EPA-670/2-75
  iM97§
Environmental Protection Technology Series
   BENCH-SCALE  HIGH-RATE  DISINFECTION
                 COMBINED  SEWER  OVERFLOWS
           With Chlorine  and Chlorine  Dioxide
                               National Environmental Research Center
                                 Office of Research and Development
                                U.S. Environmental Protection Agency
                                        Cincinnati, Ohio 45268

-------
                                      EPA-670/2-75-021
                                      April 1975
     BENCH-SCALE HIGH-RATE DISINFECTION OF

           COMBINED SEWER OVERFLOWS

      With Chlorine and Chlorine Dioxide
                      By
                Peter E. Moffa
              Edwin C. Tifft, Jr.
             Steven L. Richardson
        O'Brien & Gere Engineering, Inc.
           Syracuse, New York  13210

                      and

                James E. Smith
              Syracuse University
           Syracuse, New York  13210
       Project No. S-802400  (11020 HFR)
          Program Element No. 1BB034
                Project Officer

                 Richard Field
Storm and Combined Sewer Section  (Edison, N.J.)
 Advanced Waste Treatment Research Laboratory
    National Environmental Research Center
            Cincinnati, Ohio  45268
    NATIONAL ENVIRONMENTAL RESEARCH CENTER
      OFFICE OF RESEARCH AND DEVELOPMENT
     U.S. ENVIRONMENTAL PROTECTION AGENCY
            CINCINNATI, OHIO  45268

-------
                        REVIEW NOTICE

The National Environmental Research Center—Cincinnati has
reviewed this report and approved its publication.  Approval
does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
                             11

-------
                          FOREWORD

Man and his environment must be protected from' the adverse
effects of pesticides, radiation, noise and other forms of
pollution, and the unwise management of solid waste.
Efforts to protect the environment require a focus that
recognizes the'interplay between the components of our
physical environment—air, water 'and land.  The National
Environmental Research Centers provide this multidisciplinary
focus through programs engaged in:

     • studies on the effects of environmental
       contaminants on man and the biosphere, and

     • a search for ways to prevent contamination and
       to recycle valuable resources.

Combined sewer overflows are one of the most neglected
sources of microbial contamination of the nation's waters.
This report covers several methods of overflow treatment
that are applicable within the economic and geographic
constraints imposed by the occurrence of overflows within
urban areas.
                              Andrew W. Breidenbach, Kh.D.
                              Director
                              National Environmental
                              Research Center, Cincinnati
                             111

-------
                          ABSTRACT

A bench-scale study of high-rate disinfection of combined
sewer overflows with chlorine and chlorine dioxide was
performed to aid in the design and operation of full-scale
treatment facilities.  Four logarithm reductions in three
indicator bacteria and several common viruses were obtained
with 25 mg/1 chlorine or 12 mg/1 chlorine dioxide and two
minutes contact time.  Sequential addition of disinfectants
enhanced the process such that only eight mg/1 of chlorine
followed in 15 to 30 seconds by two mg/1 chlorine dioxide
was necessary to obtain similar reductions.

The removal of suspended solids by microscreening through a
23 micron aperture had little effect on disinfection effi-
ciency-  Disinfection increased slightly with increased
temperature, and a study of the receiving waters indicated
no bacterial or viral aftergrowth.  Adenosine triphosphate
(ATP) was found to be a possible alternative to the bacterial
indicators of disinfection efficiency and microbial contamin-
ation.  Electron spin resonance (esr) was used as a primary
standard method for quantitative measurement of chlorine
dioxide residuals.

This report was submitted in partial fulfillment of an Office
of Research and Development, U.S.  Environmental Protection
Agency demonstration grant EPA No. S-802400 (11020 HFR) by
the Division of Drainage and Sanitation, Department of Public
Works, County of Onondaga, under the partial sponsorship of
the Environmental Protection Agency.
                              IV

-------
                          CONTENTS

                                                       Page

Review Notice                                           ii

Foreword                                               iii

Abstract                                                iv

List of Tables                                          vi

List of Figures                                       viii

Acknowledgements                                        xi


Sections

     I    Conclusions                                    1

    II    Recommendations                                6

   III    Introduction                                   8

    IV    Experimental Procedures
            General Procedures                          10
            Methods for the Assessment of Disinfection  18
            Study of the Variables Influencing
            Disinfection                                34
            Impact Studies                              40

     V    Results and Discussion
            Single-stage Disinfection                   44
            Two-Stage Disinfection                      78
            ATP Correlations with Bacteria              91
            Impact Studies                              95
            Description and Operation of
              Demonstration Facilities                 106

    VI    Summary                                      125

   VII    References                                   131

  VIII    List of Publications                         139

    IX    Glossary                                     140

     X    Appendices                                   143

-------
                       LIST OF TABLES

No.
 1   Comparison of Simulated and Actual CSO             11

 2   Esr Signal Intensity from ClC>2 Standard Solutions  14

 3   Stability of ClC>2 Stock Solution                   16

 4   Effect of DDT on Total Coliforms Using MF          21

 5   Comparison of MF and MPN Results in SCSO           21

 6   Viruses Considered for Disinfection Study          25

 7   Recovery of Viruses from SCSO with Membrane
     Filters                                            27

 8   Filter Recovery Tests on Virus in Diluted Sewage   29

 9   Effect of Cl2 on ATP Standards                     32

10   Effect of C102 on ATP Standards                    32

11   Effect of Thio on ATP Standards                    33

12   Correlation of ATP and Bacteria in Untreated CSO   33

13   ATP and Bacterial Reductions Upon Disinfection     35

14   Effect of Screening on Bacterial Counts            38

15   Free C12 Concentration in SCSO                     52

16   Effect of Screening on Actual CSO                  54

17   Effect of Screening on Virus in SCSO               54

18   Survival of Viruses in SCSO Treated with C1O2      63

19   Single-stage Disinfection with C1O2 in SCSO
     Treated to Remove Demand                           69

20   Effect of Solids Removal on Disinfection with
     Cl2 and C102                                       71

21   Effect of Temperature on Disinfection of Polio-1
     with C102                                          74
                             VI

