WATER POLLUTION CONTROL RESEARCH SERIES • I608QFSNJO/ 71
  ENGINEERING METHODOLOGY FOR
  RIVER AND STREAM REAERATION
U.S. ENVIRONMENTAL PROTECTION AGENCY

-------
             WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Reports describe the results and
progress in the control and abatement of pollution in our Nation's
waters.  They provide a central source of information on the research,
development, and demonstration activities in the Office of Research and
Monitoring, Environmental Protection Agency, through inhouse research
and grants and contracts with federal, state, and local agencies, research
institutions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research Reports should
be directed to the Chief, Publications Branch  (Water), Research
Information Division, Xi&M,  Environmental Protection Agency, Washington,
D. C.  20'4-eo.
 - about our cover
The cover illustration depicts a city in which man's activities coexist
in harmony with the natural environment.  The National Water Quality
Control Research Program has as its objective the development of the
water quality control technology that will make such cities possible.
Previously issued reports on the National Water Quality Control Research
Program include:
Report Number
16080	06/69
16080	10/69

16080DRX10/69
16080	11/69

16080D0007/70
16080DVF07/70
16080	10/70

16080DWP11/70
16080DUP12/70
16080FYA03/71
16080GPF04/71
16080GGP07/71
                       Title
Hydraulic and  Mixing Characteristics of Suction Manifolds
Nutrient Removal from Enriched Waste Effluent by the
  Hydroponic Culture of Cool Season Grasses
Stratified Reservoir Currents
Nutrient Removal from Cannery Wastes by Spray Irrigation
  of Grassland
Optimum Mechanical Aeration Systems for Rivers and Ponds
Development of Phosphate-Free Home Laundry Detergents
Induced Hypolimnion Aeration for Water Quality Improve-
  ment of Power Releases
Induced Air Mixing of Large Bodies of Polluted Water
Oxygen Regeneration of Polluted Rivers:  The Delaware River
Oxygen Regeneration of Polluted Rivers:  The Passaic River
Corrosion Potential of  NTA in Detergent Formulations
Effects of Feedlot Runoff on Water Quality of Impoundments

-------
ENGINEERING METHODOLOGY FOR RIVER
       AND STREAM REAERATION
                  by
        JBF Scientific Corporation
              2 Ray Avenue
     Burlington,  Massachusetts  01803
                  for the
ENVIRONMENTAL PROTECTION AGENCY
            Project #16080 FSN
              October  1971

-------
                           EPA Review Notice


         This report has been reviewed by the  Environmental
         Protection Agency and approved for publication.
         Approval does  not  signify that the contents necessarily
         reflect the views and policies of the Environmental
         Protection Agency, nor does mention  of trade names
         or commercial products constitute endorsement or
         recommendation for use.
Forsnlo by llio Superintendent of Documents, U.S. Government Printing 0/licc, Washington, D.C. '.'WOi- I'riccSI.-.'.s
                              11

-------
                         ABSTRACT
Results of recent activities in river and stream aeration by artificial
techniques are reviewed, and a rational engineering methodology is
developed for future river and stream aeration projects.

The development of the methodology follows from a thorough review
of the oxygen dynamics in rivers  and  streams and the capabilities of
aeration systems within the present state of the art.   The report
shows how the theoretical work can be simplified considerably and
applied  to the solution of river and  stream water quality problems.
It is  assumed that aeration would only be  used as a "polishing" action
after all identifiable waste sources  have received at least secondary
treatment.

The results indicate that,  with careful consideration of  site factors,
artificial aeration can be applied  successfully to raise dissolved oxygen
to 5 ppm, using mechanical surface aerators, diffusers, downflow
contactors, and sidestream mixing.  However,  since  the transfer
of oxygen from air into water is  relatively inefficient  above 5 ppm DO,
the introduction of molecular oxygen through sidestream mixing,
U-Tubes, and possibly diffusers should be considered,  depending  on
the volume of water to be aerated.  In cases where DO may be main-
tained at levels  lower than 5 ppm, systems using air are competitive
with molecular oxygen,  depending on  site conditions.

This report was submitted in fulfillment of Project Number 16080 FSN
under the partial sponsorship of the Environmental Protection Agency.
                               111

-------
                          CONTENTS

SECTION                                                     Pag
    I       CONCLUSIONS                                        1
    II      RECOMMENDATIONS                                  3
    III     INTRODUCTION
             Scope and Purpose                                   5
             Background                                         6
             Approach                                           7
    IV     DISSOLVED OXYGEN DYNAMICS IN RIVERS
           AND'STREAMS
             Oxygen Balance                                      9
             Evaluation of Stream Parameters in Oxygen-
             Balance Equation                                  14
               Saturation Values of Cs                          14
               Aeration Constant--Ka                           14
                                   Ct
               Biochemical Oxygen Demand (BOD)               21
               Benthal Demands  S                              28
               Net Photo synthetic Oxygen Production _P          30
             Longitudinal Dispersion                            31
             Simplification of the Oxygen-Balance Equations       34
    V      AERATION SYSTEMS FOR RIVER AND STREAM
           APPLICATIONS
             Introduction                                       39
             Mechanical  Surface Aerators                        39
             Diffuser Systems                                  50
               Diffuser Systems Tests Using Air                51
               Diffuser System Tests Using Molecular
               Oxygen                                         54
               The Effect of Flow on Dispersion of
               Oxygen from Diffusers                           54
             Downflow Contactors                               55
             Sidestream Pressurization                         72
             Use of Pure Oxygen                                75
             Hybrid or Mixed Systems                           78
                                v

-------
SECTION                                                         Page

    VI     ENGINEERING METHODOLOGY FOR RIVER AND
           STREAM AERATION
             Problem Recognition                                  83
             Preliminary Assessment                              83
             Determination of Oxygen Requirements                85
             Selection of Aeration Units                            94
                Engineering Considerations Affecting the
                Selection  of an Artificial Aeration System           94
                Economic Considerations Associated with
                an Aeration System                                96
                Design and  Cost Examples                          99
                Summary                                         103
             Site Factors                                         103
             Final Selection of an Aeration System                104
    VII    ACKNOWLEDGEMENTS                               105
    VIII   REFERENCES                                         107
    IX     APPENDICES
             Appendix A - Diffusers and Mechanical Aerators      1 13
                                  VI

-------
                            FIGURES
Figure                                                         pag,
 4. 1    Typical Rate- of-Diffusion Plot                           11
 4.2    Typical Plot of Photosynthetic Oxygen Production         11
 4. 3    K  vs Flow, as Calculated by Various Prediction
          a
        Equations                                                18
 4.4    Grand River BOD Curve                                 22
 4. 5    Total Carbonaceous and Nitrogenous  BOD as a
        Function of Time                                         25
 4.6    BOD  Curve, Lansing Wastewater Treatment
        Plant, 24-hr Composite Sample,  Taken
        July^lZ,  I960                                            26
 4.7    First-Order Representation of a Nitrification
        Process                                                 29
 5. 1    Example of Velocity Profiles at Various Distances
        from a 75-hp Aerator                                    41
 5. 2    Mean Rate of Oxygen Absorption at Steady-State
        Operation  Under Standard Conditions                     46
 5. 3    Oxygen Transfer Rate vs Flow for Mechanical
        Aerator                                                  47
 5. 4    Graphical Results of Cross-Sectional Area
        Distribution                                             56
 5.5    DO Deficit in 40-Ft-Deep U-Tube                         58
 5.6    Air Blower Injection Modification of U-Tube
        System                                                  59
 5. 7    Cascade Air Injection                                    60
 5.8    Venturi Air Injection                                     61
 5.9    Schematic  Installation of U-Tube  Oxygenation of
        Stratified Impoundment Releases                          62
 5. 10   Increase in DO vs Per Cent A/W  for Diffusion of
        Oxygen and Air                                          65
 5. 11   Effect of Depth on Oxygen Transfer Economy  in
        60-in. U-Tube                                           66
 5. 12   Effect of Nominal Velocity  on Transfer Economy
        in 60-in. U-Tube, 40-ft Deep                            67
 5. 13   Comparison of Pressure Drop Across Center-Plug
        and Venturi Aspirators                                   70
 5. 14   Zone of Influence of the  Downdraft Bubble Contactor
        on the Surrounding Water                                 71
                              vn

-------
Figure                                                          Page
  5.15   Cost of Pure Oxygen                                    77
  5. 16   Total Oxygen Absorption vs Depth of Injection            79
  5. 17   Effect of Percentage Oxygen in Aerating Gas on
         the Rate  of Oxygen Transfer                             81
  6. 1     Outline of Engineering Methodology                      84
  6. 2     Example DO Profile                                     86
  6. 3     Relative  "Work Requirements for a Unit Increase
         in DO in  Water Initially  Having Different DO
         Levels                                                  88
  6.4     DO Profile Before and After Oxygen Addition at
         Three Locations                                         91
  6. 5     DO Profile Before and After Oxygen Addition at
         Two  Locations                                           92
  6. 6     DO Profile Before and After Oxygen Addition at
         One  Location                                            93
                              Vlll

-------
                             TABLES
TABLE
  4. 1     Saturation Values of Oxygen in Clean Water at One
         Atmosphere Pressure                                   15
  4.2     Aeration Parameters for Determining K                  17
                                               3,
  4. 3     Transfer Rate of Oxygen into Water from Air
         for Various Flow Conditions                              20
  4.4     Percentages of Carbonaceous BOD as a Function
         of Time for Kc  = .  I/day                                 23
  4. 5     Typical Longitudinal Dispersion Coefficients for
         Rivers                                                  32
  4.6     Sample^Calculation of Dispersion Coefficient from
         Field  Data  Test                                          35
  5. 1     Surface Aerator Pumping Rates                          42
  5. 2     Zone of Outward Aerator Influence                       43
  5. 3     Sample DO Profiles at Various Distances from Aerators  44
  5.4     Average Oxygen Transfer Efficiencies,  in lbsO£/hp-hr   48
  5. 5     Average Oxygen Transfer Rate Obtained by Different
         Investigators for Surface Aerators                       49
  5.6     Diffuser Results from the Passaic  and Delaware
         Rivers                                                  52
  5.7     Reaeration Data for the Pearl River,  Louisiana, Using
         a Double Aeration Header with 1/32-in.  and 3/61-in.
         Orifices                                                55
  5.8     Air-Water  Ratio Required for Oxygen Saturation          63
  5.9     Air-Water  Ratio Required for Nitrogen Saturation         63
  5. 10   Results of Chemical Analysis for Dissolved Nitrogen      68
  5. 11   Vertical Area Affected by Bubble Contactor               72
  5. 12   Dissolved Oxygen Added to the Pearl River at  an
         Oxygen Addition Rate of 30, 000 Ib/day by Sidestream
         Oxygenation                                             74
  5. 13   Comparison of Reaeration Studies with Air  and Oxygen    80
  6.1     Characteristic Features of Aeration  Systems           97-9!
  A. 1     List of Diffuser  Manufacturers                         114
  A. 2     Diffuser Efficiencies  at Various Locations in Tank
         Under  Standard Conditions                             116
  A. 3     List of Mechanical  Aerator Manufacturers              117
                                IX

-------
                           SECTION I

                         CONCLUSIONS
1.    A review of the present state of the art in river and stream
      aeration indicates that the performance  of aeration systems  is
      strongly related to the minimum DO level set as a standard on
      any particular river or stream.

2.    When a minimum level of 5 ppm  is set,  any system not using
      molecular oxygen will be  relatively inefficient and, unless
      unusual site conditions prevail, will probably not be suitable.
      This is particularly true in summer months, when the
      saturation concentration for DO is at its lowest value.

3.    The most efficient location for mechanical surface aerators
      and diffusers using air is at the point of maximum oxygen
      deficit.  In order to maintain some established minimum DO
      level, however,  these devices must be located wherever that
      minimum is approached,  resulting in a significant loss of
      transfer efficiency.  Based on presently available transfer rate
      data,  mechanical surface aerators and diffusers using air are
      not efficient for maintaining DO levels above 4 ppm.

4.    Downflow contactors provide a higher transfer efficiency than
      surface aerators  or diffusers but can only be used where
      sufficient water depth is available.  Nitrogen supersaturation
      may also be a problem but can be avoided by limiting the depth
      of the down leg of the tube.  Since maximum transfer efficiency
      occurs at depths greater than 40 feet, there will always be some
      loss in efficiency if nitrogen supersaturation is to be avoided.

5.    The lack of mobility and dependence on water depth limit the
      usefulness of downflow contactors  in problems requiring the
      maintenance of a  minimum DO level.  These systems may be
      considered only if an injection point for  oxygen occurs at a
      compatible location.

6.    A relatively simple methodology can be  developed for treating
      river and stream aeration problems.  In this methodology  the
      DO profile for "worst conditions" is used to determine  alternative
      locations for aeration units in order to maintain some specified
      minimum DO level.  Loss in natural aeration due to the oxygen
      addition must be taken into account.  Alternative locations  are
      required because site conditions  may preclude the use of some
      types  of aeration  systems.  There may also be cases where  a.
      trade-off is  necessary between the injection of large amounts
      of oxygen at one location versus  smaller amounts at several
      locations .

-------
 7 .     For large rivers where more than 50, 000 Ibs /day of oxygen is
       required,  the use of molecular oxygen applied through side
       stream mixing should definitely be considered.  Whether or
       not the oxygen is supplied from a generating plant located on
       the site,  or delivered to the site as liquid oxygen will depend
       on the length of time over which oxygen is  required.   For con-
       tinuous year-round  operation, gaseous oxygen should be
       generated on the site.  For intermittent periods  and  for volume
       requirements of less than 50,000 Ibs/day,  liquid oxygen trucked
       to the site may be more economical.

 8.     The transfer processes of various aeration systems  in rivers and
       streams are not as  well understood as in treatment facilities or
       even lakes and impoundments.  The superimposed flow field
       creates some uacertainty in handling  the analytical details,  and
       a wide range of efficiencies has been  reported.  The  range for
       surface aerators after conversion to standard conditions  is  1.2
       to 4.5 Ibs O2/hp-hr.  Thus,  depending on the number chosen,
       cost estimates may be off by as much as a factor of three.

 9.     The mathematical models of  stream processes are generally
       adequate  for the engineering  design of a river or stream  aeration
       system.  The major inadequacy lies in the prediction of natural
       aeration coefficients from empirical methods.  If measurements
       of the reaeration coefficient cannot be made at the site and an
       empirical method must be selected, the stream conditions from
       which that particular derivation was made  should be  reviewed to
       determine suitability for the  particular application.

10.     A review of current techniques for measuring rate  processes
       which control DO and  BOD in rivers and streams indicates that
       the present state of the art offers sufficient accuracy for the
       design of stream aeration systems.  Additional refinements can
       be made; however, the results may offer a precision greater
       than the variations in  the processes being measured.

11.     Differences in longitudinal dispersion coefficients for a stream
       have  been reported. Such differences may  be due to  whether or
       not measurements  were made with the aeration system in place
       and operating.   Knowledge of the dispersion characteristics are
       particularly important for large river systems where the  aerator
       system is  a relatively  localized source.  Measurements of the
       dispersion coefficient should  be made with the device in place.

-------
                            SECTION II

                        RECOMMENDATIONS
Artificial aeration of rivers and streams should not be used as a direct
substitute for waste treatment at the source.   There may be cases,
however,  where aeration can be used as a  "polishing" action during
periods of high temperature and low flow.  Until advanced waste treat-
ment methods are fully developed and implemented, public opinion may
require an interim solution.  Artificial aeration is at a state of devel-
opment now where it can be applied to specific river and  stream
problems.

Although artificial aeration is technically feasible, cost estimates  are
still not sufficiently accurate, due to problems in  predicting the trans-
fer efficiency of_a system in a stream.  This deficiency will probably
not be  corrected by  performing more tests.  The primary need is to
define  the conditions under which aeration  devices  should be tested if
they are to be used in a  river or stream environment. A laboratory
program should be  undertaken to standardize measurement conditions
for a variety of hydraulic parameters which might  be encountered in
the field.  The results of this program  should then be  cross-checked
to determine compatibility with field conditions.

The results of this  study have indicated that the maintenance of a 5 ppm
DO requirement will be  difficult unless molecular  oxygen is employed,
and methods for delivering molecular oxygen to the water need additional
development.  The  sidestream pressurization method  is oriented toward
large rivers where only a portion of the flow can be diverted.  Work
should be  done on using  gaseous oxygen, delivered to a site  by truck, to
supply a fine-bubble-size diffuser.   The gas would be  delivered under
pressure, and this might reduce the possibility of internal clogging.  A
system like this would have application in small, shallow rivers or
streams.

The conclusion has been made that surface aerators will not be effec-
tive in maintaining a DO level above 4 ppm.  In cases  where a level
less than  4 ppm is acceptable, the design of the aerator should be modi-
fied to promote mixing of air and water downstream of the aerator.
The symmetrical radial mixing zone of present aerators can be made
asymmetrical by adding baffle plates or flow guides.

The design of an aeration  system for a  river or stream requires con-
sideration of a number of factors.   The system can be designed suc-
cessfully  by following an orderly procedure which  highlights all of  the
considerations.  The procedures should be formalized to  the extent
that design errors due to a lack  of knowledge  about river and stream
processes are minimized.

-------
In  summary, the following areas are most in need of additional
research and development:

       I.   Uniform standards for the measurement of transfer
           efficiencies of various aeration devices under flow
           conditions.

       2.   Improved devices for the diffusion of gaseous oxygen
           in the water.

       3.   Enhancement of surface aerator performance through
           the  use  of flow guides or baffles.

-------
                           SECTION III

                         INTRODUCTION
Scope and Purpose

Present aeration technology for rivers and streams has  developed
primarily from waste treatment applications and from a limited
number of field tests in lakes and rivers.  The purpose of this study
was to review the results of recent activities in river and stream
aeration by artificial techniques and  to assess the present state of
the art.  A direct result of this review has been the development of
an engineering methodology for river and stream aeration systems.

The major objective in any aeration  system design is to  add oxygen to
the water.  In waste treatment applications  this is done to satisfy a
high biochemical oxygen demand (BOD), while in lakes and rivers one
wishes to maintain a high-enough dissolved  oxygen level to support a
healthy aquatic population.   There are other differences affecting
system design, including the need to promote mixing of suspended
solids  in an aerated lagoon versus the desire to minimize excessive
turbulence in lakes for aesthetic reasons.  In rivers and streams
natural flow  conditions provide mixing,  and new water is always
being exposed to  the aeration system.  In lakes where  there is no
natural flow  condition the most efficient aeration system is one which
provides continual recirculation with minimum power.  This results
in the exposure of new water surfaces to the air and promotes  diffusion
and mixing.

In order to efficiently design an aeration system for rivers and streams
the oxygen balance for the stream must be understood.   The primary
source of oxygen for a stream is  natural reaeration at the surface,
which is  aided  by the velocity and turbulence of the stream.  If a
stream is artificially  aerated, there will be some loss in the natural
aeration capability, and this must be compensated for  by the engineer
in estimating the additional oxygen required to meet a  specified water
quality condition.  Additional contributions come from photosynthesis,
ground water,  drainage, and flow augmentation.

The remainder of the  report is  organized as follows.   In Section III
the scope and purpose, background,  and approach methods are intro-
duced.  Section IV  includes a review of  the oxygen dynamics in rivers
and streams and a section on the measurement of parameters in
the oxygen-balance equation. Section V is a review of the use  of
surface mechanical aerators, diffusers, downflow contactors  (U-Tubes),
and sidestream mixing in river and stream  applications.  Also included
in Section V  is  a  discussion of the possible benefits of using pure
oxygen instead of air.  In Section VI the methodology for designing
the aeration  system is developed.

-------
Background

Over the last 50 years an extensive body of literature has accumulated
on the  application of aeration in waste treatment.  Recently, interest
has developed in the use of aeration technology for destratifying lakes
and reservoirs and in artificially improving the assimilative capacity
of rivers  and streams.

Serious consideration of artificial aeration for rivers and streams
appears to have been first initiated by Tyler [l] in the early 1940's.
He proposed that under certain conditions it might be economically
advantageous to treat waste in a stream rather than in concentrated
form as would  be  the case in a treatment pond.  This philosophy has
apparently interested a number of investigators, as  there have been
many tests  of this concept in the last 5-10 years.

A second  philosophy on the use of artificial stream aeration assumes
that  all possible waste material entering a stream is first treated in
a waste treatment plant,  resulting in the removal of 90-99% of the
BOD.  The  aeration system would then be used as a supplement where
DO levels on the order of 5  ppm are required to support a viable
aquatic life and its attendant recreational benefits.   In this  case the
aeration system would probably be operated on a seasonal basis when
DO levels fell below 5 ppm.  As will be shown later, much more work
is required to maintain a 5 ppm level than would be needed for a 3 ppm
level;  thus relative transfer efficiencies of aeration devices are an
important consideration.

Many of the aeration tests conducted have used site-dependent techniques,
such as turbine venting, weirs, and dams.  Although these techniques
have demonstrated reasonable transfer efficiencies, their dependence
on site conditions limits general application.  Consequently, the
emphasis in this study has  been on more flexible systems,  such as
mechanical surface aerators, diffusers, downflow contactors, and
sidestream mixing.

Since  1966, a number of tests have been conducted using mechanical
surface aerators.  A fewer number have used diffusers  and side-
stream mixing.  No tests  using U-Tubes in a flowing river were found,
although there is  reason to  believe from results  obtained in lakes and
impoundments  that these devices can be used on rivers and streams,
given sufficient water depth. In several tests  of diffusers, U-Tubes,
and  sidestream mixing, molecular oxygen has been substituted for air.

In many of the test programs the aeration systems were evaluated
under  a variety of actual or simulated stream  conditions.   Performance
data inmost cases was  converted to a set of standard conditions, with
the transfer efficiencies  stated in pounds  of oxygen transferred per
horsepower hour.

-------
Although the economics  of river and stream aeration favor a high
oxygen transfer rate,  site conditions also restrict the use of
particular aeration devices, for reasons other than cost.  Many
rivers are navigable both to shipping and recreational boating;  others
are used for recreational boating and water sports.   Where public use
is  extensive, surface  aerators using turbulent mixing would be re-
stricted.  In cases where a channel is maintained by dredging,  diffuser
systems cannot be located in the channel.  Designs of aeration systems
must also consider aesthetic conditions at the site.  Large, unsightly
structures with extensive surface agitation may not  be acceptable to
the public.
Approach

An extensive survey of the literature was conducted, with particular
emphasis on reports of field tests with mechanical  surface aerators,
diffusers,  U-Tubes,  and sidestrearn mixing.  Although necessary in
the development of a total river basin system,  reports dealing with
mathematical and simulation models were reviewed only  for background
information.  Each aeration field test was reviewed for engineering
design, continuity of results, and efficiency.

In addition to a state-of-the-art review,  emphasis was placed on the
development of an engineering methodology for river and stream
aeration which could be used by an engineer charged with the develop-
ment of a system for a particular location.

In the  course of the study it  was  evident that some  aspects of river
and stream aeration require additional  research.   Although large-scale
projects may be possible within the present state of the art,  additional
research should result in better  system effectiveness. Recommendations
for further research are included in this  report.

