be the usual choice
for controlling chemical addition to the influent.

     4.9.2 Effect of Aeration Method on Chemical Requirements

The type of oxygen transfer device chosen can have a marked effect on the chemical dose
required for pH control. As an  example, the differences between coarse bubble and fine
bubble air diffusion systems  will be examined. The following equation from bicarbonate-
carbonic acid equilibrium is useful for estimating the pH level in aeration tanks:

                                                 CO)

                                                                               (3'9)
At 20 C, the value of pKj is approximately 6.38. In using the equation, the alkalinity in the
aeration  tank  can be used to estimate the bicarbonate level (HCC>3) and the value  used
should reflect any alkalinity depletion resulting  from nitrification.  The level of H2CO3
(carbon dioxide in solution) can be estimated from Henry's Law as follows:

                           Ceq = H Pgas                                       (4-30)

where:      C    =   concentration of gas dissolved in liquid at equilibrium, mg/1,
             eq
            H    =   Henry's Law constant, mg/1/atmosphere

            P    =   partial pressure of gas in equilibrium with the liquid,
                     atmosphere
                                        4-86

-------
The Henry's law constant for carbon dioxide at 20 C is 1688 mg/1/atmosphere. To establish
Ceq for  use  in Equation 3-9,  the level of carbon dioxide  in the  aeration air must be
estimated.  In unpolluted atmospheric air,  the partial  pressure of  carbon dioxide  is only
about 0.0003 atm. However in  aeration air, carbon dioxide is generated from the oxidation
of organics and from nitrification. For the mix of organics in  municipal wastewaters, about
one mole of carbon  dioxide is produced  per mole  of oxygen consumed. Similarly for
nitrification,  about one mole of carbon dioxide is produced per mole of oxygen consumed
(c.f. Section  3.2.2). On a weight  basis, for every Ib of oxygen consumed about 1.38 Ib of
carbon dioxide is produced.  Equations 3-9  and 4-30  can  be solved simultaneously to
determine the amount of dissolved carbon dioxide in the liquid and the amount present in
the gas as well as the process operating pH.

The following example is illustrative of the procedure.  Assume a residual alkalinity of 150
mg/1 as CaCC>3 (3 x 10"3 moles/liter  of HCO~3) and an  oxygen transfer efficiency of 12
percent, under standard conditions. Equation 4-29 would indicate that 725 cu ft/lb BOD5 +
NOD is required.  Assume also that the example conditions given in Section 4.8.3 are used,
namely that 13,830 Ib of oxygen demand are contained in 10 million gallons of wastewater;
this allows  the  calculation that  1 Ib of  oxygen  demand  is  contained in  725  gallons.
Therefore the aeration rate is  1  cu  ft/gal  of wastewater nitrified  (725 cu ft/125 gal).
Assuming a pH of 7.1 (by trial  and error), Equation 3-9 allows calculation of the dissolved
carbon dioxide concentration, and for this case it is found to be 25 mg/1. Therefore at this
concentration, 725 gallons contain 0.15 Ib  of carbon  dioxide (CO2) of the total evolved
(1.38 Ib  of CO2 evolved/lb of oxygen). Equation 4-30 allows calculation of  the partial
pressure  of CO2  in the off gases; the calculated  result is 0.0149 atm. This concentration,
plus the  volume of gas required allows calculation that the gas contains 1.23 Ib of carbon
dioxide. The total is then 1.38 Ib which checks with the expected result. If the total had not
been 1.38 Ib, a new pH value level would be assumed and the procedure tried again until a
balance is obtained.

Table 4-21 presents the results  for a number of oxygen transfer efficiency values, using the
procedure outlined above. The  trends  shown are valid for municipal wastewaters, although
the absolute value of the pH may differ slightly from those indicated due to variation from
the assumed conditions.

Examining Table 4-21, it can be seen that fine  bubble  diffuser systems, at  high  oxygen
transfer efficiency, will operate at lower operating pH levels than  coarse bubble diffusers
operated  at  lower oxygen transfer  efficiencies.  If  the  same  operating  pH is  to  be
maintained,  this  will  translate into higher chemical doses for fine bubble aeration systems.
For instance, Table 4-21  shows that if the residual alkalinity is 100 mg/1, a pH of 7.0 can be
maintained  with  a coarse  bubble aeration system rated at 9 percent efficiency under
standard  conditions.  For a comparable fine bubble  system rated  at  18 percent under
standard  conditions, the residual alkalinity would have to be raised to  150 mg/1 as CaCO3 to
maintain the pH at 7.0. This would require an extra lime dose of at least 28 mg/1 as CaO.
                                        4-87

-------
                                   TABLE 4-21

               EFFECT OF OXYGEN TRANSFER EFFICIENCY AND
                   RESIDUAL ALKALINITY ON OPERATING pH
      Residual
          pH at stated
Operating transfer efficiency,  percent3
as CaCO , mg/1
O
50
75
100
125
150
175
200
Coarse bubble range
6
6.9
7.1
7.2
7.3
7.4
7.4
7.5
9
6.7
6.9
7.0
7.1
7.2
7.3
7.3
Fine bubble range
12
6.6
6.8
6.9
7.0
7.1
7.2
7.2
18
6.5
6.7
6.8
6.9
7.0
7.0
7.1
        at standard conditions
As an example of the use of Table 4-21, presume that it is desired to maintain the operating
pH  level at 7.2. Assume that an  aeration  system has been  chosen  which has an oxygen
transfer efficiency, eo, of 12 percent and residual alkalinity of 75 mg/1 as CaCO3- If it is
desired to maintain the process pH at 7.2 to prevent inhibition of nitrification rates, Table
4-21 indicates that a  residual alkalinity of 175 mg/1 is required. The difference between the
available alkalinity and the required alkalinity is  100 mg/1 as CaCO3 and this could be
supplied by a lime dose of 56 mg/1  (as CaO).

In plug flow systems, the pH will steadily decline from the influent end to the effluent end,
following the similar  decline in alkalinity occurring due to nitrification. The most severe pH
depression will be at the effluent end,  after the bulk  of the nitrogen has been oxidized.
Complete  mix systems, on the other hand, will be uniformly depressed in pH throughout
the  tank because of  the uniformity of aeration tank contents. This is a disadvantage for
nitrification with complete mix activated sludge compared to plug  flow as the pH of the
entire complete mix  tank will be  the same as the pH at the  effluent end of the plug flow
tank at the same oxygen transfer efficiency.

The preceding discussions about the effect  of carbon dioxide evolution in operating pH in
the  aeration tank are not applicable to cases where nitrification follows lime  treatment. In
                                       4-88

-------
these cases, the carbon dioxide that is evolved is used in recarbonation reactions and very
little enters the aeration air (c.f. Section 4.5.1).

4.10 Solids-Liquid Separation

In all suspended growth systems and in most attached growth systems, the nitrification stage
must be followed with a solids removal stage. Because of the complexity of the solids-liquid
separation problem, full consideration cannot be given to it within the scope of this manual.
Rather, a brief review of the problem is given with reference to the pertinent literature.

In some attached growth nitrification systems, multimedia filtration is provided following
the nitrification  stage as the level of effluent solids is low. Design criteria for multimedia
filtration are presented in the EPA Technology Transfer Publication, Waste-water Filtration,
Design  Considerations.   General design  criteria  for secondary sedimentation as well as
multimedia filtration applicable to nitrification processes are presented in the Process Design
Manual for Upgrading Existing Wastewater  Treatment Plants. 25

In suspended growth  systems, there is a strong interrelationship between operation of the
secondary clarifier and operation  of the  aeration basin. The degree to which the return
sludge can be thickened will affect the allowable mixed  liquor in the aeration tank  and
therefore the  size  of the  aeration  tank.  Dick  and  coworkers  have  presented design
relationships which are useful for analyzing clarifier-aeration tank interactions.98,99,100

Consideration of several factors is required when designing secondary sedimentation tanks.
Mechanical design of the tank is very important. Inlet turbulence must not upset the tank;
an energy dissipating  inlet well serves the purposes of distributing the flow equally across
the  tank and of providing for some mild turbulence to help aggregate  the finely divided
solids into floe and increase  the separation efficiency. Effluent launders must be sized and
placed so that currents are not created that will upset the tank and cause short-circuiting of
the solids to the effluent. The tank must also have sufficient depth to allow a sludge blanket
to form under  all conditions, especially  under those  conditions  which occur when the
system  operation is limited by the ability to obtain  a  specific return activated sludge
concentration. Overflow rates must not be so great that sludge is suspended in the "upflow"
and  carried over the weirs.  Lastly, the sludge must be quickly removed from the tank to
minimize the ocurrence and duration of anoxic (zero DO) conditions.

The  nature  of  the biological  solids  developed  in  suspended growth  systems plays an
important role in the  operation and design of the secondary sedimentation tank. First, the
ability of the mixed liquor to  be clarified  will affect the size of the required sedimentation
tank. This property is dependent  on  the  extent of disperson of the biological solids.  If a
large proportion  of the biological solids are finely divided, the separation efficiency of these
solids will be poor. If,  however, only small  portions of the biological solids are dispersed and
the bulk is incorporated into floe, the separation efficiency of the mixed liquor should be
high. Consolidation  of the small dispersed particles into the large  floe to improve effluent

                                         4-89

-------
clarity can be encouraged by a number of techniques. Chemical coagulation of the dispersed
solids has been successfully employed to enhance the flocculation of biological solids J01
Physical conditioning of the activated  sludge floe prior to secondary  sedimentation is
another technique which can be used with or without chemicals. It has been found that by
incorporating mild turbulence beweeen the oxidation tank and the secondary sedimentation
tank, effluent  clarification can  be enhanced. ^2 Mild  turbulence  can  be conveniently
accomplished by using mild  aeration in transfer channels  and by incorporating energy-
dissipating inlet wells in circular secondary sedimentation tanks.

The thickening qualities of the sludge will determine the required area for sludge thickening
in the tank, depending  on the desired or optimal MLSS or Return Activated Sludge (RAS)
levels. The limiting design consideration for thickening is usually at peak wet weather flow
(PWWF), for it is under these conditions that the solids loading on the secondary sedimenta-
tion tanks are the greatest.  Secondary sedimentation tanks must be sized such that mixed
liquor solids are  not lost in the effluent during PWWF  conditions.  Polymer feed facilities
are appropriate to allow short-term polymer dosing during wet weak peak load conditions to
enhance the thickening qualities of the sludge.

Thickening  qualities will decrease and sedimentation tank area requirements  will increase
when  wastewater temperatures  decline.  Figures  4-25  and 4-26  show  the  effect of
temperature on three different oxygen activated sludges and the temperature trends would
be  expected to  be similar for air activated  sludges. In sedimentation  tank design, solids
loadings should reflect the minimum wastewater temperatures expected. When temperatures
decline the mixed liquor levels that can be maintained under aeration also decline due to the
decreased level of thickening that can be obtained in the  sedimentation tank.

Rising sludge caused by denitrification in secondary  clarifiers has  occasionally plagued
nitrification operations.  Denitrification occurs because the  organisms in the biological sludge
in the secondary clarifier  can utilize nitrate and nitrite  as  electron acceptors to oxidize
organic compounds in the sludge layer. The formation of bubbles from nife-ogen gas evolved
in denitrification and other gasses, such as carbon dioxide, can then cause flotation. Early
experimental work by Sawyer and Bradney^4 an(} subsequent experimental  and theoretical
work by Clayfield 105 have identified the important parameters affecting denitrification. A
factor of foremost importance is the presence of adequate quantities of nitrites or nitrates in
the wastewater to cause bubble formation. In essence, the dissolved gases in  the wastewater
(nitrogen, carbon dioxide, and oxygen)  must be in sufficient concentration so that the sum
of  their partial  pressures equals or exceeds the  ambient liquid pressure.  In  the case of
Sawyer and  Bradney's  work, as little  as  4-6 mg/1 of  nitrate-N or  nitrfJTe-N would cause
flotation in a graduated cylinder,  while Clayfield found that  16-18 mg/1  nitrate-N  were
needed to cause  flotation in full scale sedimentation tanks. Differences between investiga-
tions are in part due to the greater liquid pressures in deep sedimentation tanks compared to
graduated cylinders and may  also reflect differing levels  of concentration  of other gases,
such as carbon dioxide.
                                        4-90

-------
The degree of stabilization of the sludge also has a profound effect on denitrification. It
has been shown that sludges incorporating unoxidized feed organics float more readily than
well  oxidized sludges. 104 Temperature  is  also important  as  it  affects the  rate of
denitrification and therefore affects the rate of gas and bubble formation. 104, 105

                                 FIGURE 4-25
            EFFECT OF TEMPERATURE ON THICKENING PROPERTIES
              OF OXYGEN ACTIVATED SLUDGE AT MLSS = 4000 mg/1
                          (REFERENCES 86 AND 103)
    30


    28


6   26

O   24


•*.   22
O
CO
CO   20
5
«.
O
          IB
          16
     £   *
     -J
     Uj   /O
     to
     5
     -J
     K
     K-
     kl
     CO
     LU
     S
     o
     N
     2 —
                  WASHINGTON, D.C
                                A
                                                          T
                     CITY "A"
                                              SACRAMENTO, CA.
                                         KEY
0 SACRAMENTO,CA.
& WASHINGTON, D.C.
Q CITY  "A"
                    IO        15        2O       25
                          TEMPERATURE,  C
                                                    3O
                                    4-91

-------
      Oi
      E

     I
      CO
      CO
      -J
     O
     O
     -J
     Uj
     Uj
     CO

     Uj

     O
     N
                                FIGURE 4-26


           EFFECT OF TEMPERATURE ON THICKENING PROPERTIES
            OF OXYGEN ACTIVATED SLUDGE AT MLSS = 7000 mg/1
                         (REFERENCES 86 AND 103)
© SACRAMENTO, CA.
   WASHINGTON, O.C.

GJ CITY  "A"
       WASHINGTON, D.C
                                                 SACRAMENTO,CA
                     10        15        20       25

                           TEMPERATURE,  C
The following conclusions can be drawn about denitrification in secondary clarifiers, based
on work done to date;33,104,105 (j) Rapid sludge removal can prevent sufficient time
being available for nitrogen bubble formation; (2) sludges with low SVI values are preferable
as they can be withdrawn faster; (3) since the saturation level of nitrogen is greater in deep
tanks than laboratory cylinders, bubbles will form and sludges  will float  faster in the
                                     4-92

-------
laboratory than in the field; (4) there is a minimum concentration of nitrate nitrogen below
which there is insufficient nitrogen to cause flotation. In weak wastewaters or for those
plants in  which nitrification is  suppressed, sludge flotation  will not occur; (5) a  drop in
temperature will reduce denitrification rates and may render rising sludge a problem only
under warm conditions; (6) an equivalent amount of nitrite will produce flotation  faster
than  nitrate,  because denitrification  is  more rapid when  nitrite serves as the electron
acceptor rather than nitrate; (7) a sludge with low activity or low rate of denitrification
should be less  susceptible to  flotation.  Sludge activity will vary  among the types  of
suspended growth  processes.  For  instance,  separate stage  nitrification sludge is less
susceptible to flotation than combined carbon oxidation-nitrification sludges due to the low
organic loading on the former process and resultant lower activity with respect to carbon
oxidation and  denitrification in sedimentation tanks. Among combined carbon oxidation-
nitrification alternatives, the contact stabilization and  step  aeration alternatives are most
susceptible to  sludge flotation due to the enhanced possibility in those modifications  of
influent organics being incorporated into the mixed liquor without being oxidized prior to
the clarification step. To a lesser degree, complete mix systems are also prone to sludge
flotation for the same reason.

Control measures for preventing floating sludge should be incorporated into the initial plant
design. Provision of rapid  sludge removal (vacuum pickup type) in sedimentation tank
design can prevent there being sufficient contact time  for bubble formation to occur and
cause  flotation.  Flexibility  in  influent  feed  points (e.g., allowing switching from step
aeration to plug flow in warm weather periods) can provide the operator with options in
process operation that allow him to get out of difficult operating situations. Provision  for
chlorination  of the return  activated sludge is  recommended for all suspended  growth
applications.  Recent work  done  in California has shown  that  continuous low dose
chlorination can be used for controlling sludge bulking  and reducing the sludge  volume
index, apparently without impairing nitrification (see Section 4.5.1).^9 This allows more
rapid sludge withdrawal from the sedimentation tanks. When nitrification is not required,
higher doses of chlorine can be  used to suppress nitrification  and thus avoid flotation.
Overdosing chlorine on a slug dose or continuous dose basis should be avoided, however, as
it can cause increases in the level of organics in the process effluent. 104

Control of solids retention time to values below that which will support nitrification has
been  a procedure that has been occasionally recommended for cases where nitrification is
not required and sludge  flotation is to be prevented. However, when the DO level or pH is
not limiting the nitrification rate, nitrification will proceed at solids detention times as low
as 1.30 days at temperatures equal to or greater than 20 C (Section 3.2.5.4). This renders
the control of nitrification impractical with solids retention time control in warm weather as
stable operation at such a low value for  the solids retention  time would not be possible. It
may be possible  to  suppress nitrification through control of DO to levels < 0.5 mg/1 and
thereby limit the nitrifier growth rate to levels  which result in washout of the nitrifying
organisms, but  this will  require  accurate  around-the-clock  control  of DO. Further,
maintenance of low DO can cause another operational problem, sludge bulking.

                                        4-93

-------
4.11 Considerations for Process Selection

In selecting nitrification as the process for ammonia removal, two kinds of comparisons can
be made. First, the process can be compared to the physical-chemical alternatives. Second,
alternative  nitrification  processes  can be compared. It is emphasized  that no  single
alternative  will be the best choice for all situations.

     4.11.1 Comparison to Physical-Chemical Alternatives

Several factors  dictate the choice between biological and  physical-chemical techniques for
ammonia removal. Cost  is often  the single  most influential factor in  process choice.
Ammonia removal via the nitrification process is  widely  recognized to be the least costly
ammonia removal alternative. Unless phosphorus  removal is also required, the combined
cost of lime  precipitation-air stripping is normally greater than the cost for nitrification.
Likewise,   breakpoint  chlorination  and ion  exchange  are normally more costly  than
nitrification. 106

In the majority of  situations existing facilities are utilized when treatment is upgraded,
rather than construction of wholly new facilities. The layout of the existing facility may be
more adaptable to one specific alternative or another. In many instances, it has been found
that biological nitrification has been the process most easily incorporated into the upgraded
treatment system.

Very low temperature operation ( < 10 C) may favor a  physical-chemical process rather
than a biological process as reaction rates become very low, requiring very large reactors.
Physical chemical processes are also affected by low temperatures, but to a lesser degree.

The presence of compounds  toxic  to nitrifiers may also dictate against the choice of
nitrification.  Some toxicants are resistant (e.g., nonbiodegradable solvents) to most forms of
pretreatment. Unless a very effective source control program can  be implemented for these
compounds, dependable operation of nitrification may become impractical.

     4.11.2 Choice Among Alternative Nitrification Systems

All of the factors described in Section 4.11.1 are also factors to be considered in selection
among nitrification systems. Other factors affecting choice among nitrification alternatives
are summarized in Table 4-22 as a guide for process selection.  Each of  these factors is
considered  earlier in this chapter.

Higher effluent ammonia (1-3 mg/1) in the attached growth effluents than suspended growth
effluent is  cited  as  a disadvantage of attached growth systems  in Table  4-22. However,
breakpoint chlorination' is easily appended  to attached growth systems, as the chlorine dose
for breakpoint is low ( < 30 mg/1). The addition of breakpoint chlorination puts attached
growth systems on an equal footing with suspended growth  systems with respect to ammonia
control.
                                        4-94

-------
                                        TABLE 4-22
                  COMPARISON OF NITRIFICATION ALTERNATIVES
      System Type
          Advantages
           Disadvantages
    Combined carbon
  oxidation - nitrification
    Suspended growth
     Attached growth
Combined treatment of carbon and
 ammonia in a single stage
Very low effluent ammonia
 possible
Inventory control of mixed liquor
 stable due  to high BOD /TKN
 ratio

Combined treatment of carbon and
 ammonia in a single stage

Stability not linked to secondary
 clarlfler as organisms on media
No protection against toxicants
Only moderate stability of operation
Stability linked to operation of
 secondary clarifier for biomass return
Large reactors required in cold weather

No protection against toxicants
Only moderate stability of
 operation
Effluent ammonia normally 1-3 mg/1
 (except RBD)
Cold weather operation impractical
 in most cases
 Separate stage
  nitrification

 Suspended growth
 Attached growth
Good protection against most
 toxicants
Stable operation
Very low effluent ammonia possible
Good protection against most
 toxicants
Stable operation
Less sensitive to low temperatures
Stability not linked to secondary
 clarifier as organisms on media
Sludge inventory requires careful
 control when low BOD.AKN ratio

Stability of operation linked to operation
 of secondary clarifier for biomass return
Greater number of unit  processes required
 than for combined carbon oxidation -
 nitrification

Effluent ammonia normally 1-3 mg/1
Greater number of unit  processes
 required than for combined
 carbon oxidation - nitrification
Refinement of  process choice may  require pilot studies. This  is  particularly true where
wastewater toxicity may affect the efficacy of nitrification (see Section 4.5.3).

A common issue faced by the engineer when dealing with suspended growth system design is
whether to  separate the carbon oxidation stage from the nitrification stage or whether to
provide  a combined carbon oxidation-nitrification system. In  a  recently conducted  pilot
study using  these  systems in parallel  at two wastewater temperatures, 8 C and 20C, for the
town of Cheektowaga,  New York, it was shown that when the separate nitrification  stage
system  was  operated  at  the  same  solids  retention  time  as  the  combined  carbon
oxidation-nitrification system and the same temperature, nitrification effluents of essentially
identical quality were produced. 1 "7
                                             4-95

-------
The investigators concluded that in most cases the combined carbon-oxidation nitrification
system should be chosen for the  following reasons: 107

1.   Use of the biological solids retention time concept and controlled sludge wasting make
     the combined carbon  oxidation-nitrification system  as controllable as a two stage
     suspended  growth system. The  investigators did  not agree with  the often stated
     concept  that  separating the  stages leads  to more positive  control of  the  carbon
     oxidation and nitrification  functions, as their experimental study demonstrated quite
     the opposite.

2.   The use of combined carbon oxidation-nitrification results in lower sludge quantities to
     be wasted than in  a two stage suspended growth system. This is because the first stage
     (carbon oxidation) operates at a low solids retention time (say 8 = 2 days) which
                                                                    L>
     results in less solids  destruction than when 10 or 20 days are used  in the combined
     carbon  oxidation-nitrification system.  This phenomenon has also been observed by
     others. 108,109

3.   The longer sludge retention times employed in separate stage nitrification systems ( 8
                                                                                    c
     = 10 to 20 days) results in  improved sludge settling characteristics as compared to high
     rate activated sludge systems (at 6 —,2 days).
                                    C

4.   A two stage suspended growth system appears to be more prone to control problems
     relating to sludge inventory control. The separate stage reactor's sludge inventory must
     be maintained  by shifting inventory from the first stage, or by  some other means
     (Section 4.5.2).  Further, two sets of sedimentation tanks are  required. Sedimentation
     tanks are often the most vulnerable components of the activated sludge system. "Thus
     it  is  difficult to  envision that  the  path  to increased controllability of nitrifying
     activated sludge should lead  to  a doubling of the least stable element in  the process
     configuration, i.e.;  the clarifier. Rather, if a two stage nitrification system is required, it
     appears more reasonable  to  explore  the  capabilities  of a fixed film nitrification
     reactor. . ."107

5.   Toxicants affecting nitrifiers present in the raw sewage or primary effluent are often
     cited  as a reason to provide a high rate activated sludge effluent to  act as a toxicant
     removal step ahead of a separate stage nitrification unit. First, a  detailed evaluation
     may  show that in fact  toxicity  is not a problem. "Secondly,  toxic materials might
     better be excluded from  wastewater systems by regulation rather  than relying on
     'sacrificial' biosystems to protect the nitrifying capability of the system. Thirdly, in
     most  cases it may become attractive to remove phosphorus in primary treatment by
     the addition of coagulants." 107

These conclusions are presented  in this manual as an excellent basis for consideration of the
reasons for process selection. In  many cases effective counter arguments can  be presented.
In the last analysis, the  process choice must be made by the local agency and its engineering

                                         4-96

-------
consultant or staff. As an  example, counter-arguments are listed in the same order as the
arguments previously presented:

1.   The  parallel  study of carbon  oxidation-nitrification and two sludge  systems at
     Cheektowaga was based primarily on municipal sewage, with only 10 percent industrial
     load. 107 Situations  do exist \vhere significant industrial contributors of organic load
     are  tributary to the  municipal system. For instance, seasonal  canning  industries
     tributary to California municipalities  treatment plants in some instances cause  3 to 4
     times  the non-canning season organic load at the peak of the canning season. Further,
     industrial waste production may vary from year to year depending on factors beyond
     control or accurate  prediction. In  the face of this unpredictable variation it may be
     difficult or uneconomic to design a system for combined carbon oxidation-nitrification
     due to the  sensitivity of nitrification  to solids retention time. Less difficulty  is
     experienced  in  designing for high rate  activated sludge in a two stage system, as
     production of effluent of a quality suitable for separate stage  nitrification is  not as
     sensitive to load or solids retention time as is nitrification.

2.   The phenomenon of  lower sludge production  from a two stage  suspended growth
     system compared  to  a  combined  carbon  oxidation-nitrification  system is well
     established and  cannot be  contested.   However,  from an  energy  standpoint, the
     greater sludge production  from the two  stage system may be an  advantage for two
     reasons. First,  the  lower  sludge  production of the  combined  carbon  oxidation-
     nitrification system is  obtained at the expense of greater power requirements because
     greater amounts of  oxygen must be  supplied for the oxidation of solids. Second, if
     anaerobic digestion is employed and the gas recovered for useful energy purposes, less
     energy is available from the digestion process when less solids are produced. However,
     these  two  factors may be outweighed by the increased cost  of ultimate disposal in
     some cases.

3.   It has occasionally been found that the longer solids retention times (10 to 20 days)
     also result in sludge settling difficulties.

4.   With careful monitoring, the separate nitrification stage's inventory requirements can
     be managed.  In some situations where industries are tributary, the inventory control
     problem may be easier with a two stage system (see counter-argument 1 above).

     Two sets of sedimentation tanks do present  more control requirements than one set.
     However, if the conditions for tank upset are present when two sets are provided, they
     also can be present for one set. Careful  control  of sedimentation tank operation  is
     mandatory in either case.

5.   In the case of the parallel study at Cheektowaga,  there  was no evidence of significant
     toxicants in  the primarily domestic sewage processed. 107 So perhaps it is natural to
     discount the  difficulty in dealing with this problem.

                                       4-97

-------
     The unfortunate aspect of the  presence of toxicity in wastewater is that it is often
     ephemeral in nature; i.e., it's there and  then  it's gone. Toxicants discharged on  a
     continuous basis  are  handled with relative  ease compared  to the occasional dump.
     Unless the problem is recognized in its earliest stage, the causative agent may not even
     be sampled by  plant  personnel, rendering it impossible to trace. Even if a sample is
     caught,  tracing it back through  the  system is not always possible until the next
     occurrence. Another aspect of the problem is that the dumps offer no opportunity for
     the biomass  to adapt to  the  toxicant,  whereas if it  were continuously  present,
     adaptation of the nitrifiers might be possible (see Section 3.2.9).

     In "bases where toxicants are occasionally present, the issue boils down to the need for
     plant reliability. In cases of discharge to an estuary or groundwater where mixing in the
     environment  causes dilution of  the  effluent,  occasional process failures may be
     accepted. However, where stringent regulatory requirements exist or where the water is
     reused and the  water user  demands the consistent performance expected of a water
     utility, some compensation must be made  to handle the problem of toxic upsets. This
     may be  done by  any number of means, one of which is to  provide pretreatment via
     chemical addition or by a  biological  treatment stage (Section 4.5.3).  Another is to
     provide supplemental breakpoint chlorination at the end of the system. The cost of the
     latter facility is very much affected by the degree of upset in nitrification expected.

In the  last analysis, the parallel study at Cheektowaga showed that combined  carbon
oxidation-nitrification  could be  just as reliable as  separate  stage nitrification at low
temperatures (8 C) with a primarily domestic  wastewater.107 The study provides further
proof  dispelling the  poor reputation that  the  combined  carbon oxidation-nitrification
system has acquired in the U.S. Its wide application  in England where it is coupled with a
toxicity source control program  offers additional testimony to the efficacy of the process.

Similar lengthy discussions could be prepared  which present  the pros and  cons of other
systems: e.g.,  rotating  biological  discs  versus  plastic  media trickling  filters. These
comparisons not only are beyond the scope of this manual but would not have general
validity. It is a hazardous task  to make  such an attempt as the specifics of individual
circumstances affect  the decisions to a large degree. There is no universally best nitrification
approach. Rather, the broad variety of alternatives should be viewed as a positive situation.
The  fact that there are many alternatives makes the task of adapting nitrification into waste
treatment easier, not harder. A myriad of flowsheets incorporating nitrification are not only
possible, but economically feasible and with proper design and operation, quite reliable.
                                        4-98

-------
4.12 References

  1.  Mulbarger, M.C., The  Three Sludge System for Nitrogen and Phosphorus Removal.
     Presented at the 44th Annual Conference of the Water Pollution Control Federation,
     San Francisco, California, October, 1971.

 2.  City of  Los Angeles, California, Hyperion  Treatment Plant West Battery Operating
     Reports. January through March, 1969.

 3.  Horstkotte, G.A., Niles, D.G., Parker, D.S., and D.H. Caldwell, Full-Scale Testing of a
     Water Reclamation System. JWPCF, 46, No.  1, pp 181-197 (1974).

 4.  Weddle,  C.L.,  Niles, D.G.,  Goldman, E.,  and  J.W.  Porter,  Studies  of Municipal
     Wastewater Renovation for Industrial Water. Presented at the 44th Annual Conference
     of the Water Pollution Control Federation, San Francisco, California, October, 1971.

 5.  Loftin, W.E., Annual Report,  Livermore Water Reclamation Plant, 1970.  City of
     Livermore, California, March, 1971.

 6.  Beckman, W.J.,  Avendt, R.J., Mulligan, T.J., and G.J. Kehrberger, Combined Carbon
     Oxidation-Nitrification. JWPCF, 44, No. 10, pp 1916-1931 (1972).

 7.  Stamberg,  J.B., Hais, A.B., Bishop,  D.F., and J.A. Heidman, Nitrification in Oxygen
     Activated Sludge. Unpublished paper, Environmental Protection Agency, 1974.

 8.  Heidman, J.A., An Experimental Evaluation of Oxygen and Air Activated Sludge
     Nitrification Systems With and Without pH Control. EPA report for Contract No.
     68-03-0349, 1975.

 9.  County  Sanitation Districts  of Los Angeles County, Monthly  Operating Reports,
     Whittier Narrows Water Reclamation Plant. April, 1973 to March,  1974.

10.  Greene,  R.A., Complete Nitrification by  Single  Stage Activated  Sludge. Presented at
     the 46th Annual Conference of the Water  Pollution Control  Federation, Cleveland,
     Ohio, October, 1973.

11.  Newton, D., and T.E. Wilson, Oxygen Nitrification Process at Tampa. In Applications
     of Commercial Oxygen to Water and Wastewater Systems, Ed. by  R.E. Speece and J.F.
     Malena, Jr., Austin, Texas: The Center for Research in Water Resources, 1973.

12.  Tenney,  M.W., and W.F. Echelberger, Removal of Organic and Eutrophying Pollutants
     by Chemical-Biological Treatment.  Prepared for the EPA, Report No. R2-72-076
     (NTISPB-214628), April, 1972.
                                       4-99

-------
 13.  Black,  S.A.,  Lime  Treatment for Phosphorus Removal at  the New Market/East
     Gwillimbury WPCF.  Paper No. W3032, Ontario Ministry of the Environment, Research
     Branch, May,  1972.

14.  Schwer, A.D., Letter communication to D.S. Parker — Metropolitan Sewer District of
     Greater Cincinnati, March 9, 1971.

15.  Barth, E.F., Brenner, R.C., and R.F. Lewis, Chemical-Biological Control of Nitrogen in
     Wastewater Effluent.  JWPCF, 40, No. 12, pp 2040 - 2054 (1968).

16.  Rimer, A.E.> and R.L. Woodward, Two Stage Activated Sludge Pilot Plant Operations,
     Fitchburg, Massachusetts. JWPCF, 44, No. 1, pp 101-116 (1972).

17.  Wild, H.E.,  Jr.,  Letter communication to D.S.  Parker. Briley, Wild and Associates,
     Ormond Beach, Florida, September 9, 1974.

18.  Union Carbide Corporation, "UNOX" System Study at Town ofAmherst, New York.
     1972.

19.  Linstedt, K.D.,  and  E.R. Bennett, Evaluation of Treatment for Urban Wastewater
     Reuse.  Report prepared for the Environmental  Protection Agency, EPA-R2-73-122,
     July, 1973.

20.  Linstedt, K.D., Letter communication to D.S. Parker. University of Colorado, Boulder,
     Colorado, August 5, 1974.

21.  Stenquist,  R.J., Parker,  D.S.,  and T.J.  Dosh, Carbon  Oxidation-Nitrification  In
     Synthetic Media  Trickling Filters. JWPCF, 46, No. 10, pp 2327-2339 (1974).

22.  Duddles, G.A., Richardson, S.E., and E.F. Barth, Plastic Medium Trickling Filters for
     Biological Nitrogen Control. JWPCF, 46, No. 5, pp 937-946 (1974).

23.  McHarness,  D.D., Haug, R.T.,  and P.L. McCarty, Field Studies of Nitrification with
     Submerged Filters.  JWPCF, 47, No. 2, pp 291-309 (1975).

24.  McHarness,  D.D.,  and P.L. McCarty, Field Study of Nitrification with Submerged
     Filter. Report prepared for the EPA, EPA-R2-73-158, February, 1973.

25.  Process Design Manual for Upgrading Existing Wastewater Treatment Plants. U.S. EPA,
     Office of Technology Transfer, Washington, D.C. (1974).

26.  Richardson, S.E., Pilot Plants Define Parameters  for  Plastic  Media Trickling Filter
     Nitrification. Presented at the 46th Annual Conference of the Water Pollution Control
     Federation, Cleveland, Ohio, October, 1973.

                                       4-100

-------
27.  Sampayo, Felix F., The Use of Nitrification Towers at Lima, Ohio. Presented at the
     Second Annual Conference Water Management Association of Ohio, Columbus, Ohio,
     October, 1973.

28.  Brenner, R.C., EPA Experiences in Oxygen Activated Sludge. Prepared for the EPA
     Technology Transfer Program, October, 1973.

29.  Lawrence, A.W., and P.L. McCarty, Unified Basis for Biological Treatment Design and
     Operation. JSED, Proc. ASCE, 96, No. SA3, pp 757-778 (1970).

30.  Hanson, R.L.,  Walker,  W.C., and J.C.  Brown,  Variations in  Characteristics  of
     Wastewater Influent at the Mason Farm Wastewater Treatment Plant, Chapel Hill, No.
     Carolina.  Report  No.  13,  UNC  Wastewater  Research Center,  Chapel Hill, N.C.,
     February, 1971.

31.  Lijklema, L., A Model for Nitrification  in the  Activated Sludge  Process.  ESE
     Publication  No.  303,  Department of Environmental Sciences and  Engineering,
     University of North Carolina, June, 1972.

32.  Poduska, R.A. and J.F. Andrews,  Dynamics of Nitrification in the Activated Sludge
     Process.  Presented at the  29th  Industrial Waste  Conference,  Purdue University,
     Lafayette, Indiana, May 7-9, 1974.

33.  Nagel, C.A.  and J.G. Haworth, Operational  Factors Affecting Nitrification  in the
     Activated Sludge Process. Presented at the  42nd Annual Conference of the Water
     Pollution Control Federation, Dallas, Texas, October (1969), (available as a  reprint
     from the County Sanitation Districts of Los Angeles County).

34.  Murphy, K.L., and P.L. Timpany, Design and Analysis of Mixing for an Aeration Tank.
     JSED, Proc. ASCE, 93, No. SA5, pp 1-15 (1967).

35.  Murphy, K.D., and B.I. Boyko, Longitudinal Mixing in Spiral Flow Aeration Tanks;
     JSED, Proc. ASCE, 96, No. SA2, pp 211-221 (1970).

36.  Metcalf and Eddy, Inc., Wastewater Engineering. New York, McGraw Hill Book Co.,
     1972.

37.  Gujer, W. and D. Jenkins, A Nitrification Model for the Contact Stabilization Activated
     Sludge Process. Water  Research, in press.

38.  Gujer, W. and D. Jenkins, The Contact Stabilization Process-Oxygen and Nitrogen Mass
     Balances. University of California, Sanitary Engineering Research  Lab, SERL  Report
     74-2, February, 1974.
                                      4-101

-------
39.  Sawyer, C.N., Letter communication to D.S. Parker; January 24, 1975.

40.  Sawyer, C.N., Activated Sludge Oxidations, HI. Factors Involved in Prolonging the
     High Initial Rate of Oxygen Utilization by Activated Sludge-Mixtures. Sewage Works
     Journal, 11, No. 4, pp 595-606 (1939).

41. County  Sanitation Districts of Los Angeles County, A Plan for Water Reuse.  Report
     prepared for the members of the Boards of Directors, July, 1963.

42.  City of Jackson, Michigan, Sewage Treatment Plant Operating Reports. August,  1973
     to March, 1974.

43.  Bruce,  A.M., and J.C. Merkins, Further Studies of Partial Treatment of Sewage by High
     Rate Biological Filtration. Water Pollution Control (London), pp 499-527, 1973.

44.  Grantham, G.R., Phelps, E.B., Calaway, W.T., and D.L. Emerson,  Progress Report on
     Trickling Filter Studies. Sewage and Industrial Wastes, 22, No. 7,  pp 867-874 (1950).

45.  Grantham, G.R., Trickling Filter Performance at Intermediate Loading Rates. Sewage
     and Industrial Wastes, 23, No. 10, pp 127-1234 (1951).

46.  Burgess F.J., Gilmour, C.M., Merryfield, F., and J.K. Carswell, Evaluation Criteria for
     Deep Trickling Filters. JWPCF, 33, No. 8, pp 787-799 (1961).

47.  Osbom, D.E., Operating Experiences with Double Filtration in Johannesburg. J. Inst.
     of Sew. Purif., Part 3, pp 272-281 (1965).

48.  Stones, T., Investigation on Biological Filtration at Salford.  Journal of the Institute of
     Sewage Purification, No. 5, pp 406-417 (1961).

49.  Mohlman, F.W., Norgaard, J.T., Fair, G.M., Fuhrman, R.E., Gilbert, J.J., Heacox,  R.E.,
     and C.C. Ruchoft, Sewage  Treatment at Military Installations. Sewage Works Journal,
     18, No. 4, pp 789-1028 (1946).

50.  Heukelekian, H., The Relationship Between Accumulation, Biochemical and Biological
     Characteristics of Film and Purification Capacity of a Biofllter and a  Standard Filter.
     Sewage Works Journal, 17, No. 3, pp 516-524(1945).

51.  Brown  and Caldwell, Report on Pilot Trickling Filter Studies at the Main Water Quality
     Control Plant. Prepared for the City of Stockon, California, March,  1973.

52.  Antonie, R.L., Three Step Biological Treatment with the Bio-Disc Process. Presented at
     the  New York Water  Pollution Control Association, Spring Meeting, Mantauk, New
     York, June, 1972.

                                       4-102

-------
53.  Antonie, R.L., Nitrification and Denitrification with the Bio-Surf Process. Presented at
     the Annual Meeting of the New England W.P.C. Association in Kennebunkport, Maine,
     June 10-12, 1974.

54.  Rotating  Biological Disk  Wastewater  Treatment Process — Pilot Plant Evaluation.
     Report by the Department of Environmental Sciences,  Rutgers University, prepared
     for the Environmental Protection Agency, Project No. 17010 EBM, August, 1972.

55.  Brown and Caldwell, Consulting Engineers, Lime Use in Wastewater Treatment: Design
     and Cost Data. Report submitted to the U.S. Environmental Protection Agency, 1975.

56. Process Design Manual for Carbon Adsorption.   U.S. EPA, Office  of  Technology
    Transfer, Washington, D.C. (1974).                                       *

57.  Rainwater,  F.H.  and L.L. Thatcher, Methods for Collection .and Analysis of Water
     Samples. Geological Survey Water-Supply Paper 1454, USGPO, 1960.

58.  Wood, O.K. and G. Tchbanoglous, Trace Elements in Biological Waste Treatment with
     Specific Reference  to the Activated Sludge Process. Presented at  the 29th Industrial
     Waste Conference, Purdue University, May, 1974.                          ,

59.  Eisenhauer, D.L., Sieger, R.B.  and D.S. Parker, Design of an Integrated Approach to
    Nutrient Removal.  Presented  at the BED-  ASCE Specialty  Conference, Penn. State
     University, Pa., July, 1974.

60.  Environmental Quality Analysts, Inc.,  Letter Report to Valley Community Services
     District, March, 1974.

61.  Stensil, H.D., Oases Wastewater Characterization Study, Chattanooga Moccasin Bend
     Wastewater Treatment Plant. Report prepared by Air Products and Chemicals, Inc.,
     1975.

62.  Sawyer, C.N.,  Wild,  H.E., Jr., and T.C. McMahon, Nitrification and Denitrification
    Facilities,  Wastewater Treatment. Prepared for the EPA Technology Transfer Program,
     August, 1973.

63.  Parker, D.S., Case Histories of Nitrification  and Denitrification Facilities. Prepared for
     the EPA Technology Transfer Program, May, 1974.

64.  Mulbarger, M.C., Private communication to D.H. Caldwell, 1971.

65.  Schwinn,  D.E., Treatment Plant Designed for Anticipated Standards. Public  Works,
     104, No. l,pp 54-57 (1973).
                                       4-103

-------
66.  Wilson, T.E., and M.D.R. Riddel, Nitrogen Removal:  Where Do We Stand? Water and
     Wastes Engineering, 11, No. 10, pp 56-61 (1974).

67.  Wilcox,  E.A.,  and A.A. Thomas,  Oxygen Activated Sludge Wastewater  Treatment
     Systems, Design Criteria and Operating Experience. Prepared for the EPA Technology
     Transfer Program, August, 1973.
  «.
68.  Sorrels, J.H., and  P.J.A. Zeller, Two-Stage Trickling Filter Performance.  Sewage and
     Industrial Wastes,  18, No. 8, pp 943-954 (1956).

69.  Huang, C.S., Kinetics and Process Factors  of Nitrification On a Biological Film
     Reactor. Thesis submitted in partial satisfaction of the requirements for the degree of
     Doctor of Philosophy, University of New York at Buffalo, 1973.

70.  Williamson, K.L. and P.L.  McCarty,  A Model of Substrate Utilization by Bacterial
     Films. Presented at the 46th Annual  Conference  of the Water Pollution Control
     Federation, Cincinnati, Ohio, October, 1973.

71.  Duddles, G.A. and S.E. Richardson, Application of Plastic Media Trickling Filters for
     Biological Nitrification.  Report prepared for the Environmental Protection Agency,
     EPA- R2-73-199, June, 1973.

72.  Bruce, A.M., and  J.C. Merkens, Recent Studies  of High-Rate Biological Filtration.
     Water Pollution Control, pp 113-148 (1970).

73.  Brown and Caldwell, Report on Tertiary Treatment Pilot Plant Studies. Prepared for
     the City of Sunnyvale, California, February, 1975.

74.  Antonie, R.L.,  Nitrification of Activated Sludge Effluent  with the Bio-Surf Process.
     Presented at the Annual Conference of the Ohio Water Pollution Control Association,
     Toledo, Ohio, June 7-13, 1974.

75.  Haug, R.T., and P.L. McCarty, Nitrification with  the Submerged Filter. JWPCF, 44, p
     2086(1972).

76.  Haug, R.T.,  and  P.L. McCarty, Nitrification with  the  Submerged  Filter.  Report
     prepared  by the  Department of  Civil Engineering, Stanford University, for the
     Environmental  Protection Agency, Research Grant No. 17010 EPM, August, 1971.

77.  Young, J.C., Baumann,  E.R.,  and D.J. Wall, Packed-Bed Reactors for Secondary
     Effluent BOD and Ammonia Removal. JWPCF, 47, No. 1, pp 46-56 (1975).

78.  General Filter Co., Paktor, Packed Bed Reactor, Bulletin No. 7305,  1974.
                                       4-104

-------
79.  Mechalas,  B.J., Allen,  P.M.  and W.W. Matyskiela, A Study of Nitrification  and
     Dehitrification.  A report prepared for the Federal Water Quality Administration,
     WPCRS 17010 DRD 07/70, July, 1970.

80.  Young, J.C.  and M.C.  Stewart, Advanced Wastewater Treatment with Packed  Bed
     Reactors.  Report of the Engineering Research Institute, Iowa State University,  No.
     ERI-73108,May, 1973.

81.  Gasser, J.A.,  Chen, C.L.,  and  R.P. Miele, Fixed-Film Nitrification of Secondary
     Effluent.  Presented at the EED-ASCE  Specialty  Conference, Penn. State University,
     Pa., July,  1974.

82.  Kenney, F.R.,  Letter communication to D.S. Parker. General Filter Co., Ames, Iowa,
     September, 1974.

83.  Young, J.C., Unpublished data, September 27, 1974.

84.  Smith, J., Personal communication to D.S. Parker. Environmental Protection Agency,
     Cincinnati, Ohio, March, 1974.

85.  Loftin, W.E., Personal communication  to D.S. Parker. City of Livermore, California,
     April, 1972.

86.  Sacramento Area Consultants,  Oxygen Activated Sludge  Pilot  Studies for  the
     Sacramento Regional Treatment Plant. Report prepared for the Sacramento Regional
     County Sanitation District of Sacramento County, California, July, 1974.

87.  Bishop, D.F.,  Personal  communication to D.S.  Parker.  Environmental Protection
     Agency, Washington, D.C., April, 1974.

88.  Commonwealth Department of Works, Australian Capital Territory, letter communi-
     cation to R.C. Aberley, dated June 1 and July 20, 1972.

89.  Metropolitan St. Louis Sewer District and Havens and Emerson, Ltd., Cost-Effective
     Design  of Wastewater  Treatment  Facilities  Based on Field Derived Parameters.
     Prepared for the EPA, Report No. EPA- 670/2-74-062 (PB-234 356), July, 1974.

90.  Aberley, R.C.,  Rattray, G.B. and P.P. Dougas, Air Diffusion Unit. JWPCF, 46, No. 5,
     pp 895-910 (1974).

91.  Leary, R.D., Ernest, L.A., and W.J. Katz, Full Scale Oxygen Transfer Studies of Seven
     Diffuser Systems. JWPCF, 41, No. 3, pp 459^73 (1969).
                                      4-105

-------
  92.  Nogaj, R.J., Selecting Wastewater Aeration Equipment. Chemical Engineering, April,
      1972.

  93.  City of Medford, Oregon, Sewage Treatment Plant Operating Reports. July, 1974.

  94.  San Pablo Sanitary District, California, Wastewater Treatment Plant Operating Reports.
      June,  1973, to July, 1974.

  95.  Process Design Manual for Phosphorus Removal.  U.S. EPA, Office  of Technology
      Transfer, Washington, D.C. (1971).

  96.  Process Design Manual for Suspended Solids Removal. U.S. EPA, Office of Technology
      Transfer, Washington, D.C., January, 1975.

  97.  Cleasby,  J.L., and E.R. Baumann,  Wastewater Filtration,  Design  Considerations.
      Prepared for the EPA Technology Transfer Program, July, 1974.

  98.  Dick,  R.I., Role of Activated Sludge Final Settling  Tanks. JSED, Proc. ASCE, 96, No.
      SA2,pp 423-436 (1970).
                                                  f
 99.  Dick,  R.I. and A.R.  Javaheri, Discussion of Unified Basis for Biological  Treatment
      Design and Operation by A.W.  Lawrence and P.L.  McCarty.  JSED, Proc. ASCE, 97,
      SA2, pp 234-238 (1971).

100.  Dick,  R.I. and K.W.  Young, Analysis of Thickening Performance of Final Settling
      Tanks.  Presented at the  27th Industrial Waste  Conference,  Purdue  University,
      Lafayette, Indiana, May  7-9,  1974.

101.  Tenney, M.W. and W. Stumm, Chemical Flocculation of Microorganisms in Biological
      Waste Treatment.  JWPCF, 37, p. 1370 (1965).

102.  Parker, D.S., Kaufman, W.J., and D. Jenkins, Physical Conditioning of Activated
      Sludge Floe. JWPCF, 43, No. 9, pp 1817-1833 (1971).

103.  Stamberg, J.B., Bishop, D.F., Hais, A.B., and S.M. Bennet, System  Alternatives in
      Oxygen Activated Sludge.  Presented at the 45th Annual Conference of the WPCF,
      Atlanta, Ga., 1972.

104.  Sawyer, C.N., and  L. Bradney, Rising of Activated Sludge in Final Settling Tanks.
      Sewage Works Journal, 17, No. 6, pp. 1191-1209 (1945).

105.  Clayfield, G.W., Respiration and Denitrification Studies on  Laboratory and Works
      Activated Sludges. Water Pollution Control, London, 73, No. 1, pp 51-76 (1974).
                                       4-106

-------
106. Stone, R.W., Parker, D.S., and J.A. Cotteral, Upgrading Lagoon Effluent to Meet Best
     Practicable  Treatment.  Presented  at  the 47th  Annual Conference  of the Water
     Pollution Control Federation, Denver, Colorado, October, 1974.

107. Lawrence, A.M., and C.G. Brown, Biokinetic Approach to Optimal Design and Control
     of Nitrifying Activated Sludge Systems. Presented at the Annual Meeting of the New
     York Water Pollution Control Association, New York City, January 23, 1973.

108. Stall, T.R.,  and R.H. Sherwood,  One  Sludge or  Two Sludge? Water and Wastes
     Engineering, p 41-44, April, 1974.

109. Sutton, P.M.,  Murphy, K.L., and B.E.  Tank,  Biological Nitrogen Removal -  The
     Efficacy of the Nitrification Step.  Presented  at  the  WPCF  Conference,  Denver,
     October, 1974.
                                      4-107

-------
                                     CHAPTER 5

                          BIOLOGICAL DENITRIFICATION
5.1 Introduction

The process of biological denitrification is applicable  to  the  removal of nitrogen from
wastewater  when the nitrogen is predominately in the nitrate or nitrite form. In municipal
applications, the  nitrogen  in  the raw wastewater  is  primarily present as organic and
ammonia-nitrogen and first mus,t be converted to an oxidized form (nitrite or nitrate) prior
to  biological denitrification. The biological oxidation  process used  for  this conversion,
nitrification, was described in Chapters 3 and 4.

This chapter presents design criteria for several alternative denitrification systems including
suspended growth and  attached growth systems using methanol as the carbon  source and
combined carbon oxidation-nitrification-denitrification systems using wastewater or endo-
genous carbon sources.  The basic chemistry of denitrification was described in Section 3.3.

5.2 Denitrification in Suspended Growth Reactors Using Methanol as the Carbon  Source

The suspended  growth denitrification  process is  a form of the activated sludge process.
There are several differences between its typical application for organic carbon removal and
in its use for denitrification. In common is the provision of a reactor, in which the biomass is
kept in  suspension in the liquid by  mixing. Also provided in both  applications  is a
sedimentation tank for separation of the mixed liquor solids from the effluent, allowing the
biomass to be recycled in the system and also allowing the production of a clear effluent for
discharge  or subsequent treatment. Two typical suspended growth denitrification systems
are illustrated in Figure  5-1.

There are other analogies'between suspended growth systems used for denitrification and
organic  carbon  removal.  In  organic  carbon  removal  applications,  dissolved  oxygen is
introduced  into the reactor by  aeration so  that biological oxidation of the organic  matter
can take place.  In the  process  of carbon oxidation, oxygen  is consumed as the electron.
acceptor  in the oxidation process.  In the  process of denitrification,  carbon (usually
methanol) is oxidized with nitrate or nitrite serving as  the electron acceptor (see Section
3.3.2). In denitrification  as opposed  to organics removal, it  is  the nitrate  that  is  the
pollutant  that is to be removed and the carbon source that  is added. In organics removal, it
is the carbon that is the pollutant that is  to be removed  and  the oxygen  that  is  added.
•Needless to say, only sufficient carbon (such  as methanol) is added  in denitrification to
accomplish  the  nitrate  removal, as excess dosing causes organics to appear in the effluent
unless  control measures  are undertaken. These residual organics, if left in the  effluent,
would exert BOD5 and might cause violation of effluent requirements.
                                         5-1

-------
                                             FIGURE 5-1
                     SUSPENDED GROWTH DENITRIFICATION SYSTEMS USING METHANOL
                  A. ORIGINAL  DENITRIFICATION  SYSTEM  (Reference))
 METHANOL
NITRIFIED i
EFFLUENT
                     ANOXIC MIXED
                DENITRIFICATION  REACTOR
CLARIFIER /DENITRIFIED
            EFFLUENT
                                                                  -AERATED
                                                               NITROGEN STRIPPING
                                                               CHANNEL  T = 5 min.
                    RETURN  DENITRIFIED  SLUDGE
                   B. MODIFIED  DENITRIFICATION  SYSTEM  (Reference  2,3)
METHANOL
i
NITRIFIED I
EFFLUENT




i



A NO* 1C M 1 y FTJ
nFMlTDIFIPATIf^M D IT A f* T A D










AERATED
O T A R 1 1 1 7 A T 1 /"* M
T A M Vf

T = 50 min.


^\
, ^ _ ^/nFwiTRiFirATinNJ ^
\ Cl ARIFIFR /PFNITRIFIED
/ \ / EFFLUENT
^-MILDLY AERATED
PHYSICAL CONDITIONING
CHANNEL i'
                     RETURN  DENITRIFIED SLUDGE

-------
     5.2.1 Denitrification Rates

Currently used denitrification rate data for  design of denitrification  systems  are based
upon work described in references 4,  5, 6, and  7. Data  from these  investigations are
summarized in Figure 5-2. Rather than  show individual data points, or trend lines, Figure
5-2 shows boundaries for the data so that the range in variation of denitrification rates can
be inspected. Earliest available measurements were those from Manassas, Va.4 which have
been found to be considerably higher than subsequent observations at three other locations.
The data from CCCSD, Ca., Blue Plains, D.C., and Burlington, Canada, all are in reasonable
agreement with each other, and  are all well below the Manassas rates. A possible  reason for

                                   FIGURE 5-2

             OBSERVED DENITRIFICATION RATES FOR SUSPENDED
                      GROWTH SYSTEMS USING METHANOL
 CO
 to
 -J
 5
 .a
\
 o
 0)
 k.
 a
 •o
 N
X
O
O
ft:

.o
*.
O

-------
the higher measurements at Manassas is that an acid solubilization procedure was employed
prior to  the volatile solids determination,  which may have acid hydrolized organic matter
resulting in low measurements of volatile solids (and higher apparent denitrification rates).
The  earlier  work at  Manassas^  prior to implementation  of the" acid step showed
denitrification rates closer to the observations of other locations.

Laboratory  studies  on  synthetic  nitrate  containing  wastes have  shown much  higher
denitrification  rates than are found in Figure  5-2.10,11  However, the biological solids
developed in such laboratory systems do not contain the levels  of refractory  solids  that
build up in practical systems operated under field conditions. Therefore, the data developed
from such  laboratory  studies  are  not directly  useful in  establishing  accurate  design
parameters.

Conditions maintained during the field  studies may influence field measured denitrification
rates considerably. When the denitrification reactor is continuously operated close to the
maximum growth rate  of the denitrifying  organisms, it is probable that  the denitrifying
activity of the  biomass is higher than when the system is operated with a high safety factor.
For instance, in studies in Ontario it was found that measured peak denitrification rates
were approximately 30 percent greater at a solids  retention time of 3  days than rates
measured at a  solids  retention time of 6 days. 12 Thus,  differences in measured rates  may
reflect variations in operating conditions among the various locations.
Observed  rates in Figure 5-2  are essentially the experimentally  determined values  of the
term qp using the notation presented in Section 3.3.5.2.  The peak nitrate removal rate, qrj,
is the reaction rate  when neither methanol nor nitrate  is limiting the reaction rate.  Sub-
sequent sections show how these peak nitrate removal rates are used in design calculations.

     5.2.2 Complete Mix Denitrification Kinetics

The equations presented in Section 3.3.5 are directly applicable to the design of complete
mix 'denitrification systems. The design procedure for denitrification uses the safety factor
concept to  relate peak  nitrate removal  rates,  qj), to  design nitrate  removal rates, qp.
Expressed in terms of solids retention time, the safety factor concept is:
                                   -                                            (3-29)
                                     c
where:      SF  =   safety factor,
            6    =   solids retention time of design, days, and
             C

            6    =   minimum solids retention time, days, for
                     denitrification,

                                         5-4

-------
The design nitrate removal rate, qp,  can be related to the safety factor and the peak nitrate
removal rate by using the following equations in conjunction with Equation 3-53:
where:      YD  =    denitrifier gross yield, Ib VSS grown/lb NO" - N
                      removed,

            K ,   =    decay coefficient, day" ,

            qD   =    nitrate removal rate, Ib NOl -N rem./lb VSS-day, and

            q~   =    peak nitrate removal rate, Ib NOl -N rem./lb VSS-day.

In evaluating these equations in design calculations, the specific values of Y and Kj given in
Section 3.3.5.4 may be used. Figure 5-2 may be used to arrive at estimated values of qry
Considering the range in the data in Figure 5-2, conservative practice in the absence of pilot
data would be to pick the lowest denitrification rates observed for qrj, e.g. at 10 C qrj =
0.05, at 15 C qD = 0.08,  at 20 C qD = 0.15, and at 25 C qD = 0.20. Use of these minimum
values of qjj> will result in very conservative reactor designs. Pilot plant studies may be useful
to define applicable values of qpj, as the potential for establishing higher denitrification rates
for a particular location is good; evidence of this is the range of denitrification rates among
the various locations shown in Figure 5-2.
                                                               /\             _
As a design example consider a case where the temperature is 25 C, qp_) = 0.2 Ib NC>3 rem./lb
MLVSS/day, Y = 0.9 Ib VSS/lb NO^ rem., Kj = 0.04 day"1, and KD = 0.15 mg/1. Assume
that due to diurnal variations in load (Section 5.2.2.2), a minimum safety factor of 2.0 is
adopted. Consider a 30 mgd treatment plant, where 25 mg/1 of nitrate-N must be removed.

     1.    Using Equation 3-54,  calculate the minimum solids retention time for denitrifica-
          tion:

                             -i- =0.9(0.2) -0.04 = 0.1 4
                                      0m = 7.14 days
     2.   Calculate the design solids retention time (Equation 3-29):


                                         5-5

-------
                        0  = 2.0(7.14) = 14.3 days
                         c
3.   Calculate the design nitrate removal rate (Equation 3-50):
                           ,-43=^0-0.04

                        •••    qD = 0. 1 2 Ib NO~ -N rem./lb MLVSS/day

4.   Calculate the steady state nitrate content of the effluent. The expression relating
     removal rates to nitrate level, from Equations 3-47, 3-48 and 3-49, is as follows:

      where:      K^  =   half saturation constant, mg/1 NO- -N, and
                 D,   =   effluent concentration of nitrate nitrogen mg/1.
      Evaluation of this equation for this example yields:
                                        Dl
                         0.12 = 0.20
                                     0.15+ DJ
                  D] =0.23 mg/1 NO3-N
5.   Determine  the  hydraulic  detention time  at  average  dry weather  flow. The
     equation for nitrate removal rate is useful in this calculation.
                                  D-D,
     where:      D   =    influent NOl -N, mg/1
                 D.  =    effluent NO--N, mg/1
                 Xj  =    MLVSS, mg/1, and
                 HT =    hydraulic detention time, days.


                                   5-6

-------
    The mixed liquor volatile suspended solids (MLVSS)  level is dependent on the
    mixed liquor total suspended solids, which is in turn dependent on the operation
    of the denitrification sedimentation tank (see Sections 4.10 and 5.6). Assume for
    the purposes of this example that the design mixed liquor content at 25 C is 3000
    mg/1. At  a volatile content of 80  percent,  the MLVSS is  2400 mg/1.  From
    Equation 5-2, the hydraulic detention time is:
                              = 1.99hr

6.    Determine the sludge wasting schedule.  The equations developed for wasting
     in the nitrification system are  directly  applicable  here. The necessary sludge
     inventory is:

                           I = 8.33(X1 • V)                                (4-7)

     where:      I   =   inventory of VSS in the anoxic denitrification
                         reactor, Ib,

                X.  =   MLVSS in the reactor, mg/1, and

                V   =   volume of the reactor, mil gal.

     In the example at hand:

                           I = 8.33(2400)(0.083)(30) = 49,780 Ib VSS

     From Equation 4-8,  the sludge wasting from  all sources is defined by  the
     following equation:
                               S=                                       (5-3,
                                     C

    where:      S   =    total sludge wasted in Ib/day

    For this example:

                      S = 49,780/14.3 = 3,481 Ib VSS/day

    The total sludge to be wasted each day is made up of two components, as shown:


                         S = 8.33(Q-X2+W-Xw)                         (4-6)


                                  5-7

-------
          where:      Q   =   influent (or effluent) flow rate, mgd,

                      W   =   waste sludge flow rate, mgd,

                      X^  =   effluent volatile suspended solids, mg/1, and

                      X   =   waste sludge volatile suspended solids, mg/1.
                       iV

         The sludge contained in the effluent (the term Q-X2 above) can be calculated
         assuming that the effluent VSS is 10 mg/1:

                               8.33(10)(30) = 2,499 Ib VSS/day

         By difference, the  Ib of MLVSS to be  wasted from the mixed liquor or return
         sludge is:

                               3,481 - 2,499 = 982 Ib VSS/day

     7.   Methanol requirement. From Section 3.3.2, an estimate of 3.0 Ib per Ib of nitrate
         N removed is reasonable. The methanol requirement is:

                            3.0(25 - 0.23)(8.33)(30) = 18,570 lb/day

The  sludge yield and decay values used  above  are for a case where only a short aeration
period is used prior to clarification. When an aerated stabilization step is employed, very
much lower sludge wasting is required than presented in the above example. In cases where
an aerated stabilization tank is employed, only the sludge inventory under anoxic conditions
should be considered in the sizing of the anoxic reactor for denitrification (Step 5).

         5.2.2.1 Effect of Safety Factor on Steady-State Effluent Quality

In the design example previously presented, the safety factor was assumed to be 2.0, based
on considerations presented in  Section 5.2.2.2. The effect of alternative assumptions on the
effluent nitrate  level are presented  in  Figure 5-3.  As may be seen, the assumption of the
safety factor has a marked effect on the effluent quality of complete mix denitrification
systems.

         5'.2.2.2 Effect of Diurnal Load Variations on Effluent Quality

As is the case for nitrification,  load variations have a significant effect on effluent quality.
Since upstream treatment units to some extent equalize load variations, the peak to average
load  ratio  is generally lower than  for  the  nitrification  stage.   The  effect  of  load
variations can  be analyzed in  a similar  manner  to  that used  for  nitrification (Section
4.3.3.2). By analogy  to  the  nitrification  case,  the mags balance yields the  following
expression for nitrate at any time:

                                         5-8

-------
                                  FIGURE 5-3


             EFFECT OF SAFETY FACTOR ON EFFLUENT NITRATE

                    LEVEL IN SUSPENDED GROWTH SYSTEM
                                    COMPLETE  MIX
                              SAFETY   FACTOR,  SF
                                                                       3.0
                                                                        (5-4)
where:      D   =   influent nitrate -N level at any time, mg/1,



           D   =   mass average influent nitrate -N level over

                   24 hours, mg/1,


           D^  =   mass average effluent nitrate -N level at any time, mg/1,



           D ,  =   mass average effluent nitrate -N level over 24 hours, mg/1,
            1                                              o


           Q   =   influent flow rate at any time,      °



           Q   =   average daily influent flow rate,     0



           jz   =   design denitrifier growth rate, day" , and
           jx   =   maximum denitrifier growth rate, day



                                    5-9
                                                  "

-------
Equation  5-4 has been used to evaluate the effect of the diumal load variations shown on
Figure 5-4 using the design example  conditions given in Section 5.2.2.1.  The influent
nitrate-nitrogen concentration was assumed constant at 25 mg/1, and the load variation was
assumed to be due to variation in flow only. Several trial calculations using Equation 5-4
over a 24 hour cycle were necessary to derive values of Dj. Results of these calculations for
several values of the SF are also shown on Figure 5-4. As may be seen, the safety factor, SF,
has a marked effect on  the nitrate bleedthrough occurring during peak load conditions.  In
the case examined,  a safety factor of 2.0 was sufficient to prevent excessive nitrate leakage.
The peak  to average load  for this case was  1.5. In summary, it  would appear that as a
minimum, the safety  factor should exceed  the peak to average load ratio  to  prevent
excessive nitrate leakage during peak load conditions.

     5.2.3 Plug Flow Denitrification Kinetics

The design approach  for  plug flow denitrification  reactors is  similar to the approach
developed for complete mix reactors, with the exception of the equations used to predict
effluent quality. Lawrence and McCarty's^ solution for plug flow kinetics is applicable:

                              YDqD(D  -D)
                         —              D    -Kd                           <5-5>
All these terms are as defined previously in Section 5.2.2.

The  kinetic design approach for plug flow follows that used for complete mix systems in
Section 5.2.2, excepting that at step 4, Equation 5-5 is used instead of Equation 5-1 to find
the effluent nitrate level.

Equation 5-5 can also be used to find the safety factor required to obtain any desired nitrate
level. This has been done for the example presented in Section 5.2.2.1 and plotted in Figure
5-3.  A  comparison of  the  safety factor  required to obtain  the  same nitrate  level in a
complete mix system yields the conclusion that plug flow  systems can be designed with
considerably lower safety factors while obtaining the same effluent quality.

While kinetic models have not been extended to the point where they can be expected to
describe the effect of diurnal variations on plug flow systems, it can be expected that the
effects of these loads will be similar to  those experienced in complete mix systems. This is a
result  of the  fact  that  once  the  effluent level  rises above 1  mg/1 nitrate -N,  the
denitrification rate becomes essentially zero order. For  zero order reactions, there is little
difference between plug 'flow and complete mix reaction kinetics. Therefore, the nitrate
bleedthrough in a plug flow reactor can be expected to closely approach that in a complete
mi-:  reactor under diurnal peak load  conditions. To prevent excessive nitrate leakage during
peak load  conditions,  the recommendation in  Section 5.2.2.2 should be adopted; as a
minimum, the safety factor should exceed the peak to average load ratio.

                                        5-10

-------
 The plug  flow hydraulic regime can be approximated by a series  of  complete mix
 denitrification tanks, in which backmixing is prevented. An example is provided by the case
 history of the CCCSD Water Reclamation Plant, presented in Section 9.5.2.1.

                                FIGURE 5-4

       EFFECT OF DIURNAL VARIATION IN LOAD ON EFFLUENT NITRATE
            LEVEL IN COMPLETE MIX SUSPENDED GROWTH SYSTEM
  Q
  s
  UJ
55
O
(fc ly
  Ul
  Uj

  Uj
  Uj
  I
  kl
     ISO
     100
      50
             T
                                      I
             I
I
I
I
I
I
      2400  0200  0400  0600  0800  1000  1200  1400  1600  1800 2000  2200  2400
                                    TIME,  HR
         A. ASSUMED VARIATION  IN INFLUENT NITROGEN  LOAD
      2400  0200  0400 0600  0800 1000  1200  1400  1600  1800  2000  2200  2400
                                   TIME,  HR

              B.  CALCULATED  EFFLUENT  NITRATE LEVEL

                                   5-11

-------
     5.2.4 Effluent Quality from Suspended Growth Denitrification Processes

Since denitrification technology is  new, there  is a concern on  the part of some design
engineers  that  biological  nitrification-denitrification systems are unstable and  produce
results of high variability. However, large scale tests of biological nitrogen removal have
demonstrated  over  relatively long periods that a consistently low nitrogen level can be
obtained.