-------
List of Tables (Cont'd)

No.                                                    Page

22   Average Bacterial Reductions for Single-Stage
     Disinfection                                       86

23   Average Bacterial Reductions for Two-Stage
     Disinfection                                       86

24   ATP Correlation Coefficients                       92

25   ATP Stability Experiment                           94

26   Enzyme Stability Experiment                        95

27   Aftergrowth of Virus Disinfected with Cl2         103

28   Aftergrowth of Virus Disinfected with C1O2        103

29   Breakpoint Chlorination in SCSO                   104

30   Maltbie Street Operation Schedule                 118

31   Maltbie Street Analytical Schedule                120

32   West Newell Street Operation Schedule             122

33   West Newell Street Analytical Schedule            123
                             VII

-------
                       LIST OF FIGURES

No.                                                    Page

 1   Esr Spectrum of 8 mg/1 C1C>2 Solution at -196°C     15

 2   Esr Spectrum of 8 mg/1 ClC>2 Solution at 20°C       15

 3   Static Decomposition Investigations of C102
     in SCSO                                            17

 4   Dynamic Decomposition Investigations of ClC>2
     in SCSO                                            17

 5   Effect of Blending on Bacterial Counts             23

 6   Membrane Filtration of Early Storm CSO             28

 7   Membrane Filtration of Late Storm CSO              28

 8   Single-stage Disinfection Outline                  36

 9   Two-Stage Disinfection Outline                     36

10   Aftergrowth Studies For Maximum Flow Conditions    42

11   Single-stage Disinfection of TC with Cl2           46

12   Single-stage Disinfection of FS with Cl2           48

13   Effect of Sample-to-Sample Variations on
     Disinfection of 0X174 with Cl2                     50

14   Single-stage Disinfection of 0X174 with Cl2        50

15   Single-stage Disinfection of TC with C102          56

16   Single-stage Disinfection of FS with C102          58

17   Single-stage Disinfection of 0X174 with C1O2       61

18   Single-stage Disinfection of Polio-1 with C102     61

19   Single-stage Disinfection of Several Bacteria
     and Viruses with C102                              64

20   Single-stage Disinfection of f2 with C102          64

21   Single-stage Disinfection of Polio-1 with C1O2
     in a No-Demand System                              66

                           viii

-------
List of Figures  (Cont'd)

No.                                                    Page

22   Standard Dose Curve for Disinfection of Polio-1
     with C102                                          66

23   Effect of pH on Disinfection of 0X174 with C1C>2    67

24   Effect of C102 on Cell-Associated Polio-1          67

25   Effect of Temperature on Disinfection of Bacteria  72

26   Effect of Temperature on Disinfection of 0X174
     in SCSO                                            75

27   Effect of Temperature on Disinfection of f2 in
     a No-Demand System                                 75

28   Comparison of Disinfection of f2 with Cl2 and C1C>2 77

29   Two-Stage Disinfection of Bacteria and ATP with
     C102 (I)                                           80

30   Two-Stage Disinfection of Bacteria and ATP with
     C102 (II)                                          81

31   Addition of Multiple Doses of C102 to 0X174        82

32   Two-Stage Disinfection of Bacteria and ATP with
     Cl2 and C102 (I)                                    84

33   Two-Stage Disinfection of Bacteria and ATP with
     C12 and C102 (II)                                   85

34   Comparison of Single-stage and Two-Stage
     Disinfection of 0X174 with C12 and C102            87

35   Repeatability of Disinfection of 0X174 with
     Cl2 and C1C>2                                       88

36   Two-Stage Disinfection of 0X174 with Cl2 and C102  90

37   Aftergrowth of TC in Receiving Water               96

38   Aftergrowth of FS in Receiving Water               98

39   Key for Samples Collected in Viral Aftergrowth
     Study                                             102
                             IX

-------
List of Figures (Cont'd)




No.                                                    Page



40   Maltbie Street Site  Plan                          108




41   West Newell Street Site Plan                      109



42   Maltbie Street Process Orientation                110

-------
                      ACKNOWLEDGEMENTS

These studies were co-sponsored by the U.S. Environmental
Protection Agency and the County of Onondaga.

The supervision of this project was conducted by
Richard Field, Chief, Storm and Combined Sewer Section,
Advanced Waste Treatment Research Laboratory, U.S.
Environmental Protection Agency, Edison, New Jersey;
John M. Karanik, Project Officer, County of Onondaga and
Frank J. Drehwing, Vice-President, O'Brien & Gere
Engineers, Inc., Syracuse, New York.  The studies were
performed in the laboratories of O'Brien & Gere and the
assistance of all personnel who performed countless
analyses is acknowledged.  Particular thanks are expressed
to David R. Hill and Stuart J. Spiegel of O'Brien & Gere
for their help in the design of the experiments and
literature review, respectively.  Also thanks to
James McVea of the Biological Research Laboratories of
Syracuse University for his aid in the virus studies.

Many valuable suggestions were made by Cornelius B.
Murphy, Jr., Advisor in Chemistry, O'Brien & Gere, and
Cecil W. Chambers, Special Advisor in Bacteriology, U.S.
Environmental Protection Agency, Advanced Waste Treatment
Research Laboratory, Cincinnati, Ohio.
                            XI

-------
                          SECTION 1

                         CONCLUSIONS

Any widespread application of these results should be
contingent upon verification of these concepts in the
full-scale demonstration of prototype treatment facilities.
The degrees of mixing that may occur in full-scale
facilities could not be simulated on a bench-scale.  All
these results are based on complete mixing.

1.  Total coliform  (TC) bacteria in simulated combined sewer
    overflows (SCSO) were reduced to the target levels of
    1,000 colonies per 100 ml in two minutes by the
    following disinfectant dosages:

    a.  25 mg/1 chlorine (Cl2) (only 50% of all trials)
    b.  12 mg/1 chlorine dioxide (C102)

    These same conditions reduced fecal coliform  (FC) and
    fecal streptococci (FS) bacteria to 200 colonies per
    100 ml in two minutes.