-------
                          SECTION IV

  DISSOLVED OXYGEN DYNAMICS IN RIVERS AND STREAMS


Oxygen Balance

Artificial aeration of a body of water, whether it is a lake or a stream,
will have some effect on the natural oxygen balance.  An understanding
of the natural processes controlling oxygen concentration is therefore
necessary before the design of an artificial aeration system can be
undertaken.

There are four main processes controlling oxygen concentrations  in
naturally aerated streams:

      1.    Consumption of oxygen  as a result of respiration of
            benthic and planktonic organisms, and chemical
            oxidation;

      2.    Exchange of oxygen as a result of atmospheric  reaeration;

      3.    Photosynthetic production of oxygen  during the  day by
            benthic plants  and phytoplankton;  and

      4.    Oxygen contribution from  ground water,  surface
            drainage,  and  storage.


In the first case, consumption of oxygen by respiration is expressed
as a biochemical oxygen demand (BOD)  in pounds of oxygen per unit
time.  For DO levels below  1 ppm the rate of consumption has been
found to be dependent on the DO concentration [2] ;  thus, in highly
polluted streams there may  be a diurnal variation in BOD if the
stream becomes oxygen depleted in  any particular area. For higher
DO concentrations there does not appear to be the same dependence.
Hence,  for rivers or streams in which the minimum DO levels are
on the order of 2 or 3 ppm it is sufficient to assume that the oxygen
demand is a function only of the remaining BOD.

The oxidation of the organic load may occur in the stream or on the
bottom,  depending on the physical nature of the material.  Because
the rate of demand of the finely dispersed material in the stream
differs from the demand of the larger particles,  which settle to the
bottom,  the two components  are often treated separately.

Also included in the respiration process is the oxygen uptake caused
by nitrification, which is the oxidation of ammonia and nitrites to
nitrates.   These compounds are common in the effluents of secondary

-------
waste-treatment plants.  Because nitrification is an autotrophic
process carried out by relatively few organisms, an appreciable
time lag may exist between the introduction of ammonia and the
point at which measurable nitrification occurs.  Because of the
difference in rate  and time,  nitrification is often treated separately
from BOD.

The  second process is the primary natural mechanism for oxygen-
ating a stream.  Oxygen diffuses from the atmosphere directly into
the water, with the exchange rate depending on the rate of  renewal
of new water surfaces and on the percentage saturation of oxygen in
the water. Turbulent stream conditions facilitate a higher oxygen
transfer than do quiescent conditions.  If the water becomes super-
saturated with oxygen,  the direction  of transfer  may actually be to
the atmosphere.  Figure 4. 1 shows typical diffusion rates for a
stream  in which there is significant photosynthetic oxygen production.
In this case,  oxygen is released to the atmosphere when photosynthesis
during the day results in supersaturation.

The production of  oxygen by photosynthesis is  the third process in
the oxygen balance.   In this  process,  carbohydrates  are synthesized
from carbon  dioxide and water, with  a subsequent release of oxygen.
This process requires radiant  energy from the sun and is consequently
diurnal  in nature.   The greatest rate of production occurs around noon
and drops to  zero  at night, as  is shown in Figure 4.2.  The rate and
total production  will depend  on the depth of penetration of sunlight,
which is,  in turn,  dependent on water clarity or turbidity.  Maximum
daily production occurs at about 1800 hours, as  opposed to the
maximum rate of production, which occurs around noon.

The fourth process is the accrual of  oxygen from ground water,
drainage,  and storage.  This is a site-dependent contribution whereby
incoming waters containing higher (or lower) concentrations of
dissolved  oxygen produce changes in  the stream DO.  This contribution
is usually negligible except when it involves flow augmentation from
hydroelectric storage impoundments.

The dissolved-oxygen level in a stream, therefore,  is the net  result
of several dynamic processes  occurring simultaneously.  The
processes  interact to produce  variations  in DO along the length of
a stream,  the graphic representation of which is usually called the
DO profile or "oxygen sag curve."

These processes can be quantized on an area basis,  e.g.,  g/mz/hr,
or on a  volume basis as g/m3/hr, which is also ppm/hr.   The
summation of these processes  can be expressed as follows:
                        q = d + p - r + a                        (4. 1)
                             10

-------
 o

 c
•rH


O
S-i
 O

 C
 o
•«
O
                                         Shaded Area Shows Oxygen
                                          Diffusion into the Stream
                                                1800^      2400

                                                         Time (hours)
                 Stream Becomes
                  Supersaturated
                    Figure 4. 1.  Typical Rate-of-Diffusion Plot
c
0)
81
U) .-I
O •>-'
O
-C
                    Time of Maximum
                       Production
                    600
1200         1800

Time (hours)
2400
       Figure 4. 2.  Typical Plot of Photosynthetic Oxygen Production
                                   11

-------
where

      q = rate of change of dissolved oxygen per unit volume (ppm/hr

      r = rate of respiration, including  oxygen demands by plants,
         animals,  and aerobic  bacteria (ppm/hr)

      d = rate of diffusion from the air if the concentration of
         dissolved oxygen is below saturation (ppm/hr)

      p = rate of production of oxygen by photosynthesis (ppm/hr)

      a = rate of accrual from drainage, ground water, or storage
         (ppm /hr)
If an oxygen sag curve shows low DO levels at certain stream locatio
or if DO is lower than desired over a particular reach of a stream,
artificial aeration systems may be used as a supplemental source of
oxygen.  The fate  of the added oxygen will  depend  on the natural stre
processes.

In designing an aeration system it is necessary that a new oxygen saj
curve be calculated after oxygen is  added artifically to  raise the DO
level.  The shape  of the curve will indicate how many aeration devic<
are required and their spacing, in order to keep the DO level above
some specified minimum.  In Section VI the specific techniques for
determining the number and spacing of aerators are developed.  In
this section it is sufficient to note that the  artificial aeration will hav
some effect on the natural balance,  and any decrease in the ability oJ
the  stream to reaerate naturally must be compensated for by the
aeration system.

In order to obtain  a numerical solution to the oxygen balance equatioi
a one-dimensional representation of each of the rate processes in
Equation 4. 1 is developed as expressed in  Equation 4.2.
         =DT— -U+K(C   -C)+P-KL-KN+A-S   (4.;
      at     L   2    3x    ax s     '        c      n           v
where
      A     =   rate of accrual of O? from drainage,  ground water,
               etc. (ppm /day)

      C     =   DO concentration (ppm)

      C     =   saturation value of DO (ppm)
       S
                              12

-------
      D     =  turbulent diffusion (dispersion) coefficient (ft2/day)
       JU

      K     =  aeration constant (I/day)
       a
      K     =  instream carbonaceous oxidation constant (I/day)
       c

      K     =  instream nitrogenous  oxidation constant (1/day)

      P     =  photos ynthetic production  rate (ppm/day)

      S     =  benthal demand rate (ppm/day)

      U     =  mean stream velocity (ft/day)

      t     =  time (day)

      x     =  distance along the  stream  (ft)

and

      L,  and N are given by the solutions to
           IT  =  DT
           9t      L
and
               =  D        _  u     _ KN  +N
           3t      L  gx2        3x      n      a

where

      L    = carbonaceous BOD (ppm)

      L    = uniform rate of addition of carbonaceous BOD (ppm/day^
       3.
      N    = nitrogenous BOD (ppm)

      N    = uniform rate of addition of nitrogenous BOD (ppm/day)
       3.

Equations 4.2 through 4.4 form a system of coupled, second-order,
partial differential equations,  the solution of which results in DO and
BOD profiles along a stream.  The equations are for general non-
steady-state  conditions and can be simplified for many stream con-
ditions.   Considerable simplification can be achieved, for example,
                              13

-------
when steady-state conditions are assumed or if diffusion effects are
small.  If simplification is not possible, a digital computer  greatly
eases the computational burden.

The  first step in the solution is to partition the river reach under con-
sideration into a number of sections,  each of which is assumed to be
completely mixed or homogeneous.  A finite-difference method is ther
used to express the derivatives of the dependent variables C_, L_, and 1
Details of the numerical producer can be found in references 3 and 4.

Although each of the rate terms  and constants appearing in Equations
4.2 through 4.4 may be different for each section, they are  often
essentially equal and additional simplification can be achieved.

In order to obtain numerical values for terms appearing in Equations
4.2 through 4.4, a number of laboratory and field techniques have
been developed.  These techniques are discussed in the next subsectio
along with the practical assumptions that can be made and the relative
orders  of magnitude of the measured  values.
Evaluation of Stream Parameters in Oxygen-Balance Equation

Saturation Values of C
The solubility of oxygen in water is primarily influenced by temperatu
and salinity.  According to SED-ASCE [5] ,  the  saturation value of
oxygen concentration at sea level can be expressed as
where
      C   =  14. 625  -  .41022T + . 0799T2  - . 000077774T:
        s
          = temperature in  C.
In studies conducted on the Passaic River, organic pollution did not
appear to have a discernible effect on the saturation values [6] .  It
is reasonable,  therefore,  to assume that, unless one is dealing with
extremely high organic loadings,  the values of Cs given by the above
equation are applicable.  Values of Cs for the temperature range of
0°C to 40°C  are given in Table 4. ].
Aeration Constant— K_
                     d
The natural aeration rates that occur in streams depend directly on
the amount of turbulence and consequently are related to hydraulic
parameters, such as velocity and depth of flow.  Temperature and
                               14

-------
                            TABLE 4. )

           Saturation Values of Oxygen in Clean Water
                  at One-Atmosphere Pressure
Temperature
°F °C
32. 0
35.6
39.2
42.8
46.4
50. 0
53.6
57.2
60. 8
64.4
68.0
71.6
75.2
78.8
82.4
86.0
89.6
93.2
96.8
100.4
104.4
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
C
s
14.6
13.8
13. 1
12.5
11.9
11. 3
10. 8
10.4
10.0
9.5
9.2
8.8
8.5
8.2
7.9
7.6
7.4
7.2
7. 0
6.8
6.6
the type and concentration of pollutants in solution also affect aeration
but to a lesser degree.

The aeration constant for a stream in which there is little biological
and chemical activity  can be readily evaluated.  The temperature (to
determine Cs) and DO measurements are taken at two stream locations
where  relatively steady-state conditions  exist.  An average value of
Ka can then be calculated using Equation 4.5.

The rate of atmospheric aeration has been demonstrated to be propor-
tional to the oxygen deficit (C  -C)[7],  i.e.,
Letting C  - C = D,
                              15

-------
                     -
                       dt     a


which upon integration  results in
                               = K                              (4.5)
                        At

where

      D  = upstream deficit (ppm)

      D-, = downstream deficit (ppm)

      At = time of travel between stations (days)

As mentioned previously,  if the DO concentration in a stream is  strongl
influenced by biochemical oxygen demans, photosynthesis, nitrification,
etc. ,  the  above approach is questionable.  To minimize these effects,
the time of observation between stations should be minimized;  however,
caution must be exercised in order to avoid conditions which may reflec
small differences of small numbers.

Since natural aeration is  a function of surface renewal,  which,  in turn,
depends on hydraulic parameters, a number of  empirical and theoret-
ical formulations have been proposed for determining K  .  Many of
the theoretical studies are based on transport phenomena and consider,
for example,  effects of surface  tension, molecular diffusivity,  and
turbulence.  In general, for stream applications these approaches  re-
duce  to formulations which express Ka  in terms of two readily determin^
able stream  parameters,  viz, depth and velocity.  The general form of
the equation  is
                                ,
                                  1
                                                                (4.6)
Table 4.2 contains values of the various constants in Equation 4.6
determined by different investigators.

Observation of the values given in Table 4.2 indicates  that each
formulation will result in a  different value for Ka.  It should be
realized that each formulation  has its limitations.  In the case of a
theoretical formulation [9] verification with stream conditions is
needed.  And for the remaining empirical equations caution must be
used in applying results to stream conditions  other than those from
which the observations were made.
                               16

-------
                           TABLE 4.2
            Aeration Parameters for Determining K .
                                                   cL
Investigator
Churchill et al. [8]
O'Connor & Dobbins [9]
Games on [ 1 0
Lang be in & Durham [11]
Owens et al [12]
C
11.56
1.291
23. 17
7.59
21.62
kl
. 969
. 5
.73
1. 0
.67
k2
1.673
1.5
1.5
1.33
1.85
Usually both the depth H_ and the velocity V_ can be obtained as a
function of stream flow (Equation 4.7a),  which would then permit
the aeration constant to be expressed as  a direct function of flow,
as shown in Equation 4. 7b.
                       H = fj(Q)

                       V = f2(Q)
                              (4.7a)
From Equation 4. 6
                  Ka =
                           k
                              (4.7b)
                          H
For the Passaic River [6] this type of analysis resulted in the
following  expressions for depth and velocity:
                   H = 0. 746Q

                   V = 0.020Q
                               . 398
. 524
The above expressions were used in Equation 4.6 for the various
values of C_,  kl,  and k£ given in Table 4.2  The resulting predictions
of Ka are shown in Figure 4. 3,  indicating that values of Ka differ
widely according to the various formulations.  (The precise reasons
for such a spread are not clear, although the investigators  speculated
that perhaps  it is  because  the Passaic is an unusually slow  river and
may be out of the  range of the formulations . )
                             17

-------
oo
                 K
               (day
1.00




0.90




0. 80




0.70




0.60




0. 50




0.40




0. 30




0. 20




0. 10





   0
                                                      Gameson
                                                         O'Connor & Dobbins
                                                            Owens, Edwards,  & Gibbs
                                                          Langbein & Durham
                                     Churchill,  Elmore, &  Buckingham

                                   J	I	I	I	I	I	[_
I
I
                                  100   200    300   400    500    600   700    800   900    1000


                                                        Flow  (cfs)
             Figure 4.3.  K  vs Flow,  as Calculated by Various Prediction Equations (Whipple etal. [6])
                           3.

-------
Because of the significant spread in the predicted values of the aeration
constant, a serious question arises as to the general applicability of
the formulations.   The formulation of O'Connor and Dobbins (see
Table 4.2) seems to provide an adequate description of aeration in
natural waterways,  since good agreement was  found by the authors
between measured and predicted values for several rivers [9] .  Their
formulation was also considered to be the most representative  of
aeration of the Passaic River.  Other equations, e.g. , that of Churchill
et al. ,  also provide good results for several rivers [8] .

Since the natural aeration capability is affected by many chemical, bio-
logical,  and physical factors,  it is  quite possible that no one formulation
will be  universally applicable.  Thus, the above formulations should be
used with discretion,  particularly where "white water" turbulence or
vertical stratification is  evident.   Furthermore, the predictions are
for clean water,  and if polluted water is to  be aerated, adjustments
on a percentage basis must be made, to account for the  effect of the
pollutants .

Since natural aeration is  the most  significant contributor in maintaining
the DO  level in a stream, it is advisable to measure the natural aeration
coefficient before completing the final design of an aeration system.
For preliminary design estimates,  either the method of O'Connor and
Dobbins or that of Churchill et al.  is probably  equally applicable.

Aeration transfer rates are quite sensitive  to water conditions. Table
4.3  shows the large range of aeration transfer rates that have been
measured for different water conditions.  (Values given are mass
transfer per unit of surface area per unit time. )  For still water,
values are on the order of .034 g/m2/hr; while for water conditions
involving small droplets, rates can be as high  as 34 g/m2 /hr.  Values
for flowing water lie between these two  bounds, which encompass  three
orders  of magnitude.

The aeration constant is  also a function of temperature.   Reported
values are usually given  in the form
                       K  (T) = K  (20) x
                        3,      2L
where
         K (T)  = aeration constant at temperature T
          a                                        —

         K (20) = aeration constant at 20 C
          3,

and 9 is  a factor ranging from 1.015 to 1.047. A commonly used
value is  1. 024.
                             19

-------
            TABLE 4. 3

 Transfer Rate of Oxygen into Water
from  Air for Various Flow Conditions
           (Odum [13] )
1
• Velocity
(m/sec)
Still water 0. 0
j 0. 0 ;
Moving water , :
Stirred water i
Shallow circulating trough | 0.01
1 0.01
i 0.01
| 0.013
i 0.070
! 0.119
i 0.20
Sewage in circulating !
trough ! 0. 05
! 0. 15
Stream and ponds !
New York Harbor i tidal
Tank with a wave machine '.
Sea Surface
Summer j
Winter i
Silver River, Florida |
Subtraction-of-respir - j
ation method ! 0.21
Dye-measured-turnover i
method j 0.21
Green Cove Springs, \
Florida. From •
carbon dioxide by [
respiratory-quotient !
method j 0. 3
Small rivers, diurnal
1 oxygen curve analyses
Ohio River below
Cincinnati 0.05-
0. 09
• Bubbles and drops (K given
per area of drop or bubble) ;
Air bubble :
Air bubbles
Water drops
Depth
(m)
-
-

_
0.1
0. 1
0. 1
0. 1
0. 1
0. 1
0. 1

0.45
0.45
-
-
1.8

-
-


2.77

2.77




0.23

0.5-3.

4.8



:
_
-
Temp.
f°C»
20-25
-

25
0-10
10-20
20-30
12
17
14
13

25-26
25-26
-
-


12-20
2-7


23

23




24

2

15-25



i 37
20-25
1 24
i
K
(g/m2/hr
at 10%
saturation)
.034
0. 03-0. 08

0. 09-0.74
0. 037
0.043
0.47
0. 12
0.52
1. 12
3.8

0.38
1.5 I
0. 08 '
0.23 '
0.31
i
1. 1
5.2


0. 92 :
i
1.00 j
|
)
».
i

0.55

0.6-4. 3

1.5-5. 0



13. 1
2.8-28.
22-34.
             20

-------
Biochemical Oxygen Demand (BOD)

Since the various oxygen demands  in a stream or river depend on
processes which proceed at different  rates, each one should be
described separately.  For example,  depending on the physical nature
of organic loadings, the fine particulate matter will oxidize in the
stream, while the larger particulates will tend to settle,  forming the
benthos,  and will oxidize at a different rate.  Furthermore, if the
loading contains ammonia or nitrites, oxygen demands  again occur at
different rates, and separate rate  constants are needed to describe
each of the processes,  which are commonly referred to as carbon-
aceous, nitrification,  and benthol oxygen  demand.

The oxidation of organic matter is essentially a chemical reaction,
initiated either directly by bacteria or indirectly by their enzymes.
As previously stated,  the oxygen demand  is exerted by two classes
of materials :  carbonaceous organic matter and oxidizable  nitrogen,
both of which may occur  simultaneously [ 14-1 8] . Figure 4.4 illustrates
the two processes and the clear distinction that they proceed at different
rates. It can be seen in  this case  that a period of at least  19 days was
required before oxidation was essentially completed.  The  upper curve
represents the total BOD of the water (first and second stage).   This is
readily obtained by measuring the  oxygen uptake of stream samples as
a function of time (see  reference 11).  The lower curve represents
only the carbonaceous BOD and can normally be obtained by following
the methods  given in Standard Methods [19], which call for the measure-
ment of the  oxygen uptake as methylene blue or allylthiourea.  The tests
are usually  performed  in a laboratory at 20°C.  By taking the difference
between the two curves,  a curve representing the nitrogenous BOD is
obtained.  (In both of these tests, time and temperature can be closely
controlled,  but it is difficult to simulate the dynamics of the river
environment, including the biological chain. )

The carbonaceous BOD follows a first-order reaction process,  viz,
the rate of biochemical oxidation is proportional to the  remaining con-
centration of unoxidized substance. Such a process is described
mathematically by

which yields,  upon integration,


                               -K t
                       L = LQe  C                               (4.9)

where

         t   =  time (days )

         L  = carbonaceous BOD at  time_t (ppm)
                               21

-------
Q
O
PQ
                            12      16      20     24


                        Time (days)
  Figure 4.4.  Grand River BOD Curve (Courchaine [18])
                          22

-------
         L,   = ultimate carbonaceous BOD (ppm)
        K   = c
       arbonaceous deoxygenation constant (-3	)
K  can be obtained from the slope of a semilog plot of Equation 4. 9
and is usually on the order of . 1/day at 20 C.  In order to evaluate
Kc for a stream it is not necessary  to conduct the test for 20 or more
days, since the slope of a semilog plot of Equation 4. 9 can be quite
accurately determined over a time period of 5 days. However, the
ultimate carbonaceous demand Lo must be correctly stated.  This
can be clearly illustrated by examining Table 4.4 below. It can be
seen that for Kc - . 1 /day,  at the end of 5 days only 68.4% of the
carbonaceous BOD has  been consumed; hence,  the  ultimate car-
bonaceous  BOD would be
                                  BOD
                   BOD
                        ult
                  = L  =
                     o
                                       5 day
684
                          TABLE 4.4.
             Percentages of Carbonaceous BOD as  a
                 Function of Time for K  = .  1 /day
Time (days )
0
1
2
5
10
20
Remaining BOD (%)
100
79.4
63.0
31.6
10.0
1.0
Consumed BOD (%)
0
20.6
37.0
68.4
90. 0
99.0
Both K  and L,  depend on temperature.   The following expressions
were  extracted from a recent study [54 ] .
LQ(T)  =
                         (1 + 0.0113 (T  - 20))
                                           ~ - O   0 r O .-,
                                     range 20  - 35  C
         LQ(T)  =  LQ(20) (1 + 0.0033 (T - 20))
                                     range 2° - 20  C
                             23

-------
         Kc(T) =  Kd(2Q) (0.896) (1.126T-15)
                                           ->O   ->o°r-
                                     range Z  - 32 C

                                T 20
         K (t)  =  K      (1.047     )
Kc(T)  = K  2    (1.728) (0.985T-32)
                                           ic°   ->->r~
                                     range 15  - 32 C

                                      5T-32)
                                           ->->°   ,,nO^
                                     range 32  - 40 C
The oxidation of carbonaceous matter is a heterotrophic process
carried out by a great variety of different organisms and having
optimum temperatures ranging from 18° to 25°C.  Many of the
bacteria obtain their food and energy requirements from the organic
matter present in the water and have generation times on the order of
20 to  30 minutes [14].   On the other hand, nitrification is an autotrophic
process, i. e. , one which is  carried out by specific bacteria which
obtain food and energy from  oxidation of ammonia and nitrites.  The
nitrogenous oxidation process involves two genus organisms, Nitro-
somonas and Nitrobacter,  each of which have an optimum growth
temperature  range of 25° to  28°C [20].  Furthermore,  generation
times on the  order of  31 hours are necessary for nitriting  cells to
develop [14].  Often nitrification may not occur for several days,  e..g.,
in a study on the  Passaic River [ 6]  nitrification did not begin until
3 days had elapsed and,  in the case of the Grand River,  a situation
was found in  which nitrogenous BOD did not occur until after 9  days
(see Figures 4.5 and 4.6[18].

Thus, it can  be concluded that BOD tests should be conducted over a
long-enough period to  determine if and to what  extent nitrification
processes  will occur.   This  is especially important if a  river or stream
may receive  the effluents from a secondary treatment plant.