         5.2.4.1 Experience at Manassas, Va.

The EPA conducted a 0.2 mgd (760 cu m/day) test of the "three sludge" system for eight
months at Manassas, Va. The system consisted of primary treatment and three  separate
suspended  growth  stages  for organic carbon oxidation,  nitrification  and denitrification
followed by filtration, as shown in Figure  5-5. Alum was added to the  first and  third
suspended growth stages for phosphorus removal. A dose of polymer was added to the third
reactor effluent. Performance data presented in the form of frequency distribution diagrams
show that the performance of the closely monitored system was very stable.^ A tabular
summary of denitrification effluent quality is shown in Table 5-1 for the last four months of
operation. As may be seen, an effluent very low in total phosphorus and total nitrogen was
obtained from the denitrification system and filtration provided further reductions. Further
details are available in the papers produced from the project. 4,9
                                    FIGURE 5-5

   THE THREE SLUDGE SYSTEM AS TESTED AT MANASSAS, VA (REFERENCE 4)


VI
1
Ett
BACKWASH
ALUM METHAN
LI
	 ,, 	 .,— JL
(2) © f (2) © f '
xxi Y '
RETURN SLUDGE] 1 RETURN SLUDGE 1 ( R
I 1 II
i
A
OL ACID
1.
LUM
POLYMER
I
(C) ® ^


ETURN SLUDGE |
	
1
CI2
CD ©
I
i
i
t !
BACKWASH 1
1
1
^ 	 1
                                   TYPICAL WASTE SLUDGE LINE
     7
 WASTE  TO SOLIDS HANDLING SYSTEM AND ULTIMATE DISPOSAL
PRIMARY
TREATMENT
©SEDIMENTATION
TANK
HIGH RATE
ACTIVATED SLUDGE .
(5)AERATION TANK
(3)SEDIMENTATION
^-^ TANK
NITRIFYING
ACTIVATED SLUDGE
(?)AERATION TANK
(JT) SEDIMENTATION
^ TANK
DENITRIFYING
ACTIVATED SLUDGE
§ANOXIC REACTORS
AERATED CHANNEL
(5)S£DIMENTATION TANK
POST
TREATMENT
§ MIXED MEDIA FILTERS
CHLORINE CONTACT
(JT) POST AERATION
                                        5-12

-------
                                     TABLE 5-1

             DENITRIFICATION PERFORMANCE: FINAL FOUR MONTHS
             OF OPERATION AT MANASASS, VIRGINIA (REFERENCE 4)


Parameter
S3
COD
BOD5
Total P
>
Organic N

NH+-N

N02-N
NOg-N
rf
T
O
T
r A
L
N
After final
clarification,
mg/1
2
21
4.0
0.6
1.0

0.0

0.0
0.8
+*



> 1.8


After mixed
media
filtration,
0
16
0.8
0.3
0.8

0.0

0.0
0.7
*i

mg/1







^1.5


         5.2.4.2 Experience at the CCCSD's Advanced Treatment Test Facility

In  November,  1971, the Central Contra Costa Sanitary District (CCCSD) began  the
operation of  a full-scale Advanced Treatment Test  Facility  (ATTF) at its  existing
wastewater treatment plant in California. 3 Operation of the  facility ultimately extended
over 23 months. A purpose of the test  facility  was to  obtain data on the ATTF System
sequence  of processes  (Figure  5-6)  that  had been  proposed  for  the CCCSD Water
Reclamation Plant. ^ Another purpose was to dispel the notion that the nitrification and
denitrification processes were unstable due to the erratic nitrification-denitrification results
previously obtained in a small-scale pilot study.3>15,16

The ATTF process units had capacities ranging from 2.5 mgd (9,464 cu m/day) in  the
primary  step  to  0.5 mgd  (1,990 cu m/day) in the  denitrification  facilities. Primary
clarification  follows  lime addition and preaeration and is coupled  with a separate stage
nitrification step. The use of lime in the primary stage removes much of the organic carbon
load from the  nitrification stage, thus allowing stable oxidation  of ammonia to nitrate.
Addition of lime  also enhances the removal of phosphorus, heavy metals and viruses.
Addition of lime in the initial stage of treatment, in  contrast  to  the use of lime after
conventional secondary treatment, enables the achievement of better stability in succeeding
                                       5-13

-------
                   FIGURE 5-6

  ATTF SYSTEM FOR NITROGEN AND PHOSPHORUS REMOVAL

        RAW WASTEWATER
                  LIME

                      """    POLYMER OR
                          FERRIC CHLORIDE
LIME  REACTOR
 (PREAERATION)
            I
         PRIMARY
   SEDIMENTATION TANK
   CHEMICAL
   PRIMARY
   EFFLUENT  A
            C02
       OX IDAT ION -
   NITRIFICATION TANK
            I
        SECONDARY
   SEDIMENTATION TANK
N2
       E:
                      SLUDGE TO
                      *   SOLIDS
                      PROCESSING
                      AIR

                      RETURN
                      SLUDGE

                       WASTE SLUDGE
                      	*•  TO
                              RAW WASTEWATER
                  METHANOL
    DENITRIFICAT ION
           TANK
            I
                           MIXING
         AERATED
   STABILIZATION TANK
            I
          FINAL
   SEDIMENTATION  TANK
            I
                     RETURN
                     SLUDGE
                       WASTE  SLUDGE
                       -*-  TO
                              RAW WASTEWATER
    CHLORINE  CONTACT
             I
     ADDITIONAL TREATMENT
       FOR  INDUSTRY
                      5-14

-------
treatment processes and also allows the elimination of the need for a biological treatment
step for  organic carbon removal ahead of the nitrification stage. Biological denitrification
follows nitrification, converting nitrate to nitrogen gas.

Performance data for a representative three months of operation of the ATTF are shown in
Table 5-2.3 of particular interest is the fact that the 90 percentile performance level did not
vary widely from the  median performance level for the various constituents. This provides
statistical confirmation that the nitrification and denitrification performance of the ATTF
system  was  quite stable. The concentration of organics in the nitrified  and denitrified
effluents was low, as measured  by BOD and organic  carbon.  Operation for complete
nitrification also resulted in high organic removals.  Similarly, suspended  matter  in the
nitrified  and denitrified effluents were also exceptionally low (Table 5-2). Nutrients are
effectively removed in the ATTF system.  Total nitrogen in the denitrified effluent averages
less than 2 mg/1. Total phosphorus averaged 0.5 mg/1.

5.3 Denitrification in Attached Growth Reactors Using Methanol as the Carbon Source

Denitrification  in attached growth reactors has been accomplished  in a wide  variety of
denitrification  column configurations  using  various  media  to  support  the  growth  of
denitrifiers. In part because of this  variability among systems, it is difficult to set forth
generally useful  design criteria at the present time. Several useful approaches are suggested
for characterizing  denitrification in  attached growth systems and presently available data
(1975) are analyzed by these procedures.

     5.3.1 Kinetic Design of Attached Growth Denitrification Systems

In order to size an attached denitrification reactor, knowledge is required  of the reaction
rates taking place  in  the reactor volume.  In estimating reaction rates, the level of biomass
effective in denitrification must also be known. One approach is to estimate the level  of
biomass on the  media surface and thenuse measured reaction rates per unit of biomass  to
obtain the nitrogen removal capability of a column containing the  estimated  amount  of
biomass. 17,18,19 ffas approach is limited in its usefulness in  design applications because
there is insufficient data  available at the present time to predict in advance  the level  of
biomass  that  will develop  on  the  media. Biomass  development will be  dependent on
hydraulic regime, type of media, loading, means for promoting sloughing and possibly the
temperature of operation.

An indirect procedure for consideration of denitrification rates in design  is to adopt the
approach in which denitrification rates are expressed in  terms of surface  nitrate removal
rates, e.g. Ib nitrate  -N  removed per sq ft per day.7> 18,20 Qn this basis, high surface
removal rates would  reflect extensive  biological film development, whereas  low surface
removal rates would  reflect minimal surface film development. The surface denitrification
rate varies  considerably  among the various  denitrification column  configurations and is
affected by the loadings under which the process is operated.

                                        5-15

-------
                                                 TABLE 5-2
                         ATTF PERFORMANCE SUMMARY, APRIL 16 TO JULY 15, 1972
                         (CENTRAL CONTRA COSTA SANITARY DISTRICT, CA., REF. 3)
Constituent
BOD5
TOCc
SOC
SS
Turbidity (JTU)
Settleable solids (ml/1)
Organic N
NH4- N
NOjj-N
NO§-N
Total P
Ortho P
TDS
Conductivity (105 mho/cm)
Alkalinityd
Cad
Mgd
Raw wastewater, mg/1
mean
203
122
22
214
-
13.5
_
-
_
_
9.86
9.74
583
1230
215
67.9
84.0
median
199
120
23
212
r
13.0
_
-
-
_
9.57
9.90
561
1210
218
66.0
84.0
90%a
235
152
25
295
-
16.5

-
_
_
11.19
ll.:24
750
1366
237
78.5
95.6
Chemical primary,0 mg/1
mean
57
42
27
26
12.8
.084
_
24.0
-
_
.86
.61
_
-
254
155
38.6
median
54
41
27
23
12.0
0
—
23.8
-
_
.59
.46
_
-
249
150
34.0
90%a
79
55
31
45
23
.37
_
28.5
-
-
1.85
1.75
_
-
293
184
74.0
Nitrified effluent, mg/1
mean
3.6
8.9
5.6
4.5
1.4
-
.26
.48
27
»OI5
1.04
1.00
634
1226
105
159
57.3
median
3.5
8.5
5.5
4.0
1.3
-
0
.30
27
.013
.72
.71
636
1223
106
161
61.0
90%a
6.8
11.5
6.7
7.8
2.2
-
.67
.58
32
.027
2.11
1.80
724
1394
127
187
73.0
Denitrified effluent, mg/1
mean
3.2
9.5
5.9
4.5
1.4
-
1.1
.31
.48
.009
.50
.52
537
1160
183
174
30.1
median
3.0
9.0
6.0
4.0
1.3
-
1.1
.30
0
.008
.36
.36
551
1147
187
172
30.0
90%a
4.8
11
6.6
7,7
1.8
-
2.5
.40
.79
.018
.98
.90
616
1318
217
190
48.4
a90% of observations are equal to or less than stated value.
 pH 10.2 operation to June 1, pH 11.0 thereafter.
°roC = total organic carbon, SOC = soluble organic carbon
das CaC00

-------
     5.3.2 Classification of Column Configurations

The various types of denitrification columns currently available are summarized in Table 5-3
along with calculated peak surface denitrification rates. The first type of categorization is
with respect to the condition of the void space in the  column. Until very recently,  all
denitrification work has been conducted on submerged columns wherein the voids  were
filled with the fluid being  denitrified. Very recently, a new type  of column has  been
developed in which the voids are filled with nitrogen gas,  a  product of denitrification.6>21
The  submerged  columns  can  be  further subdivided into packed  bed  and  fluidized bed
operations.

The varieties of media being employed for commercial application are also shown in Table
5-3;  the listing is not meant to exclude commercial products which may be equivalent to
those listed. For instance, other vendors of plastic media are listed in Table 4-15.

Details of  design construction and operation  of each column type  are presented in the
following sections. Also included is a comparison of column systems.

         5.3.2.1 Nitrogen Gas Filled Denitrification Columns — Packed Bed

The nitrogen gas filled column was recently developed for  use at the Lower Molonglo Water
Quality Control' Centre (LMWQCC),  currently  under construction.6»21  Details  of the
column design used for this Canberra,  Australia installation are shown in Figure 5-7. The
column media consists of corrugated plastic sheet modules of the same type used in plastic
media  trickling filters (Table 4-15). As opposed to  previously developed attached growth
processes, the present  column  system  is not submerged  with liquid; rather, the column's
void spaces are filled with nitrogen gas, a reaction product of the denitrification process.

In the denitrification column the influent wastewater is  spread  out over the top of the
media and  then the liquid flows in a thin film over the media on which the organisms grow.
These organisms maintain a balance so that an active biological film develops. The balance is
maintained by sloughing of biomass from the media, either by death or by hydraulic erosion
or both.  Sufficient  voids  are present in the media to prevent clogging and ponding. The
denitrification column must be followed by a clarification step to remove sloughed solids.
Pilot studies  for the LMWQCC  facility indicated  that effluent solids should  be  sufficiently
low so that the effluent can go directly to a tertiary multi-media filter.

Oxygen must be excluded from the  denitrification column since its presence would prevent
nitrate or nitrite in  the  applied liquor from serving as the electron acceptor in the biological
oxidation of the applied carbon (methanol). Therefore, the denitrification column is sealed
to prevent  intrusion of air.  The  units are  vented to allow  outflow of nitrogen gas while
preventing  inflow of atmospheric air. Soon after  start-up,  the nitrogen gas displaces the air
or other gases initially  present,  leaving a nitrogen gas atmosphere in the voids. This ensures
the anoxic environment that is required for  denitrification.

                                         5-17

-------
                                           TABLE 5-3

TYPES OF DENITRIFICATION COLUMNS AND MEASURED DENITRIFICATION RATES


Void
Space
Nitrogen
gas



Liquid










1
*

1
















Type
Packed bed



Packed bed
























Fluldlzed
bed





Nature of
Surface
High porosity
corrugated sheet
modules


High porosity
corrugated sheet
modules or
dumped media









Low porosity
fine media











High porosity
fine media,
sand
Activated carbon




Ref.
No.
6,
21
6,
21

22
23

24

7.
20


17,
25

18



26


24


27,
28

29
30




Media
trade name.
Munters
Plasdek



Koch
Flexlrlngs
Envlrotech
Surfpac
Koch
Flexlrlngs
Intalox
saddles


Raschlg
rings

Gravel

d50=3. 4 to
14. 5 mm
Sand
d50=0. 9mmf

Sand
d50=3 to
4 mm
Sand
d10»2.9mm
U.C.= 1.13
d50=0.8Smmf
d10=0.65mm


Specific
surface
sf/cu ft
(mW)
68
(223)

42
(138)
65
(213)
27
(89)
105
(344)
142 to 274
466 to 899)


79
(259)



245 to 85
804 to 279)

450'
(1,476)

250
(825)

270'
(886) .
130f
(426)
3908
(1,279)



Voids,
percent
~95

~95

96
94

92

70 to 78



80




28 to 37


_


40




-
_


Surface or volume denltrlflcatlon rate at stated temperature, C
Ib N rem/sf/day x 10* a (ft N rein/1000 cf/day b)
5









.32
(4.5
to
8.8)




















10









.37
(5.3
to
10.1)




















11

































12

































13




.43
(2.8)






















0.6
(16)




14

































15




.50
(3.3)




.26
(3,7
to
7.1)




















16






























_
(257)

17
6.2
(42)



.53
(3.4




























18





























55
(720)



19
18.6
(126)
































20




1.3
(8.5)




.95
(13.5
to
26)

















_
(308)

21




1.1,
(7.2)




























22





























98
(1275)
-
(292)
(388)
23


12.1
(82)
.59
(3.8)




























24






























-
(424)

25









1.1
(15.6
to
30.1)
2.4d
(19)
1.46=
(12)

















26

































27







2.02
(21)









2.3
-




2.9
(73)

4.6
(124)




Other





.93°
(2.5)














3.9
(176)










  1 kg /m2/day = 0.205 Ib/ei/day
b 1 fcgytnVday = 62.4 Ib/1000 CF/day
c 10 - 23C
  1 hour detention
6 2 hour detention
  Estimated from data In publication
8 Prior to bed expansion
  Note:  10 = effective size,  50 = median
     and U.C, = uniformity coefficient
          size,
          O
                = particle size at which 60 percent of the material Is smaller

-------
                                               982m        12.25 m
                                              TYP CELL       TYR CELL
                                                                                               SEAL WftU.
3OOmm DUMPED MEDIA
                                                                                                          WEIR
                                                                                                        METHANOL FLOW RATE
                                                                                                        CONTROLLER
                                                 ADJUSTABLE HANGER RODS
DENITRIFIED EFFLUENT
CHANNELS
DENITRIFIED EFFLUENT
COLLECTION CHAMBER
                    FIGURE 5-7


   DESIGN DETAILS OF NITROGEN GAS FILLED
   DENITRIFICATION COLUMN (REFERENCE 21)
SPLASH PLATES. APPROX
2OOmm ABOVE MEDIA

-------
Media specific surface and configuration must be carefully selected to ensure that clogging
does not occur. Clogging can occur under high loadings if the restrictions in the media are
too small. Pilot studies may be warranted for media selection when design conditions depart
from  those  previously  tested. Once the proper  media is  selected, the process has  the
advantage that backwashing is not required, considerably simplifying design, construction,
and operation. Further, the column construction is simplified since it need not be a pressure
vessel as the column is at approximately atmospheric pressure.

The denitrification column was first tested  at the Central Contra Costa Sanitary District's
Advanced Treatment Test Facility (ATTF)A21 The  test program  had three primary
objectives: the first  was  to develop confidence in the ability of the attached growth reactor
to function consistently and predictably; second, to determine the optimum specific area of
the plastic media; third, to develop criteria upon which design of the prototype columns for
the LMWQCC could be based.

The test denitrification  column consisted of a sealed  vertical 24-inch (610 mm) diameter
reinforced concrete  pipe  12 ft (3.66 m) in height and  filled with  10 ft (3.05 m) of media.
Plastic media consisted of PVC corrugated sheet modules, supplied by Munters Corporation.
The top of the column was sealed by a gasketed cover and the nitrified liquor was applied to
the top of the tower by a round pattern nozzle. After passing through the  column,  the
denitrified effluent  was collected in a  sump from where it was discharged. Provision was
made to allow pumping  of column effluent  back to the influent of the  column to test the
merits  of recirculation. A brief description of the test program  and  a summary of the results
is presented on Table 5-4. The initial media tested had a specific surface of 68 sq ft per cu ft
(223 m^/m^). This media clogged at  the high application rates  applied in May and June
1972 and was subsequently replaced with media having a specific surface of 42 sq ft per cu
ft  (138 m^/m^). Recirculation of effluent was not found to be required and was dropped
from  the test  program  because  of the high  energy costs  that would  be incurred if
recirculation was used in the full-scale plant.

Experience with the pilot column indicated that an application  rate of 5 gpm appeared to be
a reasonable flow that could be applied continuously without an objectionable buildup of
growth on the media. That value is equivalent to a rate of 245 gallons per cubic  foot per day
(33,300 l/m^/day). Because the Contra Costa wastewater has a lower nitrogen concentration
than the wastewater at the LMWQCC, the hydraulic loading rate was adjusted to ensure an
equivalent nitrogen  load was applied. The value was then further reduced to permit lower
loadings  at low wastewater temperature. Taking these  factors into  account, an ADWF
loading rate of 144 gallons per cubic foot per day (12,562 l/m^/day) was selected for design
purposes. In terms of nitrate removal rate on the basis of media surface, this corresponds to
9.9 x 10-4 Ib NC-3-N rem/sq ft/day (4.83 x 10~3 kg/m2/day).

While media with a  specific surface of 42 sq ft per cu ft (12.8  m^/m^) is believed to be  not
susceptible to  clogging in municipal  applications based on the  limited experience to date, it
would  appear prudent  to  provide  chlorine addition  capability  to  aid sloughing should
clogging occur.
                                        5-20

-------
                                             TABLE 5-4
                   SUMMARY OF OPERATION - NITROGEN GAS FILLED DENITRIFICATION COLUMN
IS)
to

Dclt6»
1972
April 24
April 27
May 2
May8
May 11
May 16
May 18
May 23
May 26
May 31
June 1
June 5
June 9
June 12
June 15
June 16
July 27
July 28
August 1
August 4
August 9
August 15
August 21
August 25
August 28
August 30
September 1
September 7
September 12
September 19
September 27
October 3
October 25
Effluent
flow rate,
gpma
4.7
5.3
2.3
2.1
2.0
2.5
2. 5/5. 0
5.0
10.0
10.0
15.0
15.0
10.0
10.0
7.0
_
-
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
Recirculation
rate
gpma
9.4
9.4
9.4
9.4
8.0
0
0
0
0
0
0
0
0
0
0
_
-
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
NOs-N, mg/1

In
27
22
28
30
24.5
28
27
27
-
22
25.5
-
27
27
27
_
-
15.3
13.5
11.8
11.8
9.8
13.2
12.5
23
30
18
27
17
24.5
18
19
20

Out
4.5
1.6
<0.2
<0.4
Nil
1.6
<0. 1
0.6
-
<1
3.4
-
1
5
6.8
_
-
5.4
2.5
2.7
1.0
Nil
4.0
5.0
17
22
3.2
0.5
Nil
Nil
0.8
. Nil
1.4
TOC, mg/1

In
26
29
37
36
40
10
9
8
-
9
_
-
7
8
7.5
_
-
46
10
73
11
-
8
8
8
8
8
-
8
8.5
8
8
6.5

Out
15.5
19
25
11
16
13
57
14
-
16
_
-
15
17
12
_
-
38
10
45
14
-
18
13
12.5
13
13
-
18.5
12
12
15.5
11

Comment

Media installed had 68 sq ft/cu ft. (20. 7 m2/m3)




Eliminate recirculation and increase flow rate.




Column being overloaded.
Severe foaming. Feed rate decreased.



Mechanical breakdowns and operation difficulties.
Test results unreliable.






Media removed.
New Media used was 42 sq. ft/cu ft.
(12. 8 m2/m3)

Denitrification reestablished.





               al gpm - 0.065 1/s

-------
The design of the denitrification column for the LMWQCC is portrayed in Figure 5-7 and
design data are given in Section 9.5.2.2. Operation will be by gravity flow, as the LMWQCC
is located on a very steep site. Fixed distribution troughs with splash plates were used to
handle the wide range in flows expected and to minimize the restrictions in the distribution
system which might be clogged by the growth of  denitrifiers. Further,  a top layer of
dumped media is placed over the corrugated media to ensure good distribution and growth
at the top of the column.

          5.3.2.2. Submerged High Porosity Media Columns — Packed Bed

Submerged denitrification  columns packed with high porosity media  have been piloted at
several locations?. 17,20,22,23 ^d tested full-scale at El Lago, Texas. 24
A typical schematic illustrating essential process elements is shown on Figure 5-8. The media
is normally contained in pressure vessels.  To  obtain  sufficient  contact time, a series
configuration of 2 or 3 vessels is employed. 7,20,24 Either an up flow or downflow column
operation may  be  used.  While  a  variety of media types have been used (Table 5-3), a

                                   FIGURE 5-8

                TYPICAL PROCESS SCHEMATIC FOR SUBMERGED
                        HIGH POROSITY MEDIA COLUMNS
TO STORAGE
M WAItK
XGE
.•NITRIFICATION
COLUMN
IPED
)IA




^
s
'

»




h



H
tFFLU
CLARI
FILTR
DENITRIFIC
COLUMN
    METHANOL-
       INFLUENT
        PUMP
                                                                       CLARIFIED
                                                                       BACKWASH
                                                                       WATER
                                                                 BACKWASH
                                                                 PUMP
                                       5-22

-------
common  characteristic  is  that a high  void  volume is maintained in  the  unit. As  a
consequence, biomass is allowed to continuously slough from the media, minimizing the
requirements for backwashing. A corollary is that the media does not build up the layers of
biomass that would develop if the void fraction were smaller such as with sand or pebble
media.23 The lower surface denitrification rates for this type of media compared to sand or
rock  (Table  5-3)  reflect  this  difference  in attached biomass development.  Most often
dumped media have been used, though there is one instance when corrugated sheet media
has been tried. 23

Backwashing, though infrequently used, is still required. At El Lago, Texas, where the media
was Koch Flexirings, the water backwash rate was 10 gpm/sf (13.5  l/s/m^) coupled with an
air backwash rate  of  10 cfm/sf (3.6 m3/m2/min). Backwashing was routinely done every
four weeks.24 Backwashing in this type of column is not required due to excessive head
losses in the column; rather, it is required to prevent the accumulated solids in the column
from  continuously sloughing into the effluent and causing high effluent suspended solids.
Others have  used  backwash rates up  to 44 gpm/sf (29.7 l/s/m^)  but did not use  an air
backwash procedure.7.20 The El Lago air-water backwash procedure is the recommended
approach  for design  purposes. As opposed to  the situation with respect to other column
designs, a fairly broad data base exists for this column type.  Surface removal rates observed
at various locations are summarized in Figure 5-9. As may be seen, most data points fit the
data correlation of Sutton, et  al. 20fOr Intalox saddles.

Figure 5-9 may be used to size the denitrification column. First, peak diurnal nitrate loading
and minimum wastewater temperature must be known. From Figure 5-9, surface removal
rate can  be determined. Then from the loading, the media surface area can be calculated.
Finally, a specific media is selected and column volume requirements are calculated.

    5.3.2.3 Submerged Low Porosity Fine Media Columns — Packed Bed Configuration

The submerged low porosity column using fine media (Table 5-3) is the column system
seeing widest commercial  application at the present time. One manufacturer's concept
(Dravo) of how to incorporate this type of column into a treatment plant is shown in Figure
5-10.27 jn this flowsheet combined carbon oxidation-nitrification is accomplished in an
activated  sludge step.  Of  course, other nitrification processes could  be employed  for
producing a nitrified  effluent.  Clarified nitrified effluent  then flows to the denitrifica-
tion column. The concept  employed in this flowsheet is that the column combines two
functions in one.  First, it serves the purpose of denitrifying  the wastewater; second,  the
column serves the purpose  of effluent filtration that normally would be required in many
plants anyway.27 A discussion of the cost-effectiveness of combining the denitrification and
filtration functions is presented at the end of this section.

The units  manufactured by the Dravo Corp. typically consist of 6 ft (1.83 m) of uniformly
graded sand 2 to 4 mm in size. Filtration rates normally recommended by the Dravo Corp.
when  removing 20 mg/1 NO3  —N from municipal wastewaters are 2.5 and 1.0 gpm/sf for

                                        5-23

-------
                                  FIGURE 5-9
 I
 i
 ki
5
CD

Uj"
K,
$
-j

o
Uj
Uj
<->
 Cfc
 ^
 CO
              SURFACE DENITRIFICATION RATE FOR SUBMERGED
                       HIGH POROSITY MEDIA COLUMNS
    2.5
    2.0
    1.5
    1.0
    0.5
         KEY
                              I
                                                                I
SYMBOL
V
~ •
A
•
LOCATION
Davis, Ca.
Hamilton,
Ontario
Firebaugh,
Ca.
El Logo,
Texas
REF.
17,
25
7,
20
22
24
MEDIA
Raschig
rings
Intalox
saddles
Koch
Flexirings
Koch
Flexirings
SPECIFIC
SURFACE
SF/CU FT*
79
142 to
274
65
105
VOIDS,
PERCENT
80
70 to 78
96
92
         '* 1 SF/Cu Ft = 3.28 m2/™3
                         <35 % CONFIDENCE LEVEL-
                         FOR ONTARIO DATA
           WEIGHTED LEAST  SQUARES FIT-
           FOR  ONTARIO DATA (Refs. 7 and 20)
           K= 90.46
                                       I kg/m2/day =0.205  Ib/sf/day

                                          I	I	I
       O          5           10          15         20          25         3O
                              TEMPERATURE, C

minimum wastewater temperatures of 21 C and 10 C respectively. 31

The procedure for backwashing the Dravo filter begins with one or two minutes of air
agitation followed by 10 to 15 minutes of air and water scouring and finally five minutes of
water  rinse. Air and water  backwash rates  recommended  by Dravo are  6 cfm/sf (1.83
m^/m^/min) and 8 gpm/sf(5.41 l/s/m^), respectively.-* 1 In addition, it has been found that
nitrogen gas accumulates in the filter during a filter run. This imposes a loss of head on the
                                     5-24

-------
                                   FIGURE 5-10


           NITRIFICATION-DENITRIFICATION FLOW SHEET UTILIZING
            LOW POROSITY FINE MEDIA IN COLUMNS (REFERENCE 27)
                     ORGANIC
                     CARBON
                                  FEED PUMPS
    COMMINUTOR
INFLUENT
0-n
                        BACKWASH RETURN
-cn
BLOWER
                  00 OOOpO
            oo
           O O  AERATION
          O O O  0TANK
           °n  °  o°
          O °
                              I  \
                                   CLARIFIER
                         ACTIVATED
                       SLUDGE RETURN
 WASTE SLUDGE
          Q-o  n O O .
          O DIGESTER OR>
            w SLUDGE  o
          O  o HOLDING  H
              TANK  f
                         -^. SUPERNATANT
                           TO INFLUENT
                 WASTE
                SLUDGE
                                                 BACKWASH
                                                   WELL
-QH
                                                     n
1                                                  BACKWASH
                                                 RETURN PUMP
 DENITRI-
 FICA.T.IQN
 COLUMN
              COLUMN
             BACKWASH
                                                                 CHLORINE
                                                         ° CHLORINE
                                                         i° CONTACT
                                                          "'  TANK
                                                                           EFFLUENT
                                                                   REAERATION
                                                                       AIR
                                       BACKWASH   BACKWASH
                                        BLOWER      PUMPS
filter and requires periodic removal of the trapped gas bubbles in the media. A "bumping"
procedure was evolved whereby a filter was taken out of service for what amounts to a short
backwash cycle. In the Dravo design, "bumping" backwash rates of 8 to 16 gpm/sf (5.4 to
10.8 1/s/m^) for one or two minutes is required every four to twelve hours.27


The period  between regular backwashes  in the Dravo filter  is  dependent on the rate of
headloss buildup. At  El  Lago,  Texas it was  found that daily  backwashing  was  required
whereas in  a pilot study  in Tampa, Florida the time between  regular backwashes ranged
from 4 to 40 days.24,32 Durmg the Tampa study, the air backwashing was found to cause a
temporary  partial inhibition of  denitrification that was not present  when  only a water
backwash was used. For  instance, with an influent nitrate-nitrogen level of 15 mg/1, the
effluent nitrate nitrogen level was 10 mg/1 one-half hour after the air-water backwashing and
reached 0 mg/1 seven and one-half hours after back washing. 3 2 in multiple filter installations,
the effluents from the recently backwashed filter would be blended with other normally
operating filters,  so the impact of this nitrate leakage would be  expected to be moderated.
Generally speaking, even the smallest of plants will require multiple filters so that an
effluent can  be  continously produced, otherwise a filter influent  storage  basin will be
required.
                                       5-25

-------
Neptune-Microfloc, Inc.  has made  available suggested design guidelines for their media
designs.33 Four media designs were tested on a nitrified effluent containing 20-30 mg/1 of
NOj-N from an extended aeration plant. Best overall performance was obtained from the
two media designs shown in Table 5-5. Basic conclusions of the study were as follows:

    "Utilizing a 36-inch (0.92m) mixed-media filter  (F-III), essentially complete
    denitrification of a highly nitfified wastewater can be achieved at filtration rates
    of  1.5  gpm/sq  ft (l.Ol/s/m^)  for temperatures of  10  C,  and  at 3  gpm/sf
    (2.01/s/m2) at temperatures of 20 C. The methanol to nitrate nitrogen ratio was
    found to be between 2.0 and 2.5. At applied nitrate nitrogen concentrations of 10
    mg/1, filter run times between 16 and 24 hours to 8 feet of headless were realized
    at a filtration rate of 3 gpm/sq ft (2.01/s/m2). At  higher applied nitrogen levels,
    filter runs were  reduced in direct relation to nitrogen concentration."^

In another study of a Neptune-Microfloc filter, a fully nitrified effluent from a trickling
filter was denitrified.  No attempt was made to determine limiting filter loadings, however.


                                   TABLE 5-5
                     NEPTUNE-MICROFLOC MEDIA DESIGNS
                    FOR DENITRIFICATION (REFERENCE 33)
      Filter Material
                                               Layer depths, inches (cm)
  F-n
F-III
      Garnet Sand
      d!0  =  0.27 mma

      Silica Sand
      d!0  ~  0.5 mma

      Anthracite Coal
      d!0  =  1.05 mma

      Anthracite Coal
      d!0  =  1.75 mma
   3
 (7.6)
   9
(22. 9)
  18
(45.1)
   3
 (7.6)
   9
(22. 9)
   8
(20. 3)
                    16
                  (40.6)
        10 = effective size
                                      5-26

-------
The filter had  three layers of media: anthracite, silica sand and garnet sand with sizes
ranging from 1.2 mm at the top to 0.2 mm on the bottom. At a surface application rate of
2.3.gpm/sf (1.61/s/m2) and influent nitrate levels averaging 8 to 9 mg/1, better than  95
percent removals were obtained. Operating temperature ranged from 16 to 18 C during this
study. 34,35

Media size is important in establishing denitrification column requirements. The relationship
between specific surface and column size was established in a pilot study at Lebanon, Ohio
using the following media sizes for three  columns: 3.4 mm, 5.9 mm, and 14.5 mmJ^
Biological film development per unit surface area was shown to be approximately the same
for each size media. Therefore, the smaller the media, the higher the media surface per unit
volume and the  smaller the column as shown in Figure 5-1 1 .
                                  FIGURE 5-1 1
          COLUMN DEPTH VS SPECIFIC SURFACE AREA (REFERENCE 18)
     Q
     iu o
     - §
     Z3 LU
     O 
-------
Care should be exercised in the design of the column underdrain system. Neptune-Microfloc
recommends that an extremely open underdrain system be employed (pipe lateral, Leopold
tile, etc.) to avoid the very real possibility that an overfeed of methanol will cause denitrifier
growth  to clog the  underdrain  as  was  experienced  in  one test with a nozzle-type
underdrain.33

Since it has been proposed to use this type of column for both filtration of suspended solids
and nitrate removal, it would be well  to examine the performance of the columnar system
for suspended solids removal. Various observations of suspended solids removal are shown in
Table 5-6. With the exception of the El Lago, Texas data, the performance of these columns
as tertiary filters falls within  the range  normally expected  for  tertiary filtration (for
comparative data see Section  9.3.2.3  of the Process Design  Manual for Suspended Solids
Removal,  an EPA Technology Transfer publication).36 Suspended solids removals will be
affected by filter design and the fact that the filter is operating as a biological treatment
system as opposed to a purely physical separation process.

In considering the use of this process  as both an effluent filter and a denitrification system,
an important design factor should be borne in mind that has considerable implications on
                                   TABLE 5-6

          COMPARISON OF SUSPENDED SOLIDS REMOVAL EFFICIENCY
          FOR SUBMERGED FINE MEDIA DENITRIFICATION COLUMNS
Location
El Lago, Texas

North Huntington
Township, Pa
Tampa, Fla.

Lebanon, Ohio





Corvallis, Or.




Midland, Ml.
Media type
Dravo
dgQ = 3 to 4mma
Dravo
d10 = 2.9mm
Dravo
d!0 = 2.9 mm
dgo = 3. 4mma

d50 = 5. 9mma

d50 = 14. 5mma

Neptune
Media F-H b
Media F-III b


Neptune
Reference
24

27,28,31

27,31

18

18

18

33

34,35



Surface
loading
gal/min/sf
(l/s/m2)
6.27
(4.23)
0.72
(0.49)
2.5
(1.70)
7.0
(4.75)
7.0
(4.75)
7.0
(4. 75)
3.0
(2. 04)
3.0
(2. 04)
*j.5
(1.70)
Depth,
ft
(m)
13
(4)
6.0
(1.8)
a

10
(3.1)
20
(6.1)
20
(6.1)
2.5
(0.76)
3.0
(0.91)
5.0
(1.50)
Influent
SS,
mg/1
37

16

20

13

13

13

25-65

25-65

13-30

Effluent
• SS,
mg/1
17

7

5

4

2

1

8

4

2-10

SS removal
efficiency,
percent
54

56

75

69

85

92

68-88

84-93

67-93

 uniformly graded

J Table 5-5
                                       5-28

-------
cost.  It has  been claimed that combining the functions  of filtration  and denitrification
reduces tankage  and equipment requirements and therefore yields cost  savings in plants
requiring filtration.27 However, it should be recognized that the column loading criteria are
different for the  functions of filtration and nitrogen removal. For effluent filtration, fairly
high hydraulic loadings can be  applied (4 to 6 gpm/sf or 2.7 to 4.1 l/s/m^).  However, for
filters 3 to 6 ft (0.9 to  1.8 m) deep acting as denitrification  columns, available data indicates
that hydraulic loading should be between 0.5 to 1.5 gpm/sf (0.34 to 1.02 l/s/m-2) at a
wastewater temperature of 10 C. Thus, to accomplish denitrification at IOC, it would  be
necessary  to have column surface areas five times as large as required for filtration alone.
Thus, an economic analysis must be  done in  each case to determine the most economic
process configuration.

          5.3.2.4 Submerged High Porosity Fine Media Columns — Fluidized Bed

The introduction of fluidized bed technology into the field of columnar denitrification is a
comparatively recent development.29,30,37,38 Figure 5-12 depicts a typical fluidized bed
reactor with its  ancillary  facilities. In the fluidized bed  unit wastewater passes upwards
vertically  through a bed of small media such as  activated carbon  or sand at a sufficient
velocity to cause motion  or fluidization of the media.  The small  media provides a large
surface for growth of denitrifiers.

High surface application rates were recently employed  in a pilot study  of the process at
Nassau County, New York (15  gpm/sf or 10.2 l/s/m2).38  Tne column had a fluidized bed
depth of 12 ft (3.7 m). The bed settled to about 6 ft (1.8 m) when the influent was shut off

                                     FIGURE 5-12

                    FLUIDIZED BED DENITRIFICATION SYSTEM
               FLUIDIZED     SAND
                 BED       SEPARATION
               REACTOR      TANK
                        BIOMASS
                    SEDIMENTATION
                         TANK
     FLUIDIZED
      MEDIA 	
     PEA GRAVEL^
METHANOL
                          y_u   I
    SAND
    CLEANI-NG
    AND RETURN
    PUMP
                                                 J
INFLUENT
 PUMP
WASTE SLUDGE
 TO DISPOSAL
                                                         /   V
                                      u
                                                                          DENITRIFIED
                                                                          EFFLUENT
                                        5-29

-------
so bed expansion during operation was 100 percent. Initially the clean bed in the column
contained 1.5 in. (38 mm) of pea gravel and 3 ft (0.91  m) of silica sand with an effective
size of 0.6 mm  and a  uniformity  coefficient of 1.5. During operation,  the media became
completely covered  with denitrifier growth and  the  individual particles  grew in size.
During the initial lab-scale test for  this  process, the media grew from 0.65 mm particles, to
particles 3 to 4 mm in  size.™ The  attached growth accounts for the greater depth of media
in the non-fluidized bed after the column had been in operation.

In a packed bed, this growth of particle size would result in high headloss, channeling, and a
loss in efficiency. In an expanded bed, however,  there are sufficient voids between the
denitrifier-sand particles to provide good liquid contact  at modest headlosses.  This greater
biological film development allows higher surface reaction rates (expressed per unit of media
surface) than for any other type of column configuration  as shown in Table 5-3. Since the
surface contained in a  unit volume is high, higher volumetric  loadings are also possible as
compared to  any other column configuration (Table 5-3). Empty bed detention time during
the recent pilot test at Nassau County was only 6.5 min. Maximum nitrate removal rates as a
function  of temperature  are shown in  Figure 5-13 and are based on the Nassau County
data.39  if diurnal variations in nitrogen  load are to  be accommodated by the column
without nitrate bleedthrough, then column volume requirements will be greater than that
determined in Figure 5-13. A provisional recommendation would be to increase the reactor
requirements  by  the ratio of the peak to average nitrogen loads.

The process has been shown to be responsive to both  diurnal load variations and to cold
temperature operation.29 Methanol feed was not automatically  controlled, so periods of
nitrate bleedthrough occurred.29  When methanol feed  was under control, 99 percent
removal of influent nitrate and nitrite  was demonstrated. Total nitrogen reductions were not
given. 29,38

While backwashing facilities are not required in  this type of column, facilities must be
provided for  managing the column media  inventory. During operation,  the denitrification
column increases in depth due to biological growth causing a continuous small loss of media
from  the system. Further,  diurnal  flow  variations  cause height variations which may
contribute to media loss. This loss can be  minimized  by provision of flow equalization
facilities  (see Chapter 3, Flow Equalization, Process Design Manual for Upgrading Existing
Wastewater Treatment Plants, an EPA Technology Transfer Publication).40

The  manufacturer of the system  suggests that for most plants subject to diurnal flow
variations, media losses and effluent solids  levels can be controlled by placing two tanks in
series with the column  as shown in Figure  5-12. The first tank would be a sand separation
tank followed by a biomass sedimentation tank  for biological  solids removal. The sand
separation tank might be very small, as a tank with an overflow rate of 13,600 gpd/sf (554
m-Vm^/day)  served satisfactorily during the pilot study. It has been suggested by Ecolotrol,
the manufacturer of the fluidized bed system, that  the  swirl concentrator which was
developed for grit removal from combined stormwater  and  wastewaters could serve as the

                                        5-30

-------
                                 FIGURE 5-13

              VOLUME DENITRIFICATION RATE FOR SUBMERGED
            HIGH POROSITY FINE MEDIA COLUMNS (REFERENCE 39)
     24OO
  Q

  I-
  U.
§
O
\
^
co
     2000
      I6OO
  Ui
  t-
  2  I2OO
  2
  O
O
u!
(t
  Ul
  Q
  Ul
  5
       BOO
     4OO
                                                 I
                                                            I
                                     I
                                                I
                       10
                                  15          2O
                                 TEMPERATURE,  C
25
30
sand separation device.^^'^2 The media settling in  the sand separation  tank would be
pumped back to the column with the:pumping action shearing the denitrifiers from the
media. This sheared  biomass would pass through the column and sand separating tank and
then settle in the biomass sedimentation tank.29 if very low levels of suspended solids are
required, the system  would have to be followed with tertiary filtration.
                                    5-31

-------
The manufacturer's cost estimate indicates that the fluidized bed system is competitive with
suspended  growth or packed bed columnar systems, but it was also stated that these must be
confirmed by larger scale tests on the fluidized bed system.29

         5.3.2.5 Comparison of Attached Growth Denitrification Systems

The  three  column systems previously described in Sections 5.3.2.1, 5.3.2.2, and 5.3.2.3 are
seeing commercial applications  at  this time and  the  fluidized bed system described in
Section 5.3.2.4 will likely see application in the near future. Therefore, the design engineer
has four alternative column systems to consider.

Where low treatment plant effluent solids are required, tertiary filtration will have to follow
all column systems except the low porosity fine media system described in Section 5.3.2.3.
In small plants, the elimination of a unit process may favor combining the denitrification
and filtration functions. In larger plants the cost trade-offs between alternatives need to be
considered.

Where space restrictions exist at a plant site, there is an  incentive to pick those systems
requiring the least land area possible. While from Table 5-3 it might appear that the fluidized
bed had the distinct advantage  because of highest volumetric loading rates, the ancillary
sand separation and  biomass sedimentation tanks diminish its advantage over the submerged
fine media column and the nitrogen gas filled column.

The  submerged high porosity  column configuration appears to offer the least attractive
alternative. Both surface and volumetric removal rates are low, requiring comparatively large
reactors (Table  5-3). Further,  the unit  must incorporate the design  features of a  filter,
without  having the advantage of low effluent suspended solids of  the submerged low
porosity columns. The  system has an  advantage for small treatment  plants in  that
backwashing is only infrequently required and can  be scheduled to coincide with plant staff
availability.

The advantages of the  nitrogen gas filled column are: (1) similar space  requirements to low
porosity submerged  columns (2) column walls need not be designed to handle hydrostatic
loads and (3)  with proper media  selection, sloughing occurs naturally and backwashing is
not required.

5.4 Methanol Handling, Storage,  Feed Control, and  Excess Methanol Removal

Methanol is a chemical not normally dealt with in wastewater treatment plant operation and
care must be exerted in the design and operation of methanol handling, storage and feeding
facilities to ensure its safe  and proper use.
                                         5-32

-------
     5.4.1 Properties of Methanol
Methanol, CH3OH, has a variety of names such as methyl alcohol, carbinol and wood
alcohol and is normally supplied pure (99.90 percent). It is a colorless liquid, noncorrosive
(except to aluminum  and lead)  at normal  atmospheric  temperature. Some important
properties of methanol are shown  in Table 5-7. Additional data is available in references 43
and 44 and manufacturer's information.

                                  TABLE 5-7

                         PROPERTIES OF METHANOL
             Property
               Value
    Density

    Vapor Density (air = 1. 00)

    Vapor Pressure     0 C
                      10 C
                      20 C
                      30 C
                      40 C
                      50 C

    Solubility

    Viscosity @ 20 C

    Combustible Limits, percent
      by volume in air at STP

    Flash Point Tag Open Cup
                Tag Closed Cup
 0. 7913 g/ml @ 20C (6. 59 Ib/gal)

 1.11

 29 mm Hg
 52 mm Hg
 96 mm Hg
159 mm Hg
258 mm Hg
410 mm Hg

Miscible in all proportions with
  water
 0.614 cps


 7.3 to 36

16 C (61  F)
12 C (54  F)
Taken internally, methanol is highly toxic. It is harmful if the vapors are inhaled or if skin
contact by liquid or vapors is prolonged or repeated. Fire and explosion are primary dangers
of methanol. Persons involved  in handling methanol  should be aware of these hazards.
Federal, state and local regulations for safety should be posted along with information from
references 43, 45 and manufacturer's data.

     5.4.2 Standards for Shipping, Unloading, Storage and Handling

The  shipping, unloading, storage  and  handling  of any flammable  chemical  including
methanol is governed by a multitude of stringent regulations which include: Federal, such as
                                       5-33

-------
the Department of Transportation  (DOT)  and the Occupational Safety and Health Act
(OSHA);  State,  which  has  various  safety  orders  and codes; municipal  ordinances;
independent associations such as the National Fire Protection Association (NFPA) and the
Manufacturing Chemists Association (MCA); and insurance requirements. It is necessary that
all of these regulations be reviewed and studied before the design of any methanol facilities,
and all such regulations must be followed.

     5.4.3 Methanol Delivery and Unloading

Methanol may be received in 55 gallon (208 1) metal drums, tank wagon, tank truck or tank
cars. Other methods of shipping, not discussed herein, are by barge, metal drums (less than
55 gal) and glass and metal cans. Tank wagons are normally 1,000 to 4,000 gal (3,785 to
15,142 1) in size, tank trucks range from 4,000 to 9,000 gal (15,142 to 34,069 1) while tank
cars are shipped in 6,000, 8,000 and 10,000 gal (22,713, 30,283, 37,854 1) capacities. Tank
cars and tank trucks are the most economical shipping mode for most plants. However, for
pilot  work and small plants, 55 gallon metal drums may be feasible.  Since  methanol is
classified as a flammable liquid by  the DOT, all shipping containers must be approved and
labeled in accordance with applicable DOT regulations.

The recommended method of unloading methanol from any container is by pumping. Some
barges and tank wagons have their own pumps for unloading. Tank cars and trucks can be
unloaded from the top or bottom and be pumped or conveyed by gravity or syphoning. The
preferred method of unloading is pumping from the top via an eductor tube. Syphoning and
gravity unloading are  only permitted when the top of the storage vessel is below the bottom
of the shipping container. Due to the increased spillage probability using bottom unloading,
it is only permitted on cars or trucks approved for bottom unloading which include valving
approved by the Association of American Railroads (AAR) and in agreement with the DOT
requirements. This valving helps contain the product  by safely controlling flow. Additional
precautions such as fusible link valves and excess flow valves may be used.

Air pressurization of the tank ("air padding") must never be used for methanol unloading.
However, top unloading using the  water  displacement method or inert  gas padding, i.e.
carbon dioxide, nitrogen, etc.,  may be used if the exact unloading procedures are as provided
by the chemical manufacturer. Unloading procedures can  be found in references 43 and 45
and ,the supplier's data.

General requirements for the design of unloading facilities for methanol  are applicable to
both tank car and truck. The unloading area should be arranged to avoid traffic areas. Also,
all facilities should be outside  due  to the fire hazard and all equipment in the vapor area
must  be explosion-proof, Class  I, Group D, Division 1 or 2 per the National Electrical Code.
Tools  should be "non-sparking." Unloading should occur during daylight hours since the
safety and  lighting   requirements  for night operation  are very  extensive.  Ample fire
extinguishers, safety blankets, deluge showers, eye washes, no smoking signs and unloading
signs are also required.

                                        5-34

-------
If top unloading will be practiced, approach platforms are required for access to the top of
the tank.46 The approach platforms used at the CCCSD Water Reclamation Plant are steel,
swinging type with  pneumatic operated drawbridges. These drawbridges provide for three
feet (0.9 m) of horizontal adjustment, 30 in. (76 cm) of vertical adjustment and a 45 degree
pivot to either side for ease in unloading both tank cars and trucks. The drawbridges fold to
the platforms when not in use providing the required railroad clearances.

In all unloading  setups, all equipment must be grounded. This includes the shipping vessel,
interconnecting  piping, pumps, approach platforms, etc. Also, bonding jumpers must be
used to provide a good continuous system. Periodic checks of the grounding system must be
made.

Static  electricity buildup  must be minimized since it is  not dangerous until at the spark
discharge level. A  spark discharge can easily start a fire or cause an explosion. Refer to
reference 47 for static electricity accumulation prevention.

Facilities for truck unloading must provide for ease in truck maneuvering, both entering and
leaving the area. Consideration must be given to the number of individual unloading spots
regarding frequency  of use, space and simultaneous unloading. For ease in truck traffic, it is
best to have parallel unloading areas for straight-through driving.

Rail unloading creates additional considerations. Unloading areas must  have derails or a
closed switch a minimum  of one car length from the car. A primary concern is who spots
the cars, the plant or the railroad. It is preferable to have private sidings so the railroad can
drop off or pick up  cars at any time without disrupting plant operations. Also, the cost of
having the  railroad spot cars at night, over weekends or holidays is high and because the
railroad cannot guarantee time of shipment, safety provisions and lighting must be provided
for night  operations. Two sidings, one for empty cars and one for full cars should be
provided. The cars can be spotted by the plant personnel with rail car movers which  operate
both on rail or streets or in the case of short distances, car spotters (winches) can be used.

By paving the railroad yard  area, both truck and rail  unloading  can  be  practiced. This is
advantageous because major  strikes affecting either kind of transport cannot cripple the
plant.  At the CCCSD Water  Reclamation Plant (described in Section 9.5.2.1), two sidings
are provided with space for ten cars on  each track for storage. Unloading sidings with two
platforms and two bottom stations will  be used for simultaneous spotting and unloading. A
rail car mover with a capacity of two cars is also provided.

Unloading equipment is normally steel but many other materials, except aluminum, are
acceptable providing they can withstand the pressure and are completely grounded. Pumps must
be "non-sparking"  such as bronze fitted steel pumps with bronze impellers. Many materials
are compatible with  methanol so seals and gaskets can be common materials. Pumps may be
either centrifugal or  positive displacement gear type. However, positive displacement pumps
must   have  relief  valves.  Because  of  the  widely varying  heads  encountered  during

                                        5-35

-------
unloading,  care  must be  taken  in  pump selection.  Piping should  have  as  few joints as
possible and should be Schedule 40 minimum. Splash guards at joints may be desired in traffic
areas. Valves may be gate, plug or diaphragm, iron or steel with bronze trim, and neoprene
plugs in the plug valve or a neoprene disc in the diaphragm valves. Refineries on the West
Coast have adopted  a standard of cast steel valves on all  flammable materials to  prevent
damage during a fire. Couplings must be leak tight and it is preferable to have a valve next to
the coupling to limit material leakage  and waste  during disconnection. If flexible hose
connections are  used, a coupling with an integral valve can be used. A strainer should be
used ahead of any pumping or storage equipment.

Care  must be exercised to not overflow the storage  vessel. A high level alarm and pump
shutoff should be utilized. Due to the increasing cost of methanol, it may be desirable to
have  a flowmeter  in the unloading piping. All vessels must be vented during unloading or
loading.

     5.4.4 Methanol Storage

In order to provide for possible  delays in methanol delivery, a  storage capability of two to
four weeks supply is recommended. The volume of storage will be determined by various
site and cost requirements; however, storage of less than two weeks is too short for expected
delivery delays and strikes. Tank truck deliveries require in-plant storage. However, with rail
deliveries, the rail cars can be used for storage, but charges (demurrage) are levied by the
carriers for time on site in excess of a fixed time. For small plants, demurrage may become
cost effective. However, carriers may have a time limit on cars or have excessive demurrage
charges.

Methanol  may be stored in vertical or horizontal tanks above  ground inside or outside, or
buried.  It  is strongly recommended  that all  methanol  equipment  and  tanks be located
outside. If interior storage is required refer to reference 43 and 44 for detailed requirements.
An exception  to this rule is drum storage which, if not stored indoors, must be shaded from
direct sunlight or constantly sprinkled with water.

Layout of methanol tanks should be in accordance with reference 48. There should also be a
dike around each aboveground tank or group of tanks to contain 125 percent of the largest
tank volume in case of rupture or fire. If the tanks are not of steel, care must be taken so that  a
fire  will not cause a rupture in the group of tanks thereby overflowing  the  dike. Fire
protection is very critical, especially  when  the tanks are near other structures. For large
volumes of methanol storage, low expansion alcohol-type foam is used for fire extinguish-
ing.  For very  small fires,  dry chemical or  carbon dioxide extinguishers can be used.
Rate-of-rise or ultraviolet detectors  may be used for sensing of fire and initiating automatic
foam release. Water should not be used, but may be used for plant area fire control.

Storage tanks  are normally of steel, but most  materials  are satisfactory except for aluminum.
Tank size is only dependent upon the capacities required and any size limitations imposed

                                        5-36

-------
by the tank material. Piping, valves,  etc. should be as described in Section 5.4.3. Ta'nk
fittings should include the following: (1)  an inlet with dip tube to prevent splash and static
electricity, (2) an anti-siphon valve or hole on the inlet to prevent back siphonage, (3) vent
pipe with pressure-vacuum relief valve^S with flame arrester, (4) an outlet connection, (5) a
drain connection, and (6) various openings for depth gauges, sample points, level switches,
etc. Manholes for access should also be provided. Also, extreme corrosion will take place if
the tank is drained  dry. The tank must  also be grounded. Due to increasing air  pollution
requirements, venting must be controlled by conservation type vents or by  maintaining a
slight negative pressure in the tank using a small ejector.

To maintain a correct inventory of tank contents, a diaphragm level sensor or float should
be used. Low and high  level alarms are  needed for protection against  overfill and settled
material at the bottom of the tank.  The high level alarm should be separate  from the tank
sensors for a fail-safe design.

     5.4.5 Transfer and Feed

Methanol  must be transferred and controlled from the storage vessel  to the point of feed.
Methanol is removed  from the tank and is fed by gravity or pumps. Normally, pumping will
be utilized  for ease of control. The transfer pumps should always have positive suction
pressure and should  be protected by  a  strainer. As with all methanol situations, it is
desirable to  mount  all equipment outside. There  are  three basic pumping  arrangements
which can be used:  (1) diaphragm chemical feed pumps using an  adjustable stroke for
volume control; (2) positive displacement pumps with variable speed drives  controlled by
either counting  revolutions to obtain flow or using a  flow meter; (3) centrifugal  or
regenerative  turbine pumps with variable speed drives  controlled by a flow meter. Each
arrangement has its own particular problems and must be studied for each installation. To
cover the widest range of feed rates,  arrangements  1 and 2 are  used due to the limited
accuracy range of flow meters.

All pumps, piping, etc.,  should be the same as noted in Section 5.4.3. All piping should be
tested for  1.5  times the maximum system  pressure for 30 minutes with  zero leakage.
Methanol  addition to the  denitrification process  is relatively  simple. In  the  CCCSD's
Advanced Treatment Test Facility, methanol was pumped into the  influent line ahead of the
denitrification reactor and the stirring of  the reactor was sufficient mixing. In the CCCSD
Water Reclamation Plant, a multi-orifice diffuser is used to evenly distribute methanol in the
channel ahead of denitrification.

     5.4.6 Methanol Feed Control

Because methanol is expensive and a methanol overdose can result in a high effluent BOD5,
it  is essential to accurately pace methanol  with oxidized nitrogen  load. Simply pacing
methanol dose against plant flow is inaccurate as it does not account for daily and diurnal
variations  in the  nitrate concentration.  Feed  forward control  utilizing plant flow and

                                        5-37

-------
nitrification effluent nitrate nitrogen is shown in Figure 5-14. Feed ratio is approximately
three parts methanol per part of nitrate nitrogen by weight (see Section 3.3.2). This method
requires continuous on-line measurement  of nitrate utilizing an automated wet chemistry
analyzer. The wet chemistry analyzer (AIT) output is proportional to nitrate concentration
in the  nitrification  effluent. The manual  control station (HIK) provides means  to select
either the analyzer  output or to enter a  manual concentration value in case of analyzer
failure. The output  of HIK is multiplied  by a signal proportional to flow from  the ratio
stations (FFIK) to obtain a signal proportional to required methanol flow ratio. This signal
may then be fed to  a chemical proportioning pump, as shown, or may be the setpoint of a
flow control  loop. FFIK provides means  to  adjust  the methanol feed  ratio.  The
dependability of this control procedure is predicated on the reliability of the automated wet
chemical analyzer. These analyzers require  very careful routine maintenance and calibration.
In a research laboratory environment methanol was paced with the use of a Technicon Auto
Analyzer for one week periods between maintenance checks. ^
                                    FIGURE 5-14

                     FEEDFORWARD CONTROL OF METHANOL
                    BASED ON FLOW AND NITRATE NITROGEN
 NITRIFICATION
 EFFLUENT
                                                               KEY
                                                               FT = Flow Transmitter
                                                             FFIK = Ratio Station
                                                              AIT = Wet Chemical
                                                                    Analyzer
                                                              HIK = Auto/Manual
                                                                    Control Station
                                                                X = Analog Multiplier
                                                          CH3OH
  TO
DENITRIFICATION
  TANKS
     5.4.7 Excess Methanol Removal
Unless specific measures are taken to provide for methanol removal, methanol addition
above  stoichiometric  requirements (Section  3.3.2) will cause methanol to appear in the
denitrification  process effluent.3>4,7, 15,16,27,29,32,49,50  in  one instance, a methanol
overdose caused an effluent BODs of 106 mg/1.3 Placing total reliance on the methanol feed
control system to prevent methanol overdoses may be unrealistic in small plants where a
trained technician's attention can be expected to be infrequent. The provision of a good
methanol control system and a methanol removal system as backup  should  allow nearly
fail-safe operation in terms of preventing effluents containing high levels of organics.
                                        5-38

-------
A recent modification of the suspended growth denitrification process,that has as one ofits
objectives preventing methanol bleedthrough,is shown in Part B of Figure 5-1.2>->l After
denitrification, mixed liquor passes to an aerated stabilization tank. In this tank, facultative
organisms "switch over" from using nitrate to dissolved oxygen and oxidize any remaining
methanol. While refinements undoubtedly could be made in determining the length of time
required for the facultative bacteria to switch over and complete methanol oxidation, it is
known that 30 minutes of aeration in an aerated stabilization tank at a Burlington, Ontario
pilot plant was insufficient,  as high effluent methanol values were periodically observed in
the system.^2 A period of about 48 minutes has been found  to be sufficient for methanol
oxidation.3 Therefore, a period of one hour aeration is recommended based on experience
to  date. Further details of this  modification  concerning solids-liquid  separation  are
presented in Section  5.6.

In attached growth  denitrification systems, the  provision of an aerated  basin after the
denitrification column would not ensure oxidation of excess methanol. This is because there
is an insufficient mass of facultative organisms in the column effluent to accomplish the
biological oxidation  of the carbon, since  the denitrifying organisms are retained on the
media and  only, a few pass  into  the column  effluent.  To  date,  excess methanol removal
systems applicable to attached growth systems have not been developed.

5.5 Combined Carbon Oxidation-Nitrification-Denitrification Systems with Wastewater and
    Endogenous Carbon Sources

The methanol price increases experienced during late 1973 have caused renewed interest in
alternative  carbon sources. Alternatives having the least chemical cost for nitrate reduction
are the organics present in domestic wastewater or endogenous respiration of the biological
sludge. The problems experienced in the past with these sources are  lower denitrification
rates  and  contamination of the effluent  with the ammonia released when wastewater
organics  or biological sludge serve as the carbon source for denitrification. The former
problem  has been mitigated by increasing  reactor detention  time. The latter problem has
been addressed by adapting  the suspended growth process configuration in specific ways to
expose the sludge  to alternating aerobic and anoxic environments so that released ammonia
is subjected to nitrification  and subsequent denitrification. The alternatives which  achieve
higher than 80 percent nitrogen  removal, while avoiding the use  of methanol, combine the
carbon oxidation, nitrification and denitrification processes in single sludge systems with no
intervening clarification steps.

     5.5.1  Systems Using Endogenous Respiration in a Sequential Carbon Oxidation-Nitrifi-
           cation-Denitrification System

When the process  uses the endogenous decay of  the organisms  for denitrification int the
system,  the  rate  limiting step  becomes the organism's  endogenous decay  rate  in an
oxygen-free environment. The process flowsheet for this system is shown on Figure 5-15; it
was developed and  first  tested  in  Switzerland^  and subsequently evaluated in other

                                        5-39

-------
locations.54'55-56'57 In the initial tests, fairly high mixed liquor solids (5200-5300 mg/1)
were employed and detention times  of 2.2 to 2.8 hours were used in the denitrification
section. Results of the various  studies are shown in Table 5-8. Deviations at the other
locations from the Swiss results are explained on the basis of insufficient reaction times in
either the nitrification and denitrification stages.5 9


                                FIGURE 5-15

     SEQUENTIAL CARBON OXIDATION-NITRIFICATION-DENITRIFICATION
 RAW
 WASTEWATER
[•»
1
COMBINED CARBON
OXIDATION-
NITRIFICATION

-*•
ANOXIC
DENITRIFICATION

_( SECONDARY
"^(SEDIMENTATION
                                   SLUDGE  RECYCLE
                                 TABLE 5-8

       PILOT TESTS OF WUHRMAN'S SEQUENTIAL CARBON OXIDATION-
             NITRIFICATION DENITRIFICATION SYSTEM (AFTER
                   CHRISTENSEN & HARREMOES, REF. 59)

Location
Switzerland

Germany

Germany
Germany
Seattle, Wa.
New York

Reference
53a

54a

55a
56b
57a
58C

Temperature,
C
13.6
17.1
16
16
12-16
20
20
-
"£)]-,, peak
denitrification rate
Ib NOs rem/Ib MLVSS/day
or kg/k.g/day
0.0168
0.041
0.022
0.026
0.038
0.048
0.026
-
Effluent
NO~-N.
2

-

3
7.8
-
1. 5 to 3. 1
Range of
total nitrogen removal,
percent
82-90

40-60

36-88
46 (average)
31-65
84-89
 ,  pilot scale
 ° lab scale              ,
   full scale, 95,000 gpd (360 mVday)
                                    5-40

-------
A modification of the concept tested in New York state involved placing a short anoxic cell
prior to the combined carbon oxidation-nitrification tank.58  An aerobic cell of 0.5 hr
detention time was also placed downstream of the anoxic denitrification, presumably to
improve settling. Total nitrogen removals of 84 to 89 percent were obtained, which are
comparable to the Swiss results.

Denitrification rates determined at the various locations are shown in Figure 5-16. Data for
Blue Plains and Pretoria, S.A. are shown also. In these two pilot plants wastewater was used
as  the  principal carbon source, but treatment stages using endogenous respiration  were
employed  also. The data shown for these two pilot plants are for those treatment stages
where endogenous respiration was predominant. When data were reported in terms of MLSS
only, volatile content was assumed to be 70 percent to allow reporting on a basis consistent
with denitrification rates given in Section 5.2.

An  envelope has been drawn around  all data points in Figure 5-16 to emphasize the
variation in measured denitrification rates. In the absence of pilot data, it would be prudent
to  establish reactor sizes on the basis  of the  lower envelope line  shown in Figure 5-16.
                                  FIGURE 5-16

       DENITRIFICATION RATES USING ENDOGENOUS CARBON SOURCES
        O.IO
C
o
o
'c
 o
 a,
    CO
    CO
        O.05
    £
    X)
               KEY
              SYMBOL
                 A
                 T
                 D
                 o
                 A
  LOCATION
Switzerland
Germany
Germany
Germany
Seattle,Wa
Blue Plains,DC
                        Pretoria, S.A.
                   ENVELOPE  FOR ALL DATA  POINTS
                          I	I             I
                         10
                                      15           20
                                   TEMPERATURE, C
                                     5-41
                                       25
3O

-------
Further, rates shown in Figure 5-16 are peak nitrate removal rates, qrj, and a safety factor
must be employed in design to ensure low nitrate levels in the effluent, as is done in Section
5.2.  It is notable that these rates fall considerably below those shown for methanol as the
carbon  source in  Figure 5-2.  For  instance,  median values at 20  C  for methanol  and
endogenous carbon are about 0.25 Ib NO^ -N rem./lb MLVSS/day (0.25 g/g/day) and 0.04
Ib NO^ —N rem./lb MLVSS/day (0.04 g/g/day). Therefore, a denitrification reactor using an
endogenous carbon source would have to be about six times larger  than a reactor with a
methanol carbon source at 20 C.

The  combined carbon oxidation-nitrification step can be designed with the criteria set forth
in Sections 4.3.3 and 4.3.5, because it has been found that the  anoxic denitrification step
has no impact on sludge activity, if the length  of the anoxic period is below 5 hours.59,62
In the calculations for the carbon oxidation-nitrification function, only the inventory under
aeration should be included  in  the  solids retention  time, growth rate, and removal rate
calculations.

     5.5.2 Systems Using Wastewater Carbon in Alternating Aerobic/Anoxic Modes

All of the systems using wastewater as the chief organic carbon source  for denitrification use
an alternating aerobic-anoxic sequence of stages, without  intermediate  clarification, to
effect  total nitrogen removal while attempting to  avoid ammonia nitrogen bleedthrough.
Some of the demonstrations of these systems have  shown that removals of 90 percent were
possible with the alternating mode concept.

         5.5.2.1 Aerobic/Anoxic Sequences in Oxidation Ditches

Pasveer was the first to investigate denitrification  in oxidation ditches and reported total
nitrogen removals of 90 percent, but few  details have been reported.59,63  Subsequent
investigators have  confirmed  Pasveer's  results  and  defined  the  conditions required for
denitrification in oxidation ditches.

Ditch design has varied  among investigators principally in the means used for aeration. Even
though aeration devices have ranged from Kessner brushes and  cage aerators to vertical
turbine mechanical aerators, all the variations can be termed oxidation ditches because all
use  the concept of a channel with aeration devices placed at localized points  as shown in
Figure 5-17. By controlling the level of aeration, the mixed liquor is exposed to alternating
aerobic and anoxic zones. The channel operates as an "endless channel" since only a portion
of the mixed liquor passing the channel outlet  is  withdrawn, with  the bulk  of the flow
remaining in the channel. In this way the mixed liquor is recirculated many times through
aerobic and anoxic zones prior to discharge from the channel.

The  largest scale test of the oxidation ditch employed for nitrogen removal has taken place
at the Vienna-B lumen thai plant  in Vienna, Austria. 64,65 -p^g plant  flowsheet  is shown in
Figure 5-18 and design  data are given in Table 5-9.  The plant consists of a pumping station,

                                        5-42

-------
                                 FIGURE 5-17

  PASVEER DITCH OR ENDLESS CHANNEL SYSTEM FOR NITROGEN REMOVAL
                                   RAW  WASTEWATER
         MIXED
         LIQUOR
RETURN  SLUDGE
                       SECONDARY  \      EFFLUENT
                  ^SEDIMENTATION
                          TANK
                                                        ROTOR  OR  OTHER
                                                        \ERATION  SYSTEM
screens, aerated grit chamber, two aeration tanks, two secondary sedimentation tanks and a
return sludge pumping station. The plant does not incorporate either primary sedimentation
or sludge handling facilities. It was constructed in 1968 at a cost of $ly623,000 (1968 U.S.
dollars).

Since the Vienna-Blumenthal plant is currently operating below design capacity, it has been
possible to  operate it in a manner encouraging nitrogen  removal.  The  two aeration tanks
have been  connected in series and the number of operational cage  aerators  varied to
encourage nitrogen removal. It has been found that  dissolved oxygen could be measured in
the  mixed  liquor  immediately  after  the  rotor, however the oxygen  demand  of the
microorganisms caused the oxygen to be depleted prior to contact with the next rotor. This
resulted in  an  alternating contact  of the mixed liquor  with  aerobic  and anoxic zones.
Nitrification took place in the aerobic zones and  denitrification  occurred in the anoxic
zones. 65
                                     5-43

-------
                      FIGURE 5-18
      VIENNA-BLUMENTHAL WASTEWATER TREATMENT PLANT
EFFLUENT
r


\


v



\


-X
^_
r.