2.  These same conditions also achieved five log reductions
    in poliovirus-1 and 0X174 bacterial virus.  Although
    target levels of viruses are not specified as part of
    federal or state water quality effluent criteria, five
    log reductions in virus populations would reduce the
    highest anticipated viral counts in actual overflows
    essentially to zero.

3.  High-rate treatment by microscreening, followed by
    disinfection, is a feasible method of reducing
    microbial contamination of combined sewer overflows  (CSO)
    to an acceptable level.

4.  The enhanced disinfection by using two-stage  (sequential)
    addition of Cl2 followed by C102 in 15 to 30 seconds may
    be due to the regeneration of CIO 2 through the interaction
    of chlorite ion (C102~) and
5.  There is no enhancement of disinfection beyond the
    expected additive effects when sequential addition of
    the same disinfectant is practiced.

6.  On a weight basis, CIO 2 is approximately twice as
    effective as Cl2 in reducing bacterial and viral
                              1

-------
     populations to target levels.

 7.   In the disinfection of contaminated waters with Cl2/ the
     initial rapid disinfection is  accomplished by free Cl2 /
     which is converted to the less potent combined Cl2
     species in one to two minutes.  C102 is converted to • the
     less potent C1C>2~ in the same  time period.

 8.   Microscreening had no measurable effect on high-rate
     disinfection with.Cl2 and'Only a slight • .positive -ef feet
     with C102« -~A possible explanation is that the increased
     rate of reaction between disinfectant and demands that
     resulted from the' shredding of particulates upon screen- '
     ing offset the increased numbers of exposed bacteria to
     yield no net increase in disinfection.

 9.   Microscreening alone ^decreased suspended solids, but, in
     some cases increased BOD5 and" bacteria counts.  These
     latter effects can be attributed to the reduction in
     protective effects and ( increased surface area - due * to • the
     shredding of the solids .

10.   An' advantage of microscreening is that by exposing previ-
     ously protected material, the  time necessary for natural
     assimilation of oxygen-demanding substances will be
     shortened.

11.   Cl2 and C102 demands can be attributed to different sub-
     stances in wastewaters.

12.   Within the dosages required for acceptable disinfection,
     Cl2 and C102 do not measurably change pH, TOC, BODs, COD,
     TKN or
13.  The temperature variations associated with the north-
     eastern climate had only a slight positive effect; on dis-
     infection of wastewaters with Cl2 and C1C>2..  This devia-
     tion from the large, positive, temperature effects observed
     in no-demand waters is most likely due to a wide variety
     of competing chemical reactions that occur- in wastewaters.

14.  The Cl2 dosages (^50 mg/1) - required to achieve breakpoint
     chlorin-ation in SCSO were far in excess of those dosages
     required for achieving target- levels of bacteria and
     virus, butt the cost is competitive with -other methods of.
     ammonia removal.

15.  Microorganism  aftergrowth was not observed to be a
     significant factor in this study. > However, the results
     may be more 'a- reflection < of the, difficulties in

-------
     simulating the conditions for aftergrowth than after-
     growth itself.

16.  Significant decreases in adenosine triphosphate (ATP)
     concentration that parallel bacterial reductions have
     been observed during the disinfection process.  The
     results of ATP measurements point to the potential of
     using this indicator parameter as an effective means
     of measuring bacterial concentration after disinfection
     or controlling disinfectant dosages.

17.  The effects of screening and temperature upon disinfec-
     tion were difficult to observe because they were of
     similar magnitude as the variations in duplicate trials.

The following conclusions are procedural in nature and
therefore not limited by the bench-scale scope of these
studies and need not be verified on a full-scale:

18.  Blending for several seconds prior to bacterial enumera-
     tion increases the counts in waters that have consider-
     able suspended matter.  Other forms of solids maceration
     are not acceptable.

19.  The levels of pesticides and heavy metals normally found
     in wastewaters do not interfere with the membrane filter
     (MF) technique for bacterial enumeration.

20.  The MF technique can be applied to disinfected CSO if
     periodic correlations are made with the most probable
     number (MPN) technique.

21.  The order of resistance of bacteria to disinfection with
     Cl2 and/or C1C>2 is FS>TC>FC.

22.  Poliovirus-1 is more resistant to disinfection than
     other pathogenic viruses studied.  Thus poliovirus-1
     would be a good indicator of viricidal efficiency.

23.  FS and poliovirus-1 show a similar resistance to disin-
     fection,  a fact that indicates that FS may be a better
     indicator of disinfection than the current choice, TC.
     It still remains to be established that the presence of
     FS in drinking waters is an indication of significant
     fecal contamination.

24.  The tendency for both bacteria and viruses to resist
     disinfection may be related to the lipids (fats) in the
     cell membrane or coating.

-------
25.   The particulate fraction of CSO is responsible for some
     of the disinfectant demand of free C12 and
26.  The normal variations in CSO from one storm to another,
     and the chemical and microbial changes of samples stored
     for more than a few days necessitated the use of an SCSO
     in order to evaluate the effects of the variables that
     influence disinfection within the defined project period.

27.  An SCSO that consists of a mixture of raw sewage
     collected at the same time each day and distilled water
     will minimize any effects due to variations in chemical
     composition.

28.  It is necessary to increase natural virus levels in CSO
     by seeding with known populations in order to observe
     the effects of disinfection of viruses.

29.  The lack of correlation between ATP and TC in different
     non-disinfected CSO is attributed to the wide variations
     in microbial composition.

30.  The technique of electron spin resonance  (esr) offers a
     primary standard method for the detection and
     quantification of C102.

31.  The DPD method is the preferred colorimetric technique
     for the measurement of free Cl2, mono-, di- and tri-
     chloramine, C102 and C1O2~.

32.  A fifty percent decrease in the C102 concentration in
     stock solutions occurs in two weeks if the solution is
     refrigerated and kept dark.  The same decrease is
     observed in twenty-four hours at 20°C under fluorescent
     light.