The ultimate oxygen demand due to  nitrification is directly limited by
the amount of oxidizable nitrogen available.  The process involves
the oxidation of ammonia to nitrite, which is carried out by the genus
Nitrosomonas, and the second phase,  the conversion of nitrite  to
nitrite,  which is  carried out by the  genus Nitrobacter.  The  overall
stoichiometric relation is given below:
                    NH3  +

            (M. W.) —14     64      62
By comparing molecular weights, one finds that 1 part of NH3 will
consume 4. 57 parts of O2  by weight.  Thus,  1 mg/1  of ammonia is
equivalent to 4. 57 mg /j.  of BOD,  and 1 mg/1  of nitrate produced by
                              24

-------
                        22
IV
Ul
                         18
                      (D
                      4-)
                      a
                      D
                      0)
                      Cud
                      >>
                      x
                     O
                         14
10
                                                                               Carbonaceous +

                                                                                 Nitrogenous
                                                                         Carbonaceous
                                                                                J	L
                                                      6810

                                                          Time (days)
                                                        12       14      16
                                    Figure 4. 5.  Total Carbonaceous and Nitrogenous BOD

                                                 as a  Function of Time     (Whipple  [6])

-------
           40
ro
           32
           24
      BOD

     (mg/1)
           16
                                                                                   Carbonaceous BOD
                                                             10
12
14
16
18
20
                               Figure 4. 6.  BOD Curve, Lansing Wastewater Treatment Plant,
                                           24-hr  Composite  Sample,  Taken  July  12,  I960.

-------
the oxidation of ammonia is  equivalent to 1.03 mg/1 of BOD.  It is
expected that some variation from  the above ratio will exist because
of fixation by carbon dioxide, lowering the ratio of 4. 57 to perhaps
4. 3 or 4.4 [21] .

The rate of oxygen consumption by ammonia and nitrogenous matter
is directly related to the multiplication of the nitrifying bacteria.  It
can clearly be  seen from Figure 4.6 that the nitrogenous BOD curve
consists of two phases.   In the  first phase the nitrifiers are lagging;
the bacteria are undergoing  multiplication,  building up to  a maximum
population. Once this maximum is reached, the process enters the
second phase (corresponding to the point of inflection), which is
similar in shape to curves obtained for carbonaceous  BOD.  Such a
process is clearly not first  order.  However, for mathematical
expediency in predicting  oxygen sag curves for a stream,  the process
is frequently assumed to be  a first-order reaction [6, 22, 23],  i.e.,
one in which
                        = K  N = K (N  - N.)                   (4. 10)
                    dt     n     nv  o    r                   l      '

where

         N   =  nitrification demand (ppm)

         t    =  time (day)

         K   =  nitrification constant (day  )

         N   =  ultimate nitrification demand (ppm)

         N   =  nitrification demand at time_t (ppm)

and which integrates to

                                 -K t
                    N = NQ(1 - e  n ) '                       (4. 31)


In order to take into consideration the lag time associated with the
nitrification process,  Whipple et al. [6] found it advantageous  to
include a lag-time  term t  in the following manner
                        3,

                                 -K  < t  - t >
                    Nt = NQ(1 - e  n       a  )                (4.12)


where the triangular brackets represent the singularity function such
that
                    < t - t  >  =0      f o r t < t
                         a                    a

                    < t - t  >  = t-t   for t > t
                         a          'a         a
                                27

-------
An important point when evaluating the various constants is the fact
that the time lag ta can be a function of where and when a stream
sample is taken.  If oxidizable nitrogen was introduced near the point
at which the sample was taken,  long lag times can be expected.
However, if the sample was taken a considerable  distance downstream
from  the introduction of nitrogenous matter, short lag times may
result,  since  incubation may have initiated during the transit time.

By describing the nitrification process as a first-order process, an
upper bound is obtained on the BOD, since such a representation
results in BOD  values larger than those actually occurring (see
Figure  4.7).

As mentioned earlier, the effect of temperature is quite significant.
According to O'Connor [24] , the  temperature dependence is given as
                    Kn = Kn(20)  1-°9    "                      (4. 13)


More detailed investigations of the twofold nitrification process have
been conducted by several authors [ 25-28] ,  in which the  rate of change
of nitrogenous BOD is expressed  in terms  of ammonia oxidation and
nitrite oxidation.  The expressions are relatively complex,  nonlinear
in nature,  and,  to be of practical use, they often require the assistance
of a digital computer.  In view of the many environmental conditions
affecting nitrification (temperature,  pH, chemical composition of the
water, the specific genus of nitrifying organism present,  etc.) and
the variability of determining  other terms  in the equations used to
predict the oxygen profile of a stream (Equations  4.2 through 4.4), it
may be concluded that the effort involved in obtaining the precision
offered by the more refined approaches  is" questionable.  For stream
aeration design purposes, the use of Equation  4. 12 will result in oxygen
demands in excess of what actually is  required for nitrification.
However,  in view of the many variables affecting  stream aeration,  its
predictions are perhaps as good as any others.
 Benthal Demands S
 One of the best methods of determining benthal demands S_ is to measure
 the oxygen consumption of the deposit "in situ." This can be effectively
 performed by using an opaque respirometer which is only open to the
 stream's bottom surface.  By monitoring DO levels and subtracting
 the oxygen consumed  by carbonaceous oxidation and nitrification at
 the same time and temperature,  the net oxygen demand due to  the
 benthos  can  be  determined.  Given the time period over which  the
 consumption occurred, one can determine the benthal oxygen demand
 on a rate basis, usually expressed as ppm/day, or, when multiplying
 by the stream height H_,  as g/m2 /day.  Having the  results in the latter
 form and knowing the percentage of the river bottom covered with
                               28

-------
BOD
(ppm)
                    First-Order
                   Approximation
                                  Time (days)
      Figure 4. 7.  First-Order Representation of a Nitrification Process
                                 29

-------
similar mud deposits permit the determination of the river demand on
a daily basis.  This approach gives an oxygen demand rate which is
indicative of the steady-state consumption of the  benthos.  Making
these measurements during  summer months  as opposed to other times
of the year eliminates the need for temperature corrections when
determining  the needs of an  artificial aeration system,  even though
such biological uptakes are temperature dependent.  Since benthal
demand rates are generally  fairly constant and low,  the "steady-state"
value associated with the higher consumption rates during the summer
months  is a safe value to use for aeration system design.
Net Photo synthetic Oxygen Production P
The production of oxygen by photosynthesis in streams has been con-
sidered by several investigators [6,  13, 29-35] .  Depending on the
type of stream and associated climatological conditions, the contribution
can be significant.  Relevant parameters include presence of aquatic
plants, especially free-floating and benthic algae;  stream clarity;
intensity of sunlight;  stream temperature;  and stream velocity.  A
study conducted on the Passaic River in New Jersey[6] indicated that
the eutrophic zone was found to be limited to 50 centimeters, while
investigations conducted on streams in Oklahoma  [31, 35] which have
relatively low turbidity have indicated that photosynthetic productivity
in the benthos is  considerable and at times in excess of that  which is
free-floating.  Other studies,  conducted on the Sirsloch River [29]
in the U.S.S. R. and on the Rhine River in Germany  [36]  , have also
substantiated the finding that photosynthetic  production of oxygen can
be significant.  In general, it can be concluded that the contribution of
photosynthetic aeration should be considered when designing an aeration
s ys t em .

Photosynthetic oxygen,  by its nature, is diurnal in production and conse-
quently is reported as a net production per 24-hour period, usually
expressed  in mg / /day.   A widely known method for measuring photo-
synthetic oxygen is the light-dark bottle technique.   The measurement
procedure  requires one  clear and one opaque bottle filled with stream
water, which are incubated at several depths in the stream.   At the
end of the incubation  period,  which may run from one to several days
depending on the  rate of oxygen production,  the difference in oxygen
concentration between the water in the two bottles  is assumed to be
equal to the photosynthetic oxygen production.

The light-dark bottle technique requires the assumption that biological
processes  in the  bottle are the same as those in the  stream.  This is
not a completely  valid assumption because the quiescence imparted to
the samples affects the natural exchange of food and waste products to
and from the organisms, and also because a higher surface-to-volume
ratio is introduced in the sample bottles.  In addition, this technique
neglects photosynthetic oxygen from benthic algae  or attached plants,
an assumption which  can only be made in relatively deep or  turbid
                             30

-------
streams, where sunlight cannot penetrate to the  benthos.  If measure-
ments are being made on a relatively clear,  shallow stream, they
could easily be  in error.  For this type of stream, Stay et al. [31]
have developed  a method which uses  three plastic chambers  placed
over the stream bottom.   Two sets of measurements are required, the
first with plastic bottom plates covering the  benthos and the second set
without plates.   Each of the three chambers  is employed differently.
One is clear and closed to the atmosphere,  the second is clear and
open to the atmosphere, and the third is black and closed.  The oxygen
concentration in the closed, clear chamber is affected by photo-
synthesis and respiration,  while the  oxygen in the black chamber is
affected only by respiration.  The clear chamber open to the atmos-
phere  is a control.  The tests are run with and without bottom plates,
in order to account for respiration and photosynthesis of the benthos.
The differences  in concentration between the clear and dark chambers
is the  photosynthetic contribution.

The oxygen production should  be expressed in terms of daily average
rate (ppm/day). A more precise expression can be derived to  account
for the variation in sunlight intensity during  the day.  This precision
is not required, however, when the object is to design an aeration
system.
Longitudinal Dispersion

Longitudinal dispersion is the process of spreading and distributing
a mass of pollutant or oxygenated water along the length of the stream.
The spreading is a result of two diffusion processes, molecular
diffusion  and  eddy  diffusion.  In turbulent flow the effect of molecular
diffusion  is negligible compared to  that of eddy diffusion and conse-
quently can be neglected.

In aeration system design,  the rate of spreading and the resulting
distrubition are important.  The dispersion characteristic of flowing
water has been found to vary greatly [36-42].   Table 4.5 contains
values of dispersion coefficients Dj_, determined  by  various investigators
for different rivers.

Examination of the table  shows that longitudinal diffusion coefficients
are larger for estuaries  than for a  conventional stream or river.
This is primarily due to  the mixing effects that occur in estuaries.
An interesting point is  the considerable increase in the  diffusion
coefficient of the Delaware River as a result of using a  diffuser aerator
(141,000 ft2 /sec,  compared with 1075 ft2 /sec).  The investigators
speculated that the large value of the dispersion  coefficient was a
consequence of turbulence caused by the large bubbles  emitted by
the diffuser,  and that a small-bubble diffuser might give smaller
values.

Several approaches have been devised to predict the magnitude of the
longitudinal diffusion coefficient. Some of the schemes  have only
                            31

-------
                              TABLE 4.5

         Typical Longitudinal Dispersion  Coefficients for Rivers

Delaware (estuary) [40]
James (estuary) [49 ]
Hudson (estuary) [40 ]
Delaware (estuary) [37]
Delaware (estuary) [44]
Copper Creek, Virginia
[42]


Clinch River, Tenness ee
[42]


Powell, Tennessee [42]
Coachella Canal,
California [42]
Depth
(ft)

	
	
	
	
1.6
2.8
1.6
1.3
2.8
7. 0
6.9
1.9
2.8

5. ]
Width
(ft)
_ _
--
--

--
52
60
53
61
154
195
175
118
111

80
Discharge
(ft3/sec)
	 _
	
	
	
	
54
300
48
483
323
3000
1800
240
140

950
DL(ft2/sec)
195-350
325
4, 850-16, 100
1075
141, 000*
210
230
102
106
150
580
500
87
102

103
*with artificial aerator in operation
been applied  to laboratory channel conditions, while others have been
verified with natural stream data.  The "routing procedure"  as
described by Fischer [42] appears to give the best predictions when
compared to  dye-concentration measurements.  However,  this approach
requires the  use of tracer dyes and is mathematically involved to  the
extent of requiring the use of a computer.  If cost and personnel
limitations do not  permit use of the routing technique, the disperson
coefficient can be  estimated from field measurements of velocities and
water-surface slope.  This technique has also been developed by
Fischer [4]] and will only be briefly described here.   (If one is  faced
with evaluating the diffusion coefficient of a stream,  references 41 and
42 should be  consulted.)

The procedure involves measuring the slope of the stream and the cross-
sectional  geometry and velocity distribution at one or more typical cross
sections.   For many stream situations these field measurements  can
usually be completed in a day by a three-or four-man crew.

Equation 4. ]4 is used  to calculate DT  for streams having a width-to-
depth ratio equal to or greater than o.
                               32

-------
_!_
A
                                f
q'(z)d(z)
(4.14)
where
                                d(z;
                       q'(z)=  J   u'(y,z)dy

                               o
         E       =  .23 dU  transverse turbulent mixing coefficient [42]

         u'(y,z)  =  velocity at any point relative to the mean flow
                    velocity, i.e.  , u'(y,z) = u(y,z) - u",  where
                    u(y,z) is the actual velocity

         q'(z)    =  depth of flow at any point on  the cross section (ft)

         A       =  area of flow (ft* )

         b       - width (ft)

         y       =  vertical coordinate

         z       -  transverse coordinate
          *              1/2
         U       =  (grSe)    function velocity,  where  g = acceleration
                    of gravity,  r = hydraulic radius, and Se = slope of
                    the energy gradient
Although the calculation of D^ appears  complicated and evaluation by
hand is tedious, the equation can easily be programmed for a computer.
Normally, depth and velocity measurements are taken at a minimum
of 20 vertical sections across the stream.  The integrals in Equation
4. 14 are then approximated by summations, viz,
                     n
                    k=2
                                                        Az
                               33

-------
where
         q'.   =  i(d. + d.   )u'.
           i      Z %  i    i+l   i

         u.    =  velocity in the i   vertical slice (ft/sec)

         u1.   =  u. - u (ft/sec)

         u    =  mean velocity of the flow within the entire cross
                 section (ft/sec)

         d.    =  depth at the i  vertical slice (ft)

         Az   «  width of vertical slice (ft)

         n    =  number of vertical slices (usually more than 20)
                        #                                        th
         E    =  . 23d.U  transfer coefficient between i-1 and the i
          z.          i  .  n.
           i      vertical slice
Table 4.6 has been reproduced from reference 42 and illustrates the
procedure to be followed.


Simplification of the Oxygen Balance Equations (4.2-4.4)

Equations 4.2 through 4.4 describe the temporal and spatial distribution
of BOD and DO.  The flow volume and  the cross-sectional area are
functions  of both space and time. In order to simplify the treatment,
the assumption is usually made that the most severe condition for
design purposes  is the  one which can be made to approach steady state.
Although steady-state conditions are seldom,  if ever, attained,  the
errors  introduced by this assumption are minor compared to variations
occurring in the  physical stream and in measurement procedures.

If,  in addition to the steady-state conditions, the following assumptions
are applicable:
         1.    Stream flow is uniform.

         2.    Carbonaceous and nitrogenous BOD are first-order
              reactions .

         3.    Photosynthetic oxygen,  benthal demands, and accrual
              processes are uniform along a stream section and are
              determined on an average daily basis.

         4.    Vertical and transverse variations are negligible.
                               34

-------
                                                         TABLE 4.6

                           Sample Calculation of Dispersion Coefficient from Field Data Test [41]
Data





Vertical
slice
(1)
)
2
3
4
5
6
7
8
9
JO
) 1
12
]3
14
15
16
17
18
19
20
21
22
23
24
25
i 	

Distance


from left Depth at ]
bank to
left side
start of I of slice
slice (ft) d. (ft)
(2) l(3) i
0.0 : 0.00
2.5 i 1.00
5.0 2.15
7. 5
10.0
12.5
15.0
17.5
20. 0
22.5
25.0
27.5
30.0
32. 5
35.0
37.5
2.61
3.30
3.41
3. 32
3. 14
3.00
3.01
3. 10
2.99
2.78
2.64
2.59
2.66
40.0 2.90
42.5 ! 2.98
45.0 '• 2.84
47.5 2.78
50.0
2.72
52.5 2.39
55.0
57.5
J.82
0.84
60.0 0.34
Computed Quantities




Mean velocity
in s lice p. .
(ft/sec)1
(4)
0. 15
0.25
0.35
0.42
0.65
J. 35
1.80
2.30
2. 35
2.40
2.50
2.55
2.70
2.75
2.65
2.45
2.30
2. 15
1.85
1.50
1. 10
0.70
0. 40
0.20



Discharge through
slice relative
to mean velocity
q! &z (ft/sec)
1 (5)
-2. 00
-5.89
-8. 31
-9.80
-9.20
-3.33
Cumulative
relative
discharge
J
Z q1. A z
i=l. l
(ft3)/sec)
(6)
-2. 00
-5.89



k A7 J-1
•*- "^ •^ ,-,< i\ -
Z E7 d ^ qi
j=2 Z) j i=l
(ft)
(7)
0. 0
-65.7
-16.20 { - J 2 1 . 9
-26.00 -200.2
-35. 19
-278.9
-38.53 -378.5
0.43 -38.09 ' -493.7
4.25
-38.84 -620.9
4.53 -29.31 : -744.8
4.99 ! -24.32 -851.2
5.74 ' -18.58 : -934.2
5.80 -12.79 • -1,003.2
6.46 -6.33 ; -1,057.7
6.56
5.93
4. 90
4. 07
2. 94
0.73
-1.69
-4. 13
-5.5]
-4. 48
-2.28
0.24 ; -1,087.6
6.17 : -1, 086. 4
1 1.05
15. 12
18.06
-1, 057.7
-1 , 014. 4
-958. 3
18.79 : -884.6
17.09 . -804.5
12.96 ! -728.4
7.46 j -653.6
2.98 | -579.5
0.70 i -440. 3
0. 10 -0.70 0. 00 ' -241. 0
1
^ ~Z[ column (5) x column (7)
34,800 ft4 /sec _.„ ft2 ,_
LJ — 	 	 — 	 — s 	 — u J'-l it / ^ C <_
total area 149 ftz
IjO
Ul

-------
then considerable simplification of Equations 4.2 through 4.4 results:
    )   ^-^  + K  (c  - C) -  U^ -KL+P-KN+A-S=0       (4.16)
    L. j - 2     a  s         dx    c         n                   v     ;
       dx^
       TT
       dx
   D        _ U     _  K N+ N = 0                               (4.18)
    L2      dx
Dobbins [45] has shown that,  for freshwater streams, neglecting the
longitudinal  diffusion terms results in a prediction which is at most
a few per cent different from a solution which includes the effects of
                           U2
diffusion.  If the value of -^  p— is on the order of 100 or more,
                            a L
neglecting the diffusion term  results in a negligible difference.   It
should  be noted, however,  that the value of DL should include the
effects  of the aerator system (see  Table 4.5.)

                                                U2
If one is dealing with an estuary, the ratio  of ?       may be quite small,
                                                a L
meaning that the diffusion is significant. This would be indicative of a
relatively slow-moving body of water which experiences good mixing
(e.g. ,  tidal  effects).  If one  concludes  that the diffusion terms are
negligible, Equations 4. 16  through 4. 18 reduce to
                -K(C   -C) + KL + KN-P-A+S             (4.19)
           dK     av  s     '     c      n                          v
                +  KL-L  =0                                   (4.20)
             x      c      a
                +  KN-N=0                                   (4.21)
           dx     n      a


 Equations 4.20 and 4.21 can be integrated to give
                              36

-------
                          KC
                       ~  "zrr  X.     JL/
              T  - T  0    U     4-   a
              JL = _L e          t  ——
                    o              K
                                    c
                          K
                       - ~nx    N
              N = N e    U     + T-r-^
                    o             K
                                    n
which can be substituted into 4. 19 to give
                          K               K
                           c               n
       K                 ~  TI x,          ~ T T x
  dC    a,r                u   + K N e   u   =P-L-N+A-S
  te~ TT(Ca-C) +KcLoe         n  °                a    a
                                                                 (4.22)
or
                       K                K
       K                c                n
  dD ,  a  p.  v ,   "UX+KNe   "  U     . T   ._,    _   A^c
  j— + -7-rD=KLe            no          +L+N-P-A+S
  dx   U       c o                             a    a
                                                                 (4.23)

where D represents the oxygen deficit (C  - C).

If nitrogenous BOD is small compared to carbonaceous BOD,  the  two
can be combined in a single  expression for BOD.  Hence, Equations
4.4, 4.18, and 4.21 may no't be needed in many stream analyses .  The
effects of nitrification would be contained  in Kc,  L,  and La and would
not be differentiated from  carbonaceous consumption of oxygen.  This
has often  been done in many of the analyses  that appear in the open
literature.
                             37

-------
                           SECTION V

                AERATION SYSTEMS  FOR RIVER
                 AND STREAM APPLICATIONS
Introduction

Various types of mechanical surface aerators, diffusers, and side-
stream mixing systems have been used in experimental river and
stream aeration  projects.  Downflow contactors (U-Tubes),  although
not used previously for river and stream aeration,  have been used
successfully in impoundments and lakes and should be adaptable to
river and stream applications.

In this section the aeration systems which could be used for  rivers
and streams are described, and results of recent tests using these
systems  are evaluated.  The specific application of any of these
systems  is  extremely site dependent,  thus  site considerations are
included in  the discussion of each system.  In most cases, aeration
devices designed specifically for waste treatment have been  used
without modification in the stream application.  Some  improvements,
particularly for surface aerators,  can  be made to enhance their per-
formance when used in stream environments,  since strong pumping
action is not as  important there  as it is in waste treatment.

Downflow contactors,  diffusers,  and sidestream mixing systems can
be used with both air and pure oxygen,  and the relative merits of  the
two approaches  are discussed in this section.  Since the efficiency of
diffuser systems depends  strongly on the diffusion mechanism for
small  bubbles, this subject is included in the discussion of diffusers.
Mechanical Surface Aerators

A variety of designs for mechanical surface aerators are available
from a large number of manufacturers.  There are no designs of
surface aerators specifically for river and stream aeration, available
systems being used only in waste-treatment applications.  A compre-
hensive description of surface aerators can be found in Manual of
Practice No.  5, published by the Water Pollution Control Federation
[47 ] . Some of that information is included in Appendix A.

For river and stream applications, it is desirable that  the system be
float-mounted, so that a wide range of flow conditions  can be accommo-
dated.

The following types of surface aerators have been developed. (See
Appendix A for more details:
                             39

-------
         Rotating Plate   -  Creates a peripheral hydraulic jump that
                            accomplishes oxygen transfer through
                            entrainment.

         Updraft          -  Pumps large quantities of water at the
                            surface at relatively low heads.  Aeration
                            efficiency related closely to efficiency
                            as a pump.

         Downdraft       -  Oxygen is supplied by air self-induced
                            by negative head produced by rotating
                            element.

         Combination     -  A rotating element and a sparge ring are
                            combined to transfer oxygen by dispersing
                            compressed air fed below the  surface to a
                            rotating agitator or turbine.

         Brush           -  A horizontal revolving shaft with attached
                            brush-like elements extending slightly
                            below the surface.


The transfer of oxygen by surface aerators to a body of water is a
direct result of breaking  the water into( small droplets and inducing
turbulent mixing  on or near the water surface.  Each process causes
new unsaturated water surfaces to be exposed to the air,  resulting in
their  becoming saturated and subsequently mixed with the bulk of the
water.

A significant difference between using surface aerators in streams and
using them in lakes or still bodies of water is the superimposed effect
of the  stream current.  The current provides the aerator with a supply
of water low in DO compared to water in the  mixing zone of influence
of the  aerator.  Hence, the DO deficit will tend  to be greater, enhancing
the transfer  of oxygen.  Furthermore,  since part of the energy  con-
sumed by surface aerators is  in the pumping of water,  aerators which
expend relatively low pumping energy should be considered for  stream
applications.  The pumpage associated with surface aerators  should
be supplied by the manufacturers.  For example,  the pumping rates
published by two manufacturers for various horsepower units are
listed  in Table 5.1.