*


v
>>
/

t




	 PAIR OF MAMMC
AERATION
TANKS
RETURN SLUDGE
PUMP STATION
                         EXCESS

                         SLUDGE
                       TO SEWER
                               /i
 GRIT  CHAMBER
            SETTLING
              TANKS
                        INFLUENT
                                    SCREEN
                                    PUMP STATION
\
   r
       OPERATION
       BUILDING
                         5-44

-------
                                      TABLE 5-9

                   DESIGN DATA FOR THE VIENNA-BLUMENTHAL
                        TREATMENT PLANT (REFERENCE 65)
   Population equivalents                                           150,000a
                                                                  SL      3
   Average dry weather flow (ADWF)                           22.8 mgd  (1.0 m /sec)
   Aerated grit removal tanks
      Number                                                      1
      Volume           .                                  9,500 cu ft (270 m )
      Detention time, ADWF                                       4.5. min
   Aeration tanks
      Number                                                      2
      Passes/tank                                                  2
      Length/pass                                              492 ft (150m)
      Width/pass                                                29 ft (8. 5m)
      Depth                                                    8. 2 ft (2.5m)
      Volume - total                                       420, 000 cu ft (12, 000m )
      Detention time, ADWF                                         3.3hr
      Cage aerator pairs/tank                                         6
      Horsepower each, rotor, pair                                    56
   Secondary sedimentation tanks
      Number                                                      2
      Diameter                                                148 ft (45m)
      Average depth                                            10 ft  (3. Om)
      Overflow rate, ADWF                               670 gpd/sf (27.2m3/m2/day)
    Estimate current connections, ultimate capacity estimated to be 300,000 population equivalents

The results of special 24-hr tests at the Vienna-Blumenthal plant are shown in Table  5-10.
While in some cases the available data on total nitrogen are incomplete, the ammonia and
nitrate nitrogen data indicate that nitrogen  removals of 80 to 90 percent are obtainable.
Special attention should be given to the operating conditions for these tests. While aeration
detention time was only 5.5  to  8.4 hr, the mixed liquor solids were maintained at relatively
high values, 5,300 to 6,700 mg/1. These high MLSS levels were  made possible due to the low
overflow rates maintained in the sedimentation tanks (280 to 410 gpd/sf or 11.4 to 16.6
m^/m^/day). U.S. practice would be to use higher overflow rates in sedimentation, reducing
the ability to maintain high MLSS levels under aeration  and consequently requiring higher
aeration detention times (see Sections 4.10 and 5.6).

Alternative aeration patterns were evaluated  in a recent  South African Study.66 Disc-type
aerators were arranged  in  a  series of four concentrically arranged channels.  Effluent was
monitored over a full year of weather conditions in the South African study.

One problem that was eventually overcome in the South African work was floating sludge in
the sedimentation tank. This condition was especially prevalent when an anoxic channel
section preceded the sedimentation tank. The condition was corrected by ensuring that each
channel had sufficient aeration to maintain DO in the channel after the disc. Before the disc

                                        5-45

-------
                                   TABLE 5-10

    OPERATION AND PERFORMANCE OF THE VIENNA-BLUMENTHAL PLANT
                  24-HOUR INVESTIGATIONS (REFERENCE 65)


Q
Influent flow rate, mgd (m /sec)
Aeration detention time, hr
MLSS, mg/1
Return sludge SS, mg/1
Average sedimentation tank
Overflow rate, gfid/sL
mVmVday
Aeration tank temperature, C
Operating rotors, tank 1, tank 2
Sludge loading,
Ib BODg/lb MLSS/day
Influent concentrations, mg/1
BOD.
0 •
COD
TOC
Total - N
Ammonia - N
Effluent concentrations, mg/1
BOD.
o
COD
TOC
Total - N
Ammonia - N
Removal efficiencies, percent
BODC
5
COD
TOC
Total - N
Ammonia - N
Sept. 2,
1971
9.6(0.42)
8.0
6,700
10, 000

280
11.4
18
4,2

0.11

268
475
155
36
21.8

13
49
14
4
3.8

95
90
91
88
83
Feb. 17,
1972
14.0(0.61)
5..S
6,200
10,300

410
16.6
12
3, 3

0.13

200
384
126
24
9.0

13
50
13
4
2.7

93
87
90
82
70
June 6,
1974
12.4(0.54)
6.1
5,300
8,000

360
14.7
18
3, 3

0.20

268
463
128
a
21.8

10
39
12
a
6.3

96
92
91
a
71
July 16,
1974
10.9(0.48)
7.0
5,900
9,000

320
13.0
20
4, 2

0.19

a
574
143
a
16.6

a
39
14
a
3.9

a
93
90
b
76
July 31,
1974
9.5(0.42)
8.0
5,400
7,900

280
11.4
19
4, 3

0.16

a
515
134
a
17.2

a
35
13
a
2.4

a
93
90
c
86
Aug. 7,
1974
9.0(0.40)
8.4
5,400
8,400

270
10.9
20
4, 4

0.24

a
778
208
a
21.2

a
35
15
a
2.7

a
92
93
d
87
  aNot available
   Not available, but no nitrate - N In effluent
  °Not available, but only 0.3 mg/1 Nitrate - N In
  dNot available, but only 2.4 mg/1 Nitrate - N In
effluent
effluent
aerator, DO was allowed to drop to zero. Since mixed liquor was withdrawn after the disc in
a fresh condition, a sparkling clear effluent was produced.

Table 5-11  summarizes the performance data for the successful configurations of aeration in
the South African study. Nitrogen removal averages 79 percent. A detailed 4-day analysis
during a period when 86 percent nitrogen removal was observed showed that 40 percent of
the influent nitrogen was incorporated in the sludge, 45 percent was nitrogen gas lost to the
atmosphere and 14 percent appeared in the effluent.
                                       5-46

-------
                        TABLE 5-11
       OPERATION AND PERFORMANCE OF OXIDATION DITCH
OPERATED FOR NITROGEN REMOVAL IN SOUTH AFRICA (REFERENCE 66)

Feed flow rate, gpm (1/h)
Retention time, hr
Beturn-sludge ratio
MLSS, mg/1
Water temperature, C
Number of discs used, total
Dissolved oxygen: mg/1
Before aeration discs
After aeration discs
SVI, ml/gram
COD, Ib COD/lb MLVSS/day
(g/g/day)
Influent
COD, mg/1
Kjeldahl-N, mg/1
Ammonia-N, mg/1
Effluent
COD, mg/1
Kjeldahl-N, mg/1
Ammonia-N, mg/1
Nitrite-N, mg/1
Nltrate-N, mg/1
COD removed, percent
N removed, percent
Bun 1
21.4 (4800)
23; 2
2:1
4330
21-22.5
5 (on single shaft)
Virtually always 0
Banging 0. 3-0. 8
213
0.155
749
39.5
21.3
28.2
4.5
3.4
Trace
96; 2
85.2
Bun 2
35.2 (8000)
13.9
1.5:1
4280
20-22
11 (on two shafts)
Positive DO with
only Infrequent
zero readings
Higher by only 0. 2-
0. 3 mg 1 than
before the discs
212
0.28
791
39.4
20.7
30i9
4.05
3.3
0,3
2.65
96,0
81.6
Bun 3
22 (5000)
22; 2
2:1
3720
14-17
5 (on single shaft)
Usually zero or slightly
positive. 0-0- 3 in
channel 1
Oi3-0;6 in chs. 2 and
3. 0.6-1.0 in chs.
1 and 4
263
0:20
678
43; 1
28:2
34.4
6;4
314
Trace
94;6
77.3
Bun 4
26.5-28.6 (6000-6500)
17.1-18.5
3660
14-15. 5
7 (on single shaft,
3-2-1-1)
0-0-4 in first 3 chs.
0-6-2-0 inch. 4.
0.3-1.3 in chs. 1,2 and
3. 1.2-2.3 inch. 4.
268
0.26
723
45.0
29.2
37.4
5.0
2.1 1
0.6
7.1
94.5
72.0

-------
The sludge in the South African study was always in a bulking condition, with SVI levels
ranging from 212  to 268  ml/g. This bulking tendency may be a general property of
alternating aerobic-anoxic processes, as bulking sludges have developed in other alternating
aerobic/anoxic systems (see Sections 5.5.2.3 and 5.5.2.4).

It has been found that the cage aerators which are typically employed in the oxidation ditch
are not well  suited to nitrogen removal applications.67 The cage aerator is not capable of
simultaneously mixing and maintaining DO control; too much oxygen is imparted to allow
development  of alternating aerobic  and anoxic zones while maintaining sufficient ditch
velocities (1 fps or 0.30 m/s) for prevention of settling of solids in the ditch. In one case, the
problem was solved by providing separate submerged propellers for mixing which allowed
the cage aerators  to be managed for DO control alone. 67 An aeration system has been
developed that can  both control the  mixing level and  the level of aeration simultaneously.
Vertical shaft aerators are placed at the channel turning points in restricted areas so that the
fluid rotation generated by the aerator is marshalled to move the flow around a semicircular
turn. Power input and  the number of on-line aerators can be varied to control DO level
while maintaining sufficient mixing in the system. Nitrogen removals of 80 to 87 percent are
reported.68

         5.5.2.2 Denitrification in an Alternating Contact Process

The novel alternating contact process shown in Figure 5-19 has been tested in Denmark in
both lab-scale and full-scale tests at wastewater flows up to 1.5 mgd (0.067 m-Vsec).69 The
process is similar to the oxidation ditch approach  in that alternating  aerobic and anoxic
residence periods are provided in contact with raw wastewater. The means of accomplishing
this alternating contact are very  different from an oxidation ditch, however.

The operational  sequence consists of the four phases shown in Figure 5-19. Aeration air is
controlled  to provide alternating aerobic and anoxic  conditions.  When anoxic conditions
were required, the  aeration air was merely turned  down to  a level sufficient to keep the
sludge  anoxic but in suspension. Raw wastewater is alternately directed to one or the other
of the two tanks, according to a predetermined cycle. The cycle that an individual tank
passes  through is depicted on Figure 5-20. The operational phasing of wastewater addition
and aeration  are shown in addition to the  phasing  of nitrification and denitrification.
Multiple liquid exposure to anoxic and aerobic zones is done by limiting the sequence time
so that an average of 6 to 15 cycles are completed prior to liquid discharge. The effluent is
discharged  from the tank to sedimentation  only when the tank is under aeration.

The stoichiometric  carbon to nitrogen ratio for denitrification was defined in terms of a 7
day BOD  and found to be BODy/NO^ —N = 5.2. Denitrification rates are summarized in
Section 5.2.6. Effluent nitrate  levels of 2.0 to 5.0 mg/1 appeared obtainable with proper
selection of design and operating mode. A full description of the mathematical model used
for process design can be found in reference 69.
                                        5-48

-------
                                  FIGURE 5-19

                       ALTERNATING CONTACT PROCESS
 Phase 1 - Denitrif ication/Nitrification      Phase 3- Nitrification/Denitrification
 RAW
 WASTEWATER
                                            RAW
TEWATER_
J
I
SL
AN
^
UDGE RETURN
 Phase 2 - Intermediate  Aeration
 RAW
 WASTEWATEF^
Phase 4 - Intermediate  Aeration
RAW
WASTEWATER
1
SI
1
A
s
-UDGE RETURN
 KEY
 AN =  Anoxic conditions ,  A = Aerobic conditions, S = Sedimentation
         5.5.2.3 The Bardenpho Process

A recent South African development  for nitrogen removal using both wastewater and
endogenous carbon for  denitrification is shown in Figure 5-21. Termed the "Bardenpho"
process by its  developer,  the system is a  combination of two previously developed
processes.7® Mixed liquor containing nitrate is recycled from the second (aerobic) tank to
the initial (anoxic) tank for denitrification.71 Appended to the first two tanks is a third
(anoxic) tank for removing the nitrate remaining in the effluent from the second (aerobic)
tank.  In this third tank endogenous respiration is used for denitrification in the manner
described in Section 5.5.1.  Finally, a period of aeration is provided to improve sedimenta-
tion.

Initial lab scale tests of the Bardenpho process showed that nitrogen removals of 93 percent
were possible at recycle rates of 4:1 and 1:1 of mixed liquor and return sludge, respectively.
                                       5-49

-------
                            FIGURE 5-20
              OPERATIONAL SEQUENCING OF ONE OF TWO
         AERATION TANKS IN ALTERNATING CONTACT PROCESS
                            TSEQ
                                         -H
                0.5 (TSEQ)
                             0.5  (TSEQ)
TLP-r
IA
TLP+
                                                   IA
           Wastewater Add i t ion  No Wastewater Add i tion
             Effluent to other
              Reaction  Tank
           No Aeration
                        Eff I. to Sedimentation
                                    Tank
                         Aeration
                                                             i>
Aerobic



Anoxic





Aerob

ic



Anoxic
^^
 Denitrification
                                        Nitrification
    KEY TO SEQUENCING  TIMES

      TSEQ = Overall  Anoxic-Aerobic Sequence  Time
      TLP-S. = Lag phase  of  Denitrification

      TLp+ = Lag phase  of  Nitrification

      TIA   = Intermediate Aeration  Time

Under this condition, 5 to 7 mg/1 of total nitrogen appeared in the effluent.70 A pilot test
of 9 months duration  was then conducted to determine the long-term performance of the
system on a scale of 26,000 gpd (100 m3/day). During the last three months of the study,
nitrogen removals were in the range of 80 to 90 percent.61 Observations of denitrification
rates for the first (anoxic) rank are summarized in Section 5.5.2.5 and denitrification rates
for the third (anoxic) tank where  endogenous carbon is employed are presented in Section
5.5.1.
                                 5-50

-------
                                 FIGURE 5-21

             THE BARDENPHO SYSTEM - SEQUENTIAL UTILIZATION
            OF WASTEWATER CARBON AND ENDOGENOUS CARBON
             MIXED LIQUOR RETURN
RAW
WASTEWATER'
PRIMARY
EFFLUENT



ANOXIC
DENITRIFI-
CATION
TANK




AEROBIC COMBINED
OXIDATION-
NITRIFICATION
TANK
RETURN
S

-^
LUC
ANOXIC
DENITRIFI-
CATION
TANK
-*

AEROBIC
TANK
>GE
/ SECON
— ^SEDIME
                                                                        EFFLUENT
The sludge  tended to be in bulking condition, settling very little in the standard one-liter
cylinder SVI test. Even with  stirring a relatively high SVI was obtained (150 ml/g). This
bulking sludge  condition appears to be typical for alternating  aerobic/anoxic systems, and
care must be employed in design and operation to deal with this problem.

The Bardenpho process has been tested in a second 26,000 gpd (100 m^/day) pilot plant at
Pretoria, South Africa.72 While full test results were not available, test data from the initial
few weeks of operation are shown in Table 5-12. These  data demonstrate that relatively high
nitrogen removals are obtainable with the Bardenpho process.
                                  TABLE 5-12
                PERFORMANCE OF THE "BARDENPHO" PROCESS
                AT PRETORIA, SOUTH AFRICA (REFERENCE 72)
Parameter
Influent COD, mg/1
Effluent COD, mg/1
Percent COD removal
Influent TKNa, mg/1
Effluent TKN, mg/1
Percent TKN removal
Effluent nitrate - N, mg/1
Influent total nitrogen , mg/1
Effluent total nitrogen , mg/1
Percent total nitrogen removal
Period
Jan. 7 to Jan. 31,
1975
226
46
79
21.1
1.9
91
2.6
21.1
4.5
79
Feb. 2 to Feb. 14,
1975
176
48
73
15.9
1.4
91
1.7
15.9
3.1
81
   TKN = total Kjeldahl nitrogen
   Assuming influent oxidized nitrogen is zero
   Assuming effluent nitrite nitrogen is zero
                                     5-51

-------
         5.5.2.4 Alternating Aerobic/Anoxic System Without Internal Recycle

Investigators at  the EPA Blue Plains  pilot plant conceived of still another way to achieve
alternating aerobic and  anoxic environments with the system shown in"Figure 5-22.60 A
two pass aeration tank  was provided with separate aeration and mixing facilities. In each
basin, 2  mechanical  mixers  were employed  to keep the mixed  liquor in suspension,
independent of the aeration system. Air was supplied alternately to each basin; first to one
basin and then to the other in a 30 minute cycle.6® Dissolved oxygen level in the pass under
aeration was  controlled between 2 and 3 mg/1, while the anoxic pass decreased to zero
rapidly after cessation of aeration. The pilot process was typically operated at a flow rate of
50,000 gpd( 189 m3/day).

                                  FIGURE 5-22

        BLUE PLAINS ALTERNATING ANOXIC AEROBIC SYSTEM (REF. 60)

                               D C.  RAW  WASTEWATER

                            FeCI3-
                                PRIMARY
                    AIR
   i
   C\
j  U
                            o  fx,   Q
                            A         I
                       NITRIFICATION DENITRIFICATION
                                  REACTOR
                            WASTE —
                                      RETURN SLUDGE
                                 SECONDARY
                                 —I        I—
                                  CLARIFIER
                                ALUM
                                      FILTERS
                                  FINAL EFFLUENT
                                      5-52

-------
A summary of operating and performance data for the 9 months of test work is shown in
Table  5-13.  During  9  months  of  operation,  the  plant was  operated  at  a F/M  of
approximately 0.1 Ib BODs/lb MLVSS/day (0.1 g/g/day), expressed on the basis of the total
inventory in the reactor. This F/M was sufficiently low to permit  the development of a
mixed culture of organisms for carbon oxidation, nitrification and denitrification.

Nitrogen removals during the study varied from 54 to 84 percent, but operational problems
contributed to  the lower reported removals. These problems may be avoidable in full-scale
operation. For  instance,  in July and  August,  ferric chloride  addition  in the primary
treatment stage reduced the COD/TKN ratio from 10 to about 7.5 to 8.0, and resulted in a
situation in which the lack of a sufficient organic  carbon source  limited the  degree  of
denitrification  obtainable. The lower removals experienced in April and May were due  to
the fact that during a portion of each month, the alternate aerobic/anoxic regime was
altered to a full aerobic mode to rid the system of filamentous growth.

During September, performance  deteriorated for an unexpected reason. The flow sheet in
Figure 5-22 was  modified to include two more  reactor stages prior to the  clarifier. In the
first added stage,  an  anoxic one,  methanol was  added  to cause denitrification of residual
nitrite and nitrate. An aerobic stabilization step was added as the last stage. It was found
that methanol addition caused an immediate ammonia increase in the process. Subsequent
studies showed  that methanol is toxic to nitrifiers.

Since  all operating problems  can be  explained, it  can be concluded that the  system is
capable of 84  percent nitrogen removal in the summer (23 C)  and 75 percent nitrogen
removal in the winter (14 C).

Like in other  alternating aerobic/anoxic studies, it was found  that a severe filamentous
bulking condition developed in the sludge,  limiting wintertime  clarifier  overflow rates  to
about  300 gpd/sf.60 Bulking  sludge has been observed  at low temperatures at Blue Plains
and at low F/M  operation at  Blue Plains and at other plants. Whatever the cause of the
bulking  problems, it appears that  clarifier  operation  will  limit  operation of this
aerobic/anoxic system, as it will limit the other aerobic/anoxic systems.

Kinetic data was obtained during the study. On and off aeration of samples of the pilot
plant mixed liquor was employed over several cycles to simulate operation of the pilot unit.
Nitrification and denitrification rates determined by this procedure are shown in Table 5-14.
Nitrification rates were similar through all cycles, while denitrification rates decreased as the
batch reaction  continued. Denitrification rates measured in the first cycle represent peak
rates possible when a readily available carbon source is available during denitrification. By
the third or fourth cycle,  the rates represent  denitrification when the readily  available
carbon is depleted and an endogenous carbon source is used. Denitrification rates for cycle 1
are also presented in Section 5.5.2.5 with measurements of other observers, while rates for
cycle 4 are presented in Section 5.5.1.
                                        5-53

-------
                                               TABLE 5-13
                      SUMMARY OF OPERATION AND PERFORMANCE FOR THE BLUE PLAINS
                          ALTERNATING AEROBIC/ANOXIC SYSTEM (REFERENCE 60)
Month
1973
Jan

Feb

March

April

May

.Tune

July

August

Sept

Det.
time
hr
12i3

12.3

12.3

12.4

10.5

8.8

6.8

6.6

8.7

F/M
lb BOD applied/
tb MLVSS/day
or g/g/day
0.072

0.066

0.10

0.081

0.089

0.105

0.093

0.089

0.11

MLSS
mg/1
(%
volatile)
3510
(74)
3980
(73)
2950
(73)
3540
(67)
4170
(69)
4010
(69)
3040
(64)
3200
(57)
3700
(65)
SVI,
sl
g
245

250

330

277

227

188

133

134

-

COD/
TKN
ratio
9.6

9.9

10.5

10.6

10.0

10.3

7.9

7.5

10.0

Temp. ,
C
14.0

14.2

15.5

-.

-

23.0

25.0

25.5

26.0

Influent quality mg/1
BOD5
96.5

99

110

98.8

115

107

51

44.2

99

SS
110

108

128

120

109

112

153

197

110

Total
Kjeldahl
Nitrogen
25.7

23.2

24.8

21.7

23.3

24.0

15.0

14.9

22.6

Effluent quality, mg/1 a
BOD,.
o
20.4

14.0
L
6.5b

5.3b

3.3b

3.2b
t.
3.8b
U
2.6b

7.2b

SS
15.4

14.3

15.0

13.0

11.8

7.8

9.0

10.0

16.0

Total
Kjeldahl
Nitrogen
2.28

1.52

4.20

5.20

1.36

1.51

2.14

1.23

10.2°

NOg and
NOJ - N
3.99

4.41

2.30

6.03

8.25

2.30

2.72

3.74

0.22

Removals, percent
BOD5
79

86

94b

95b

97b

97b

93b

94b

93b

SS
85

87

88

89

89

93

94

95

85

Total
Nitrogen
76

75

74

49

59

84

68

67

54

 Prior to filtration
bNltrificatlon inhibited
 Ammonia level was 9.4 mg/1 as N (see text)

-------
                                  TABLE 5-14
        OBSERVED NITRIFICATION AND DENITRIFICATION RATES FOR
            BLUE PLAINS ALTERNATING ANOXIC/AEROBIC SYSTEM
Mode
Aerobic-
Nitrification



Anoxic-
Denitrification



Temp. ,
C
15.5
25.0
27.0
26.5
15.5
25.0
27.0
26.5
Peak nitrification or denitrification
removal rate, Ib N/lb MLVSS/day
Cycle 1
0.032
0.083
0.11
0.12
0.032
0.055
0.042
0.026
Cycle 2
0.042
0.095
0.11
-
0.029
0.030
-
0.0075
Cycle 3
0.016
-
-
-
0.021
0.033
-
-
Cycle 4
0.026
-
-
-
0.019
0.030
-
-
Cycle 5
0.035
-
-
-

-
-
-
         5.5.2.5 Kinetic Design of Alternating Aerobic/Anoxic Systems

The four factors which can limit denitrification process efficiency in alternating aerobic/
anoxic systems using wastewater as the carbon source are as follows:^

     1. Nitrification

    2. Denitrification

    3. Carbon-nitrogen ratio

    4. Operational mode (process hydraulics)

The third factor  has been evaluated by  several investigators (see  Sections 5.5.2.2  and
5.5.2.4) and further discussion here is unnecessary.
                                      5-55

-------
To  evaluate nitrification limitations on the system, nitrogen loads and nitrification rates
must be taken into account. Most investigators agree that the design  of the combined carbon
oxidation-nitrification functions of the aerobic phase can be separated from the anoxic
phase. 61,69,73  jj has been found that anoxic  periods up to 5 hours have no impact on
aerobic sludge activity. 62,5 9 Therefore, the carbon oxidation and nitrification calculations
for the aerobic periods  can be virtually identical to those advanced for combined carbon
oxidation-nitrification in Sections 4.3.3 and 4.3.5. In the calculations of nitrifier solids
retention time,  nitrifier  growth rate,and removal rates in the aerobic residence periods, only
the solids inventory under aeration is employed. This is because the environment must be
aerobic for nitrifier growth to occur. As always, a safety factor must be employed.

Sizing  of denitrification steps must consider nitrite  load  and nitrate removal rates and
consideration of the safety factor concept in design. Of the two models formulated for these
systems,  the safety factor concept is used only  for the nitrification step in the Bardenpho
design^ but for  some unapparent reason not for denitrification. 61 The safety factor concept
was not used for nitrification or denitrification in  the alternating contact process  design
(Section 5.5.2.2).69 it must be emphasized that unless a safety factor is incorporated in the
design, nitrogen removal will deteriorate under peak load conditions.

Most of the alternating  processes employ both wastewater carbon  and endogenous carbon
for denitrification  at some  point  in the  system.  Observed  denitrification  rates for
endogenous carbon have been summarized  in  Section 5.5.1. Experimentally determined
denitrification rates in alternating aerobic/anoxic systems with wastewater as the carbon
source are shown in Figure 5-23. When data were reported in terms of MLSS only, volatile
content was assumed to be 70 percent to allow reporting on a basis consistent with the
denitrification rates shown in Section 5.2. These rates are peak nitrate removal rates, and are
expressed as q^, using the terminology developed in Section 3.3.5.2. As can be seen from
Figure 5-23,  there  is a wide variation in measured denitrification rates in  systems using
wastewater as the organic carbon source. As a result, it may not be a conservative practice to
use the denitrification rates given in Figure 5-23; rather, it would appear prudent to conduct
pilot investigations  to verify  design parameters for  denitrification  when wastewater is the
carbon source for denitrification.

The rates for denitrification with wastewater as the carbon source fall below those found for
methanol as the carbon source shown in Figure 5-2. Median rates at 20 C for methanol and
wastewater carbon are about 0.25  Ib NO^ -N removed/lb MLVSS/day (0.25 g/g/day) and
0.07 Ib NO3  -N removed/lb MLVSS/day (0.07 g/g/day), respectively. A denitrification
reactor using a  wastewater carbon  source would have to be about three and one-half times
larger than a denitrification reactor using methanol as the carbon source.

One cause of the difference  in reaction rates  between methanol and wastewater carbon
relates to biological availability. Methanol is a simple, easily degraded compound, whereas
wastewater contains a mix  of easily degraded and hard to degrade compounds. Wastewaters
may vary in  the  relative distribution of easily  degraded and hard to degrade compounds,
thus causing variations in denitrification rates between locations.

                                        5-56

-------
                                 FIGURE 5-23
             EFFECT OF TEMPERATURE ON PEAK DENITRIFICATION
                  RATES WITH WASTEWATER AS CARBON SOURCE
U. 10
O
\
to
to
^
5

e
u o.io
2»

Qj
N
•^
O
•Q
Qj
^.
O
tfc
c 0.05
.0
o
O
lj
M •
U
H—
c
0)
Cl
>0*
n
i 1 1 1
KEY
SYMBOL LOCATION REF.

• Blue Plains 60
• Austin, Texas 74
D Pretoria, S. A. 61
A Denmark 69 D

~ D ~
a
a
a



•

•
— —

A •

•
•


1 1 1 1
                         IO
  15           20
TEMPERATURE,  C
25
30
The hydraulic mode of operation significantly affects the kinetic design procedure. In all of
these systems,  relatively  complex and lengthy mass balances are necessary to describe the
system. None-the-less, such descriptions are possible and have been developed for two of the
alternating aerobic/anoxic systems.61.69 These models are presented in sufficient  detail in
the literature to allow their modification for use in design. Iterative solution of equations is
required, and the digital computer has proven a useful design tool.69 The limitation of these
models is that generally applicable kinetic rate data are not yet available.
                                       5-57

-------
5.6 Solids-Liquid Separation

The considerations for design of sedimentation tanks for denitrification systems are  the
same as those discussed for nitrification systems in Section 4.10 and those points common
to both will not be repeated herein.

Rising  sludge has occasionally  plagued denitrification  systems,  depending  on system
design.^A?, 15,49,50 jo remedy this and other problems, the original suspended growth
denitrification system (using methanol)  was modified by  placing an aerated stabilization
step between the anoxic denitrification  reactor and  the  denitrification clarifier (Figure
5-lB).2>3,21,51  This  step was taken because it was found that the "conventional" design
was basically an unstable process. In the conventional system, the methanol: nitrogen ratio
(M:N ratio) had to be kept at precisely the optimum level (2.5 to 3.0). When methanol  was
overfed, the effluent BOD5 would rise. When methanol was underfed, nitrate would bleed
through to the clarifier, and denitrification would proceed in the clarifier using the sludge as
the  carbon source. Floating  sludge, buoyed up by nitrogen gas bubbles, caused a severe
deterioration in effluent quality.

The coupling of an anoxic residence period and an aerobic residence period in the modified
system  (Figure  5-1)  is based  on the  recognition  that  dissimilatory  denitrification is
accomplished by facultative bacteria using biochemical pathways that are almost identical to
aerobic biochemical pathways. The main difference in the biochemical pathways lies in the
electron transport system where the terminal enzyme is changed and nitrate replaces oxygen
as the  final  electron acceptor. 75 These facultative bacteria can shift rapidly  from using
nitrate to using oxygen and vice versa. In the aerobic tank, the excess methanol is oxidized
and the mixed  liquor solids are aerobically stabilized. The aerobic tank also serves  the
purpose of stripping supersaturated nitrogen  gas from solution so that nitrogen gas bubbles
will not form during sedimentation.

A mildly aerated physical conditioning channel transfers the denitrification mixed liquor to
the final clarifier. Recognizing that the very turbulent conditions in the aerated stabilization
tank causes floe breakup and dispersed fines in suspension, the purpose of the channel is to
allow these dispersed particles to be incorporated into floe under mild turbulence conditions
that favor aggregation over breakup. 76

Tests of the original and modified systems  were conducted at the Central Contra Costa
Sanitary District's  Advanced Treatment Test Facility.2>3,21,51 Comparing performance
with stabilization  to performance  without it  (Tables  5-15  and  5-16),  indicates  the
substantial merits of the aerated stabilization tank. The test  period without stabilization  was
one in which daily adjustments were made in the methanol feed rate, hence there was little
if any methanol bled into the effluent. This is reflected in the low soluble BODs of 5 mg/l.
During  the test  period  with aerated  stabilization, less careful control was exerted in
methanol feed which  resulted in a fairly high M/N ratio of 3.3. Despite  this overfeeding of
methanol,  the effluent soluble BODs remained 5  mg/l.  In other words,  the aerated

                                         5-58

-------
   stabilization tank formed a favorable environment for the oxidation of the excess methanol.

   One significant factor contributing to less suspended solids and turbidity in the effluent was
   the  development  of  a  ciliate and rotifer population in the culture. Previously these
   organisms were not abundant. With a significant aerobic residence period, these organisms
                             TABLE 5-15
  EFFECT OF STABILIZATION TANK ON DENITRIFIED EFFLUENT AT THE
            CENTRAL CONTRA COSTA SANITARY DISTRICT'S
        ADVANCED TREATMENT TEST FACILITY (REFERENCE 3)
Constituent
Nitrate as N
Total BODg
Filtered BOD,
o
Suspended Solids
Turbidity
Total organic carbon
Soluble organic carbon
Temperature
Mean effluent quality
without stabilization, mg/1
(Feb. 13 to Mar. 13, 1972)
0.5
37
5
14
5a
18
7
16 to 17b
Mean effluent quality
with stabilization, mg/1
(Mar. 28 to April 20, 1972)
0.7
6
5
4
1.3a
8.6
5
16 to 19b
JTU
b
 degrees C
                             TABLE 5-16
DENITRIFICATION PROCESS PARAMETERS AT THE CENTRAL CONTRA COSTA
  SANITARY DISTRICT'S ADVANCED TREATMENT TEST FACILITY (REF. 3)
Parameter
Flow, mgd
Residence time, hr
reactor
stabilization tank
MLSS, mg/1
SVI, ml/g
Nitrogen rem Ib/lb MLVSS/day
Methanol/ nitrate - H ratio
Without stabilization
Feb. 13 to
Mar. 13, 1972
.46

.85
0
3000
143
.23
2.8
With stabilization
Mar. 28 to
April 20, 1972
.47

.82
.79
2500
242
.18
3.3
                                     5-59

-------
could flourish and clarify the liquid.  Sulfide odors in the sludge were  eliminated by the
modification.

In addition to stabilizing the liquid, the sludge is stabilized as well. Without the stabilization
tank, solids appearing in the effluent contained about 2.3 Ib BOD5 per Ib of SS. With
stabilization, effluent solids contain less BOD5 with a BOD5/SS ratio of 0.25. The net effect
of this is to reduce the effluent BOD5 from 37 to 6 mg/1. This reduction in BOD5 value of
the solids is very likely due to enhanced endogenous respiration in the stabilization tank.

Another indication of sludge  stabilization in the  aerated stabilization tank is the reduction
in denitrification rates  observed  in the  modified system  compared to the conventional
system. Prior to the modification, denitrification rates were as high as 0.3 to 0.56 Ib Nitrate
-N rem./lb MLVSS/day at 16 to 18 C and these rates were below the peak limiting rate, qpj,
as effluent nitrate was always low. This range in rates is roughly twice the  range in rates
shown for the  system employing  aerated stabilization at CCCSD as shown in Figure 5-2.
While one impact of  aerated  stabilization is to decrease denitrification rates  and therefore
increase anoxic reactor requirements, the other effect of this impact is to render the sludge
less active and more resistant to the rising sludge  problem in the denitrification clarifier flue
to denitrification. This is in agreement with the conclusions drawn in Section 4.10, where it
is shown that the tendency for rising sludge was related to denitrification rates and sludge
residence time in the  clarifier. Rapid sludge removal equipment, such as the vacuum pickup
type, should be provided in all sedimentation tanks to minimize sludge residence time and
reduce the likelihood of rising sludge.

Similar to  the  findings  with methanol  based  denitrification, an  aerobic step  has been
usefully  employed in  combined  carbon  oxidation-nitrification-denitrification systems.
Generally, a 1 to 2 hr residence period is provided prior to mixed liquor separation in the
sedimentation step.

Reactor-clarifier interactions  should  not be ignored. For instance design  examples are
presented in the  literature for alternating aerobic/anoxic  systems where the mixed liquor
solids are assumed to  be  5000 mg/1 (at 14 C) and 7000 mg/1 (at 10 C).61>69 Current U.S.
practice is to limit mixed liquor  levels below 3000 mg/1 unless clarifier overflow rates are
reduced to account for the need to thicken and  return the  sludge.40 Operation at 5000 to
7000 mg/1 would require very large clarifiers to ensure that solids are not lost at peak wet
weather flow conditions. Bulking  sludge tends to occur in the  combined carbon-oxidation-
nitrification-denitrification systems which mandates even greater conservatism in allowable
mixed liquor levels and clarifier design than would normally  be the case.

5.7 Considerations for Process Selection

The choice of denitrification for nitrogen  removal mandates the process sequence of
nitrification-denitrification  for nitrogen removal. Two kinds of comparisons in denitrifica-
tion process selection can be made. First, the nitrification-denitrification sequence can be

                                         5-60

-------
compared to  the  physical-chemical  alternatives.  Second, if nitrification-denitrification is
chosen, comparisons  must be  made among the denitrification alternatives to make the
process selection.

     5.7.1 Comparison to Physical-Chemical Alternatives

The  comparison  made  between the  nitrification  portion  of the  sequence  and  the
physical-chemical alternatives in Section 4.11.1 need not be repeated here.

Total dissolved solids (TDS) increment in the nitrogen removal system will have a bearing on
process selection in many situations. Nitrification-denitrification leads to a net reduction in
wastewater alkalinity  and no change in  the total mineral content of the water. Both the
breakpoint chlorination and the selective ion exchange process lead to TDS increases.

     5.7.2 Choice Among Alternative Denitrification Systems

Many  of the  considerations  presented  in Section  4.11  are applicable to  denitrification
system selection  and  are not repeated  here.  Other  factors affecting process  choice are
summarized in Table 5-17. Most of these factors were considered earlier in this chapter.

In some treatment plants, both nitrogen  and phosphorus removal has been mandated. The
combined carbon oxidation-nitrification-denitrification systems are somewhat restricted in
the sequencing of the phosphorus removal step. Chemical addition in the primary treatment
stage cannot be employed,  as this  would not leave sufficient organic carbon influent to the
process to complete denitrification.

Another factor listed in Table 5-17 is stability of operation and degree of nitrogen removal.
Several long-term  tests of  denitrification  systems  using  methanol  have successfully
demonstrated  consistently  high  levels of nitrogen removal.  Equivalent  operational exper-
ience with the combined carbon oxidation-nitrification-denitrification  systems will soon be
obtained, as large-scale experimental  work is currently underway in the U.S., Denmark, and
South  Africa.  When the results  of this  work  are available,  fair comparisons can  be made
between the two  types of systems. Based upon presently available data, it appears  that the
combined carbon  oxidation-nitrification-denitrification systems are  capable of 75 to 90
percent nitrogen  removal;  in comparison, methanol based systems can achieve 90 to 95
percent nitrogen removal.

One other kind of comparison  can be made between the systems using methanol and the
combined systems.  Since the rate  of denitrification with wastewater as the carbon source in
the  combined system is  lower than  with  methanol  as the carbon  source, greater
denitrification reactor sizes are  required for the  combined  system. This issue can best be
analyzed with an example comparing the alternative systems. A design  example for the
Bardenpho process has been presented in the  literature that can be usefully employed for
the comparison.61 Specific loading criteria are not important in this example; for these the

                                        5-61

-------
                                          TABLE 5-17
                COMPARISON OF DENITRIFICATION ALTERNATIVES
     System Type
               Advantages
       Disadvantages
 Suspended growth using
   methanol following a
   nitrification stage
Denltrification rapid, small structures required
Demonstrated stability of operation
Few limitations in treatment sequence options
Excess methanol oxidation step can be easily
  incorporated
Each process in the system can be separately
  optimized
High degree of nitrogen removal possible	
Methanol required
Stability of operation linked
  to clarifier for biomass return
Greater number of unit processes
  required for nitrification-de-
  nitrification than in combined
  systems
 Attached growth (column)
   using methanol following
   a nitrification stage
Denitrification rapid, small structures required
Demonstrated stability of operation
Stability not linked to clarifier as organisms
  on media
Few limitations in treatment sequence options
High degree of nitrogen  removal possible
Each process in the system can be separately
  optimized	
Methanol required
Excess methanol oxidation process
  not easily incorporated
Greater number of unit processes
  required for nitrification-
  denitrification than in combined
  system
 Combined carbon oxi-
   dation-nitrification-
   denitrlfication in sus-
   pended growth reactor
   using endogenous
   carbon source
No methanol required
Lesser number of unit processes required
Denitrification rates very low;
  very large structures required
Lower nitrogen removal than in
  methanol based system
Stability of operation linked to
  clarifier for biomass return
Treatment sequence options
  limited when both N and P
  removal required
No protection provided for
  nitrifiers against toxicants
Difficult to optimize nitrification
  and denltrification separately
 Combined carbon oxi-
   dalion-nitrification-
   denltrlflcation in
   suspended growth
   reactor using wastewater
   carbon source
No methanol required
Lesser number of unit processes required
Denltrification rates low; large
  structures required
Lower nitrogen removal than In
  methanol based system
Stability of operation linked to
  clarifier for biomass return
Tendency for development of
  sludge bulking
Treatment sequence options
  limited when both N and P
  removal required
No protection provided for
  nitrifiers against toxicants
Difficult to optimize nitrification
  and denitrlficatlon separately
reader is referred to reference 61. Reactor residence times are as provided in reference 61 at
a temperature of 14 C excepting that the  MLSS value  has been downwardly  adjusted from
5000 to 3000 mg/1, according  to U.S. practice. The effect of this is to increase by the ratio
of  5/3  the detention  times in the  reactors  in the  design example. The  adjusted residence
times are shown in Figure 5-24. A methanol-based system useful for comparison purposes is
also shown in Figure 5-24. In this sytem,  a  combined carbon  oxidation-nitrification step is
chosen,  since if a  combined  operation provides acceptable treatment  for the Bardenpho
                                                5-62

-------
INFLUENT

WASTEWATER
                              FIGURE 5-24

                 COMPARISON OF DENITRIFICATION SYSTEMS
               RETURN SLUDGE
        ANOXIC

     DENITRIFICATION

       TANK (4HR)
           I
         AEROBIC

     COMBINED CARBON

        OXIDATION-

      NITRIFICATION

          TANK


          (10 HR)
           I
UJ
oc


(£.
O
ID
O
         ANOXIC

     DENITRIFICATION

       TANK   (5HR)
           I
   AEROBIC TANK (2 HR)
           I
     SEDIMENTATION

          TANK

          (8 HR)
                           CO
              INFLUENT

              WASTEWATER
                   COMBINED CARBON

                     OXIDATION-

                    NITRIFICATION

                        TANK



                        (10 HR)
                     UJ
                     o
                     o
                     ID
                     _l
                     V)
                     UJ

                     QL
                    INTERMEDIATE

                    SEDIMENTATION

                      TANK  (4 HR)
              METHANOL
                   DENITRIFICATION
                     TANK  (2 HR)
AEROBIC  TANK (l HR)
                         I
                       FINAL

                   SEDIMENTATION

                    TANK  (4 HR)
                                      UJ
                                      o
                                      o
                                      3
                                      _l
                                      CO
                                                               UJ
                                                               cc
      DENITRIFIED
      EFFLUENT

  A. Bardenpho Process (29 hr)

                              5-63
                    DENITRIFIED
                    EFFLUENT


               B.  Alternative Methanol

                  Based  System (21 hr)

-------
process it would also work effectively for the comparative case. Solids retention time (and
hydraulic detention time)  of the nitrification step would be the same as for the nitrification
tank in the Bardenpho process.

Since denitrification rates are more rapid, the denitrification tank in the methanol based
system  can  be  proportionately  smaller than in  the  Bardenpho process.  Interpolating
denitrification rates from Figures 5-2  and  5-23,  about 2  hours would be required. A
residence time of 4 hours  is assumed for sedimentation tanks  in the methanol-based system,
whereas 8 hours is assumed for the Bardenpho process due to the bulking tendency of the
sludge. Comparing the two alternative systems, it can beseen that the alternating aerobic/anoxic
system requires greater tankage than the methanol-based system (29 hours compared to 21
hours). In terms of economics, differences  in the systems  can be seen as  a trade-off of
capital cost (tankage) with operating cost (methanol).

5.8 References

 1.  Barth, E.F., Brenner, R.C., and R.F. Lewis, Chemical-Biological Control of Nitrogen in
     Wastewater Effluent.  JWPCF, 40, No. 12, pp 2040-2054 (1968).

 2.  Parker D.S., Zadick,  F.J., and K.E. Train, Sludge Processing for Combined Physical-
     Chemical-Biological Sludges. Prepared for the EPA, Report No. R2-73-250, July, 1973.

 3.  Horstkotte, G.A., Niles, D.G., Parker, D.S., and D.H. Caldwell, Full-Scale Testing of a
     Water Reclamation System. JWPCF, 46 No. 1, pp 181-197 (1974).

 4.  Mulbarger,  M.C., The Three Sludge Systems  for Nitrogen  and Phosphorus Removal.
     Presented at the 44th Annual Conference of  the Water  Pollution Control Federation,
     San Francisco, California, October, 1971.

 5.  Sawyer,  C.N., Wild,  H.E.,  Jr., and T.C. McMahon,  Nitrification and Denitrification
     Facilities, Wastewater Treatment. Prepared for the EPA Technology Transfer Program,
     August, 1973.

 6.  Parker D.S., Case Histories of Nitrification and Denitrification Facilities. Prepared for
     the EPA Technology Transfer Program, May, 1974.

 7.  Murphy,  K.L., and P.M. Sutton, Pilot Scale Studies on  Biological Denitrification.
     Presented at  the 7th International Conference on  Water Pollution Research, Paris,
     September, 1974.

 8.  Bishop,  D.F.,  Personal  communication to D.S.  Parker.  Environmental Protection
     Agency, Washington,  D.C., April, 1974.

 9.  Mulbarger,  M.C.,  Nitrification  and Denitrification in Activated  Sludge Systems.
     JWPCF, 43, No. 10, pp 2059-2070 (1971).

                                        5-64

-------
10.  Dawson, R.N.,  and K.L. Murphy, Factors  Affecting Biological Denitriflcation  in
     Wastewater. In Advances in Water Pollution Research, S.H. Jenkins,  Ed., Oxford,
     England: Pergamon Press, 1973.

11.  Stensel, H.D., Loehr, R.C., and A.W. Lawrence,   Biological Kinetics of Suspended
     Growth Denitriflcation. JWPCF, 43, No. 2, pp 249-261 (1973).

12.  Sutton, P.M., Murphy, K.L., and R.N. Dawson, Continuous Biological Denitriflcation
     of Wastewater. Environmental  Protection Service (Canada), Report EPS 4-WP-74-6,
     August, 1974.

13.  Lawrence,  A.W., and P.L. McCarty, Unified Basis for Biological Treatment Design and
     Operation.  JSED, Proc. ASCE, 96, No. SA3, pp 757-778 (1970).

14.  Brown and Caldwell, Project Report for the  Water Reclamation Plant. Report to the
     Central Contra Costa Sanitary District, November, 1971.

15.  Weddle,  C.L.,  Niles,  D.G., Goldman, E.,  and J.W. Porter, Studies  of Municipal
     Wastewater Renovation for Industrial Water. Presented at the 44th Annual Conference
     of the Water Pollution Control Federation, San Francisco, California, October, 1971.

16.  Horstkotte, G.A.,  Jr., Pilot Demonstration Project for Industrial Reuse  of Renovated
     Municipal Wastewater. Prepared for the EPA, EPA-670/2-72-064, August, 1973.

17.  Requa, D.A., and E.D. Schroeder, Kinetics of Packed Bed Denitriflcation. JWPCF, 45,
     No. 8, pp 1696-1707(1973).

18.  Smith, J.M., Masse, A.N., Feige, W.A., and L.J. Kamphake, Nitrogen Removal from
     Municipal  Wastewater by  Columnar Denitriflcation. Environmental Science and Tech-
     nology, 6, p 260 (1972).

19.  Jewell,  W.J.,  and  R.J.  Cummings,  Denitriflcation  of Concentrated  Wastewaters.
     Presented at the Water Pollution Control Federation, Cleveland, October,  1973.

20.  Sutton,  P.M.,  Murphy,  K.L.,  and R.N.  Dawson, Low  Temperature Biological
     Denitriflcation of Wastewater. JWPCF, 47, No. 1, pp 122-134 (1975).

21.  Parker, D.S., Aberley, R.C., and D.H. Caldwell, Development and Implementation  of
     Biological  Denitriflcation for  Two  Large  Plants.  Presented at  the Conference  on
     Nitrogen as a  Water  Pollutant, sponsored by  the IAWPR, Copenhagen, Denmark,
     August, 1975.

22.  Denitriflcation  for Anaerobic  Filters and Ponds,  Phase  II.  Robert S. Kerr Water
     Research Center, EPA WPCRS 13030 ELY 06/71-14, June, 1971.

                                       5-65

-------
23. Denitrification for Anaerobic Filters and Ponds. Robert S. Kerr Water Research Center,
    EPAWPCRS 13030 ELY 04171-8, April, 1971.

24. Description  of the El Logo, Texas.  Advanced  Wastewater Treatment Plant. Seabrook,
    .Texas: Harris County Water  Control and Improvement District Number 50, March,
    1974.

25. Requa, D.A., Kinetics  of Packed Bed Denitrification. Thesis  submitted  in  partial
    satisfaction  of the requirements for the degree of Master  of Science in  Engineering,
    University of California at Davis, 1970.

26. English, J.N., Carry, C.W., Masse, A.M., Pitkin, J.B., and F.D. Dryden, Denitrification
    in Granular Carbon and Sand Columns. JWPCF, 46, No. 1, pp 28-42 (1974).

27. Savage, E.S.,  and J.J. Chen, Operating Experiences with  Columnar Denitrification.
    Pittsburgh, Pennsylvania: Dravo Corporation, 1973.

28. Lamb, G.L., Nitrogen Removal Utilizing Single Stage Activated Sludge and Deep Bed
    Filtration. Presented at the Joint Meeting of the Sanitary Engineering Section, ASCE,
    and the Metropolitan Section, NYWPCA, New  York, N.Y., March, 1972.

29. Ecotrol,  Inc., Biological Denitrification  Using Fluidized  Bed Technology. August,
    1974.

30. Jeris, J.,  Beer, C.,  and J.A.  Mueller, High Rate Biological Denitrification Using a
    Granular Fluidized Bed. JWPCF, 46, No. 9, pp  2118-2128 (1974).

31. Chen,  J.J., Letter communication to D.S. Parker, Dravo Corporation, November, 1974.

32. Kapoor,  S.A.,  and T.E. Wilson, Biological Denitrification on Deep Bed  Filters at
    Tampa, Florida.  Unpublished paper, Greeley and Hansen, Engineers, Chicago, 111., (no
    date).

33. Hansen, S.E., Letter communication to D.S. Parker. Neptune MicroFloc Inc., Corvallis,
    Oregon, September 11, 1974.

34. Duddles, G.A., Richardson, S.E., and E.F. Earth, Plastic Medium Trickling Filters for
    Biological Nitrogen Control. JWPCF, 46, No. 5, pp 937-946  (1974).

35. Duddles, G.A. and S.E. Richardson, Application of Plastic  Media Trickling Filters for
    Biological Nitrification.  Report prepared for the Environmental Protection Agency,
    EPA-R2-73-199, June, 1973.

36. Process Design Manual for Suspended Solids Removal. U.S.  EPA, Office of Technology
    Transfer, Washington, D.C., January, 1975.
                                       5-66

-------
37.  Jens, J.S., and F. J. Flood, Plant Gets New Process. Water and Wastewater Engineering,
     pp 45-48, March, 1974.

38.  Jeris, John S., and R.W. Owens, .Pilot Scale High Rate  Biological Denitrification at
     Nassau County, N. Y. Presented at the Winter Meeting of the New York Water Pollution
     Control Association, January, 1974.

39.  Owens, R., Letter communication to D.S. Parker, Ecolotrol Corp., October,  1974.

40.  Process Design Manual for Upgrading Existing Wastewater Treatment Plants. U.S. EPA,
     Office of Technology Transfer, Washington, D.C. (1974).

41.  Ecolotrol, Inc., Hy-Flo Fluidized Bed Denitrification,  Ecolotrol Technical Bulletin, No.
     123-A, November, 1974.

42.  Sullivan, R.H., Cohen, M.M., Ure, J.E., and F. Parkinson, The Swirl Concentrator as a
     Grit  Separator  Device.  Prepared  for the EPA by  the American Public Works
     Association, EPA-670/2-74-026, June, 1974.

43.  Manufacturing Chemists  Association, Properties  and Essential Information for Safe
     Handling and Use ofMethanol, Chemical Safety Data  Sheet SD-221, 1970.

44.  Austin,  George  T.,  Industrially Significant Organic Chemicals — Part 7.  Chemical
     Engineering, 81, No. 13, pp 152-153 (1974).

45.  Manufacturing  Chemists Association, Recommended Practice-Unloading Flammable
     Liquids from Tank Cars, Manual Sheet TC-4. 1969.

46.  Manufacturing Chemists  Association,  Tank Car  Approach Platforms, Manual Sheet
     TC- 7. 1965.

47.  National Fire Protection Association, Static Electricity 1966, NFPA No. 77. 1966.

48.  National Fire Protection Association, Flammable and Combustible Liquids Code 1972,
     NFPA No. 30, 1972.

49.  Smith, A.G., Denitrification Reactor Studies in a Lime  Treated Sewage Plant. Paper
     No. 2029, Ontario Ministry of the Environment, Research Branch, July, 1972.

50.  Tenney,  M.W., and W.F. Echelberger, Removal of Organic and Eutrophying Pollutants
     by Chemical-Biological  Treatment.  Prepared  for the  EPA, Report No.  R2-72-076
     (NTISPB-214628), April, 1972.
                                      5-67

-------
51.  Parker, D.S., and D.G. Niles, Full-Scale Test Plant at Contra Costa Turns Out Valuable
     Data On Advanced Treatment. The Bulletin (of the California Water Pollution Control
     Association), 9, No. 1, pp 25-27 (1972).

52.  Sutton, P.M., Murphy, K.L., and B.E. Sank, Nitrogen Control: A Basis for Design With
    Activated Sludge  Systems. Presented at the Conference  on  Nitrogen  as a Water
     Pollutant, sponsored by the IAWPR, Copenhagen, Denmark, August, 1975.

53.  Wuhrmann,  K., Nitrogen Removal In Sewage  Treatment Processes. Verh. Int. Ver.
     Limnol., 15, pp 580-596 (1964).

54.  Hunerberg and Sarfert, Experiments  on the Elimination  of Nitrogen  from Berlin
     Sewage. Gas u. Wasserfach, Wasser-Abwasser, 108, pp 966-969 and 1197-1205 (1967).

55.  Hamm, The Influence on Denitrification of Phosphate  Precipitants.  Z.  Wasser and
     Abwasserforschung, 2, pp 180-182 (1969).

56.  Schuster,  Laboratory  Experiments  on Removal of Nitrogen  Compounds  from
     Domestic Sewage. Fortschr. Wasserchem. Greng., 12, pp 139-148 (1970).

57.  Carlson, D., Nitrate Removal from Activated Sludge Systems. Report for  OWRR,
     Project AO26, July, 1970.

58.  Beer,  C., Hetling,  L.J., and R.E.  McKinney, Nitrogen Removal and Phosphorus
    Precipitation  in a  Compartmentalized Aeration Tank.  Technical Paper 32, New York
     State Department of Environmental Conservation, February, 1974.

59.  Christensen,  M.H.,  and P.  Harremoes, Biological  Denitrification  in  Wastewater
     Treatment. Report 2-72, Department of Sanitary Engineering, Technical University of
     Denmark, 1972.

60.  Bishop,  D.F.,  Heidman,  J.A.,   and  J.B.  Stamberg,  Single  Stage  Nitrification-
     Denitrification. Presented at  the 47th Annual Conference of the Water Pollution
     Control Federation, Denver, Colorado, October 6-11, 1974.

61.  Barnard, J.L., Cut P and N Without Chemicals. Water and Wastes Engineering, 11, No.
     7, pp 33-36 (1974) and No. 8, pp 41-94 (1974).

62.  Wuhrmann,  K., Effect of Oxygen Tension on Biological Treatment Processes.  Proc.
     Third Conference Biological Waste  Treatment, Manhattan College, N.Y., 1960,  pp
     27-38.

63.  Pasveer, A., Contribution on Nitrogen Removal from Sewage.  Muncher Bertrage zur
     Abwasser-Fisherei-and Flussbiologie, Bd 12, pp 197-200 (1965).

                                       5-68

-------
64.  Matsche,  N., The Elimination  of Nitrogen  in  the  Treatment Plant of  Vienna-
     Blumenthal. Water Research, 6, No. 4/5, pp 485-486 (1972).

65.  Matsche,  N.F., and G. Spatzierer, Austrian Plant Knocks Out Nitrogen. Water and
     Wastewater Engineering, 12, No.  1, pp 18-24 (1975).

66.  Drews,  R.J.L.C.,  and  A.M. Greeff, Nitrogen Elimination by Rapid Alternation of
     Aerobic/"Anoxic" Conditions  in "Orbal" Activated Sludge Plants. Water Research, 7,
     pp 1183-1194(1973).

67.  Halvorson, H.O.,  Irgens,  R., and  H.  Bauer,  Channel Aeration Activated Sludge
     Treatment at Glenwood, Minnesota. JWPCF, 44, No. 12, pp 2266-2276 (1972).

68.  Zemaitis,  W.L., Letter communication to D.S.  Parker.  Envirobic Systems, Inc., New
     York, New York, September 27,  1974.

69.  Christensen,  M.H., Denitrification  of Sewage by  Alternating Process Operation.
     Presented at the IAWPR, Paris, September, 1974.

70.  Barnard, J.L., Biological Denitrification.  Presented  at  the monthly  meeting of the
     South African Branch of the Institute of Water Pollution Control, August, 1972.

71.  Ludzack,  F.J., and M.B. Ettinger, Controlling Operation to Minimize Activated Sludge
     Effluent Nitrogen. JWPCF,  34, No. 9, pp 920-931 (1962).

72.  van Vuuren, L.R.J., Letter communication to D.S. Parker, February, 1975.

73.  Gujer, W. and D. Jenkins, The Contact Stabilization Process-Oxygen and Nitrogen Mass
     Balances.  University of California, Sanitary Engineering Research Lab, SERL Report
     74-2, February, 1974.

74.  Balakrishnan, S.,  and W.W. Eckenfielder, Nitrogen Relationships in Biological Waste
     Treatment Processes — III, Denitrification in the Modified Activated Sludge Process.
     Water Research, 3, pp 177-188 (1969).

75.  Schroeder, E.D.,  and  A.W. Busch,  The  Role  of Nitrate Nitrogen  In Bioxidation.
     JWPCF, 40, No. 11, pp R445-R457 (1968).

76.  Parker,  D.S., Kaufman,  W.J., and  D. Jenkins, Physical  Conditioning  of Activated
     Sludge Floe. JWPCF, 43, No. 9, pp 1817-1833 (1971).
                                       5-69

-------
                                    CHAPTER 6

                          BREAKPOINT CHLORINATION
6.1 Process Chemistry

When chlorine is added to dilute aqueous solutions containing ammonia nitrogen, reactions
occur which  may  lead ultimately to oxidation  of  the  ammonium ion to  end products
composed predominantly of nitrogen gas. When such chemical processes are performed in
water and wastewater treatment  for  the  purpose  of ammonia  nitrogen  removal,  the
procedure is  termed breakpoint  chlorination.  This chapter  discusses  the  theoretical
stoichiometry  of breakpoint chlorination,  presents  the practical process considerations
which influence actual  chemical consumption,  reaction end  products and rate  of  the
reaction, and presents process design criteria.

Recent work at the Blue Plains  wastewater treatment pilot plant in  Washington, D.C.I>2>3
has confirmed breakpoint chlorination reaction products. Gas chromatography was used at
Blue Plains to  identify  breakpoint reaction products from  buffered  aqueous wastewater
samples in laboratory tests. Further confirmation of  breakpoint reaction end products was
obtained in pilot scale investigations with wastewater effluent of different qualities.

Breakpoint chlorination tests on domestic wastewaters at the Blue Plains pilot plant showed
that 95  to 99 percent of the ammonia nitrogen in solution is converted to nitrogen gas.2>3
No breakpoint reaction intermediate  compounds of N2O, NO or NO2 were detected. The
oxidized ammonia  nitrogen fraction which did not appear as nitrogen  gas was found to be
made up of nitrate and nitrogen  trichloride.

     6.1.1 Chemical Stoichiometry

When chlorine gas is dissolved in water, hydrolysis of the chlorine molecule occurs according
to the following relationship:

                         C12 + H20 ?=Z HOC1 + H+ + Cf                     (6-1)
                                        (hypochlorous acid)

The active (oxidizing)  forms  of chlorine  in  solution are hypochlorous  acid and  its
dissociation product, hypochlorite ion.


               HOC1 5^=r  OC1~ + H+       K = 3.3 x 10~8                     (6-2)
                                               at20C
                                           where: K = dissociation constant

                                        6-1

-------
The fraction of the total chlorine residual in a sample which is made up of hypochlorous
acid  and hypochlorite ion is termed the "fiee  available" chlorine residual. The rate of
dissociation of HOC 1 is very rapid and equilibrium proportions are maintained even when
HOC1 is continuously being reacted. The equilibrium relationship between HOC1 and OC1~
in relation to solution pH is shown in Figure 6-1.
                                 FIGURE 6-1

                  RELATIVE AMOUNTS OF HOC1 AND OCf AT
                    VARIOUS pH LEVELS (REFERENCE 4)
                  too
                                                            too
                                               10    It    12
The oxidizing capability of the free available chlorine residual is manifest in the chemical
transformation of hypochlorous acid to chloride ion (Cl~). This transformation involves a
gain of two electrons and a valence change of the chlorine atom from "+1" to "-1".

Ammonia nitrogen  concentrations  of 10  mg/1  to  40 mg/1  may  be found  in typical
municipal wastewater treatment plant effluents. The source of ammonia nitrogen typically
includes direct discharge from industrial processes and release following hydrolysis of urea
and biological degradation of amino  acids  and other organic  derivatives of ammonia
nitrogen. The actual chemical form  of ammonia nitrogen in solution is pH and temperature
                                       6-2

-------
dependent. The  relative distribution of ammonia nitrogen and ammonium ion may be
defined according to the equation below:
                             NH,
       K = 5.0x10
          at20C
                                                   -10
(6-3)
This relationship is indicated graphically in Figure 6-2 in relation to pH of the solution.
Reactions  between  chlorine and  ammonium  in  dilute  aqueous  solution  can  proceed
according  to  a number of competing pathways. Formation  of chloramines, termed
"combined residual" and nitrate can occur in the following manner:
                    + HOC1
NH2C1 (monochloramine) + F^O + H
 (6-4)
                  NH2C1 + HOC1
    NHCL, (dichloramine) + HO
         ^                 2
(6-5)
                                 FIGURE 6-2
            EFFECTS OF pH AND TEMPERATURE ON DISTRIBUTION
                 OF AMMONIA AND AMMONIUM ION IN WATER
                 IOO
                                             10    II   12
                                                         IOO
                                     6-3

-------
              NHC12 + HOC1  - - NC13 (nitrogen trichloride) + H2O            (6-6)

                                     and

                  NH* + 4HOC1 - - HN03 + 5H+ + 4Cf + H2O               (6-7)


The  reactions  are  dependent  upon certain process variables, including pH, temperature,
contact time, and the initial chlorine to ammonia nitrogen ratio
Breakpoint chlorination occurs  when sufficient  chlorine has been added  to  a  water or
wastewater sample  to  cause  the chemical  oxidation of the ammonium in solution to
nitrogen gas and other end products.

Significant aspects of breakpoint chlorination process chemistry which were studied during
the Blue Plains pilot work included identification of the predominant end products of the
breakpoint reaction. Tests on municipal wastewater and wastewater treatment plant effluent
at Blue Plains indicated that 95  to 99 percent of the ammonia nitrogen in solution was
converted  to nitrogen  gas. Nitrate and nitrogen  trichloride account  for  the remaining
fraction. The overall reaction between the ammonium ion and chlorine leading to formation
of nitrogen gas may be expressed in terms of the simplified equations below:

                      NH* + HOC1 - - NH2C1 + H2O + H+                   (6-4)


            NH2C1 + 0.5 HOC1  - - 0.5 N2 + 0.5 H2O + 1 .5 H+ + 1 ,5 Cf        (6-8)
             NH* + 1.5 HOC1  	^ 0.5 N2 + 1.5 H2O + 2.5 H+ + 1.5 Cl          (6-9)


Stoichiometrically, the  breakpoint  reaction  of Equation 6-9 requires a weight ratio of
chlorine to ammonia nitrogen at the breakpoint of 7.6:1, as shown below:
                                               v

                  Molecular Weight HOC1 = 70.9 (expressed as CU)

                  Molecular Weight NH. = 14.0 (expressed as N)

                  Therefore, weight ratio of CLrNH, -N at breakpoint:

                  C12:NH+-N = (1.5X70.9): (0(14.0) = 7.6:1
                                       6-4

-------
Therefore 7.6 parts of chlorine are theoretically required to chemically oxidize one part of
ammonia nitrogen in aqueous solution. In practice, the actual weight ratio of chlorine to
ammonia nitrogen at breakpoint has ranged from about 8:1 to  10:1. Many of the process
variables which are known to affect the total chemical requirement for this  process have
been identified and those factors are discussed in subsequent sections.

     6.1.2 The Breakpoint Curve

The  breakpoint chlorination curve is a graphical representation of chemical relationships
which  exist as varying amounts of chlorine are added  to dilute solutions  of ammonia
nitrogen. An investigation in 1939 led to the discovery that increasing the chlorine dose in
certain waters resulted in an overall reduction in the chlorine residual measured in the water
sample  following contact. ^ The point of  maximum  reduction of chlorine residual was
termed the "breakpoint".

The  theoretical breakpoint curve shown in Figure 6-3 has several characteristic features. The
characteristics of the breakpoint curve shown in Zone 1 include  principally  the reaction
between chlorine and ammonium indicated in Equation 6-4. The hump  of the breakpoint
curve occurs, theoretically, at a chlorine to ammonia nitrogen weight ratio  of 5:1 (molar
ratio of 1:1). That  ratio corresponds to the point at which the reacting molecules are
present in solution in equal numbers.                       "

The  chemical equilibria of Zone 2 favor the formation of dichloramine (Equation 6-5) and
the  oxidation  of ammonium  according  to  Equation  6-9. These  reactions proceed in
competition to, theoretically, a Cl2:NH4-N weight ratio  of 7.6:1. At the breakpoint, the
ammonium concentration is minimized.

To the right of breakpoint, Zone 3 chemical equilibria include the build-up of free chlorine
residual as well as the presence  of small quantities of dichloramine (Equation 6-5), nitrogen
trichloride (Equation 6-6), and nitrate (Equation 6-7)1 The free chlorine residuals which
result from  dosages beyond breakpoint are known to be considerably more bactericidally
potent than the combined residuals found at lower chlorine dosages (See Section 6.2.7).

6.2 Process Application Considerations

The  basic theoretical background  chemistry must be combined  with application  funda-
mentals if proper process designs and operations are to be achieved in full scale wastewater
treatment practice. Much background work on breakpoint chlorination has been done on a
laboratory "pure  system" basis and  in potable water. This information is useful, but not
always applicable to wastewater treatment considerations. Generally, the presence of high
concentrations of ammonia nitrogen and  other amino  substances as well as the presence of
other chemical constituents in wastewater effluents contribute to the discrepancy between
some data  available  from laboratory testing and that  collected in  actual  wastewater
treatment applications.

                                         6-5

-------
                                 FIGURE 6-3
             THEORETICAL BREAKPOINT CHLORINATION CURVE
                                                      TC TAL
                                                      CHLORINE
                                                      APPLIED
                                                         MEASURED
                                                         CHLORINE
                                                         RESIDUAL
                            BREAKPOINT-
                      IRREDUCIBLE
                      MINIMUM CHLORIN
                      RESIDUAL
                                                                         1

                                        5              7.6

                      CI2:NH^-N WEIGHT  RATIO
     6.2.1 Chlorine Dosage Requirement

The total amount of chlorine which must be added to wastewater to achieve breakpoint is
affected by the chemical nature of the wastewater and by the conditions which exist in the
zone where the reacting species come into contact. Several important factors which should
be considered in breakpoint chlorination process application and design are discussed below.

           6.2.1.1 Effect of Pretreatment

The  degree  of treatment which  a  wastewater stream receives prior to breakpoint
chlorination effects both the chlorine dosage and end-product distribution. The relationship
between pretreatment and nitrate and nitrogen trichloride formation is discussed in Section
6.2.2.
                                     6-6

-------
The chlorine demand of a wastewater treatment plant effluent sample is the total chlorine
oxidative capacity consumed during a given period of time by substances in solution which
do  not result in measurable chlorine residual or chemical oxidation of ammonia nitrogen.
This is the chlorine oxidative capacity which is essentially "lost" from participation in the
desired breakpoint reaction. Chlorine demand may be exerted by a number of substances
commonly present in wastewater,  including S^— ,HS~SC)2~ ,NO~ Fe2+,  phenols, amino
acids,  proteins and carbohydrates. 9

Generally,  as the degree  of pretreatment of wastewater is increased, the chlorine demand
exerted by the  substances noted above is reduced. One particularly important factor is that
if a  treatment process  employing  anoxic  conditions  precedes a chlorination  facility,
substances  in solution may be  converted from an  oxidized to a reduced form and the
chlorine demand may be substantially increased.

Laboratory and pilot  plant studies  of breakpoint chlorination at  the Blue Plains pilot
plant ^>2 and at Sunnyvale 10»H have shown  that increasing levels of pretreatment decrease
the amount of chlorine required to  achieve breakpoint. Table  6-1 shows  that laboratory
tests on buffered distilled water containing only ammonia nitrogen reached breakpoint at a
Cl2:NH4-N ratio of 8:1, a level  near that predicted by chemical stoichiometry (7.6:1). In
comparison,  raw wastewater required  a Cl2:NH4-N  of 9:1 -  10:1  to reach breakpoint.
Ammonia nitrogen  concentrations in  the samples  following breakpoint were found to be
consistently  in  the  range  of 0.2 mg/1 or  less.  Pilot plant scale  testing of breakpoint
chlorination processes  has  confirmed chlorine  dosages predicted through laboratory work.

         6.2.1.2 Effect of pH and Temperature

                               1 9
Laboratory studies at Blue Plains >  in which buffered distilled ammonia nitrogen solutions
of  20 mg/1 concentration  were subjected to  breakpoint  chlorination dosages  showed a
definite optimum pH for breakpoint  in the range of pH 6 to  7. The chlorine dosage at
optimum pH levels was found to  be Cl2:NH4-N of 8:1. Breakpoint tests conducted outside
the apparent optimum  range  of  pH  6 to  7 showed  an appreciably  higher chlorine
requirement for breakpoint and slower reaction rates.