33.  Viruses may be recovered from CSO by absorption on a
     filter pad and subsequent elution.

34.  In the dosages required for high-rate disinfection, Cl2,
     C1O2 and thiosulfate did not significantly interfere
     with the bioluminescent determination of ATP.

35.  A standard solution of ATP showed only a slight decompo-
     sition after four days if properly buffered and kept in
     the dark.

-------
36.  The luciferin-luciferase enzyme reagent showed
     significant decomposition after twenty-four hours
     unless frozen.  It should be noted that this reagent
     consists of purified material.  If the reagent is
     prepared under less than pure conditions, this
     evaluation must be repeated.

37.  The development of an instrument to monitor ATP in
     unattended operation is a realistic venture primarily
     contingent upon the stability and cost of the enzyme
     reagent.

-------
                         SECTION II

                       RECOMMENDATIONS

It is recommended that:

 1.  The results of the operation of the full-scale demonstra-
     tion units be correlated with the bench-scale results to
     determine the validity of using these bench-scale
     studies for certain design parameters.

 2.  The role of suspended solids in disinfection of combined
     sewer overflows (CSO)  be evaluated on a full-scale to
     determine the effect of screening on the disinfection
     process.

 3.  The comparison of the effects of disinfection between
     the swirl concentrator and the various microscreens
     receive particular attention in the operation of the
     full-scale facilities in view of the conclusions about
     microscreening previously mentioned.

 4.  The full-scale facilities be operated in different seasons
     to evaluate the effects of temperature on disinfection.

 5.  Consideration be given to the development of a remote
     adenosine triphosphate (ATP) monitor to control the
     addition of disinfectants.

 6.  The effect of chlorite (C102~) on receiving waters be
     investigated before the widespread use of chlorine
     dioxide  (C102) is implemented.

 7.  C102 be considered as a disinfectant pursuant to the
     previous recommendation.

 8.  The effects of mixing on high-rate disinfection be
     thoroughly investigated on a full-scale study.

 9.  Two-stage disinfection with chlorine (Cl2)  and C102 be
     investigated to determine the mechanism through and
     conditions under which enhanced disinfection occurs.

10.  Future virus studies be conducted to assess their role
     in CSO and the validity of using bacterial species

-------
     and/or ATP as a measure of viral contamination.

11.  Microbial aftergrowth  studies be conducted in a natural
     environment  following  high-rate disinfection.  The pro-
     blems of simulating  such  a situation make the laboratory
     evaluation of aftergrowth extremely difficult.

12.  For bench-scale comparisons of the factors that affect
     the disinfection of  CSO,  a simulated CSO  (SCSO) be used.

13.  The procedural difficulties in running bench-scale
     studies such as these  be  recognized.  The greatest
     care must be taken to  preserve the intended experimental
     conditions and to maintain sample integrity.

14.  Electron spin resonance  (esr) be studied further in
     establishing a primary standard method for the determina-
     tion of C102 in wastewater.

15.  Additional studies be  performed to determine the validity
     of the membrane filter (MF) technique for enumerating
     bacteria in  CSO.

16.  Blending be  adapted  as a  preliminary step in the
     bacteriological examination of waters that contain
     significant  amounts  of particulate matter.  Because of
     the differences in individual blenders, a study of
     bacterial counts vs. blending time should be performed
     to determine the optimum  time for each model.

17.  Further work be done to determine an acceptable viral
     indicator organism.  This should include comparative
     studies with the traditional bacterial indicator
     organisms as well as pathogenic viral species.

18.  A standard procedure for  the recovery of viruses from
     sewage be established.

19.  The relationship between  ATP and total bacteria counts
     be studied.

20.  A standard method for  the measurement of ATP in sewage
     be adopted.

21.  Breakpoint chlorination be considered as a possible
     method of ammonia removal in CSO.

-------
                         SECTION III

                        INTRODUCTION

BACKGROUND

The County of Onondaga, New York, through its Department of
Public Works, Division of Drainage and Sanitation, in con-
junction with the U.S. Environmental Protection Agency, has
undertaken a number of studies to define the sources and
extent of pollution of Onondaga Lake.  Studies completed thus
far include the following:

    - County comprehensive sewerage study!

    - A survey of the major industries in the county^

    - A limnological study of Onondaga Lake to determine its
      present condition and project the effects of future
      programs on the lake3

    - A continuous lake monitoring program to measure the
      effects of changing conditions on the lake4

As a result of the above, the County has embarked upon a major
program to increase the hydraulic capacity of the existing
wastewater treatment facilities and upgrade the degree of
treatment from primary to tertiary.  It is projected that
these improvements will result in significant reductions of
pollutants discharged to the lake.

OBJECTIVE

Although these new facilities will result in significant
improvements to the lake, the inadequate capacity of the
existing sewer interceptor system results in combined sewer
overflows (CSO) to the tributary creeks of the lake.  The
microbial contamination of these overflows precludes the use
of the lake for any contact recreation.  Therefore, various
methods of treating these overflows were evaluated in the
county comprehensive sewerage study .  It was suggested that
the CSO be conveyed to a centralized treatment facility
adjacent to the existing Metropolitan Sewage Treatment plant.
The high construction costs of this solution led to the
investigation of point-source treatment of CSO.  Demonstration
grant EPA No. S-802400 (formerly 11020 HFR), was awarded to

-------
Onondaga County to investigate several methods of suspended
solids removal and high-rate point-source disinfection at two
outfall sites on the CSO system.  The solids removal step is
primarily for the purpose of optimizing disinfection.  The
term high-rate disinfection is used in this study to indicate
contact times of two minutes or less, as opposed to more con-
ventional contact times of 15 minutes or greater.

In order to establish a basis of design and operation for
full-scale demonstration facilities, bench-scale studies
were conducted to determine effective ranges of disinfectants
to be demonstrated.  The effects of temperature, screen mesh
size, disinfectant dose and type, sequential addition of one
or more disinfectants, and contact time on the disinfection
process were evaluated.  The limitations of relying exclusively
on indicator bacteria as a measure of microbial contamination
were recognized in this study.  Consequently, the degree of
disinfection was measured not only with indicator bacteria
but also with several viruses.  The possibilities of using
adenosine triphosphate  (ATP) as a measure of disinfection
and as an indicator of microbial population were also eval-
uated.  Efforts were made to determine the feasibility of
developing an ATP remote monitor for the ultimate purpose of
controlling the addition of disinfecting agents.