In a recent study, McKeown [47] measured velocity profiles and DO
concentrations in stabilization basins as affected by surface aerators.
Although the study dealt with basins and not streams,  the results are
characteristic of surface aerators and can provide preliminary informa-
tion for  stream applications.  The investigation included a  number of
different sizes of surface aerators.  Figure  5. 1  shows the  extent of
influence of a 75-hp unit.  The profiles shown are typical of surface
aerators .
                             40

-------
          Distance from 75-hp Unit
                                                                                125 ft
                                                    Velocity-Depth-Vector
Figure 5. 1.  Example of Velocity Profiles at Various Distances from a 75-hp Aerator
                                          (McKeown [47])

-------
                             TABLE 5. 1

                   Surface Aerator Pumping Rates
Horsepower.
5
7.5
10
15
20
25
30
Manufacturer A
"Aqua -Jet"*
3400
3800
5000
6] 00
8300
9800
12,500
Manufacturer B
"AQUARIUS"**
2000
3400
4000
5000
6500
7500
8000
             Aqua Aerobic Systems,  Inc., Rockford,  Illinois
             Keene Corporation, Aurora,  Illinois
Table 5.2 [47] lists for various size aerators the depth of penetration
of outward flow,  the radius at which this depth is reached,  and also
the radius  at which all discernible outward flow  at the surface ceases.
The position of the aerators was such that there was no interference
from adjacent aerators or basin banks.   It is interesting to note that
the depth of penetration of the crossover point rarely exceeded 50% of
the bas in depth.

Table 5. 3 [47]  contains  data on influence of flow and the net  DO con-
centration  changes at each  foot of depth and at varying distances from
the aerator.

It was concluded that the maximum zone of influence for different-size
aerators was  300-400 feet for 100-hp units, 200-250 feet for 75-hp
units,  150-250 feet for 40-to 60-hp units , and out to 150 feet for
1 0- to 25 -hp units.

Attempts to compute the  pumpage associated with the various aerators
were not successful, primarily because  the flow associated with the
surface  aerators is three-dimensional and the data collected did not
consider the vertical dimension.

In another  study conducted  by Burns et al.  [48]  , two 15-hp surface
aerators were used to raise the DO in the Jackson River near Covingtoi
Virginia.  It was reported that,  by placing the aerators at  sag points,
transfer efficiencies on  the order of 2.2 pounds  O?/hp-hr (20°C and
0 upstream DO) were obtained.  Average stream velocities were
approximately 3/8 ft/sec.

Kaplovsky et al. [ 49] conducted an investigation on  aerating  the foreba^
of a canal  at Lockport, Illinois,  using two model 100 Hi CoWave
                             42

-------
                          TABLE 5.2
                                          Vf.
         Zone of Outward Aerator Influence (McKeown [47])
Type
NPHP
f
L
L
L
H
JL
L
L

L-
H
H
H
L
L
L
L
10
10
DHP

8
8
10 8
25 2]
40 i 41
40
40

40
50
27
34

26
47
50 47
50 i 50
50 1
50
60 52
60 52
L ; 60
L ' 60
L 75
L 75
L 75
49
49
54
64
55
H 75 64
L i 75 1 59
L 100 j -
L ' 100
L
100
-
-
D
(ft)

11
12
1 1
10
8
8
8

8
13
14
9
12
1 1
11
14
14
12
8
14.
14
10
10
18
18
18
\ '
d
(ft)

5
3
4
2
2
2
3

2
7
6
3
5
6
6
3
3
4
2
5
4
2
5
9
9
7

r}
(ft)

50
50
75
40
r2
(ft)

110
80
100
60
50 , 80
50
50

50
50
50
50
60
60
75
50
75
100
100
75
75
40
60
150
100
100

80
80 i
1
80
100 !
80
60
110
110
110 j
75
110 i
125
110
125
no
60
100
200 !
150 ;
175
i
	 *
NOTE:    (1)  Nearest restriction (bank or aerator) at least 2.5r->
              distant from aerator reported.
          (2)  Symbols:
                 L  = low speed aerator
                 H
             NPHP
              DHP
                 D
                 d

                rl
high speed aerator
name plate horsepower
drawn horsepower
basin depth
maximum depth of penetration of outbound flow
radius from aerator shaft to point of maximum
depth (d)
radius from aerator shaft to point where
surface outward flow blends with background
                          43

-------
                         TABLE 5. 3

Sample DO Profiles  at Various Distances from Aerators (McKeown [47])
                    (values shown  inmg/1)
i
60-hp Unit i
	 	 1_
i
!
1
100-hp Unit |
1


25-hpUnit


50-hp Unit


10 -hp Unit




RadialDistanc e i
Depth 	 [
0 !
1 i
2
3 :
4
5
6 :
7 ;
8-10
11-13 :
0 '
1
2
3 ;
4 ;
5-12 i
!
1 !
2 !
3
4
5
6-10
0
1
2
3
4
5
6
7-10
0
1
2
3
4
5
6-10
25 ft
2.4
2 . 4*
1. 9
1.8 !
1.5
1.4
1.4
1.2
1.2
1.2
2.0
1.4
0.2
0.0
0.0
0. 0
2.7
2.3
2.2
2.2
1.9
1.9
1.4
2.5
1.7
1.5
1.4
1. 1
0.8
0.6
1. 0
2.0
2.0
1.9
1.6
1.5
1.5
1 1'5
-.
50 ft i 75 ft
2.1 i 2-0
2.0 ; 1.8
2.0 1.7
1.8 ; 1.7
1.5 ; 2.0
174" 1.6
1.4 1.9
1.4 1.5
1.4 1.4
1.3 1.2
1. 0 0.4
0.4 0. 1
0.2 ; 0.0
6.0 j 0.0
0.0 i 0.0
0.0 : 0.0
2.7 ' 2.2
2.6 ' 2.1
2.4 : 2.0
2.3 ; 1.9
2.4 ' 2.0
2.2 i 1.9
1.9 ;' 1.7
1.5 \ 1.6
1.5 1.4
1.3 1.4
1.3 1.3
1.1 1.3
1.0 1.3
1.0 1.3
1.0 1.3
1.9 ' 1.7
1.7 1.5
1.6 1.6
1.5 1.5
1.5 1.5
1.5 1.5
1.5 • 1.4
100 ft
1.7
1.5
. 4
1.3
L_o
1.2
1.2
1.2
1.2
1.2
0.0
0. 0
0. 0
0. 0
0.0
o.o


1.9
1.4
1.5
1.4
1.4
1.4
1.3
1.3
1.6
1.5
T7T
1.4
1.5
K5
i I-3
Velocity crossover  zones are underlined
                          44

-------
Aerators  (73 hp each).  Flow rates varied from 800 cfs to 5600 cfs,
and the average depth of the bay was 15 ft.

It was  found that at low  flow rates (1000-3000 cfs),  transfer efficiencies
on the  order of 1. 8 Ibs  Oz/hp-hr were obtained, while for high flow
rates (4000-6000 range),  efficiencies increased to on the order of 4.4 Ibs
G>2/ hp-hr (see Figure  5.2).  The values quoted above have been correc-
tec to standard conditions,  using the average of upstream and  downstream
concentrations.

Figure 5.2 illustrates the differences in oxygen transfer at different
rates of flow.  The rapid transition shown is likely to occur if flow
conditions changed,  e.g., laminar to turbulent.  However,  from the
data presented in the report no evidence of such a phenomenon could be
found.   It is quite possible  that a more realistic curve of increasing
efficiency would show a gradual increase, rather than an inflection.

The authors speculated that the appreciable increase in efficiency might
be attributable to increased shearing of bubbles in the upstream region
or greater retention time of the bubbles in  the water.  However, it is
more likely that, as the stream velocity increases,  more unoxygenated
water is fed to the aerators,  resulting in a greater O-> deficit and a
greater rate of transfer.

Whipple et al. [ 6] also found from studies  using surface aerators on
the Passaic River (average width 100 ft , average depth 7 ft in low
flow) that an increase in efficiency occurred with flow rate, as shown
in  Figure 5.3.  The efficiency increased from about 1.2 lb  C>2  /hp-hr
at  120 cfs to approximately 2.8 lb/O2/hp-hr at standard conditions.
These values are consistent with those of Kaplovsky.  However,  the
transition to higher efficiencies  is  more gradual and perhaps more
representative of the actual situation.  Some question remains as to
what happens to the efficiency when the discharge approaches zero.

In  contrast to the results of Kaplovsky and  Whipple, Susag et al. [50]
found in a series of laboratory tests that very little  difference in
oxygen transfer efficiencies occurred between flow and non-flow
conditions.  His average  results, corrected to standard conditions
(according to the aeration equation) for a 9-in.  and a 12-in,  unit,
are contained in Table  5.4.

In  the authors' opinion  the values associated with the 9-inch unit were
more representative, since very little splashing on the sides of their
test tank occurred during that test.  The velocity range investigated
was from about . 1 ft/sec to .65  ft/sec for the flow-through tests.

In  addition to his tests  on the Passaic River [6] , Whipple directed a
series of  tests of surface aerators and bottom diffusers on  the Delaware
River near Philadelphia in  order to determine the practicality of
oxygenating a deep,  navigable river [44] .   The intent of the study was
                             45

-------
a
-C
—.
 N!
O


-------
4.0
200       400
                                  600
800       1000



Discharge (cfs)
1200       1400       1600  1700
         Figure 5. 3.  Oxygen Transfer Rate vs Flow for  Mechanical Aerator (Whipple [6])

-------
                          TABLE 5.4

             Average Oxygen Transfer Efficiencies,
                 in Ibs O2/hp-hr (Susag [50]  )
Turbine Size
9 inch
12 inch
Flow
3. 73
4. 11
No Flow
3.89
4.86
to test systems on a wide,  deep,  navigable stream to determine
efficiency, economy, and operating characteristics and to prepare
prototype designs and cost estimates  for installations appropriate
to such rivers.  This is in contrast to the study on the Passaic  River,
which is a relatively small river.

The oxygen deficiency on the Delaware extended over a distance of
about 40 miles,  including areas containing heavy industry.  The river
is characteristically 2000  to 2500 ft   wide above the confluence of
the Schuylkill and somewhat wider below this  point.  The main channel
is 40  ft  deep, with adjacent anchorage areas of about 30 ft.  Normal
velocity range is ] to 1-1/2 ft  /sec.

Oxygen transfer rates were reduced to standard conditions.   The  a
and p  values  were found to be  equal to ].  Before  being reduced to
standard conditions, the surface  aerator transfer rates varied from
1.18 Ibs Oz/hp-hr to 3. 78 Ibs Oz/hp-hr, with an average of 2. 56  Ibs
Oz/hp-hr; after conversion, the range became 1.29 to 4.50,  with an
average of 3.06  Ibs  Oz/hp-hr.

These values are substantially higher than the average transfer rates
for the mechanical aerator tests  conducted on the Passaic River [6].
The average  field  condition of 2. 56 Ibs Oz/hp-hr on the Delaware
compared to  1.04 Ibs Oz/hp-hr on the Passaic, and under standard
conditions the averages were 3.06 Ibs Oz/hp-hr and 2. 12 Ibs  Oz/hp-hr,
respectively.  The above results are  within the range of values obtained
by other investigators for flow conditions (Table 5. 5).
In a  study conducted by Brookhart [5] on the  Miami River, floating
mechanical aerators were located at an oxygen  sag point occurring
in an impoundment.  Four 20-hp high-speed  surface aerators were
selected (Welles Products Company, Roscoe, Illinois),  and were
placed across the channel  so that each aerator would handle  equal
amounts of discharge.  The channel was approximately 300 ft wide
and the depth about 7  ft.  Water temperatures were approximately
25°C.   The average transfer rate,  computed for tests in which the
incoming water had a DO of approximately 3 ppm, was  1. 0 Ibs
O2/hp-hr (corrected only to standard temperature) in a discharge
range of 400 to 800 cfs  (velocity . 1/3 fps).
                             48

-------
                                                     TABLE 5. 5

                                      Average Oxygen Transfer Rate Obtained by
                                      Different Investigators for Surface Aerators
River
Delaware River
Passaic River
Laboratory Channels
Chicago Canal
Miami River
Jackson River
Flambeau River
Investigator
Whipple et al. [ 44]
Whipple et al. [6]
Susag et al. [50 ]
Kaplovsky et al. [49]
Brookhart [51 ]
Burns et al. [48 ]
Lueck et al. [53 ]
Standard Conditions
(Av. Ibs O2/hp-hr)
3. 16
2. 1
-4.0
1.5—4. 5
= 1.0*
2'2
.44 —.90
Velocity Range
(fps)
1 —1-1/2
] /5 —2 /5
1/10-2/3
1/4—1/2
1/3
3/8
•j~
Only corrected to standard temperature
##Not at standard conditions (T= 25°C, Incoming DO = 2 ppm)
*>.

-------
In a surface-aerator study similar to the one conducted by Brookhart,
McKeown installed three 50-hp surface aerators in Gulf Island Pond
on the Androscoggin River in Maine [52]  .  The pond resembled a
sluggish river, and the BOD^ was generally between 2 and 4 ppm. The
three 50-hp aerators  were used with incoming DO levels  in the water
of less than 3  ppm.  Transfer efficiencies of  1.7  Ibs C>2/hp-hr were
obtained.  It was speculated  that if the spacing between aerators had
been increased, somewhat higher efficiencies would  have resulted.

A comparison of spray aerators and turbine venting aeration was made
in studies conducted on the Flambeau River in Wisconsin [ 53 ] .  The
spray aerators were  furnished by Welles  Products Company and were
known as "Aqua-Lators . "  Four Aqua-Lators were installed in the
tailrace of the Pixley Dam powerhouse and used to create a  spray by
pumping river water  through a slotted disc. Each unit covered an
area approximately 35 ft in diameter and  sprayed water in the vicinity
of 8 to 10 ft high.  It  was  reported that droplets in the spray were
rather large.

The efficiencies of the spray aeration device are  relatively low and are
compared to results obtained from turbine  venting at the same site
(Pixley  Dam).

The turbines added 1 . 9 to  2. ] Ibs C>2/hp-hr with oxygen saturation
levels ranging from  3% to  12% and stream flow of 464 to  517 cfs,
while for similar conditions  the aerators  only added  from .44 to  .90
Ibs O2/hp-hr.

In the tests  only about 7. 5% of the total flow was pumped by  the aerators,
and, except for minor leakage,  all of the flow passed through the
turbines.

From the limited number of tests,  the turbines were 2-1/2 to 5  times
more efficient than the spray aerators.

A summary is presented in Table 5.4 of the transfer efficiencies for
surface aerators found by  several of the investigators mentioned in
this section.  Most of the tests were conducted with  updraft-type
aerators, thus the variations in results do  not appear to  be attributable
to the type of  aerator used.  All of the transfer efficiences have been
corrected to standard conditions,  except for the results reported by
Brookhart [ 5] ] ,  which have only been corrected to standard temperature,
and those of Lueck [53 ] , which are reported for a  temperature of 25°C
and an incoming DO concentration of 2 ppm.


D iff user Systems

In diffuser systems,  air or molecular oxygen is piped to a distribution
system  where  it is introduced directly into the water through porous
ceramic heads (usually silicon dioxide or alumina),  finely perforated
                             50

-------
tubing,  networks of nozzles, orifices, or jets.   The diffuser heads
can be installed at various depths below the water surface.  Increasing
the height of water above the system provides greater contact time
between the gas bubbles  and the water, thus increasing  oxygen
absorption.  However, by increasing the height of water,  one also
increases the  hydrostatic head against which the gas must be pumped.

If an air diffuser system  is being used, higher operating pressures
are then required, resulting in an overall  decrease in transfer
efficiency measured in Ibs O2/hp-hr absorbed.   If, on the other hand,
the diffuser system utilizes gaseous  oxygen,  an  increase  in the hydro-
static head does not present a problem,  since the source would be under
high pressure.  Furthermore,  when using oxygen instead of air,
absorption rates will  increase by up to a factor of five because of the
increase in  partial pressure.

In general,  the porous-ceramic and perforated-plastic-tubing diffusers
produce smaller bubbles  than the other system types and  consequently
allow for a greater percentage of oxygen to be transferred to the water,
due to the increased area-to-volume ratio of smaller bubbles.  However,
associated with these systems are greater  head  losses, which for air
diffusers produce lower  system efficiencies as measured in Ibs  C>2/hp-hr
Furthermore,  the fine-pore diffusers are more  susceptible to clogging.

The total rate of oxygen  transfer is  associated with three different
components: bubble formation,  bubble ascent, and the breaking of the
bubble at the water surface. For small diffuser openings and low
air-flow rates,  oxygen transfer during bubble formation is appreciable.
Conversely, for large aperture  diffusers with high air-flow rates,
little  transfer occurs during bubble  formation, primarily because the
bubbles are formed while rising  in the water.  The aeration during
bubble bursting at the water surface is primarily due to turbulence.
The oxygen  transfer during ascent depends on such factors as size
of the bubble,  depth,  terminal velocity,  etc.


Diffuser System Tests Using Air

Diffuser tests using air have been conducted as early as 1943 in the
Flambeau River at pixley,  Wisconsin. Those tests  have  been de-
scribed by Tyler [54] and Wiley et al. [55]  .  Noticeable improvement
in DO concentration was obtained for several miles along the river by
introducing  air through carborundum plates and  porous  tubes located
under  ]2 ft of water in the headrace and tailrace waters of the Pixley
Dam.  Absorption efficiencies were on the order of 7%.   (The
efficiencies for diffuser  systems are often  reported  as the percentage
of oxygen absorbed by the water. ) Results of subsequent tests on the
Flambeau River, using drilled pipe  (1 /8-in. holes) as diffusers  under
4 to 5 ft of water,  were reported by Palladino [56].  Considerably lower
absorption efficiencies (1.7%) were  obtained for  incoming water having
DO levels from 2 to 3 ppm.  Operating efficiency was about .38  Ibs
O2/hp-hr.
                             51

-------
Low absorption efficiencies have also been reported by Bohake for fine
and coarse diffusers using air to aerate the Lippe River at Heil,
Germany[ 57]  .  Diffuscr depth was about 3 ft, and the incoming water
was  nearly depleted of oxygen.  The river depth was approximately
]2 ft at the installation.  An absorption efficiency of 1.7% was obtained
with the fine-bubble diffuser while the coarse-bubble diffuser ( = 1/4-in,
nozzles) yielded an absorption efficiency of only ) . 0%.  Subsequent
tests using diffusers located on the stream bottom increased the
absorption efficiency to  J. 5% and  the system efficiency to approximately
] . 65 Ibs Q£/hp-hr  [73].  The importance of depth was emphasized
in  each of the  above river applications;  the greater the depth of the
diffuser the greater the  absorption efficiency, i.e. , the percentage of
oxygen absorbed.

More recent studies using air diffusers  have been conducted in the
Passaic [&1 and Delaware [44] Rivers in New Jersey.  Similar diffuser
systems were used in each  case.   The system consisted of two 8-in.
underwater heads with a total of ]60 diffuser nozzles,  each having
twelve 5 /32-in. ports .  The  tests were conducted at various  depths in
the Delaware and for a  single depth in the Passaic River.  Table 5.6
summarizes the results.

                                TABLE 5.6

            Diffuser Results from the Passaic and Delaware Rivers
Mean
Depth
(ft)'
7.2
7.2
7.2
13.2
25.0
38.3
Location
Passaic River
Passaic River
Passaic River
Delaware River
Delaware River
Delaware River
Approximate
Mean Water
Velocity
(ft/sec)

1/4

1
1
1
Ail-
Flow
]&
J V . 6
9
12
12
12
Average
Absorption
of Op
%
2.7
2.0
4.2
5.0
6.3
7.0
Average
Efficiency
(Std. Cond. )
(Ib Oz /hp-hr)

1.20

1.36
.93
.68
In comparing the results,  it should be realized that the diffuser systems
are essentially the same,  the air flow rates being quite similar, and
the only significant difference other than diffuser depth is  the lower
water velocity of the Passaic River.  The conclusion can be reached
from the table that oxygen absorption increases substantially with depth.
However, the operating efficiency  in Ibs O2/hp-hr increases from the
7.2-ft depth to the 13.2-ft depth, after which it decreases.

The above comparisons for different depths can be made because the
diffuser systems were essentially the same.  However,  there are a
large number of factors affecting absorption of oxygen from a diffuser
                               52

-------
system,  and caution must be exercised when conclusions on diffuser
performance are made by comparing results of entirely different
systems.  For example, one might consider diffuser test results from
two different types of  systems in which the air flow rates,  water
velocities, and depths of submergence were the same.  The oxygen
absorption efficiency and system efficiency in terms of Ibs O2/hp-hr
might differ considerably,  since the mean air bubble sizes emerging
from the diffusers could be quite different.  The bubble size would,
among other factors,  depend on the size and  number of diffuser
openings, which would not  be the same  unless similar equipment was
used.

Additional useful results could have  been obtained in the Delaware
River study if the air-flow  rate had been varied at the different depths.
This would have provided information for optimizing system efficiency
in terms of Ibs CU/hp-hr and air flow versus the number of nozzles.

The efficiency of diffuser systems should not be a strong function of
stream velocity unless large eddies  entrain the air  bubbles so as to
increase the air-water contact time.  If air bubbles require longer
time to surface because their mean ascent velocity  is lowered by eddies,
then higher absorption efficiencies would be expected.  However, if
flow conditions significantly affect rise time,  the stream would, in all
probability, have  adequate  natural aeration to correct problem conditions

For the types of diffuser systems used  in the above river studies,  the
maximum absorption efficiency was  7.0% (Table 5.7).  Test results
reported by Imhoff [58] , using finely perforated (.02  in-to .028 in. -
diameter holes) tubing in the Ruhr River (West Germany),  indicated
the following absorption efficiencies:


                  8% @  8-foot depth
                17% @  16-foot depth
                15% @ 20-foot depth


Even higher absorption efficiencies have been obtained from laboratory
tests using 1/2-in, polyethylene tubing having die-formed slits 1-1/2
in- on center.     Tests were conducted for low air flows and non-flow
conditions at 23°C.  Results have shown that, for 1-cfm air flow with
100 ft of tubing, absorption efficiencies  in oxygen-depleted water have
ranged from 14 to 44% at 3- and 10-ft depths, respectively [ 59] .

In view of the relatively high efficiencies obtained by Imhoff for stream
conditions and the even higher values obtained  with  the 3/2-in.
polyethylene tubing for non-flow situations, it is possible that a poly -
ethelene tubing system might be adaptable to stream applications.
Further  study would be required  to determine whether or not the system
would have similarly high absorption efficiencies for flow conditions
as it does for non-flow conditions .  Specifically, further investigation
                              53

-------
of its performance as  a function of air flow, water velocity,  and depth
is needed.   In addition, it would be of interest to use this type  of
system with molecular oxygen.  The following discussion considers
previous tests of diffuser systems with gaseous  oxygen.
Diffuser System Tests Using Molecular Oxygen

The results of a series of experiments using pure oxygen in a shallow
diffuser system have been reported by McKeown [60] . Fifteen ceramic
diffusers were evenly spaced along a 4-in*  diameter header,  100  ft
long, which was submerged 3 ft below  the stream surface.  Dissolved
oxygen  levels of incoming water varied from 0-1  pprn, and  stream
depth in the vicinity of the tests varied from 3 to  5 ft.  It was  found
that at a feed rate of 300  scfrn  absorption was only  1. 1%, while at
10  scfm absorption efficiency was increased to 12%.