Comparable tests carried  out with filtered secondary effluent did not show a clearly defined
relationship  between  pH  and Cl2:NH4-N  to  reach breakpoint.  Formation   of  other
nitrogenous residuals (NO~3 and  NC13) was considered to  be  the  controlling  criteria in
selection of the optimum  pH operating range of pH 6 to 7 (See Section 6.2.2).

There is no evidence that  ordinary variations  in the  temperature of wastewater effluents
affect the C^NHiJ-N to reach breakpoint.
         6.2.1.3 Initial Mixing of Chlorine

The significance of initial mixing in certain unit processes of sanitary engineering has been

                                         6-7

-------
                                          TABLE 6-1
                      EFFECT OF PRETREATMENT ON Cl^NHj-N BREAKPOINT RATIO
Sample


Buffered water
Raw wastewater
Lime clarified
raw wastewater
Secondary effluent
Lime clarified
secondary effluent
Ferric chloride
clarified raw
wastewater-
carbon adsorption

Filtered secondary
effluent
Lime clarified raw
wastewater-filtered
Alum clarified
oxidation pond
effluent-filtered
Breakpoint
PH


6-7
6.5 - 7.5

6.5 - 7.5
6.5 - 7.5

6.5 - 7.5



3.2


6-8

7.0 - 7.3


6.6
Initial
NH . -N
(mg/D

Final
NH4 -N
(mg/1)

Irreducible
minimum
residual
(mg/1 as C12)
Laboratory Tests
20
15

11.2
8.1

9.2



10.2
0.1
0.2

0.1
0.2

0.1



0.1
0.6
7

7
3

4



20
Pilot Plant Tests

12.9 - 21.0

9.7 - 12.5


20.6

0.1

0.4 - 1.2


0.1

2 - 8.5

-


7.6
Breakpoint
ratio
C12:NH+ -N
(weight basis)

8:1
9:1 - 10:1

8:1-9:1
8:1-9:1

8:1



8.2:1


8.4:1 - 9.2:1

9:1


9.6:1
Ref


2
2

2
2

2



12


2

2


11
oo

-------
amply demonstrated.  Tests 13  have  shown  significantly  improved  alum  coagulation
efficiency as a direct result of increasing the level of turbulence of mixing in the zone of
alum  application to a water sample. Improved disinfection efficiency in laboratory tests was
also noted when the application of chlorine to a wastewater effluent was accomplished at
increased levels of turbulent mixing.

Recent  data from Blue  Plains^  have shown  the total chlorine dose required to reach
breakpoint  is not affected by initial mixing  conditions  to the  degree  which  was first
reported. A number of pilot plant tests with secondary effluent in which  the chlorine was
dosed into a mechanical mixer showed no difference in the efficiency of chlorine utilization
whether  or  not  the  mixer was operating. Other laboratory studies  have shown  the
Cl2:NH4-N  ratio at breakpoint to be  unaffected by the degree of mixing in the reaction
zone.

In plant-scale  design  of breakpoint chlorination facilities, a  quantity  of hydraulic  or
mechanical  energy  sufficient  to  facilitate rapid and  thorough blending of the chlorine
solution, pH adjustment  chemical and process influent should be provided. The blending of
chemicals with the process influent initiates the breakpoint  reactions and allows completion
of the breakpoint reaction within the contact zone, provides the basis for containment of
process  odors and assures process consistency, a necessary prerequisite  for the  feedback
element of process control (Section 6.3).

     6.2.2 Residual Nitrogenous Materials

Nitrate (NO^) and nitrogen trichloride (NC13)  are occasionally found in the effluent from
the breakpoint  chlorination process.  Both  compounds can be  found  in varying  con-
centrations, depending upon the degree of pretreatment and pH in the reaction zone. The
total  concentration of these residuals seldom exceeds  10 percent of the influent  ammonia
nitrogen concentration.2

Nitrogen  trichloride is a particularly volatile  compound which  exhibits  a  very strong
chlorinous odor. It is an extremely strong-oxidizing agent, having been used for many years
in the bleaching of flour. Formation of NC13 in breakpoint chlorination, even in fairly small
concentrations, is undesirable because of the obnoxious and dangerous characteristics of the
compound in the gaseous form. Nitrate formation in breakpoint  chlorination should  be
avoided since it represents a reaction product which has consumed considerable amounts of
chlorine (Equation 6-7) and because it reduces the overall nitrogen removal capability of the
breakpoint process.

Table 6-2 presents  data  on  the  effect  of wastewater pretreatment  on  the formation of
residual nitrogenous materials at breakpoint. * >2 Although the investigators concluded that
NC13  formation decreased with decreasing pretreatment, it appears that very highly treated
effluent may be  breakpointed with the production of negligible levels of NC13.  Nitrate
production was similar at each treatment level.

                                         6-9

-------
                                   TABLE 6-2

                EFFECT OF PRETREATMENT ON FORMATION OF
         NITROGENOUS RESIDUALS AT BREAKPOINT (REFERENCE 1, 2)
Sample
Raw Wastewater
Lime Clarified and
Filtered Raw Wastewater
Secondary Effluent
Lime Clarified and
Filtered Secondary
Effluent
Initial
Ammonia-N Cone .
(mg/1)
15.0
11.2
8.1
9.2
NC13 Cone, at
Breakpoint
(mg/1 as N)a
0.0
0.25
0.13
0.0
NO^ Cone . at
Breakpoint
(mg/1 as N)a
0.3
0.45
0.24
0.2
 pH Range = 6.5  to  7.5


Laboratory and pilot-scale tests have confirmed the pH sensitivity of NC13 formation in
breakpoint chlorination. Pilot-scale tests of filtered secondary effluent showed 0.33 mg/1
NC13 (as N) after breakpoint chlorination at pH 6; at pH 7 and above, NC13 concentration
was reduced to about 0.05 mg/1. Nitrate formation was found to be slightly affected by pH
at  breakpoint. Nitrate concentration in pilot tests of filtered secondary effluent ranged from
about 0.7 mg/1 (as N) at pH 6 to 1.0 mg/1 (as N) at pH 8.

Figure 6-4 shows the consequences of chlorine dosages beyond breakpoint on the formation
of nitrogenous residuals. Nitrate  production  following breakpoint chlorination of lime
clarified filtered secondary effluent (breakpoint  at Cl2:NH4-N of  8:1) showed a linear
increase in concentration at increased chlorine  dosages. Nitrogen trichloride formation was
noted beginning at  C^Ntfy-N  ratios exceeding 9:1. Close control of chlorine dosage levels
in  breakpoint is shown to  be an important  factor in minimizing production of  these
residuals.

Pilot testing of breakpoint chlorination at Blue Plains showed that variations in temperature
from 5 to 40 C did not affect the  final distribution of nitrogenous residuals in the effluent.

Awareness of the pH sensitivity of formation of NO3 and NC13 has led to formulation of
two specific recommendations  for design of plant-scale breakpoint chlorination facilities.
First,  the pH adjustment chemical  (Section  6.2.3) should be added to the chlorine solution
prior to application of the chlorine solution to the breakpoint chlorination process influent.
Without premixing of chemicals,  disproportionate mixing of chlorine  solution and pH
                                       6-10

-------
                                     FIGURE 6-4

                EFFECT OF Cl^NH^ -N ON NITROGEN RESIDUALS IN
            LIME CLARIFIED FILTERED SECONDARY EFFLUENT (REF. 1)
     O.8
     0.7
^   °6
 *   0.5
 o
«   O4


\   0-3

Uj
%
AMMONIA-N CONC. = 9.2 mg/t

  pH  RANGE  6.5-7.5
Cl2:NHj-N  DOSAGE AT BREAKPOINT = 7:1 to 8:1
     O.2
     O.I
     O.O £*	&-
                                      NITRATE
                     -o-
                     -^
                                                   NITROGEN
                                                   TRICHLORIDE
                                                         8
                                                   IO    II
12
                                CL2--NH4~N WT. RATIO
 adjustment chemical in the breakpoint reaction zone can result in "pockets" of liquid in the
 breakpoint reaction zone not at the desired pH level. Occurrence of the breakpoint reaction
 in such "pockets" could lead to formation of excessive concentrations of NO^ and NC13.

 A second design recommendation is that the operating pH for breakpoint chlorination
 should be maintained at pH 7. This pH allows optimal Cl2:NH4-N ratios and reaction rates
 and also reduces  the NC13 production and attendant odor production to about 0.2 mg/1 or
 less, as long as careful control over chlorine dosage is maintained.

 In summary, the  design engineer should be cautioned that a poorly designed or maintained
 system for breakpoint chlorination  pH  and chlorine dosage  control  can result in the
 generation of chlorinous odors  due to  the formation  of nitrogen trichloride. However,  it
 appears that if appropriate care is exercised in the design and maintenance of pH and
 chlorine dosage systems,  the breakpoint chlorination process  can  function satisfactorily
 without the need  for odor control facilities.
                                       6-11

-------
     6.2.3 Alkalinity Supplementation

In the  addition  of chlorine  to  a wastewater  sample for the  purposes of breakpoint
chlorination, acidity is generated through the hydrolysis of chlorine gas (Equation 6-1) and
oxidation of ammonium (Equation 6-9). Theoretically, four moles of hydrogen ions are
generated for every mole of ammonia nitrogen which is oxidized with chlorine gas.


                 l.SCU+l.SILO - +• 1.5HOC1+1.5 H++ 1.5 Cf             (6-1)
              NH* + 1 .5 HOC1 - - 0.5 N2 + 1 .5 H2O + 2.5 H+ + 1 .5 Cf         (6-9)
                      1.5 C12 + NH* 	- 0.5 N2 + 4H+ + 3C1                 (6-10)


The  buffering capacity (alkalinity) of the breakpoint process influent is consumed by the
acidity  of Equation 6-10.   Stoichiometrically, 14.3  mg/1 of alkalinity (as CaCO3)  is
consumed  for each  1.0  mg/1 of  ammonia nitrogen which is  oxidized in breakpoint
chlorination.  In actual practice, around  15 mg/1 alkalinity is consumed due to the hydrolysis
of chlorine needed beyond that predicted by stoichiometry.

It is  apparent that if the ammonia nitrogen concentration in the breakpoint process influent
is high,  or if  the wastewater alkalinity is relatively low, insufficient buffering  capacity will
be available to maintain the process pH at a reasonable level. Any alkaline substance may be
used to supplement the natural alkalinity for pH control but sodium hydroxide (NaOH) and
lime (CaO) are  the  compounds most  commonly used. If  all of  the acidity  generated in
breakpoint must be neutralized through chemical addition,  1.50 parts of sodium hydroxide
(NaOH) would be needed for each part of chlorine, or 1.05 parts of lime (CaO) would be
needed  for each part of chlorine.  In practice, the alkalinity  of the process  influent may
supply a portion of the total buffering capacity needed under breakpoint conditions. If a
highly alkaline stream is  to  be treated (such  as might result from a lime  precipitation
process), alkalinity supplementation would not be needed and, perhaps, acid addition would
be needed to achieve the recommended operating pH of 7.  In situations where breakpoint
chlorination  is  used for removing low  ammonia  nitrogen residuals (< 3 mg/1)  from
nitrificaton process  effluents, there  may be sufficient remaining alkalinity  to  buffer the
water without the need for addition of neutralizing chemicals.

      6.2.4 Reaction Rates

The  reaction rate  for breakpoint chlorination has not been  measured  quantitatively, but
several investigations^! 1 have noted that the reaction is very rapid. According to reaction

                                        6-12

-------
rates established by Morris,6>7 the reaction between ammonium and hypochlorous acid
(Equation 6-4) to form monochloramine is 99 percent complete in 0.2 seconds at pH 7 and
           of 0.2:1. The  optimum reaction pH  was shown to be pH 8.3, allowing 99
percent conversion to  monochloramine  in 0.069  seconds. Obviously,  the  rate of this
reaction is important in breakpoint since it is the first of several sequential reactions.

Laboratory studies have shown the breakpoint reaction rate to vary, depending upon pH. At
pH levels between 6 and 7, the breakpoint reaction was found to proceed to completion in
less than 15 seconds when secondary effluents were tested. When reaction rates at pH levels
outside  the  optimum  range  of pH 6-7 were tested, the rate  was  found to be slowed
considerably.  At pH 3.5, for example, the breakpoint chlorination reactions were not
complete following two hours of contact.

The rapid rates noted for breakpoint reactions conducted in the pH range of 6-7 lead to the
design  recommendation  that a breakpoint  chlorination contact period of one minute is
sufficient  for plant-scale applications. The design of the  contact basin should provide, as
closely as possible, a plug flow contact regime. The blending of breakpoint chemicals with
the process influent should be carried out as indicated in Section 6.2. 1.3.

     6.2.5 Effect on Total Dissolved Solids
In many instances where high levels of wastewater treatment are required, total dissolved
solids (TDS)  limitations may be  a controlling  criterion in the  selection  of alternative
treatment processes.  Breakpoint chlorination can involve the addition of large quantities of
chemicals to solution. The TDS increment attributable to each of several chemicals which
may be utilized in breakpoint chlorination is summarized in Table 6-3.

                                  TABLE 6-3
                  EFFECTS OF CHEMICAL ADDITION ON TOTAL
               DISSOLVED SOLIDS IN BREAKPOINT CHLORINATION
      Chemical Addition
TDS Increase :  NH4~N Consumed
 Breakpoint with chlorine gas

 Breakpoint with sodium
      hypochlorite

 Breakpoint with chlorine gas  -
      Neutralization of all
      acidity with lime  (CaO)

 Breakpoint with chlorine gas  -
      Neutralization of all acidity
      with sodium hydroxide  (NaOH)
               6.2  :  1
               7.1  :  1
              12.2  :  1
              14.8  :  1
                                      6-13

-------
If breakpoint chlorination is contemplated on a wastewater effluent stream which contains
20 mg/1 ammonia nitrogen, the increase of TDS in solution following addition of chlorine in
the gaseous form would amount to 124 mg/1. If all of the acidity generated in the hydrolysis
of chlorine and  the oxidation of ammonium is neutralized  with lime (CaO), the total
increase in TDS would amount to 244 mg/1.

     6.2.6 Reactions with Organic Nitrogen

Studies at Blue Plains^ found only "slight reduction in organic nitrogen within the two hour
contact  time." The data of Lawrence, et al. ^ showed a decrease in organic nitrogen from
3.2 mg/1-3.5  mg/1 to levels of 0.2  mg/1 - 0.4 mg/1.   More recent data^ collected in
Sunnyvale, California also show an apparent decrease in soluble organic nitrogen following
breakpoint (Table 6-4).

                                   TABLE 6-4

                   EFFECT OF BREAKPOINT CHLORINATION
              ON SOLUBLE ORGANIC NITROGEN (REFERENCE  1 l)a

Date, 1973

8/30
9/4
9/11
9/12
Soluble Organic Nitrogen
before Breakpoint
(mg/1 as N)
2.7
2.8
4.6
4.7
Soluble Organic Nitrogen
after Breakpoint
(mg/1 as N)
1.0
1.7
1.3
2.0
  Soluble organic nitrogen determination conducted on filtrate
  from 0.45/x membrane filter.  Wastewater treated was tertiary
  treated oxidation pond effluent
        reported in 1953 that the concentration of unsubstituted ammo nitrogen of many
common amino acids was reduced slowly by reaction with chlorine. Organic nitrogen in the
more complex form of proteins  was practically unaffected by chlorine over a period of
several days. The  degree of organic nitrogen reduction through breakpoint chlorination is
likely a function of the relative proportion of proteins to the simpler hydrolytic products
(including amino acids) of the protein molecules. In summary, the true reductions in organic
nitrogen with breakpoint chlorination are difficult to predict.

The  presence of organic nitrogen in solution at breakpoint has also been shown to effect the
shape of the breakpoint curve. ^ It has been noted that waters containing  a  mixture of
ammonia nitrogen and organic nitrogen did not display the classic "dip" of the breakpoint
curve. 17,18,19 The irreducible minimum residual was found to be appreciably greater when
                                       6-14

-------
organic nitrogen was present than when the sample contained only the inorganic ammonia
nitrogen form.

     6.2.7 Disinfection

Under normal conditions, disinfection of a non-nitrified wastewater effluent stream with
chlorine is mainly  accomplished by  the  form of  combined chlorine residual known as
monochloramine (NH2C1 at pH 7.0). When contact for disinfection is provided downstream
from  the breakpoint chlorination process, the free chlorine residual in solution following
breakpoint will provide much  higher bactericidal potential than  with  monochloramine
alone.

Figure 6-5 is  a comparison of the germicidal efficiency of hypochlorous acid, hypochlorite
ion and monochloramine.  Figure  6-1 shows the  ionization  characteristics of the hypo-
chlorous acid at various pH levels. The data of Figure 6-5 show that hypochlorous acid is a
far more  effective germicidal  agent than either hypochlorite ion or monochloramine.
Formation of  hypochlorous  acid  following  breakpoint  chlorination  will,  therefore,
considerably  enhance the  capability  of a wastewater disinfection  system if an efficient
contacting system is .provided.

6.3 Process Control Instrumentation

Both  South African^! and American researchers2>22 have reported that if a continuously
functioning breakpoint chlorination process is to be a consistent and reliable environmental
engineering unit process, the system  must be capable of responding rapidly to changes in
ammonia nitrogen concentration, chlorine demand, pH, alkalinity and flow.

     6.3.1 Process Control System

Failure of the chlorine dosage control system to respond adequately to changes in  process
conditions  can result in a substantial  loss  in nitrogen removal  capability as  well as
potentially adding  significant overdoses of chlorine to the process. Overdoses of chlorine to
the system are a direct waste of this chemical, they result in increased difficulty of pacing
the dechlorination equipment,and can cause the direct discharge of high concentrations of
chlorine residuals to the receiving water.

         6.3.1.1 Chlorine Dosage Control

A  function diagram of the breakpoint chlorination process control system is shown in
Figure 6-6. This system is the  same  as that  used  in  the pilot plant testing of breakpoint
chlorination at the Blue  Plains pilot plant in Washington, D.C.  The  control of chlorine
dosage is accomplished by a combination feed foreward and feedback control loop. The
feed foreward component utilizes ammonia concentration and flow  signals  together with a
manually  selected  multiplier  to establish an approximate  chlorine  dosage to  achieve

                                        6-15

-------
                     FIGURE 6-5

       COMPARISON OF GERMICIDAL EFFICIENCY OF
      HYPOCHLOROUS ACID, HYPOCHLORITE ION, AND
     MONOCHLORAMINE FOR 99 PERCENT DESTRUCTION
              OF E. COLI AT 2-6 C (REF. 20)
                                 MONOCHLORAMINE
                 HYPOCHLORITE
                 ION
                     HYPOCHLOROUS ACID
O.OOI
                   IO             IOO
                      TIME, MIN.
IOOO
                        6-16

-------
                                              FIGURE 6-6
                      BREAKPOINT CHLORINATION CONTROL - FUNCTIONAL SCHEMATIC
   FEED -
 FOREWARD
 CONTROL
COMPONENT
      AMMONIA -N
    CONCENTRATION
   (PROCESS INFLUENT)
                         PROCESS
                         FLOWRATE
(MULTIPLY)
    i
                                                                 SIGNAL FOR AMMONIA MASS FLUX
                                                PRESELECTED
                                                 MULTIPLIER
                                                     I
                                                 (MULTIPLY)
                                                     I
                                          (CORRESPONDS TO CI2:NH4 - N)
                       SIGNAL FOR COMPUTED
                       CHLORINE REQUIREMENT
   FEED-
   BACK
 CONTROL
COMPONENT
      MEASURED
    FREE CHLORINE
RESIDUAL CONCENTRATION
  (PROCESS EFFLUENT)


     PRESELECTED
    FREE CHLORINE
RESIDUAL CONCENTRATION
(SUBTRACT)
                                                           ..J
ELECTRONIC
FUNCTION
GENERATOR
1
i
	 1
^-CHLORINATION
PACING
SIGNAL
                                                                    SIGNAL INDICATING DEPARTURE
                                                                    FROM DESIRED FREE CHLORINE
                                                                    RESIDUAL IN PROCESS EFFLUENT

-------
breakpoint.  The  manually  preselected multiplier corresponds  to the C^.'NH^-N ratio
required for breakpoint and varies from values of about 8 to 10.

The feedback control loop involves measurement of the free chlorine residual in the process
effluent. The level of measured free  chlorine residual is compared to a setpoint value
(usually 2-4 mg/1)  using  a  standard process controller and a signal is generated which
provides a "trimming" of the chlorine dosage to the system.

          6.3.1.2 pH Control

Under most circumstances, a base is added to the breakpoint process to neutralize a portion
of the  acidity generated in  the chlorine  addition (Equations 6-1  and  6-9).  The base
requirements at  a particular installation are related to  wastewater alkalinity, individual
treatment processes employed prior to breakpoint chlorination as well as pH or alkalinity
restrictions which might be imposed upon the breakpoint effluent.

pH control  in  the breakpoint chlorination process  may be  effectively achieved using a
combination feed foreward and feedback system. The feed foreward component of the
system  involves pacing the base addition  directly on the chlorine application rate. This
accomplishes neutralization of a preselected portion of the chlorine acidity.  A feedback
loop is employed to "trim" the system to a designated pH  level. Accurate pH control of the
process promotes efficient utilization of chlorine and can reduce or eliminate undesirable
end products of the reaction.

     6.3.2 Process Control Components

Effective  control over breakpoint chlorination requires utilization of accurate and reliable
sensory equipment. In this  regard two components deserve special attention; namely the
ammonia nitrogen and free chlorine monitoring elements.

The ammonia nitrogen and free  chlorine  monitoring device which was used during the Blue
Plains  pilot  testing of breakpoint chlorination was an automated wet chemical analyzer
using  colorimetric techniques  to measure the  ammonia  nitrogen  and free  chlorine
concentrations  in small individual samples of the process influent. A continuous-duty free
chlorine residual  analyzer was also successfully used at Blue Plains to provide a feedback
signal for breakpoint chlorination control.

6.4 Dechlorination Techniques

Dechlorination  is a unit process  which eliminates  the  active chlorine residual in solution
prior to effluent discharge or additional treatment steps. Current emphasis on elimination of
chlorine residual  from  wastewater treatment  effluents  has resulted  in dechlorination
requirements in many cases. Frequently, dechlorination  will be required as a companion
process to breakpoint chlorination. Dechlorination techniques using sulphur  dioxide and
activated carbon may be used.
                                        6-18

-------
     6.4. 1 Sulphur Dioxide Dechlorination

Sulphur dioxide is a colorless, odorless gas which hydrolyzes in aqueous solution to form
sulfurous acid, a strong reducing agent. When added to a sample containing active chlorine
residual (oxidizing agent), the chlorine residual is converted to a non-toxic form, normally
chloride ion.

          6.4.1.1 Stoichiometry

Sulphur dioxide reacts with both free and combined forms of chlorine residuals according to
the expressions noted below:

a. Reaction with free chlorine residual (i.e. HOC1)
                           SO2 + H2O  - ~ HSO~ + H                       (6-11)
                        HOC1 + HSO~ - - Cf + SO^ + 2H+                  (6-12)
                           HOC1 + HO	 Cf + SO4 + 3H+                (6-13)
 h. Reaction with combined chlorine residual (i.e. NH2C1)
                           SO2 + H2O  	^ HSO3 + H+                      (6-11)
                  NH2C1 + HSO3 + H2O 	^ Cf + SO4 + NH* + H+           (6-14)
                 S02 + NH2C1 + 2H20	 Cf + S04 + NH+ + 2H+           (6-15)
Several important observations can  be  made  from the chemical reactions presented in
Equations 6-11 to 6-15.  First, both  combined and free chlorine residuals  are reduced to
chloride ion following the reaction. The sulfite ion is correspondingly oxidized to.the sulfate
ion. Both chloride and sulfate ions are usually present in abundance in wastewater,. so that
the additional increment added in dechlorination is usually insignificant.
                                        6-19

-------
The predictions of sulphur dioxide dosages required for dechlorination (Equations 6-13 and
6-15) suggest an SO2:Cl2 molar ratio of 1:1. This corresponds to a weight ratio of 0.9:1. In
practice, 0.9 to 1.0 parts of SO2 are required for dechlorination  of 1.0 part of chlorine
residual (expressed as <
The acidity generated in dechlorination, while appearing to be significant on a molar basis, is
seldom an important factor in practice. Roughly 2 mg/1 of alkalinity are consumed for each
1.0 mg/1 of sulphur dioxide applied. The chlorine residual remaining in solution  following
breakpoint chlorination and disinfection  contact is usually low enough (approximate range
of 1 mg/1  to 8 mg/1) so that the alkalinity consumption and pH change resulting from
sulphur dioxide addition may usually be neglected.

         6.4.1.2 Reaction Rates

The  rate of  reaction  between  SC>2 in  solution  and  chlorine  residual is practically
instantaneous. The high specificity of SC>2 in solution for the chlorine residual and the rapid
reaction  rate tend to reduce to very low levels competing side  reactions which could result
in wastage of the sulphur dioxide.

         6.4.1.3 Significance of Sulphur Dioxide Overdose

Sulphur dioxide dechlorination is most often accomplished to eliminate the chlorine induced
toxicity  in  a  wastewater effluent  prior to discharge. Studies on primary and secondary
effluents23,24 have  shown that effluent toxicity following chlorination-dechlorination, as
measured by fish bioassay, is reduced to a level slightly below that measured on comparable
unchlorinated effluent. These studies indicate that sulphur  dioxide dechlorination  can
eliminate acute toxicity related to chlorine residual. Tests in an aerated continuous-flow fish
bioassay  apparatus at sulphur dioxide overdoses of about 70  mg/1 (as 862)  beyond that
necessary for dechlorination showed no adverse effects on the test fish.

Large overdoses of sulphur dioxide should be avoided in plant  scale dechlorination systems
because of chemical wastage and the oxygen demand exerted by the excess sulphur dioxide
in solution.  Excess sulphur  dioxide reacts slowly with  dissolved  oxygen  in  solution
according to the following equation:

                        HSO^ + 0.5O2 —-SO4+H+                         (6-16)

The net  result of this reaction can be a reduction  in wastewater effluent  dissolved oxygen
and an  increase in  the measured  effluent COD  and BOD  levels. Proper control of the
dechlorination system is essential to minimize these adverse effects.

         6.4.1.4 Process Application and Control

Chemical dechlorination using sulphur dioxide can  be  a rather inexpensive process for

                                         6-20

-------
reducing effluent toxicity related  to chlorine residual. Much of the process equipment,
chemical handling and  chemical  addition components of a sulphur dioxide dechlorination
system  are identical  to those  used  in  a chlorination system.  For  example, gaseous
chlorinators  are  used  to feed  sulphur  dioxide, evaporaters may be used practically
interchangeably between chlorine  and sulphur  dioxide, and piping,valves  and gaseous
injectors are identical in the two systems. The speed of the reaction precludes  the necessity
for providing a separate contact basin as in chlorine contact for disinfection. The  dosage of
sulphur dioxide required is only that amount needed to dechlorinate the effluent  following
the chlorine contact period, a level substantially less than the total chlorine dose.

There are no analytical instruments  currently  on the market which have demonstrated a
capability to reliably and accurately measure the excess sulphur dioxide in solution following
dechlorination. This is  why the process control  schemes which have been  devised do not
have the  feedback self-corrective aspects  which provide a  high  degree of consistency to
effluent quality.  Control is usually achieved by  pacing the  dechlorination equipment on
signals of flow  and chlorine residual measured directly upstream from the point of sulphur
dioxide application. Operator surveillance and frequent manual adjustments to the system
are needed to maintain a slight overdose of sulphur dioxide, say 0.5 mg/1.

A more complicated  control  scheme which  does incorporate  positive control features
involves application of a preselected chlorine dosage to a small side stream of dechlorinated
effluent. If, for example, a preselected chlorine dose of 1 mg/1 is applied and dechlorination
is paced to maintain a chlorine residual of 0.8 mg/1 in the "biased" side stream, the net
effect  of the  control  system would be  to provide  a controlled overdose  of  0.2  mg/1
(expressed as Cty of sulphur dioxide.

With either control scheme, the sulphur  dioxide overdose  in the final effluent may be
controlled to 0.5  mg/1 (as Cty or less. Given that the reaction rate of Equation 6-16 is very
slow, the effect of the overdose on plant effluent dissolved oxygen concentration  would be
minimal, probably less than 0.2 mg/1 DO depletion.  In such a case,  aeration following
sulphur dioxide dechlorination would seldom be warranted.

     6.4.2 Activated Carbon Dechlorination

Dechlorination  of wastewater  effluent streams using activated  carbon can serve several
functions other than removal of chlorine residuals. In the case  of breakpoint chlorination,
the activated carbon can effectively catalyze the chemical reactions, serving as a "reaction
bed." Some removal of soluble organics is also accomplished through adsorption.

          6.4.2.1  Stoichiometry

Activated carbon (C*)  reacts with  both free and  combined chlorine residuals in the
following manner:
                                        6-21

-------
    a.   Reaction with free chlorine residual
                       C* + 2HOC1 	- CO2 + 2H+ + 2Cf                  (6-17)
    b.   Reactions with combined chlorine residual
                  C* + 2NH2C1 + 2H2O  	^ CO2 + 2NH*  + 2C1              (6-18)
               C* + 4NHC12 + 2H20 	^ C02 + 2N2 + 8H+ + 8Cf           (6-19)
Carbon dioxide is formed in each case following reaction with chlorine residual. Ammonia is
returned  to  solution following reaction of carbon with monochloramine, but dichloramine
has been observed to decompose to nitrogen gas following contact with activated carbon.

These chemical pathways have been confirmed by several recent studies.25,26

         6.4.2.2 Process Application

Studies by Stasiuk  et al. 25 showed complete  dechlorination of both  free and combined
chlorine  residuals  following  carbon contact  times  of   10  minutes. Studies on  the
dechlorination  characteristics of granular activated carbon  have been  conducted27,28,29
which show  a variation in the dechlorination capacity of activated carbon  as a function of
the hydraulic  application rate and particle  size. Those data suggest  the formation of a
dechlorination  intermediary compound, nascent oxygen, on   the surface of the carbon
which builds up and causes a gradual loss of dechlorination efficiency. Regeneration of the
carbon was  accomplished by heating in the absence of air at 400 C  or higher. Smaller
activated carbon particles were found to give enhanced dechlorination capacity. One cubic
foot of activated carbon was found to  dechlorinate  0.55  million gallons at 2 gpm/sq  ft
hydraulic  application  rate and  5 mg/1 free  chlorine  residual. Approximately 3.6 million
gallons of the same  water could be dechlorinated by the activated carbon when applied at a
rate of 1 gpm/sq ft.

6.5 Design Example

As  an example,  consider a  10 mgd conventional activated  sludge plant  that  must be
upgraded to  meet an effluent nitrogen limitation of 2 mg/1. Present effluent quality is 1 mg/1
organic nitrogen, 20 mg/1 ammonia nitrogen, 15 mg/1 of suspended solids, BOD5 of 25 mg/1
and pH of 7.0.  The  peak to average nitrogen load ratio  is 1.9 The breakpoint chlorination
design should provide capacity for removal of all ammonia nitrogen in the plant effluent.
                                        6-22

-------
1 .    Calculate the average ammonia nitrogen oxidized daily.
                             = 8.33-Q(NQ-N1)                         (4-24)

 where:   NT  =   ammonia nitrogen oxidized, Ib per day

          Q  =   average daily flow, mgd

         N   =   influent ammonia nitrogen, mg/1

         N,  =   effluent ammonia nitrogen, mg/1

 For this example, the result is:

                   NT = 8.33(10)(20-0) = 1,666 Ib/day

2.    To calculate the average daily chlorine consumption, a Cl2:NH4-N of approxi-
     mately 9:1 would  be appropriate. The  average daily chlorine consumption would
     be:

               Average daily O, = (9)(1 ,666) = 14,994 Ib/day
3.   The peak  rate of chlorine utilization, that rate used to size the chlorine feed
     system, may be computed by multiplying the peak to average nitrogen load ratio
     by the average daily chlorine consumption as follows:

                  Peak C12 = (1.9)(14,994) = 28,489 Ib/day


4.   To calculate the average daily quantity of alkalinity supplementation, assume that
     sodium hydroxide  is to be  used. Since the effluent pH is at  pH  7  prior to
     breakpoint chlorination,and the breakpoint process itself should be conducted at
     pH 7, all acidity generated in breakpoint chlorination must be neutralized by the
     sodium hydroxide  added.  Since 1.50 Ib. of NaOH are  required per Ib of C\2
     added in breakpoint, the average daily NaOH added per day would be:

               Average daily NaOH = (1.50)(14,994) = 22,491 Ib/day


5.   The TDS  increment added to the plant effluent  as  a  result of  breakpoint
     chlorination can  be  computed  from  data   of  Table  6-3.   For  breakpoint
     chlorination with chlorine gas and neutralization  of all acidity  with sodium
                                  6-23

-------
         hydroxide,  the TDS increase per mg/1 of ammonia nitrogen consumed is  14.8
         mg/1. The total TDS increase in this example would be:

                          TDS increase = (14.8)(20) = 296 mg/1.

     6.   To  compute  the  average  daily  consumption  of sulphur  dioxide  needed  to
         completely  dechlorinate the plant effluent prior to  discharge,  assume that the
         average chlorine residual following breakpoint chlorination and disinfection is 5
         mg/1. The average daily sulphur dioxide consumption would be:

                   Average daily SO2 = (8.33)(10)(5)(1.0) = 416 Ib/day.


6.6 Considerations for Process Selection

The breakpoint chlorination process  offers a  number of  advantages which should  be
considered when evaluating alternative nitrogen removal processes:

     1.   Ammonia  nitrogen removal  may  be accomplished  in a  one-step  process  to
         concentrations less than 0.1 mg/1 (as N). The major end product of the breakpoint
         reaction is nitrogen gas which is evolved to the atmosphere.

     2.   Breakpoint  chlorination is  free from the  toxic  upsets and  acclimation periods
         which can affect biological nitrification  and denitrification processes. Ammonia
         nitrogen may  be  successfully  removed from solution regardless of upstream
         treatment processes. Breakpoint chlorination is also rather insensitive to changes
         in process temperature.

     3.   The  breakpoint process may be  operated intermittently or in  split-stream
         arrangements as needed to meet individual  receiving water  nitrogen  limitations.
         The process can be used as a  polishing step for other ammonia removal processes
         such as nitrification to provide low ammonia nitrogen  concentrations.

     4.   Breakpoint chlorination is  reliable  and consistent in  terms of process perfor-
         mance.

     5.   Disinfection of a wastewater effluent is enhanced following  breakpoint chlorina-
         tion due to the presence of free available  chlorine residuals (HOC1 and OC1~).

     6.   The  low spacial requirement of the breakpoint process makes it  particularly
         suitable for certain applications, including addition to an existing facility, where
         nitrogen removal is required but where space constraints exist.
                                         6-24

-------
     7.   The cost of physical  facilities for breakpoint chlorination is much less than for
         biological nitrification-denitrification facilities.

A list of potential disadvantages  of the breakpoint chlorination process includes:

     1.   The breakpoint chlorination process is usually quite high in operating costs.

     2.   The addition  of chorine,  pH  adjustment chemicals and  sulphur dioxide  all
         contribute  to the level of total dissolved solids in the wastewater effluent. Some
         treatment  plants have TDS limitations which  would limit the applicability of
         breakpoint chlorination.

     3.   Dechlorination will be required  in  many cases  to remove the potentially toxic
         chlorine residual. It should be noted, however, that this is a disadvantage of any
         chlorination process.

6.7 References

 1.  Pressley,  T.A.,  Bishop,  D.F., and S.G. Roan, Nitrogen Removal By Breakpoint
    Chlorination. Report prepared for the Environmental Protection Agency, September,
    1970.

 2.  Pressley, T.A., Bishop, D.F., Pinto, A.P., and A.F. Cassel, Ammonia-Nitrogen Removal
    by  Breakpoint  Chlorination.  Report prepared for  the  Environmental Protection
    Agency, Contract Number 14-12-818, February,  1973.

 3.  Pressley, T.A., et al, Ammonia Removal by Breakpoint Chlorination. Environmental
    Science and Technology, 6,  No. 7, 662, July,  1972.

 4.  White, G.C., Handbook of Chlorination. New York, Van Nostrand Reinhold Company,
    1972.

 5.  Morris,  J.C., and I. Weil, Chlorine-Ammonia Breakpoint Reactions: Model Mechanism
    and Computer Simulation. Paper, annual meeting Am.  Chem. Soc., Minneapolis, Minn.,
    April 15, 1969.

 6.  Morris,  J.C., Weil,  I., and R.H. Culver, Kinetic Studies on  the  Break-Point with
    Ammonia and Glycine. Unpublished copy from senior author, Harvard Univ.  (1952).

 7.  Morris,  J.C., Kinetic Reactions between Aqueous Chlorine and Nitrogen Compounds.
    Fourth  Rudolphs Res. Conf., Rutgers Univ. (June 15-18, 1965).

 8.  Griffin, A.E., and N.S. Chamberlin,  Some  Chemical Aspects of Breakpoint Chlori-
    nation.  J. NEWWA, 55, pp 371, 1941.

                                       6-25

-------
 9.  Weber,  W.J., Jr., Physio chemical Processes for Water Quality Control.  New York,
    Wiley-Interscience, 1972.

10.  Stone, R.W., Parker, D.S., and J.A. Cotteral, Upgrading Lagoon Effluent to Meet Best
    Practicable  Treatment.  Presented at the  47th Annual Conference  of the Water
    Pollution Control Federation, Denver, Colorado, October, 1974.

11.  Brown and  Caldwell, Report on Tertiary Treatment Pilot Plant Studies. Prepared for
    the City of Sunnyvale, California, February, 1975.

12.  Stearns and Wheeler, Waste-water Facilities Report. Cortland, New York, May, 1973.

13.  Stenquist, R.J., and WJ. Kaufman, Initial Mixing in Coagulation Processes.  Prepared
    for EPA, Project EPA-R2-72-053, November, 1972.

14.  Schuk, W., personal communication with D.S. Parker. EPA-DC Blue Plains Pilot Plant,
    September 9, 1974.

15.  Lawrence, A.W.,  et al, Ammonia Nitrogen Removal from  Wastewater Effluents by
    Chlorination.  Presented  at the  4th  Mid-Atlantic  Industrial  Waste  Conference,
    University of Delaware, November, 1970.

16.  Taras, M.J., Effect  of Free Residual Chlorine on Nitrogen Compounds in  Water.
    JAWWA,45,47(1953).

17.  Griffin,  A.E.,  and  N.S.   Chamberlin,  Some Chemical Aspects  of  Break-Point
    Chlorination. J. NEWWA, 55, pp 371 (1941).

18.  Griffin, A.E., Chlorine for Ammonia Removal. Fifth Annual  Water Conf. Proc. Engrs.
    Soc. Western Penn., pp 27 (1944).

19.  William, D.B.,  private communication with G.C. White, Brantford, Ontario, Canada
    (1967).

20.  Clarke, N.A.,  et al,  Human  Enteric Viruses   in   Water: Source, Survival and
    Removability. International Conf.  Water Poll. Research, Pergamon  Press, London
    (September, 1962).

21.  van  Vuuren, L.R.J., et al, Slander Water  Reclamation Plant: Chlorination Unit
    Process. Project Report 21, Pretoria, So. Africa, November, 1972.

22.  Pollution: Engineering  and Scientific Solutions.  E.S. Barrekette Ed.,  New York,
    Plenum Publishing Corp., pp 522-547.
                                       6-26

-------
23.  Esvelt,  L.A., Kaufman, W.J.,  and R.E. Selleck,  Toxicity Removal from Municipal
     Wastewater.  University of  California,  Sanitary  Engineering  Research Laboratory
     Report 71-7, 1971.

24.  Stone,  R.W., Kaufman, W.J.,  and A.J. Home, Long-Term Effects of Toxicants and
     Biostimulants on the Waters of Central San Francisco Bay. University of California,
     Sanitary Engineering Research Laboratory Report 73-1, May, 1973.

25.  Stasuik, W.N.,  Hetling, L.J., and W.W.  Shuster, Removal of Ammonia Nitrogen by
     Breakpoint  Chlorination  Using an  Activated  Carbon  Catalyst.  New  York  State
     Department of Environmental Conservation Technical Paper No. 26, April, 1973.

27.  Hagar, D.G., and M.E. Flentje, Removal of Organic Contaminants by Granular Carbon
     Filtration. JAWWA, 57, 1440 (November, 1965).

28.  Kovach, J.L., Activated Carbon  Dechlorination.  Industrial Water  Engineering, pp
     30-32, October/November, 1973.

29.  Calgon Corporation, Bulletin 20-44.
                                       6-27

-------
                                    CHAPTER 7

             SELECTIVE ION EXCHANGE FOR AMMONIUM REMOVAL



7.1 Chemistry .and Engineering Principles

The basic concepts of the ion exchange process apply to its use for ammonium removal and
these concepts are discussed in the following paragraphs.

     7.1.1 Basic Concept

The use of conventional ion exchange resin  for removal  of nitrogenous material from
wastewaters has not proven attractive because of the preference of these exchangers for ions
other than ammonium or nitrate ions. In addition, the regeneration of conventional ion
exchange resins results in regenerant wastes which are difficult to handle. The non-selective
nature of conventional resins is unfortunate because as a unit process, ion exchange is easily
controlled to achieve almost any desired product quality. The efficiency of the process is
not significantly impaired at temperatures usually encountered and ion exchange equipment
can be automatically  controlled,  requiring only occasional  monitoring, inspection, and
maintenance.

The above limitations of conventional resins may be largely overcome by using an exchanger
which  is  selective  for ammonium. The  exchanger currently  favored for this  use  is
clinoptilolite, a zeolite, which occurs naturally in  several extensive deposits in the western
United States. It is selective for ammonium relative to calcium, magnesium and sodium. The
removal  of the ammonium  from  the  spent  regenerant permits regenerant  reuse. The
ammonium  may  be removed  from the regenerant and released to the atmosphere  as
ammonia (in certain situations) or  nitrogen gas or it may  be recovered as an ammonium
solution for use as a fertilizer. Figure 7-1 is a simplified schematic of the process.

The wastewater is passed downward through a bed of clinoptilolite (typically 4-5 ft or 1.2
to 1.5  m of 20 x 50 mesh particles) during the  normal service cycle.  When the effluent
ammonium  concentration increases to an  objectionable  level, the clinoptilolite  is regen-
erated  by passing a concentrated salt solution through the exchange bed. By  removing the
ammonia from the spent regenerant, the regenerant may be reused, eliminating the difficult
problem  of brine disposal  associated  with conventional exchange resins. Some of the
regenerant recovery techniques, as discussed later in detail, remove  the ammonia as nitrogen
gas which is discharged to the atmosphere while others remove and recover the ammonia in
solution form for potential use as a fertilizer.
                                        7-1

-------
                             FIGURE 7-1
                 SELECTIVE ION EXCHANGE PROCESS
     INFLUENT
                   SPENT REGENERANT
                                        "I
  cLiNOPTiLOLiTE
        BED
                                     REGENERANT
                                       RECOVERY
                  AMMONIA OR
                  NITROGEN GAS
                 'OR RECOVERED
                  AMMONIA
                  SOLUTION
                ________
                   FRESH REGENERANT
     EFFLUENT
    7.1.2 Ion Exchange Principles

Ion exchange  is a process in which ions held by electrostatic forces to charged functional
groups on the surface of a  solid are  exchanged for ions of similar charge in a solution in
which the solid is immersed. It is a stoichiometric, reversible exchange of ions between a
liquid and solid which produces no significant changes in the  structure of the solid. The
mass action equilibria expression provides a useful model for ion exchange  behavior. In a
binary system, the reaction,
                        bA+a + aBZ,
 bAZ +aB
     a
          +b
(7-1)
expresses the  reversible  equilibrium where a and b are the valences of ions A and B,
respectively, and Z is the exchange site in the solid. * This reaction may be expressed as the
equilibrium constant,
                                  (a)
                              K =
                                    AZ
(a)
  B
                                  (a)
(a):
                                                                         (7-2)
                                            B
    Zb
                                     7-2

-------
in which (a)^, (a)AZa> etc- are the activities of the various spc-cies.  Because of the difficulty
in measuring activities, especially in the solid phase, it is convenient to use  concentrations
uncorrected for activity.  In doing so, the equilibrium constant in Equation 7-2 varies with
concentration and has been termed the "selectivity coefficient,"
where q is the solid phase ionic concentration in milliequivalents per gram (meq/g) and c is
the solution phase concentration in meq/1. Alternatively, the selectivity coefficient can be
expressed in terms of dimensionless concentration,
                                        a-b   /  \b  /  \a
                                           =(4(f)B

These variables are expressed in terms of the total solution concentration, Co, in meq/1 and
the total exchange capacity, Q, in meq/g. Thus, x = c/Co and y = q/Q.

The preference of an ion exchange for one ion relative to another in binary systems is often
expressed as the "separation factor,"

                           ocA = /-9-U-}  =(!-}  (+-}                     (7'5)
                              B    KH   ^'AH
Because the numerical  value of  the  separation factor is not affected  by the choice of
concentration  units, equilibrium  data are  often expressed in  this way. If  the  equivalent
fraction of ion A in the solid phase, y^, is plotted against the equivalent fraction of A in the
solution, XA, three cases can be identified corresponding to oc<.l,   ot = 1, and ex. > 1 as
shown  in  Figure  7-2.  Isotherms which are concave upward, l, are referred to as "favorable" isotherms since  the solid prefers ion A to ion
B. Ion exchange operations almost always are concerned. with systems in which  the ion of
concern has a separation factor greater than unity during the service cycle.

The basic principles of ion exchange can be used to determine the capacity of clinoptilolite
for ammonium and an excellent example is  available  in Reference 2. However,  such
calculations are lengthy and  rather complex and a later section (7.2.9) presents a simplified
technique for  quickly approximating the ammonium capacity of clinoptilolite for varying
concentrations of competing cations.

     7.1.3 Properties of Clinoptilolite

          7.1.3.1  Selectivity

Isotherms demonstrating the selectivity of clinoptilolite for ammonium over other cations
have  been  reported in the  literature. 3   An example  is  the  comparison  of  Hector

                                         7-3

-------
                                  FIGURE 7-2
                  GENERALIZED ION EXCHANGE ISOTHERMS
              0.0   O.I   0.2  0.3   0.4   0.5   0.6    O.7   0,8   0.9   t.O
           EQUIVALENT  ION  FRACTION IN SOLUTION PHASE, XA

Clinoptilolite  and a  strong acid polystyrene resin, IR 120, shown in Figure 7-3. The total
equilibrium solution normality was constant at O.IN. The terms (Ca)z, (NH4)Z = equivalent
fraction of calcium or ammonium on the zeolite. The terms (Ca)^, (NH|>N = normality of
calcium or ammonium in the equilibrium solution. The ammonium capacity of IR 120 was
4.29 meq/g of air-dried resin. A comparison of the generalized isotherm presented in Figure
7-2 with Figure  7-3 clearly  shows that the IR 120  prefers calcium to ammonium ions.
Clinoptilolite, on the other hand, prefers ammonium to calcium ion and is of greater utility
for ammonium ion removal from wastewaters containing calcium.
                                      4
The ion exchange equilibria for the systems NHj-Na+, NH4-K+, NH4-Ca+2 and NH|-Mg+2,
with Clinoptilolite and other zeolites, are also  available in the literature.2 Plots of the NH^
selectivity coefficients vs.  the solution concentration ratios of the  cations are shown in
Figures 7-4  and 7-5. .Using  these  data in  conjunction  with  published  calculation
techniques, 2  it is possible to accurately predict the  ammonium capacity of Clinoptilolite in
the presence of various concentrations of other cations.
                                       7-4

-------
                                   FIGURE 7-3

   THE 23 C ISOTHERMS FOR THE REACTION, (Ca)z + 2(NH*)N = 2(NH^)z + (Ca)N
                   WITH HECTOR CLINOPTILOLITE AND IR 120
          1,0
                                            Hector
                                       Clinoptilolite
             0           0.2
               EQUIVALENT
     0.4         0.6
ION  FRACTION   IN
      0.8          1.0
SOLUTION PHASE
Section 7.2.9 presents a  curve useful in approximating the ammonium exchange capacity
which can be employed in sizing the ion exchange beds. The equilibrium isotherms for
ammonia and other cations which are present as macrocomponents in wastewaters are
shown  in Figure 7-6J  These isotherms  illustrate  that  clinoptilolite is  selective for
ammonium relative to all of the listed ions except potassium.

         7.1.3.2 Mineralogical Classification

The  zeolites  are classified  as  a family   in  the  silicate group.  They  are hydrated
alumino-silicates of univalent and bivalent bases which can be reversibly dehydrated to varying

                                      7-5

-------
                                  FIGURE 7-4
     SELECTIVITY COEFFICIENTS VS. CONCENTRATION RATIOS OF SODIUM
     OR POTASSIUM AND AMMONIUM IN THE EQUILIBRIUM SOLUTION WITH
            HECTOR CLINOPTILOLITE AT 23 C FOR THE REACTION
     -*•<*•
    UJ
    u,
    Lu
    o
    o
    O
    UJ
    -J
    UJ
    CO
              (Y)
                                                  (Y)
                                                     N
          IOOO
           100
10
           O.I
                       I Hill  I I  I ll Illl   I 1 I I I I II I   I  I   I Illll
O.OI
             O.I
                                               10
IOO
IOOO
                      CONCENTRATION  RATIO,
degrees without undergoing a change in crystal structure  and are capable of undergoing
cation  exchange.   The  general  composition of  zeolites  is given  by  the  formula
(M,N2)O-Al2O3-nSiO2'mH2O where  M and  N are,  respectively,  the  alkali  metal  and
alkaline earth counter ions present in the zeolite cavities.

Clinoptilolite is a common material  found in bentonite  deposits in the western United
States.  The largest known deposit of clinoptilolite in the United States is found in southern
California within a deposit of bentonite called Hectorite because of its proximity to Hector,
California. The U.S. deposits are predominantly in the sodium form. Although a widely
occurring material, not all deposits produce a clinoptilolite of adequate structural strength
to withstand the handling which occurs in a columnar operation.
                                      7-6

-------
                                   FIGURE 7-5
    SELECTIVITY COEFFICIENTS VS. CONCENTRATION RATIOS OF CALCIUM
    OR MAGNESIUM AND AMMONIUM IN THE EQUILIBRIUM SOLUTION WITH
             HECTOR CLINOPTILOLITE AT 23 C FOR THE REACTION
                         (X)z + 2(NH+)N = 2(NH+)  + (X)
   xl
    u.
    u.
    iu
    o
    o
    I-
    o
    Lu
           10''
           10'
           10'
            10
                 I . ill...I  I I . I....I  I I lill.J  I 1 llll.j I  I II.1..I  I  I .(...I II ll.l.J I  I lllll.i
               /      10     10'
I05    IO6    I07    I08
                       CONCENTRATION  RATIO,
                                                      (X)N
         7.1.3.3 Total Exchange Capacity

Although the total  ion  exchange capacity of a  material  is by no means a complete
description  of its ion exchange properties, it is an indication of the applicability of the
substance for process use. For example, New Jersey greensand, which was widely used in
water softening before  the development of organic exchangers, has a total exchange capacity
of 0.17 meq/g. 1 In comparison, the exchange capacities of strong acid cation exchangers are
usually 4 to 5 meq/g. The total exchange capacity of clinoptilolite as measured by several
different investigators ranges from 1.6 to 2.0 meq/g and is slightly lower than the average
for zeolites. With typical cation concentrations encountered in municipal wastewaters, the
capacity for NH$ is typically 0.4 meq/g (see Section 7.2.9).

         7.1.3.4 Chemical Stability

The instability of natural clays and zeolites toward acids and alkalis is known as  these
materials are widely used in water softening. However, clinoptilolite is considerably  more
                                      7-7

-------
                                  FIGURE 7-6

        ISOTHERMS FOR EXCHANGE OF NH* FOR K+, Na+, Ca**, AND Mg"1
                             ON CLINOPTILOLITE
                   '      I      I

            _ Magnesium
                                      Potassium
                                                   Total  Solution
                                                   Concentration 0.1  N-,
                                                   Temperature 23 C
           EQUIVALENT FRACTION OF  NH$~N  IN  SOLUTION  PHASE,
                                  *NH+-N
acid  resistant  than  other zeolites. ^ Very high strength (20% NaOH)  caustic solutions
produce significant chemical attack on the clinoptilolite. However, at the lower solution
strengths encountered in systems which use a caustic regenerant, physical attrition is more
significant than chemical attack. This will be discussed later in this section.

        7.1.3.5 Physical Stability

When crushed, sieved,  and thoroughly washed with agitation to remove fines, clay, and
other impurities, 20 x 50 mesh Hector clinoptilolite gives a wet attrition test of 3 percent. 4
                                     7-8

-------
The wet attrition test determines the amount of fines (less than 100 mesh) generated by 25
grams of the granular zeolite  during rapid mixing with 75 milliliters of water on a paint
shaker  for 5  minutes.  Commercial zeolites,  such  as erionite and chabazite, which are
powdered,  mixed with clay binder, extruded, and fined, will generally give a wet attrition
test of  about 6 percent or twice that of the Hector  clinoptilolite. Low wet attrition is
important to minimize losses of clinoptilolite in an ion exchange column operation.

         7.1.3.6 Density

Clinoptilolite (20 x 50 mesh) has been reported to have a wet particle specific gravity of
1.5 9 and a bulk density of 0.74-0.79 g/cc1.

7.2 Major Service Cycle Variables

The factors which have a major effect  on process efficiency include: pH, hydraulic loading
rate, clinoptilolite size, pretreatment, wastewater composition, and bed depth.

     7.2. IpH.

Within an influent pH range of 4 to 8, optimum ammonium exchange occurs. 1 As the pH
drops below this range, hydrogen ions  begin to compete with ammonium for the available
exchange capacity. As the pH values increase above  8, a shift in the NH3-NH4  equilibrium
towards NH3 begins. Operation outside of the pH range of 4 to 8 results in a rapid decrease
of exchange capacity and increased ammonium leakage.

     7.2.2 Hydraulic Loading Rate

Variations in column loading rates within the range  of 7.5-20 Bed Volumes (BV)/hour (7.5
BV/hr is equivalent  to 0.95 gpm/cu ft or 2.15 l/m3/sec) have been shown to  produce  no
significant  effects on the  ammonium removal efficiency of 20 x 50 mesh  clinoptilolite.3
Ammonium concentrations in the clinoptilolite effluent of 0.22-0.26 mg/1  were  produced
throughout the  above  range in one set of tests. 1  When rates  exceed 20 BV/hour,  the
exchange kinetics suffer as demonstrated by a significant leakage of ammonium early in the
loading  cycle. The effects  of loading rate as a function  of clinoptilolite size are discussed in
the next section.

     7.2.3 Clinoptilolite Size

Mine-run clinoptilolite is typically 1-2  inch (25 to 51 mm) chunks which must be ground
and screened to the size desired for column operation. As would be expected, the smaller
the clinoptilolite size, the  better the kinetics of  the exchange  reaction. This  effect is
illustrated by data that show that 20 x 50  mesh clinoptilolite kinetics begin to suffer (see
Section  7.2.2) at rates of 20-30 BV/hour while  50  x 80 mesh kinetics do not  suffer until
rates of 40 BV/hour are reached. However, the improved rate of exchange is accompanied

                                         7-9

-------
by the disadvantage of higher head loss. It appears that 20 x 50 mesh clinoptilolite (about
the size of typical filter sand) offers an adequate compromise between acceptable headless
and exchange kinetics. At a loading of 15 BV/hour in a 3 ft (0.9m) deep bed (5.6 gpm/sq ft
or 3.8 l/m^/sec), the headloss is 2.1  ft (0.64m) with 20 x  50 mesh clinoptilolite. Lower
headlosses  could  be  obtained by lower rates.  Use of deeper beds would result in greater
headlosSji.e., 6  ft (1.8 m) depth would have a headloss of 4.2 ft (1.28 m) at 5.6 gpm/sq ft
(3.8 l/m^/sec). These headloss values do not include losses in inlet and outlet piping or in
the underdrain system.

     7.2.4 Pre treatment

To avoid excessive headloss, the clinoptilolite influent must be relatively free of suspended
solids — preferably  less than 35 mg/1. Tests with clarified and filtered raw wastewater
indicate  no  problems  with organic  fouling.  Biological growths  which  occurred  were
adequately removed in the regeneration cycle.2 Additional data on pretreatment effects are
presented in the next section.

     7.2.5 Wastewater Composition

As noted earlier, although clinoptilolite prefers ammonium ions to other cations, it is not
absolutely selective and other cations  do compete for the available exchange capacity. Pilot
tests conducted at several locations illustrate the effects of wastewater composition on the
useful capacity of the  clinoptilolite.^ Tests at three locales that span a wide  range  of
wastewater compositions are shown in Table 7-1.

The equilibrium Nffy-N bed loading computed for each of the wastewaters listed in Table
7-1  was 4.1 g/1, 3.9 g/1, and 4.3 g/1, respectively, forTahoe, Pomona, and Blue Plains. Figure
7-7  presents equilibrium bed loading in  an alternate way. The minimum bed  volumes
required to attain equilibrium Nlfy-N loading are expressed  as  a function of the Nlfy-N
concentration in the influent wastewater. The equilibrium  bed volume values given in Figure
7-7 normally represent the 50 percent breakthrough point (the effluent concentration is 50
percent of the feed concentration). The water lowest in competing cations (Blue Plains) had
the greatest ammonium removal capacity. In the 10-20 mg/1 influent Nlfy-N range,  the
lower  competing  ion  concentrations  at Blue Plains resulted in the useful ammonium
exchange capacity being about 33 percent greater than that for Pomona with its higher TDS
water. The lower degree of pretreatment at Blue Plains (i.e., no biological pretreatment) did
not impair the effectiveness of the clinoptilolite for ammonium removal.

     7.2.6 Length of Service Cycle

To illustrate the determination of a permissable length service cycle for a given wastewater,
ammonia breakthrough  curves  for a single 6 ft deep  (1.8m) bed  of  clinoptilolite  are
illustrated in Figure 7-8 for Tahoe tertiary effluent with flow rates varying from 6.5 to 9.7
                                        7-10

-------
                                   TABLE 7-1

           INFLUENT COMPOSITION FOR SELECTIVE ION EXCHANGE
                    PILOT TESTS AT DIFFERENT LOCALES
Parameter, mg/1
NH*-N
Na
K
Mg
Ca
PH
Range
COD
IDS
Activated sludge plant effluent
Tahoe
Carbon treated
15
44
10
1
51
7-8b
11
325
a
Pomona
16
120
18
20
43
6.5-8.2b
10
700
Clarified raw wastewater
Blue Plains
(Washington, D
12
35
9
0.2
30
7-9b
50
250
.C.)







      Approximately half of the runs at Pomona were made with carbon treated
      secondary effluent and the others with alum coagulated secondary effluent.
      pH units

BV/hour (bed volumes per hour) with 15 to 17 mg/1 NH^-N in the feed stream. These curves
indicate a throughput value of 150 bed volumes should be used for this wastewater for
design for effluents requiring  a high degree of ammonia removal.  Although the effluent
concentration had reached 2-3  mg/1 at 150 BV, the average concentration produced to this
point in the cycle was less than 1 mg/1. Breakpoint chlorination would be more economical
for removing the 1 mg/1 residual, if required, than would provision of greater ion exchange
column  capacity. The  average ammonium concentration for a breakthrough  curve  is
obtained by  integrating the area under the breakthrough  curve and dividing by the total
flow. For example, integrating the area under the curve for 8.1 BV/hr in Figure 7-8 indicates
an average NH^-N concentration of 0.67 mg/1 for 150 BV.

     7.2.7 Bed Depth

The  effect of bed depth on ammonia breakthrough at 9.7 BV/hr is illustrated in Figure 7-9.
In general, the 3 ft (0.9 m) bed of clinoptilolite was not as effective for ammonium removal
as the 6 ft (1.8  m) bed at the same  bed volume rate.  The shallow bed has a lower flow
velocity because a 9.7 BV/hr flow in a 3 ft (0.9 m) deep bed corresponds to 3.6 gpm/sq ft
(2.5  l/m2/sec)  while in a 6  ft (1.8  m) deep bed it corresponds  to 7.2 gpm/sq ft (4.9
                                      7-11

-------
                                FIGURE 7-7
           MINIMUM BED VOLUMES AS A FUNCTION OF INFLUENT
              NH*-N CONCENTRATION TO REACH 50 PERCENT
              BREAKTHROUGH OF AMMONIUM (REFERENCE 2)
    401
'a:
LU
Cj
Uj
-J
u.
    30
20
     10
                                                  D  TAHOE
                                                  X  POMONA
                                                  O  BLUE  PLAINS
                  100          ZOO         300
                            MINIMUM  BED VOLUMES
                                                    400
500
l/m^/sec). The lower velocity might increase the likelihood of plugging of portions of the
bed. Plugging would cause poor flow distribution and lower bed efficiency. As discussed in
the design examples in Chapter 9, full-scale designs are using bed depths of 4-ft (1.2 m) with
a high degree of pretreatment (coagulation and filtration) which will minimize plugging of
the clinoptilolite bed.

    7.2.8 One Column vs. Series Column Operation

Operation to the  150 bed volume throughput value (Figure 7-9) to maintain an average
NH4—N concentration at or below 1 mg/1 uses  only 55  to 58 percent of the zeolite's
equilibrium capacity,  The number of bed volumes throughput per bed can be  increased
while maintaining low NH4—N effluent concentrations with semi-countercurrent operation,
using two beds in series. Semi-countercurrent operation is achieved by first operating the
                                     7-12

-------
                                   FIGURE 7-8

      AMMONIUM BREAKTHROUGH CURVES FOR A 6 FT CLINOPTILOLITE
                        BED AT VARIOUS FLOW RATES
 ¥
 Ul
 -J
 u.
 u.
 10
 Q
 UJ
 00
           OPERATING  CONDITIONS:

           ZEOLITE GRAIN SIZE: 20x50  MESH
           BED  VOLUME: 50 FT3  (l.4l'6<)
           FEED:  TAHOE TERTIARY  EFFLUENT
                            Avg. Influent
      - Symbol   Flow Rate    NH^-N.mg/J
                                              _L
             20     4O      60      80      IOO
                               BED  VOLUMES
120
140
160
180
columns in a 1-2 sequence and then placing column 2 into the lead position, after the first
regeneration, with column 1 becoming the polishing column. A column is removed from the
influent end when it becomes loaded while simultaneously adding a regenerated column, at
the effluent end. This procedure in effect moves the beds counter-current to liquid flow by
continually shifting the more saturated beds closer to the higher influent concentrations.
Beds can  be loaded nearer to capacity with this procedure than  with single column or
parallel  feed multi-column operation. The  most highly loaded  column is always at the
influent end backed up by one (if two in series) or more columns having decreasing loadings
and NH4-N concentrations at locations progressively nearer the end of the series. Removal
of a column is not decided by applying a breakthrough criterion to the column's own effluent
but by  breakthrough at the end of the series. Tests have indicated that the  ammonium
loadings could be increased from 55-58 percent of the equilibrium capacity to 85 percent by
using two  columns in series. 2 Average throughputs for the Tahoe example discussed earlier
increased from 150 to 250 BV/cycle. However, such a two column operation requires three
columns (two on stream while the third is being regenerated) and more complicated valving
and piping than a parallel column operation. Because of the added capital costs involved in a
                                      7-13

-------
                                   FIGURE 7-9

      EFFECT OF BED DEPTH ON AMMONIUM BREAKTHROUGH AT 9.7 BV/HR

     6 i	T	1	1	1	1	1	1	1	
 Oi
 6
  i»


f
 K-
 Uj
 3
 -J
 U.
 U.
 UJ
 Q
 UJ
 OJ
OPERATING  CONDITIONS:

ZEOLITE  GRAIN  SIZE: 20x50 MESH
BED VOLUMES:  3  FT  DEPTH = 25 FT3, 6 FT DEPTH = 50 FT3
              (0.9 m =708 «)          (l.8m = 14164)
AVG INFLUENT  NH^-IM: 17 mg/l
LOCATION: TAHOE
RATE: 9.7 BV/Hr
                                                      6 FT  DEPTH
                                      _L
             20     4O      60      8O      IOO
                                 BED VOLUMES
                                           I2O
140
I6O
ISO
 series  system, all of the full-scale systems currently under design or in operation utilize
 parallel single beds. By blending the effluents from several parallel columns, each of which is
 in a different stage of exhaustion, improved utilization of the available exchange capacity is
 also achieved. That is, if equal volumes of effluent containing 2 mg/l  NH4—N from one
 column are blended with effluent containing 0.6 mg/l from another, some added throughput
 through the  more heavily loaded column  could be achieved while still  meeting an overall
 standard of no more than 2 mg/l NH4-N.

     7.2.9 Determination of Ion Exchanger Size

 In order to calculate the size of the ion exchange unit needed, the ammonium capacity of
 the  clinoptilolite must be determined from the characteristics of the influent water. The
 ammonium  capacity of clinoptilolite can be estimated from Figure 7-10 if the cationic
 strength of the wastewater is known. The data used to plot Figure 7-10 were determined in
 several experimental  runs where the influent ammonium nitrogen  concentration was
 16.4-19.0 mg/jl. Although the curve is empirical  and  is a simplification of the complex
 effect of competing cation concentrations on  ammonium capacity, it illustrates this effect
 and serves as a useful tool in sizing the exchange bed.
                                       7-14

-------
                     FIGURE 7-10

   VARIATION OF AMMONIUM EXCHANGE CAPACITY WITH

       COMPETING CATION CONCENTRATION FOR A

      3 FT DEEP CLINOPTILOLITE BED (REFERENCE 1)
»*
 Ul
 o

 o
 o
 o
 UJ
 V)
 <*
 a:
 Q.
 O
 CO
    0.7
    0.6
  -  O.5
    O.4
0.3
    O.2
    0.1
    0.0
                  \
                                   I
                       Total Ammonium Exchange

                       Capacity
        •Effective  Ammonium Exchange Capacity-
        (to 1 mg/£  NHj-N in effluent)
                                       I
0.015
                                           0.02
                                m,zf,
               0.005      0.01

          CATIONIC  STRENGTH,  1/

          Where-, m = concentration of the cation  species  i

                z = valence of the cation species i

                        7-15

-------
Assuming that the influent water has a cationic strength of 0.006 moles/1, the breakthrough
ammonium capacity of the clinoptilolite will be approximately 0.25 meq/g for a 3-ft (0.9
m)  bed;  the capacity to  saturation  will be approximately 0.44 meq/g. A greater effective
ammonium capacity can be realized by increasing the depth of the zeolite bed. The use of a
6-ft (1.8  m) bed would result in greater ammonia capacity per unit of exchanger and while
requiring a deeper structure, the additional cost would be nominal. Assuming that 3 ft (0.9
m) of the zeolite bed will have  an ammonium exchange capacity equal to 0.25 meq/g and
that the  remaining 3 ft (0.9 m) will have a capacity equal to 90  percent of the saturation
capacity  or 0.40 meq/g, the  6 ft (1.8 m) bed will have an effective capacity of 0.32 meq/g
[equivalent to 6.6 eq/cu ft (236  eq/m3) and 5.1 kgr/cu ft (182 kgr/m3)].

The zeolite volume required  to  treat a 10 mgd (0.44 m3/sec) waste flow at 15 BV/hr(1.9
gpm/cu ft or 4.3  1/sec/m3) is 3650  cu ft (102  m3).  Assuming complete  removal of
ammonium, the  throughput  to ammonium breakthrough would then be  165 BV with a run
length  of 11 hr.  Allowing 2 hr down time per cycle for regeneration and rinsing, the zeolite
volume would be increased  proportionately to 4300 cu ft (120.4m3) to accommodate the
total design flow. Using four units, each having the dimensions  12 ft x 15 ft x 6 ft deep
(3.66 m x 4.6 m  x 1.8 m), the total zeolite volume would be 4320 cu ft (121 m3).1

7.3 Regeneration Alternatives

The key  to the applicability  of  this process is the method of handling the spent regenerant.
The following paragraphs discuss available alternates.