The effects of the discharge of treated CSO were assessed by
a study of aftergrowth in the receiving waters.  Breakpoint
chlorination was studied for the removal of ammonia  (NH3N),
an important nutrient, and residual chloramines, which may
be harmful to aquatic organisms.

-------
                         SECTION IV

                   EXPERIMENTAL PROCEDURES

GENERAL PROCEDURES

In order to assess the effects of the variables that
influence disinfection, a series of standardized procedures
for the characterization of samples, source and analyses of
disinfectants, and measurement of degree of disinfection
were developed.  Research of this type has been done in the
past, to a large extent in pure cultures of various bacteria
and viruses in buffered solutions.  However, this study
required that the investigations be conducted on combined
sewer overflows to aid more effectively in the design of
prototype treatment facilities.

Simulated Combined Sewer Overflows

The number of experiments necessary to determine the effects
of all the variables extended over a period of several months.
In order to ensure valid comparisons from one experiment to
the next, samples that were as similar as possible in
chemical and microbial characteristics were necessary.
However, there were several reasons why this could not be
done.  Storage of combined sewer overflows  (CSO) for more
than two days resulted in chemical and microbial
deterioration of  the sample.  The intermittent and
unpredictable occurrence of CSO precluded the collection of
a fresh sample each week.  Additionally, the storm-to-storm
variations in the chemical and bacterial composition of CSO
 (Appendix A) were judged to be too great to enable valid test
comparisons to be made.

Therefore, a simulated combined sewer overflow  (SCSO) sample
was made each week by mixing equal parts of the influent to
the Onondaga County Metropolitan Sewage Treatment Plant with
distilled water.  The results in Table 1 show a comparison
of the composition of an average of three SCSO with data
randomly selected from a six-month characterization of three
CSO sites.

The use of SCSO as a substitute for CSO eliminates all the
previously described problems and seems justifiable based on
the above data.   The composition of each SCSO was checked
and no significant variations were observed.
                              10

-------
Table 1. Comparison of Simulated and Actual CSO	

                            Maltbie    Rowland    Newell
Overflow         SCSO       Street     Avenue     Street
PH                 7.1           6.4        7.5       9.6
Five Day Bio-
chemical Oxygen
Demand  (6005)-
mg/1             130            82         25        24
Total Suspended
Solids  (TSS)-
mg/1             156           282        482        32
Volatile Sus-
pended  Solids
 (VSS)-mg/!         92           146        140        30
Total Coliforms
 (TC)-IOO ml   11,400,000   3,640,000  33,200,000  6,000,000
Ammonia Nitrogen
 (NH3N)-mg/l N      5.41          0.66       2.28      4.39
Total Inorganic
Phosphate  (T-IP)-
mg/1 P             1.60          0.83       1.28      2.40
Source and Analysis of Chlorine

The disinfectants evaluated in this  study were chlorine
and chlorine dioxide  (C102).  It was necessary to obtain the
materials in pure form and to determine their strength and
residual concentrations after disinfection using a standard
procedure.

A stock solution of Cl2 was obtained as a five percent solu-
tion of sodium hypochlorite, known commercially as Chlorox®.
The stock solution was found to lose less then ten percent
of its available Cl2  content over a  period of three months
if stored in the dark and refrigerated.  The strength of
different bottles varied less than five percent from 50,000
mg/1 as available Cl2•

The strength of the Cl2 stock solutions was determined by a
sodium thiosulfate  (thio) titration  of iodine liberated from
potassium iodide using a starch indicator to detect the end-
point^.  This technique does not differentiate between Cl2
and other materials that will oxidize iodide such as chlorite
ion (C102~) or chlorine dioxide (C102).  The DPD Ferrous
Colorimetric Method^  has not only the desired specificity
but can also detect free Cl2f mono-, di- and tri-chloramine.
                             11

-------
No other accepted colorimetric technique has the capability
of the DPD method.  The cost of the equipment necessary to
perform the amperometric technique5, which has the same
capabilities as the DPD method, did not seem warranted.
Replicate measurements of the Cl2 stock solutions by the DPD
method indicated that this technique lacked the precision of
the thiosulfate method.  Therefore, both tests were performed
on Cl2 stock solutions: the DPD test to determine the quality
and the thiosulfate test to measure the strength.  In all
cases, the solutions were found to contain at least 99
percent Cl2«

Cl2 is consumed in three ways by SCSO, namely, by reaction
with ammonia (NI^N) to form chloramines, by oxidation reac-
tions with organic materials to form organic chlorides, and
by decomposition to chloride (Cl~) as catalyzed by normal
light conditions.  These reactions generally cause the
complete disappearance of free Cl2 in less than five minutes
and combined Cl2 in less than three hours in SCSO.  Therefore,,
in the bench-scale studies which required some Cl2 determina-
tions at 30-second intervals, the time-consuming titrimetric
methods were not possible.  Under these conditions, the less
accurate orthotolidine method^ was used to obtain approximate
Cl2 concentrations.  When time permitted, the free and
combined Cl2 levels during and after disinfection were mea-
sured with the DPD technique.

Source and Analysis of Chlorine Dioxide

Methods of Generation—

C102 cannot be obtained as a prepared compound, but must be
generated as the demand arises.  Tests showed that when a
stock solution was kept refrigerated in a container wrapped
in metal foil, C1C>2 had a half-life of two weeks, undergoing
a disproportionation reaction to Cl~ and chlorate (C1C>3~) .
In the demonstration of full-scale facilities, C1C>2 will be
generated upon demand, with no storage time involved.  It was
desired to generate C102 for bench-scale study in the same
manner as in the full-scale demonstration, but the technology6
was not available at the time of this study.