In a somewhat limited study conducted by Amberg et al.  [6l]  on the
Pearl River  near Bogalusa,  Louisiana, molecular oxygen was diffused
through a multiport feeder placed on the  bottom.  In this  test the
oxygen  was produced in a pilot plant  on the site.

The multiport diffuser consisted of two 4-in. pipes,  36 in. on center,
and  equipped with Walker Process Company's "Sparjets."  The
Sparjets on one pipe had  1 /32- in.  holes and those on the  other had
3/64-in. holes.

During  the tests, water flow rate  over the diffuser section  was
approximately 2440 cfs,  with an average depth of 19 ft, and a tempera-
ture of  22  C. The incoming water was moving at an average of 2.  14
ft/sec and  had a DO of 7.7 ppm.   A summation of five trials is presented
in Table 5.7. Oxygen efficiencies varied from 14. 6% to 2 1. 5%.  It was
found that  a single  header with 1/32-in. orifices was  as  effective as
the multiple  header with  two different-size orifices.  The observation
was made  that,  even with the small orifice size and low  aeration rates,
a great  number  of oxygen bubbles were breaking at the water surface,
indicating  waste of oxygen.

The efficiency might have been increased by using carborundum  or
Saran-wrapped diffusers, commonly used in sewage treatment.
However,  such systems may become clogged when immersed  in  a
river and not used  on a continuous basis.
The Effect of Flow on Dispersion of Oxygen from Diffusers

A study of the effects  of flow on dispersion has been conducted by
Whipple et al.[ 44 ] ,  in which Rhodamine-B dye was injected through
a diffuser system into the Delaware River.  By measuring the  dispersion
of the dye as it travelled downstream  and using curve-fitting techniques,
the longitudinal dispersion coefficient was obtained.  According to
                            54

-------
                              TABJLE 5.7

            Reaeration Data for the Pearl River,  Louisiana
             Using a Double Aeration Header with 1/32-in.
                  and  3/6]-in.  Orifices (Amberg[6]] )
Oxygen
scfm
100
200
300
400
feed rate
lo/day
11, 900
23, 800
35,800
47,700
Oxygen
increas e
in stream
(Ib/day)
2 550
4250
5230
7850
Oxyg en
absorption
efficiency
(%)
21.4
17.8
14.6
16.5
Single Header (1/32-in. Orifices)
100
11, 900
2620
22.0
one-dimensional-dispersion theory, the coefficient was found to be
8.5 x 10  ft^ / min or 1.41 x 10^ ft2/sec.  Data on the surface aerator
was found to be incomplete, due to equipment failures.  This value
is an  order of magnitude larger than the largest "natural" coefficient
for a  similar river.  It was felt that the large  value was primarily
due to turbulent mixing induced by the aeration system.  Furthermore,
although this value  is characteristic of a given installation,  the
authors believed that it constitutes a decent first-order approximation
to other reaches in similar rivers being aerated.  It was also concluded
that the DO disperses in a manner similar to the  dispersion of the dye.
Such an assumption is reasonable, since changes in the stream charac-
teristics due to such a dye or oxygen are negligible.

A transverse-dispersion coefficient was not reported.  However,  the
transverse   spreading effect was presented in an equation relating
the area affected as a function of distance  downstream. The equation
was developed  from data on the dye cloud  geometry and is shown in
Figure 5.4.

It was noted  that, from similar tests using surface aerators, lateral
dispersion was observed to be approximately the  same as for the
diffuser  s ystem.
Downflow Contactors

The downflow contactor category includes the more widely known U-Tube
System iirst reported in the Netherlands by Bruijn and Tuinzaad[62]
                              55

-------
    11
-   10
o
o
o
-a
O
                        I
I
I
                      1000                    2000



                 Distance Downstream from Diffuser  (ft)
                     3000
              Figure 5. 4.  Graphical Results of Cross-Sectional

                              Area Distribution (Whipple [44])
                                56

-------
and more recently in this country by Speece[63] .   Aeration by systems
in this category is accomplished by temporarily pressurizing an air-
water mixture as it is forced downward by a slight  head over a vertical
tube.  More oxygen is transferred near the bottom  of the  tube by virtue
of increased pressure and lower temperature, which increases  the
DO deficit.  This relationship is illustrated in Figure 5.5.  In a U-Tube
the mixture is released from the tube near the surface, whereas in a
straight downflow contactor  it would be  released near the bottom.

Several different types of U-Tube systems can be designed[ 65] . In
Figure 5.6 a system  is illustrated where oxygen is  injected into the
inlet water by a blower.   Since  only low pressures  are required at the
point of injection, centrifugal blowers may be used, which reduce cost
and maintenance requirements.  Good control over  the air-water ratio,
which ranges  between 0 and  20% (by volume), is maintained by adjusting
the air-intake lines.

A second  type of U-Tube system is  shown in Figure 5.7.  in this system
the head is provided by a natural cascade.  The  system shown has been
used in a  fish hatchery where 400 gpm of water at 75% DO saturation
cascades  3 ft  into the inlet of a  40-ft-deep U-Tube and emerges at
120% DO saturation [63].

In still another type  of U-Tube, air is introduced by a venturi which is
vented to  the air as shown in Figure 5. 8.  The venturi system has a
somewhat larger head loss than the blower system  described above.
The  air-water ratio  can  be controlled by valving the air-intake port.
This system has the advantage of  not requiring  external power.

Where stratified impoundments are a problem,  U-Tubes can effectively
reaerate the water by selective withdrawal of the hypolimnion water.
Such a system has been proposed  for the Snake  River and  is  shown in
Figure 5.9.  This system is capable of  raising  the  DO from  40% to 95%
saturation.  The U-Tube would  be a 40-ft-deep  trench,  160 ft long  and
10 ft wide, on each side of the center baffle.

The  transfer of  oxygen to water is more effective the deeper the U-Tube;
however,  the  saturation  of dissolved nitrogen (DN)  also increases.  This
may be a  critical factor  since fish are adversely affected  when nitrogen
super-saturates.  For example, it is cited in reference 63 that a
tolerable  dissolved nitrogen  level for salmon is about 105% saturation
at the water surface.  If U-Tubes  are restricted to  a water depth of
10 ft, the maximum  possible super saturation  is 106%,  and acceptable
nitrogen levels would exist.   The  mixture of air and water could then
be passed through several ]0-ft stages  which would increase the DO
concentration  but not the DN.

Any  time  U-Tubes are considered for an aeration application, attention
must be given in the design to the avoidance of nitrogen supersaturation.
The  prime design consideration, of course, is the DO concentration
                              57

-------
Ul
oo
 DO

(mg/1)
                         Inlet
                                                                           DO Saturation Value
                                                                           DO Value in Tube
                                                                                   Outlet
                                  Figure 5. 5.  DO Deficit in 40-Ft-Deep U-Tube (Speece [65])

-------
   Air Blower
Figure  5. 6.    Air  Blower Injection Modification of
               U-Tube System      ( Speece  [65])
                    59

-------

Figure 5.7.  Cascade Air Injection (Speece [65])
                 60

-------
                       Air
Figure 5.8.  Venturi Air Injection (Speece [65])
               61

-------

Figure  5.9.  Schematic Installation of U-Tube Oxygenation of
             Stratified Impoundment Releases  (Speece [63])
                         62

-------
entering  and leaving the tube.  However,  in achieving a desired DO
level the effluent should be checked for possible supersaturation of
nitrogen.

In general, high air-water ratios and deeper depths are required to
obtain DO saturation, as is  shown in Table 5.8 [64].   According to
gas transfer theory, the associated air-water  ratios that will saturate
the  water with nitrogen are  those given in Table 5. 9.
                          TABLE 5.8

                   Air-Water Ratio Required
                   for Oxygen Saturation (%)
                          (Speece  [64]  )
Inlet DO U-Tube Depth (ft)
(% saturation) 20 30 40 50 60
0
20
40
60
80
_ _
—
—
22
14
23
20
18
14
9
18
16
14
11
8
15
13
12
9
7
13
11
10
8
5
                          TABLE 5.9

                   Air-Water Ratio Required
                   for Nitrogen Saturation (%)
                          (Speece [64] )
Inlet DO U-Tube Depth (ft)
(% saturation) iQ 30 40 50 60
0
20
40
60
80
_ _
—
—
—
16
— — .
23
21
16
10
21
18
16
13
9
17
15
14
10
8
15
13
11
9
6
                               63

-------
If the required air-waler ratio to saturate  the water with oxygen
exceeds the air-water ratio for nitrogen saturation, then nitrogen
s upersaturation will occur.

Another possibility for avoiding  nitrogen s upersaturation is to use
pure oxygen in place of air.  With such a system, initial DN would not
be altered, but the DO would increase.  Another advantage of an
oxygen injection system is that the change  in DO is approximately five
times that of air (Figure  5. 10).

Tests conducted bySpeece[65] using a variety of parameters for
4-in. diameter U-Tubes resulted in the following significant trends:

         1.   Increasing the air-water  ratio increases the change in
             DO at a diminishing rate.

         2.   Increasing the depth at which air is introduced results
             in a  reduction of  change in DO for  a given air-water
             ratio.

         3.   Increasing the depth at which air is introduced reduces
             the head  loss due to air injection in the  system for a
             given air-water ratio.

         4.   There is an equilibrium air-injection depth at which the
             air injection head loss is  zero.

         5.   Higher water velocities reduce the  head loss due to  air
             injeclion and decrease the change in DO through the
             U-Tube.   These changes  are  a consequence of the more
             nearly equal bubble residence times in both legs of the
             tube and  the reduced time for  oxygen transfer to occur.

         6.   The head loss due to air injection is proportional to
             the U-Tube depth.

         7.   Minimum diffuser submergence gave maximum  transfer
             economy.

The authors extrapolated their findings to a 60 in.-diameter U-Tube
and found that transfer economies in excess of 3 Ibs O^/hp-hr could
be obtained (Figure 5.11).  Furthermore,  it was determined  that
minimum air injection submergence resulted in  maximum transfer
economy.  Figure 5. 11 shows  that associated with each  U-Tube depth
and velocity there is an optimum outlet DO.  The optimum outlet  DO
was found to be dependent on the  velocity,  as indicated in Figure  5. 12.

Figure 5. 12 also  indicates that the lower velocity of 3.5 ft/sec  gives
rise to more efficient oxygen transfer.   However, if one is interested
in transferring a  given amount of oxygen into water,  It may be more
economical to use smaller U-Tubes with higher velocities, in which
lower operating efficiencies are offset by lower capital expenditure.
                              64

-------
GO
o
Q
    10 h
  Diffused Gas - Oxygen


O Diffused Gas - Air
                     6    8    10  12    14   16   18   20


                             % A/W
          Figure 5. 10. Increase  in DO vs Per Cent A/W

                       for Diffusion of Oxygen and Air
                          65

-------
            U-Tube Depth (ft)
                     40
                     30
                     20
                     10
      Nominal Velocity = 5.5 ft/sec
      Inlet DO = 0. 2 mg/1
 ro
O
en
X!
                          I
I
                         4         6
                      Outlet DO  (mg/1)
                  10
       Figure 5.11.  Effect of Depth on Oxygen Transfer
                     Economy in 60-in.  U-Tube
                      66

-------
From the work of Speece and Adams [65]  it has been concluded that
initial  bubble  size has no noticeable effect on the change in DO in a
U-Tube.  The diffusers  used in the tests consisted of a nylon cloth
and perforated units with 1/32-in. and ]/4-in. holes.

An exhaustive U-Tube testing  program has  recently been completed
by Rocketdyne,  in which a 2-in.-diameter U-Tube system was  used [66]
Variables investigated included depth, water velocity, air-water ratio,
and aspirator configuration:
             Depth
             Velocity
             Air-Water
             Aspirator
9 to 45 ft
1 .4 to 3. 4 ft/sec
0 to . 2 (volume ratio)
center-plug and venturi type
Quantitative analyses for rate of nitrogen transfer relative to that of
oxygen were conducted by a combination of vacuum degassing and mass
spectrometry.   The results are interesting in that the ratios of DN to
that of DO were found to have an average value of 2. 4 .  Table 5.10
contains the results from the samples  investigated.
                             TABLE 5.10

                   Results of Chemical Analysis for
                  Dissolved Nitrogen (Rocketdyne [66 ]

Superficial water velocity (ft/sec)
Air/water at 1 atm (68°F)
DO concentration (mg/1):
Entrance
Exit
DN concentration (mg / 1):
Entrance
Exit
DN change /DO change
Run Number
293 375
1.9
5.8
2.2
7 .4

3.8
16.8
2.5
1.4
7.0
1. 1
5.6

2.6
12.9
2.3
This result  is in contrast to the findings of Speece [ 65] which,  for the
conditions investigated,  indicated that the nitrogen gas transfer out of
the bubbles  was insignificant.
                              68

-------
The discrepancy between Rockefcdyne's and Speece's investigations lies
in the fact that the  test water used by the former was subjected to
vacuum degassing and, in so doing, nitrogen as well as oxygen was
stripped from the water,  which resulted in a large nitrogen deficit and
consequently a large nitrogen transfer.  In the case of Speece's work,
well water was used, which is characteristically high in DN.   He-nce,
only a small amount can be transferred at a relatively low rate.

Usually, in a stream or river the DN level is  expected to be near
saturation,  since there are virtually no nitrogen sinks and the water
surface quickly establishes equilibrium with the atmosphere.  Hence,
the nitrogen-oxygen transfer results in the Rocketdyne report can be
misleading when applied to a stream or river  situation.

These tests also indicated that pressure losses  in a venturi aspirator
were  considerably  lower than losses from center-plug aspirators
(Figure 5. 13),  and  also that the pinimum flow passage for the venturi
remains higher than for the center-plug aspirator, thus  reducing the
chance for plugging.

A modification of the U-Tube into a straight doxyaiflow contactor has
been investigated recently by McKeown in the  Androscoggin River in
Maine [ 52],   The modification consisted essentially of removing the
return leg of the U-Tube,  which then permitted  the oxygenated water
and remaining  air bubble to rise freely.

Extensive testing involved collecting DO and velocity profiles in the
vicinity of the contactor,  which defined the zone of influence.  The
site was a deep pond,  formed by a power dam, with a span of about
1300 ft.  Water depth was  over 60  ft,  and the general character
resembled a sluggish river.

The aeration system consisted of an axial flow pump with intake 4 ft
below the surface and whose  turbulent discharge,  containing large
quantities of entrained  air, flowed through a header box  into an 3 8-in.
corrugated pipe. Nominal liquid velocities in the pipe were  on the
order of 6 ft/sec, and depths up to 40 ft were  investigated.  Since
there was no direct method of calculating how much air became en-
trained,  the  discharge  from an auxiliary air blower was also fed  into
the head box.  Sparge rings or diffusers were not used to form small
bubbles, since sufficient turbulence existed in the header box, and
the associated  sheer force was adequate to produce  small  bubbles.

The velocity profile associated with a  10-ft downdraft bubble contactor
is shown in Figure  5.14.

Mixing was observed to occur some  15 ft below  the discharge, and
the representation  of the rising bubbles (Figure 5. 14) is based on
visual observation.  The vertical area zone of influence  was estimated
to be  approximately 800 ft2 (40 ft wide by 20 ft  deep).  The major
                              69

-------
 
-------
                 20'
                 10'
     10'
     15'
     20'
     25 !
     30'
     35!
     40' u-

                  0.08
    o;o?
                                            /\
                                0. 16
    0.^10  ^


       \    X   '  <'
    0.05 \   '   0. 15
            \
                  0. 12
                  V
    0. 05
                                0. 06
horizontal velocity component
     given in knots (kn)
    0. 06
0. 04
Figure 5. 14.   Zone of Influence of the Downdraft  Bubble
                 Contactor on the Surrounding Water
                          (McKeown [52])
                       71

-------
difference in the area of influence for different depth tubes was that
of depth.

The effects  are shown in Table 5.11.
                              TABLE 5.11.

               Vertical Area Affected by Bubble Contactor
Depth of Tube
(ft)
10
20
40
Vertical Area Affected
Width Depth
40 ft x 20 ft ( 800 ft2 )
45 ft x 35 ft (1575 ft2 )
50 ft x 60 ft (3000 ft2 )
In general,  McKeown's findings can be summarized as follows:

        1.   The blower produced no increase in DO or economy.

        2.   The average efficiency for all the tubes is above 1 . 4 Ib
            O? /hp-hr (converted to standard conditions).


Sidestream  Pres s urization

Sidestream  pressurization is a  technique for oxygenating river water,
in which a small percentage of the flow volume is drawn off, mixed
with oxygen under pressure, and the resulting supersaturated mixture
diffused back  into the river. There have been very few tests of this
technique, although the concept is promising,  particularly for large
rivers where  the oxygen requirement is great  enough to justify the
cost of constructing on site a gaseous-oxygen generating plant.  In
the preceding subsections,  cost has not been considered in the dis-
cussions, primarily because site conditions have a significant affect
011 cost.  For Sidestream mixing, however, the cost of oxygen is a
major factor and is discussed in'this  subsection.

In one of the Sidestream pressurization studies conducted by Amberg
et al. [61],  water entering the Pearl River was pumped through an
oxygen-diffusion system operating at  a pressure of 68 psig.  The
system was designed to pump 10, 000  gal/min under a 204-ft head
through a 150-ft-long,  14-in. -diameter pipe where oxygen was added
through spargers.  In this process the water became  supersaturated
with oxygen and was returned to the river through a diffuser header
placed across the bottom of  the river. The diffuser header was
equipped with twelve  2-in. nozzles and fifteen 1-1/2-in. nozzles,  and
was tapered from a 14-in.-to an 8-in.  diameter.
                              72

-------
 Initially,  considerable trouble was experienced from clogging of
 pumps and diffuser nozzles with debris carried in the river. However,
 these difficulties were overcome with appropriate modifications
 (screens  and increasing nozzle sizes to 1. 5 in. ).   Table 5. 12 illus-
 trates the effects of adding 30, 000 Ibs of oxygen.   It can be seen that
 about 2 ppm or  16,400 Ibs of oxygen was added to the Pearl River at
 the first station,  where mixing was considered to be complete.  The
 oxygen absorption efficiency for the system when aerating  25 cfs of
 the total stream flow (1. 64% of the total) was 54. 6%.

 The  oxygen was  released in very small, discrete  bubbles through the
 return header.  The small bubbles permit considerable  oxygen
 absorption in the water  as  they rise.  The average head of water over
 the diffuser was  9 ft.

 The  daily cost of adding  16, 400 Ibs of oxygen  (54. 6% of 30,  000 Ibs)
 based on $30 /ton  (delivered) was:

                     $450 for oxygen
                     $ 59 for power ($. 005 /kw-hr)
This resulted in.  adding  1.5  to 1.6 ppm oxygen over a 9 -mi.  stretch.

The above costs  do not include capital, maintenance, etc., and thus do
not provide a true  picture.  In addition, the system study was only
conducted for a limited  set of conditions which were probably not
optimum.

The above sides tream oxygenation system was designed by the Linde
Division of Union Carbide Corporation and is  referred to as  the
"LINDOX" System.

A smaller system  was  tested in Brewton, Alabama.  The system was
designed to inject 3000 Ibs of oxygen per day into the effluent of a paper
mill of the Container Corporation of America.  Union Carbide  claims
that injection efficiencies of from 55 to 75% were achieved [67]  .

According to Union Carbide  Corporation,  preliminary capital cost
estimates for sidestream aeration can be made on  the basis  of $4000
per daily ton of oxygen injected.  This price includes pump installation,
concrete pad, c ontrol panel, injection thimble,  dispersion header , etc.

Oxygen  costs will vary according to the installation and depend on such
factors  as quantity used, transportation costs , etc. Typical prices ,
according to  Union Carbide  Corporation (Spring 1971), vary  from
$35 to $50 per ton.  They also suggest that power costs for pumping
can be estimated to be  approximately 15% of the total oxygen cost.

In other investigations  conducted by Linde,  parameters have been
established for the following process variables:
                              73

-------
                TABLE 5. 12
 Dissolved Oxygen Added to the Pearl River
 at an Oxygen Addition Rate of 30, 000 Ib /day
by Sidestream Oxygenation (Amberg et al. [6 l]
(Water temperature was 25. 5 C)
Station
Lakeview (above
mill)
1 . 5 miles (below
mill)
3. 0 miles (below
mill)
6. 0 miles (below
mill)
9. 0 miles (below
mill)
Flow
(cfs)
1528
1528
1528
1528
1528
Before oxygenation
DO DO
(ppm) (Ib/day)
7.5
5.0
4.2
3. 1
2.7
61, 600
41, 100
34, 600
25, 500
22,200
After oxygenation
Flow DO DO
(cfs) (ppm) (Ib/day)
1524
1524
1524
1524
1524
7.5
7.0
6.0
5.0
4.3
61,600
57, 500
49,250
41, 100
35, 300
DO increase
(ppm) (Ib/day)
...
2.0
1.8
1.9
1.6
• • •
16, 400
14,650
15, 600
13, 100

-------
        1.   Equilibrium oxygen concentration in water versus C>2
            partial pressure.

        2.   DO at various pressures in the pumped stream versus
            efficiency.

        3.   Bypass line pressure and input oxygen concentration versus
            overall efficiency.

        4.   Velocity (Reynolds Number) in bypass  versus overall
            efficiency.

        5.   Contact time and oxygen input concentration versus DO
            in pumped stream.

        6.   Fraction bypassed and Input oxygen concentration versus
            overall efficiency.

        7.   Cost  of pumping.
Use of Pure Oxygen

Pure (molecular) oxygen can be used to replace air in several aeration
systems,  such as U-Tubes, diff user systems, venting of turbines, and
sidestream pressurization.  The form  of the  oxygen can either be
gaseous or liquid, although oxygen in the gaseous form is  more easily
injected.  The following discussion deals with various applications and
the relevant findings .

The only liquid oxygen (LOX) test  reviewed  in this study was conducted
by Midwest Research Institute  and reported in July, 1970  [68].  In that
study,  the investigators dealt with injecting LOX into both static and
flowing water which was  vacuum-stripped of oxygen.  The water was
then stored under a nitrogen blanket.

The rationale behind using LOX was the following:

       1.  LOX is  the most economical form for transportation of
           oxygen.

       2.  LOX is  more dense than water, hence it sinks and prevents
           loss es.

       3.  Evaporation of LOX in water could be at rates that would
           produce high-pressure bubbles.

       4.  The cooling is  localized, thus increasing the driving force.

       5.  LOX introduction imparts turbulence.
                             75

-------
The program encountered considerable difficulty, and consequently a
definitive assessment of the aeration process was not able to be given.
However, based on the work carried out,  certain trends were observed,
and a limited number  of conclusions were drawn by the investigators.
Among those conclusions were the following:


       1.   Based on mass  transfer, LOX is at  least as attractive
            as gaseous oxygen if the quantities are less than those
            necessary to require  construction of a production plant
            on s ite.

       2.   If the oxygen consumption is such that on-site production
            must be considered, then gaseous oxygen would be more
            economical.

       3.   When the flow was turbulent, absorption appeared to be
            more effective, and DO concentrations up to 30 pprn were
            obtained.