     7.3.1 Basic Concepts

As noted earlier, after about 150-200 bed volumes of normal strength municipal waste have
passed  through the  bed,  the capacity of the clinoptilolite has been used to the point that
ammonium  begins  to leak  through the  bed. At  this point,  the  clinoptilolite  must be
regenerated so that  its capacity  to remove ammonium is restored. The resin is regenerated
by passing concentrated  salt solutions through the exchange bed when  the ammonium
concentration in the solid phase has reached the maximum desirable level. The  ammonium-
laden spent regenerant volume is about 2.5 to 5 percent of the throughput treated prior to
regeneration. By removing the ammonium from the spent regenerant, the regenerant may be
reused. The alternative approaches available for regenerant recovery are:

                           •  air stripping
                           •  steam stripping
                           •  electrolytic treatment

These alternatives for regenerant recovery will be discussed  following a discussion of the
regeneration process.
                                        7-16

-------
     7.3.2 Regeneration Process

The ammonium retained on the clinoptilolite exchange sites may be eluted by either sodium
or  calcium  ions contained in a  regenerant solution. While  the normal service  cycle is
downflow, regeneration is carried out by passing the regenerant up through the clinoptilolite
bed.

         7.3.2.1 High pH Regeneration

The approach originally studied for wastewater applications was to use a lime slurry (5 gm/1)
as the regenerant so that the ammonium stripped from the bed during regeneration would
be converted to gaseous ammonia  which could then be removed from the regenerant by air
stripping. 3  It was found that elution with  lime could be speeded  up by the addition  of
sufficient NaCl  to render the regenerant 0.1./V with respect to NaCl. 3

In addition to converting the ammonium ion to ammonia so it can readily be removed from
the regenerant, the volume of regenerant required for complete regeneration has been found
to decrease with increasing regenerant pH. 1 However, high pH regeneration was found to be
accompanied  by  an  operational  problem of  major  significance.^  Precipitation  of
magnesium hydroxide and calcium carbonate  occurs  within the exchanger  during the
regeneration  cycle. This leads to plugging of the exchanger  inlets  and outlets,  as well as
coating  of the clinoptilolite particles. Violent backwashing of the clinoptilolite was found to
be necessary  to remove these precipitants from the clinoptilolite particles, which resulted in
increased mechanical attrition of the clinoptilolite.  Chemical attrition  also increases  at
elevated pH values. 1

Substantial data have been collected on high pH regeneration and are available in references
1, 2 and 3 if this approach is considered.  However, the practical problems of scale control
are major limitations which can be overcome by using neutral regenerants. The use of closed
loop regenerant recovery  processes negates  the disadvantage  of higher regenerant volumes
required at lower regenerant pH values (See Section 7.3.2.2).

         7.3.2.2 Neutral pH Regeneration

Two  of the  largest municipal  wastewater installations under  design which will use
clinoptilolite  are the Upper Occoquan (Virginia) Regional Plant (15 mgd or 0.66 m-^/sec) as
described in Section 9.5.4.1 and the Tahoe-Truckee (California) Sanitation Agency plant (6
mgd or 0.26 m^/sec), both of which will utilize a regenerant  with a pH  near neutral. The
active portion of the regenerant will be a 2  percent sodium chloride solution. Calcium and
potassium will  be eluted as well as ammonia and will  build up in  the regenerant until they
reach equilibrium: The typical elution curve for ammonium with a neutral pH regenerant is
shown  in Figure 7-11. Approximately 25-30  BV  were  required before the ammonium
concentration reached equilibrium.^  Although greater regenerant volumes are required than
                                        7-17

-------
with a high pH regenerant (10-30 BV), this is not a major disadvantage if the regenerant is
recovered and reused in a closed loop system.

Variations in  regenerant flow rates of 4-20 BV/hr do not affect regenerant performance.
Higher rates result in less ammonia removed per volume of regenerant. Typical design values
are  10 BV/hr  which  insures efficient use of the regenerant while keeping headless values at
low levels. Provisions should be made for backwash at rates of 8 gpm/ sq ft (3.9 l/m^/sec)
and surface wash of the contactor prior to initiation of the regenerant flow. Additional
details on neutral pH regeneration are contained in Section 9.5.4.
                                   FIGURE 7-11

                 AMMONIUM ELUTION WITH 2  PERCENT SODIUM
                    CHLORIDE REGENERANT (REFERENCE 5)
    1000
                                 BED
                                      7-18

-------
          7.3.2.3 Effects on Effluent TDS

Effluent Total Dissolved Solids (TDS) is an important consideration in many plants. When a
2-3 percent solution of salt is used for regeneration, elution of this salt remaining in the bed
after regeneration at the start of the service cycle may result in an increase in TDS of about
50 mg/1. The increment would be greater with stronger regenerants. The TDS effect is much
less than for the breakpoint process, however.

     7.3.3 Regenerant Recovery Systems

Ammonia may be removed from the regenerant so that the regenerant may be reused. Air
stripping of a high pH  regenerant, air  stripping of a neutral regenerant, steam  stripping,  or
electrolytic treatment may be used.

          7.3.3.1  Air Stripping of High pH Regenerant

In the original pilot  work on this approach to regenerant recovery, a stripping tower packed
with 1 inch (2.54 cm)  polypropylene saddles was used.^3 Because the regenerant volume is
only a small portion of the total wastewater flow, it becomes feasible to heat the air used in
the stripping process. The regenerant was normally recycled upflow through the zeolite bed
at a flow  rate of 4.8-7.1  gpm/sq  ft (3.3-4.8  l/m^/sec)  until the NH3—N approached a
maximum  concentration. The regenerant  was then  recycled through both the zeolite bed
and the air stripper until the NH3—N was reduced to about 10 mg/1. The liquid flow rate to
the stripper was  normally 2 gpm/sq ft (1.36  l/m^/sec) with an  air/liquid ratio of 150
cu ft/gal  (1.1  m^/l).  Ammonia  removal  in the air stripper generally averaged about  40
percent per cycle at  25 C. Calcium carbonate scaling occurred on the polypropylene saddles,
but the scale could generally be removed by water sprays. The headloss through the 1  inch
(2.54 cm) pilot plant  saddles caused  the  power requirements for  the air  stripping to  be
excessive.  It was suggested  for a full-scale  design that the ammonia stripping tower be sized
to treat the contents of an elutriant tank in 8 hours, using two passes through the tower at
85 percent removal per pass at an air-to-water ratio  of  300 cfm/gpm (2.2 m^/1), and a
loading of 3.5 gpm/ft^(0.63 l/m2/sec)..2  The tower would be a modified cooling tower
with low differential pressure across the tower as discussed in Chapter 8.

An example design  of a 7.5 mgd (0.33 m^/sec) system illustrates how the  air stripping
system can be integrated into an overall system.^ A schematic diagram of the ion exchange
beds, lime elutriant  system, and ammonia  air stripping system is shown in Figure 7-12. For
design  flows, nine beds (12 ft or 3.65  m diameter and 8 ft  or 2.4 m deep)  would be in
service and three beds in regeneration. The  direction of flow for the beds in service would  be
downflow. All beds would operate in  parallel. When a  given volume of  wastewater has
passed through a  set of three beds,  for example, beds  1, 2, and 3, the set of beds would  be
taken off line for regeneration.  At this time elutriant tank A would contain elutriant water
from a  previous regeneration with a very high ammonia nitrogen content  (say 600 mg/1);
tank B would  contain elutriant water with a low ammonia nitrogen content (say 100 mg/1);

                                        7-19

-------
                                                              FIGURE 7-12
                          EXAMPLE ION EXCHANGE - AIR STRIPPING SYSTEM FOR HIGH pH REGENERANT
                                                INFLUENT HEAOtR ,
                                                                     INFLUENT FEED
ION
EXCHANGE
BEDS	
           9 x
ri
O
                                                                      r
                               O x
          Ts       Tf       TJ       TT       T*
          j T  FINAL  j T TBEATCP J T HOOUCT  j T  MIAOtR  ^ T
                                                                                                     10
 N)
 O
             'REGfNCRANT SOLUTION
              l*ltT MCADEft
              KIGCNEKANT
                                           REOIMEHANT SOLUTION SUCTION LINC
             BACKWASH Sumv LINI
                                           TOWK INFLUfNT
                                                  _L
                                                                      REGENERANT STORAGE TANKS


AMMONIA
STRIDING
T01KER

<

)
TOWtR
STORAGE
TANK
i) (£> 1 ' "
fS ^rS TOWIH INFLUCNT LINE
® r
\ TOWtR IFFLUINT LINE __ :



f 1
^— TOMER
HECVCLI
LINE
                                                                                              NoTes

                                                                                              1 BEOS 1.943 ARE SHOWN «M REGENERATION CYCLE
                                                                                              2 BEDS 4 . 12 ARE SHOWN IN SERVICE CVCLE
                                                                                              3 TOWER SHOWN STRIPPING FROM REGENERANT
                                                                                               TANK A. SINGLE CVCLE
                                                                                               TANKS B 4 C ARE IN REGENERATION CVCLE.
                                                                                              * ® INDICATES VALVE CLOScD
                                                                                              5 X INDICATES VALVE OPEN

-------
and tank C would contain nearly ammonia-free elutriant water (say 10 mg/1). The contents
of tank A would be air stripped during the regeneration of exchange beds 1, 2, and 3. The
regeneration would proceed as follows:

     1.   Exchange beds 1, 2 and 3 would be drained to the final effluent.

     2.   Low  ammonia content elutriant  water  from tank B (100 mg/1) would  be
         recirculated upflow through the three exchange beds and back through tank B to
         the exchange beds until the concentration of ammonia in the elutriant began to
         approach  a  maximum  value  (say  600  mg/1). Throughout the  recirculation,
         make-up lime and salt would be added. A pH of about 11.5 would be maintained.
         About 4 BV  will elute 75 percent of the NH^-N.

     3.   After an allotted time (long enough  for elutriant from tank B to approach a
         maximum ammonia concentration), the elutriant would be changed to recircula-
         tion to and from tank C through beds 1, 2, and 3. Tank C with its ammonia-free
         elutriant  water would be recirculated for an allotted time (long enough for
         elutriant from tank C to reach about 100 mg/1) which will bring the elution up to
         more than 90 percent. About 4 BV are required. At this stage of the elution, the
         small amount of ammonia left on the zeolite would be distributed uniformly
         throughout the bed.   Tank A with nearly ammonia-free water (10 mg/1 Nrfy-N,
         water stripped during the  regeneration of beds 1, 2, 3) would be pumped once
         upflow through the bed to further polish the lower portion of the  bed and
         prevent leakage of ammonia during the  downflow service cycle.

         The elutriant tank B (600  mg/1 — Nrfy—N) would  be held for air stripping  during
         the regeneration of the next set (say beds 4, 5, and 6) of ion exchange beds. Tank
         C with 100 mg/1 elution water would become the lead tank for this next set of ion
         exchange beds.  Tank A  with ammonia-free elution  water (10  mg/l-NH^-N,
         water stripped  during  the regeneration of beds 1, 2, 3) would be used  at the
         polishing tank for beds 4, 5, and 6.

     4.   Once the elution of beds  1, 2, and 3 was completed, the three beds would be
         drained back to tank A.

     5.   Beds 1, 2, and 3 would then be filled slowly from the bottom to remove trapped
         air with product water from the other nine beds in service.

     6.   After the beds were  filled with product  water, more product water would  be
         pumped at a high rate through beds 1, 2, and 3 in sequence. The backwash water
         would be returned to the wastewater treatment plant.

     7.   After backwashing was completed, ion exchange beds 1, 2, and 3 would be  placed
         in service and beds 4, 5, and 6 would be taken off line for regeneration.

                                       7-21

-------
Ammonia in the elutriant solution would be removed by air stripping at a pH of about 11.5.
In the preceding example, during the regeneration of beds  1, 2, and  3, the very high
ammonia  nitrogen content (600 mg/1) in the elutriant solution of tank A was  to  be air
stripped. The following procedure would be used:

     1.   The contents of tank A would pass through the tower down into the recycle basin
         below the tower.

     2.   The contents of the recycle basin  would then be pumped back up through the
         tower once  again. This  time, however, the effluent from the tower would flow
         back to tank A.

     3.   The  contents of tank A would now contain about 10 mg/1  of ammonia,  and
         would be ready to  serve as  the polishing volume during the regeneration of ion
         exchange beds 4, 5, and 6.

By  using  the  above  batch countercurrent  recycle technique,  it is  possible to achieve
complete regeneration with only about 4 BV requiring  stripping per cycle. This is a key to
making steam stripping (as discussed later) practical.

         7.3.3.2 Air Stripping of Neutral pH  Regenerant

As previously discussed, the use of high pH regenerant  is accompanied by scab'ng problems
within the ion exchange beds.  Thus, as discussed in Section  7.3.2.2, a regenerant with  2
percent sodium chloride  as  the  active  agent  and pH nearer neutral has been used to
overcome  the  scaling problem. This regenerant  may  also be recovered for  reuse by air
stripping. Figure 7-13  is a schematic diagram of such a system. 6 In this system,  the stripping
tower off-gases are not discharged to  the  atmosphere  but are instead passed through an
absorption tower where  the  ammonia in  the  off-gases is absorbed in sulfuric acid. The
stripping gases are recycled to  the tower. This approach eliminates discharge of ammonia to
the atmosphere and recovers  the ammonia  in  a form suitable for fertilizer usage. The
stripping-absorption approach is applicable  to high pH  regeneration systems as well. It also
reduces scaling  problems in the stripping tower by limiting the CO 2 content of the stripping
air. This system (the Ammonia Removal and Recovery Process — "ARRP") is also discussed
in Chapter 8 (see Section 8.4.1  and Figure 8-7).

In the system shown  in Figure 7-13, batch-countercurrent regenerant flow similar to that
described above for the high pH regenerant is practiced to reduce the amount of regenerant
which must be  stripped per cycle. The first 11 BV of spent regenerant are discharged to the
spent regenerant tank  for stripping. The second and third 11 BV batches are stored and used
as the first 22 BV of regenerant flow in the next regenerant cycle. The last 6-11 BV batch of
regenerant is mixed with the  11 BV of stripped regenerant  for use as the final regenerant
flow in the next cycle.  Thus, although 40-44 BV of regenerant  are passed through the
exchanger per cycle, only 11 BV are actually renovated by air stripping per cycle.

                                        7-22

-------
                     FIGURE 7-1 3
     FLOW DIAGRAM OF NEUTRAL pH REGENERATION
             SYSTEM USING AIR STRIPPING
FIRST 11 BED VOLUMES
                        11 BED VOLUMES

"~[SEC
VOL
CLINO
BED
'
J
1

SPENT
REGENERANT
TANK
DND 11 BED
UMES~~J

INTERMEDIATE
REGENERANT
TANK
TIRST 11 BED
[VOLUMES J
i LT.HI
•
•
[LA;
tlED VOLUMES
RD 11 BED
lURCS"^

INTERMEDIATE
REGENERANT
TANK
>T 6-11
7 \

'
RECOVERED
REGENERANT
TANK
1
PER R

EGENEf
I

NATION
^ 3 	 .... NoOH
^1 .. 	 MAKEUP
NaCI

i
CLARIFIER
Y
SLUDGE
Mg(OH)2
+
X*^ _ H,S04
f ^ —
HI
STRIPPER ABSORBER "
o
V J *"


AHHK
 LAb I  1 / — if.   I
"BED"~VOLUMET
                           7-23

-------
Regenerant  stored  in the  tanks  shown  in  Figure 7-13  varies in ammonium-nitrogen
concentration from  about 250 mg/1 in the spent tank to 50 mg/1 in the recovered regenerant
tank. The intermediate tanks have intermediate concentrations.  The regenerant pH varies
from about 9.5 in the recovered regenerant tank to about 7.0 in the spent tank. As discussed
earlier, higher pH values produce more efficient  regeneration but near-neutral pH levels
avoid problems with magnesium hydroxide precipitation in the bed during regeneration and
attrition of the clinoptilolite caused by high  pH. Media attrition has been insignificant in
pilot studies under these pH conditions."

A  typical  ammonium elution  curve  is  shown  on Figure  7-14  with  the  background
concentrations in each regenerant storage tank also shown. At the end of the cycle, the last
portion of spent  regenerant is discharged  to  the recovered regenerant tank. This has the
effect of neutralizing the alkaline pH from the ARRP process. ARRP effluent is normally at
a pH of 10.7 to 11.0. This is  reduced to 9.5±  by recycling the last portion of spent
regenerant. In this manner, pH is controlled without use of acid.

When spent regenerant is accumulated to a predetermined amount,  the recovery portion of
the process is activated. This system  operates at a flow rate of approximately  1/13  of
average plant flow since the regenerant concentration is about 13 times as concentrated as
plant waste. Initially, sodium hydroxide is  added to the spent regenerant to achieve a pH of
about  11.  Sodium  chloride is also added  because of some salt  loss from the regenerant
solution in  sludge removal  and  bed rinsing.  Following pH adjustment,  the regenerant is
clarified and any magnesium hydroxide formed is removed, the clarified regenerant at pH =
11 is then  pumped to the  ARRP  process  for ammonia removal  and recovery. The ARRP
effluent flows to the recovered regenerant  tank where it is mixed with the last 6-11 BV of
spent regenerant for pH adjustment prior to reuse.

This system is being used in the  design of plants for the Tahoe-Truckee Sanitation Agency
(California)  and for the  Upper  Occoquan (Virginia) Regional Plant discussed in Section
9.5.4.1.

         7.3.3.3 Steam Stripping

Steam  Stripping of regenerant is being practiced at the 0.6 mgd (0.026 m^/sec) Rosemount,
Minnesota physical-chemical plant. 7)8  This process  is economically  feasible only with high
pH regenerant.

The  higher  regenerant volumes  resulting from the neutral regenerant approach are not
economically treated by this approach. This process is feasible only if the regenerant volume
requiring stripping is held to 4 BV per cycle which is achievable with the high pH regenerant
batch recycle system discussed in Section 7.3.3.1.9 In this case, the necessary portion of the
spent regenerant is stripped in a distillation tower in which steam  is injected countercurrent
with the regenerant.  An air cooled  plate-and-tube condenser  condenses the  vapor for
collection in a  covered tank as a one  percent aqueous ammonia solution which could  be

                                        7-24

-------
                                  FIGURE 7-14
                           TYPICAL ELUTION CURVE
     500
     40O
     300
  I

      200
      IOO
                    CONCENTRATION  BEING
                    ELUTED  FROM  BED
BACKGROUND  CONCENTRATION
IN  FIRST  REGENERANT  TANK
                                      SECOND  REGENERANT
                                       TANK
                           RECOVERED
                           REGENERANT
                           TANK
                      10          20           30
                          TOTAL  BED  VOLUMES
                                40
50
used as a fertilizer. A stripping tower depth of 24 feet (7.3 m) and a loading of 7  gpm/ sq ft
(21 l/m^/sec) are being used at Rosemount. Ceramic saddles are used rather than wooden
slat packing because wood is not a suitable packing in a high pH- steam environment.

Heat exchangers are used to transfer heat from the stripped regenerant to the incoming, cold
regenerant. Heat transfer to the incoming regenerant from the condenser used to condense
the stripped  regenerant may also be attractive. Provisions for scale control in the heat
exchangers should be provided.  The steam requirements have been estimated to be 15
pounds per 1,000 gallons (1.8 g/1).  Added information may be found in the Rosemount
design example, Section 9.5.4.2.
                                     7-25

-------
         7.3.3.4 Electrolytic Treatment of Neutral pH Regenerant

In this approach, ammonium  in the regenerant solution is converted to nitrogen gas by
reaction with chlorine which is generated electrolytically from the chlorides already present
in the neutral pH regenerant solution. The regenerant solution is rich in NaCl and CaCl2
which provide the chlorine produced at the anode of the electrolysis cell. A diagram of the
regeneration system is presented in Figure 7-15. The regeneration of the clinoptilolite beds
is accomplished with a  two percent sodium chloride solution. The spent  regenerant is
collected in a  large  holding tank and then subjected to soda ash treatment for calcium
removal. After the soda ash addition, the regenerant is clarified and transferred to another
holding  tank where the  regenerant is recirculated through electrolysis cells  for ammonia
destruction.
                                 FIGURE 7- 15

                SIMPLIFIED FLOW DIAGRAM OF ELECTROLYTIC
                      REGENERANT TREATMENT SYSTEM
                        Na2C03
                         NaOH
ZEOLITE
   BED
                           I
                                        H2,N2
   SPENT
REGENERANT
  HOLDING
    TANK
                 CLARIFIER
                                                      t
RENOVATED
REGENERANT
  HOLDING
    TANK
                                SLUDGE
                      REGENERANT
                                                          REG
                                                           IN
                                                   ANODE
                                             REG .1   I
                                             OUT
                                                                CO
                                                                CO
                                                                > CO
CC QJ
•- o
o
UJ
-I
UJ
                                                        CATHODE
                                                                RECTIFIER
                                     7-26

-------
During the regeneration of the ion exchange bed, a large amount of calcium is eluted from
the  zeolite along  with the ammonia. This  calcium tends to scale  the cathode of  the
electrolysis cell, greatly reducing its life. Calcium may be removed from the spent regenerant
solution  by a soda ash softening process prior to passing the spent regenerant through  the
electrolytic cells. High flow velocities through the electrolysis cells are required in addition
to a low concentration of MgCl2 to minimize scaling of the cathode by calcium hydroxide
and  calcium carbonate. The effects of flow rate are well illustrated by pilot test data. 5 Using
the system shown in Figure 7-14, the flow rate through the cell was initially set at velocities
of 0.13 to 0.16 ft per second (0.04-0.05 m/sec) and a thin buildup of scale was observed on
the cathode at the bottom cell inlet end after 160 hours of operation. After 230 hours of
operation, the flow velocity was reduced  to 0.06 ft per second (0.018 m/sec) and very light
scale buildup  was observed depositing over the entire cathode area. Scale was removed from
a one-square inch (6.45 cm2) area of the cathode and the flow velocity through the cell was
increased to  0.21 ft per second (0.064 m/sec) to determine the effect of scaling at higher
cell velocities. At this increased flow which was maintained for most of the period of this
study, no new scale was deposited on the  cathode. Visually it appeared that from 25 to 50
percent of the previously  deposited scale was removed. These observations suggest that
scaling within the cell can be controlled by sufficient flow velocities.  Acid flushing of  the
cells is necessary  to  remove this scale  when the cell  resistance becomes  too  high  for
economical operation.

In pilot  tests  of the  electrolytic  treatment  of  the  regenerant at Blue  Plains,  about 50
watt-hours of power were required  to destroy  one  gram of ammonia nitrogen.2 When
related to the treatment of water containing 25 mg/1 Nlfy—N, the energy consumed would
be 4.7 kwh/1,000 gallons (1.2 watt-hrs/1).  Tests at South Tahoe also indicated that a value
of 50 watt-hours per gram is reasonable for design.5

The  electrolytic process also results in about 56 cu ft (1586 1) of hydrogen gas being evolved
per pound of ammonia nitrogen  destroyed.  Provisions  must be made to vent,  burn, or
otherwise adequately control the hydrogen gas evolved in  the electrolytic process.

The  major disadvantage of the electrolytic approach is the substantial  amount of electrical
energy required. The electrical requirements of the air stripping (ARRP) system described in
the preceding section are only about 10 percent of that required by the electrolytic process.

7.4 Considerations in Process Selection

The  selective  ion exchange process has the advantages of high efficiency,  insensitivity to
temperature fluctuations,  and removal of ammonium with a minimal addition of dissolved
solids.  It may also be used with regenerant recovery systems which enable the recovery of
the nitrogen removed from the wastewater in  a reusable  form.  Its  major disadvantage is its
relatively  complex operation. The process should be controlled  by a system which will
automatically  initiate and program the regeneration cycle and return the ion exchangers to
normal service.

                                         7-27

-------
The process is particularly attractive for those cases requiring year-round high level removal
of nitrogen and where effluent TDS is of major concern. Although the effluent TDS is
increased by the process  (see Section 7.3.2.3), the overall increase is much less than for the
breakpoint chlorination process. It must be recognized in the sizing of the upstream process
capacities  that there will be backwash wastes returned from the ion exchange process. The
capacity of the clinoptilolite  may be predicted accurately, based on the concentrations of
ions present in the wastewater, minimizing the need for pilot tests for defining ion exchange
capacity. Pilot tests of the overall ion exchange-regenerant recovery system may be useful,
however, in evaluating physical and economic aspects of the proposed system design as
applied to  a specific wastewater.

7.5 References

  1. Koon, J.H. and W.J. Kaufman, Optimization of Ammonia Removal by Ion Exchange
    Using Clinoptilolite.  Environmental  Protection  Agency  Water Pollution  Control
    Research Series No. 17080 DAR 09/71.

  2. Battelle Northwest and the South Tahoe Public Utility District, Wastewater Ammonia
    Removal by Ion Exchange.  Environmental Protection Agency Water Pollution Control
    Research Series No. 17010 ECZ 02/71, February,  1971.

  3. Battelle  Northwest,  Ammonia Removal from Agricultural Runoff and Secondary
    Effluents by Selective Ion Exchange.  Robert A.  Taft Water Research Center Report
    No. TWRC-5, March, 1969.

  4. Mercer,  B.W., Clinoptilolite in Water Pollution Control. The  Ore Bin,  published by
    Oregon Dept. of Geology and Mineral Industries, p. 209, November, 1969.

  5. Prettyman,  R.,  et  al,  Ammonia  Removal by  Ion  Exchange  and  Electrolytic
    Regeneration. Unpublished report, CH2M/Hill Engineers, December, 1973.

  6. Suhr,  L.G., and L. Kepple, Design of a Selective  Ion Exchange System for Ammonia
    Removal.  Presented  at  the  ASCE Environmental Engineering Division  Conference,
    Pennsylvania State University, July, 1974.

  7. Physical-Chemical Plant  Treats Sewage Near the Twin  Cities. Water and Sewage Works,
    p. 86, September, 1973.

  8. Larkman, D., Physical-Chemical Treatment. Chemical Engineering, Deskbook Issue, p.
    87, June 18, 1973.

  9. Personal communication, B.W. Mercer, Battelle Northwest, December 14, 1973.
                                       7-28

-------
                                    CHAPTER 8

                    AIR STRIPPING FOR NITROGEN REMOVAL
8.1 Chemistry and Engineering Principles

The ammonia stripping concept is based on very simple principles. Because of its simplicity,
it offers a reliable means of ammonia removal when applied under appropriate conditions.
The following section describes the basic concept.

     8.1.1 Basic Concept

The equilibrium equation for ammonia in water is represented by:
                              NH*   i   " NH3*  +  H+                      (8-1)

At  ambient temperatures and pH 7, the  reaction is nearly  complete to the left and only
ammonium ions are present.  As the pH,is increased above 7, the reaction is driven to  the
right, and the fraction of dissolved ammonia gas increases until at pH values of  10.5-11.5,
essentially all of the ammonium is converted to NH3 gas (see Figure 6-2). The gaseous .form
may be removed by stripping.

The ammonia stripping process itself (Figure  8-1) consists of:  (1) raising the pH of the water
to values in the range of 10.8 to 11.5 generally with the lime used for phosphorus removal,
(2) formation and reformation of water  droplets in a stripping tower,  and (3)  providing
air-water contact and droplet agitation by circulation of large quantities  of air through the
tower. The towers used for ammonia stripping of municipal wastewaters closely resemble
conventional cooling towersJ Countercurrent towers,  as  opposed to cross-flow towers,
appear best suited to ammonia stripping  applications.

Detailed discussions of mass and enthalpy relationships and theoretical mathematical models
of the stripping process are available in references 2, 3 and 4. However, these models are not
normally used for stripping tower design, and an empirical design procedure is used.

Before addressing detailed design considerations, the general environmental impacts of the
stripping process must be evaluated.  It is obvious from Figure 8-1 that  ammonia is being
discharged  into the  atmosphere. Does the process  solve  a water pollution problem while
creating an air pollution problem? What  is  the fate of the ammonia in the atmosphere?
These questions  must  be satisfactorily addressed  prior to the selection of air stripping for
ammonia nitrogen removal.

8.2 Environmental Considerations

There are three major potential environmental impacts which must be evaluated if use of the

                                        8-1

-------
                                FIGURE 8-1
                       AMMONIA STRIPPING PROCESS
                                            AIR
                                       ^^   OUT    s&

                                      JOL
           HIGHpH
           INFLUENT
                                             PACKED
                                             TOWER
                                          / AIR IN X
                        EFFLUENT TO pH ADJUSTMENT
                         AND OTHER AWT PROCESSES

ammonia stripping  process is  proposed:  air  pollution, washout of ammonia from the
atmosphere, and noise. If these three-concerns cannot be favorably resolved for any given
situation, then the potential process advantages of simplicity and low cost may become only
academic.

    8.2.1 Air Pollution

At an air flow of 500 cu ft per gallon (3.7 m^/1)  and at an ammonia concentration of 23
mg/1 in the tower influent, the concentration of ammonia in the stripping tower discharge is
about 6 mg/m^. As the odor threshold of ammonia is  35  mg/m^, the process does not
present a pollution  problem in this respect. Concentrations of 280-490 mg/m^ have been
reported to cause eye, nose, and throat irritation. 5 Concentrations of 700 mg/m^ can have
adverse effects on plants. Concentrations of  1,700-4,500 mg/m^ must be reached before
                                      8-2

-------
human  or animal toxicities begin  to occur. Ammonia discharged to the atmosphere is a
stable material that is not oxidized to nitrogen oxides in the atmosphere. 6 Ammonia can
react with sulfur dioxide and water to form an ammonium sulfate aerosol. However, for this
consideration to be a limitation, the stripping tower would have to be located adjacent to a
point source of sulfur dioxide.

The production and  release of ammonia  as part of the natural nitrogen cycle is about
50,000,000,000  tons  per year.  Roughly 99.9  percent of the atmosphere's  ammonia
concentration is produced by natural biological processes, primarily the bacterial breakdown
of amino acids. 5.6 Although they are relatively insignificant sources, burning of coal and oil
produces measurable  quantities of ammonia. 6 The background levels of ammonia in the
atmosphere have been observed to vary from .001 mg/m-* to 0.02 mg/m^  with a value of
0.006 mg/m^ being typical.6

Available diffusion technology can be used to estimate the  atmospheric concentration of
ammonia at any point downwind of the stripping tower. 7 Calculations were made for the
Orange  County, California stripping tower for low mixing conditions (wind speed 1 m/sec).
The resulting surface concentrations at the center of the downwind discharge zone including
natural  background levels were as follows:

                  Distance from Tower             Surface Air Concentration
                     ft.       m                   of Ammonia, mg/m^

                     300      91                         5.2
                    1,000    305                         1.6
                    1,600    488                         0.6
                    3,200    975                         0.2
                  16,000   4,877                         0.0006

Background levels of ammonia are reached within 3  miles (4.8km). No  U.S. ammonia
emission standards have  been established by regulatory agencies because there are no known
public health implications at concentrations normally encountered.

The American Conference of Governmental Industrial Hygenists recommended in 1967 an
occupational threshold  limit of 35  mg/m^.S The permissible limit for  ammonia in a
submarine during a 60 day dive is 18 mg/m^.S The Navy's Bureau of Medicine and Surgery
has recommended an ammonia threshold limit for  1 hour of 280 mg/m^. All of these values
are above the 6 mg/m^ which will typically occur  at the tower discharge. As noted above,
no ambient air  quality standards for ammonia exist for the United States.  However, such
ambient air standards exist for Czechoslovakia, the  U.S.S.R., and Ontario, Canada, as shown
below: 5
                                        8-3

-------
                                          Basic Standard               Permissible
                                       mg/mr   Averaging         mg/rn^    Averaging
     Location                                      Time                      Time

     Czechoslovakia                      0.1        24 hr            0.3       30 min
     U.S.S.R.                            0.2        24 hr            0.2       20 min
     Ontario, Canada                    3.5        30 min

     8.2.2 Washout of Ammonia from the Atmosphere

There is a large turnover of ammonia in the atmosphere with the total ammonia content
being displaced once per week on the average. Ammonia is returned to the earth through
gaseous deposition (60 percent), aerosol deposition (22 percent),  and precipitation (18
percent). 5 Although not the most significant mechanism for removal of ammonia from the
atmosphere, precipitation does provide one pathway for the return of atmospheric ammonia
to bodies of water and to soil. In rainfall, the natural background ranges from 0.01 to 1 mg/1
with the most frequently reported values  of 0.1 to 0.2 mg/1. The amount of ammonia in
rainfall  is  directly  related to  the concentration of ammonia in the  atmosphere. Thus,  an
increase in the ammonia in  rainfall would occur only in that area where the stripping tower
discharge increases the natural background ammonia concentration  in  the  atmosphere.
Calculations for the ammonia washout in the rainfall rate of 3 mm/hr (0.12 in./hr) have been
made for the Orange County, California project with the following results:

                                                  Peak Rainfall Ammonia
                  Distance From Tower                 Concentration, mg/1
                      ft        m

                                                            60
                                                            18
                                                            11
                                                             5
                                                             0.5

The  concentrations of ammonia in the  rainfall would approach natural background levels
within 16,000 ft (4.8 km) of the tower. The ultimate fate of the ammonia which is washed
out by rainfall within this 16,000 ft (4.8 km) downwind distance depends on the nature of
the surface upon which it falls. Most soils will retain the ammonia. That portion which lands
on paved areas or directly on a stream will appear in the runoff from that area. Unless the
stripping tower is located upwind in close proximity to a lake or reservoir, the direct return
of ammonia to the aquatic environment by  atmospheric washout should not make a
significant contribution to the  total ammonia discharged to the  aquatic environment.
However, this is a factor which must be carefully evaluated for each potential application.
300
1,000
1,600
3,200
16,000
91
305
488
975
4,877

-------
     8.2.3 Noise

There  are three potentially significant noise sources in  an ammonia stripping tower:  (1)
motors and fan drive equipment; (2) fans; and (3) water splashing. The following control
measures are available:

         •  Motors   - proper installation, maintenance and insulation
         •  Fans     - reduction in tip speed and exhaust silencers
         •  Water    — shielding of the tower packing and air inlet plenum

Based on sound level measurements from the tower at Lake Tahoe, the expected noise level at
the tower is calculated to be about 64 decibels (dBA). This noise level can be reduced to 46
dBA at 600 ft (183 m) from the towers by control measures. The Orange County project (Sec.'
9.5.5.2) includes several specific noise control measures. Before construction of the plant,
ambient nighttime noise levels in the residential neighborhood around the Orange County
plant were 40-45 dBA.

8.3 Stripping Tower System Design Considerations

The major factors affecting design and process performance include the tower configuration,
pH, temperature, hydraulic loading,  tower packing depth and spacing, air flow, and control
of calcium carbonate scaling.

     8.3.1 Type of Stripping Tower

There  are two basic types of stripping towers now being used in full-scale applications:
countercurrent towers and  cross-flow towers (see Figure 8-2). Countercurrent towers (the
entire airflow enters at the bottom of the tower while the water enters the top of the tower
and falls to  the bottom) have been found  to be the most efficient. In the crossflow towers,
the air is pulled into the tower through its sides throughout the height of the packing. This
type of tower has been found to be more prone to scaling problems (see Section 8.3.7).

     8.3.2 pH

The pH of the water has a major effect on the efficiency of the  process. The pH must be
raised to the point that all of the ammonium ion is  converted to  ammonia gas (see Section
8.1.1). If phosphorus removal is required, the use of lime as the coagulant will generally
enable the necessary pH elevation to be achieved concurrent with phosphorus removal.  If
pH elevation  does not occur in some upstream processes,  then the economics  of the
stripping process are adversely  affected since the  costs of pH elevation must  then be
incurred solely for ammonia stripping.
                                        8-5

-------
                                    FIGURE 8-2
                 TYPES OF STRIPPING TOWERS (REFERENCE 8)
                                               ^-COLLECTION BASIN
                                   CROSS-FLOW TOWER
                                   FAN
                                          1 AIR
                                          [OUTLET
                        AIR INLET
                                                   DRIFT
                                                   ELIMINATORS
                                                    DISTRIBUTION
                                                    SYSTEM
                                                     -AIR INLET
                                                  WATER
                                                  COLLECTING BASIN
                                  COUNTERCURRENT TOWER

     8.3.3 Temperature

A critical factor is the air temperature. The water temperature reaches equilibrium at a value
near the  air temperature in the  top few inches of the stripping  tower.  As the water
temperature decreases, the solubility  of ammonia in water increases and it becomes more
difficult  to  remove  the  ammonia by stripping. The amount of air per gallon  must be
increased to maintain a given degree of removal as temperature decreases. However, it is not
practical to supply enough air to fully offset major temperature decreases. For example, at
20  C,  90-95  percent removal of ammonia is typically achieved. At 10 C, the maximum
practical removal efficiency drops to about 75 percent. Data collected in pilot tests by EPA
at the Blue Plains plant in Washington, D.C. well illustrate the temperature effects and are
shown  in Figure 8-3.9  in  warm  weather  tests at pH  11.5, with inlet air  and water
temperatures averaging 25.5 C and 26 C respectively, air stripping cooled the outlet water
temperature by evaporation of the liquid within the tower to an average of 22.2 C. In a
                                         8-6

-------
                                  FIGURE 8-3

        EFFECT OF TEMPERATURE ON AMMONIA REMOVAL EFFICIENCY
           OBSERVED AT BLUE PLAINS PILOT PLANT (REFERENCE 9)
   too
    80
 UJ
 o
 UJ
 a.  6O
 uj  4Q
 oc
 Z
 O
 s  20
 «t
         TOWER  EFFLUENT  WA.TER
         TEMPERATURE = 22.2 C


         /'     •                        LIQUID RATES
    >/'*                                   gpm/sf
    /     TOWER EFFLUENT WATER           B   2 44
 /     TEMPERATURE =5C                A   2 QQ
                                            •   1.50
Metric Conversion:                          *   '-O0
1  gpm/sf = 0.68 4/m2/sec
1  cu ft/gal  = 7.48 m3/m3
      0               200              400              6OO             8OO
                              CU FT AIR /GAL  LIQUID

similar test with the inlet air temperature averaging 6 C and the inlet water temperature
averaging 16 C, the air stripping cooled the outlet water to an average temperature of 5 C.
Data from both the 22.2 C and 5  C conditions are shown in Figure 8-3. The decrease in
efficiency from the warm to cold temperatures was approximately 30 percent over a wide
range of air to water flows.

When  air temperatures reach freezing (or when the wet bulb temperature of the air within
the tower reaches  0 C), the tower  operation must generally be shut down due to icing
problems.  The very large  volumes of air required  for the stripping process make it
impractical to heat the air in cold climates. Waste heat from potential on-site sources such as
sludge incinerators is typically only a small percentage of that needed.

    8.3.4 Hydraulic Loading

The hydraulic loading rate of the tower is an important factor. This is typically expressed in
terms of gallons/minute applied to  each square foot (or l/m^/sec) of the plan area of the

                                      8-7

-------
tower packing.  When  the  hydraulic loading rates become too high,  the  good droplet
formation needed for efficient stripping is disrupted and the water begins to flow in sheets.
If the rate is too low, the packing may not be properly wetted resulting in poor performance
and  scale accumulation.  Data collected in  pilot tests at South  Tahoe  illustrate this
relationship and are shown in Figure 8-4.^ In optimum summer conditions, tthe pilot data
indicate that  a  flow rate of 2 gpm/sf (1.4 l/m^/sec) is compatible with efficient tower
operation at 20-24 ft (6.1-7.3 m) packing depths. Adequate flow distribution over the entire
packing area is a critical factor. Full-scale towers at Orange County and Pretoria,  South
Africa are based on tower loadings of 1-1.13 gpm/sf (0.68-0.771/m^/sec).
                                  FIGURE 8-4

          PERCENT AMMONIA REMOVAL VS. SURFACE LOADING RATE
             FOR VARIOUS DEPTHS OF PACKING (REFERENCE 10)
   100
          Metric Conversion:
           1 ft  =  0.305m
           1 gpm/sf = 0.68 ^/m2/sec
                I	I	I
Z
UJ
o
cc
UJ
Q.
 - 60 -
5
O
5
UJ
Q:
5
5
   2O-
                         2.O        3.0        4.0         5.0
                         SURFACE  LOADING RATE, gpm/sf
                                      8-8

-------
     8.3.5 Tower Packing

         8.3.5.1 Packing Depth

The  depth of tower packing required for maximum ammonia removal will depend on the
tower packing selected. Most stripping tower designs are based on  the use of an open,
cooling tower-type packing (horizontal packing members spaced about 2 in (5.1 cm) apart
both horizontally  and vertically) to minimize the power required to move adequate air
quantities through  the tower (see Sec. 8.3.6). If maximum  removals are desired, tower
packing depth should be at least 24 ft (7.3 m) with this type of packing, unless pilot plant
data indicate that a lesser depth of a specific packing will accomplish the required removal.
Packing with members spaced more than 2 inches (5.1  cm) apart may require greater depths,
and  pilot tests  should be  run to determine the required depth if  greater spacings are
proposed.

         8.3.5.2 Packing Material and Shape

Both wood (Lake Tahoe) and plastic packings (Orange County) have been used in full-scale
towers. The smooth plastic surfaces appear to be one factor accounting for reduced calcium
carbonate scaling at the Orange County facility. Plastic packing has an advantage in that it
does not suffer from the delignification that occurs with wood at elevated pH values.

Pilot studies at Orange County evaluated three different types of packing: Vi inch (1.27 cm)
diameter PVC pipe, triangular shaped splash bars, and vertical film packing like that used in
cooling towers. * 1 With vertical packing the water  moves in a thin film down vertical sheets
of packing rather  than moving as droplets as occurs in packing composed of horizontal
splash bars. The film packing was found to provide only 50 percent or less ammonia removal
and was eliminated from  consideration early in the tests. Film packing fails to provide the
repeated droplet formation and rupture needed for  efficient stripping.

Since repeated splashing and droplet formation is a key parameter in ammonia stripping, a
triangular shaped splash bar was tested. It was thought that it  might provide two points of
droplet formation compared to only one for a round splash bar. It was observed that droplet
formation throughout the tower still occurred at only one point. The water flowed down
the sides of the triangle and around the corners, where it collected on  the base and dripped
from a  single point.  Air flow and pressure  drop measurements were made on both the
circular and triangular packing. The static pressure drop (24 ft  or 7.3 m packing  depth) was
0.40-0.44 in  (1-1.1 cm)  of water when the triangular packing was  used, compared to
0.36-0.40 in (0.9-1  cm) of water when  the circular packing was used.  No  significant
differences in ammonia removals were noted between the round and triangular shaped
packings.
                                        8-9

-------
         8.3.5.3 Packing Spacing and Configuration

Figure 8-5 is a sideview of a typical packing configuration using wood or plastic slats. In this
example, the slats are spaced 2 in. (5.1 cm) apart (center to center) on the horizontal and
1.5 in. (3.8 cm) apart vertically. This spacing is referred to as  1.5 x 2 in.(3.8 x 5.1  cm).
Figure 8-6 shows that the spacing of the tower packing is important in determining the air
requirements for ammonia stripping. The 1.5 x 2 in.(3.8 x 5.1 cm) packing has 2.66 more
slats for droplet formation and coalescing than does a 4 x 4 inch (10.2 x  10.2 cm) packing.
Although spacing the packing members closer than 1.5 x 2 inches (3.8 x 5.1 cm) would
improve performance, the  increased pressure drop  would greatly increase power costs (see
Sec. 8.3.6).

Tests at  Orange County indicate that  packings in which alternate layers of packing are
placed at right angles, rather than the parallel position shown in Figure 8-5, maintains better
flow distribution and may be less susceptible to scale accumulation.

                                  FIGURE 8-5

                  ILLUSTRATIVE PACKING CONFIGURATION
                     5.1  cm
                                                         3.8 cm
f.5
                                      I
                                        8-10

-------
                                 FIGURE 8-6
             EFFECT OF PACKING SPACE ON AIR REQUIREMENTS
              AND EFFICIENCY OF AMMONIA STRIPPING (REF. 1)
                                 I       'I         '        I         I
                            '/2 x  2QIN. PACKING (REDWOOD SLATS)    o
                            4x4  IN. PACKING (PLASTIC  TRUSS BARS)
                                    NOTE:  24 FT  PACKING DEPTH
              500     1000     /500    2000    2500    3OOO    3500   40OO
                     CU FT  AIR/QALLON  TREATED
    8.3.6 Air Flow

Gas transfer relationships indicate that an increase in ammonia removal'can be achieved by
increasing the air flow for a given tower height (see Figures 8-3 and 8-6). However, there is a
practical limit on air flow rate due to the increase in air pressure drop with increasing flow
rate. This results in higher capital investment for fans and increased power costs. The air
pressure drop in a countercurrent tower is given as: 2
where:
             P = f-z-Q .


Pressure drop, in. of water
Fanning friction factor
Air flow rate, cu ft/min/sq ft (m /min/m )
                                                                        (8-2)
            air
           z   =   Packing height, ft
                                     8-11

-------
Pressure drop increases exponentially with air flow rate. In general, air velocities of 550 cu
ft/min/sq  ft (1600  m^/min/m^) are considered to  be the  practical  upper limit for
countercurrent  towers. The  friction factor should be obtained from the packing manufac-
turer. General guides for wood grids are available in reference 3.

Figure 8-3 reflects the effects of the ratio of air to wastewater as observed at the Blue Plains
pilot  plant.  These data are in general agreement with similar pilot data collected at South
Lake  Tahoe (Figure 8-6). For warm weather conditions, typical air requirements are about
300 cu ft/gal (2240 m3/m3)  for 90 percent removal and 500 cu ft/gal 3740 m3/m3 for 95
percent removal.  In  cold weather conditions, the air requirements to achieve maximum
tower efficiency  increase substantially.  Full-scale  data  at Tahoe indicate that,  for  their
packing design, air flows of about 800 cu  ft/gal (5980 m3/m3) would be needed to achieve
90 percent removal at an air  temperature of 4 C. However, reliance solely on the stripping
process in cold weather conditions is usually not practical, and most designs are  based on
moderate to warm weather conditions. Typical air design quantities for 90 percent removal
are as follows: Orange County, California - 400 cf/gal (2990  m3/m3); South Lake Tahoe -
390 cf/gal (2920 m3/m3); Pretoria, South Africa - 338 cf/gal  (2530 m3/m3).

The required air quantities are usually provided by a fan located on top of the tower. Two
speed  fan motors may be used to better  match air supplied to the actual requirements.
Because of the low pressure  drops associated with the types of packings typically used (less
than  1 inch or 2.5 cm water),  the horsepower requirements  for the fans  are not great for
these  large quantities of air.  For the 15 mgd (0.65  m3/sec) Orange County plant, the total
installed fan brake horsepower is 1380 HP.

     8.3.7 Scale Control

A factor  which may have an  adverse effect on tower efficiency is scaling of the tower
packing resulting from deposition  of calcium carbonate from the unstable, high pH water
flowing through the tower. Scaling potential can be minimized by maximizing the extent of
completion  of the calcium carbonate reaction in the lime treatment step. Using a high level
of solids recycle  in  the clarification  step will ensure more complete reaction. Another
approach is to eliminate CO2  from the air (see Section 8.4).

The original crossflow tower at the South Lake Tahoe plant has suffered a severe scaling
problem. The severity of the scaling problem was not anticipated from  the pilot studies in
which a countercurrent tower was used. As a result, the full-scale cross-flow tower packing
was not designed  with access for scale removal in mind. Thus, portions of the tower packing
are inaccessible for cleaning. Those portions  which were accessible were readily cleaned by
high pressure hosing.

The severity of the scaling problem has varied widely. Perhaps the most severe case is that
reported at  the Blue Plains pilot plant. 9 When operating at a pH of 11.5, a heavy scale of
                                         8-12

-------
calcium carbonate formed on the crossflow tower packing (polypropylene grids). The scale
was crystalline, hard, and could not be removed by a high pressure water hose. In contrast,
several months of operation of the countercurrent  pilot  tower at Orange  County  at pH
values above 11 resulted in only a thin coat of calcium carbonate  scale  on  the 0.5 in.(l-3
cm)  diameter PVC pipe packing. The thin coat of scale stabilized and did not continue to
accumulate. The scale was  friable and easily removed by water hosing. The Orange County
pilot  tower was later moved to South Tahoe where several months of operation indicated no
scale  buildup. The differences in operation between  the pilot tower  at Tahoe and the
full-scale Tahoe tower were as follows: packing of plastic rather than wood, packing shape
round rather than rectangular, countercurrent tower rather than parallel.  The relative
importance of these factors in eliminating the scaling noted in the  full-scale  Tahoe tower is
uncertain. The experience with the  full-scale (1 mgd) countercurrent  tower at Pretoria,
South Africa is similar to that at the  full-scale Tahoe tower in that the scale formed  can be
readily flushed from the packing by a water jet. ^ The feeding of a scale-inhibiting polymer
to the tower influent may  also offer a means of scale control and such provisions are being
made in the Orange County facility.

In light  of the  as-yet unpredictable  nature  of factors contributing to scaling of tower
packing,  it is prudent to conduct pilot tests for 3-6 months on  the specific wastewater
involved  with the specific tower configuration proposed. The pilot vs. full-scale experience
at Tahoe and  the independent  pilot tests at Orange County  indicate that the use of
countercurrent rather  than crossflow towers will reduce scaling  problems.  Also,  access
should be provided to all of the tower packing for cleaning (see Sec. 9.5.5.2 for a discussion
of the Orange County plant).

8.4 Ammonia Recovery or Removal From Off-Gases

As noted earlier, the calcium carbonate scaling problem can be minimized or eliminated by
removing carbon  dioxide  (CO2) from  the  stripping air.  This section  describes  two
approaches which accomplish this goal by removing ammonia  and CO2 from the off-gas
from  the tower and recycling the air.

     8.4.1 Acid Systems

An approach to  overcoming the limitations of the stripping process is currently being
developed. 13 it appears that  the process overcomes  many limitations of the stripping
process and has the advantage of recovery of ammonia as a byproduct.

The  improved process is shown  diagramatically  on Figure 8-7. The process includes an
ammonia stripping unit and an ammonia absorption unit. Both of these units are  sealed
from  the outside air but are connected  by  appropriate ducting. The stripping gas,  which
initially is  air,  is  maintained  in  a closed cycle. The stripping unit operates  in  the same
manner as is now being or has been done in a number of systems with the exception that the
gas stream is recycled rather than outside air being used in a single pass manner.

                                        8-13

-------
                                 FIGURE 8-7
            PROCESS FOR AMMONIA REMOVAL AND RECOVERY
 WASTEWATER
 CONTAINING 	
 DISSOLVED
 AMMONIA (NH3)
                                     FAN (TYPICAL)
RECYCLE   ,  .
ALTERNATE  11
    PUMP
VS STREAM WITH
/IMONIA INCREASED
                A,  AA  A
                   STRIPPING
                     UNIT
                                       Vj^TS ^V
                                       /TVi/   \
                                                    DUCTING (TYPICAL)
                                                   V
                                               A
                                          ABSORPTION
                                             UNIT
                                   RECYCLED
                                   ABSORBENT
                                   LIQUID
                          GAS STREAM-AMMONIA
                          REDUCED BY ABSORPTION
                                                                    ACID AND
                                                                    WATER MAKEUP
                                                                AMMONIUM SALT
                                                                SLOWDOWN (LIQUID
                                                                OR SOLID), OR
                                                                DISCHARGE TO STEAM
                                                                STRIPPER FOR AMMONIA
                                                                GAS REMOVAL AND
                                                                RECOVERY
                          WASTEWATER STRIPPED OF NEARLY
                          ALL OR PART OF AMMONIA (NH3)
Most of the ammonia discharged to the gas stream from the stripping unit is removed in the
absorption unit. Because of the favorable kinetics of the absorption reaction, the absorption
unit may  be reduced in size by about one third from that required for the stripping unit.
The absorbing liquid is maintained at a low pH to convert absorbed and dissolved ammonia
gas to ammonium  ion. This  effectively traps the ammonia and  also has the effect of
maintaining the full driving force for absorbing the ammonia since dissolved ammonia gas
does not build up in the absorbent liquid. The absorption unit can be a slat tower or packed
tower using sprays similar to the  stripping unit,  but will usually  be smaller due to the
kinetics of the absorption process.

The absorbent liquid initially consists of water with acid to obtain low pH (usually below
7). In the simplest case, as ammonia gas is dissolved in  the absorbent and converted to
ammonium ions, acid is added to maintain the desired pH. If sulfuric acid is added, as an
                                        8-14

-------
example, an ammonium sulfate salt solution is formed. This salt solution continues to build
up in concentration and the ammonia is finally discharged from the absorption device as a
liquid or solid (precipitate) blowdown of the absorbent. With shortages of ammonia based
fertilizers, a saleable byproduct may result. Ammonia sulfate concentrations of 50 percent
are obtainable.

Mist eliminators are necessary between the absorber and stripper to prevent carryover of the
ammonia laden moisture from the absorber to the stripper effluent. Because of the headless
in the mist eliminators and absorber packing, total headless  for the air approaches 2 inches
(5.1  cm).  It is believed that the usual scaling problem associated with ammonia stripping
towers will be reduced by the improved process, since the carbon dioxide which normally
reacts with the calcium and hydroxide ions in the water to form the calcium  carbonate scale
is eliminated from the  stripping air during the first  few passes. The freezing  problem is
eliminated due to the exclusion of nearly all outside air. The treatment system will normally
operate at the temperature of the wastewater.

As  discussed in Chapter 7, this approach  is being used for  stripping of  ammonia  from
selective ion  exchange process regenerant. Because of the higher power requirements (as
compared to single-pass  stripping), the use of this process may be limited to regeneration of
the brines from the selective ion exchange process. Full-scale designs of this application are
underway for a 22.5 mgd (1.0 m^/sec) plant serving the Upper Occoquan Sewage Authority,
Virginia and for a 6 mgd (0.26 m^/sec) plant for the Tahoe-Truckee Sanitation Agency,
California. Design criteria for this application are presented in  Sec. 9.5.4.1 which describes
the Upper Occoquan plant.

     8.4.2 Nitrification-Denitrification

Another approach to  achieving many of the same objectives of the acid system described
above is also being developed. 14  The  system,  termed the Ammonia Elimination System
(AES), is shown in Figure 8-8. Basic elements of the process are  as follows:

     1.   Ammonia Stripping Tower. Ammonia is  transferred in this tower from the liquid
         stream to the gas stream at high pH.

     2.   Ammonia Absorption-Oxidation Tower. Ammonia  is transferred from the gas
         stream to the liquid stream at about pH 8. Ammonia is oxidized to nitrate by
         nitrifying bacteria in this tower. The  following reaction summarizes the reactions
         in this tower:

                                                                               (8-3)

     3.   Denitrifying Reactor. Nitrate is reduced to nitrogen gas in this reactor (methanol
         or a waste carbon source is added to the reactor) as  shown by the reaction:
                                        8-15

-------
N0~ + 1 16
5/6 CH3(OH)
1 /2
8/6
                                                   HCO
                                                                             (8-4)
    4.   Solids Separation. Denitrification organisms are settled from the process stream
         and are returned to the denitriflcation reactor or wasted.

    5.   The overall reaction in the AES system can be found by adding equations 8-3 and
         8-4:
2O2 + 5/6 CH3(OH) — •- 1/2 N2 f + 7/3 H2
-------
     1.    Removal of nitrogen from the main flow stream by a physical process (which has
          advantages over biological systems in terms of ease of process control).

     2.    Isolation of the nitrification stage from most agents in the plant influent which
          are toxic to nitrifying organisms.

     3.    Isolation of the ammonium oxidation stage from the carbon oxidation or removal
          stage.

     4.    Nitrogen is eliminated from the system as inoffensive nitrogen gas.

In addition to the above advantages, the AES has the following potential advantages not
shared by either ammonia stripping or the three sludge system:

     1.    The insulation  and heating of the liquid recycle streams, oxidation column,
          denitrification  reactor, and  clarifier(s)  to increase  process efficiency  becomes
          economically feasible.

     2.    A waste carbon source may be used in place of me than ol, since denitrification is
          accomplished on a side stream.

The AES shares the following advantages of the acid system described in Sec. 8.4.1:

     1.    Free ammonia gas is not discharged to the atmosphere.

     2.    Water vapor is not discharged to the atmosphere in large volumes.

     3.    The heating of air to prevent tower  freezing  and increase process efficiency
          becomes economically feasible due to gas recycle.

     4.    Gas recycle reduces scaling problems in  the stripping tower as recycle gas will be
          very much lower in CC>2 content than atmospheric air.

From Equation 8-4, it can be seen that there is no net requirement for acid, as in the acid
system described in Section 8.4.1.

8.5 Stripping Ponds

The South Tahoe system has been modified to reduce the impact of temperature and scaling
limitations encountered  at the plant. ^  Basically, the modified process consists of three
steps  (See Figure  8-9):  (1) holding  in high pH, surface agitated ponds, (2) stripping in a
modified, crossflow forced draft tower through water sprays installed in the tower, and (3)
breakpoint chlorination (see Sec. 9.5.5.1 for  pilot plant results). This system was inspired by
observations in Israel of ammonia nitrogen losses from high pH holding ponds. 1°

                                        8-17

-------
                                    FIGURE 8-9
                       AMMONIA STRIPPING POND SYSTEM
                              AlR SPRAYING OF RECYCLED POND WATER
                                 IN THE SECOND OF TWO PONDS
IN SECOND POND. TWO
 RECYCLE PUMPS
 34 mgd CAPACITY
414 TO 13» RECYCLES
  CLARIFIED LIME
TREATED WASTEWATER.
    pH- 11.0
            TWO HIGH pH PONDS IN SERIES
           7 TO 18 HOURS DETENTION TIME
                                                            FLOW VARIES. 2.5 TO 7.5 mgd
                  New High  pH  Flow  Equalization  Ponds
   EXISTING CROSS FLOW AMMONIA
       STRIPPING TOWER
               Stripping  Tower  Modified  with New Sprays
         COZ OR







••


•



/ EXISTING 2 STAGE
/ RECARBONATION
^^/ BASIN


i










^^N-


.^-*— -










TO FILTERS AND
CARBON COLUMN
NEW BREAKPOINT EXISTING 1 MG
CHlORINATION BALLAST POND
CHAMBER FOR CHLORINE CONTACT
                        Breakpoint  Chlorination
                                       8-18

-------
Pilot tests at South Tahoe indicated that the release of ammonia from high pH ponds could
be accelerated by agitation of the pond surface. In the modified Tahoe system, the high pH
effluent from the lime clarification  process flows to holding ponds. Holding pond detention
times of 7-18 hours are used in the modified South Tahoe plant. The pond contents are
agitated and recycled  4-13 times by pumping the pond contents  through vertical spray
nozzles into the air above  the ponds. The holding pond detention time and number of
recycles vary with plant flow with the time and cycles decreasing as plant flow increases. At
least  37 percent  ammonia removal in the  ponds is anticipated,  even in cold weather
conditions. The pond  contents are  then sprayed into the forced draft tower. The packing
has been removed from the  tower and the entire area of the tower is equipped with water
sprays. At  least 42 percent removal of the ammonia in the pond effluent is anticipated,
based on pilot  tests, from this added spray in cold weather that includes recycling of the
pond effluent through  the tower to achieve 2-5 spraying cycles. The ammonia escaping this
process is removed by  downstream  breakpoint chlorination. It appears that stripping ponds
offer an approach that  takes advantage of the low  cost and simplicity  of the stripping
process for removal of the bulk of the nitrogen, making  breakpoint  chlorination more
attractive for complete removal of ammonia.

8.6 Considerations in Process Selection

One of the great advantages  of air stripping is its  extreme simplicity. Water  is merely
pumped to the top of the tower at a high pH, air is drawn through the fill, and the ammonia
is  stripped from the water droplets. The  only control required  is the proper  pH in the
influent water. This simplicity of  operation  enhances the reliability of the process.  The
process costs are  also significantly  less than any alternate method of nitrogen removal,
assuming that the needed pH elevation occurs in  conjunction with  upstream phosphorus
removal.

The major  engineering  limitations on application of the process result from its sensitivity to
temperature variations and from potential scale accumulation on the tower packing.  The
first of these limitations is that of temperature as discussed in Section 8.3.3.  Ammonia
removals decrease  as air and water temperatures decrease. Although increased air flows can
offset temperature effects to some  degree, it is not practical to supply enough air to offset
major temperature drops. It has not  been practical to operate stripping towers at ambient air
temperatures below 0 C. Of course, this  is not  a limitation in climates where freezing
temperatures do not occur or for plants where nitrogen  removal is not required during cold
or freezing weather. Modifications (Section 8.4) of the ammonia stripping process are being
developed which may eliminate temperature limitations.

For applications in cold weather where a high degree of nitrogen removal is required, the
stripping process itself will generally not be adequate. It is usually not practical to heat the
large quantities of air required for the stripping process unless one is so fortunate as to have a
large source of waste heat in  proximity to the stripping tower (see Section 9.5.5.2 for a
description of the use of waste heat from a desalting plant to heat stripping tower air while

                                        8-19

-------
providing  needed cooling of desalting plant water and wastes). The use of the stripping
process supplemented as  needed in cold weather by breakpoint chlorination is  a process
combination that may be attractive in some cases where efficient cold weather operation is
needed and cold weather conditions do not persist for prolonged portions'of the year.

A potentially serious problem is the formation and accumulation of calcium carbonate scale
on the tower packing. Designs should anticipate this problem and provide for easy access for
cleaning the packing as has been done in the Orange County tower (Sec. 9.5.5.2). It appears
that scaling is not as severe in a countercurrent tower as in a crossflow tower and that scale
does not adhere as tightly to the smooth, hard surface of plastic packing as it does to the
rough, soft wood surface. The more open "criss-cross" type of packing developed at Orange
County may  also  be  more resistant  to  scale accumulation than  the parallel packing
arrangement used at Tahoe. Experience to date indicates  that if adequate provisions are
made in the design of stripping towers for access to packing, in many cases scale removal
can be accomplished by  water  sprays without the use of chemicals or mechanical means.
However,  this is a factor  that deserves  special consideration and investigation at  each
location, since the scaling characteristics of different wastewaters may differ markedly. In
light of the as-yet unpredictable nature of causes contributing to scaling of tower packing, it
would be  prudent to conduct pilot tests for 3-6 months on the specific wastewater involved
with the specific tower configuration proposed.

8.7 References

  1.  Slechta, A.  F., and G. L. Gulp, Water Reclamation Studies at the South Tahoe Public
     Utility District. JWPCF, 39, No. 5, pp 787-814 (1967).

  2.  Roesler, J.  F.,  Smith,  R., and  R. G. Eilers, Mathematical Simulation of Ammonia
     Stripping Towers for Wastewater Treatment.  U.S. Department of Interior — FWPCA,
     Cincinnati, Ohio, January, 1970.

  3.  Roesler, J.F., Smith, R., and R.G.  Eilers, Simulation of Ammonia Stripping From
     Wastewater. JSED, Proc. ASCE, 97, No. SA 3, pp 269-286 (1971).

  4.  Perry, J.H., Chemical Engineering Handbook. McGraw Hill Book Co., New York, NY.

  5.  Miner, S., Preliminary Air Pollution Survey of Ammonia. U.S. Public Health Service,
     Contract No. PH22-68-25, October, 1969.

  6.  Nitrogenous Compounds  in  the Environment. EPA  Report SAB-73-001, December,
     1973.

  7.  Pasquill, F., Atmospheric Diffusion. D. Nostrand Co., Ltd., London, 1962.

  8.  Gulp, R.L., and G.L. Gulp, Advanced Wastewater Treatment. Van Nostrand Reinhold,
     New York,  1971.
                                       8-20

-------
 9.  O'Farrell, T.P., Bishop, D. F., and  A.F.  Cassel, Nitrogen Removal by Ammonia
    Stripping. EPA Report 670/2-73-040, September, 1973.

10.  South Tahoe Public Utility District, Advanced Wastewater Treatment as Practiced at
    South Tahoe. EPA Report 1701OELQ  08/71, August, 1971.

11.  Wesner, G.M., and D.G. Argo, Report on Pilot Waste Water Reclamation Study. Orange
    County Water District Report, July, 1973.

12.  van Vuuren, L.R., personal communication, May 30, 1972.

13.  Kepple, L.G., Ammonia Removal and  Recovery Becomes Feasible. Water and Sewage
    Works, p. 42, April, 1974.

14.  Brown  and  Caldwell, Application for a.  Research  Contract  for the  Ammonia
    Elimination  System. Submitted to the Office of Research and Development, Water
    Quality Office, EPA, March, 1971.

15.  Gonzales, J.G., and R.L. Gulp. New Developments in Ammonia Stripping. Public Works,
    104, No. 5, p 78 (1973) and No. 6, p 82 (1973).

16.  Folkman, Y., and A.M.  Wachs, Nitrogen Removal Through Ammonia Release From
    Ponds.  Proceedings, 6th Annual International Water Pollution Research Conference,
    Tel Aviv, Israel, June 18-23, 1972.

17.  Gulp, G.L., Physical-Chemical Techniques for Nitrogen Removal. Prepared for the EPA
    Technology Transfer Program, July, 1974.
                                     8-21

-------
                                    CHAPTER 9

                              TOTAL SYSTEM DESIGN
9.1 Introduction

With a defined set of effluent quality objectives, the environmental engineer must develop a
cost-effective treatment system that is suited to  the local situation. A variety of nitrogen
removal options are presented in this manual that may form a part of the total treatment
system. No one combination of processes is universally applicable.

The main thrust of this chapter is to present specific examples of treatment systems that
have been selected and implemented.  Design concepts that have  evolved  to suit local
circumstances are emphasized.

9.2 Influence of Effluent Quality Objectives on Total System Design

The effluent  quality obtainable from a treatment  system has the most significant impact on
its  selection  or rejection.  When  only ammonia  removal (or conversion)  is required,  as
opposed   to  total   nitrogen  removal,  cost considerations  would  dictate  selection  of
nitrification over any other method of ammonia removal in most cases. Exceptions would
be  low temperature ( < 5 C) situations  or where toxicants which  cannot be effectively
removed  by  a source  control program  are  present in  the wastewater.  Other possible
exceptions are  cases  wherein influent   nitrogen  levels  are so low that  they  can be
economically removed by breakpoint chlorination and where the carbon to nitrogen ratio is
such that all the nitrogen  can be assimilated  into  biomass  during biological oxidation  of
organics.

In Section 2.5.5 the effectiveness of each type of process in removing the various nitrogen
species was summarized. Tables 9-1, 9-2  and  9-3 contain representative nitrogen data for
final effluents from systems incorporating nitrification-denitrification, ion exchange, and
breakpoint chlorination. Ammonia stripping has not been included, since it usually must be
backed up by breakpoint chlorination to  achieve consistently low effluent  nitrogen levels.
These systems are capable of producing average effluent nitrogen  levels of 2.0 to 3.0 mg/1,
whether the nitrogen removing processes are biological or physical-chemical in nature.

Each nitrogen removal system should be assessed from the standpoint of its reliability  in
meeting effluent objectives. Factors resulting in the failure of a system to achieve nitrogen
removal objectives are as follows:

     1.    Toxicity upsets

     2.    Overloading
                                         9-1

-------
    3.   Design deficiencies

    4.   Poor operation

    5.   Mechanical failures

    6.   Changes in influent quality

                                TABLE 9-1
          EFFLUENT NITROGEN CONCENTRATIONS IN TREATMENT
       SYSTEMS INCORPORATING NITRIFICATION - DENITRIFICATION
Type and process sequence
Lime treatment of raw sewage.
nitrification,
denitrification
Primary treatment.
high rate activated
sludge, nitrification.
denltrlfication.
filtration0
Primary treatment.
roughing filters ,
nitrification.
denitrification ,
filtration
Location
CCCSD.Ca.


Manassas,
Va.



El Lago,
Texas



Ref.
1


2




3




Scale
mgd
(mVsec)
0.5
(0.022)

0.2
(0.0088)



0.3
(0.0132)



Period,
days
90


120




55




Average effluent nitrogen, mg/1
Organic-N
1.1


0.8




0.8




4
0.3


0.0




0.9




NO~-N
0.5


0.7




0.6




NO~-N
0.0


0.0




0.0





Total
N
1.9


1.5




2.3




 CCCSD = Central Contra Costa Sanitary District
 bSysteni  1, Fig. 9-1
 CSystem 3B, Fig. 9-2
 Coarse media

                                TABLE 9-2
                EFFLUENT NITROGEN CONCENTRATIONS IN
           TREATMENT SYSTEMS INCORPORATING ION EXCHANGE


Type and process sequence
Lime treatment of
raw wastewater,
two-stage recarbonation.
filtration, activated
carbon, ion exchange
Lime treatment of
raw wastewater,
recarbonation ,
filtration, activated
carbon, ion exchange


Location
Blue Plains,
D.C.



East Bay
Municipal
Utilities
District,
Ca.