Therefore, C102 was generated in a manner suitable for pro-
duction of small (one or two liter) quantities .  A one
percent solution of sodium chlorite (NaClC>2) was acidified
to liberate C102 gas that contained impurities such as Cl2-
The mixture was passed through a  scrubber solution of
saturated NaCl02 and the resulting purified C102 was bubbled
through distilled water to produce a stock solution of
                              12

-------
approximately 1000 mg/1.  The reaction was carried out with
considerable care in a hood because of the mildly explosive
properties of NaC102 and the toxic vapors of C1C>2.  The stock
solution was stored as above and discarded after two weeks.
Upon use, the scrubber changes from light green to dark
brown, at which point it loses its effectiveness, and must
also be discarded.

Methods of Analysis—

Colorimetric—Although there is no universally accepted
method for quantitative measurement of C1C>2, if a pure
solution of C1C>2 could be obtained, the thiosulfate titration
mentioned previously would yield satisfactory results.  An
extensive literature review' indicated that the DPD method
as modified by PalinS is specific for C1C>2 in the presence of
free and combined Cl2 and ClO^"-  The concentration of C102
in the solution that is actually titrated must be less than -^
two mg/1.  The strength of the ferrous ammonium sulfate
solution used for titration should be diluted by a factor of
ten so that a quantity of titrant sufficient to obtain good  ,
precision is consumed.

The combination of thiosulfate and modified DPD techniques
enabled standardization of the C102 stock solution.  The
solution was always tested immediately prior to use, and if
the strength dropped below 500 mg/1, or if impurities were
found, the solution was discarded.

Electron Spin Resonance—C1C>2 also dissipates rapidly in
SCSO but its reactions are not as easily defined as those of
Cl2-  The lack of a rapid, specific test for C1C>2 in sewage
has prevented any detailed studies of the behavior of C102
during the disinfection process.  The fact that C1C>2 has an
odd number of electrons (23), and therefore an unpaired
electron, led to the conclusion that it could be studied by
electron spin resonance (esr)9.  The theory and instrumenta-
tion of esr, which is a recently developed, highly sophisti-
cated analytical technique, are beyond the scope of this
report and are adequately covered in the literature10'11.
The technique is free of any interferences, gives a
continuous instantaneous value and is non-destructive toward
the sample.

A brief summary of the theory is as follows:  energy is
absorbed from an applied magnetic field at a frequency that
is unique to the compound being studied.  As the esr
spectrometer scans through the different frequencies, a
peak or series of peaks that correspond to the compound or
                              13

-------
compounds in the sample can be observed on a recorder.  The
frequency at which the peak is observed can be related to
the structure and environment of the molecule.  The height
or area of the peak is directly proportional to the concen-
tration of the material.  If standard solutions are used to
calibrate the instrument, quantitative determinations can be
made.

The esr spectra were obtained using a Varian E-9 instrument.
The general standards of operation include a microwave
frequency of 9292 GHz, a field of 3300G, a scanning range of
1000G and a modulation frequency of 100K. . All spectra are
presented in the customary derivative form.

Samples of C1O2 in water investigated at -196°C (the boiling
point of liquid nitrogen) revealed a single broad line (see
Figure 1) with full width between points of maximum slope
equal to 85G+5G with g equal to approximately 2.02.  The
more concentrated solutions showed increasing line broadening,
most likely due to uneven distribution of C102 molecules
within the polycrystalline ice sample.

The esr spectrum of C102 in water at room temperature
revealed a partially resolved quartet having a hyperfine
splitting of 17G+1G and a full width between points of
maximum slope of approximately 12G (Figure 2).  A moderately
strong signal was observed from a one mg/1 concentration
while a very intense signal was obtained from a 15 mg/1
concentration.

Having ascertained that C102 gives a measurable signal at
room temperature when present in mg/1 quantities, the esr
spectra of a series of standard solutions were determined.
This standardization procedure was required for each
individual instrument operation.  The results in Table 2
show that quantitative measurement of C1C>2 in this manner is
feasible.

Table 2 .  Esr Signal Intensity from CIO? Standard Solutions

                                         Esr Intensity
C102 mg/1
0.0
1.0
3.0
(Arbitrary Units) •
0
24
58

The stability of C102 was checked by placing aliquots of the
same solution into identical quartz esr tubes stored in the
                             14

-------
             MAGNETIC FIELD STRENGTH, gouss-
                    MAGNETIC FIELD STRENGTH, gouss-
Figure I. Esr spectrum of 8mg/l
         CI02  solution at-!96°C
Figure 2. Esr spectrum of 8mg/l
          CI02 solution at 20° C

-------
dark.  One was stored at -196°C and analyzed at room tempera-
ture, while the other was left at room temperature throughout
the experiment.  After the large initial drop that is most
likely due to traces of impurities remaining in the esr tubes
which are difficult to clean thoroughly, analysis at 1 hour,
5 hours, 1 day and 2 days indicated only a slight change in
signal intensity (see Table 3).

Table 3. Stability of C102 Stock Solution	

                      Esr Intensity
   Time (Hours)	(Arbitrary Units)	% Change
0
1
5
17
24
48
73
56
64
66
50
64
—
-23.3
-12.3
- 9.6
-31.6
-12.3
Background spectra of SCSO at -196°C and room temperature
revealed no significant interfering peaks.  Several small
absorptions were infrequently observed at low concentrations
(0.5 mg/1).  However, they can be attributed to divalent
manganese, an ion common to urban sewage having a large
industrial component.

The stability or reactivity of C1C>2 in SCSO was measured in
two experiments designated as the static and dynamic studies
as shown in Figures 3 and 4.  The static experiment involved
the addition of a quantity of C102 to SCSO sufficient to
create a concentration of 8 mg/1.  The strength of the C102
dosing solution was determined by the thiosulfate and DPD
methods.  Samples were pipetted from the disinfected SCSO
into closely matched quartz esr tubes at 0, 15, 30, 45, 60,
90, 120, 180 and 300 seconds and immediately immersed in
liquid nitrogen at -196°C.  These samples and a series of
freshly prepared C102 standards in the range 0-10 mg/1 were
analyzed by esr in the described manner.  The resulting
quantitative C102 values were plotted against time to give
the decay curve shown in Figure 3.  The results show that all
the C102 had reacted within five minutes, most likely yielding
C1O2~.  The results were further supported by the DPD method
which showed the sum of C102~ and C102 to be constant in each
sample.