       4.   Increasing the contact time of the LOX with water increased
            the efficiency of absorption but not the mass-transfer
            coefficient.

       5.   Varying the water temperature from 7  to 30°C had no
            apparent effect  on the absorption efficiency.

       6.   The initial DO content of the water did not affect the
            mass-transfer  coefficient significantly.


These results would tend to indicate that the injection of LOX into water
is a difficult problem,  and  a better alternative would be to transport
oxygen as LOX but deliver  it to the water  in gaseous form.

Since the cost of pure oxygen is based on  rate of consumption and
distance from the source, the prediction was made that, when the
consumption rate exceeded  25 tons/day, a separation plant  should  be
erected at or near the site.  If the demand exceeds 25 tons/day, the
oxygen should be piped as gaseous oxygen, and if the rates  were smaller,
the oxygen should be transported  as LOX.  This  conclusion is based on
pricing schedules  from four major industrial gas suppliers  for  supplying
oxygen to ten different locations.   Figure  5. 15 illustrates the results,
indicating that the four companies quoted  similar prices.  The prices
include delivery within a 100-mi.  radius of  the separation plant.

The minimum price of pure oxygen is directly a  function of the  cost
of power.  To produce one  ton of  oxygen,  350 kwh are required.  It
was pointed out  that a lower bound would be about $7/ton of oxygen.
This is  based on an industrial rate of slightly under  $. 02/kwh.  Con-
sumption rates of 1000 tons /day would  be  required to drop the cost to
$10/ton.  Such a consumption rate is atypical, since this rate would
be indicative of  the domestic waste requirements of metropolitan
New York City.
                             76

-------
10,000,000
 1, 000,000
   100,000


    50, 000

1^
n!
-c

U]
,£>

—  10,000
c
o
•r-l
•«->
ex
 c
 o
o
 X
o
     1, 000
       100
        10

                                          On-Site Plant
                          LOX Transport
                        0
                 t      t       t      t     t Econom:

                     r                            ~
                                                Breakpoint
O  Producer No. 1

D  Producer No. 2

V  Producer No. 3

A  Producer No. 4

A  Minimum Cost Based on Thermodynamics 1  KWH =
                                   J	L
                                       J	L
                    2    3     4567     89    10   11


                      Cost of Pure Oxygen (cents per pound)




                  Figure 5. 15.   Cost of Pure Oxygen  (Both [60])
                             77

-------
An interesting and valuable theoretical study has  been conducted by
Speece [69]  to predict the transfer of oxygen out of a bubble of oxygen
and the transfer of nitrogen from water into an oxygen bubble.  The
results of this study are important, since a knowledge of conditions
which enable efficient oxygen absorption is necessary in order to
design an economically competitive oxygen injection system.  The
significant findings of his study can be  summarized as follows:


       1.   When small bubbles of approximately 2-mm diameter are
            released  from depths in excess  of 100 ft,  essentially all
            of the  oxygen is transferred.

       2.   The initial concentration of DO in the water has very
            little effect on the absorption of pure  oxygen.

       3.   The absorption of oxygen is greater for 2-mrn  bubbles
            than it is  for 4-mm  bubbles.  At 40 ft there is approxi-
            mately a  100% difference.

            Findings  1 and 3 above are illustrated in Figure 5. 16.

In another test concerned with the feasibility of using pure oxygen,
Amberg  et al. [70] added substantial quantities  of oxygen to water
passing through a  power turbine at Willamette Falls on the Willamette
River in Oregon.  The turbine was vented with  pure oxygen and also
with air.  In this test the goal was to achieve a 5  ppm concentration of
DO to meet  a state standard.  It was found that aeration with air did
not offer a practical solution.  Through the  use of a sparge ring  and
pure-oxygen absorption, efficiency on the order of 40% was obtainable.
The  relative cost of air and oxygen for this  test are shown in  Table 5. 13,
where  it can be seen  that at high DO levels of the intake water  the use
of pure oxygen is more economical.

In a  study conducted by Pfeffer  and McKinney [71]  using oxygen-enriched
air to aerate industrial wastes, it was  indicated that the rate  of oxygen
transfer is  considerably increased as the  oxygen  content of the gas
increases.  Upon examination of Figure 5. 17 it can be seen that  at a
given DO level the rate of oxygen transfer (slopes of the curves)
increases significantly with increase in oxygen content  of the  gas.

In oxygen-absorption tests conducted by Carver [72] it was found that,
when pure oxygen was used for  aeration, the rate of oxygen transfer
was  independent of the DO content in the liquid  between 0 to 12 mg/1.


Hybrid or Mixed Systems

A "mixed"  system would be two different types of systems in  combination.
Such a system might  be economically feasible when large changes in  DO
are needed,  when  considerable  shifting  of the DO sag point occurs with
                             78

-------
     100
G
01
o
o
in

-------
                                                   TABLE 5.13


                                Comparison of Reaeration Studies with Air and Oxygen
Location
Willamette Falls
Willamette Falls (a)
Willamette Falls (b)
Gas
Used
Air
Oxygen
Oxygen
DO to
Turbine
(ppm)
7 . 8-8.0
7.2
7.8
Oxygen
Absorption
Efficiency (%)
6.0
33.9
39.0
Oxygen
Exchange
(Ib/kw-hr)
0.53
13.9
9.5
Power Cost
per 1000 Ib
Oxygen
Dissolved
(dollars)
9.52
0. 36
0. 53
(a) Trial No. 1, Single -Opening Vent to Turbine
(b) Trial No. 2, Sparge Ring to Turbine
00
o

-------
o
Q
M
o
3
0"
T)
HI
X
                                       44. 3% 02
                         10        15        20

                         Aeration Time (min)
25
   Figure 5. 17.  Effect of Percentage Oxygen in Aerating Gas on the
                Rate of Oxygen Transfer (Pfeffer & McKinney  [71])

-------
different seasons, or if unusual conditions are associated with a
stream.  For example, if the water level in an impoundment where
a U-Tube is normally used drops  significantly, surface aerators
might be used to temporarily continue oxygen delivery until  water
depth again increases .

If a mixed  system is to be considered,  there are several natural com-
binations which can be ma'de.  Reasonable combinations  would be units
which work efficiently at  low DO levels followed by units  which operate
well in surface aerators or air diffusers acting as the primary system,
and the U-Tube,  sidestream pressurization,  or diffusers using pure
oxygen as a secondary fixed system.

When considering a mixed system, it should be kept in mind  that a
system which operates well in the 3 to 5  ppm DO range will  also work
well in the 1 to 5 ppm  DO range.  Hence,  although in a particular
application it may be found that a mixed system is more economical,
in general,  a non-hybrid  system offers fewer complications  and will,
inmost cases, be less expensive.
                              82

-------
                           SECTION VI

                  ENGINEERING METHODOLOGY
               FOR RIVER AND STREAM AERATION

In the previous sections the present state of the art in river and stream
aeration was reviewed.  The  oxygen balance in a river or stream has
been shown to be relatively complex.   In this section a methodology is
developed for the design of an aeration system.   It will be shown that
from an engineering viewpoint much of the rigor required in the under-
standing of aeration and of oxygen balance can be simplified considerably
in the system design.

An outline  of the steps required in the  design process  is shown in
Figure 6. 1.  In the following  paragraphs each one of the steps will be
discussed, with supporting calculations and examples  given where
necessary.  It should be noted that additional refinement  can be added
to the procedure,  but in this section it is  only intended to present the
framework of the methodology.


Problem Recognition

The first step is a very obvious one, but one which may be difficult,  in
that artificial aeration of rivers and streams is  not a generally accepted
practice.  This step requires the recognition that a problem in river
water quality  exists and that the problem  can be solved by artificial
aeration.

In some cases all  known sources of industrial and municipal waste may
have been treated  to the degree that 90 to 99% of the BOD is removed.
During some periods of the year,  however,  there may still be times
when fish die  off or there is a noticeable decrease  in some of the more
desirable forms of aquatic life.  This is most likely to occur in the
summer months, when flow volume decreases and  temperatures rise.
The condition may also develop during the winter,  when ice cover on
a river prevents natural reaeration through the  surface.  Extensive
eutrophication is an indication that oxygen-depletion conditions may
develop.  In these cases the assimilative  capacity of the river or stream
may be increased  by artificial aeration,  and recognition of this fact-
constitutes  the first step in the problem solution.


Preliminary Assessment

For a preliminary assessment of the problem,  a profile along the
critical  reach in the river should be obtained.  If profiles are available
from past measurements and no  significant changes in BOD loading or
flow volume have  occurred, these profiles can be used.  The profiles
should be for  the worst case,   i.e. , low flow and high  temperature.  If
                              83

-------
                    PROBLEM RECOGNITION

           Dissolved  oxygen levels in river or stream
           are below  acceptable levels.
                   PRELIMINARY ASSESSMENT

    Dissolved oxygen profile is obtained along critical reach.
    Profile should be obtained for worst conditions, i.e., low
    flow during summer months or possibly under ice in winter.

    Temperature,  flow velocity,  and depth  are measured along
    critical reach.  From these measurements flow rates are
    calculated,  and reaeration coefficients  are estimated from
    those empirical formulations most suited to the particular
    river or stream.
                                _L
                      OXYGEN REQUIREMENTS

    Select firstpoint where DO decreases below the accepted mini-
    mim level (5 ppm).

    Determine step increase  in DO necessary to bring DO above
    minimum requirements.
    Calculate daily loss of natural reaeration.

    Plot new DO profile from step increase data,  old profile, and
    calculated loss of natural reaeration.

    Examine new profile for points where DO may again decrease
    below minimum.
    Trade off costs of making larger  initial step increase of DO at
    one location versus  smaller increases at several locations.
                    T
             EQUIPMENT SELECTION

  Convert transfer efficiencies of several
  possible aeration systems from standard
  conditions to conditions present in the
  particular application at hand.

  Determine size and number  of units
  necessary to supply required amount
  of oxygen.
  Determine cost per year for each
  system.
  Compare costs.
                      SITE CONSIDERATIONS

                    Consider site factors which
                    may eliminate any particular
                    aeration systems from
                    further consideration.
        Figure 6. 1

Outline  of Engineering
     Methodology
     FINAL SYSTEM DESIGN
Compare cost and site factors  and select system.

Refine system design.
                                 84

-------
no profiles are available,  or if the ones that are available are
questionable,  a survey of DO should be made over the critical reach.
In either case, measurements should definitely be made of the flow
velocity, temperature, and depth.  Thes e measurements are required
for flow rate calculations and for estimates of reaeration coefficients
which are used in the determination, of oxygen requirements and in
compensation for loss in natural aeration.

The flow rate is  simply the velocity times  the cross-sectional area.
For the preliminary assessment the cross-sectional area can be
estimated from the width and average depth.  Whenever significant
variations occur in c ross-s ectional area, a new flow rate should be
calculated for that section.  The aeration coefficient can be measured
as described in Section IV, or it can be calculated using whichever of
the empirical methods  best fits the river or stream.  The methods of
Churchill [8] and O'Connor [9] appear to  have wide application.


Determination of Oxygen Requirements

Once it has  been determined that  the DO  levels  in a stream or river
must be increased by artificial aeration and a preliminary  assessment
of the problem has been made,  the next task is  to determine how much
oxygen must be added  and  where to add it.   To answer these questions it
is first assumed that DO profiles of the river in question have been
obtained  for the worst set  of conditions.  The profile used to  determine
the critical  reach of a river may not be a recently measured  one but
may be one  associated with a one-in-twenty or one-in-thirty year
low-flow situation.

In order  to illustrate the methodology, we  assume that the  DO profile,
upon which an artificial system will be designed,  is the one given in
Figure 6.2.  For the purpose of illustrating a case of a "polishing"
action, this profile shows  a low point of only  about 3 ppm.

Assuming it is required that the DO level should not drop below  5 ppm
in the stream,  it is then necessary to begin aeration artificially no
further downstream  than the one-mile station.  Assuming then that
artificial aeration would begin at  this station, it is necessary to
calculate fche amount of oxygen  that should  be added.  Since it has been
assumed that the minimum DO  level would be 5 ppm, it might seem
logical to raise the DO level at this point by 2 or 3 ppm.  However,
the higher the DO level is  raised,  the less  efficient the transfer of
oxygen becomes,  since the rate of transfer of oxygen is  proportional
to the oxygen deficit, i. e.  ,
                        =  KLa(Cs - C) =(KLa)D                  (6.]}
                               85

-------
        7


        6


        5
 DO
(ppm)
        3


        2


        1
        0     123456789   10    11    12

                               Distance (miles)




                    Figure  6. 2.  Example  DO  Profile
                                86

-------
where

      C     = the Oj concentration of the water (ppm)

      C     = the saturation level (ppm)

      K, a  = the overall transfer coefficient of an aerator (1/hr)

      t      = the time (hrs)

      D     = the oxygen deficit (ppm)
Equation 6. 1 shows  that the rate of change in oxygen concentration at
any time is proportional to the magnitude of the deficit at that time.
Hence,  at low initial DO levels the operating efficiencies of a system
are greater than for conditions where DO levels in the water are
initially high.  Figure 6. 3 illustrates the work expenditure for step
oxygen increases of  1 ppm,  in water where the saturation value is
9.2 ppm.  The work  required is proportional to
                             C
                               s
                          C   - C
                           s    m
where C   = mean DO concentration at aerator
        m
                         0<  C  <  C
                              m    s
This behavior is indicative of artificial aeration systems operating at
a nominal pressure of 1 atmosphere, viz,  surface aerators and shallow
diffuser systems.  When initial DO is on the order of 6 ppm or higher,
systems  which transfer oxygen from air into water become relatively
inefficient.   For instance, increasing the DO level from 8 to  9 ppm
requires approximately 13 times  as much work as raising  the DO from
0 to  1 ppm.

In order to illustrate the methodology for determining oxygen  require-
ments  and the spacing of aeration systems, the following case has been
selected:
                              87

-------
    10. 0
     9.2
     7.
     6.
DO  5.
     3.
     2.
     1.
                                      10
15
20
                         Relative Work Requirements
    Figure 6.3.  Relative  Work Requirements for a Unit Increase in
                 DO  in Water Initially Having  Different DO Levels
                              88

-------
                      DO Profile  - Figure  6.2

                      Flow Rate   - 1000 cfs

                      Velocity     - 0.5 mph

                               T  - 24°C

                             K   - 0.4/day
                               cL


The values listed above are assumed to be  constant over the 12-mile
reach.  This assumption is made for the purpose of illustrating the
methodology rather than for indicating  exact  stream conditions within
the section.

As discussed above,  unless pure oxygen is being dissolved directly,  the
work  required to transfer  oxygen from air  increases  significantly for
DO levels above 6 ppm.  Since the objective is  to prevent DO from
decreasing below 5 ppm, the first aeration devices must be placed in
the stream at mile 1.  For a first cut,  a step increase of 1 ppm is
made at this  point, raising the DO to 6 ppm.  The loss in natural
aeration due to the artificial addition must  then be calculated.

As a consequence  of increasing the DO level  of the stream from 5 ppm
to 6 ppm,  the natural aeration rate will decrease.  The decrease can
be computed as follows;
                     ~       =  K (C  - 5. 0)
                     dt ..          av  s      '
                        5 ppm
                               =  K (C  - 6. 0)
                      dt ,           av  s      '
                        6 ppm
Subtracting the above two equations,
               .  dC    dC         dC       _ . n T,.
               AdT  =  dT,     ~  dTV      ~ 1-u  a
                         6 ppm     5 ppm
Since the loss in natural aeration depends only on the increase of DO,
an upper bound on the loss of oxygen from natural aeration can be
established by assuming that the maximum loss  occurred over the  entire
length.   Hence,  the upper bound for the loss  in natural aeration at  the
12-mile station  is

      12 mi.x (I/. 5 mi. /hr) x (1/24 hr/day)  x 1. 0 ppm x .4/day =  . 4 ppm
                             89

-------
In Figure 6.4 the effect of the loss in natural aeration is  shown at
mile 12, where it is seen that instead of a ].0 ppm step increase
over the old DO profile there is only a 0.6 ppm  increase.  Now it is
also noted that the new DO profile crosses the 5 ppm minimum level
at mile 4.  At this point a second set of aerators must be added and
the loss in natural aeration again calculated.  At mile 7 there is another
crossover,  but at this point.it appears that a .5 ppm  increase will be
sufficient.  The new profile shows that this  is true.   Thus a total of
2.5 ppm of DO has baen added at  three separate locations.   The oxygen
requirement for this  case is;
         Ibs O2 =  ]000 ft3 /sec x 62.4 lbs/ft3 x 3600 sec/hr

                      x 24 hr/day x 2. 5 x 10"

                =  13, 350 Ibs  O2/day


By making a larger initial increase in DO at mile 1,  it may be possible
to eliminate the need for additional sites downstream.  It is  also known
that the work  required to raise DO from 5 to 7 ppm is greater than twice
that required  to raise it from 5 to 6.

To pursue this further,  two more  sets of calculations are made.  In
Figure 6. 5 the initial DO is raised from 5 to 7 ppm, and it is shown
that a second  set of aerators will be required at mile 5.7.  The  step
increase required at this point  is greater than .5, thus, for  convenience,
1 . 0 is selected as the value.

The total oxygen needed in this case will then be 16,020 Ibs/day, but
only two locations are required.  It is  not known at this point which
alternative is better, since the cost of constructing and maintaining three
sites must be weighed against  the  transfer efficiency of the aerators
employed in the two-site case  illustrated in Figure 6. 5.

There is still another alternative, which is  to raise the DO from 5 to
8 ppm at mile  ] , as shown in Figure 6. 6.  The only feasible way of
doing this would be  to use  pure oxygen, since conventional surface
aerators and diffusers become  quite inefficient in this DO range.  The
costs associated with this  alternative are discussed in Section V and
also later in this section.

The above example  illustrates the basic procedure for determining
oxygen requirements and the spacing of aeration systems.  There  are
several important factors  which will affect the calculations.   These
include:

      3 .     Value of aeration constant. A high value will lead to the
            requirement for a  high initial step increase or more
            aerator sites downstream.
                              90

-------
 DO   5
(ppm)
       4
                                                   Mean Values of Parameters

                                                     Flow Rate = 1000 cfs
                                                     Velocity   = . 5 mph
                                                     T         = 24°C
                                                     Ka        = 0. 4/day
                 Initial DO
        0     1    2    3    4    5    6   7    8   9   10    11    12    13

                                  Distance (miles)
   Figure 6.4.  DO Profile Before and After Oxygen Addition at Three Locations

-------
vO
        7


        6

 DO    5
(ppm)

        4

        3


        2


        1
                                                                    Mean Values  of Parameters

                                                                       Flow Rate = 1000 cfs
                                                                       Velocity   = . 5 mph
                                                                       T
                                                                       K
                                                               = 24°C
                                                               = 0. 4/day
                                    Initial DO
                                                     56789

                                                       Distance (miles)
                                                          10    11    12   13
                      Figure 6. 5.  DO Profile Before and After Oxygen Addition at Two Locations

-------
sO
                DO
               (ppm)
                                                                  A-lean Values of Parameters

                                                                     Flow Rate =  1000 cfs
                                                                     Velocity  =  .5  mph
                                                                     T
                                                                     K,
= 24°C
= 0. 4/day
                             Initial DO
                        0123456?     8   9    10   11   12    13

                                                Distance (miles)
                   Figure 6.6.  DO Profile Before and After Oxygen Addition at One Location

-------
      2.     Flow rate.  Large flows require a high oxygen input to
            increase the DO initially.  Loss  in natural aeration is
            less significant for large flow rates.

      3.     Pure oxygen versus  air.  If pure oxygen is available on
            the site,  it is possible to make large initial increases in
            DO without suffering a significant loss  in transfer efficiency.
As can be seen from the above  discussion, the transfer efficiencies of
the aeration devices play a significant role in system design.  In the
following step in the design methodology, the characteristics of specific
aerators affecting the equipment selection are discussed.   This  is
followed by a discussion of site factors,  which may eliminate certain
types of systems from consideration, depending  on specific conditions
at the site.
Selection of Aeration Units
As discussed previously in Section V, the currently available aeration
systems are surface aerators,  diffusers, downflow contactors (including
U-Tubes), and sidestream  mixing devices using molecular oxygen.
There are also several other types which may be  used with one of the
above  systems to form a "hybrid" system.  The hybrid approach has
not been explored in great detail  but offers some promise with additional
development.

The selection of a particular aeration system depends  on a number of
factors, including transfer efficiency,  cost,  maintainability, and site
suitability. Site factors are very important,  and  it is  quite likely that
even if a system is  suitable by  virtue of  its transfer efficiency and
cost,  it might be aesthetically displeasing within a particular area.


Engineering Considerations Affecting the Selection of an Artificial
Aeration System

According to the laws governing the transfer of oxygen to water,  it is
desirable to locate the aeration, system at the point  of maximum oxygen
deficit.  This is particularly true when oxygen is  being transferred from
air,  since the partial  pressure of oxygen is then much less than it would
be if pure  oxygen were being transferred.

Unfortunately,  it is  not possible to locate the aeration  system at this
point  if some minimum DO level  must be maintained.  If,  for instance,
the lowest DO concentration is  4  ppm and a level of 5 ppm must  be
maintained, the aeration system  must be located where DO first drops
below  5 ppm.  This  type of consideration tends to favor a system using
pure oxygen, since  the system  using air  cannot be located at its most
efficient operating point.
                              94

-------
In order to compare transfer efficiencies  of the various aeration systems
for  the application at hand,  they must first be converted from standard
conditions to field conditions.  For systems using air the field transfer
rate can be  calculated using Equation 6.2.
                        TR[PC        C  ]9
                 TRf=  -    S7^ - ^ -          (6.2)
                                   C  t
                                    st

where

      TR   =   field transfer rate (Ibs/hp-hr)

      TR   =   transfer rate at standard conditions (Ib/hp-hr)

      C     =   DO level  seen by aerator (ppm)

      C     =   saturation DO level at stream water temperature (ppm)
       s

      C     =   saturation concentration at standard conditions  = 9. 17 ppm

      P     =   pressure at which the system operations (mm of Hg)

               rate  of transfer of Oz in stream water
      00     ~   rate  of transfer of O^ in clean water

      a     _   saturation concentration of O;? in stream water
               saturation concentration of O? in clean water

      6     =   temperature-dependent  coefficient (average value is  1.024)
For a given river or stream cc > P >  6. and C  would be fixed, and
                                          S
                 TR =k. [k, ™  - C  ]                        (6.3)
                        1 L 2 760     mJ                        v    '
where k^ and k£ would be calculated  constants.  Thus, the transfer rate
for a system will depend on the DO in the water and on the operation
pressure.

The exact evaluation of C^ is  difficult to obtain,  consequently, the trans-
fer rate cannot be precisely determined.  Susag et al. [ 50] have proposed
three methods for estimating  Cm.  The first method assumes Cm = Cu,
the upstream DO value;  the second considers Cm =  1 /2 (Cu + C^), i. e. ,
the average of upstream and downstream DO levels;  and  the third uses
a logarithmic average based on the aeration equation.  Based on
                             95

-------
laboratory tests using surface aerators, the proposers of these methods
found the third technique to give the best results.  However, this area
warrants further investigation.  The application of these methods for
diffusers,  U-Tubes,  and sidestream pressurization  systems should be
verified.