Ref.
4




5




Scale,
mgd
(mVsec)
0.05
(0.0022)



Pilot
scale




Period
days
90




na




Average effluent nitrogen, mg/1

Organic-N
a
na




2.4





NH4-N
3.6b




0.5





NO'-N
na




d





NO~-N
na




d




Total
N
4.5°'




2.96




 na = not available
 Intermittent operator attention only
°System 2A, Fig. 9-1
 Assumed negligible
Estimated
9-2

-------
                                     TABLE 9-3

       EFFLUENT NITROGEN CONCENTRATIONS IN TREATMENT SYSTEMS
                  INCORPORATING BREAKPOINT CHLORINATION
Type and process sequence
Lime treatment of
raw waste water,
two-stage recarbonation.
filtration, breakpoint
chlorination , activated
carbon"
Lime treatment of
raw wastewater.
filtration, activated
carbon, breakpoint chlorin-
ation, and dechlorlnation
by activated carbonb
Primary treatment.
oxidation ponds.
algae removal by
alum-flotation.
filtration, breakpoint
chlorlnatlon
Location
Blue Plains,
D.C.

Same,
digital
control
Owosso,
Michigan




Sunnyvale,
Ca.




Ref.
4


6


7





8





Scale
mgd
(mVsec)
0.05
(0.0022)

0.05
(0.0022)

0.02
(0.0009)




0.01
(0.0004)




Period,
days
120


9


11





2





Average effluent nitrogen, mg/1
Organlc-N
naa


na


0,58





2.6





NH+-N
na


na


1.42





0.2





NO~-N
na


na


c





0.4





NO~-N
na


na


c





0.0





Total
N
3.3


1.6


2.0d





3.2





  na = not available from publication
  System 2A, Figure 9-1
  Assumed to be negligible
  Estimated

Toxicity upsets affect only biological nitrogen removal processes. A degree of protection
against  toxicity  upset can  be provided for the nitrification-denitrification system  by
providing pretreatment processes as described in Section 4.5.3.  Several case histories are
presented in this manual which show a very high stability for the biological processes, due in
part to pretreatment.  There are classes of toxicants, such as nonbiodegradable solvents,
which are not  effectively removed by pretreatment. Reliability under this circumstance is
dependent on  source  control. Under some circumstances,  the reliability of the source
control program may be not effective enough to allow dependable nitrification.

The other factors listed affect both physical-chemical and  biological systems to varying
degrees.  Overloading  can  be  defined as  operation which exceeds  design conditions.
Obviously  both  physical-chemical and biological processes  can  be expected  to lose
effectiveness when overloaded.

Theoretically sound processes can  fail to meet  objectives due  to  design deficiencies,
mechanical breakdown or poor operation. This is true for both physical-chemical processes
and biological processes.

Some of the nitrogen  removal systems are  more sensitive to the form of nitrogen in the
influent than others. Generally,  the physical-chemical processes are geared  to a specific
                                        9-3

-------
chemical form of  nitrogen; for instance, urea  cannot be removed by air stripping or
breakpoint chlorination. Biological systems have the inherent  capability to  transform a
multitude of nitrogenous compounds to ammonia for subsequent conversion to nitrate, a
form suitable  for denitrification. Thus, if changes in the distribution of influent nitrogen
compounds  occur  with time,  the biological processes may be more able to adapt to
treatment of the new compounds than the physical-chemical processes.

In sum, the  issue  of relative reliability of the various approaches is mixed, and it cannot be
claimed that some  specific approach has a  clear advantage  over  the  others for general
application.  When stringent regulations require enhanced reliability, it is relatively simple to
provide breakpoint  chlorination  for effluent polishing. Since breakpoint chlorination has no
effect on nitrate or nitrite, it cannot make up for deficiencies in the denitrification process.
However, in the nitrification-denitrification system, it is the nitrification step that is most
susceptible to upset, and the breakpoint process provides full backup for it.

9.3 Other Considerations in Process Selection

Costs of the  alternative nitrogen removal systems  are specific to each situation  and
time-frame and generalizations  about  the  alternatives  are  difficult to  make.  Long-run
operating costs are of  interest, but the long-term  prices  of chemicals and energy are
particularly difficult to estimate.

Total dissolved solids (TDS) in  the process effluent is sometimes a consideration. Biological
nitrification-denitrification results in little change in TDS, whereas both ion exchange and
breakpoint chlorination result in an increase in TDS.

Low liquid  temperatures ( < 10 C)  often  favor physical-chemical systems because the
tankage requirements for biological nitrification-denitrification become very large. Biologi-
cal nitrification-denitrification becomes less cost-effective below 5 C.

Receiving water  standards  or  effluent  requirements may  dictate  intermittent  nitrogen
removal. This requirement may  favor breakpoint chlorination. An example is the nitrogen
removal facility for the Sacramento Regional Treatment Plant, described in Section 9.5.3.1.
In this  case, breakpoint  chlorination was chosen  because  its relatively  low capital cost
avoided the higher fixed costs of the other alternatives.

9.4 Interrelationships with Phosphorus Removal

Phosphorus  removal is the subject of another publication of the EPA Office of Technology
Transfer.9 However,  experience indicates that  the  majority of treatment plants being
designed for nitrogen removal also have a requirement for phosphorus removal. As these two
requirements have very different influences on treatment plant design, some consideration
needs to be given to how nitrogen and phosphorus removal are interrelated in treatment
system design.

                                        9-4

-------
     9.4.1 Alternative Systems

Figures 9-1 and 9-2 are summaries of the five general approaches currently being considered
or implemented for cases where  high  degrees of  phosphorus and  nitrogen removal  are
required. There are few exceptions to the listing; one  exception concerns nitrogen removal
from  oxidation pond effluent, but these approaches tend to be very case specific and  are
difficult to generalize.

All flow diagrams in Figures 9-1 and 9-2 are capable of achieving low effluent levels of nitrogen
and phosphorus. These levels are taken as averaging 2-3 mg/1 of total nitrogen and 0.1 to  0.3
mg/1 of total phosphorus. Multipoint chemical addition and filtration are shown to achieve
low phosphorus levels in the final effluent.  If lesser degrees of phosphorus removal  are
required, then some of these steps may be eliminated. Also, in each case, it is assumed that
effluent BOD5 objective is  on the  order of 10-15 mg/1.  If  further organic reduction is
required, supplemental treatment is also required. Figures 9-1  and 9-2 also  show variations
within the five flowsheets where substitute processes are possible. The flowsheets are general
process arrangements;  for example, a block showing  denitrification could  mean either an
attached growth process or a suspended growth process with a sedimentation tank.

System No. 1 is an integrated chemical-biological system using lime or other metal salt in
the  primary treatment  stage to reduce  phosphorus  and  organic loads  followed  by
nitrification and  denitrification stages  and filtration. By  moving  lime treatment  to  the
primary treatment stage, as opposed to  tertiary applications, several advantages are gained.
First, moving lime treatment to the primary stage  causes enough organic reduction in  the
primary tanks to eliminate the need  for a separate carbon removal step. Second, lime dose
can be adjusted to elevate  the pH in the nitrification step to the optimum range  for
nitrification as well as to compensate for any  alkalinity depletion due to nitrification.
Lastly, protection for the nitrifiers against most toxic heavy metals is provided.

Systems 2A and 2B are the independent physical-chemical treatment sequences  incorporat-
ing physical-chemical  nitrogen removal. A coagulant such as lime, or  a  metal salt and
polymer, is used in the  primary step  for organics and phosphorus reduction. Activated
carbon  is  provided  for further  organics reduction.  In System 2A,  either  breakpoint
chlorination or ion exchange is usual for nitrogen removal. In System 2B, ammonia stripping
is used for nitrogen removal. Filtration is placed ahead  of carbon adsorption in both
variations  of System  2; however, it may be  placed  after carbon  adsorption in certain
instances. For  considerations in arrangement  of the  adsorption component of physical-
chemical  systems,  the  reader  is referred to  the Process Design Manual for  Carbon
Adsorption, a publication of the EPA Office of Technology Transfer. 10

Another  approach to integration of biological and physical-chemical treatment  is provided
by Systems 3A and 3B. System 3A takes advantage of the favorable effect of alkalinity
depletion in nitrification on  reducing lime dose in the chemical precipitation step as lime
dose is directly affected by  alkalinity. ^ > 12 System 3B shows a slightly different way of

                                         9-5

-------
                           FIGURE 9-1
   ALTERNATE PROCESS SEQUENCING FOR SYSTEMS YIELDING COMBINED
           NITROGEN AND PHOSPHORUS REMOVAL - SYSTEMS
       WITH COAGULANT ADDITION TO PRIMARY SEDIMENTATION

                        SCREENED $  DEGRITTED
                        RAW WASTEWATER

                                     LIME and or
                                     METAL SALT
                           PRIMARY
                         SEDIMENTATION
           C02(WITH LIME
           r2 EFFLUENT)
 NITRIFICATION
           •C02 (WITH LIME
                EFFLUENT)
  FILTRATION
          METHANOL
DENITRIFICATION
        (LIME
        EFFLUENT
         ONLY)
 AIR STRIPPING
                                              METAL
                                               SALT "
 BREAKPOINT OR
 ION  EXCHANGE
                   •METAL
                    SALT
  FILTRATION
                  (OPTIONAL)
  DISINFECTION
       I
                                  •C02
  FILTRATION
                              I
                        BREAKPOINT FOR
                       RESIDUAL AMMONIA
ACTIVATED CARBON
  ADSORPTION
 FINAL  EFFLUENT
      I
                          DISINFECTION
                              T
   SYSTEM 1
   INTEGRATED
CHEMICAL-BIOLOGICAL
   TREATMENT

EXAMPLES I CONTRA COSTA
         CANBERRA
  FINAL EFFLUENT

  SYSTEM  2A
SYSTEM  2B
            INDEPENDENT
       PHYSICAL -CHEMICAL
            TREATMENT
        EXAMPLE: ROSEMONT
                              9-6

-------
                         FIGURE 9-2
ALTERNATIVE PROCESS SEQUENCING FOR SYSTEMS YIELDING COMBINED
       NITROGEN AND PHOSPHORUS REMOVAL - SYSTEMS WITH
        COAGULANT ADDITION AFTER PRIMARY TREATMENT

CARBON
1 (eg ACTIV
LIJe. Moy be I
SALT 1 «t«P
I -
CHEMICAL V MITR
PRECIPITATION NITR


DENITRIFICATION DENITF


FIL

DISIN
SCREENED 1 DEGRITTEO
RAW WASTEWATER
1
PRIMARY
SEDIMENTATION

1


OXIDATION SALT COMBINED CARBON CARBON OXIDATION
.TED SLUDOE, "^ 'J^^KlVVo. <" »™"» *""»*
3ti only)
LIME and /or LIME and/or
METAL SALT . METAL SALT
CHEMICAL CHEMICAL
FICATION PRECIPITATION PRECIPITATION
— METHANOL METHANOL
* FOR RESIDUAL,
NITRATE
IFICATION *i DENITRIF
* fSALT
• ^ 1 («. TeBUATFs}

1 	 C02 (L,ME EFFLUENT ONLY) « 	 COZ(*ITH
< METAL SALT 1 « 	 METAL SALT
i°AT*NN) A'R STRIPPING FILTRATION
. — co2
1 	 METAL SALT
TRATION DISINFECTION FILTRATION ""EXCHANGE

«^S02 FOR
OECHLORINATION IF
< BREAKPOINT EFFLUENT
M«.L t^LUtN, BREAKPO.NT FOR DIS.NFECTION
FECTION RESIDUAL AMMONIA (IF NO BREAKPOINT)
I U 	 S02 1
& 1 for Dechlorination 1
FINAL EFFLUENT FINAL EFFLUENT FINAL EFFLUENT
SYSTEM 3A SYSTEM 38 SYSTEM 4 SYSTEM 5A SYSTEM 5B
INTEGRATED
BIOLOGICAL-CHEMICAL
COMBINED SYSTEM BIOLOGICAL TREATMENT
WITH TERTIARY WITH TERTIARY PHOSPHOROUS AND
PHOSPHORUS AND PHYSICAL - CHEMICAL NITROGEN REMOVAL
RESIDUAL NITRATE
REMOVAL
EXAMPLES: EL LAOO
      BLUE PLAINS
                        EXAMPLES: NONE
EXAMPLES: SOUTH LAKE TAHOE
     ORANGE COUNTY
     MONTGOMERY COUNTY
     UPPER OCCOQUAN

-------
accomplishing the same objective. In this case metal salt is added to treatment units such as
the carbon  oxidation step and  the  filtration step, avoiding the need  for  the  separate
precipitation step of System 3A.

System 4 uses  a  combined carbon  oxidation-nitrification-denitrification  sequence. The
organic carbon in the primary effluent serves as the carbon source for the  removal of the
bulk of the nitrogen  in the influent wastewater. This  effluent is polished  with a tertiary
phosphorus  removal step. Residual nitrates are removed in the wastewater filter. System 4
has lower operating costs than System 1 or 3 because of  the elimination of the bulk of the
methanol costs incurred in  the latter systems. On the other hand, the phosphorus removal
step must be placed after the biological treatment step  in System 4  for two reasons. First,
precipitation of phosphorus in the primary step would reduce the organics  in the primary
effluent  to  the  point where denitrification could  not be supported. Second, metal salt
addition to  the  biological step  would add so many solids  as to render unmanageable the
simultaneous operation of carbon oxidation, nitrification and denitrification.

System 5 was the first system implemented in the U.S. and was used at South Lake Tahoe
(in  the System 5A configuration). It consists of conventional biological treatment followed
by  tertiary steps for  phosphorus  removal and physical-chemical nitrogen removal. System
5A  employs air stripping with polishing by breakpoint chlorination while System 5B uses
either breakpoint chlorination or ion exchange for nitrogen removal.

In conjunction with  the system  descriptions  in Figures 9-1  and 9-2 are  listed  the case
examples presented  in this chapter which generally fit the system description. In most cases
the  examples do not precisely follow the system  description because local requirements have
dictated lesser or greater degrees of treatment. However, the case examples are close enough
to be fitted into system categories.

    9.4.2 Considerations in System Selection

Each of the systems outlined in Section 9.4.1 has its inherent advantages and disadvantages
that need to be considered by  the  treatment  plant designer in each  individual situation.
Some of these considerations are described in this section.

          9.4.2.1 Phosphorus Removals Obtainable

Perhaps because of  the long experience accumulated with System 5A (Figure 9-2)  at South
Lake Tahoe and the more recent development of alternative systems,  System 5A or SB  were
thought to have had a decided advantage in terms of low phosphorus residual over any of
the integrated systems (System 1, 3A and 3B). It  has been suggested that the integrated
system, even with proper coagulant dosage, is limited to reduction of effluent phosphorus
levels  to 0.5 mg/1 total phosphorus and  that when lower phosphorus levels are required,
System 5A or 5B should be  employed, l^
                                        9-8

-------
In actuality, the degree of phosphorus removal is not primarily affected by the system
selected but by the pattern of chemical addition, the nature and doses of the chemicals used
and  the sophistication of process control. Regardless of the system selected, low effluent
phosphorus levels are made possible by multiple phosphorus removal steps; this may be
achieved by multipoint chemical addition or chemical addition in conjunction with other
phosphorus removal methods such as tertiary filtration.

An example is provided  by the well  documented operation of the South Lake Tahoe
plant. 13 A flowsheet for the  plant  as it  existed in  1970 is  presented in Figure  9-3.
Multipoint chemical addition was practiced with lime treatment  of the secondary effluent
plus alum treatment at the tertiary filtration step. Phosphorus removal resulted at both these
steps.  In addition, phosphorus uptake occurred in the activated sludge step and possibly
some removal occurred in the carbon adsorption step. The lime treatment alone reduces the
phosphorus level to about 0.6 mg/1 entering the filtration step. About 30 percent phosphorus
removal occurred in the filter  without alum addition.  With alum addition, the effluent
phosphorus level  was reduced to 0.1 mg/1 in a special one month test. 13  The phosphorus
residuals fora one year period averaged 0.17 mg/1 in the plant effluent as shown in Table 9-4.
Alum dosage required to boost phosphorus removals by the filters from 30 to 90 percent was
only 10 mg/1.13

From the Tahoe example it can be seen that the key to obtaining a low phosphorus residual
is multiple removal steps. Multiple phosphorus removal steps have been included in all of the
treatment systems portrayed in Figures 9-1 and 9-2. For instance, the  option of metal salt
addition to tertiary filters is available to all of the systems,  not just System 5A or 5B, and
comparable performance can be expected in each application. System performance is given
in  Table 9-4 for all systems except System 4. Favorable examples  were  chosen for each case
in  Table  9-4  and other  cases  for each  system could be found  with higher  effluent
phosphorus  values. These  examples are illustrative of what good design, operation,  and
control can produce.

         9.4.2.2  Impacts on Sludge Handling

In Systems 3A, 4, 5A and 5B, (Figure 9-2).-chemical precipitates can be kept  separate from
organic sludges. In Systems 1,  2A, 2B and 3B, (Figures 9-1 and  9-2) chemical sludges are
combined with  organic sludges.  Separation of sludges allows  the plant  designers more
options in sludge handling. For instance, chemical  sludges  can  be subjected to coagulant
recovery operations while  processing of organic sludges can  proceed  without  the hindering
effects of the inert chemical sludges.

It  has also been suggested that tertiary phosphorus removal (System 5A or 5B) may have an
advantage in that lime recovery can be practiced.^ However, System 1, 2, 3A and  4 all
possess an advantage in common with 5A or 5B; namely, the ability to recover lime if it is
the chosen coagulant.
                                        9-9

-------
                                                                    FIGURE 9-3
                       SCHEMATIC FLOW DIAGRAM - SOUTH LAKE TAHOE, CALIFORNIA PLANT (1970)
                                    SECONDARY
                                   TREATMENT
                  PRIMARY
                 TREATMENT
                                     CHEMICAL TREATMENT
                                            AND
                                     PHOSPHATE REMOVAL
MAJOR TYPES
OF TREATMENT
  PROVIDED
                                                                         NITROGEN
                                                                         REMOVAL
                                   (BIOLOGICAL
                                   TREATMENT)
                  (SOLIDS
                 SEPARATION)
                                                                                                                         60 MG EMERGENCY
                                                                                                                         HOLDING POND
                                                                                                                                              RECLAIMED
                                                                                                                                              WATER ro
                                                                                                                                              INDIAN
                                                                                                                                              CREEK
                                                                                                                                              RESERVOIR
                    PARSHALL FLUMES-
                    FLOW MEASUREMENT
                    AND DIVISION
              BMMINUTORS
                                                                                                 CHLORINE
                                                                                                 APPLICATION
                                                                                                                         RETURN TO
                                                                                                                         SECONDARY BALLAST
                                                                                                                         POND AT PLANT
                                                           RECARBONATION
                                                           (STAND an
                                                             AMMONIA
                                                             STRIPPING
                                                              TOWER
                                                                                                                                   LUTHER PISS .
                                                                                                                                   BOC/STER PUMP
                                                                                                                                   STATION
WASTE WATER
    FLOW
  THROUGH
    PLANT
                            SECONDARY
                            CLARIFIERS
                                                                            RE-
                                                                            CARBONATION
                                                                            BASIN
                                                                                       TERTIARY
                                                                                       PUMP
                                                                                       STATION
PLANT INFLUENT
FORCE MAINS
                PRIMARY
                SLUDGE PUMPS
                                       SCCONDARYX
                                       SLUDGE PUMPS
                                      SLUDGE FLOW
                                      DIVISION BOX
                                   WASTE ACTIVATED
                                   SLUOCC
   SOLIDS
  HANDLING
  LIME AND
   CARBON
RECLAMATION
                       BIOLOGICAL
                        SLUDGE
                       FURKACt
                                                     RCCAUWCO LIME
                                                     TORE-USE
                JTEHILE ISM
                TO DISPOSAL
                                                                                                     REGENERATED
                                                                                                     CARBON TO
                                                                                                     RE-USE

-------
                                            TABLE 9-4
                         EFFLUENT PHOSPHOROUS CONCENTRATION FROM
                                     ALTERNATIVE SYSTEMS
Type and process sequence
System 1
Lime (with recalcination and
recycle), nitrification.
denitrification, without
filtration
System 2A
Lime with iron in second . .
settling tank, filtration.
ion exchange or break-
point, filtration
System 3B
Carbon oxldation-
nitrif ication-denitrif ication ,
filtration with alum addition
to carbon oxidation and
denitrification
System 5 A
Carbon oxidation, lime.
ammonia stripping',
recarbonation , filtration
with alum addition, carbon
ad sorption
Location

CCCSD, Ca.a




Blue Plains,
b.c. •



Manassas,
Va.




South Lake
Tahoe, Ca.



Ref.

14




. ,4




2





13




Scale,
mgd
(mVsec)

0.5 to
(0.022)



0.05 .
(0.0022)



0.02
(0.0009)




2.4
(0.11)



Period,
days
W
10b




480




120





365C




Average effluent phosphorus,
mg/1
Total P

0.04




0.14




0.3
/
/
' /


0.17




PO"-P

0.01




na

/x

X
./ na





na




Average effluent
total suspended
solids, mg/1

3.0




4.0




0





0




aCCCSD = Central Contra Costa Sanitary District

bAugust 1 to 10, 1973

C1970; representative year, the plant has been operational since 1968.

-------
          9.4.2.3 Reliability

Factors affecting the reliability of nitrogen removal processes have already been described in
Section 9.2. Most of these same factors affect phosphorus removal and will not be repeated
here.

The longest record of reliable experience in obtaining low phosphorus residuals is at the
South  Lake Tahoe Plant, operational since 1968 (Table 9-4). Low values have also been
obtained consistently for long periods with  physical-chemical  systems (System 2A). In the
latter  case, iron  has been  used  in  a  second  stage settler after lime  treatment  and
recarbonation,  achieving further phosphorus  removals.  Less  experience is available with
Systems 1, 3A  or 3B, but the limited testing to date indicates very low phosphorus residuals
can be consistently obtained.

          9.4.2.4 Flexibility of Operation in Multipurpose Treatment Units

The systems portrayed in Figures 9-1 and 9-2  incorporate varying degrees of integration of
process function into the various treatment units. Systems 4, 5A and 5B represent extremes
in terms of combining functions; in System  5A or 5B the tendency is for individual steps to
perform a minimum of  purposes while in System 4 many functions  are accomplished in
parallel in each step.

The argument can be made that the level of integration in a plant can affect its flexibility of
operation in terms of adjustability of the system  to meet varying loads  or in  terms of
providing redundancy for possible process failures. The degree  of integration possible is best
studied by examination of pilot or full-test results. There have  been full-scale tests that have
shown that nothing has  been lost  in terms of flexibility or performance with a degree of
process  integration.  Examples  are provided by the Manassas and  CCCSD  experience
(Systems 1  and 3B) described in Section 5.2.4. No long-term test results are available for
System 4, which  is unfortunate, since a high degree of integration of function is provided in
the combined carbon oxidation-nitrification-denitrification step.

          9.4.2.5 Cost

Cost is an essential factor in  process selection.  It is widely recognized that the integrated
approaches hold  a potential of  cost savings over the biological-tertiary approach (System
5) 1,12,15 jhe reality of this cost saving will be  determined in individual situations by local
factors  and must be specifically evaluated  in each  case by cost-effective analyses  of the
alternative systems.

9.5 Case Examples

Fourteen case examples  of nitrogen control are presented, each showing how  the various
nitrogen removal systems described in this chapter have been applied. These include:  four

                                        9-12

-------
examples   of nitrification  for  ammonia  reduction,  four  examples  of  nitrification-
denitrification for nitrogen removal, and two examples each of breakpoint chlorination, ion
exchange and air stripping for nitrogen removal.

     9.5.1 Case Examples of Nitrification for Ammonia Reduction

Four examples of how biological nitrification has been implemented are presented in the
following  discussion.  The Jackson,  Michigan  plant  design was oriented to reducing the
nitrogenous oxygen demand (NOD) of the plant effluent in the receiving waters. The designs
of the Valley Community Services District plant and the City of Livermore's plant were
oriented to satisfying the very low coliform requirements set by the State of California. In
these cases, ammonia reduction allowed efficient disinfection after breakpoint chlorination.
In the  design of  another California plant, operated by the San Pablo  Sanitary District,
nitrification was included so that effluent toxicity requirements could be met.

Other case  examples of nitrification were  presented previously in Section 4.3.8. These
included the Whittier  Narrows  Reclamation Plant  in California  which  is  oriented to
groundwater recharge and the Flint, Michigan plant, designed for NOD removal.

         9.5.1.1  Jackson, Michigan

The  City of Jackson is operating a 17 mgd (0.74 m-^/sec) activated sludge plant that is
designed to nitrify year-round. Nitrification is provided for removal of nitrogenous oxygen
demand so that the receiving water, the Grand  River, can be maintained at dissolved oxygen
levels of about 4.0 mg/1. Since the  implementation  of full nitrification  at this plant, this
requirement has been consistently met in the zone of influence of the plant's discharge.

The  City of Jackson, Michigan, with an equivalent population of 60,000 has been served by
a conventional activated sludge treatment plant since 1936.16 The current upgrading of the
plant was completed in 1973 and resulted in the plant depicted in Figure 9-4 with the design
data shown in Table 9-5. During this upgrading, the following facilities were added: (1) new
primary effluent pumps, (2) a stormwater retention basin, (3) stormwater pumps for filling
the  retention basin, (4) additions  to the aeration  tanks, (5)  additions  to the secondary
sedimentation tanks, (6) new return activated sludge pumps, (7) three new blowers and a
blower  building for the  secondary treatment additions, and (8) a new plant electrical and
control system.17 This work was bid in December, 1970 and totaled $3,200,000 including
legal, engineering and contingency  costs.  Operation and maintenance costs for the entire
plant for  fiscal year 1973/1974 totaled  $464,159  for 5255  million gallons treated or
$88/milgal.18

Several features have been  incorporated into this  plant  that have been  stressed in this
manual. First, the activated sludge system is operated in the conventional  or plug  flow
manner to  gain highest efficiency of nitrification even at the coldest temperature conditions
(as low as  8 C).  Coarse bubble  aeration  is utilized.  Another feature of the plant is the

                                        9-13

-------
                             TABLE 9-5

                          DESIGN DATA
   JACKSON, MICHIGAN WASTEWATER TREATMENT PLANT
Average dry weather flow (ADWF)
Peak dry weather flow (PDWF)
Peak wet weather flow (PWWF)

Raw wastewater quality at ADWF

   BOD5
   SS
     17 mgd (0.74 m3/sec)
     22 mgd (0.96 mVsec)
     30 mgd (1.31 m3/sec)
    145 mg/1
    200 mg/1
Primary sedimentation tanks

   Number
   Length
   Width
   Depth
   Overflow rate at ADWF
   Detention time at ADWF

Retention basin

   Volume

Air blowers

   Number
   Discharge pressure
   Capacity - total

Aeration tanks

   Old tanks

     Number tanks
     Passes per tank
     Width
     Length/pass
     Depth
     Volume (4 tanks)

   New tanks

     Number of tanks
     Passes per tank
     Width/pass
     Length/pass '
     Depth
     Volume (2 tanks)

   Detention time at ADWFa

Final sedimentation tanks

   Old tanks

     Number
     Diameter
     Sldewater depth
     60 ft (18.3 m)
     24 ft (7.3 m)
     11 ft (3.4 m)
   ,970 gpd/sf (80.2 m3/m2/day)
      1 hr
     12 mil gal (45,430 m3)
    6.5 psig (0.46 kgf/cm2)
 33,000 cfm (940 m3/min)
      4
      1
   25.5 ft (7.8 m)
    240 ft (73.2 m)
   14.5 ft (4.4 m)
355,000 cu ft (10,000 m3)
      2
      2
     25 ft (7.6 m)
    150 ft (45.7 m)
   14.5 ft (4.4 m)
217,500 cu ft (6,159 m3)
               t
    6.0 hr
     70 ft (21.3 m)
     11 ft (3.4 m)
                                 9-14

-------
                             TABLE 9-5

                            DESIGN DATA
         JACKSON, MICHIGAN WASTEWATER TREATMENT PLANT
                            (CONTINUED)
   New tanks

     Number
     Diameter
     Sidewater. depth

   Overflow rate at ADWF a
              at PDWF a
   Detention time at ADWFa

Chlorine contact chamber

   Number of passes
   Detention time at ADWF

Sludge digestion

   Number digesters
   Volume

Sludge drying beds

   Surface area, sf
     80 ft (24.4 m)
     12 ft (3.7 m)

    670 gpd/sf (27.3 m3/m2/day)
    865 gpd/sf (35.2 m3/m2/day)
    3.0 hr
      4
     47 min
297,410 cu ft (8,422 m3)
328,000 sf (30,489 m2)
Including both old and new tanks
                             FIGURE 9A

 JACKSON, MICHIGAN WASTEWATER TREATMENT PLANT FLOW DIAGRAM
                      WASTE  ACTIVATED SLUDGE
REENEO
tw
kSTEWATER


GRIT
REMOVAL

DRYING BEOS
1
'
SUPER-
NATANT
RETURN


r PRIMARY

RAW ond
WASTE
SLUDGE
SLUDGE \
DIGESTION]
^^
\

1
RETURN SLUDGE

/^ ^\
AERATION- /SECONDARY \ CHLORINE
TANKS V TANKS / EFFLUENT
\ / ' " TO
\ 	 ' RECEIVING
WATER
LINED
RETENTION
BASIN
                                 9-15

-------
primary effluent retention basin. Rather than designing the secondary facilities for the full
wet weather flows, flows above about 22 mgd are pumped to the retention basin. This flow
is brought  back by gravity to the secondary facilities when storm flows subside. It is also
anticipated that the  retention basin will  serve as flow equalization  storage  during dry
weather, once  plant flows reach design capacity. This will be done to  prevent ammonia
bleedthrough during peak flow periods. 1 ^

The retention basin is lined, but has no provision for mixing. Actual operation indicates very
satisfactory performance without odor development during storage. Except for completely
draining the basin, cleaning is limited to once per year. ^

Performance of the plant has been exceptionally stable and was previously summarized in
Section 4.3.8.3. It should be noted that  the  plant is not yet being operated  at its design
flow.

          9.5.1.2 Valley Community Services District, California

The Valley Community Services District  (VCSD)  Wastewater Treatment Plant at Dublin,
California, is  treating an average daily  flow of 3.7 mgd  (0.16  m^/sec) from  a largely
residential service'  area.  The original plant,  consisting of raw wastewater screening, grit
removal,  primary  sedimentation,  activated  sludge  aeration, secondary sedimentation,
digestion and sludge lagooning, was constructed in 1960 for approximately $1.5 million. In
1972, an additional primary sedimentation tank and appurtenances, an additional secondary
clarifier, a digester, dual media filters, and chlorination  and dechlorination facilities were
constructed for about  $3 million. These  facilities raised  the average dry weather flow
capacity of the plant to 4 mgd (0.17 m^/sec). The design data for the plant are shown in
Table 9-6.19

Waste discharge requirements mandate effluent filtration, as the State of California requires
that any effluent that may be used for water contact sports must be coagulated and filtered.
Requirements also dictated that the  median coliform content must not exceed an MPN of
2.2 per 100 ml. This  plant incorporates nitrification for ammonia reduction so that residual
ammonia can be economically breakpointed. This allows disinfection to proceed with a free
chlorine residual so that the stringent bacteriological requirements may be met.

A flow diagram of the existing facilities (Fig. 9-5)  shows a holding basin  which is routinely
used for flow equalization after primary  treatment. Figure  9-6 is  a photo of the holding
basin. The  holding basin is asphalt-lined and is equipped with a sprinkler system for washing
out accumulated solids when the basin is drained. The sprinkler system was added by plant
staff and was found  to be very effective for odor control. The basin is emptied daily. A
peak-to-average flow  ratio of 3.4:1 is equalized in the  holding  basin to  maintain  stable
biological treatment  conditions.  It was found prior to the 1972 additions to the plant that
operation  without  flow  equalization resulted in  ammonia bleedthrough and eventual
complete loss of nitrifying capability. The aeration  tank is generally operated with the first

                                         9-16

-------
                                     TABLE 9-6
           VCSD WASTEWATER TREATMENT PLANT DESIGN DATA
Population
Average dry weather flow (ADWF)
Peak dry weather flow (PDWF)
Peak wet weather flow (PWWF)
Raw wastewater quality at ADWF
   BOD5
   Suspended solids
Primary treatment
   Preaeratlon and grit removal tanks
     Number
     Detention time at ADWF
   Primary sedimentation tanks
     New tanks
          Number
          Length
          Width
     Old tanks
          Number
          Length
          Width
     Average water depth
     Detention time at ADWFa
     Overflow rate at ADWF a
   Assumed removals
     BOD5
     Suspended solids
Activated sludge
   Aeration tanks
     Number
     Passes per tank
     Width each pass
     Length each pass
     Average water depth
     Detention time, based on ADWF
     BODs loading
   Aeration blowers
     Number
     Capacity per blower
     Discharge pressure
Secondary sedimentation tank
   Old  tanks
     Number
     Diameter
     Sidewater depth
   New tanks
     Number
     Diameter
     Sidewater depth
   Detention time, based on ADWF8
   Overflow rate at ADWF a
Dual media filters
   Number
   Area each filter
   Filtration rate at PWWF
   Anthracite media
     Depth
     Effective size
           48,000
    4 mgd (0.17 m3/sec)
    8 mgd (0.33 mVsec)
    12 mgd (0.50 m3/sec)

          330 mg/1
          330 mg/1
             2
          24 min
       100 ft (30.5 m)
        19 ft (5.8 m)
       110 ft (33.5 m)
        19 ft (5.8 m)
        9 ft (2 .74 m)
           1.7 hr
960 gpd/sf (38.9 m3/m2/day)

         40 percent
         70 percent
             1
             2
        30 ft (9.2 m)
       210 ft (64.0 m)
        15 ft (4.6 m)
           8.5 hrs
      35 lb/1000 cf/day
    2,900 cfm(82 m3/min)
    7.5 psig (0.53 kgf/cm2)
       65 ft (19.8 m)
        9 ft (2.7 m)

             1
       90 ft (27.4 m)
        14 ft (4.3 m)
          5.3  hrs
410 gpd/sf (16.6 m3/m2/day)
     728 sf (67.66 m2)
3.8 gpm/sf (1.36 l/m2/sec)

     36 inches (0.91 m)
        2.4 - 4.8 mm
                                         9-17

-------
                                     TABLE 9-6
             VCSD WASTEWATER TREATMENT PLANT DESIGN DATA
                                  (CONTINUED)
      Sand media
        Depth
        Effective size
      Pea gravel size
        Depth
      Backwash water rate (max.)
   Chlorine contact tanks
      Number
      Volume
      Detention time @ ADWF
   Sludge digestion
      Digester loading
        Primary solids
        Biological solids
        Total solids
      Digesters
        Number
        Diameter
        Sldewater depth
        Total volume
   Sludge disposal
      Sludge lagoons
        Number
        Volume
     18 inches (0.45 m)
       0.8 - 1.0 mm

     8 Inches (203 mm)
 20 gpm/sf (13.6 ]/m2/sec)
    22,500 cu ft (637 m )
         1.0 hr
 7,700 Ib/day (3,500 kg/day)
 3,800 Ib/day (1,730 kg/day)
11,500 Ib/day (5,230 kg/day)
      55 ft (16.8 m)
      33 ft (10.1 m)
  157,POO cu ft (4,500 rn )
  350,000 cu ft (9,.920 m )
  Including both old and new tanks

half of the  first  pass being used for reaeration of the return sludge, and with the primary
effluent being step fed in increments to the remainder of the first pass. Fine bubble aeration
is employed.
The VCSD.  plant  has consistently  nitrified  the influent ammonia as shown  by the
performance data in Table 9-7.20 j^g nitrogen figures are from once monthly 24-hour
composite  samples, while the  BOD and  suspended solids data is the  average  of  daily
composite samples. For the first ten months of 1974, the VCSD waste water treatment  plant
averaged 98.6 percent  BOD removal and 99.3 suspended solids removal. The ammonia
nitrogen concentration  in  the effluent  was typically less than  1 mg/1 and has been at this
level since August;  1 973, apart from a notable process upset caused by an industrial spill in
March,  1974. The nitrate-nitrogen concentration in the effluent has generally been around
24  mg/1 and this is about 99 percent of the nitrogen in the effluent. For several months
before August, 1973, the aeration capability was limited to two on-line blowers which were
not enough  to sustain complete nitrification. The data given in Table 9-8 shows the change
in process performance after mechanical difficulties were overcome and a third blower was
started  up. 21 Before the aeration capacity was increased,  the average ammonia nitrogen
concentration in the effluent composite samples was 3.9 mg/1 with the nitrate nitrogen level
averaging 13.9 mg/1. A dissolved oxygen level of 2  to 4 mg/1 is now maintained in the last
portion of the aeration tank, whereas before August, 1973, the level was often- less than 2
mg/1.
                                        9-18

-------
                          FIGURE 9-5

           VALLEY COMMUNITY SERVICES DISTRICT (CALIF.)
           WASTEWATER TREATMENT PLANT FLOW DIAGRAM
RAW
WASTEWATEF
                         TO
                      ALAMO CANAL
                              9-19

-------
                                             TABLE 9-7
                NITRIFICATION PERFORMANCE AT VALLEY COMMUNITY SERVICES DISTRICT
                            WASTEWATER TREATMENT PLANT, CALIFORNIA
Month
1974
January

February

March

April

May

Tune

July

August

September

October

Raw
waste-
water
flow,
mgd
(m /sec)
4.09
(0.179)
3.64
(0.159)
3.99
(0.175)
4.12
(0.181)
3.64
(0.159)
3.76
(0.165)
3.49
(0.153)
3.41
(0.149)
3.56
(0.156)
3.55
(0.156)
Mixed
liquor
recycle
ratio
0.49

0.41

0.46

0.42

0.42

na

0.55

0.43

0.45

0.47

Temp.
C
nab

na

16

19

22

25

28

29

27

26

MLSS,
mg/1
na

5,512

6,130

1,533

4,759

7,696

4,811

5,132

4,771

4,872

Sludge
volume
Index
119

119

126

73

81

82

73

77

85

103

Solids
retention
time,
days
10.5

11.9

15.2

8.8

9.6

10.9

11

10.9

8.1

8.3

BOD,., mg/1
Primary
effluent
na

231

167

140

147

142

106

116

164

183

Secondary
effluent
na

10.0

21.0

14.7

16.0

8.8

6.5

8.0

7.7

10.0

Final
effluent
2.5

5.6

3.7

3.8

5.6

3.0

3.0

2.5

2.0

1.6

Percent
removal
na

98.6

98.5

98.8

96.0

98.9

98.4

98.4

99.4

99.4

Suspended solids, mg/1
Primary
effluent
91

87

99

86

88

84

90

84

93

91.7

Secondary
effluent
19.2

9.6

8.2

9.2

9.0

10.3

9.4

12.2

15.7

10.6

Final
effluent
2.5

1.3

1.7

1.9

2.8

1.6

2.5

l.S

1.4

1.3

Percent
removal
99.0

99.5

99.3

99.1

98.7

99.4

99.0

99.4

99.6

99.7

Nitrogen,8 mg/1
Effluent
NH4-N
0.23

2.4

16

0.17

0.06

0.84

0.06

0.22

0.78

0.11

Effluent
NO~-N
17.0

21.8

6.8

24.4

26.6

24.9

23.1

21.9

28.9

21.9

Results of one 24-hour composite sample per month.

na = not available.

-------
                        FIGURE 9-6

      HOLDING BASIN AT THE VALLEY COMMUNITY SERVICES
     DISTRICT (CALIFORNIA) WASTEWATER TREATMENT PLANT
                          TABLE 9-8

 NITROGEN ANALYSES ON 24 HOUR COMPOSITE EFFLUENT SAMPLES AT
   THE VALLEY COMMUNITY SERVICES DISTRICT TREATMENT PLANT
Date
sampled,
1973
June
July3
August
September*3
October13
November*3
December'3

Nitrate nitrogen
as N, mg/1
9.9
16.5
9.5
14.0
26.7
26.0
27.1

Ammonia nitrogen
as N, mg/1
6.3
2.0
4.7
0.39
<0.06
0.30
<0.06
aTwo blowers on-line.
 Three blowers on-line.
                             9-21

-------
The VCSD staff feels that this type of nitrification system is particularly subject to upset
due to  toxicants in  the primary effluent. One recurring  loss of nitrifying  ability  was
ultimately  traced to  the  periodic discharge of a  solvent,  trichlorethylene.22 Once the
industrial discharger was located and the spills ceased, the problem disappeared.

Operation and maintenance costs have averaged $350 per million gallons since the  1972
additions.

         9.5.1.3 Livermore, California

Since 1967, the City of Livermore has operated the 5 mgd (0.22 m^/sec) Water Reclamation
Plant  whose flow diagram is shown in Figure 9-7.  Before 1967, the original plant included
primary  treatment, trickling filters,  secondary sedimentation,  polishing treatment  with
oxidation ponds and sludge digestion. During the 1967 enlargement, existing structures were
rearranged  in the flowsheet and additional facilities were added.23 After the enlargement,
the plant consisted of preliminary treatment and primary sedimentation, roughing filters,
activated sludge  for  nitrification, and chlorination.  The existing oxidation ponds  were
converted to emergency holding basins. Sludge  disposal is by digestion with digested sludge
being applied to sludge lagoons. Drying beds are used intermittently.

The plant layout was oriented to meeting discharge requirements mandating an effluent that
contained no more than 20 mg/1 of BOD5, 20 mg/1 of SS, and  a  five-day-median  total
coliform of 5 MPN per 100 ml. The low bacteriological requirements dictated that the plant

                                    FIGURE 9-7

  CITY OF LIVERMORE WATER RECLAMATION PLANT (CALIF.) FLOW DIAGRAM
                                        9-22

-------
be designed to  dependably nitrify on a year-round basis. A free  chlorine residual was
thought to be required for disinfecting to such low levels and that the ammonia level must
be minimized in the secondary effluent so that disinfection through breakpoint chlorination
could be economically practiced.

The  trickling  filters  were retained  in the plant  layout to act as  pretreatment for  the
nitrification step, to reduce organic loads and to moderate any shock loads that might upset
nitrification. The sloughed solids from the roughing filter pass directly to the aeration tanks
without any intermediate clarification.

Plant design data are shown in Table 9-9 and operating data for the year 1971 in Table
9-10.23,24 The  plant has consistently met effluent requirements and demonstrated stable
year-round operation. Wastewater temperatures are favorable for nitrification, with a range
of 15 to 24 C. Even when ammonia breakthrough has occurred in the secondary effluent, it
has been effectively reduced by breakpoint chlorination.  Recently,  the plant has not had
sufficient aeration capability to completely nitrify during peak nitrogen loan conditions, but
this  condition is being rectified in plant modifications currently underway. Coarse bubble
aeration is used  in the plant. A photograph of the aeration tank is shown in Figure 9-8. The
activated sludge exhibits  very good  settling  properties with sludge  volume index (SVI)
measurements consistently below 100 ml/g.

On the few occasions  when  plant  upsets  occur,  making it  likely that  bacteriological
requirements will not be met, the effluent is directed to the emergency holding basins. The
chlorine contact tank has also served as a supplementary settling tank and yielded further
reductions  of BOD5 and  SS. The chlorine  contact tank must be occasionally drained to
remove accumulated solids.

Twenty to twenty-five percent of the effluent is used for irrigation purposes at a nearby golf
course and on agricultural land. Effluent is also used at the golf course  to fill several small
lakes.

The  initial plant had a construction  cost of $900,000 (1957 dollars), which included land
purchase,  while  the cost of the 1967 plant expansion was $1,300,000 (1968  dollars).
Operational expenditures for 1971 were approximately $275,000 which, when expressed on
a unit basis, is $224 per million gallons treated.

         9.5.1.4 San Pablo Sanitary District, California

The  San Pablo Sanitary District, California, operates a 12.5 mgd (0.55 m^/sec) wastewater
treatment plant  designed for year-round complete nitrification. The original plant consisted
of a  primary treatment plant with effluent chlorination and digestion for solids processing.
Additions completed in  1972 resulted in the plant flow diagram shown in Figure 9-9 and
included additional treatment  facilities,  a new roughing trickling  filter, new  aeration-
nitrification tanks, new  secondary  clarifiers, an additional chlorine  contact tank, new

                                         9-23

-------
                                        TABLE 9-9
          DESIGN DATA - LIVERMORE WATER RECLAMATION PLANT
Population
Average dry weather flow (ADWF)
Peak dry weather flow (PDWF)
Peak wet weather flow (PWWF)
Raw wastewater loadings
   BOD5
   Suspended solids
Primary treatment
   Preaeratlon and grit removal tanks
      Number
      Detention time at ADWF
   Primary sedimentation tanks
      Number
      Length
      Width
      Average water depth
      Detention time at ADWF
      Overflow rate at ADWF
Emergency holding basin
   Volume
Roughing filters
   Number
   Diameter
   Depth of media
   Total volume of media
   Media type
   Media size
   Reclrculation ratio at ADWF
   BODs load
Air blowers
   Number
   Discharge pressure
   Capacity  - total
Aeration-nitrification tank
   Number
   Passes  per tank
   Length/pass
   Width/pass
   Depth
   Detention time at ADWF
   BODs load
Secondary sedimentation tank
   Number
   Diameter
   Sidewater depth
   Overflow rate at ADWF
Chlorine contact tank (Breakpoint chlorination)
   Number
   Passes  per tank
   Detention time at ADWF
Anaerobic digestion
   Number
   Volume
   Loading, total solids
   Volatile matter, percent
Sludge disposal
   Digested  sludge lagoons
      Number
      Total volume
   Sludge drying beds
      Number
      Total area
                 62,500
           5 mgd  (0.22 mVsec)
          10 mgd  (0.44 m3/sec)
          18 mgd  (0.79 m3/sec)

      12,500 Ib/day (5,670 kg/day)
      12,500 Ib/day (5,670 kg/day)
                   2
                36 mln
             124 ft (37.8 m)
              19 ft (5.8 m)
               9 ft (2 .7 m)
                 1.5 hr
      1,050 gpd/sf (42.8 m3/m2/day)

         31 mil gal (117,000 m3)
             110 ft (33.5 m)
             4.25 ft (1.3 m)
          80,152 cf (2,270 m3)
                  Rock
                2 to 4 in.
               3.0 to 1.0
100 Ib BODs/1,000 cf/day (1.62 kg/m3/day)

                   3
          7.5 psi (0.53 kgf/cm2)
          6,000 cfm (170 m3/min)

                   1
                   2
             160 ft (48.8 m)
              30 ft (9.2 m)
              15 ft (4.6 m)
                 5.2 hr
28 Ib BOD5/1,000 cf/day  (0.45 kg/m3/day)

                   1
              90 ft (27.4 m)
              12 ft (3.7 m)
       767 gpd/sf (31.2 m3/m2/day)

                   1
                   2
                  1 hr
            27,500 cf (779 m3)
        11,800 Ib/day (5,350 kg/day)
                  75
           320,000 cf (9,060 m3)
           22,400 sf (2,080 m2)
                                              9-24

-------
dissolved air flotation thickeners, and two new digesters. This flowsheet is very similar to
that used at the Livermore plant, described in Section 9.5.1.3, except that plastic media is
used in the  roughing filter in place of rock media. Design data for the plant are shown in
Table 9-11.25'26
                                FIGURE 9-8

     AERATION TANK AT THE LIVERMORE WATER RECLAMATION PLANT
    (CALIFORNIA) WITH ROUGHING TRICKLING FILTERS IN BACKGROUND
                                    9-25

-------
                                              TABLE 9-10
                                    NITRIFICATION PERFORMANCE AT
                               THE LIVERMORE WATER RECLAMATION PLANT
Month
1971
January

February

March

April

May

Jane

July

August

September

October

November

December

Flow,
mgd
(mVsec)
3.3
(0.14)
3.3
(0. 14)
3.2
(0.14)
3.2
(0.15)
3.4
(0.15)
3.5
(0.15)
3.3
(0.14)
3.4
(0.15)
3.6
(0.18)
3.5
(0.15)
3.4
(0.15)
3.3
(0.14)
Recycle*
ratio
0.35

0.34

0.35

0.34

0.35

0.38

0.44

0.43

0.44

0.42

0.42

0.44

MLSS,
mg/1
1803

1756

1702

1748

1743

2112

2189

2160

2123

2157

2275

2316

SVI.
ml/g
99

66

79

82

91

79

78

79

74

64

58

52

9c.
days
4.3

4.7

4.2

4.3

5.0

5.4

7.7

7.2

8.0

9.6

5.8

5.0

HT,
hr
7.8

7.8

8.1

8.1

7.6

7.4

7.8

7.6

7.2

7.4

7.6

7.8

Air use
MCF/day
(1/eec)
6.4
(2100)
6.4
(2100)
5.9
(1900)
8.0
(2100)
6.3
(2100)
6.8
(2200)
6.9
(2300)
7.2
(2400)
7.3
(2400)
7.3
(2400)
7.1
(2300)
6.5
(2100)
Primary effluent
BOD«.
mg/r
140

128

156

125

125

110

118

108

114

106

146

147

ss.
mg/1
80

85

84

74

98

81

71

81

60

75

63

94

Roughing filter effluent
BOD*,
mg/T
129

79

121

107

112

74

81

46

82

66

115

104

SS.
mg/1
137

122

128

96

118

110

102

118

115

74

89

107

NHd-N
mg/r
33.5

41.0

41.9

33.3

31.0

27.1

23.3

25.2

22.5

32.6

35.6

44.4

Secondary effluent
BOD.,
mg/1
8.9

11.8

16.9

8.7

11.6

6.6

8.1

5.2

5.4

10.6

11.3

12.6

SS,
mg/1
20

17

24

15

19

24

19

31

19

18

31

27

NHi-N,
mg/1
0.86

1.07

6.73

0.94

0.76

0.48

0.88

0.88

1.75

1.03

0.46

4.32

NO£-N,
mg/1
0.07

0.32

0.78

0.02

0.05

0.02

0.04

0.12

0.17

0.03

0.10

0.13

NC£-N,
mg/1
17.8

17.6

19.3

18.8

20.2

16.5

18.3

19.4

18.2

18.2

18.9

16.6

Final effluent
BODg.
mg/1
9.3

6.3

5.2

3.1

8.5

6.8

7.3

8.1

6.1

7.8

12.4

7.0

SS.
mg/1
19

13

16

8

12

12

9

17

11

15

12

17

NHj-N.
mg/1
<0.1

<0.1

1.3

<0.1

<0.1

<0.1

<0.1

<0.1

<0.1

<0.1

<0.1

0.4


Organic N,
mg/1
2.0

3.1

1.9

2.0

<0.1

7.8

1.0

1.3

0.4

1.7

4.0

1.3

Cotlform
MPNper
100ml
1.5

0.6

3.4

7.0

3.0

3.8

0.4

0.5

3.3

1.9

3.4

1.7

o\
   "Return activated sludge

-------
Current discharge requirements  are  essentially  those  defined by EPA for municipal
secondary treatment plants with additional requirements set on effluent toxicity.26  \n
addition, the effluent pH is restricted to the range of 6.5 to 8.5. The acid production from
nitrification and subsequent chlorination normally forces the effluent pH below 6.5. This
has necessitated the addition of caustic to the final effluent to raise the pH to or above 6.5.
Toxicity requirements  state that fish bioassays must be run on the undiluted effluent and
that 90 percent of a series of 10 consecutive tests must show 70 percent fish survival for
96 hr. Experience at this plant has indicated that the requirements cannot be met without
removal of ammonia through nitrification.26

A primary design consideration in laying out the plant for nitrification was the presence of a
significant volume fraction (11 to 13  percent) of potentially toxic industrial wastes in  the
influent wastewater. Tank truck washing residues and  the waste from a manufacturer of
organic peroxide and  phenol  formaldehyde are  the major industrial waste sources. The
roughing  filter is used in the treatment plant  to protect the nitrifying organisms from
influent wastewater toxicity. Toxic  dumps have caused severe sloughing and loss of growth
on the media in the roughing filter, but nitrification remained unaffected.
                                    FIGURE 9-9

              SAN PABLO SANITARY DISTRICT TREATMENT PLANT
                         (CALIFORNIA) FLOW DIAGRAM
                                                                        CAUSTIC
                                                                        ADDITION
                      SU8NATANT  /DISSOLVED
                      TO HEADWORKS /  AIR   1_  WASTE ACTIVATED SLUDGE
                                FLOTATION
                                THICKEN WO
                            m SLUDGE TO DRYING BEOS
                            ^^ CENTRIFUGES (FUTURE)
                                        9-27

-------
                                     TABLE 9-11
DESIGN DATA, SAN PABLO SANITARY DISTRICT TREATMENT PLANT
      Average dry weather flow (ADWF)

      Peak wet weather flow (PWWF)
      Raw wastewater quality
        BODs
        SS
      Primary sedimentation tanks
        New (Dry and wet weather use)
           Number
           Diameter
           Sidewater depth
           Detention time  (ADWF)
           Overflow rate (ADWF)

        Existing (Wet weather use only)
           Number
           Diameter
           Sidewater depth
      Roughing filter
           Number
           Diameter
           Media type
           Depth of media
           Volume of media
           Media specific  surface
           Recirculation ratio
           BOD5 load
      Air blowers
           Number
           Discharge pressure
           Capacity - total
      Aeration-nitrification  tanks
           Number
           Passes/tank
           Length/pass
           Width/pass
           Depth
           Volume (total)
           Detention time (ADWF)

      Secondary sedimentation tanks
           Number
           Length
           Width
           Depth
           Overflow rate at ADWF
                       at PDWF
           Detention time at ADWF
      Chlorine contact chambers
           Number
           Passes/tank
           Length/pass
           Width/pass
           Depth
           Detention ADWF
      Caustic addition (Na(OH))
           Dose
      Waste activated sludge thickening
           Number
           Diameter
           Sidewater depth
           Solids loading
      Anaerobic Digestion
           Number
           Volume - total
      Sludge drying bed
           Surface area
 12.5 mgd (0.55m3/sec)
 30 mgd (1.32m3/sec)


 340 mg/1
 300 mg/1
 70 ft (21.3m)
 10 ft (3.1m)
 2 hr
 1624 gpd/sf(66/m3/m2/day)
 1
 100 ft (30.5m)
 10 ft (3.1m)
 52 ft (15.6 m)
 Plastic corrugated sheet modules
 18 ft (5.5 m)
 38,200 cf (1080 m3)
 29 sf/cu ft (95 m2/m3)
 2.4 to 1.0
 350 lb/1000 cf/day (5.6 kg/m3/day)
 6.5 pslg (0.46 kgf/cm2)
 24,000 cfm (629  m3/min)


 2
 1
 252 ft (76.9m)
 50 ft (15.2m)
 15 ft (4.6m)
 378,000 cu ft(10,700m3)
 5.4 hr
. 2
 180 ft (54.9m)
 60 ft (18.2m)
 8 ft (2.4m)
 580 gpd/sf(23.6 m3/m2/ day)
 1390  gpd/sf(S6.6 m3/m2/day)
 2.5 hr
 2
 2
 110 ft (33.5m)
 15 ft  (4.6m)
 9 ft (2.7m)
 0.85  hr

 20 mg/1

 1
 35ft  (10.7m)
 8 ft (2.4m)
 48 Ib ds/sf/day (235 kg/m2/day)
 367,000 cu ft (10,400m3)

 158,000 sf (14,700m2)
                                          9-28

-------
Consistent year-round nitrification is obtained in this plant as is shown in Table 9-12. While
only  once-monthly  analyses of  nitrogen  are  shown  in  Table 9-12,  the  consistency of
nitrification in the plant is confirmed by daily ammonia nitrogen analyses of grab samples
which normally  show less than 0.2 mg/1 ammonia nitrogen. Wastewater temperatures are
favorable  for nitrification,  typically  dropping  to  only  17 C,  the average monthly
temperature in January.

The treatment plant is currently operating at only one-half of its design capacity. To test the
nitrification portion  of the system at  close to its design condition, one of the two aeration
tanks was taken out  of service during May, June and  July of 1974. Both secondary
sedimentation tanks  remained in service during this period. Operating conditions and plant
performance for this period  are summarized in Tables 9-13  and  9-14 respectively.2? Full
nitrification was maintained throughout this special test period.

Examination of  Table 9-14 provides some insight into the operation of the roughing filter.
Total BODs and total COD remained relatively unaffected by the roughing filter operation.
Evidence of treatment, however, is provided by the soluble BOD and soluble COD tests and
the total suspended solids (TSS) determinations.  A reduction in soluble BODs an(* soluble
COD  occurred coincidentally with an  increase in TSS.  This indicates organic removal with
associated growth of biomass. In this plant's operation, the roughing filter converts influent
organic  matter to biological organisms. Subsequent treatment in the aeration-nitrification
tank provides further oxidation and nitrification.

Construction cost of the added facilities totaled $4,900,000, including legal, engineering and
contract administration. The contract was  awarded  in October 1970 and  construction
essentially completed by October, 1972. Treatment plant operating costs totaled $397,500
for fiscal year 1973/1974, during which time a total of 2,600 million gallons (9.8  million
m^) were processed  through the secondary treatment facilities. On a unit basis, treatment
plant O  & M costs total $153/mil gal for fiscal year 1973/1974.26

     9.5.2 Case Examples of Nitrification-Denitrification for Nitrogen Removal

Four  examples of the  incorporation of biological nitrification-denitrification in treatment
plants for nitrogen removal are presented in this section. The Central Contra Costa Sanitary
District's plant design is oriented  towards reuse of the reclaimed water by nearby industries
as  well  as  meeting discharge requirements.  The designs  of the Canberra, Australia,
Washington, D.C., and El  Lago, Texas plants are laid out so that  nitrogen and phosphorus
are removed to protect the receiving waters.

          9.5.2.1 Central Contra Costa Sanitary District, California

The Central Contra Costa  Sanitary District (CCCSD) has under construction a new 30 mgd
(1.31  m^/sec) Water  Reclamation Plant near Concord, California. Due to go on-line in 1976,
the plant is designed to produce water for  sale to the Contra Costa County Water District

                                        9-29

-------
                                                  TABLE 9-12
                                    NITRIFICATION PERFORMANCE AT THE
                              SAN PABLO SANITARY DISTRICT TREATMENT PLANT
Month
7

8

9

10

11

12

1

2

3

4

5

6

Year
1973

1973

1973

1973

1973

1973

1974

1974

1974

1974

1974

1974

Flow,
mgd
( mVsec)
6.3
(0.28)
5.6
(0.25)
5.7
(0.25)
5.9
(0.26)
9.7
(0.43)
8.4
(0.37)
9.1
(0.40)
6.6
(0.29)
8.3
(0.36)
7.4
(0.32)
6.2
(0.27)
6.2
(0.27)
Return
sludge
ratio
0.30

0.34

0.25

0.31

0.24

0.29

0.22

0.29

0.19

0.23

0.44

0.44

Temp.
C
22.2

23.6

24.4

23.2

20.3

18.5

17.0

17.8

18.2

19.0

22.0

22.0

MLSS
(% vol-
atile)
1403
(79)
1497
(80)
1758
(76)
1765
(76)
1609
(72)
1545
(73)
1481
(74)
1415
(78)
1522
(74)
1581
(70)
e

2833
(78)
SVI,
ml/
gram
73

68

48

66

73

85

87.

96

101

69

82

77

ec,
days
12.3

13.9

15.4

11.8

9.1

12.1

7.4

9.2

7.5

6.8

e

6.0

HT, -
hr
10.8

12.1

11.9

11.5

7.0

8.1

7.5

10.3

8.2

9.2

e

5.5

Air
use
tfCF/day
(I/sec)
naa

na

20.0

19.4

19.4

20.0

na

na

18.3

na

na

15.3

Roughing filter effluent
BODSb
mg/1
121

125

134

131

88

81

91

92

107

111

140

123

CODb
mg/1
279

306

283

281

212

214

249

240

314

247

332

327

ssb.
mg/1
8.2

97

67

95

59

73

96

79

135

95

130

118

Secondary effluent
BODs°,
mg/1
16

4

6

6

7

4

3

4

4

5

3

2

CODC,
mg/1
78

56

55

62

54

59

53

S3

61

54

50

46

SSC,
mg/1
8

7

8

6

3

7

4

4

6

7

4

4

Organic-ltf1,
mg/1
7.8

3.8

3.1

3.4

3.4

4.8

4.8

2.2

2.8

4.8

5.1

3.6

NH}-NC,
mg/1
<0.2

<0.2

<0.2

<0.2

0.7

0.1

<0.2

<0.2

<0.2

<0.2

<0.2

0.3

NO3~-Nd;
mg/1
18.6

19.8

18.2

20.4

24.8

17.6

17.0

IS. 8

11.8

11.2

18.7

19.0

NO2-NC,
mg/1
0.05

0.02

0.02

0.02

0.26

0.05

0.22

0.02

0.09

0.10

0.06

0.05

 na = not available
 grab sample at peak flow each week day
 composite sample, once per week
 composite sample, once per month
e on May 7, one aeration tank taken out of service

-------
                              TABLE 9-13
  AVERAGE PROCESS LOADING CONDITIONS AT THE SAN PABLO SANITARY
           DISTRICT TREATMENT PLANT DURING SPECIAL TEST,
                      MAY 19TH TO JULY 8TH, 1974
 Flow, mgd (mVsec)
 Temperature, C
 Roughing filter
   BOD loading,
      Ib BOD5/1,000 cf/day (kg/m 3/day)
   Hydraulic loading
      gpm/sf (mVmVmin)

 Aeration-nitrification tanks
   MLSS, mg/1
   Percent .volatile
   SVI, ml/gram
   Average detention time, hr
   BOD load
      Ib BOD5/1,000 cf/day  (kg/m3/day)
      Ib BOD5/lb MLVSS/day (kg/kg/day)
   Solids retention time , days

 Secondary sedimentation tanks
   Average overflow rate, gpd/sf (mS/
   Average return ratio
   Average solids load, Ib/sf/day (kg/m2/day)
   Return activated sludge, mg/1
6.30(0.28)
23.0
199 (3.19)

2.1 (0.086)


3070
78.2
80.2
5.4

35.8 (0.57)
0.24 (0.24)
6.6


292  (11.89)
0.48
11.0 (53.7)
6835
                              TABLE 9-14
    PERFORMANCE SUMMARY FOR THE SAN PABLO SANITARY DISTRICT
TREATMENT PLANT DURING SPECIAL TESTING, MAY 19TH TO JULY 8TH, 1974
Characteristic
Total BODs, mg/1
Soluble BODs, mg/1
Total COD, mg/1
Soluble COD, mg/1
TSS, mg/1
Ammonia-N
Nitrate -N
Nitrite-N
Raw
waste-
water
220
nab
na
na
191
na
na
na
Primary
effluent
145
97.5
322
191
86.8
na
na
na
Roughing
filter
effluent a
129
52.8
334
137
121
19.8
na
na
Secondary
effluent3
3.3
na
47.5
40.7
4.9
<0.2
19.0
0.04°
 aComposite sample each weekday, except as indicated
  na = not available
 c
  Grab sample at peak flow
                                 9-31

-------
which will  resell  the  water to five large industries for cooling and process water. The
reclamation contract calls for production of a water containing less than 10 mg/1 of BOD5,
1 mg/1 total phosphorus and 5 mg/1 total nitrogen. 1

The liquid processing flowsheet for the CCCSD Water Reclamation Plant is shown in Figure
9-10 and design data  are in Table 9-15.29 primary treatment follows  lime addition and
preaeration and is followed with a separate stage nitrification step. The  use of lime in the
primary  treatment  stage removes the bulk  of the organic carbon  before  nitrification,
resulting in a very stable oxidation of ammonia to nitrate. Addition of lime also enhances
the  removal  of  organic  nitrogen,  phosphorus,  heavy metals  and  viruses.  Biological
denitrification follows nitrification, converting nitrate to nitrogen gas. Multimedia filtration
will also be provided  prior to distribution  of reclaimed water to industry. Not shown in
Figure 9-10 is  a 140 million gallon (530,000 m3) storage basin that can be used to store
primary  effluent to reduce  peak wet weather loads on the nitrification and denitrification
units. Also, the filtration facility has been provided with a 5 million gallon (18,900 m3)
storage basin for equalizing filter influent and a 30 million gallon (114,000 irP)  clear well
for storage of filtered water before pumping it into the distribution system.

There is  no intervening  recarbonation  stage  between  the primary  clarification  and
nitrification stages. External carbon dioxide (CO2) is added directly to the first pass of the
aeration-nitrification tanks as needed. External requirements are minimal  as the chief source
of CC>2 is not the external source, but the CC>2 generated in the process. Carbon  dioxide is
derived from the oxidation of both organic carbon and ammonia.   The in-process CC>2
generation capability is  an  example of lime clarification-nitrification  compatibility.   The
nitrification pH is  also kept in the 7.0 to 8.5 range, which is optimal for nitrification.

Based upon tests by the  City of Milwaukee in the 1960's^0,31 ancj testing performed at the
South Eastern Purification Plant in Melbourne, Australia,32  the decision was made to use
flat  porous  plates arranged  uniformly over the bottom of the aeration-nitrification tanks.
This method of fine bubble aeration gives an oxygen transfer efficiency of between 14 and
20 percent under standard conditions. The porous plates are  14 in. (36 cm) in diameter by
\1A  in. (3.2 cm) thick and are secured in polypropylene holders as shown in Figure 9-11.
Thirty plates are arranged in a single precast panel; there are 40 panels  per pass. Each panel
has an inverted  channel  shape and is grouted to the tank.  By using an inverted channel
shape, the channel forms an air duct with the bottom slab of the aeration-nitrification tanks.
Each channel has a manual drain so that each pass may be drained  during start-up and
shutdown  operations.  Four  panels are fed by  each downcomer pipe,  thereby allowing
throttling of the downcomer pipes and a tapered aeration operation. Air for the nitrification
tanks, channel aeration,  and preaeration is provided by two  60,000 scfm (1,698 m^/min)
steam turbine driven single-stage centrifugal blowers and is  filtered in large compartment-
type bag filters, to achieve  a particulate  concentration of less than 0.09  mg/1,000  cu ft
(0.32 mg/100m3) of air.
                                         9-32

-------
                               FIGURE 9-10
 LIQUID PROCESS FLOW SHEET - CCCSD WATER RECLAMATION PLANT (CALIF.)