The time required to collect and freeze the samples was
sufficient to give a degree of uncertainty to the results
from samples taken in the first minute.  The dynamic
                              16

-------
in
"c
3
 3
 k_
 o
 Lkl
 CC
 01
 UJ
               6O     120     150     240
                       TIME, seconds
                                             70  —
                                             60  —
                                             30
                                         Z   40
                                            30
                                             20
                                              10
                                                                    	 TRIAL I
                                                                    	TRIAL 2
                                                       \
                                                          \
                                                           \
                                        300
                                                       50
 IOO       150
TIME , seconds
                                                                                 20O
                                                Figure 4. Dynamic  decomposition
Figure 3. Static decomposition investigations            investigations  of ClOo in
          of CI02 in SCSO                              SCSO

-------
experiment was designed  to circumvent  this  difficulty  by
mixing the C1C>2 and SCSO in an esr  solution mixing  cell as
the C1C>2 peak was being  monitored.  The  spectrometer was
calibrated with C1C>2 standards in the  usual manner  and the
mixing cell  inserted in  the instrument. " The  flow rates and
strengths of the C1C>2  solution and  SCSO  were  adjusted  to
give  a concentration of  eight mg/1  C102  in  a  cell.  The
solutions were then allowed to flow for  several  seconds to
establish a  dynamic equilibrium while  the spectrometer was
tuned to the C102 absorption peak-.  The  effluent stopcock
on the cell  was abruptly closed to  initiate the  experiment.
It was estimated that  no more than  five  seconds  elapsed from
the initial  mixing of  the solutions to the  time  of  first
observation  of the esr signal.  The only concern is that
adequate mixing might  not occur, in spite of  the fact  that
the cell is  designed to  minimize these effects.

The results  of this experiment are*" shown in Figure  4 and
indicate that C102 has a half-life  of  60-90 seconds.   This
is in agreement with the decrease in bactericidal and
viricidal efficiency observed in other phases of this  study.

"This  application of esr  should be of great  value in
establishing a primary standard method for  the determination
of C102 in wastewater.   The sophistication  of the technique
and the expense of the equipment prevent the  widespread
application  of esr for the average  environmental chemist.
However, the method can  be used to  evaluate the  colorimetric
techniques and the reactions and kinetics of  C1C>2 in
wastewater.

METHODS FOR  ASSESSMENT OF DISINFECTION

The parameters used to evaluate the disinfection capabilities
of Cl2 and C102 in SCSO  were bacteria, viruses,  adenosine
triphosphate (ATP) and various chemical  parameters. The
traditional  indicators of fecal contamination have  been the
coliform bacteria, but there is some doubt  as to their
ability to serve as an indicator of viruses!2.   Therefore,
in this study, actual  viruses as well  as bacteria were
measured in  order to establish the  validity of using bacteria
in disinfection studies  of CSO.

The shortcoming of direct bacteria  or  virus enumerations  is
that  the test procedures require incubation periods of 24
hours or longer.  This delay can be a  serious problem  when
the question of human  safety is at  stake.  ATP measurements
were  made to determine the feasibility of using  this
parameter, the measurement of which takes only a few minutes,
                              18

-------
as a more useful indicator of microbial contamination.

Bacteriology

Indicator Bacteria—

Because the direct measurement of pathogenic bacteria and
viruses is beyond the scope of most laboratories, certain
bacteria that are known to originate from the same sources
are used to indicate the possible presence of pathogens.
The coliform group, consisting of both fecal and non-fecal
bacteria, was first used as the indicator organism partially
because of the relative ease with which the MPN test5 could
detect its presence.  Even today this group is used in many
water quality standards even though it lacks the specificity
of some more recently developed tests.  For these reasons,
the total coliform  (TC) group was used as the primary
indicator organism in this study.

Recent developments show that a part of the coliform group,
namely, fecal coliforms (FC), may be a better indicator of
fecal pollution than the total group13.  Therefore, this
test was included in the studies whenever possible.  However,
there have been reports that indicate that both fecal and
total coliforms are less resistant to disinfection than some
of the pathogenic viruses1^.  Fecal streptococci (FS) have
been advocated as indicators of fecal pollution15 and
possibly may be more resistant to disinfection than either
the coliforms16 or many viruses.  Therefore, FS determinations
were carried out whenever time permitted.

Method of Bacteriological Analysis—

The membrane filter (MF) technique5 offers a rapid means of
detecting all the above bacteria.  There have been some
questions raised about the validity of the MF technique on
chlorinated wastes and on samples containing large amounts
of particulate matter.  These issues are treated in the
following paragraphs.

Toxicity—Standard Methods, 13th ed., states very explicitly
on page 678 the limitations of the MF technique in the
sentence "Experience indicates that the membrane filter
technique is applicable to the examination of saline waters
but not chlorinated wastewaters".  Although recent studies
indicate that with proper precautions the above statement
may not be true1?'1®, the specific objections to the MF
technique, as first outlined several years ago,1^ are
considered in detail.   The primary objections to the use of
                              19

-------
the MF technique in the bacterial examination of chlorinated
wastewaters arise from the need to filter large amounts of
sample to collect sufficient colonies to obtain a confident
count.

The suspended solids (TSS) collected on the filter pad may be
sufficiently great to clog the pores thus preventing the'
required amount of sample from being filtered.  Also, the
layer of TSS on the filter pad may provide a barrier to the
nutrients adsorbed in the pad and thereby hinder bacterial
growth.  Finally, the TSS may absorb disinfectants, heavy
metals and other toxic materials so that when collected on
the filter pad, the TSS may have a bactericidal effect.