For a surface aeration unit at or near sea level,  the pressure is
essentially 760 mm,  and the transfer rate will depend inversely on Cm
(the upper  limit of Cm  = (P Cs) = k£).  Hence, if it is desirable to raise
the DO level C_ of the stream  to a high value, the overall  efficiency
would drop off rapidly.  However,  if a  pressurized system or a pure-
oxygen system were  used, high transfer efficiencies could be achieved.
For example, if an air pressurized system were used at  70 psig,  transfer
efficiencies would increase significantly.  If such a  system used molecular
oxygen in lieu of air, the transfer  efficiency should  increase by a factor
of five compared to that of air, and perhaps on the order  of twenty com-
pared to the surface  aerator system.

Each of the aeration  systems  has particular advantages and disadvantages
relating to the river  and stream application.  A summary of these con-
siderations is presented in Table 6. 1


Economic  Considerations Associated with an Aeration System

One of the  most significant factors associated with an aeration system is
transfer efficiency (Ibs O2/hp-hr).  It has been shown in  this  section how
to calculate oxygen requirements for a particular  application.   From
these calculations it was considered that there are several possible
alternatives and that the final selection will depend on transfer  efficiencies
of the available devices and on site factors at the point where oxygen
should be added.

The cost of adding oxygen at one point,  such as the case shown in Figure
6.6, will depend on the transfer efficiency of a device attempting to
increase DO from 5 to  8 ppm.  It will be shown in a  design example that
this can be achieved  using pure oxygen but that the cost of using surface
aerators would be prohibitive at this location.

Other alternatives are  shown  in  Figures 6.4 and 6.5.  In  these cases,
the cost of constructing additional  systems at more  than one location must
be considered.  These  costs will depend on specific  conditions at the site,
such as:


         3. Type of soil for supporting structures

         2. Availability of power

         3. Accessibility for  maintenance
                              96

-------
                            TABLE  6. 1

             Characteristic Features of Aeration Systems
Type of
Aerator
                      Features
                                         Disadvantage s
Mechan-
   ical

Surface

Aerator
         Wide choice of commercially
         available units.

         High-speed, lightweight, electrical
         units can provide portability.

         Most units do not require direct
         vertical support,  i. e. , they float.
         However, provision must be made
         for sudden increases in stream
         depth due to sudden storms.

         Ideal operating DO level is under
         6 ppm.

         Range of instream transfer rates:
                1 — 4 Ibs  O2/hp-hr

         Range of cost: 10-hpunit,  $ 3,400
                        75-hp unit,  $18, 500

         Total cost increases as the number
         of sites increases, because  of
         electrical service and connections.
                                    Mooring cables may re-
                                    strict boating.

                                    Northern climates may
                                    cause freezing problems.

                                    Their presence may pre-
                                    sent aesthetic and/or
                                    noise problems.

                                    Being exposed, they are
                                    vulnerable to vandalism.
                                              Internal clogging from
                                              dust particles in the air.

                                              Except for plastic tubing,
                                              installations are gener-
                                              ally fixed.

                                              Cannot be used in a
                                              channel maintained by
                                              dredging.

                                              Long distance between
                                              compressor and diffuse?
                                              results in large head
                                              los ses.
Diffuser
Can be used with air or molecular
oxygen.

Does not provide any surface
obstructions.

Not vulnerable to vandalism.

Operates with a minimum of noise
and aesthetic upset.
Diffuser heads are often porous
ceramic,  however,  perforated
plastic tubing is also available.
Range  of absorption efficiencies
        (with  air) :
  >orous ceramic: 3 — 10%
  perforated tube : as high as 40%
        (when flow rate is low)

Range  of instream transfer rates
(porous ceramic  only):
     . 7 —  1. 4 Ibs O2/hp-hr

Ideal operating range is  usually
under 6 ppm when used with air.
Ideal operating depth: air, 10-15 ft
     pure  oxygen,  stream depth

                       97

-------
                                TABLE 6. 1

                                (continued)
 Type of
 Aerator
              Features
     Disadvantages
 Side-
 stream

 Mixer
Works efficiently with high initial
DO in water.

Does not interfere with boating.

Oxygen-absorption efficiencies are
normally over 50%.

Is not seriously affected by cold
weather  climates.

Requires a supply of pure oxygen.
Requires sophisticated
equipment and continual
maintenance.

Relatively high initial cost

Small installations are
expensive to operate be-
cause of high  cost of
oxygen.

Not portable.

Limited number of com-
mercial units available.
 Down-
  flow

  Con-
 tactor

(U-Tube)
If natural head is available and
sufficient stream depth, this sys-
tem offers maintenance-free
operation.

Can be located to the side of a main
channel; does not restrict boating.

Can be used with air or molecular
oxygen.

Ideally suited for high DO levels
(near saturation).
Not portable.

High initial cost compared
to other systems.

Requires depth  of at least
10 ft, 30 to 40 ft for high
efficiency.

Possible problems with
nitrogen saturation when
used at depths in excess
of 10 ft.

Usually requires a cus-
tom installation.
                                  98

-------
After the sizes and number of  units required to meet a given oxygen
requirement have been determined, the capital,  construction,  and
engineering costs can be computed.  This cost is converted to an annual
amortization rate to which are added operating and maintenance  costs.
Where there are  several possible types of systems, a cost analysis
should be performed  for each one.  The final selection may be based
on factors  other than cost,  but it is important to have the cost infor-
mation available.
Design and Cost Examples

The following two examples are offered as illustrations of the procedure
for estimating costs and also to show some of the differences in approach
using surface aerators and sidestream mixing with pure oxygen.

Figures 6.4 and 6. 6 are graphic illustrations of the two examples.  In
Figure 6.4 oxygen is injected at three different locations,  and  in
Figure 6. 6 the same DO profile is treated  at only one location.  It is
fairly obvious from  the previous discussions in this section that pure
oxygen can be injected efficiently  at a point where the DO level  in the
water is  fairly high.   The  transfer efficiency of a surface aerator,
however,  is best at  a  low DO concentration.

For the two cases, surface aerators will be used at each of the three
locations in Figure 6.4  while a sidestream mixing system will be used
at the single location shown in Figure 6.6.

The following design conditions apply to both examples :


         Velocity                    . 5 mph
         Discharge rate              1000 cfs
         Length of critical  reach      12 mi.
         Temperature                2 4°C
         Pressure                   1 atmosphere
         oc =  6                       .95
         Aeration constant, K        .4/day
                             3,

In the first example, surface aerators are used at  two locations to
raise DO from 5 to 6 ppm  and at a third to raise DO from 5 to  5. 5 ppm.
The oxygen requirements at each  location are
    = 1000 ft3 /sec x 624 lbs/ft3  x 3600 sec/hr x 24 hr/day
l-2              A
      x 1 . 0  x 10   = 5, 380 Ibs O2/day = 222 Ibs
  Ibs O2/day
  Ibs O2/day     = 1000 ft3 /sec x 624 lbs/ft3 x 3600 sec/hr x 24 hr/day

                    x 0.5 x 10"6 = 2,695 Ibs O2/day = 112 Ibs O2/hr
                              99

-------
To determine  how much horsepower is required at each location,  the
standard transfer efficiency for a surface aerator is converted to  a
field transfer  rate using Equation 6.Z.
                          - C
      TR   =	—     m
                          C  f
                           st
For this  example, surface aerators with a transfer efficiency of 2 . 2 Ibs
O2/hp-hr are selected.  There will be two field transfer rates,  one for
the case  where DO is increased from  5 to 6 ppm and the other when DO
is  raised from 5 to 5.5 ppm.
                 Tofoc   or    760   5+6i    i A ~O   -   in    nc
                 2.2  .95 x 8.5 x  =-7-77	T— I  x  10   x *.  10 x . 95
                     L            760     ^

         -j 2 =                      9. 17 x 10-6



             =  . 64 Ibs O2/hp-hr
      TR
                 oornc   oc    760    5 + 5 . 5 i   ,n-6   i  i o   nc
                 2. Z [ . 95 x 8.5 x  yr-pr  -  - s - j x 10   x 1. 10 x . 95
         3                          9. 17 x ]0
             = .71 Ibs  O2/hp-hr


The hp requirement at locations  1 and 2 will be
                 ,         222
                 KPl,2  =
At location 3
                 hp3    =777 =  158
Cost estimates for 75-hp electrical surface aerators have been reported
in a recent study[6] .   The estimates include all costs necessary to
install and operate the systems (cables, piling,  lights,  specially
reinforced frames, etc.) and are given for single-mounted units  and
                               100

-------
aerators mounted in clusters of two and three.  (Prices are based on
335 clays of annual use.)
      Arrangement                       Single        Double     Triple

      Total Horsepower                   75           150        2Z5

      Equipment and  Construction        57,000        92,700    130,500

      Engineering and Contingencies      11. 400        18,500     26, 100
                                        $68, 400    $]11,200   $156,600

      Operation and Maintenance          20,000       26,500     33,000

      Amortization & Interest             7. OOP       1 1, 500     ]6. 200
         (15 years @  6%)                 $27, 000     $38, 000    $49, 200
Assuming that the above costs are representative of the costs associated
with the given example,  the yearly expense of a system for the three
locations can be estimated as follows;


      At the two sites  requiring 347 hp, one triple cluster and one
      double cluster could be used at each location,  while at the
      third site one double cluster could be installed.  This results
      in a slight excess at the first and second locations and a slight
      deficit at the third location.  The total provision is on the
      conservative side.

      The total annual costs would then be

        2 x $49,200 + 3 x $38,000 = $212,400

In the second example, illustrated by Figure 6.6, the design conditions
given for the previous example also apply.   The oxygen in this case will
be injected by sidestream pressurixation at one location, raising DO
from 5 to  8 ppm .

Based on correspondence with the manufacturer [ 67] ,  it  is estimated
that installation costs  will be approximately $4000 per daily ton of
oxygen injected.   This includes pump installation, concrete pad for
oxygen supply system,  fabrication of oxygen control panel, injection
thimble, and dispersion header.

The operating costs  for supplying oxygen at the site would be  about
$35 to $50 per ton.  The power required to pump water at 100 psig is
about 50 kw per 1000 gpm.  (The  oxygen and water are mixed at
approximately  1 00 psig.) The injection efficiencies for the system
                              101

-------
should be on the order of 55-75% when 1 to 3% of the total flow is
pumped to the sidestream unit.

The oxygen  requirement to  raise DO by 3 ppm at the one location is;


      lbsO2/day=  1000  ft  3-/sec x 62. 4  Ibs /ft3  x 3600 s ec /hr x 24hr/day

                    x 3 x 10~6

                 - 16,020

Assuming an injection efficiency of 65%,  the injection system should
deliver
                    16, 020 Ibs         !•> 7 <.     ^ /j
                 —TT—-> AAA iu—77	 =  12.3 tons  O,/day
                 . 65 x 2000 Ibs /ton               2   *
The installation costs would then be

                   J2. 3 tons x $4000/ton  = $49,200
This cost can be amortized over 15 years at a 6% interest rate, yielding
an annual cost of $4,700/year.

The operating costs are calculated as follows:

      Assuming the cost of O;?  is $45/ton and that 3% of the river flow is
      pumped to the mixer:

            O? costs (135 days ) = 12. 3 tons x $45/ton x 135 day/yr
                               = $74,800/yr

The cost of pumping water to the mixer over a 135-day period is:

      Power required  =  1 000  ft3 /sec x 7. 48 gal/ft3 x 60 min/sec  x . 03
                         x 50  kw/1000 gal/rnin = 675 kw

      Power cost  = 675 kw x  24 hr/day x 135 day/yr x $.015 /kw-hr
                  = $32,800

Maintenance costs are estimated as follows:

      Personnel - 2 men @ $8, 000/yr =  $16, 000
      Equipment                           5, OOP
                                        $21,000
                             102

-------
The total annual cost is:          $   4,700
                                    74, 800'
                                    32,800
                                    21,OOP
                                 $ 133, 300
Summary

The cost of using surface aerators in this application is more than
1-1/2 times as high as that of using the sidestream system.  The basic
reason for this is  the  low transfer efficiency of the surface aerator
when working with an  incoming DO level of 5 ppm.   The oxygen-injection
system  is not affected by the relatively high DO and thus performs the
job at a much lower cost.

These two  examples serve only to illustrate the methodology.  For  each
case,  the cost calculations  should be made, but site  factors will also
influence the final result.  These factors are  considered next.
Site Factors

Depending on the uses of a  stream or river,  the application of certain
aeration systems may be precluded or severely limited.  For example,
if the waterway is navigable and is used for shipping and  pleasure
boating,  the placement of obstructions such as surface mechanical
aerators or U-Tubes in the river would present an obvious  problem.  In
this case,  the tendency would be to select a d iff user or sidestream
pressurization system.   This situation does not completely eliminate
the possible use of surface aerators or U-Tubes,  since they can be
installed outside of a main  channel;  although,  this would not permit
optimum utilization of the equipment.  On the other hand, if a river
channel is dredged periodically, the positioning of diffuser units or
outflow lines from a sidestream pressurization system would also be
constrained.

Aesthetic considerations and noise levels may also  carry considerable
weight when selecting a system.  If the installation  is to be near a
town or city, opposition may arise if the units produce considerable
foaming and frothing  or a continuous whine or  roar.

In some cases  there may be shifting of the low points in the DO profile,
and consideration must be given to a system which affords portability.
Such a situation would favor the selection of high-speed surface aerators,
which are  relatively small  and  light  compared to low-speed units.
U-Tubes,  sidestream pressurization,  and diffuser systems  are generally
not portable or easily moved.

If artificial aeration is necessary during winter months,  freezing may
be a problem with surface aerators.  When the surface of a river
                               103

-------
freezes, the aerator may be tilted or lifted and consequently may
become ineffective. In this case, a diffuser or sidestream pressuri-
zation system would be favored, since they would not be affected by
conditions on the surface.

Another factor that must be considered from a practical viewpoint
is the vulnerability of the system to vandalism.  For example,  with
a diffuser system,  accessibility to any of the components  can be
minimized;   the diffusers are under water,  the feed lines  and mani-
folds  are submerged or buried,  and the compressors can  be  located
in a blockhouse.  If sufficient natural head and water depth exist in a
stream, a U-Tube  system is also relatively impervious to vandals.
The system in this case would essentially be a submerged concrete
structure  whose cross-sectional shape corresponded to that of  a
U-Tube.
Final Selection of an Aeration System

As discussed  above,  site considerations play a major role in the final
selection of an aeration system.  In the design examples,  several
possible solutions were shown for a selected DO profile, each having
a different cost.   The final choice must involve a trade-off between
the costs of a system,  the cost of construction and maintenance, and
the aesthetic considerations at the possible sites.

For  each of the possible technical solutions, the corresponding  site
problems should be listed.  The site problems can be characterized as
those affecting cost and those affecting aesthetic  qualities.  From each
combination there will emerge a top candidate system.  If several
combinations  appear to be satisfactory,  the final  selection will
obviously be the  one  with the lowest estimated cost.  In the examples
discussed previously,  the use of pure oxygen in a sidestream mixing
system emerged as a top choice on a cost  basis and would probably
also be a better choice aesthetically.

However, further examination might have  indicated that diffuser
systems using plastic pipe were both less  expensive and more acceptable
aesthetically.  Unless  there are very obvious reasons not to do  so,
calculations and  a trade-off should be performed  for each possible
alternative.

In this report it has been, assumed that the minimum acceptable DO
level is  5 ppm.   The results  of the design  examples may differ con-
siderably if this  standard  is lowered, for example,  to 3 or 4 ppm.
The  methodology, however, would not be affected.
                             104

-------
                  SECTION VII

             ACKNOWLEDGEMENTS

The support and assistance of the Project Officer,
Dr. Curtis C. Harlin, Jr., of the Robert S. Kerr
Water Research Center, EPA, is acknowledged
with sincere  thanks.  In addition, the assistance of
Mr.  Lowell Leach, of the  Robert S.  Kerr Water
Research Center.and of Mr.  James  Basilico and
Mr.  Charles  Myers,  of the EPA, has been appreciated.
                     105

-------
                        SECTION VIII

                        REFERENCES
[1]  Tyler, R. G. ,  "Accelerated Reaeration, "  Sew. Works J. ,  _14,
    4, 834, July,  1942.

[2]  ZoBell, C. E. ,  "The Effect of Oxygen Tension on the Rate of
    Oxidation of Organic Matter in the Sea, " J.  Mar. Res. ,  3_,
    211-223,  1940.                                          ~

[3]  Dresnack, R. , and Metzger, I.,  "Oxygen  Response and Aeration
    in Streams, " Proc. 23rd Industrial Waste Conference, Purdue
    Univ., Lafayette,  Ind. , May,  1968.

[4]  Dresnack, R., Transient Dissolved Oxygen Profiles in Streams,
    Doctoral Thesis,  New York Univ. ,  19667

[5]  American Society of Civil Engineers, SED, Twenty-Ninth Progress
    Report of the  Committee on Sanitary  Engineering Research,  86,
    SA 4,  41, I960.

[6]  Whipple,  W. , et al. ,  Instream Aeration of Polluted Rivers,  Water
    Resources Research Institute,  Rutgers Univ., August, 1969.

[7]  Littleton Research and Engineering Co. , An Engineering Economic
    Study for  the Development of an Optimum Mechanical Aeration Sys-
    tem for Quiet Rivers  and Ponds,  Report #G-182,  July, 1970.

[8]  Churchill, M. A. ,  et  al. ,  "The Prediction of Stream Reaeration
    Rates, "  J. San. Eng. Div. ,  Proc. American Society of  Civil Eng. ,
    July, 1962.

[9]  O'Connor, D. J. , and Dobbins, W. E. ,  American Society of Civil
    Eng. ,  123,  641,  1958.

[10] Gameson,  A. L. H. , and Truesdale,  G.  A. , J. Institute of Water
    Eng. ,  _1_3,  175-87, 1959.

[11] Langbein,  W. B. ,  and Durrem,  W. H. ,  Geological Survey Circular
    542, U. S. Dept.  of the Interior, Washington, D. C.

[12] Ownes, M. , Edwards, R. W. , and Gibbs, J. W. , International J.
    of Air and Water  Pollution,  _8, 469, 1964.

[13] Odum, H. T. , "Primary Production in Flowing Waters,  "  Limnol.
    Oceanog. , !_, 102-117,  1956.

[14] Buswell,  A.M., VanMeter,  I. , and  Gerke,  J. R.,  "Study  of Nitri-
     fication Phase of the  BPD Test,  "   Sewage  and Industrial  Wastes,
     22,  4, 508,  February,  1950.
                                107

-------
[15] Gaffney, P. E. ,  and Heukelekian, H. ,  "Oxygen Demand Measure-
    ment Errors in Pure Organic Compounds -Nitrification Studies, "
    Sewage and Industrial Wastes,  30,  4,  503,  April,  1958.

[16] Sawyer, C. N. , and Bradney, L. ,  "Modernization of the BOD
    Test for Determining the Efficacy of Sewage Treatment Processes, "
    Sew. Works J. ,  1_8_, 6,  1113, November, 1946.

[17] Ruchhoft,  C. C. ,  Placak,  O. R., and Ettinger, M. B. ,  "Correction
    of BOD Velocity Constants for  Nitrification, " Sew. Works  J. ,  20,
    5, 832, September, 1948.

[18] Courchaine, R. J. , "Significance of Nitrification in  Stream
    Analysis-Effects on the Oxygen Balance, "  30th Annual Conference,
    Michigan Water Pollution  Control Assn. , May, 1962.

[19] American Public Health Assn. , Standard Methods for the Examina-
    tion of Water,  Sewage and  Industrial Wastes,  13th Edition, New
    York, 1971.

[20] Salle, A. J. ,  Fundamental Principles of Bacteriology, 3rd Edition,
    McGraw-Hill Book Co., New York, 1948.

[21] Velz, C.  J. , Applied Stream Sanitation, John Wiley  & Sons,  New
    York, 1970.

[22] O'Connor, D. J. ,  "The  Temporal and Spatial Distribution  of Dis-
    solved Oxygen in Streams, " Water Resources Research,  _3>  1 •  65,
    1st Quarter,  1967.

[23] Gameson, A. L. H. , "Some Aspects of the  Carbon,  Nitrogen, and
    Sulphur Cycles in the  Thames Estuary,  Part II, " in Effects of
    Pollution on Living Material, pp 47-54, Institute of  Biology, London,
    1959.

[24] O'Connor, D. J. ,  "Stream and Estuarine Analysis,"  Manhattan
    College Summer Institute  in Water Pollution Control, New York, N. Y.

[25] Buswell,  A. M. , Shiota, T. , Lawrence, N. , Van Meter, I. , Applied
    Microbiology,  2^,  21-5,  1954.

[26] Garrett,  M. T. , Proc.  Industrial Water and Waste Conference,
    Rice Univ.,  Houston,  Texas, 1961.

[27] Knowles,  G. ,  Downing, A. L. ,  Barrett, M. J. ,   J.  General  Micro-
    biology, 3J3, 263-78,  1965.

[28] Stratton,  F. E. , and McCarty, P. O. ,  "Prediction of Nitrification
    Effects on the Dissolved Oxygen Balance of Streams," Environmental
    Science and Technology, J_,  5, May, 1967.
                                108

-------
[29] Winberg, G. G. , and Sivko, T. N. ,  "The Significance of Photo-
    synthetic Aeration in the Oxygen Balance of Polluted Waters, "
    International J. of Air/Water  Pollution, 6, 267-75,  1962.

[30] Innjatovic,  L. R. ,  "Effect of Photosynthesis on Oxygen Saturation, "
    J. Water Pollution Control Fed. , May, 1968.

[31] Stay, F. S. , et al. , The Components of Oxygenation in Flowing
    Streams, U. S. Dept.  of the Interior, FWPCA, April,  1967.

[32] Hull, C. H.  J. ,  "Photosynthesis as  a Factor  in the Oxygen Balance
    of Reservoirs, " Symposium on Streamflow Regulation for Quality
    Control. , PHS Pub. No.  999-WP-30,  June,  1965.

[33] Hull, C. H. J. , "Discussion of 'Effects of Impoundments on Oxygen
    Resources' by Churchill, " Oxygen Relationships in Streams,  W58-2,
    pp 124-29,  1958.

[34] Copeland, B. J. ,  and  Duffer, W. R. ,  "Use of a  Clear Plastic Dome
    to Measure Gaseous Diffusion Rates in Natural Waters, "  Lira no 1.
    Oceanog. . _9_, 4, 494-499, October, 1964.

[35] Hornuff,  L. E. ,  A Survey of Four  Oklahoma Streams with Reference
    to Production,  Oklahoma Fishery Research  Laboratory Reference to
    Production  Report 62,  June,  1957.

[36] Kn'opp,  H. , "Investigation of the Oxygen Production  Potential  of
    River Plankton, "  Hydrology,  22, 152-66, I960.

[37] Paulson, R. W. ,  "The Longitudinal Diffusion Coefficient in the
    Delaware River Estuary as Determined from a Steady-State Model, "
    Water Resources Research, _5,  1,  59,  1969.

[38] Harleman,  D.  R. F. ,  and Ippen, A. T.,  "The Turbulent Diffusion
    and  Convection of Saline Water in an Idealized Estuary, International
    Assn. Sci.  Hydro. Publ. , _5_1,  I960.