                   RAW WASTEWATER

                            •PRECHLORINATION
                   SCREENING
INFLUENT
PUMPING
 WASTE
 BIOLOGICAL'
 SLUDGES
 LIME REACTOR
AND PREAERATION
                    PRIMARY
                 SEDIMENTATION
                      I
-POLYMER
 and/or FeCI3
SLUDGE _
RECLAIMED LIME
               PRIMARY EFFLUENT
                    PUMPING

                                                            MAKEUP  LIME
                                               .ASH TO
                                                DISPOSAL
NITROGEN GAS
    TO
AERATION-
NITRIFICATION
|
SECONDARY
SEDIMENTATION
AIR


RETURN
SLUDGE

                                                                SLUDGE
                                                          TO  PREAERATION
              METHANOL
ATMOSPHERE|_ 	
STABILIZATION
POSTAERATIO
POSTCHLORINA
DENITRIFICATION
REACTOR
AERATED
STABILIZATION
r.
1
i
FINAL
SEDIMENTATION
N AIR——*.
1

t
FINAL EFFLUENT
PUMPING
^



-MIXING
RETURN
.SLUDGE
WASTE SLUDGE
*TO PREAERATION
EFFLUENT TO
" SUISUN BAY


                                   CHLORINATION
                                          INDUSTRIAL SYSTEM
                                              PUMPING
                                                 T
                                           RECLAIMED WATER
                                            TO  INDUSTRY
                                  9-33

-------
                                     TABLE 9-15

                 CENTRAL CONTRA COSTA SANITARY DISTRICT
                  WATER RECLAMATION PLANT - DESIGN DATA
Population
Average dry weather flow (ADWF)
Peak dry weather flow (PDWF)
Peak wet weather flow (PWWF)
Raw wastewater quality at ADWF
   BODs
   SS
   TKN
   Total phosphorus as P
Primary solids separation
   Chemical addition
      Lime dose as CaO
      Ferric chloride dose as Fed,
      PH                       3
   Preaeration, flocculation, grit removal tanks
      Number (one existing)
      Volume, ea.
      Detention time,  ADWF
   Primary sedimentation tanks
      Number (two existing)
      Length
      Width
      Average depth
      Overflow rate at ADWF
   Detention time at ADWF
Air blowers  - steam turbine driven
   Number
   Discharge pressure
   Capacity, ea.
   Horsepower/ea. turbine
Aeration-nitrification tanks
   Number
   Passes per tank
   Length/pass
   Width/pass
   Depth
   Detention time, ADWF
Secondary sedimentation tanks
   Number
   Diameter
   Sidewater depth
   Overflow rate, ADWFa
Denitrification tanks (anoxic contact)
 and aerobic stabilization)
   Number
   Length/tank
   Width/tank
   Depth
   Detention time, ADWF
   Reactors/tank
   Reactors used for stabilization (max)
   Methanol to Nitrate - N ratio
        310,000
 30 mgd (1.31 m3/sec)
 48 mgd (2.10 m3/sec)
140 mgd (6.13 m3/sec)

        216mg/l
        240 mg/1
         30 mg/1
         11 mg/1
        303 mg/1
          14 mg/1
         11.0
 27,700 cu ft (785 m3)
        30 min
     254 ft (77.4 m)
      38 ft (11.6 m)
     9.5 ft (2.9 m)
780 gpd/sf (31.8  m3/m2/day)
          2.2 hr
 8.0 psig (0.56 kgf/cm2)
      60,000 cfm
       2,750

            2
            4
     270 ft (82.3 m)
      35 ft (10.7 m)
      15 ft (4.6m)
          6.8 hr
     115 ft (35.1 m)
      16 ft (4.9 m)
720 gpd/sf (29.3 m3/m2/day)
     315 ft (96.0 m)
      30 ft (9.1 m)
      15 ft (4.6m)
          102 min
            9
            4
           3.0
                                        9-34

-------
                                       TABLE 9-15

              CENTRAL CONTRA COSTA SANITARY DISTRICT WATER
                RECLAMATION PLANT - DESIGN DATA (CONTINUED)
  Final sedimentation tanks
     Number
     Diameter
     Sidewater depth
     Overflow rate, ADWFa
  Effluent filtration13
     Number
     Total rated capacity
     Total hydraulic capacity
     Media depth
       Anthracite
       Sand
  Filtration rate at rated capacity
  Water backwash rate,  max
  Surface wash rate
  Solids disposal and Lime reclamation
     Sludge thickening
       Primary sludge thickener (converted digester)
          Number
          Diameter
          Sludge to thickener
          Solids loading
     Centrate thickener  (converted digester)
          Number
          Diameter
          Solids loading
     Centrifugation
          Number
          Type
          Max feed rate
          Max g force, G
          Cake solids cone., first stage
          Cake solids cone.,  second stage
     Furnaces
          Number
          Diameter of hearth
          Number of hearths
          Rated capacity
           Sludge burning duty
           Rec ale ination-duty
          Recycled lime fraction
           115 ft (35.1 m)
           20 ft (6.1 m)
     720 gpd/sf (29.3 m3/m2/day)
      36 mgd (1.46 m3/sec)
      54 mgd (2.20 m3/sec)

             2 ft (0.61 m)
             1 ft (0.30 m)
     4.0 gpm/sf (2.72 VmVsec)
      25 gpm/sf (17.0 l/m2/sec)
     0.75 gpm/sf (0.51 l/m2/sec)
            62 ft (18.9 m)
  243,000 Ib/day (110,500 kg/day)
  80 Ib DS/sf/day (390 kg/m2/day)
            62 ft (18.9 m)
  42 Ib DS/sf/day (205 kg/m2/day)
                 t
       vertical, solid bowl
      260 gpm (0.016 m3/sec)
              3,100
            55 percent
            14 percent
            22 ft (6.7 m)
                11

 70,000 Ib DS/day  (32,000 kg/day)
150.,000 Ib DS/day  (68., 000 kg/day)
         60 - 65 percent
    Loading applied to sedimentation tanks at PWWF can be equalized by a primary effluent holding basin

   Data on filtration from reference 28
Nitrified  mixed  liquor flows  to four  circular  sedimentation tanks.  The  tanks have a
center-feed,  peripheral discharge  arrangement  with sludge removal by vacuum-type sludge
collectors.  Sludge  return  rate  is  controlled by the blanket  level in each secondary
sedimentation tank. Solids retention time is controlled by use of waste activated sludge flow
meters and return sludge concentration.
                                            9-35

-------
                                    FIGURE 9-11

               NITRIFICATION-DENITRIFICATION SYSTEM AT THE
               CCCSD WATER RECLAMATION PLANT (CALIFORNIA)
                                REnjRJTNITRIFIED
                                  DOE (RNS) PUMPS
                                       (TYP.)
                                     	NITRIFIED MIXED
                                     \LIQUOR CHANNEL
                                     \   VRNS CHANNEL
                                                              PASS WIDTH  35'-O
                                                                    AIR PLENUM CHANNEL
                                                                    TYP OF 40 PER PASS
                                                                    31'-6
DENITRIFICAT|ION
TANKS
EFFLUENT FRO»
PRIMARY
SEDIMENTATION
                             Nl TRIP ICATION
                                                        -jL^j^U^-:-.-
   FINAL
   SEDIMENTATION
   TANKS
                                       NITRIFIE
                                    (RDS) PUMPS
                                                           SECTIONt-1

                                                      <—POROUS PLATE
                                                                 POLYPROPYLENE HOLDER
                       -POSTAERATION CHANNELS

            PLAN AND  FLOW  DIAGRAM
                                         9-36

-------
Suspended growth denitrification  is  used for nitrate  removal.  Uncovered reactors are
employed, as initial testing showed them to  be acceptableJ Two parallel denitrification
tanks  consist  of nine  completely mixed reactors in series. This  arrangement allows
approximation of a plug flow hydraulic regime. Each reactor is equipped with a low-shear
turbine type mixer to keep the mixed liquor solids in suspension.  The last four reactors in
the series are equipped  with spargers for aeration. With this arrangement, the first five cells
in the series can be used for anoxic denitrification reactors  and the last four cells can be
used for denitrification or aerated stabilization depending on whether or not aeration  is
used. The volume devoted to each function  can be varied to  meet seasonal loads  and
temperatures. The anoxic reactors  are designed  to operate  at 0.1 Ib  NC>3 — N rem./lb
MLVSS/day at MLSS levels ranging from 3000 to 4000 mg/1. The gates between the last five
reactors allow  positive  prevention of backmixing  of oxygen  between the reactors selected
for denitrification and the reactors selected for aerated  stabilization. The gates are open at
the top to prevent trapping of floating solids in any reactor. The  denitrified mixed liquor
channel between the denitrification tanks and the final sedimentation tanks is aerated for
further stabilization. Final sedimentation  tanks are similar to the secondary sedimentation
tanks.

All waste sludge produced in the CCCSD  Water Reclamation  Plant is eventually cycled into
the primary  sedimentation tanks and appears in the primary underflow.  Sources  of sludge
include the suspended solids associated with the raw wastewater, the solids that are wasted
to the primary sedimentation tanks from the subsequent biological treatment stages, and the
inorganic sludges that are precipitated due  to chemical action.

To maintain a pH of  11.0 in the primary sedimentation tank, large quantities of lime,
approximately  400 mg/1  as Ca(OH>2,  must be used. The  need  for such a  large dosage
predicates the economical recovery and reuse of the lime. The solids flowsheet is shown in
Figure 9-12.  The heart of the  system  is a  two-stage centrifuge process using vertical
solid-bowl centrifuges, where primary sludge is separated into two components, sludge cake
rich in calcium carbonate (CaCO3)  and centrate containing most  of the organic  material,
magnesium and phosphorus. The first stage centrifuge  cake  is approximately 70 percent
CaCO3- This cake is recalcined in a multiple hearth furnace, subsequently dry classified and
the lime returned to storage. The lime recovery  is expected to be approximately 60-65
percent of the lime used.33

The first stage  centrate  is thickened before being clarified in  centrifuges  identical to those
used for primary sludge classification. The resulting cake is  reduced in  a multiple-hearth
furnace (MHF) to a sterile ash which is used for landfill. The MHF is identical to that used
for re calcining.

Energy in the hot off-gases from each MHF is reclaimed via waste heater boilers. Recovered
steam is used to run the  turbines which power the aeration blowers.
                                         9-37

-------
                          FIGURE 9-12




SOLIDS FLOW DIAGRAM AT THE CCCSD WATER RECLAMATION PLANT (CALIF.)
MAKEUP
LIME.
RECYCLE LIME I
1 1
LIME
STORAGE
EQUIP.
M
EXHAUST
GAS TO
ATMOSPHERE
WET
SCRUBBER
STEAM 1
I r~ *-^yWASTE
1 — ^ JHEAT
BOILER
1 U -i
DRY! P
SCRU8BER\/
1 RECALCINE
FURNACE
WASTI
BIOLOG
LIME SOLIC
FEED
EQUIP. OV
4 R
T
SCREENED
WASTEWATEft
r
ICAL POLYMER and/or F»CI3
ERF LOW
ETURN
\ i
P0IUAP
REACTOR PRIMAK
tfPREAE*' SEDIMENTA
__» ATION TANKS



T.ON EFFLUENT
_-— J
f THICKENER

SLUDGE
THICKENER)
THICKENER UNDERFLOW |
J
PIO«?T ^TAffF CE»
CENTRIFUGE
(WET
V-CLASSIFICAJ
XTION)/'
|CoC03
CAKE

RECALCINEO
ASH
\
_ RECYCLE LIME

DRY LIME
CLASSIFICATION
' REJECTS

UTRATE
T THICKENER
OVERFLOW
CfcNTRATE ~~~ 	 	 ~~*
THICKENER
S^
(THICKENED
ICENTRATE
LJ SECOND U 	 POLYME
STAGE
ICENTRIFUGEJ Hg
' » m
SLUDGE

IASH

1 TO PRIMARY
INFLUENT
EXHAUST
GAS TO
ATMOSPHERE
WET
CRUBBER
(
STEAM
/ 	 ^ f
c \_j
WASTEV ^
AT BOILER

-GRIT L /DRY
YSCRUBBER
-SCUM
^ ASH TO
" DISPOSAL
                             9-38

-------
Operational control of the CCCSD Water Reclamation Plant will be divided into four areas,
each of which will be manned by a senior operator, operator, and various maintenance men
during each shift. The four areas are: (1) primary treatment, (2) biological treatment, (3)
filtration, and (4) solids handling and conditioning. Because of the highly automated control
system, cathode ray tubes (CRT) are provided in each operator control room, eliminating
the need for lighted control panels. In addition to the main CRT's in the computer room,
CRT's are located throughout the plant for data monitoring. Because of the very exacting
standards required of industrial water, this  plant is designed to prevent potential plant
upsets by incorporating a direct digital control (DDC) dual-computer system to monitor and
control all process functions.34 Operation of the on-line computer is by a "management by
exception" basis which means that as long as the process status is within normal limits,
information is not printed, displayed, or needlessly alarmed.

While this plant will  not go on-line until 1976, the process was rigorously monitored in a
full-scale test at the existing CCCSD plant for a 23-month period. Portions of the test data
are summarized in Section 5.2.4.2.

The construction contract for the first phase  of work, excluding effluent filtration, totaled
$47,000,000 and was let in mid-1972. The construction contract for the effluent filtration
and appurtenances  totaled approximately  $14,000,000 and was let in the fall of  1974.
Operation and maintenance costs were estimated for fiscal year 1976/1977 at $300/mil gal
based on an average dry weather flow of 30 mgd.

          9.5.2.2 Canberra, Australia

The Lower Molonglo  Water Quality Control  Centre  (LMWQCC), an advanced wastewater
treatment plant under construction, is designed  to serve the City of Canberra, Australia's
national capital. Present discharges from Canberra's existing plants are causing algal growth
problems  in  the receiving water and in a downstream reservoir.   To circumvent  these
problems, it was established that the LMWQCC should produce an effluent that contains:35

     1.    Substantially no settleable solids, turbidity, color or odor.

     2.    BOD5 and suspended solids concentrations of less than 5.0 mg/1.

     3.    A median fecal coliform content less than 50 per 100 ml and a 90 percentile value
          of less than 400 per 100 ml.

     4.    Total nitrogen  not exceeding 2.0 mg/1 as N and  total phosphorus not exceeding
          0.15 mg/1 as P.

     5.    Detergent concentrations less than 0.5 mg/1 of MBAS.

     6.    No substances toxic to aquatic biota.

                                        9-39

-------
Unit processes employed at the LMWQCC include raw wastewater screening, lime addition,
grit removal, flocculation, primary sedimentation, nitrification, secondary solids separation,
denitrification, effluent filtration, effluent disinfection by chlorination, and dechlorination
prior to discharge. Figure 9-13 is the process flow diagram for the treatment of the liquid
fraction. Design data adopted and used for the LMWQCC is presented in Table 9-16.36 The
solids processing flowsheet is very similar to that shown in the case history for the Central
Contra Costa Sanitary District's Water Reclamation Plant in Section 9.5.2.1 and will not be
duplicated here.

Very  steep site  conditions and  confinement of the  plant to a limited area mandated an
unusual arrangement of  treatment structures. An example is the  nitrification tanks which
step down the hillside as shown in Figure 9-14. A total of 4 parallel tanks are provided. Each
tank is subdivided into 8 compartments. A very close approach to plug flow is provided by
these tanks since backmixing is prevented because the only way mixed liquor can pass along
the tank is by overflowing weirs between compartments. The plug flow arrangement is the
recommended process configuration for separate stage nitrification when very low residual
ammonia nitrogen levels are required. Aeration air is provided to each compartment through
porous plate diffusers spread across the tank floor. Diffuser arrangement is very similar to
that described in Section  9.5.2.1  for the CCCSD plant. Carbon dioxide (in furnace stack gas)
is added to the first two compartments on a continuous basis according to pH level in the
nitrification tanks.

Due  to  site  restrictions,  an  attached  growth  reactor system was chosen  for the
denitrification unit. The reactor chosen was specifically developed for this project and is the
nitrogen gas filled denitrification column described in Section 5.3.2.1. Design details of the
column are shown in Figure 5-7.

Bids for the LMWQCC were received  in February,  1974  and the winning tender was $A
27,000,000 ($US 35,600,000). The plant is expected to be completed by late 1976. O&M
costs calculated  in  February  1974, exclusive of amortization,  were estimated at the
equivalent  of $US 266/mil gal.36

         9.5.2.3 Washington, D.C.

In  1969, regulatory agencies established stringent effluent standards for treatment plants
that discharge into the Potomac River in the vicinity of Washington, D.C. These standards
required upgrading the  Washington,  D.C.  Blue Plains  Plant to  provide phosphorus and
nitrogen removal as well as improved BOD and SS removals.37,38,39,40 jn early  1975 the
EPA announced that the construction  of the denitrification portion of the Blue Plains plant
would be  delayed for  two years.  This  decision  came  after study  of  the energy and
construction costs for  the facility.  During  the  postponement  period, water quality
improvement  due  to  phosphorus  removal and  other treatment will be  evaluated to
determine  if denitrification is necessary  to achieve eutrophication control  goals.41 -phe
                                        9-40

-------
                                       FIGURE 9-13


             PROCESS FLOW DIAGRAM FOR THE LOWER MOLONGLO WATER QUALITY
                         CONTROL CENTRE (CANBERRA, AUSTRALIA)
            CI2
                                          PREAERATION
                     PRIMARY SOLIDS         AND GRIT
                    SEPARATION TANKS    REMOVAL TANKS
LOME
BYPASS IN EXCESS
   OFSxADWF
                                                                                     RAW
                                                                                    WASTEWATER
   BIOLOGICAL
  NITRIFICATION
     T.
FINAL CLARIFIERS
                                       BIOLOGICAL
                                       DENITRIFICATION
                                       COLUMNS
                                          EFFLUENT FILTER
                                                               CI2      Clg    DE-CI2
                                                                   CONTACT
                                                                    TANKS
          TO NITRIFICATION TANKS

-------
                                       TABLE 9-16

                  LOWER MOLONGLO WATER QUALITY CONTROL
                        CENTRE (AUSTRALIA), DESIGN DATA
Population
Average dry weather flow (ADWF)
Peak dry weather flow (PDWF)
Peak wet weather flow (PWWF)
Raw wastewater quality at ADWF
   BOD,.
   SS  5
   NH^-N
   TKN
Primary solids separation
   Chemical addition
     Lime dose as CaO
     Ferric chloride dose
     PH
   Flocculatlon and grit  removal tanks
     Number
     Volume (each)
     Detention time (ADWF)
   Primary sedimentation tanks
     Number
     Length
     Width
     Depth
     Overflow rate at ADWF
     Detention time at ADWF
Air blowers
   Number
   Discharge pressure
   Capacity - total
Biological nitrification tanks
   Number of tanks
   Compartments per tank
   Width
   Length/compartment
   Depth
   Volume (4 tanks)
   CO- required
   Detention time at ADWF
   BOD load
Final Sedimentation tanks
   .Number
   Diameter
   Sldewater depth
   Overflow rate at 3 x ADWF
Biological denitriflcatlon columns
   Number of cells
   Width/cell
   Length/cell
   Media depth
   Total media volume
   Application rate
   Methanol to nitrate - N ratio
Effluent filtration
   Number
   Width
   Length
   Media depth
     Anthracite
     Sand
   Filtration rate at 3 x ADWF
   Water backwash rate
            269,000
     28.8 mgd (1.26 mVsec)
     43.3 mgd (1.90 mVsec)
    144.0 mgd (6.30 mVsec)

            221 mg/1
            242 mg/1
            35 mg/1
            59 mg/1
            230 mg/1
             20 mg/1
              11.0
      27,700 cu ft (785 m )
             20 mln
         212 ft (64.7 m)
        38.4 ft (11.7 m)
         8.9 ft (2.7m)
   880 gpd/sf (35.7 m3/ni /day)
            1.8 hr
               3       7
    7.5 psig (0.53 kgf/cm )
    90,000 cfm (2,550 m3/min)
               4
               8
          35 ft (10.5 m)
          39.4 ft (12 m)
         16.2 ft (4.95 m) ,
      565 cu ft (19,960 m )
   3,600 Ib/day (13,600 kg/day)
            4.4 hr
   0.18 Ib BOD /lb MLVSS/day
         120 ft (36.6 m)
        20.6 ft (3.28 m)
  1,635 gpd/sf (66.6 nvVmVday)

               8
         32.2 ft (9.8 m)
        40.2 ft (12.3 m)
          20 ft (5.8m)     .
     196,750 cu ft (5,572 m )
146 gal/cu ft/day (19.0 m3/m /day)
              2.8
        23.75 ft (7.24 m)
         105 ft (32.00 m)

          3.5 ft (1.06 m)
          1.5 ft (0.45 mi
    6.0 gpm/sf  (4.1 1/mVsec)
   18.6 gpm/sf (12.65 l/m2/sec)
                                            9-42

-------
                                      TABLE 9-16

                  LOWER MOLONGLO WATER QUALITY CONTROL
                CENTRE (AUSTRALIA), DESIGN DATA (CONTINUED)
 Effluent chlorlna tion-dechlorination
   Number of tanks
   Contact time at ADWF
 Solids disposal and lime reclamation
   Primary underflow solids
   First stage centrifugallon (classification)
     Number of centrifuges
     Calcium carbonate recovery
     Cake solids contraction
     Total solids capture
   Second stage centrlfugation (clarification)
     Number of centrifuges
     Cake solids concentration
     Total solids capture, minimum
   Furnaces
     Number
     Diameter of hearth
     Number of hearths
     Rated capacity
        Sludge burning duty
        Recalclnatlon  duty
     Reclaimed lime to storage
     Recycled lime fraction
              1
            50 min

  198,450 Ib/day (90,000 kg/day)
          90 percent
         50-60 percent
          60 percent
         12-18 percent
           70 percent
          22 ft (6.7m)
              9

 70,400 Ib DS/day (32,000 kg/day)
237,600 Ib DS/day (108,000 kg/day)
  47,960 Ib/day (21,800 kg/day)
           68 percent
  Assumed influent nitrate nitrogen: 28 mg/1
discussion which follows includes the design of the denitrification facilities for this plant as
it provides an example of how denitrification can  be incorporated in a. large plant. Most of
the discussion in this  section is drawn from  the Process Design Manual for Upgrading
Wastewater Treatment Plants, an EPA Technology Transfer publication.37

In  1969,  very little  performance  data were available on the alternative phosphorus and
nitrogen removal methods that might be used in this situation. Through the cooperation of
the Joint EPA-DC Pilot  Plant,  it was possible to pilot and evaluate  several alternative
nutrient removal treatment sequences. Based on these studies, two-point addition of a metal
salt was selected for phosphorus  removal, and it was determined  that nitrogen removal
would be best achieved through biological nitrification  and denitrification processes. The
pilot  studies  also indicated that to consistently meet the established effluent standards,
multimedia filtration was required. Anticipated performance data for the upgraded plant are
presented in Table 9-17. Figures 9-15, 9-16 and 9-17 show, respectively, the flow diagrams
for the primary  and  secondary systems, nitrification  and  denitrification  systems and
filtration and disinfection systems of the plant.

The existing secondary system consists of four aeration tanks and 12 sedimentation units.
To handle the anticipated  increase in  plant design flow  from 240 mgd to 309 mgd, (10.5
        to 13.6 m^/sec) the existing secondary  system will be enlarged with two additional
                                         9-43

-------
                                             FIGURE 9-14

                         SECTION THROUGH NITRIFICATION TANKS AT THE LMWQCC,
                                        CANBERRA, AUSTRALIA
                                                                              LEVEL
                                                                             CONTROL  PRIMARY SOLIDS
                                                                              GATE    SEPARATION TANK
                                                                                       EFFLUENT
                                                                                      COLLECTOR PIPE
MIXED  ±
LIQUOR
CLARIFIERS
                                       COMPARTMENT (TYPICAL)
    RETURN ACTIVATED  SLUDGE
    DISTRIBUTION CHANNEL
CARBON DIOXIDE
(STACK GAS)
                          •AIR (TYPICAL)
                PLUG FLOW NITRIFICATION  TANK (ONE  OF FOUR)

-------
                                   TABLE 9-17
        ANTICIPATED PERFORMANCE DATA AND EFFLUENT STANDARDS
                      BLUE PLAINS PLANT (REFERENCE 39)


BOD, mg/1
Total phosphorus , mg/1
Nitrogen:
Organlc-N, mg/1
NHj-N, mg/1
NOj + NO J-N , mg/1
Total N, mg/1

Influent
206
8.4

8.6
13.7
0
22.3
Secondary
Effluent
35
2.0

3.0
14.8
0.2
18.0
Nitrification
Effluent
10
1.0

1.0
1.5
11.1
13.6
Denltrification
Effluent
6
0.5

1.0
1.0
1.0
3.0
Filtration
Effluent
4
0.2

0.5
1.0
0.5
2.0
Effluent
Standard
5
0.22

-
-
-
2.4
aeration tanks and 12 additional final sedimentation tanks. The aeration tanks are designed
for a volumetric loading of  120 Ib BOD5/1,000  cu ft/day (1.92 kg/m3/day),  an organic
loading of 2.4 Ib BOD/lb  MLSS/day and a MLSS concentration of 1,300 mg/1. Since the
increased design loadings require more air per unit volume than the existing aeration system
can deliver, the existing aeration capability will be increased. This system will be modified
from a  coarse-bubble, spiral-roll  system to a coarse-bubble,  spread-pattern  system  to
improve oxygen transfer efficiency. The secondary system air capacity has been designed to
provide 0.54 Ib C>2/lb BODs removed.

Alum or ferric chloride will  be added to the mixed liquor of the secondary system and is
expected to remove approximately 70 percent of the phosphorus  contained in the plant
influent. The addition of metallic salts to the secondary system is also expected to improve
the BOD removal in the secondary system from 75 percent to 85 percent. This will ensure a
secondary effluent BOD  concentration of less than 40 mg/1, which was found during pilot
testing to be desirable for successful nitrification in the second stage.  To remove most of the
remaining phosphorus, metallic salts will also be added to the nitrogen release tanks.

Biological nitrification facilities are designed for oxidation of 0.066 Ib NH^-N/lb MLVSS/day
at minimum  wastewater  temperatures  and a MLVSS  concentration of 1,700  mg/1.
At the stoichiometric oxygen requirement of 4.6 Ib O2/lb NH^-N  oxidized, one hundred
and twenty - 75 hp turbine aerators are  required.  Maximum air supply to the turbines will
be 88,000 cfm (2500 m /min). The turbines were selected in this instance because, due to
the limitations of the site, the nitrification tanks are designed  to have depths  of 30 feet
(9.15 m) to  obtain the required volume.  The turbines will provide adequate  mixing to this
depth and are capable of supplying a range of oxygen to the system as required  by varying
ammonia-nitrogen influent loads and varying wastewater temperatures. Lime will be added
to the nitrification reactor to maintain a minimum  wintertime pH of 7.5.
                                       9-45

-------
                                                FIGURE 9-1 5
ON
                              WASHINGTON, D.C. BLUE PLAINS TREATMENT PLANT
                            FLOW DIAGRAM OF PRIMARY AND SECONDARY SYSTEMS


|CL2
RAW ' '
WASTEWATER
[£E
RAW 1 '
WASTEWATER
CG
SLUDGE -
WASTEWATER _
CHEMICALS _
(INC. AIR)
S3
ANACOSTIA ^1 '
FORCE MAIN
ruup 1 AERATED
^ rump i^^^ r. m rHAM
~ MAIIONl^^^* "K" "•"*""
o
s| PSl
-• n + G><
< Z "* THICK
2 o
O u
Ul
EXIST 1 EXIST
fc, PUMP 1 	 kr.,V™Il?.
* STATION 1 "-BKII «.M«WI!
LEGEND
NTINUOUS INTERMITTENT
FLOW FLOW



ALT.
SPENT
WASHWATER

I FERRIC OR ALUM!
-
-. \r^\
\\ '**/ PRIMARY V
lERlT T%l SEDIMENTATION V
| *A TANKS r
r
i
d

TO
kVITY
ENERS
r;

ERl??f\s



SLUDGE
PROCESSING »
RECYCLE 1



GRAVITY '
THICKENING ,
OVERFLOW 1



EXIST. \
PRIMARY 1
EDIMENTATIOM |__
TANKS 1

L
r

i r
12


_^ DISINFECTION | EXCESS FLOW TO RIVER



SECONDARY| ^ SECONDARY |
1
t
RSI
A
EXIST. |T
SECONDARY 11
REACTORS H
|AIR|
-*
li^/ •
[^•2] rQ5tI*!!!iJ
f FERRIC OR ALUM J

t ^ 1
• •
! POLYMER |
FERRIC" oR~AiuM]
WSl

EXIST. | EXIST. |
SECONDARY! SECONDARY!

k
RSL

1 fyMLl
rpbl'YMETI
FERRlC~6T Al1T«r]
. . *Sl
1
W
^mm^fim
t
•HB^^H
N
DE
r
NITRIFICATION
| INFLUENT*
1
i
ALT. TO
FILTERS
OR
OUTFALL
ITC'IT"
	 £__
ALT. TO
ITRIFICATION
OR
NITRIFICATION
r
SL
-*•
                                                                            TO FLOTATION THICKENERS
              PSL
                   PRIMARY SLUDGE
                                   RSL
                                         RETURN SLUDGE
                                                        WSL
                                                              WASTE SLUDGE

-------
                          FIGURE 9-16


       WASHINGTON, D.C. BLUE PLAINS TREATMENT PLANT

FLOW DIAGRAM OF NITRIFICATION AND DENITRIFICATION SYSTEMS

LIME
1
SECONDARY
EFFLUENT |
i
i
i_
POLYMER AIR METHANOl
SPENT
WASHWATER
it t
NITRIFICATION! !

NITRIFICATION! 1 ...... | I
SED. BASINS |-^$TPAUTT0PNU
1 t !'
•» 1 ! 1
i
ALT. WSL FROM
SECONDARY
OR
DENITRIFICATION

M
DENITRIFICATIONli
REACTORS U
FERRIC OR ALUMJ

AIR 1 ' '
i
NITROGEN
. RELEASE
r TANKS
POLYMER

"\
DENITRIFICATIONI TO MULTI-MEDIA.
^ 	 1 FILTERS
t -"
; RSI
rwSl 1 [ALTj i , |WSL
1 i

ALT. WSl FROM
SECONDARY
OR
NITRIFICATION

r
Cl
SL
WASTEW
LEGEND
CONTINUOUS INTERMITTENT
FLOW FLOW

         TO
    FLOTATION THICKENERS
CHEMICALS

(INC. AIR)


    RSL


    WSL
                                                        RETURN SLUDGE


                                                        WASTE SLUDGE

-------
                                                  FIGURE 9-17

                               WASHINGTON, D.C. BLUE PLAINS TREATMENT PLANT
                           FLOW DIAGRAM OF FILTRATION AND DISINFECTION SYSTEMS
             SPENT WASHWATER
            TO NITROGEN RELEASE
OO
                            LEGEND

                          CONTINUOUS  INTERMITTENT
                            FLOW       FLOW
FLUSHING. SERVICE AND
   DILUTION WATER
                 WASTEWATER

                  CHEMICALS

-------
The nitrification sedimentation tanks are designed for average and peak hydraulic and solids
loadings of 580 and 1,210 gpd/sq ft (23.6 m3/m2/day and 49.3 m3/m2/day) and 17.4 and
36.6 Ib/sq ft/day (85  and 177 kg/m2/day), respectively. The sludge  return system  is
designed to provide return of 40 percent  of peak flow. However, the  system will normally
be operated to return 30 percent, of the average flow. Continuous monitoring of the DO
content of the nitrification effluent will be provided to ensure  that the influent DO to the
denitrification system is minimized.

The biological denitrification system is laid out to include reactors, nitrogen release tanks
and sedimentation tanks. The reactors have been designed for removal of 0.0425 Ib NO^
-N/day/lb MLVSS at  a design  MLVSS  concentration of 2,100 mg/1, with up to 4.5  Ib
methanol  added/lb NO^ —N applied.  The reactors will be  44 ft (13.4 m) deep  and be
equipped with forty-eight - 75 hp mixers,  and will be covered but not airtight.

The nitrogen release tanks were designed to serve three functions: (1) to strip supersaturated
nitrogen gas,  (2)  to  provide  mixing  for second-stage  metal  salt  addition  for residual
phosphorus removal and ,(3) to  provide an. aerobic zone for removal of excess methanol.
These tanks will furnish a  20-minute detention period at average  flow and  12 minutes  at
peak flow.40

The denitrification sedimentation tanks are designed for hydraulic loadings of 670 gpd/sq ft
(27.3  m3/m2/day) at average flow, and  1,410 gpd/sq ft (57.4 m3/m2/day) at peak flow.
Solids loadings are 25.6 and 54.0 Ib/sq ft/day (125 and 264 kg/m2/day) at average and peak
flows, respectively.

The 36  multimedia filters are designed for filtration rates of 3.0 gpm/sq ft (2  l/m2/sec)  at
average  flow and 6.2 gpm/sq ft (4.2 l/m2/sec) at peak flow. Backwashing will be done  at
intervals of 24 hours at a rate of 25 gpm/sq ft (17 l/m2/sec). The backwash water will be
equalized  in conduits and may be returned upstream of either the secondary reactors or the
nitrogen release tanks.

Provision has been made to chlorinate either upstream or downstream of the filters with 24
minutes detention provided in contact tanks following the filters.

Sludge  processing facilities will include  gravity thickening  of primary  sludge, flotation
thickening of secondary and advanced treatment sludges, vacuum filtration and  sludge
incineration.

          9.5.2.4 El Lago, Texas

El Lago, Texas, is a small suburban community of 3,000 persons located near the Lyndon  B.
Johnson Space Center. The operating agency for wastewater treatment is the Harris County
Water Control and Improvement District #50. This district  currently operates a 0.3 mgd
(0.013 m3/sec) treatment plant. In 1969, the District received an order from the Texas

                                        9-49

-------
Water Quality Board that mandated protection of Clear Lake from excessive eutrophication.
Two means were available for compliance with this order at that time; export of wastewater
or providing  nutrient removal prior to  discharge to Clear Lake. The second option was
chosen and the District obtained a grant from the EPA to demonstrate full-scale nitrogen
and phosphorus removal.

The  original plant consisted of a rock trickling filter plant  with anaerobic sludge digestion
for solids processing. The modified flowsheet incorporating nutrient  removal is shown in
Figure 9-18.  Added facilities are identified by asterisks and  include new aeration-nitrifi-
cation  tanks, new  denitrification  columns, new tertiary filtration,  and  facilities for
metal salt, polymer and methanol addition. All existing structures were incorporated into
the upgraded plant. Design criteria for the modified plant are  shown in Table 9-18.  Two
separate types of denitrification columns were provided so that alternative designs could be
compared. One  set of columns was of the submerged high porosity media type described in
Section 5.3.2.2 and was filled with Koch Flexirings (coarse media column in Table 9-18).
The  other type was the submerged low porosity media type described in Section 5.3.2.3 and
as supplied by the Dravo Corp. (Fine media column  in Table 9-18.) Both column types are
shown on Figure 9-19.

Tables 9-19 and 9-20 are tabular summaries of the initial performance with the fine media
and  coarse media respectively.  Phosphorus removal during the fine media  evaluation was
erratic due to  the  cessation  of iron addition during storm conditions. The  fine  media
columns produced an effluent containing 17  mg/1 of suspended solids  because the columns
are backwashed with nitrified effluent rather than clear tertiary filter effluent. Since tertiary
filtration is also  provided, this has not been a problem at El Lago.

Table 9-21 reproduces  a month of operating  data for the  fine  media  denitrification
columns. This data shows the improvements  in all parameters  of effluent quality obtained
after one year of experience in plant operation.

The  coarse media  denitrification columns also performed well during the investigation.
While the fine  media columns  required at  least  daily backwashing  to prevent excessive
headlosses, the  high void volume of the coarse media allowed operation without frequent
backwash. The  routine procedure was to backwash every 4 weeks. This proved to be an
important difference to the plant operators, who found that the coarse media units required
much less attention  than the fine media units.43

The  capital costs for the modifications to the El Lago Facility were incurred over a two-year
period (1971-1973) and totaled  $312,365 including change orders. This cost includes the
provision of dual denitrification facilities; had only one type of denitrification system been
included it is estimated that  the total would be about $75,000 less. The only  increase in
                                                                            o
operating costs  has been for chemicals and power and this has totaled $96/mil gal.   It was
found that the existing plant operators could adapt to advanced waste treatment processes
and no increase  in staff was required.

                                       9-50

-------
                       FIGURE 9-18
 EL LAGO, TEXAS WASTEWATER TREATMENT PLANT, FLOW DIAGRAM
        Fed
t+polymer* ^
RAW WASTEWATER
     m  Digester Supernatant
             INFLUENT PUMPING
                     1
                 PRIMARY
              SEDIMENTATION
                     SLUDGE
                 ROUGHING
             TRICKLING FILTERS
                                ANAEROBIC
                                DIGESTION
                                    WASTE
 FeCI,
        Methanol
r
TION-
CATION
NKS*
1
JDARY
JTATION
NKS
1 	 	
i
SLUDGE
RETURN
SLUDGE

I

                                              SAND
                                             DRYING
                                              BEDS
BACKWASH
   TO
HEADWORKS
 DENITRI-FICATION
     COLUMNS
  (LARGE MEDIA)
                                      DEN IT RI.F 1C AT ION
                                         COLUMNS
                                        (FINE MEDIA)
      J    I    TERT
      "^         FILTR
          I ARY
          ATION*
                                            KEY

                                            *New facility
            CHLORINE CONTACT
                       FINAL  EFFLUENT
                       TO RECEIVING  WATERS
                          9-51

-------
                                         TABLE 9-18
       DESIGN DATA, EL LAGO, TEXAS WASTEWATER TREATMENT PLANT
Population
Average dry weather flow (ADWF)
Peak dry weather flow (PDWF)
Peak wet weather flow (PWWF)
Raw wastewater quality
   BOD,.
   SS  5
   Total Kjeldahl nitrogen
Primary sedimentation tanks (existing)
   Number
   Detention time (ADWF)
   Overflow rate (ADWF)
Roughing trickling filter (rock, existing)
   Number
   Depth
   Rock size
   Volume of media
   Organic load (ADWF)
   Recirculation rate
Air blowers (new)
   Number
   Discharge pressure
   Capacity - total
Aeration-nitrification tanks (new)
   Number
   Arrangement
   Volume - Total
   Detention time (ADWF)
   MLVSS
   Solids retention  time
Secondary sedimentation tanks (existing)
   Number
   Detention time (ADWF)
   Overflow, rate at  ADWF
   Overflow rate at  PWWF
Denltrification columns
   Coarse media
     Number (series)
     Type
     Diameter
     Media depth
     Media type
     Specific surface
     Voids
     Empty bed contact time
     Surface  application rate at ADWF
     Air backwash  rate
     Water backwash rate
   Fine media
     Number (series)
     Type
     Diameter
     Media height
     Media type
     Specific surface
     Voids
     Empty bed contact time
     Surface  application rated at ADWF
     Air backwash  rate
     Water backwash rate
                 3,000
         0.3 mgd (0.013 mVsec)
         0.5 mgd (0.022 m3/sec)
         1.0 mgd (0.044 m  /sec)

                 161 mg/1
                 195 mg/1
                37.5 mg/1

                   2
                 1.6hr
        440 gpd/sf (30 mVmVday)
              6.5 ft (2.0 m)
              4 in (100 mm)
         20,900 cu ft (590  m3)
12 Ib BOD /1,000 cf/day (0.192 kg/rn/day)
         t>.3 mgd (0.013 mVsec)
       6.5 psig (0.46 kgf/cm2)
          900 cfm (25.2 mVmin)

                   2
                 Series     _
          10,100 cu ft (302 m )
                 6.1 hr
               1,000  mg/1
                10 days

                   2
                 5.4 hr
       320 gpd/sf (13 m3/m2/day)
       1,060  gpd/sf (42 m3/m2/day)
                   £,
   50 psig (3.5 kgf/cm2) pressure vessel
               10 ft (3 m)
               10 ft (3 m)
             Koch Flexirings
         105 sf/cu ft (346 m2/m3)
               92 percent
                  1 hr
        2 . 5 gpm/sf (1.6 1/m  /sec)
        10  cfm/sf (3.1  mVmVmin)
        20  gpm/sf (13.5 l/m2/sec)
   50 psig (3.5 kgf/cm2) pressure vessel
               6 f t (1. 8 m)
             6.5 ft (2.0 m)
         3 to 4 mm uniform sand
        250 sf/cu ft (825 m2/m3)
               40 percent
                0.25hr  „
         7.4 gpm/sf (5 1/m /sec)
        8 cfm/sf (2.4 mVmVmin)
        20 gpm/sf (13.5 l/m2/sec)
                                             9-52

-------
                                 TABLE 9-18
DESIGN DATA, EL LAGO, TEXAS WASTEWATER TREATMENT PLANT (CONTINUED)
 Tertiary filtration
   Number (parallel)
   Type
   Height
   Diameter
   Media height and type
   Media support base
   Surface application rate at ADWF
   Water backwash rate
 Chlorine contact tank
   Detention time (ADWF)
 Anaerobic digestion
   Volume
 Sand drying bed-
   Area
               o
30 pslg (2.1 kgf/cm ) pressure vessel
          8 ft (2.4 m)
         3.5 ft (1.15 m)
  3 f t (1.0 m) of 0 .3 to 0 . 8 mm sand
       0.5 ft (0.15 m) gravel
    2.2 gpm/sf (1.5 l/m2/sec)
     15 gpm/sf (10 l/m2/sec)

             1 hr

       8,830 cf  (2,472 m3)

        6,300 sf (580 m2)
                                 FIGURE 9-19

         EL LAGO, TEXAS DENITRIFICATION COLUMNS, COARSE MEDIA
                TYPE ON RIGHT AND FINE MEDIA TYPE ON LEFT
                                     9-53

-------
                                   TABLE 9-19

     INITIAL PERFORMANCE OF FINE MEDIA DENITRIFICATION COLUMNS
                 AT EL LAGO, TEXAS - JUNE 4 TO JULY 6, 1973
Constituent
Total P
Soluble P
SS
NH+-N
TKN
NO~-N6
BOD5
COD
Temperature
Mean value, mg/1 at indicated sample location
RaW a,b
wastewater
12.8
10.3
113
18.7
42.6
175
297
26.5
Primary
Influent
15.4
4.7
289
21.7
38.6
222
488
-
Primary
effluent
8.4
4.1
72
21.5
30.2
181
-
Nitrified
effluent
7.3
3.4
37
0.9
3.7
15.2
65*
121
-
Denitrified
effluent0'
6.6
5.5
17
0.8
2.4
2.6
9
72
-
Final
effluent
4.8
3.6
3
0.6
3.3
2.3
9
51
-
  Average flow to plant: 0.307 mgd (0.013 m /sec)
  Peak dally flow to plant:  1.0 mgd (0.044 m /sec)         3
  Average flow to denltrlflcatlon columns: 0.254 mgd (0.011 m /s,ec)
  Peak dally flow to denltrlficatlon columns: 0.420 mgd (0.018 m /sec)
  Nitrite - N always less than 0.2 mg/1
  Includes methanol

  Degrees C
                                  TABLE 9-20
   INITIAL PERFORMANCE OF COARSE MEDIA DENITRIFICATION COLUMNS
               AT EL LAGO, TEXAS - JULY 8 TO AUGUST 31, 1973
                      Mean value, mg/1 at Indicated sample location
Constituent
Total P
Soluble P
SS
NH4-N
TKN
N0~-Ne
BOD
O
COD
Tempera tureg
Raw
wastewater '
12.3
10.3
102
16.3
29.7
143
248
27.2
Primary
Influent
13.1
3.1
231
14.6
31.8
156
336
-
Primary
effluent
6.7
2.4
63
14.4
26.7
87
167
-
Nitrified
effluent
-
-
43
0.9
2.6
13.6
43 f
107f
-
Denitrified
effluent0'
-
-
19
1.2
2.5
0.9
15
52
-
Final
effluent
2.8
2.3
4.5
0.9
1.7
0.6
8
38
-
!| Average flow to plant: 0.320 mgd (0.014 m /s.ec)
  Peak daily flow to plant: 0.900 mgd (0.039 m /sec)        3
° Average flow to denitrlflcatlon columns: 0.315 mgd (0.014 m /s^ec)
  Peak dally flow to denltrlflcatlon columns: 0.632 mgd (0.028 m /sec)
6 %T I i.—I f. _  TlT _l...«..n lAnn ^Vi^r\ n  O rvirv/l
e Nitrite - N always less than 0.2 mg/1
* Includes methanol
9 Degrees C
9-54

-------
                                     TABLE 9-21

    SUBSEQUENT PERFORMANCE OF FINE MEDIA DENITRIFICATION COLUMNS
          AT EL LAGO, TEXAS - OCTOBER 1 THROUGH OCTOBER 31, 1974
Constituent
Total P9
Soluble P
SS
NH|-N
TKN
NO3-N
BOD5
COD
Temperature
Mean value, mg/1 at indicated sample location
Primary .
Influent3'
12
1.8
295
-
-
-
~
-
Primary
effluent
3.6
1.0
51
18
24
-
~
-
Nitrified
effluent
3.1
0.41
81
0.4
2.6
15
62e
113e
21
Denitrified
effluent0 -d
Column No. 1
-
-
51
-
-
1.9
16
44
-
Denitrified
effluentc-d
Column No. 2
-
-
44
0.4
1.5
0.9
12
36
-
Final
effluent
0.41
0.40
1
0.3
0.9
0.6
3
19
-
 3 Average daily flow to plant: 0.301 mgd (0.013 mVsec)
  Peak daily flow to plant:  0.47 mgd (0.021 m3/sec)
 ^Average flow to denitrification columns:  0.282 mgd (0.012 m3/sec)
  Peak daily flow to denitrification columns: 0.470 mgd (0.021 m /sec)
 , Includes methanol
  Degrees C
 gReduction of P due to forric chloride addition to primary and nitrification step (37 mg/1 as Fe),  Also polymer,
  DOW A-23, added to primary at 0.23 mg/1 and to tertiary filter at 0.17 mg/1.
     9.5.3 Case Examples of Breakpoint Chlorination for Nitrogen Removal

Two examples of the use of breakpoint chlorination for nitrogen removal are presented in
this section. The Sacramento  Regional County Sanitation District's plant will incorporate
breakpoint chlorination of approximately one-half of the plant  effluent to achieve the
partial nitrogen removal dictated by plant effluent requirements. The Montgomery County,
Maryland facility is  designed for breakpoint chlorination of the entire flow to meet rigid
limitations on total nitrogen set on the plant's effluent.

          9.5.3.1 Sacramento, California

The  proposed  Sacramento Regional  Wastewater  Treatment  Plant  will be  owned and
operated by  the Sacramento  Regional County Sanitation District to serve the City and
County of Sacramento. The Regional Plant is designed for 125 mgd (5.43 m /sec) average
seasonal dry  weather flow. It consolidates 23 existing plants presently discharging to the
Sacramento  and American Rivers  above  Sacramento  and provides for discharge  to the
Sacramento River,  downstream.   At minimum river flows maintained by upstream dam
development, the  125 mgd average daily  flow will be  about 2.7 percent of the total river
flow at the plant discharge site.
                                         9-55

-------
Effluent from  the  plant will comply with waste discharge requirements adopted by the
California Regional Water Quality Control Board on October 25, 1974.  Effluent quality
requirements require BOD and suspended solids to average less than 30 mg/1 on a monthly
basis. Total nitrogen is limited to 15 mg/1 when Sacramento River flow is below 12,000 cfs
(340 m /sec) at a specified gauging station. The  15 mg/1 total nitrogen requirement under
low flow conditions in the  receiving water is to reduce algae blooms. Based on long-term
rainfall  and  river flow data, it is anticipated  that nitrogen removal will be required for an
average  of 67 days per year.   »4"  One consideration in developing design criteria for the
regional plant was to find the most cost-effective solution for intermittent nitrogen removal.

In the regional facility mixed municipal and cannery waste will undergo treatment steps
including  prechlorination,  preaeration,  grit removal,   primary  sedimentation,  oxygen
activated sludge treatment, secondary sedimentation, post aeration to strip carbon dioxide
and raise pH,  chlorine disinfection, and sulfur dioxide dechlorination.  For intermittent
nitrogen removal at  Sacramento,  breakpoint chlorination  was most cost-effective. In
comparison  to  biological nitrification-denitrification,  breakpoint chlorination  was cost-
effective if the  total nitrogen limitation  did  not exceed a duration of about 300 days per
year.  Breakpoint via on-site production of  hypochlorite solution from  direct mixing of
liquid chlorine and caustic was  cost-effective in comparison to electrolytic generation of
hypochlorite if the  duration of the  nitrogen limitation did not exceed about 200 days per
year.46

Most  significant  in  the cost  analysis is  the capital expenditure to  meet  the maximum
chlorination capacity. Calculations indicate the maximum required chlorination capacity for
breakpoint will be  170 tons per day (154,000 kg/day). This figure represents treating the
entire plant flow. The average required chlorination capacity for breakpoint is 78 tons of
chlorine per day (70,700 kg/day). One alternative for meeting maximum required capacity
would be to provide 42  standard 8,000  Ib/day  (3,600 kg/day)  vacuum chlorinators.
Alternatively, generation of hypochlorite using the  electrolytic  process would require 51
units  of 3.3 ton/day  (2,990 kg/day) capacity as well as brine  and  salt  storage. Another
alternative was reviewed, that employed  by the Los  Angeles County system which utilizes
liquid chlorine and base mixed  in a water stream.   This system appeared  to provide the
required maximum operating flexibility with minimum investment cost. Thus, it was decided
to feed liquid  chlorine and caustic soda into a recirculated effluent water stream directly
forming hypochlorite.

Liquid chlorine is available in rail tank cars from San Francisco Bay area producers as well as
Washington  state. The rail facilities allow flexibility of plant deliveries and eliminate the
need for permanent on-site chlorine  storage tanks.

Design criteria for  the breakpoint facilities  at Sacramento are summarized in Table 9-22.
The  breakpoint system is designed as two  complete and separate units  each capable of
chlorinating half of the plant flow.  This allows breakpoint chlorination of only the portion
of the plant flow required to meet the 15 mg/1 total nitrogen limitation. It is estimated that

                                        9-56

-------
                                       TABLE 9-22

        DESIGN CRITERIA FOR HYPOCHLORITE PRODUCTION FACILITY
          SACRAMENTO REGIONAL WASTEWATER TREATMENT PLANT
Plant Loading
   Flows
     Maximum hourly
     Average dry weather
     Average seasonal dry weather
   Total nitrogen (Influent)
     Maximum hourly
     Average dry weather
Chemical requirements
   Chemical ratios
     Ammonia nitrogen to total nitrogen
     Chlorine to ammonia nitrogen ratio
        Maximum
        Average
     Caustic to chlorine
        Maximum
        Average
   Chemical feed rates
     Chlorine
        Maximum
        Average8
     Caustic
        Maximum
        Average a
Chemical storage and delivery
   Chlorine storage
     Number single unit railroad tank car spots
     Maximum capacity, each unit
     On-line storage capacity
     In-plant storage capacity
   Sodium hydroxide storage
     Number tanks
     Size, dla. x height
     Capacity, each tank (25% cone.)
     Number single unit railroad tank car spots
     Maximum capacity, each unit (50% cone.)
   Sodium hydroxide feed pumps
     Number
     Capacity, each unit (25% cone.)
   Padding air compressors
     Number
     Capacity, each unit
     Discharge pressure
Sodium Hypochlorlte Generators
   Number
   Maximum capacity, each unit
   Channel mixers
     Number
     Capacity
   Ammonia analyzers
   Chlorine analyzers, total
   Chlorine analyzers, free
   Water supply pumps
     Number
     Capacity, each unit
150 mgd (6.57 rnVs)
110 mgd (4.82 m3/s)
125 mgd (5.48 mVs)

 37 mg/1
 33 mg/1
0.75 : 1.00

10.0 : 1.0
 9.0 : 1.0

 1.3 : 1.0
 1.0 : 1.0
170 tons/day (154,195 kg/day)
 78 tons/day ( 70,750 kg/day)

221 tons/day (200,450 kg/day)
 78 tons/day ( 70,750 kg/day)
 12
 90 tons (81,630 kg)
  5 days
  7 days
40 ft x 20 ft (12.2m x 6.1 m)
188,000 gal (711,580 1)
  2
 90 tons (81,630 kg)
 75 gpm (4.73 1/s)

  2            3
130 scfm (61.3 m /s)
220 psig (15.2 bar)
 85 tons/day (77,100 kg/day)
 40 hp (29.8 kW)
  2
  2
  2
800-2100 gpm (50.5-132.5 1/s)
 Represents breakpoint chlorlnatlon of three-quarters of average plant flow.
                                            9-57

-------
both  units  will be  required  to be operational  at  certain  times  to maintain effluent
requirements. Figure 9-20 is a simplified schematic of the breakpoint chlorination system
and its control instrumentation. The process consists of producing sodium hypochlorite by
inline  mixing of  liquid  chlorine,  water,  and  caustic at a chlorine  concentration of
7,000-8,500 ppm and a pH of .7.5. Maintenance  of proper pressures in the chlorine feed
system is the key to successful operation of the hypochlorite generation system. Experience
at Los Angeles County White Point plant indicates a necessary pressure of 40 psig (2.8
kgf/cm2) in the mixing tee and greater than 110 psig (7.73 kgf/cm  ) at the chlorine feed
valve. These pressures require  pressurizing ("padding") the chlorine railcars with 175 psig
(12.30 kgf/cm2) dry air.

A chlorine reserve tank is used  to provide chlorine during periods when empty chlorine tank
cars are being replaced with full cars and can provide about a one-hour supply of chlorine.

The feed rate of liquid chlorine is measured with a steel tube rotameter and controlled with
an automatic valve. Pressure drop from the chlorine feed line  to the mixing tee and initial
mixing of chlorine, caustic and water is  partially accomplished using a plug  injector. To
improve mixing immediately downstream of the mixing tee, an inline powered mechanical
mixer is provided. The hypochlorite solution line then includes three taps for pH metering
and a  back  pressure valve to  maintain 40  psig at the mixing  tee. The pH meters are
duplicated for each unit. The duplication is for improved reliability of the pH system and
the multiple taps are for additional flexibility in selecting the  point which is monitored for
control purposes.

The related feeds of caustic and water are provided by pumping from the chemical storage
area and the effluent channel.  Caustic is fed  from storage tanks through centrifugal pumps,
automatic control  valves, and  magnetic flow meters. The control signal is proportional to
the chlorine feed with feedback control provided by the pH meters. Manual caustic feed is
provided for breakpoint start-up and for pH adjustment of final effluent.

Plant  effluent is used  for sodium hypochlorite  solution  water  and is  provided by two
variable speed pumps located as shown in Figures 9-20 and 9-21. The control of the solution
flow (via pump speed) is proportional to the chlorine feed. The solution water flow is
measured by magnetic flow meters just ahead of the mixing tee.

The  plan  view of Figure  9-21  also  shows the  application points for the breakpoint
chlorination located at the end of each battery  of secondary sedimentation tanks. To insure
rapid mixing, two spargers are  installed with orifices that insure a minimum exit velocity of
about 10 fps (3.1 m/sec). Automatic valving provides one sparger for lower  feed rates and
two when flows cause back pressure to reach  the control limit  of the hypochlorite back
pressure valve. Immediately  downstream  of the breakpoint spargers is a mechanical mixer
followed by a submerged overflow weir. The mixing channel and the post-aeration channel
following are covered. The exhaust can be passed through mobile activated carbon filters to
remove odors, such as nitrogen trichloride,  if  necessary. The activated carbon filters are

                                        9-58

-------
                                   FIGURE 9-20


             HYPOCHLORITE GENERATION SCHEMATIC - SACRAMENTO
                   REGIONAL WASTEWATER TREATMENT PLANT
                                     CONTROL   PRESSURE INDICATOR ^
                                            f TRANSMITTER Q-H
                               TOTAL Clj RESIDUAL
                                BREAKPOINT
                               MONITOR ANALYZER
                                                                                        CHLORINATED
                                                                                        EFFLUENT
SECONDARY
 tFFLUEN
                      BREAKPOINT
                      MIXER
DISINFECTION
MIXER

-------
                                                          FIGURE 9-21

                                 PLAN AND SECTION OF THE BREAKPOINT FACILITY AT THE
                                 SACRAMENTO REGIONAL WASTEWATER TREATMENT PLANT
                    BATTERY JL
                    SECONDARY
                    EFFLUENT
                            BATTERY I
                            SECONDARY
                            EFFLUENT
VO
ON
o
                     _._.
                              -BREAKPOINT MIXER
              KEY
                                                                     -BREAKPOINT MIXER
                                                                                               EXHAUST GAS SYSTEM

                                                                                                   DISINFECTION MIXER
                                                                                                     /-WATER SUPPLY
                                                                                                     /   PUMPS
                                                       o
                                                                                                 7
                                                                                                          CHLORINATED
                                                                                                           EFFLUENT
              [A]  CI2 RESIDUAL ANALYZER
TO AMMONIA-
NITROGEN    40 HP
ANALYZER   BREAKPOINT
           MIXER
              PLAN VIEW-POST AERATION CHANNEL
                            NO SCALE
                                                                                          DISINFECTION
                                                                                          TOTAL CI2
                                                                                          RESIDUAL ANALYZERS-
BOO-2100 GPM
WATER SUPPLY
  PUMP
                                                     33/75 HP DISINFECTION
                                                           MIXER
                                            MOBILE
                                            VENTILATION^^.
                                 BREAKPOINT
                                 TOTAL f FREE CI2
                                 RESIDUAL ANALYZERS
            SECONDARY
            EFFLUENT  BREAKPOINT
                     SPARGER
 -DETENTION. TIME = 25 Sec

kAT 62.5 MGO
                                                                        DISINFECTION/ ZDETENT|ON TIME , 20 S.C*
                                           SCHEMATIC  SECTION- POST AERATION CHANNEL
                                                            NO SCALE

-------
mobile units that are identical to those used throughout the plant in various applications.
When the carbon is exhausted, the mobile units are replaced with duplicate units and taken
to a central carbon handling facility for bed replacement. Breakpoint chlorination control is
closed loop. Feed forward control includes ammonia nitrogen concentration, effluent flow,
and a ratio proportioner of chlorine to ammonia nitrogen. Feedback control is achieved by
free chlorine residual measured after the breakpoint reaction. Ammonia nitrogen samples
are pumped continuously from  each battery upstream of the hypochlorite sparger to
ammonia nitrogen analyzers. Chlorine  residual analyzers are of the amperometric type and
are in duplicate. One analyzer measures free chlorine residual and is used for control; the
other analyzer  measures total chlorine residual and is used as a monitor. Separate provision
is made for feeding chlorine for disinfection when breakpointing is not being practiced. This
was done because different magnitudes of chemicals are  involved  when dealing  with
breakpoint than with disinfection.  Estimates for capital  and operating  costs  for the
breakpoint chlorination facility are in Tables 9-23 and 9-24.
                                   TABLE 9-23
       CAPITAL COST BREAKDOWN FOR BREAKPOINT CHLORINATION AT
        THE SACRAMENTO REGIONAL WASTEWATER TREATMENT PLANT
Item
Breakpoint generation equipment
Outside piping
Chlorine unloading facilities
Caustic storage and pumping
Railroad
Air padding facilities
Subtotal
Engineering and contingencies
Total
Estimated cost9
$ 95,000
240,000
110,000
162,000
244,000
55,000
$ 906,000
272,000
$1,178,000
aCost basis: October, 1974
                                  TABLE 9-24
   TOTAL ANNUAL COST BREAKDOWN FOR BREAKPOINT CHLORINATION AT
       THE SACRAMENTO REGIONAL WASTEWATER TREATMENT PLANT

Chemical cost '
Labor cost
Subtotal
Amortization of capital :
Total annual costc
1,000/yr
1,265
30
1,295
101
1,396
$/mil gal treated
172.00
4.00
176.00
13.50
189.50
$/mil gal - annual average
31.50
3.00
34.50
2.50
37.00
  Costs are based on chlorine @ $144/ton and caustic soda @ $168/ton, October 1974 prices;
  average plant flow of 110 mgd; and'breakpoint chlorination required 67 days/aver, yr. at
  an average of three quarters of the plant flow.
  Power costs negligible
 Q
  Capital recovery at 7 percent and 25 years
                                      9-61

-------
          9.5.3.2 Montgomery County, Maryland

This new 60  mgd  (2.6  m3/sec) facility is now being designed. ° It will be owned and
operated by the Washington Suburban Sanitary Commission to serve the Maryland suburbs
of Washington, D.C. The plant  will discharge to the Potomac River above the raw water
intakes for the Washington, D.C. water treatment plants.  " At minimum  river flows, the
effluent will  make  up about 15 percent of the water volume at the water plant intakes.
Because of the critical nature of the downstream water use, the following effluent goals have
been established:

                   Parameter                                  Value

                   BOD5, mg/1                                1.0
                   Suspended solids, mg/1                      0
                   Total nitrogen (as N), mg/1                  2.0
                   Total phosphorus (as P), mg/1                0.1
                   Chloride, mg/1                           200-350
                   Total dissolved solids, mg/1               850-1,120
                   Coliform bacteria - MPN/100ml             2.2
                   Fecal coliform bacteria - MPN/100ml       2.2

Figure 9-22 is a flow diagram of the treatment process. The final effluent will be stored in a
reservoir with a 10  day holding capacity before discharge to the Potomac. The lime sludges
will be recalcined and reused and the granular carbon will be regenerated on-site for reuse.

The breakpoint process was selected over alternate approaches for several reasons.  It was
felt that the results of the first scale-up of the selective ion exchange process underway at
the nearby Upper Occoquan (see Sec. 9.5.4.1) plant in Virginia should be available before a
60 mgd (2.6 irr/sec) selective ion exchange facility was attempted.

 The  prolonged cold weather periods  made  ammonia stripping inadequate for substantial
 portions  of the year. The biological approach seemed to offer no significant cost savings and
 was prone to occasional inefficiencies which would be particularly significant because of the
 effluent quality goals. The effluent TDS additions from  the breakpoint process were not a
 limitation in  this case because the effluent does not enter a reservoir or confined watershed
 where  the solids would  be recycled and continue to build up but enters the Potomac River
 shortly before it flows into the Potomac Estuary.

 Concern  over the hazards of transporting and storing large quantities of chlorine gas led to
 the decision to use sodium hypochlorite for the breakpoint process. An investigation of the
 availability  and cost of sodium  hypochlorite  led to  a  decision to install  an  on-site
 hypochlorite  generation facility of the  membrane  cell  type.51>52 For  this electrolytic
 process of sodium  hypochlorite generation, the raw materials required are electrical power,
 salt,  and water. The  membrane cells and the overall system  are shown schematically in

                                         9-62

-------
                                    FIGURE 9-22
      FLOW DIAGRAM OF THE MONTGOMERY COUNTY, MARYLAND PLANT

1
1
1
1
Primary
Settling


Watte Organic
Sludge
Incineration

T"
1
1
t

Activated
Sludge




"t
Lime
Recovery

Secondary Litr" Coagulation
Settling f Phosphor ut'Removol
1 	
— — Recvcle —

Recarbonat ion
* to pH 9,3
1
	 t
Plant
Effluent
Retervoir
(Optional)


Supplemental
Chlorination
(at needed)


Granular
Carbon
Adtorption
-
BP Chlorination
for
NrlJ-N Removal
               Oitchorge
              to Potomac
                River
 LI
  Carbon
Regeneration
Figures 9-23 and 9-24. Saturated brine is fed to the anode compartment of the cells. At the
anode surface, chlorine gas is generated. The effluent from the cathode compartment is sent
to a gas-liquid separator in which  the  hydrogen is  removed from the caustic solution.
Chlorine  and caustic are fed to the reactor, with the caustic in slight excess, to produce
sodium hypochlorite.  The reactor is water cooled to avoid  decomposition of hypochlorite
and formation of oxygen. Insoluble gases,  mainly oxygen, are  removed in a gas-liquid
separator before the  hypochlorite is sent to product storage. Brine  feed for the system
comes  from a salt storage tank which is used for salt storage and production of saturated
brine. The  cell typically produces hypochlorite  with 8  percent available chlorine at 1.7
kwh/lb-(0.77-kwh/gm) available chlorine.

The  design  criteria for the hypochlorite production facility at Montgomery County are
summarized in Table 9-25. Concentrated brine  is created by combining softened  water
(softened to minimize potential for scaling of the membrane with calcium) and solid salt to
achieve a saturated brine solution (approximately 26 percent salt by weight). Both the solid
salt storage and  salt solution are  contained in lixators  which are  fitted  with a brine
collection system and a brine  level  control system. Delivery of the solar-type salt to the
lixators is by 22 ton (19.95  metric tons) dump trucks. The brine is pumped  from the
lixators to a brine treatment system, consisting of successive addition and mixing of caustic
soda (NaOH)  and soda  ash (Na2CO3)  followed by  settling  to  precipitate  calcium  and
magnesium. The brine is then  filtered through a rapid sand filter and a cartridge filter to
remove suspended solids and pumped to storage. Storage for a one-day supply of treated
                                        9-63

-------
                                 FIGURE 9-23
         MEMBRANE CELL USED FOR HYPOCHLORITE PRODUCTION
      CHLORINE*
T BRINE





ANODE 	 »»
MEMB
i

BRIMF ——
RANE'^
i

,


HYDRC

^CATHODE
N
i


t



WATER
                                                                    OR
                                                          DILUTE  BRINE
brine is provided. The direct current source required by the membrane cells is provided by a
solid state rectifier, with side-mounted controls. Safety features in the control panel include
automatic shutdown in case of high reactor temperature, high DC voltages, blower failure,
or high or low DC current.  The hydrogen from  the electrolytic cells can be either (1)
diluted  with air as it  is formed to maintain a hydrogen  concentration of less than 0.25
percent by volume, the explosive limit, or (2) compressed and piped to the solids processing
building for use as a fuel. Sodium hypochlorite leaves the reactors by  gravity flow to pumps
from which it is sent to storage.

The  breakpoint reaction is accomplished by adding a sodium hypochlorite solution to the
wastewater at a dosage slightly in excess of the stoichiometric requirement for oxidation of
the ammonia nitrogen to  gaseous nitrogen. Figure 9-25 illustrates the design of the system
and Table 9-26 presents the design criteria. Filter effluent flows by gravity to the breakpoint
chlorination reactors.
                                       9-64

-------
                                  FIGURE 9-24

 OVERALL SYSTEM USING MEMBRANE CELLS FOR HYPOCHLORITE PRODUCTION
                                        CHLORINE
BRINE-
 WATER-
                  CELL
                 STACK
                                  SEPARATOR
                            SPENT BRINE
HYDROGEN
                                                           VENT

                                                            TlNERTS
                                                        REACTOR
r
                                              WASTE
                                             VENT
                                 SEPARATOR
                                         CAUSTIC
                                   COOLER
                                       PRODUCT
The wastewater first passes through two in-line mechanical mixers in series. The sodium
hypochlorite is added to the  first mixer along with sodium hydroxide if needed for pH
adjustment. The second mixer is provided for protection if the first mixer malfunctions. The
second mixer is normally operated to provide thorough mixing. The mixer was sized so that
a single mixer would provide violent mixing (G = 1,000 sec~l) to insure instant and
complete mixing of the hypochlorite and the wastewater.

Wastewater then flows to breakpoint reactors  (closed concrete tanks) where it is air mixed
to complete the chemical reaction  and where 30 minutes contact time for disinfection is
provided. Flow is distributed over the first  half of the length of each basin through a
multiport header. Distribution of flow in this  manner minimizes the decrease in pH caused
by the reaction of sodium hypochlorite with ammonia nitrogen (by avoiding a single point
of injection into the basin of  the wastewater-hypochlorite mixture) and thereby minimizes
the formation of nitrogen trichloride. Air is diffused into the wastewater over the bottom of
                                      9-65

-------
                                          TABLE 9-25

                DESIGN CRITERIA FOR HYPOCHLORITE PRODUCTION
                 FACILITY AT THE MONTGOMERY COUNTY FACILITY
 Average dry weather flow
 Sodium chloride storage and treatment
    1.  NaCl required @ 60 ton/day C12 production8
    2.  Salt storage
        Liquid level
        Storage capacity
        Maximum saturated brine production
    3.  Type of salt used
    4.  Brine treatment system
       A.   Soda ash  requirements
           3 minute rapid mix
       B.   NaOH requirements
           3 minute rapid mix, 3 hour settling
       C.  Rapid sand filters and cartridge filtration
            follows  settling
    5.  Treated brine storage, fiberglass tanksb

    6.  Truck delivery
        7 trucks/day/5 day week
 Sodium hypochlorite generators
    1.  Generating capacity
    2.  Modules required @ 3.3 tons each
        On  line
        Redundant
    3.  Rectifiers required
    4.  Power required
    5.  Building required
        Builder provides for housing, modules, and rectifiers.
        Monorail with 5 ton capacity provided for cell stack
          maintenance.
        Rectifiers separated from modules by a glass wall to
          prevent corrosion.
    6.  Generator module dimensions
    7.  Rectifier dimensions
 Sodium hypochlorite storage
    1.  Storage sized  to provide 1 day storage for power outage
        plus storage for 7 days @ 90  mgd (3.94 m3/sec)
    2.  Storage capacity
           60 mgd (2.63 m3/sec)

     105 tons/day (95.2 metric tons/day)

4 @ 22 ft x 33 ft x 18 ft (6.7 m x 10.1 m x 5.5 m)
        1,470 tons (1333 metric tons)
        60 gpm (3.8 I/sec)  each (min.)
                 Solar

          450 Ib/day (204  Kg/day)
3 ft x 3 ft x 3 ft swd (0.91 m x 0.91 m x 0.91 m)
          360 Ib/day (163  Kg/day)
3 ft x 3 ft x 3 ft swd (0.91 m x 0.91 m x 0.91 m)
     6 @ 12 ft x 12 ft (3.66 m x 3.66 m)
         16,000 gal each (60,560 1)
      22 tons/load (19.95 metric tons)
       60 tons/day (54.4 metric tons)
                  20
                  18
                   2
                   7
               10,000 KVA
 87 ft x 96 ft x 20 ft (26.5 m x 39.2 m x 6.1 m)
   4 ft x 16 ft x 8 ft (1.2 m x 4.9 m x 2.4 m)
   4 ft x 6 ft x 8 ft (1.2 m x 1.8 m x 2.4 m)
           700,000 gal (2650 m3)
     230 tons equivalent (209 metric tons)
   6-117,000 gal ground level tanks (443 m3)
     6-30 ft x 22 ft tanks (9 .1 m x 6.7 m)
  Assumes 90 percent utilization
  Each tank will provide 4 hours storage at full production

the second half of the basin to strip any nitrogen trichloride from solution. Additional air
may be introduced  in the space between the liquid surface and the basin cover to further
dilute any nitrogen  trichloride  that might be  present to  prevent the development  of an
explosive  concentration that occurs at 0.5 percent by volume. The diffusion of air into the
breakpoint reactor contents also strips gaseous  nitrogen and carbon dioxide from solution.
The latter will result in a desirable increase in pH.

Exhaust gases from the reactors are recirculated to provide mixing with some of the gas bled
off to  the recalcining furnace for thermal decomposition of the  nitrogen trichloride. The
alkaline environment in the recalcining  furnace  will avoid discharge  of hydrochloric acid
(HC1) to the atmosphere that might otherwise occur.
                                              9-66

-------
                                                 FIGURE 9-25

                               SCHEMATIC OF MONTGOMERY COUNTY, MARYLAND
                                     BREAKPOINT CHLORINATION PROCESS
o\
      ISOLATION VALVE

      METER 	
      BALL VALVE
      FROM
FILTERS  /
(One of Six/
 lines shown)
                                                              C|RCULA|ON
                        GAS
                	^BLEED-OFF TO
                        RECALCINERS
                EXHAUST
                GAS     /--EFFLUENT
                        ^ WEIR
                                            INLET DIFFUSER
AIR HEADERS


         ISOLATION
                                            INLET GATES
                                                                              n
                                                                           N >*1t
                                                                           -^"^
                                             ^-
                                               One of  Six Basins shown
                                                                                         UJ
                                                                                         z
                                                                                         0
                                                                                   " J
                                                                                  •3f
                                                                                   u. L»-
                                                                                   UJ
                                                                                              TO
                                                                                              CARBON
                                                                                              PUMP
                                                                                              STATION

-------
                                           TABLE 9-26

                   BREAKPOINT CHLORINATION DESIGN CRITERIA
                      FOR THE MONTGOMERY COUNTY FACILITY
 Average dry weather flow
 In-line mixers
   1 .   Lines normally in service
   2 .   Lines normally on standby
   3.   In-line mixers
        Size
        Mixers per line
        HP
        G
   4 .   Flow rate through mixer with 5 in service
        Average
        Maximum
        Peak instant
 Chemical feed
NaOCl
Feed rate @ 8% available
 service
Total feed rate per basin
                                and with 5 units in
       NaOHa
       Feed @ 20% NaOH with 5 units in service (total)
3.
Feed rates based on:
Influent NHj-N
C12 : NHj-N
Reaction basins and air mixing
1.
2.
3.

4.
Basins normally in service
Basins normally on standby
Basin dimensions
Volume
Theoretical mix time
       Diffuser air requirements
        Total
        Per basin
       Air headers and diffusers
        Main headers
        Cross headers
        Diffusers
                                                            60 mgd (2.63 m3/sec)

                                                      5, 3 ft dia. (91.4 cm) influent lines
                                                                     1

                                                               36 in. (91.4 cm)
                                                                2 (in series)
                                                   3 HP/Mixer (2.23 Kw), 6 HP/Line (4.46 Kw)
                                                                 1000 sec'1

                                                           12.81 mgd (0.56 m3/sec)
                                                           19.21 mgd (0.84 mVsec)
                                                           21 .34 mgd (0.93 mVsec)
     210,000 gpd @ 90 mgd (552 1/min @ 3.94 mVsec)
      32.5 gpm @ 100 mgd (123 1/min @ 4.38 m3/sec)
      30.0 gpm @ 90 mgd (113 1/min @ 3.94 m3/sec)
      20.0 gpm @ 60 mgd (75.7 1/min @ 2.63 mVsec)

                   15 mg/1 average
              52 mg/1 max. w/no alkalinity
   5500 gpd @ 90 mgd @ avg. (14.5 1/min @ 3.94 mVsec)
 21,000 gpd @ 100 mgd @ maximum (55 1/min @ 4.38 mVsec)

                      18.7 mg/1
                    10:1 by weight

                          5
                          1
     20 ft x 120 ft x 15 ft swd (6.1mx36.6mx4.6m)
                  36,000 gal (136 m3)
                  5 basins in service
                30 min  @ average flow
                  20 min @ max. flow
             30 scfm/1,000 cu ft (30.3 1/m3)
       5,400 scfm (150 m /min) - 5 basins in service
                1,080 scfm (30 mVmin)

          2 (one at each end) 8 in. dia. (20.3 cm)
              540  scfm normal (15 m3/min)
              1,080 scfm max. (30 m3/min)
  24, 2.5 in.  dia. (6.4 cm)  455 scfm (12.6 m3/min) normal
              90 scfm max. (2.5 mVmin)
 11/header, 264/basin, 2 ft (0.61 m) O.C. on cross headers,
4.1  scfm (0.11 m3/min) normal,  8.2  scfm max. (0.22 mVmin)
7 . Blowers
Normally in service
Normally on standby
Capacity
HP
8. Inlet diffuser
Diameter
Material
Length
Inlet ports

2
1
3,000 cfm (85 mVmin) @ 8 psi (0.60 kgf/cm2) ea.
150 HP each (112 Kw)

48 in. (121.9 cm)
FRP
42 ft 6 in. (12.95 m)
8 @ 3 ft 0 in. (0.9 m) O.C.
NaOH needed only when alkalinity is 150 mg/1 and no air stripping.

Basins covered to contain product gases and lined with corrosion protection membrane to 1 ft (30.5 cm) below water surface.
                                                 9-68

-------
The breakpoint process is  controlled by pacing the sodium hypochlorite feed rate to the
influent flow and influent ammonia nitrogen concentration. The pH is also monitored and
controls the addition, when necessary, of sodium hydroxide to maintain an optimum pH for
the breakpoint reaction.

Ammonia nitrogen in the breakpoint effluent is monitored to determine efficiency of the
process. Free and combined chlorine residuals and pH are also continuously monitored in
breakpoint reactor effluent.

The  estimated costs of the Montgomery County facility are shown in Table 9-27. The
hypochlorite generation  facility  will   also  provide hypochlorite  for uses other than
breakpoint although its entire cost has  been shown for the breakpoint process. As noted in
Table 9-27, the  hydrogen liberated in the  breakpoint process has a potential  value of
$10.00/mil gal ($0.0026/m3) if it is collected and used as a fuel.


                                  TABLE 9-27

              ESTIMATED COSTS OF BREAKPOINT CHLORINATION
                    AT THE MONTGOMERY COUNTY PLANT
  Capital
        a,b
     Breakpoint reaction basins
     Operations building (including off-gas
      treatment and blowers)
     Hypochlorite pla nt
       Storage
       Generation
       Salt dissolution
     Total
$2,300,000

   565,000

   817,000
 3,780,000
   705,000

$8,167,000
  Operation and maintenance
     Power
       Hypochlorite production
       Mixing, stripping
     Salt
     Labor
     Subtotal

     Amortized capital, $8,167,000, 20 years
      @ 7% @ 60 mgd
$   26.20/mil gal
    18.00
    25.20
    12.60

$   82.00  ($0.022/m3)
    35.20  ($0.009/m3)

$  177.20  ($0.031/m3)
   Not including contingencies and engineering
   December,  1974 cost levels
                                       9-69

-------
     9.5.4 Case Examples of Selective Ion Exchange for Nitrogen Removal

Two examples of the incorporation of ion exchange into wastewater treatment plant layouts
are presented  in this section. In the case example for Upper Occoquan Sewage Authority,
ion exchange  is  used in a tertiary treatment step  following biological treatment. The
objective of this plant is to meet an effluent limitation of 1 mg/1 total  nitrogen. In the
Rosemont case, ion exchange  is used in a physical-chemical flowsheet to meet an effluent
limitation of 1 mg/1 ammonia nitrogen.

         9.5.4.1 Upper Occoquan Sewage Authority, Va.

This new regional plant now under construction will replace 11 small secondary plants
which  discharge into  tributaries of a water supply reservoir which serves as the raw water
source for water treatment plants serving about 500,000 people in the Virginia suburbs of
Washington, D.C. Information about the  water reuse aspects of the project is  available in
reference 49. The effluent eventually reaches a water  supply reservoir in which nitrogen is
believed to be one of the principal eutrophication factors. The effluent standards are shown
below:

                   Parameter                                  Value

                   BOD5, mg/1                               1.0
                   COD, mg/1                               10.0
                   Suspended solids                           Unmeasurable
                   Phosphorous, mg/1                          0.1
                   Methylene Blue Active Substances
                   (MBAS), mg/1                              0.1
                   Turbidity, JTU                             0.4
                   Coliforms, total/100 ml                       2
                   Nitrogen, total                             1.0 mg/1
The main processes  which are included in this plant are shown in Figure 9-26.

The initial plant capacity will be 22.5 mgd (0.99 m^/sec), although an initial daily capacity
of 10.9 mgd (0.48 m3/sec) is all that will be initially certified by the State of Virginia so as
to provide 100 percent  complete backup facilities in the initial  operation. Provisions are
included in the Virginia  State Water Control Board regulatory policy to increase the rated
capacity to 15.0 mgd (0.66  m^/sec) after a one year satisfactory demonstration period.
Backup facilities would then constitute approximately 50 percent of the rated capacity.

The methods of nitrogen  removal of biological, ammonia  stripping,  and  breakpoint
chlorination  were evaluated  before  selection  of the selective ion exchange process. This
process was  selected primarily  because  of  its inherent reliability and efficiency and  the

                                        9-70

-------
                                   FIGURE 9-26

  FLOW DIAGRAM - UPPER OCCOQUAN SEWAGE AUTHORITY PLANT (VIRGINIA)
, 	 -1
1
LT-T
1 1
1
Preliminary 	 Primary ^
Treatment *" Settling

Plant
Effluent «
Reservoir
1
Discharge to
Bull Run

^ Sludge
Digestion
— Recycle 	 •
1
Activated Secondary L'm"
' Sludge fc Settling f * ^
1
1 	
Chlorination Selective
NHj Removal Exchange
i
Regenerant
Ns^""™
' Recovery

Coagulation Two-Stage
tiling for » Recarbonation
torus Removal
1
- Recycle 	 *
Ballast
Pond
i
Granular
Adsorption

Carbon
Regeneration
minimal effect on total  dissolved solids (TDS). The selective ion exchange process was
located after the  carbon  columns to take advantage of the available head remaining after
pumping through  the pressure carbon columns. Also, as an incidental benefit of this process
sequence, the clinoptilolite will serve as a final polishing filter to remove the small amount
of carbon fines or other suspended solids which may be present in the carbon column
effluent.

Carbon fines and  other solids trapped in the clinoptilolite bed are removed by backwashing
before the regeneration cycle. These backwash  wastes will be returned to the treatment
process, typically  to the chemical coagulation process where these solids will be removed in
the precipitation process and be trapped in the chemical sludge.

The selective ion exchange regenerant recovery was initially planned to be accomplished by
breakpoint chlorination of the ammonium using electrolytic cells (see Chapter 7). A delay in
the project of about one year occurred following  final design while awaiting the project
funding to develop. During this period,  the ammonia removal  and recovery process (ARRP)
described in Sections  7.3.3.2 and 8.4.1  was developed by the design engineers. Based on
pilot plant results, it was concluded that the annual operating cost for the plant could be
                                       9-71

-------
reduced by $375,000 at a flow of 15 mgd (0.66 m3/sec) ($0.07/1,000 gal or $0.18/m3). In
addition, the electrical energy requirement would be only 10 percent of the electrolytic cell
breakpoint chlorination process needs. Also, a byproduct would be obtained in the form of
ammonium  sulfate, a  common  chemical  fertilizer.^3 Based on  this information,  the
Authority authorized a redesign  of the  regeneration  facilities to incorporate the ARRP
process.  The  following paragraphs  describe the full-scale  ion  exchange and regenerant
recovery facilities.

Eight ion exchange beds operate in parallel  and are separated into two independent trains,
each with four beds and a common  manifold. Each bed is a horizontal steel pressure vessel,
10 ft (3.05 m) in diameter by 50 ft (15.24 m) long containing  a  four  ft  (1.2 m) deep bed
of  clinoptilolite (see  Figures 9-27  and   9-28).^4  Each  parallel train  is completely
independent,  including  piping, instrumentation  and  control, and  electrical supply.  In
addition, certain backup facilities are available in each train such as key instrumentation and
control.  Such measures were necessary  to  conform  to  the design  policy  for  reliability
established by the Virginia State Water Control Board.

Table 9-28 is a summary of the design criteria for the ion exchange process at  the future
anticipated rating of 15 mgd (0.66 m3/sec). The system will  be entirely  automated using
automatic valves in  a manner similar to most larger water treatment plant filtration facilities.
Regeneration  will  be initiated either on a run time  basis, volume throughput basis, or
manually. Backwashing will be done before each regeneration.  Backwash  water will be'
returned to the wastewater process,  typically to the chemical coagulation process, or to the
plant headworks.
                                     TABLE 9-28

    DESIGN CRITERIA SELECTIVE ION EXCHANGE PROCESS FOR AMMONIUM
             REMOVAL AT THE UPPER OCCOQUAN PLANT (VIRGINIA)
 Flow rate
 Beds in service
 Beds in regeneration
 Beds - backup capacity
 Flow per bed
 Bed loading rate

 Backwash rate
 Bed volumes to exhaustion
 Average ammonia removal efficiency
 Average influent ammonia nitrogen concentration
 Average effluent ammonia nitrogen concentration
 Normal concentration of ammonia nitrogen at initiation
  of regeneration
 Clinoptilolite size
 Clinoptilolite depth
  15 mgd (0.66 m3/sec)
         4
         2
         2
 3.75 mgd (0.16 mVsec)
  10.82 bed volume s/hr
5.25 gpm/sf (3.6 l/m2/sec)
 8 gpm/sf (5.4 l/m2/sec)
        145
        95%
       20 mg/1
       1 mg/1

       2.5 mg/1
    20 x 50 mesh
     4  ft (1.22 m)
                                         9-72

-------
                                    FIGURE 9-27
           PLAN AND SECTION OF ION EXCHANGE BEDS AT UPPER OCCOQUAN PLANT
SURFACE WAJSH SURFACE
f HEADER SUPPORT y-CLINO SURFACE
/" 'J
/ '

                        PVC SURFACE

                        , HEADER
                                             I I  I  I
                                 WASH
                                             i  I
                                                            QUICK OPENING MANHOLE
\
                                                                  ii i  i i  ,  i i  i  i
                                                                 : LATERAL SUPPORTS
-3T.
?r-E
                           18" WSP-
                                           PLAN
                               -£
                                                        •TYPICAL LATERAL - 49 REQ'D

                                                         PER CLINO BED
                                                               COMBINATION

                                                               AIR

                                                   18 WSP INFLUENT HEADER
1

H_ja~i 	 1 	 1 	 i 	 1 	 <^ 	 i 	 1 	 1_^ 	 =jt,
v SURFACE WASH HEADER !
n\8 8 8 8888888888888888888 8,8 8 8 ¥8 888
II lit Ie"
.11 « - 1 j 1
25'-0" J

WSP EFFLUENT HEADER t=^
1 Ik \/
^PIPE SU
                                                                                 SUPPORT

-------
                            FIGURE 9-28

    ADDED DETAILS - ION EXCHANGE BEDS AT UPPER OCCOQUAN PLANT
3" PVC
8 HOLES-7 SPACES @ 5'/2"
8-'/4" OIA. HOLES IN LATERAL
FACING DIRECTLY DOWNWARD
                                       3" PVC
                                       THREADED TEE
                                        • — 3" EXTRA HEAVY
                               STEEL PIPE

                              I8"WSP

                              SYMMETRICAL  ABOUT
             TYPICAL  UNDERDRAIN  LATERAL
Z ROWS OF l^2" OIA. HOLES
@, 6 l/2" O.C. FOR INFLUENT
DISTRIBUTION, EACH ROW 30°
OFF VERTICAL, 86 HOLES EACH
SIDE , 172 HOLES TOTAL
                            18  WSP
SUPPORTING HIGH
DENSITY GRAVEL
     GRADED GRAVEL
                                  QUICK OPENING
                                  MANHOLES
                                    10* DIA. STEEL
                                    CLINO BED

                                    UNDERDRAIN LATERAL
                                            CONCRETE  FILL
                               9-74

-------
The beds will be regenerated with a 2 percent sodium chloride solution. The regeneration
process  will be  that  shown in  Figure  7-13  and as described  in Section 7.3.3.2. The
regenerant recovery system consists of four 375,000 gallon (1420 m3) tanks, a regenerant
pumping system  and associated automatic valves, two 35-ft (10.7 m) diameter clarifiers for
magnesium hydroxide removal, and 18 ARRP modules. Figures 9-29 and 9-30 illustrate the
ARRP module design. The ARRP units will be shop  fabricated. The basic tower units are
12-ft (3.66 m) diameter fiberglass tanks. All materials will be fiberglass or PVC. Each tower
has a  25 HP (18.6  kw) fan. The air rate is approximately 34,000 cfm/tower (952 m3/min)
at an  air to liquid ratio of 566 ft3/gal (4.15 m3/!). Tower air velocities are 300 fpm (91.4
m/min). Knitted mesh mist eliminators prevent moisture carryover from  tower to tower.
The total  system head loss is about 2.5-3.0 inches (6.4 cm -  7.6 cm) of which about 1.5
inches (3.8 cm) is in the media. The media is 2-inch (5.1 cm) diameter polypropylene plastic
packing (Tellerette). A summary of the regeneration system design  criteria is shown in Table
9-29.

                                   FIGURE 9-29

           PLAN VIEW OF ARRP MODULE - UPPER OCCOQUAN PLANT
  BOTTOM ACCESS
  HATCH
    4'-6" DIA
    FRP DUCTING (TYP)
                                                                       LADDER
                                                                       SUPPORT
                                                                       PLATFORM
                                                                       ABOVE
                                                                      SEE
                                                                      FIGURE 9-30
                                                                      FOR
                                                                      SECTION A-A
12' DIA. FRP
TOWER (TYP)
                                       9-75

-------
                   FIGURE 9-30
SECTION OF ARRP MODULE - UPPER OCCOQUAN PLANT
                                 DUCT
                                 SUPPORTS
                                 TYP
             KNITTED MESH
             MIST ELIMINATOR
                            V /  HATCHES
                             SERVICE AND
                             SUPPORT
                             PLATFORMS
         ACCESS HATCH
         BEYOND
                                PIPING DUCT
                                BEYOND
                    REGENERATION
                    BASIN  BELOW
                       9-76

-------
                                    TABLE 9-29

        REGENERATION AND REGENERANT RECOVERY SYSTEM DESIGN
            CRITERIA AT THE UPPER OCCOQUAN PLANT (VIRGINIA)
 Regeneration system
   Number of regenerant tanks
   Size of each tank
   Number of beds regenerated at once
   Number of regeneration cycles per day
   Regeneration bed volumes
 Regenerant recovery system3
   Recovery system flow rate
   Operation time per day
   Clarifiers
     Number of units
     Diameter
     Overflow rate
 Ammonia removal and recovery process3
   Number of ARRP modules
   Liquid loading rate
   Air to liquid loading rate
   Media height
   Removal efficiency at 10  C
                  at 20  C
  375,000 gals (1420 mj)
          2
         3.58
         39-44

   1 ,080 gpm (68 I/sec)
         16 hr
     35 ft (10.66 m)
800 gpd/sf (32.6 m3/m2/day)

          18
760 gpd/sf (31.1 m3/m2/day)
  566 cf/gal(4234 m3/m3)
     7.S ft (2.29 m)
          90%
          95%
  At 15 mgd flow rate


The average  ammonia-nitrogen concentration from ion exchange beds will be about one
mg/1. The  organic  nitrogen is expected  to  be 0.5-0.8 mg/1 and the  nitrate nitrogen  is
expected to be from 0.1-0.2 mg/1. Thus, the total nitrogen leaving the ion exchange process
will be  about 1.6-2.0 mg/1. Since the discharge standard is  1.0  mg/1, additional  nitrogen
removal  is necessary. This will be  accomplished by  breakpoint chlorination of the ion
exchange effluent.  A dosage  of approximately 8-10  mg/1  will result in nearly complete
removal  of ammonia nitrogen. The final effluent is then expected to have a total  nitrogen
concentration of less than 1 mg/1.

The estimated costs are shown in Table 9-30. Since the initial constructed capacity of 22.5
mgd (0.99 m^/sec) may be operated at no more than 15 mgd (0.66 m^/sec) because of state
requirements for backup  capacity, costs are shown on the basis of operation of two-thirds of
the constructed capacity and at the full constructed  capacity, which would be the  more
generally applicable circumstance. The income from sale of ammonium sulfate is based on
the lowest wholesale price in effect at the time of this writing,$43/ton ($39/metric ton).
In some areas, the wholesale price is as high as $65/ton ($59/metric ton).

          9.5.4.2 Rosemount, Minnesota

This new 0.6 mgd (2271  cu.m/day) plant operated by  the Metropolitan Sewer Board of the
Twin Cities area (Minneapolis-St. Paul) provides independent physical-chemical treatment of
a municipal wastewater.55,56 This  plant is the first full-scale physical-chemical plant  to be
                                         3-77

-------
                                TABLE 9-30

              ESTIMATED COSTS OF SELECTIVE ION EXCHANGE
                AT THE UPPER OCCOQUAN PLANT (VIRGINIA)
Item
a
Operating and maintenance
Chemicals
NaOH
NaCl
H2SO4
Income from sale of (NH^ SO4 @ $43/ton
Net chemical cost
Power, 18 HP/mil gal @ $0.0192/kwh
Labor
Total, O & M
Capital3
$4,470,000, 20 years @ 7%
Total annual costa
Estimated costs, $/mil gal
at 15 mgd
(0.66m3/sec)


$ 26.80
7.10
9.80
$ 43.70
$(12.60)
31.10
6.90
17.70
$ 55.70

$ 77.22
$132.92
($0.035/m3)
at 22.5 mgd
(0.99 m3/sec)


$ 26.80
7.10
9.80
$ 43.70
$ (12.60)
31.10
6.90
17.70
$ 55.70

$ 51.59
$107.29
($0.027/m3)
  August, 1974 costs

placed in operation in the U.S. A schematic of the process is shown in Figure 9-31. The
entire plant is enclosed in a single  14,500 sq ft (1347 m^) steel building. Final effluent
standards are as follows:
                 Parameter
                      , mg/1
                 Suspended solids, mg/1
                 COD, mg/1
                 Ammonia — Nitrogen, mg/1
                 Phosphorus, total (P), mg/1
                 PH
Value

 10
 10
 10
  1
  1
8.5
                                    9-78

-------
                                   FIGURE 9-31
                SCHEMATIC OF ROSEMOUNT, MINNESOTA PLANT
         CHEMICALS
RAW
WASTEWATER
               SOLIDS
               CONTACT
              CLARIFIER
 DUAL
 MEDIA
FILTER
   BAR
   SCREEN
                    SETTLED
                    SOLIDS
                                           <  a:
DU
MED
FIL
1
AL
1 A
TER
I





I
AM
EX(
               SLUDGE
            CONCENTRATOR
  SLUDGE
  DISPOSAL
     SPENT
     CARBON
REGENER-
ATED
CARBON
STORAGE
Selective ion exchange is accomplished by clinoptilolite  in 2 of 3 downflow columns in
series, each containing a 6-ft (1.82 m) depth of clinoptilolite. When the ammonia-nitrogen
reaches  1 mg/1 in the effluent from the polishing column, the lead column is removed from
service and regenerated (upflow)  and the third column placed on line. The steam process
discussed in Section 7.3.3.3 is used for clinoptilolite regeneration and ammonia recovery.
Brine is stored at 77 C and is cooled to 27 C while passing through a heat exchanger on the
                                      9-79

-------
way to the column. The waste brine leaving the column is passed through the other side of
the heat exchanger to elevate its temperature to 71 C before entering the stripping process.
Brine temperature in the storage tanks is controlled by steam supplied to internally mounted
coils. Waste  brine is  collected  in a mixed storage tank before being stripped.  Soda ash is
added to the storage  tank to elevate  the  pH  to 12, which results in the precipitation of
calcium  carbonate and magnesium hydroxide.  Mixing is discontinued 20 minutes after  the
pH has reached 12 to allow these precipitates to settle.  The sludge is then pumped from  the
bottom  of the waste  brine  storage tank.  The waste brine is then pumped to the  steam
stripper at a  rate of 53 gpm (3.34 I/sec) and a steam flow of 3000  Ib/hr (1498  kg/hr). The
stripper  operates at an equilibrium temperature  of 104 C and the  condensor at 38 C. The
time  required for a  stripping/reclaiming  operation is about  5  hr. Design  criteria  are
summarized  in Table  9-31. The plant  was in the start-up phase at  the time  of  this writing
and no operating data were available.

                                      TABLE 9-31

          ROSEMOUNT (MINNESOTA) ION EXCHANGE DESIGN CRITERIA
  Ammonium exchange columns (Two trains of 3)
     Loading rate8
     Clinoptilolite capacity
       per unit volume
       per column
     Ammonia nitrogen loading rate
     Ammonia removal
     Clinoptilolite depth per column
     Clinoptilolite size
     Normal operation
     Backwash rate
  Regeneration system
     Brine  solution to columns'3
       Hydraulic application rate
       Volume
       Strength
       Temperature
       PHC
     Brine  solution regeneration
       Regeneration cycle length
       Hydraulic loading rate to steam stripping tower
       Tower depth
       Caustic soda added
     Bed rinse3
       Rinse rate
       Time
     Ammonia recovery
       Aqueous ammonia strength
       Aqueous ammonia volume
     Ammonia stripper
       Steam @ 10 psig
       Throughput
       Size - diameter
            height
4.2 gpm/sf (2.85 l/m2/sec), 5.6 BVAr

      0.3 Ib/cu ft (4.86 kg/m3)
          90 Ib (40.8 kg)
      50 Ib/day (22.7 kg/day)
               95%
           6 ft (1.83 m)
           20 x 50 mesh
  2 columns in series, 250 BV/cycle
      8 gpm/sf (5.43 l/m2/sec)
     2.0 gpm/sf (1 .36 l/m2/sec)
              4.5 BV
             6% NaCl
              71 C
                11

              5 hr
      7 gpm/sf (4.75 l/m2/sec)
           24 ft (7.3 m)
          3 Ib/lb NH^-N

       300 gpm (18.9 I/sec)
             70  min

                1%
       1,000 gpd  (3,785 Ipd)

     3,300 IbAr (1,498 kg/hr)
       53 gpm (3.34 I/sec)
           3 ft (0.91 m)
           18 ft (5.5 m)
   Downflow
   Upflow
   Elevated with caustic soda
                                          9-80

-------
     9.5.5 Case Examples of Air Stripping for Nitrogen Removal

Two plants  which utilize air  stripping for nitrogen removal are described in this section.
Both of these, South Lake Tahoe and Orange County, use the biological-tertiary approach in
which ammonia stripping is used after biological treatment. At Tahoe, nitrogen removal is
incorporated on an experimental basis, as no nitrogen removal requirement exists. At the
Orange County Water District plant, nitrogen is removed to permit recharge of groundwater.

         9.5.5.1 South Lake Tahoe, California

At South Tahoe an experimental full-scale ammonia stripping tower built to handle half of
the total plant  design flow of 7.5 mgd (0.33 m^/sec) has been operated on an intermittent
basis since  1969.13,57,58,59  jhe piant flow sheet is shown  in Figure 9-3. The stripping
process  was installed  at South Lake Tahoe  as an EPA research  and demonstration
installation .and not as the  result of any requirement to remove nitrogen from the plant
effluent. In  the absence of any need for full time operation, the purposes of this stripping
tower have included: (1)  demonstration of full-scale tower efficiencies as compared to pilot
plant  test results; (2) determination  of cold  weather  operating limitations and other
operating problems with investigation  of solutions to these problems and (3) collection of
data for design  purposes for future expansion of this process to full plant design capacity as
well as for use in planning similar facilities at other locations. The large tower capability to
remove ammonia almost exactly  duplicates the results of pilot plant operation, reaching 95
percent removal in warm weather. The  cold weather operating limitations and recommended
tower design improvements have  been determined as discussed in Chapter 8. The new design
criteria  obtained  from operating the  tower have already been used to prepare plans for
stripping towers to treat wastewaters  at Orange County, California*^ and the operating
experience has also provided the direction and the basis for expansion of the Tahoe nitrogen
removal facilities to meet anticipated future requirements for nitrogen removal. As discussed
later in this  section and in Section 8.5, the  original packed tower system has recently been
modified to  provide year round, full-scale nitrogen removal.

The design data for the original packed tower  are given in Table 9-32.  It  is a crossflow
cooling type tower modified for ammonia stripping. The overall dimensions of the tower are
32 ft (9.75 m) x 64 ft  (19.5 m)  x 47 ft (14.3 m) high. Water, at pH 11, was pumped to the
top of the  tower  by  either  or  both of two  constant-speed pumps.  These pumps were
backflushed  two or three times daily to minimize buildup of calcium carbonate scale in the
pump units.  When the plant inflow was less than the rate at which the pumps were delivering
water to the tower, some water was recycled from the tower effluent back to the pump
suction well. This avoided the need for variable  speed pump control, and at the same time
provided some .recirculation through the tower,  which improved ammonia removal. At the
top of the tower, the influent water entered a covered distribution box  and overflowed to a
distribution  basin.  The distribution basin is a flat deck with a  series  of holes fitted with
plastic  nozzles.  Further  distribution  of the inflow was  provided by  diffusion decks
                                        9-81

-------
immediately below the distribution basin. Three other diffusion decks were provided at 6 ft
(1.82 m) vertical intervals in the fill. The tower fill provided, theoretically, 215 successive
droplet formations as the water passed down through the tower. The tower effluent fell into
a concrete collection basin  which also formed the base for the tower structure. From the
collection basin, the tower effluent passed through a Parshall measuring flume into the
first-stage recarbonation chamber, where excess pumpage returned through a flap gate into
the tower pump sump to recirculate through the tower.
                                 TABLE 9-32

                 DESIGN DATA, AMMONIA STRIPPING TOWER
                    AT SOUTH LAKE TAHOE, CALIFORNIA


 Nominal capacity:  3.75 mgd (0.164 m3/sec)
 Type:
 Fill:
 Air flow:
Cross flow with central air plenum and vertical air discharge
through fan cylinder at top of tower

Plan area, 900  sq ft (83.6 m2)
Height, 24 ft (7.3 m)
Splash bars:
   Material, rough  sawn treated hemlock,size,
   3/8 in. x 1-1/2  in. (0.95 cm x 3.8 cm)
   spacing, vertical 1.33 in. (3.37 cm)
            horizontal 2 in. (5.08 cm)

Fan, two-speed, reversible, 24-ft diameter, horizontal
Water
gpm
(I/sec)
1,350
(85)
1,800
(114)
2,700
(170)
rate
gpm/sf
(l/m2/sec)
1.0
(0.68)
2.0
(1.36)
3.0
(2.04)
Air
cfm
(mVmin)
750,000
(21,000)
700,000
(19,600)
625,000
(17,500)
rate
cf/gal
(m3/m3)
550
(4,115)
390
(2,918)
230
(1,720)
 Tower structure:
 redwood
 Tower enclosure:    corrugated cement asbestos

 Air pressure drop:   1/2 in. of water at 1 gpm/sf
                    (1.27 cm @ 0.68 l/m2/sec)
                                     9-82

-------
Air entered the tower through side louvers, passed horizontally through the tower fill and
drift eliminators (or air flow equalizers), and entered a central plenum. At the top center of
the plenum is a  24  ft  (7.3 m) diameter, six-bladed, horizontal fan. Fan blades and fan
cylinder are both made of glass-reinforced polyester. The fan takes suction from the plenum
and discharges to the atmosphere through the fan cylinder. The fan has a maximum capacity
of about 750,000 cfm (21,000 m^/min). It is equipped  with a two-speed reversible 75 HP
(56 kw) motor.

The limitations of the packed tower at Tahoe are caused by the cold winter temperatures
and the scaling of the tower packing. Because the scaling problem was not anticipated from
the pilot studies, the access to the tower packing needed to remove the scale with water jets
was not provided. As a result, portions of the original hemlock packing became hopelessly
fouled with calcium carbonate scale, thereby decreasing effective tower packing area.
The costs of the operation of the packed tower at Tahoe at a 2 gpm/sq ft (1.36 l/
rate as estimated for a 7.5 mgd (0.33 m^/sec) scale for continuous operation are  listed in
Table 9-33. Operating  labor included  backflushing  of tower  pumps and cleaning the
distribution deck to remove CaCC>3 precipitation, process inspection, lubrication, and daily
determination  of tower ammonia removal efficiency. Maintenance costs reflect the cost of
cleaning tower fill.

Because of the cold weather limitations of packed tower inherent to the Tahoe location and
the reduced efficiency of the existing tower due to scaling of the tower packing, a research
program  was  initiated  at  the  South  Lake Tahoe  plant to develop alternate,  low-cost
techniques which could  be applied  to the full-scale plant while using as much of the existing
facilities  as possible. Although  there are  no current regulatory agency requirements for
nitrogen removal for exported wastes, future requirements for effluent reuse or disposal in
the Lake  Tahoe  basin will probably include nitrogen removal.  At the current time, the
States of  California and  Nevada require that all  wastewaters, regardless of the degree of
treatment be exported from the Lake Tahoe basin. The South Tahoe effluent is exported to
Alpine County, California where it is used to form the 1 billion gallon (3.785 billion liters)
Indian Creek Reservoir.59 An excellent trout fishery has been established in this recreation
reservoir and  it is necessary to control pH and  ammonia  concentrations to prevent fish
toxicity. Thus, anticipated future regulatory agency requirements and current effluent reuse
practices required that a  system for full-scale nitrogen removal be developed for the South
Lake Tahoe plant.

Inspired by observations of ammonia release from holding ponds in Israel," 1 research was
undertaken  at Tahoe to  improve the release of ammonia from the high pH effluent. The
results led to the following three steps being applied to a full-scale modification of the
              69
Tahoe system.

     1 .    Holding in high pH ponds (with surface agitation in one pond).
                                        9-83

-------
     2.   Stripping in a modified crossflow forced draft tower through air sprays installed
         in the tower.

     3.   Breakpoint chlorination.

Design data are given in Table 9-34 and the entire system is depicted in Figure 8-9. Because
ammonia removals by stripping vary so much with temperature, and since low temperatures,
high ammonia  concentrations, and high flows never occur simultaneously at this location,
design data are presented for two sets of conditions which are expected to occur: low flow
(2.5 mgd or 0.11 m^/sec), low water temperature (3 C), and low ammonia content; and high
flow (7.5 mgd or 0.33 m^/sec), high water temperature (22 C), and high ammonia content.
                                 TABLE 9-33

               OPERATING COSTS FOR AMMONIA STRIPPING FOR
    CONTINUOUS OPERATION OF TAHOE AIR STRIPPING TOWER AT 7.5 MGD
Operating cost per day
Electricityb
Operating labor
Maintenance labor
Repair material
Instrument maintenance
Total operating cost
Total cost per mil gala
Operating
Capital
Total
$/Dgy
60.78
4.63
5.17
.78
.94
72.30
$/mil gal
9.64
8.00
17.64
  1970 dollars

  Average cost per day at 7.5 mgd from 10 months of continuous operation, i.e.,
  May, 1969, through September, 1970.
                                     9-84

-------
                                  TABLE 9-34
             DESIGN DATA AND ESTIMATED NITROGEN REMOVALS
   FOR ALL-WEATHER AMMONIA STRIPPING AT SOUTH TAHOE, CALIFORNIA




Description


Influent, pH = 11.0
Holding ponds
Detention = 7 hr
Detention = 18 hr
Air spraying in second pond
Turnovers = 4-j
Turnovers = 13|
Stripping tower, air spraying with
forced draft
Recycle turnovers = 1.6
Recycle turnovers = 5
Overall removal
NH3~N remaining to be removed by
by breakpoint chlorination, mg/1
Chlorine required for breakpoint
chlorination, Ib/day
Flow 2 . 5 mgd
(0.11 m3/sec)
Water Temp. 3 C
NH3-N
Estimated
reduction,
percent



10


30



42
64




Remaining,
mg/1

15


13.5


9.5



5.4


5.4

1,130
Flow 7 . 5 mgd
(0.33 mVsec)
Water Temp. 22 C
NH3-N
Estimated
reduction,
percent


15


28



23

40




Remaining,
mg/1

35 •

30



21.5


16



16

10,000
It is difficult to predict accurately the performance of the full-scale ponds and natural
wind-vented  sprays  from the results  of laboratory  and  pilot plant data, because the
efficiency of ammonia removal by  these  methods  depends  to  a great extent on the
immediate removal of the released ammonia gas from the vicinity of the air-water interface.
The  efficiency of air sweeping in removing ammonia declines as the area of the ponds
increases. In the  absence of full-scale plant data, the removal efficiencies obtained in the
laboratory and pilot plant*>2 were heavily discounted in estimating the performance of the
full-scale plant shown in Table 9-34. The following paragraphs describe the modified Tahoe
stripping process.
                                      9-85

-------
The two high pH ponds also serve to equalize flow to the modified stripping tower and
breakpoint facilities. This reduces design capacity requirements for these two processes and
improves their operating characteristics. The total surface area of these ponds is about
64,000  sq  ft  (5946 m2). The  water depth varies from 3.0 to  7.25 ft (0.9 m  - 2.2 m).
Detention time ranges from 7 hr, at 7.5  mgd (0.33 m^/sec) to 18 hr at 2.5 mgd (0.11
m^/sec). The second pond in the series is provided with surface agitation in order to increase
the ammonia removal. The system of surface agitation consists of sprinkling with 34 mgd
(1.5 m^/sec) of recycled pond  water through 2  pumps and a spray system consisting of
about 75 vertical nozzles each delivering about 320 gpm (20.1 I/sec). At the 2.5 mgd (0.11
m^/sec) plant  flow, the spray system recycles the pond water 13 times, and at the 7.5 mgd
(0.33 m-Vsec)  flow, 4 recycles occur. The spray nozzles are 4 in.  x 2^ in. (10.2 cm x 6.35
cm) female  pipe reducers each fitted with a 2l/2 in. x \Vz in. (6.35 cm x 3.8 cm) bushing.
Each spray orofice is about 1-7/8 in. (4.8 cm) in diameter.  Nozzles having interval vanes or
other obstructions introduce air containing carbon dioxide  to the water in the nozzle. This
causes deposition and rapid buildup of calcium carbonate within the nozzle. Such nozzles
are unsatisfactory  because of the resulting plugging and flow restriction problems. Only
nozzles with unobstructed clear opening, and without internal vanes, should be used.

A  major modification has been made to  the existing stripping tower. The existing packing
has been removed, and the entire area of the tower equipped with water sprays. The existing
trays at the  top of the tower distribute part of the flow, and 4 nozzle equipped headers in
the bottom  of the  tower spray water upward into the tower. The pump capacity to the
tower is 11.8 mgd (0.52 m^/sec). At cold weather plant flow  rates of 2.5 mgd (0.11
m^/sec), this flow  will provide  a recycle rate of 5 through the sprays in the tower. The
capacity and type of nozzle used in the  tower is similar to the nozzles used in the ponds.
Based on plant scale tests of this spray system, with the induced draft fan operating at high
speed, it is anticipated that at least 42 percent of the ammonia in the pond effluent will be
removed in the tower under cold weather operating conditions.

Chlorine may be added at  two points  in  the process (Figure 8-9).  The first  point of
application is in  the primary recarbonation chamber at a pH of 11.0. Only enough chlorine
is  added to  reduce  the pH to about 9.6, thus eliminating the need for addition of carbon
dioxide (CO2) at this  point. About 65 mg/1 is required to reduce the pH from  11.0 to 9.6.
The balance of the  chlorine needed to reach the breakpoint for complete ammonia removal
is  added in a chamber immediately downstream from the secondary recarbonation chamber.
At times sufficient chlorine is added at this point to reduce  the pH to 7.0 or less, so that no
CO2 will be required. A dose of approximately 160 mg/1 of chlorine is required  to reduce
the pH from 11.0 to 7.0. When the breakpoint is reached with a lesser dose of chlorine, then
some CO2 must be  added in the secondary  stage of recarbonation to produce a pH of 7.0.
About 10 mg/1 of chlorine is required for each mg/1 of NIfy —N to reach breakpoint. After
breakpoint chlorination treatment, water from the ballast pond is pumped to  the existing
filters and carbon columns. The carbon columns remove any excess chlorine. The modified
Tahoe process was being placed in operation at the time of this writing and full-scale data
are not yet available.

                                       9-86

-------
          9.5.5.2 Orange County Water District, California

The Orange County Water District (OCWD) at Santa Ana, California has under construction
a 15 mgd (0.66 m-Vsec) wastewater reclamation plant and a 3 mgd (0.13 m^/sec) seawater
desalting  plant at the same  site.60 The water from the two plants will be  blended  and
pumped into a line of injection wells which are located on the land side of a line of seawater
pumping  wells to  form a barrier against seawater  intrusion into the fresh water aquifer.63
The injection system also serves to replenish the  supply of groundwater available for  use.
The OCWD  water reclamation plant will take 15 mgd (0.66  m^/sec) of trickling filter
effluent from the  secondary  treatment plant of the Orange County Sanitation District. It
will be subjected to high lime treatment at pH of 11.0 and clarification in a basin equipped
with settling tubes.  The clarified  high pH water will be  pumped to two countercurrent
ammonia  stripping towers. In the  climate at this location, freezing temperatures are  not
experienced, and waste heat from the desalting plant operation will be used to heat the inlet
air  to the stripping towers for increased ammonia removal efficiency. The design also fully
realizes scaling problems encountered elsewhere  and  incorporates provisions for scale
control even though scaling was not a problem in pilot tests at this site.

The plant is designed with two ammonia stripping/cooling towers each equipped with six 18
ft (5.49 m) diameter fans. An end section view of a tower is  shown in Figure 9-32 and an
overall view  is shown  in Figure  9-33.  The stripping sections are  designed for a hydraulic
loading of 1.0 gpm/ft2 (0.68  l/m2/sec) and an air flow of 400 cu ft/gal (2,990 m3/m3).  The
cooling sections are designed  to cool the desalting process water and brines from 46-49 C to
27-29 C and will raise the air temperature to the  stripping section to 31-33 C.  Splash-bar
packing will be used for ammonia stripping and a film packing located in the air inlet plenum
will be used for cooling.

A prime  design  criterion  was that  the  ammonia stripping packing be accessible  and
removable for cleaning because scaling of the packing might reduce air flow and ammonia
removal efficiency. Provisions have been made to feed a scale inhibiting polymer, if needed,
to  the  tower influent.  The warm  saturated air exhausted from the cooling sections
theoretically would permit the tower packing to be less than the design height of 25 ft (7.6
m). Because during the first few  years of operations the desalting  plant will operate on an
intermittent basis,  it was not considered prudent to reduce the packing depth.

The tower fill or packing is made from l/i in. (1.27 cm) diameter  Schedule 80 PVC pipe  at 3
in. (7.6 cm) centers horizontally and with alternate layers placed at right angles and at 2 in.
(5.1 cm)  centers vertically (see Figure 9-34). The fill was factory prefabricated in modules
which are about 6 ft (1.82 m) by 6 ft (1.82 m)  by 4 ft (1.22 m) high.  Each module is
supported within its  own steel or fiberglass frame, so that it is easily removable if necessary
for cleaning. An overhead hoist and moveable dolly are provided to assist in packing removal
for cleaning. However, the access corridors and the removable air baffle panels in the tower
should make it possible to reach all  of the fill in  its normal operating position  within the
tower for hosing down to remove  any excessive calcium carbonate scale which may form.

                                        9-87

-------
Based on the results of extensive pilot plant tests at  the OCWD, it is expected  that the
towers will  remove more than 90 percent of the ammonia from  the wastewater. The
ammonia remaining in the tower effluent will be removed by breakpoint chlorination using
about 10 mg/1 of chlorine for each mg/1 of ammonia nitrogen still present. No difficulty is
anticipated  in meeting the limits set by the regulatory agency  of 1.0 mg/1 of ammonia-
nitrogen in  the injection  water. The Orange County design offers an example of a design
with adequate scale control provisons incorporated.

                                    FIGURE 9-32
           ORANGE CO. AMMONIA STRIPPING/COOLING TOWER SECTION
                                           AIR OUTLET
              2 SPEED MOTOR
                                            FAN
                                               :\
                                                FAN  STACK

                                                 WATER INLET FLOW
                                                 CONTROL  VALVE
i j
s
K>
.1
' ]

<: — r

* I !
1 •
>!

^
-WHI


r
t
j
i
o


,4if 3

J
J


'•
I—I .
r- V
AIR
el


.
OUT
!_„

_ll_


_JU
nn
Hi
H
_^AIR IN



fs"
i
1 V\
» HOIST^
INTERIOR
ACCESS
AREA^
>
AMMONIA
REMOVAL
FILL
DOLLY-j
r—L

S
1

r
a.--

T
i
1
1
1
1
1
t
1
1
1
I
j
i
i
^WASTE
DRAIN

j
XWATER
DRIFT
ELIMINATOR -j
o1 '

r ^
13" NOZZLE'
WATER




•-•- -!-.-
IN




-"


!— '-—
WATER 0
UT
R 	 f-
i
I
., 	



j
^-AMMONIA
REMOVAL
FILL BUNDLES
^-EXTERIOR
ACCESS
AREA
\\ AIR
i'i i''i "' V 1
     6
POOLING  FILL
 AND AIR  INLET
WASTEWATER-
COLLECTION
CHANNEL
                                                           COOLED PROCESS
                                                           WATER OR  BRINE
                                                           COLLECTION CHANNEL
                                       9-88

-------
The estimated costs (1974 costs) for the Orange County ammonia stripping towers are as
follows:
                           Capital         $42/mil gal
                           Operating       $29/mil gal
                             Total        $71/mil gal

These costs are substantially higher than those reported for the Tahoe facility. The difference
results from the following factors: Orange County towers were designed at 1 gpm/ft^ (0.68
1/m^/sec) loading rate while Tahoe towers were designed  at 2 gpm/ft^ (1.36 l/m^/sec);
extensive provisions for  tower cleaning were included at Orange County; heat exchange
facilities were included in the Orange County design but no credit taken in the tower design;
Orange County costs were for 1974 vs the 1970 basis for the Tahoe costs.
                                FIGURE 9-33

     OVERALL VIEW OF THE ORANGE COUNTY WATER DISTRICT (CALIF.)
   (DESALTING PLANT IS IN RIGHT BACKGROUND, CHEMICAL CLARIFIER IN
     RIGHT FOREGROUND. NOTE WALLS AT BASE OF TOWERS TO SHIELD
                           NOISE AT AIR INLET.)
                                    9-89

-------
               FIGURE 9-34

  STRIPPING TOWER PACKING MODULE AT THE
ORANGE COUNTY WATER DISTRICT PLANT (CALIF.)
                    9-90

-------
9.6 References

 1.  Horstkotte, G.A., Niles, D.G., Parker, D.S., and D.H. Caldwell, Full-Scale Testing of a
     Water Reclamation System. JWPCF, 46, No. 1, pp 181-197 (1974).

 2.  Mulbarger, M.C., The  Three Sludge System  for Nitrogen  and Phosphorus Removal.
     Presented at  the 44th Annual Conference of the Water Pollution Control Federation,
     San Francisco, California, October, 1971.

 3.  Description of the El Lago, Texas, Advanced Wastewater Treatment Plant.  Harris
     County Water Control and Improvement District Number 50, Seabrook, Texas, March,
     1974.

 4.  Bishop,  D.F.,  O'Farrell,  T.P.,  Cassel,  A.F., and A.P. Pinto, Physical-Chemical
     Treatment  of Raw Municipal  Wastewater. Report  prepared  for the Environmental
     Protection Agency, Contract Number 14-12-818, February,  1973.

 5.  Koon, J.H., and W.J. Kaufman, Optimization of Ammonia Removal by Ion Exchange
     Using Clinoptilolite.   Environmental Protection  Agency Water Pollution  Control
     Research.Series No. 1708009/71, 1971.

 6.  Bishop, D.F., Schuk, W., Yarrington, R., Bowers, J.F., Fein, E.D., and H.W. Treupel,
    Physical-Chemical Wastewater Treatment Under Direct Digital Control.  Presented at
     the International Workshop: Instrumentation, Control and Automation for Wastewater
     Treatment Plants, London, September 15-20, 1973.

 7.  Barnes,  R.A.,  Atkins,  P.F.,  Jr.,  and  D.A.  Scherger, Ammonia  Removal  in  a
     Physical-Chemical Wastewater Treatment Process:  Prepared for the EPA, Report No.
     R2-72-123, November,  1972.

 8.  Brown and Caldwell, Report on Tertiary Treatment Pilot Plant Studies.  Prepared for
     the City of Sunnyvale, California, February, 1975.

 9. Process Design  Manual for Phosphorus Removal.  U.S. EPA,  Office of Technology
     Transfer, Washington, D.C. (1971).

10.  Process Design  Manual for  Carbon  Adsorption.   U.S. EPA,  Office of Technology
     Transfer, Washington,  D.C.  (1973).

11.  Brown and Caldwell, Consulting Engineers, Lime Use in Wastewater Treatment. Report
     submitted to the U.S. Environmental Protection Agency, 1975.

12.  Culp, G.L., Wesner, G.M., and R.L. Gulp, 7974 Lake Tahoe Advanced Wastewater
     Treatment  Seminar Manual. Clean Water Consultants, El Dorado Hills,  California,
     1974.
                                       9-91

-------
13.  South Tahoe Public Utility District, Advanced Waste Treatment as Practiced at South
    Tahoe. EPA-WPCRS 17010ELQ8/71, August, 1971.

14.  Brown and  Caldwell, Lime Sludge Recycling Study. Report to the Central  Contra
    Costa Sanitary District, June, 1974.

15. Battelle Memorial Institute, Pacific Northwest Laboratories, Evaluation of Municipal
    Sewage  Treatment Alternatives. Prepared  for the Council on Environmental Quality
    and the Environmental Protection Agency, Contract EQC 316, February, 1974.

16.  Greene, R.A., Complete Nitrification by Single Stage Activated Sludge. Presented at
    the 46th Annual Conference  of the Water Pollution Control Federation, Cleveland,
    Ohio, October, 1973.

17.  McNamee, Porter and Seeley, Consulting Engineers, Plans for Sewage Treatment Plant
    Additions,  City  of Jackson,  Michigan. Submitted to the Michigan Water Resources
    Commission, October, 1970.

18.  Greene, R.A., Personal Communication to  Parker, D.S., City of Jackson, Michigan,
    September,  1974.

19.  Brown and  Caldwell, Project Report  for the  Wastewater Treatment Plant Stage 3
    Improvements. Valley Community Services District, September, 1972.

20.  Valley Community Services District, Wastewater Treatment Plant Operating Reports.

21.  Johnson, James L.,  Personal  Communication to G.A.  Carthew. Valley Community
    Services District, November, 1974.

22.  Environmental Quality Analysts, Inc., Letter Report to Valley Community Services
    District, March, 1974.

23.  Brown and Caldwell, Contract for Treatment Plant Enlargement — 1965. Prepared for
    the City of Livermore, December, 1965.

24.  City of Livermore, California,  Water Reclamation Plant Operating Reports,  1971.

25.  John Carollo  Engineers, Contract Documents for  San  Pablo  Sanitary  District
    Wastewater Treatment Plants - 1970 Additions. 1970.

26.  Kennedy, Bill, personal communication with D.S. Parker. San Pablo Sanitary District,
    November, 1974.

27.  San Pablo Sanitary District, California,  Wastewater Treatment Plant Operating Reports.
    June, 1973 to July, 1974.
                                      9-92

-------
28. Weddle, C.L., Personal communication to D.S. Parker, Bechtel Corp. October, 1974.

29. Eisenhauer, D.L., Sieger, R.B. and D.S. Parker, Design of an Integrated Approach to
    Nutrient Removal.  Presented  at  the  EED-ASCE  Specialty Conference, Penn.  State
    University, Pa., July, 1974.

30. Leary, R.D., et al,  Effect of Oxygen-Transfer Capabilities on Wastewater Treatment
    Plant Performance.  Journal of the Water Pollution Control Federation,  40, p  1978
    (1968).

31. Leary, R.D.,  et al. Full Scale Oxygen Transfer Studies of Seven Diffuser Systems.
    Journal of the Water Pollution Control Federation, 41, p 459 (1969).

32. Aberley, R.C., Rattray, G.B. and  P.P. Dougas, Air Diffusion Unit. JWPCF, 46, No. 5,
    pp 895-910 (1974).

33. Parker, D.S., Carthew, G.A. and G.A. Horstkotte, Lime Recovery and Reuse in Primary
    Treatment. BED of the ASCE, in press.

34. Flanagan, M.J., Direct Digital Control of Central Contra Costa Sanitary District Water
    Reclamation Plant. Presented at  IAWPR Specialized Conference, London, England,
    September, 1973.

35. Caldwell Connell Engineers, Design Report, Lower Molonglo Water Quality Control
    Centre. Report to the National Capital Development Commission, April, 1971.

36. Caldwell  Connell  Engineers, Revisions to  Design Report,  Lower Molonglo Water
    Quality Control Centre. Report  to  the National Capital Development Commission,
    May, 1972.

37. Process Design Manual for Upgrading Existing Wastewater Treatment Plants. U.S. EPA,
    Office of Technology Transfer, Washington, D-C. (1974).

38. Schwinn,  D.E., Design Features of the District of Columbia's Water Pollution Control
    Plant.  Presented at the Sanitary  Engineering Specialty Conference, ASCE, Sanitary
    Engineering Division, Rochester, New York,  June, 1972.

39. Schwinn,  D.E., and G.K. Tozer, Largest Advanced Waste Treatment Plant in the U.S.
    and in the World. Environmental Science and Technology, 8, No. 10, (1974).

40. Sawyer, C.N., Supplementary Comments on  Nitrification and Denitrification Facilities.
    Prepared for the EPA Technology Transfer  Seminar, Denver, Colorado, November 13,
    1974.
                                       9-93

-------
41. EPA Postpones Denitrification.  Reporter, Interstate  Commerce Commission on the
    Potomac River Basin, 32, No. 2, February, 1975.

42. Harris  County Water Control and  Improvement District No. 50, Monthly Progress
    Report for October 1974, on Project 11010 GNM, prepared for the EPA.

43. Barth,  E.F., EPA Technology Transfer Design Seminar on Nitrogen Control Tech-
    nology. Presented at Denver, Colorado, November 13, 1974.

44. Brown and Caldwell/Dewante and Stowell, Feasibility Study for the Northeast-Central
    Sewage Service Area. Prepared for the County of Sacramento, November, 1974.

45. Sacramento Area Consultants, Outfall  Project and Environmental Impact Report.
    Prepared for the Sacramento Regional County Sanitation District, December, 1974.

46. Sacramento Area Consultants, Chemical Report, Sacramento Regional  Wastewater
    Treatment Plant. January, 1975.

47. White, G.C., Personal Communication to E. Appel, June, 1974.

48. CH2M/Hill, Design  of Montgomery  County,  Maryland Plant for the  Washington
    Suburban Sanitary Commission, 1915.

49. Gulp,  G.L., Gulp, R.L.  and C.L. Hamann, Water Resource Preservation  by Planned
    Recycling of Treated Wastewater. JAWWA, 65, No. 10, pp 641-647 (1973).

50. CH2M/Hill, Wastewater Treatment Study, Montgomery County, Maryland. November,
     1972.

51. Cloromat. Ionics, Inc. Brochure

52. Michalek, S.A., and F.B. Leitz, On-Site Generation of Hypochlorite. JWPCF, 44, No. 9,
    pp 1697-1712(1972).

53. Suhr, L.G., and L. Kepple, Design of a Selective Ion Exchange System for Ammonia
    Removal. Presented  at the ASCE  Environmental Engineering Division Conference,
    Pennsylvania State University, July 1974.

54. CH2M/Hill, Design of the Upper Occoquan Sewage A uthority. (1974).

55. Physical-Chemical Plant Treats Sewage Near the Twin Cities. Water and Sewage Works,
    p. 86, September, 1973.
                                      9-94

-------
56. Larkman, D., Physical-Chemical Treatment:  Chemical Engineering, Deskbook Issue, p.
    87, June 18, 1973.

57. Gulp, R.L., Nitrogen Removal by Air Stripping. Proceedings, Wastewater Reclamation
    and Reuse Workshop, SERL, Univ. of California, LakeTahoe, Ca., June 25-27, p. 128,
    1970.

58. Gulp, R.L.  and G.L. Gulp, Advanced Wastewater Treatment. Van Nostrand Reinhold
    Co., New York, p. 51, 1971.

59. Gulp, R.L., and H.E. Moyer, Wastewater Reclamation and Export at South Tahoe. Civil
    Engineering, p. 38, June, 1969.

60. Wesner, G.M.,  and R.L. Gulp, Wastewater Reclamation and Seawater Desalination.
    JWPCF, 44, No. 10, pp 1932-1939 (1972).

61. Folkman, Y., and A.M.  Wachs, Nitrogen Removal  Through Ammonia Release From
    Holding  Ponds. Proceedings, 6th Annual International  Water Pollution Research
    Conference, Tel Aviv, Israel, June 18-23, 1972.

62. Gonzales, John G., and R.L. Culp, New Developments in Ammonia Stripping, Public
    Works, 104, No. 5, p. 78 (1973) and No. 6, p. 82 (1973).

63. Wesner, G.M.,  and  D.C. Baier, Injection of Reclaimed  Wastewater Into Confined
    Aquifers. JAWWA, 62, No.  3, pp 203-210 (1970).
                                     9-95

-------
                                   APPENDIX A

                            GLOSSARY OF SYMBOLS
Symbol

A



ADWF

(a)

8


BHP

C

Cl
Cb
 m
D
D
D
Definition

air  flow  per unit of tank volume, standard
cu ft per min per  1000 cu ft; dimensionless
number in  contact  stabilization  calculations

average dry weather flow

activity of an ion

fraction of sludge in stabilization tank; oxygen
required  for carbonaceous  oxidation, mg/1

brake horsepower

fraction  of  total   sludge in  contact tank

process dissolved oxygen level, mg/1

denitrifier biomass production, mg/1

required methanol concentration, mg/1

oxygen saturation in water at temperature T,
mg/1

concentration of nitrate nitrogen, mg/1; axial
dispersion coefficient, sq ft/hr

influent NO^-N, mg/1

effluent NO~ -N, mg/1

mass average influent NO~-N level over 24 hr,
mg/1
                                       A-l

-------
Symbol                                   Definition

Dj                                        mass average effluent NOl-N level over 24 hr,
                                          mg/1

DO                                       dissolved oxygen, mg/1
                                                             •

e                                         aerator  rated  oxygen transfer efficiency at
                                          standard conditions, percent

F/M                                      food to  microorganism ratio

f                                         nitrifier fraction  of the mixed liquor solids;
                                          fanning  friction factor

HT                                       hydraulic detention time

I                                         inventory of VSS under aeration, Ib
  A
KJJ                                       selectivity coefficient  for ion exchange equili-
                                          bria.
                                          half  saturation  constant  for  nitrate,  mg/1
                                          NO~-N
K,                                        "decay"  coefficient,  day

K^                                       half  saturation  constant for methanol, mg/1
                                          of methanol

Kj^                                       half  saturation  constant for  oxidation of
                                          ammonia nitrogen, mg/1

Kn                                       half  saturation  constant for oxygen, mg/1
   2
K                                        half saturation constant = substrate concentra-
                                          tion, mg/1 at half the maximum growth rate

L                                        tank length, ft

M                                        methanol concentration, mg/1

M                                        mass of heterotrophs grown through oxidation
                                          of organic carbon

                                        A-2

-------
Symbol
Definition
MLSS

MLVSS

MPN

N

N
N
  1
N
N,
NOD

NT

PDWF

PWWF

Q

Q
mass of nitrifiers grown through oxidation of
ammonia

mixed liquor suspended solids, mg/1

mixed  liquor  volatile suspended solids, mg/1

most probable number

NH. -N concentration, mg/1

TKN in the influent, mg/1

NH£ -N in the effluent, mg/1

24 hr-average influent TKN, mg/1

24 hr-average effluent NH.-N, mg/1

NOl -N level in the contact tank, mg/1

NO--N level in the stabilization tank, mg/1

nitrogenous oxygen demand, mg/1

ammonia nitrogen oxidized, Ib/day

peak dry weather flow

peak wet weather flow

influent flow rate, mgd

mean flow rate (ADWF), mgd

air flow, cfm

rate of substrate removal, Ib BOD (or COD)
removed/lb VSS/day
                                      A-3

-------
Symbol

°lr>
 D
A
qN
JN

A
rN


R

S


SCFM

S
  o

Sl

SF

SS

SVI

T

TKN

At
Definition

nitrate removal rate, Ib NOu-N rem./lb VSS/
day
peak nitrate removal rate, Ib NO~-N rem./lb
VSS/day
                                        ammonia oxidation rate, Ib NH.-N oxidized/
                                        Ib VSS/day
peak  ammonia  oxidation rate,  Ib  NtL-N
oxidized/lb VSS/day

nitrification  rate,   Ib  NH.-N  oxidized/lb
MLVSS/day

peak nitrification rate, Ib NH .-N oxidized/
Ib MLVSS/day

recycle flow rate, mgd

growth limiting substrate concentration, mg/1;
total sludge wasted in Ib/day

standard cubic ft per minute

influent total BOD (or COD), mg/1

effluent soluble BOD (or COD), mg/1

safety factor

suspended solids, mg/1

sludge volume index, ml/g

temperature, degrees C

Total Kjeldahl Nitrogen

time increment
                                      A-4

-------
Symbol                                  Definition

u                                        mean displacement velocity, ft per hr

V                                        volume  of aeration  tank or reactor, mil  gal

V                                        volume of contact tank, mil gal
 c

V                                        volume of stabilization tank, mil gal
 o

VSS                                     volatile suspended solids

W                                        waste sludge flow rate, mgd; tank width, ft;
                                                                                 -
                                         wase suge   ow ra, mg;      w,
                                         total  oxygen  demand, mg/1; oxygen  trans-
                                         ferred under process conditions, Ib/day

                                         oxygen transferred under standard conditions,
                                         Ib/day
                                         a coefficient  in oxygen transfer calculations

                                         MLVSS, mg/1

                                         effluent VSS, mg/1
X                                       contact tank MLVSS, mg/1
  C

X                                       stabilization tank MLVSS, mg/1

X                                       waste sludge VSS, mg/1

Y.                                       heterotrophic yield coefficient, Ib VSS grown
                                         per Ib of substrate removed

Y.                                       net yield of VSS of heterotrophs per unit of
                                         carbon (BOD5 or COD) removed

YD                                      denitrifier gross yield, Ib VSS grown/lb NO~-N
                                         rem.

YD                                      denitrifier net yield, Ib VSS grown/lb NO~-N
                                         rem.
                                      A-5

-------
 Symbol                                 Definition
                                         organism yield coefficient, Ib Nitrosomonas
                                         grown (VSS) per Ib NH^-N removed
CX „                                     selectivity coefficient for ion exchange equili-
                                         bria

 9                                       solids retention time, days
  c

 9                                       solids retention time of design, days
  C

.0                                       minimum solids  retention time,  days,  for
                                         nitrification at given pH, T and DO

 M                                       growth rate of microorganism, day
 A
 IJL                                       maximum  growth  rate  of microorganism,
                                         day"1

 M,                                      net growth rate of heterotrophic population

 ^c                                      growth rate of nitrifiers in contact tank, day"

 MT-V                                     denitrifier growth rate, day

 *                                                                          1
  D                                     maximum denitrifier growth rate, day

 Mpv                                     design denitrifier growth rate, day

 Mxr                                     Nitrosomonas growth rate, day"
 A                                                                        _1
 MXT                                     peak Nitrosomonas growth rate, day
 •
 MN                                     maximum possible nitrifier growth rate under
                                         environmental conditions  of T, pH and DO,
                                         and N >> KN

 n
  s                                      growth rate of nitrifiers in stabilization tank,
                                         day"1
                                        A-6

-------
                                  APPENDIX B
                            METRIC EQUIVALENTS

                            METRIC CONVERSION TABLES
Recommended Units
Description
Length






Area








Volume







Mass




Time





Force








Unit
meter

kilometer
millimeter
centimeter
micrometer

square meter

square kilometer

square centimeter
square millimeter
hectare



cubic meter

cubic centimeter

liter



kilogram
gram
milligram
tonne


second
day

year


newton








Symbol
m

km
mm
cm


m'

km2

cm*
mm*
ha



m3

cm3

1



kg
g
mg
t


s
day

yr or
a

N








Comments
Basic S/ unil











The hectare (10,000
m2) is a recognized
multiple unit and
will remain in inter-
national use.




The liter is now
recognized as the
special name for
the cubic decimeter

Basic SI unit

1 tonne = 1,000 kg


Basic S/ unit
Neither the day nor
the year is an SI unit
but both are impor-
tant.

The newton is that
force that produces
an acceleration of
1 m/s2 in a mass
of 1 kg.





English
Equivalents
39.37 in. = 3.28 ft =
1.09yd
0.62 mi
0.03937 in.
0.3937 in.
3.937 X 10'3=103A

10.744 sq ft
= 1.196 sq yd
6.384 sq mi =
247 acres
0.155sqin.
0.00155 sq in.
2.471 acres



35.314 cu ft = .
1 .3079 cu yd
0.061 cu in.

1. 057 qt = 0.264 gal
= 0.81 X 10-4 acre-
ft


2.205 Ib
0.035 oi-tS.43 gr '
0.01 543 gr
0.984 ton (long) •
1.1 023 ton (short)







0.22481 Ib (weight)
» 7.5 poundals








Description
Velocity
linear






angular


Flow (volumetric)


Viscosity


Pressure








Temperature








Work, energy,
quantity of heat




Power


Application of Units

Description
Precipitation^
run-off.
evaporation






River flow


Flow in pipes.
conduits, chan-
nels, over weirs.
pumping

Discharges or
abstractions.
yields



Usage of water

Density





Unit
millimeter








cubic meter
per second

cubic meter per
second

liter per second

cubic meter
per day

cubic meter
per year

liter per person
per day
kilogram per
cubic meter




Symbol
mm








m3/s


m3/s


l/s

m3/day


m3/yr


I/person
day '
kg/m3





Comments
For meteorological
purposes it may be
convenient to meas-
ure precipitation in
terms of mass/unit
area i kg/m 3).
1 mm of rain =
t kg/sq m

Commonly called
the cumec






1 l/s = 86.4 m3/day







The density of
water under stand-
ard conditions is
1,000kg/m30r
1.000g/l
English
Equivalents









35.314 cfs





15.85 gpm

1.83X 10'3gpm





0.264 gcpd

0.0624 Ib/cu ft





Description
Concentration


BOD loading



Hydraulic load
per unit area;
e.g. filtration
rates




Hydraulic load
per unit volume;
e.g. biological
filters, lagoons

Air supply



Pipes
diameter
length


Optical units

Recommended Units
Unit

meter per
second
millimeter
per second
kilometers
per second

radians per
second

cubic meter
per second
liter per second

poise


newton per
square meter

kilonewton per
square meter

kilogram (force)
per square
centimeter

degree Kelvin
degree Celsius







joule




kilojoule
watt
kilowatt
joule per second
Symbol

m/s

mm/s

km/s


rad/s


m3/s
l/s

poise


N/m2


kN/m2

kgf/cm2



K
C







J




kJ
W
kW
J/s
Comments











Commonly called
the cumec





The newton is not
yet well-known as
the unit of force
and kgf cm2 will
clearly be used for
some lime. In this
field the hydraulic
head expressed in
meters is an accept-
able alternative.
Basic SI unit
The Kelvin and
Celsius degrees
are identical.
The usa of the
Celsius scale is
recommended as
it is the former
centigrade scale.
1 joule «1 N-m





1 watt = 1 J/s


English
Equivalents

3.28 fps

0.00328 fps

2.230 mph





15, 8 50 gpm
* 2.120 cfm
15.85 gpm

0.0672/lb/
sec ft

0.00014 psi


0.145 psi

14.223 psi



5F
-- - 17.77







2.778 X ID'7
kwht "
3.725X10-'
hp-hr = 0.73756
10-4 Btu
2.778 kw-hr



Application of Units

Unit
milligram per
liter

kilogram per
cubic meter
per day

cubic meter
per square meter
per day





cubic meter
per cubic metir
per day


cubic meter or
liter of free air
per second


millimeter
meter


tumen per
square meter

Symbol
mg/i


kg/m3 day



m3/m2 day







m3/m3 day




m3/s

l/s


mm
m


lumen/m2


Comments







If this is con-
verted to a
vetocity.it
should be ex-
pressed in mm/s
(1 mm/i o 86.4
m3/m2 day).
















English
Equivalents
1 ppm


0.0624 Ib/cu-ft
day


3.28 cu ft/iq ft

















SBJS.1!-
3.28ft

0.092 ft
andle/sq ft
                                      B-l

-------