[39] Orlob,  G. T.,  "Eddy Diffusion in Homogeneous Turbulence, Proc.
    American Society Civil Eng. ,  85, No.  HY 9,  September,  1959.

[40] O'Connor, D. J. ,  "Oxygen Balance of an Estuary, "  Proc. American
    Society Civil Eng. , _86, SA 3,  May,  I960.

[41] Fischer, H.,  "The Mechanics of Dispersion in Natural Streams, "
    j. Hydraulics  Div. , Proc. American Society Civil Eng.,  p  187,  1967.

[42] Fischer, H. , "Dispersion Predictions  in Natural Streams," J. San-
    itary Eng. Div. , Proc. American Society Civil Eng.  , p 927, 1968.

[43] Elder,  J. W. ,  "The Dispersion of Marked Fluid  in Turbulent Shear
    Flow,"  J.  Fluid Mechanics,  5,  544-60,  1959.
                              109

-------
[44] Whipple, W. ,  etal. ,  Oxygen Regeneration of Polluted Rivers:
    The Delaware River,  Environmental Protection Agency Program
    No.  16080 DUP,  December,  1970.

[45] Dobbins,  W. E. ,  "BOD and Oxygen Relationships in Streams, "
    J. Sanitary Eng.  Div. , Proc. American Society Civil Eng. ,
    pp 53-78,  June,  1964.

[46] Aeration in Waste-water Treatment,  Water Pollution Control
    Federation Manual of Practice No. 5, 1971.

[47] McKeown, J.  J. ,  and Buckley, D. B. , "Mixing Characteristics of
    Aerated Stabilization  Basins," TAPPI  8th Water and Air Conference,
    1971.

[48] Burns, O. B.  , etal. ,  "Pilot Mechanical Aeration Studies of the
    Jackson River in  Covington,  Virginia, "  Proc.  21st Industrial Waste
    Conference,  Purdue  "Univ. , Lafayette, Ind. , 799,  1966.

[49] Kaplovsky, A. J.  , etal. ,   'Artificial Aeration of Canals in Chicago, "
    36th Annual Meeting  of the Water Pollution Control Federation,
    Seattle,  Washington,  October, 1963.

[50] Susag,  R. H. , Polta,  R.C.,  and Schroepfer, G. J. ,  "Mechanical
    Surface Aeration  of Receiving Waters, "  J.  Water Pollution  Control
    Fed. ,  38, 1,  January, 1966.

[51] Brookhart, N. M. ,  Mechanical Aeration Project Performance and
    Feasibility Study, Miami Conservancy District Report, November,
    1969.

[52] National Council  of the Paper Industry for Air and Stream Improve-
    ment,   Results of a Cooperative Field Study of a Downflow Bubble
    Contactor and a Conventional Surface Aerator, Techn.  Bull. #237,
    June,  1970.

[53] JLueck, B. F. , etal., Evaluation of the Spray Type "Aqua-Lator"
    for River Aeration, State of Wisconsin Commission on Water Pol-
    lution, Bulletin No. WP-109, March, 1964.

[54] Tyler, R. G. , "Polluted Streams  Cleared up by Aeration, "  Civil
    Eng. ,  16, 348, August,  1946.

[55] Wiley, A. J. ,  et  al. , "River Reaeration, " Paper  Trade J.,  124,
     12,  123,  1947.

[56] Palladino, A. J. ,  "Investigation of Methods of Stream Improve-
    ment,  "  Industrial Water and Wastes, _6,  3, 87,  1961.

[57]  Bohnke,  B. ,  "Effect  of Organic Wastewater and Cooling Water on
     Self-Purification of Waters, " Proc. 22nd Industrial Waste Conference,
     Purdue Univ. , Lafayette,  Ind., 752,  1967.
                              110

-------
[58] Imhoff, K. R. , "Oxygen Management and Artificial Reaeration in
    theArea of Baldeney Lake and the  .Lower Ruhr River, " Das Gras
    und Wasserfoch,  109, Germany, 1968.

[59] Sellner,  E. P. ,  "Am-Aqua Aerated Lagoons, " presented at the
    38th Annual Conference of the Arizona Water and Pollution  Control
    Assn. , April, 1966.

[60] National Council  of the Paper Industry for Air and Stream Improve-
    ment,  Artificial Reaeration of Receiving Waters,  Tech.  Bull. #229,
    New York, May,  1969.

[61] Amberg,  H. R. ,  et al. , "Aeration of Streams with Air  and  Molecular
    Oxygen, "

[62] Bruijn, J. ,  and Tuinzaad,  H. ,  "The Relationship Between Depth of
    U-Tubes and the  Aeration Process, " American Water Works Assn.
    _,L, _50, 879-885, July,  1958.

[63] Speece, R. E. , "U-Tube Stream Reaeration, "  presented at the 7th
    Annual Sanitary and Water Resources Engineering Conference,  Van-
    derbilt Univ. , May,  1968.

[64] Speece, R. E. , "U-Tube Oxygenation for Economical Saturation  of
    Fish Hatchery Water, " presented at the American Fisheries Society
    Meetings, September 1968.

[65] Speece, R. E. , and  Adams, J. L. ,  "U-Tube Oxygenation Operation
    Characteristics,  " Proc. of the 23rd Industrial Waste Conference,
    Purdue Univ. , Lafayette,  Ind. , May, 1968.

[66] Rocketdyne (Division of North American Rockwell Corp. ), The U-Tube
    for Water Aeration,  Final Report, Project R - 8043,  Canoga Park,
    California, March,  1970.

[67] Union  Carbide Corporation, private  communication  dated March
    1971.

[68] Both,  T.  D. ,  et al. ,  Oxygenation of Aqueous  Bodies Using Liquid
    Oxygen - Loxination,  Midwest Research Institute, Kansas City,
    Missouri, March, 1970.

[69] Speece, R. E. , "The Use  of Pure Oxygen in River and Impoundment
    Aeration, "  Proc. of the 24th Industrial Waste Conference, Purdue
    Univ., Lafayette, Ind., May,  1969.

[70] Amberg,  H. R. , etal. ,  "Reaeration of Streams with Molecular
    Oxygen, " Industrial Water Engineering, pp 15-20,  February,  1967.
                             Ill

-------
[7l]Pfeffer, J.  T. ,  and McKinney, R. E. ,  "Oxygen-Enriched Air
    for Biological Waste Treatment,  "  Water and Sewage Works,
    p  381,  October, 1965.

[72] Carver, C. E. , "Absorption of Oxygen in Bubble Aeration, "
    Biological Treatment of Sewage and Industrial Wastes, Volume I:
    Aerobic Oxidation, Reinhold Publishing Corp., New York,  1956.

[73] Bohnke, B. , Possibilities of Artificial Aeration of Streams and
    Waters, Illustrated by the Lippe  as an Example,  Lippe Associa-
    tion,  Essen, Germany.
                               112

-------
                           SECTION IX

                          APPENDICES


           A.  Diffusers and Mechanical Aerators [46]


Diffusers

Diffusers are devices  which introduce air into liquids.  They are
installed in various  locations below the liquid surface on either fixed
or retractable mountings.   Air is furnished by  blowers, which are
usually in a central  location.   The blowers operate at a pressure
sufficient to overcome the static head of  liquid  above the diffusers and
the distribution losses.  In many large plants,  gas engines operated
from  the digester gas  are  used to drive the blowers.

Diffusers may be classified as porous or nonporous (Table A. 1).
Porous diffusers in the form of plates  or tubes are  either of the
ceramic type, constructed  of silicon dioxide or aluminum  oxide grains
held in a porous  mass with a ceramic binder, or of  the non-ceramic
type,  consisting  of plastic-wrapped tubes or plastic-cloth  tubes.

Nonporous diffusers may be of the nozzle,  orifice,  valve,  or shear
type.   Nozzle and orifice-type diffusers are  constructed of metal or
plastic, have  larger openings, and release larger bubbles  than  the
porous-type diffusers. Valve-type diffusers have a disc or valve
which closes when the air  supply is shut  off. They release larger
bubbles than do  porous diffusers.

Shear-type  diffusers provide for the  reduction of the bubble size by
the shearing force of the water entering the diffuser at  the  open top in a
counter-flow direction to the upflowing air.  These diffusers are  square
in shape.

Other diffuser arrangements include water jets, which consist of com-
bined air units,  and perforated or  slotted pipes, which  are used
occasionally as  a temporary expedient or for unusual conditions.  The
perforated or slotted-pipe diffusers  are not offered for deep-submergence
diffusion, show  a low  oxygen absorption efficiency,  and are readily
clogged or corroded.  Many of the diffuser applications in the past have
been in treatment tanks, and much of the following  discussion is in
relation to that type of usage.

Diffuser mountings may be either fixed or retractable.  Portable hoists
may be used for  raising the headers  out of the tank for  servicing.  Some-
times porous  plates are mounted in plate holders installed  on the floor
of the  tank.  The location of the diffuser  in an aerator tank has  a  con-
siderable impact on the efficiency  of the  device.  Data are  available from
studies on the oxygen  transfer of diffusers at various tank locations,
                              113

-------
                             TABLE A-l

                    List of Diffuser Manufacturers
    Manufacturer
 Diffuser  Type
          Description
Aer-O-Flo Div.
 Clow Corp.

Carborundum Co.
Chicago Pump
 FMC Corp.
Dorr-Oliver,  Inc.
Eimco Corp.
  An Envirotech Div.
Filtros Plant
  Ferro Corp.

FMC Corp. Link-Belt
  Division
Fuller Co. ,  Infilco
  Products

Hinde Engr.  Co.
Keene Corp. (formerly
  American Well Works
Norton Co.
Ray Products Co.
Rex Chainbelt Inc. ,
  Pacific Flush
  Tank Div.
Walker Process
  Equipment Div.
  Chicago Bridge  &c
  Iron Co.
Nonporous

Porous
Porous plastic
Porous plastic
flexible media

Nonporous

Porous

Nonporous

Nonporous

Nonporous


Nonporous

Porous



Nonporous



Porous plastic

Nonporous

 Nonporous
 Porous
 Porous
 Nonporous
 Nonporous
Stainles s-steel-nozzle type with
  check valve
Ceramic plates and tubes
Saran wrapped media, metal  core
  with integral end caps and
 control orifice
Saran cloth flexible media, metal
 frame with integral  end cap
 and orifice
Plastic nozzle type with integral
 control orifice
Ceramic tubes with cast-iron
  end caps  and control orifice
Plastic and metal valve type with
  integral control orifice
Nonporous metal shear type with
 control orifice
Tubular metal grid,  nozzle-type
  diffusion,  0. 5 to 1 m below
  water surface
Plastic base with elastomer cover

Ceramic plates and ceramic  tubes
  with and without integral end
  caps  and control orifice

Metal nozzle type,  adjustable,
  with 4 to 12 openings; ball
  check valve

Porous plastic media, plastic
  pan-type holder
Plastic aeration  tubing

Variable-flow, multiple-orifice type,
  made of cast bronze or aluminum
  magnesium
Ceramic plates and tubes
Plastic media, various mountings
Nonporous-metal-valve  type  with
  plastic ball  valve

Plastic-nozzle type
                                    114

-------
water depths, numbers of diffusers, tank width,  diffuser spacings, and
air rates.  Table A. 2 reports  the results of these studies.  Similar
studies  have also been made of nonporous diffusers.
Mechanical Aerators

Over the past decade mechanical aerators have been widely used to
supply oxygen to treatment plants treating a broad range of flows  and
organic loads,  providing high removal efficiencies at comparable
power cos ts .

Mechanical aerators  can transfer atmospheric oxygen to liquid  by surface
renewal and interchange;  and,  when properly designed, mechanical aer-
ators meet the mixing requirements at various single and multiple-unit
installations for a broad range of tank sizes and configurations.  Mechan-
ical aerators can also transfer atmospheric oxygen  by dispersing
compressed air fed below  the surface to a rotating agitator or turbine.
In the former case for an updraft-type aerator, oxygen is introduced
to the tank contents by lifting large volumes of liquid above the  "water
surface and exposing it in  thin films to the atmosphere.  With the
plate-type aerator a high degree of turbulence  is generated.   Both
types of aerators transfer oxygen through incorporation and dispersion
of air into the liquid because of high surface agitation.

In the latter case, air bubbles are discharged from  a pipe or  sparge
ring beneath the turbine and are broken up by the hydraulic shearing
action created  by the high-speed rotating blades of the turbine moving
through the liquid.  The sparge ring is fed by compressed air;  hence,
this system is  actually a combined mechanical diffused-air system.

In the downdraft system oxygen may be  supplied by air, self-induced
from the negative head produced by the  rotor.  With this system
external blowers or compressors are not required.

Brush-type mechanical aerators consist of a horizontal revolving shaft
with combs,  blades,  or  circular discs with T-shaped bars attached,
extending below the water  surface.

Table A. 3 lists some of the manufacturers of mechanical aerators.

The current trend is  toward the use of mechanical surface aerators in
larger basins.  Today there are many installations throughout the world
utilizing multiple units  in large tanks. Mechanical aerators have been
used in both small and large activated sludge plants and aerated lagoons
treating domestic and industrial wastes.

Efficiencies up to 7 Ib O2/hp-hr have been attained for  various  aerators.
Actually it has been found  that very high efficiencies can be achieved
on small pilot aerators but that for practical application,  under normal
conditions, the larger aerators have an efficiency range of 2  to 4 lb/O2
                            115

-------
                TABLE A, 2

      Diffuser Efficiencies at Various
Locations in Tank Under Standard Conditions
Location
Diffusers mounted on a
header on the wall side


Diffuser s mounted on both side s
of a header near tank well




Diff users mounted on both
sides of a header located
near both tank walls
Diffuser s mounted on both
sides of multiple
headers
Type
Porous



Porous





Porous


Porous

Depth
(ft)
12. 75



12.4





12.4


12.4

Air Rate
4 cfm/ft
8 cfm/dif.
12 cfm/ft
8 cfm/dif.
8 cfm/ft
8 cfm/dif.
16 cfm/ft
8 cfm/dif.
20 cfm/ft
10 cfm/dif.
16 cfm/ft
8 cfm/dif.

72 cfm/ft
8 cfm/dif.

Effi-
ciency
(%)
9.7


11.5

12

12.8

12.5

14

16

Tank
Width
(ft)
24




24



24

24

24

                      116

-------
                             TABLE A-3

              List of Mechanical Aerator Manufacturers
  Manufacturer
                           Type
Chicago Pump
  FMC Corp.

Dorr-Oliver Inc.
Eimco Corporation,
  An Envirotech
  Division
Fuller Co. , Infilco
  Products
Keene Corp. (formerly
 American Well
 Works)
Lakeside Engineering
 Corp.
Mixing Equip.  Co.
 Inc.

Permutit Co. ,  Div.
 Ritter Pfaudler
 Corp.
Vogt Manufacturing  Co.

Walker Process
 Equipment,
 Div. Chicago  Bridge
 & Iron Co.
Welles Products Corp.
Yeomans -Clow
                         Updraft
                         Combination
                         Updraft
                         Plate
                         Plate
                         Updraft
                         Combination
                         Downdraft
                         Updraft
                         Updraft
                         Updraft
    Aerator Characteristics


"Chicago"- propeller-driven flow
 discharged against diffuser cone
 at top
"D-O Aerator" -induced by sub-
 surface  rotor and compressed
 air
"Simcar "- induced by rotating
 impeller at top

"Vortair"-induced by horizontal,
 radially vaned impeller at top
"Aer o-Accelator "-induced by
 sub-surface rotor and com-
 pressed air
"American"-induced by horizontal
 vaned impeller at top
                                           Horizontal rotating
                                           "Lightnin1 "-induced by rotating
                                            impeller at top with rotor below
                                            surface
                                           "Permaerator "-induced by surface,
                                            sub-surface, and compressed
                                            air
                                           "Aer-O-Mix" - Produced by impeller
                                            in tube; with radia] inlet troughs.
Induced by rotating updraft impeller

"Aqua-Later" -either submersible
 or non-submersible pump dis-
 charge through vertical tube
"YeoCone"-induced by spiral
 vaned revolving cone at top and
 draft tube
Sigma-induced by scoop-type
 revolving blades at top
                                  117

-------
transferred under standard conditions per horsepower hour.  This
efficiency range normally covers the entire range of aerators
marketed today.
Plate Types

Typical plate types are the "Vortair," manufactured by Infilco/Fuller,
and the "American Aerator," manufactured by American Well Pump.
The "Vortair"  consists of a circular flat plate with vertical blades
attached at the periphery of the plate.  The "Vortair"  uses a standard
motor and gear-drive unit.  The plate rotates in a horizontal plane a
short distance  below the normal water surface.   When the aerator is
in operation, the top of the plate is clear of water.   The performance
of the aerator depends on  establishing proper design relationships to
determine the effects of diameter, blade sizes,  number of blades,
speed of rotation, submergence, and other variables that  affect horse-
power requirements and oxygen transfer.


Updraft Types

There are a number of updraft  types available,  including the Yeomans-
Clow  "Sigma," the Eimco "Simcar," the Mixco  "Lightnin," the Welles
Products  " Aqua-Lator, " the Walker "Intens -Aer, " and one manufactured
by Chicago Pump,  all of which  are described  below.

The Yeomans-Clow "Sigma" aerator operates on the updraft principle
with an impeller located at the  surface of the  liquid, designed to pump
large quantities of liquid at a low head.   Individual  blades, attached to
the rotating drive ring, are designed specifically to insure maximum
hydraulic efficiency.  The use of individually  mounted blades  allows
for great flexibility in changing the capacity of the unit on the job site
at any desired  future date. The unit is driven by a standard motor
and gear drive and is adaptable to  the use of a variable-speed drive.

The Eimco "Simcar" aerator consists of a standard motor and gear-drive
unit,  supported by walkway beams spanning the  aeration tank.  The
impeller is  a cone-shaped disc  with square-bar blades radiating outward
from  the center.  The blades are at or just below the  surface of  the
liquid.  Liquid is drawn upward and outward by  the rotating impeller
into a center cone and then is propelled outward in  a low trajectory.

The Mixco "Lightnin'' aerator is a pitched  blade, open-style turbine
utilizing a motor and gear-drive unit.  The aerator  is located at the
surface of the liquid and operates  on the  updraft principle. The  aerator
usually consists  of four blades  pitched at a 45-degree  angle.  Operating
speeds are generally between 30 and 60 rpm.  At times, a smaller rotor
is installed  on an extended shaft near the bottom of the tank to increase
turbulence and maintain solids  in suspension.
                              118

-------
The "Aqua-Lator" (Welles Products Corporation) is a relatively simple
aerator, consisting  of a special submersible or nonsubmersible pump,
a riser tube, a fiberglass-covered float,  and an orifice-diffuser
assembly.   Power is supplied from an on-shore control station by
means of a submersible power cable.  The "Aqua-Lator" is supported
in the liquid by its own integral float, which automatically adjusts  to
water-level changes.

The Walker "Intens-Aer" aerator consists of an updraft unit pumping
large volumes of liquid at low head by means of a slinging-blade
impeller located near the surface of the liquid.  Oxygen transfer ossurs
from  the large interfacial area created by the impeller discharge
pattern, from high turbulence and surface wave action,  and from en-
trained air bubbles  carried down with the circulating tank liquid by
the high tank-turnover rates  established.

A mechanical aerator manufactured by Chicago Pump,  FMC Corporation,
consists of an impeller on a vertical shaft located near the top of a
vertical draft tube,  which is  in the  center of the tank.  This impeller
lifts the liquid up the tube and discharges it against a diffuser cone,
which deflects it horizontally and down to the surface.
Downdraft Types

The Vogt Manufacturing Company makes the "Aer-O-Mix," in which
an impeller in a vertical tube forces liquid from the top down through
the tube to the bottom of the tank.  The liquid entrains the  air from
above.   The mixture expelled from the bottom of the tube rises through
the tank. This  system, in principle, is a  "downflow contactor"  as well
as a surface aerator.
Combination Types

The "D-O" turbine (Dorr-Oliver, Inc. ) is powered by a standard,
direct-connected motor and gear drive that is secured to a beam
structure spanning the tank.  Two or more  turbine impellers can be
used,  one usually located near the bottom of the tank above a sparge
ring, and another located about 30 in. below the surface.  The upper
impellers, or any intermediate impellers located below the surface,
serve  as shearing devices for the compressed air released through
the sparge ring.

The "D-O" aerator transfers oxygen through the emission of air at the
sparge ring and the shearing action  on the bubbles  by the rotating
impellers located above the sparge ring.

The Permutit " Permaerator"  uses two motor-driven impellers  and a
sparge ring at the lower impeller.   The Permutit aerator sparge ring
is designed to release air within the diameter of the adjacent impeller.
The upper impeller on the Permutit unit is  located near the surface and
functions as  a surface aerator.
                               119        olI.S. GOVERNMENT PRINTING OFFICE: 19" 484-482/201-3

-------
1

5

2

Sub/iM'1 Fii-lriSr. Group
O5G
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
JBF Scientific CorDoration
       2 Ray Avenue, Burlington, Massachusetts 01803
    Title
       ENGINEERING METHODOLOGY FOR RIVER AND STREAM REAERATION
1Q Authors)

       Murro, Ronald P.
       Yeaple, Donald S.
                                   16
                                      Project Designation
                                                 16080 FSN 10/71
                                  91  Note
22
    Citation
23
     Descriptors (Starred First)

        *Water quality control, *Aeration, Dissolved oxygen, Oxygenation,
         River Basins.
25
     Identifiers (Starred Firsl)
        *Induced aeration, ^Oxygenation efficiency, Air distribution,
         Molecular oxygen, Surface aerators,  Diffusers,  Downflow contactors,
        U-Tubes, Sidestream mixing,  Oxygen balance.
27
    Abstract
       Results of recent activities  in river and stream aeration by artificial techniques
       are reviewed,  and a  rational engineering methodology is developed for future
       river and stream aeration projects.

       The development of the methodology follows from a thorough review of the oxy-
       gen dynamics in rivers and  streams and the capabilities of aeration systems
       within the present  state of the art.  The report shows how the theoretical work
       can be simplified  considerably and applied to the solution  of river and stream
       water quality problems.  It is assumed that aeration would only be used as a
       "polishing" action after all identifiable waste sources have received at least
       secondary treatment.

       The results indicate  that,with careful consideration of site factors,artificial
       aeration can be applied successfully to raise dissolved  oxygen to  5 pprn,  using
       mechanical surface aerators, diffusers, downflow contactors, and sidestream
       mixing.  However, since the transfer of oxygen from air into water is relatively
       inefficient above 5 pprn DO,  the introduction of molecular oxygen through side-
       stream  mixing, U-Tubes, and pos sibly diffusers should be considered,  depending
       on the volume of water to be aerated. In cases where DO may be maintained at levels
       lower than 5 ppm, systems using air are competitive with molecular oxygen, depend-
                                                          ing on site conditions.
A bstrnctor                       Institution
Donald S. Yeaple, Sr StffEngJTRF firip.ntifir. Corporation
 WR:t02 
-------
ENVIRONMENTAL PROTECTION AGENCY
   Publications Distribution  Section
   Route  8, Box 116, Hwy. 70, West
   Raleigh, North Carolina  27607
       Official Business
    POSTAGE AND FEES PAID
ENVIRONMENTAL PROTECTION AGENCY

-------