The question of toxicity was explored by examining the over-
flows for heavy metals and chlorinated hydrocarbons and
conducting some controlled tests.  Several samples were
examined by atomic absorption spectroscopy for chromium,
cadmium, lead, nickel, zinc and silver but only trace amounts
of less than 0.1 mg/1 were found.  Gas chromatography showed
that several chlorinated species similar to the common pesti-
cides were present in the overflows at levels which sometimes
reached 1 mg/1  (Appendix A).

The effect of chlorinated hydrocarbons on bacteria counts
using the MF method was investigated by adding 10 mg/1 of DDT
to an overflow sample for one hour and dosing the sample with
chlorine for contact times of 30, 60, 120, 180 and 300 seconds*
The results of this investigation are presented in Table 4.

A sample containing 10 mg/1 of DDT was compared to a sample
containing no DDT, with no significant difference in TC levels
observed using the MF technique.  Since this could be the
result of high dilution factor, the test was repeated on a
disinfected sample in order to reduce the dilution factor and
increase any potential effects.  Again, no significant
difference was observed upon the addition of DDT.

Following these tests, Cl2 was added to give a concentration
of 20 mg/1 in samples that had DDT levels of 0 mg/1 and
10 mg/1, respectively.  Chlorine contact times were varied
but no significant differences in TC levels were found in
any case as. shown in Table 4.

In addition to the studies conducted on SCSO, a brief deter-
mination was performed on pure cultures of TC.  Confirmed
colonies of TC were lifted from the plates and grown in
nutrient broth.  Aliquots of the inoculated broth were placed
in buffered solutions, and the stuides were performed on
these solutions.
                              20

-------
Table 4.  Effect of DDT on Total Coliforms Using MF
Sample
 DDT     Cl2      Contact    Total Coliform
(mg/1)  (mg/1)  Time  (sec.)  (counts/100  ml)a
Overflow 10
0
10
0
10
0
10
0
10
0
10
0
10
0
Pure Culture 10
0
0
0
0
0
20
20
20
20
20
20
20
20
20
20
0
0
0
0
0
0
30
30
60
60
120
120
180
180
300
300
0
0
6,300,000
8,700,000
12,000,000
9,400,000
410,000
560,000
135,000
130,000
12,000
32,000
9,400
6,300
5,300
1,700
552
535
   Eacn  count  is  the  average  of  three blended  replicates

To further  assess  the  validity of  using  the MF technique  for
bacterial measurements of  CSO, the bacterial counts  of  a
series of untreated and chlorinated SCSO were  determined  by
both the MF and MPN methods.  The  results  in Table 5 indicate
very little difference in  the two  methods.

Table 5.  Comparison  of MF and MPN Results in  SCSO	
Sample   Bacteria
                 MPN            MF
   C12     (counts/100 ml)  (counts/100 ml)
SCSO(l)
SCSO(l)
SCSO(l)
SCSO(l)
SCSO(2)
SCSO(3)
SCSO(4)
SCSO(5)
SCSO(l)

SCSO (2)
SCSO (3)
SCSO(4)

SCSO (5)
TC
TC
TC
FS
TC
TC
TC
TC
TC

TC
TC
TC

TC
0
0
0
0
0
0
0
0
15 mg/1 for
one minute
n
n
25 mg/1 for
one minute
n
9,000,000
9,000,000
7,000,000
900,000
7,000,000
5,000,000
7,000,000
9,000,000

1,200
1,400
700

7
21
11,200,000
8,600,000
10,800,000
766,000
6,800,000
7,200,000
7,700,000
8,500,000

960
1,100
840

13
19
   All  samples  were  blended
                             21

-------
These investigations indicated that the objections to the MF
method voiced in Standard Methods, 13th ed., do not apply to
SCSO.  Since the SCSO in this study was chosen to represent
the first flush or most heavily contaminated CSO, this
conclusion may be generalized to include CSO.  However, in
CSO there is the possibility of a slug of toxic material or
TSS higher than the values considered in these studies.
Therefore, both MF and MPN methods will be used periodically
during the remainder of the studies, particularly in long-
term composite samples.

Blending—A second problem in the bacterial analysis of
untreated and disinfected wastewaters is the protection of
the bacteria by harboring within the interstices of solid
particles and grease.  This results in bacterial counts
that may not reflect the actual number of bacteria in the
sample.

Sonication and blending were considered as methods for
exposing the protected bacteria prior to analysis of the
samples.  A literature search of the methods of solids
maceration with the use of ultrasonic waves indicated that
under the influence of ultrasonic fields with frequencies
commonly used for maceration, cell rupture may occur.
Significant kills of various strains of FC were observed
when subjecting the bacteria to a sound field for a period
of twenty minutes20.  Early studies showed that cell rupture
could occur within 0.0005 seconds under similar ultrasonic
fields2^.  Therefore, sonication could cause bacterial kills
before enumeration and did not appear acceptable.  It was
decided to conduct a preliminary laboratory study on the
effects of mechanical blending.

Samples containing both high (>1,000,000) and low(<100)
counts/100 ml as measured by the MF procedure were blended
for varying lengths of time to obtain an optimum blending
time.  The low counts were obtained by disinfecting a sample
of SCSO with 15 mg/1 of Cl2 at a contact time of five
minutes.  A conventional household blender, Hamilton Beach
Model 50, was run at maximum speed for zero, two, four, six,
eight, ten, twenty, thirty and sixty seconds.  The results
plotted in Figure 5 indicate that six seconds was the
optimum blending time for this model.  A similar study
should be made for any blender used for this purpose.

On the basis of the results from the MF and blending tests,
the MF technique was used to measure all bacteria unless
otherwise stated with a minimum requirement of three
dilutions per sample and all. samples were blended.  When the
                             22

-------
IV)
U)
                 E  lo'r-
                    •••I-  ?"
                         ;
                             /
                                                             o. e
                                                             Si P
DISINFECTED WITH l5mg/ICI2   ^ «
  lor 5 minut«s            O z
                     1«1 o
                     ^8
                                        \
                                         v--*
                         BLENDING TIME. iecond»
                                                                          DISINFECTED WITH 15 mg/l
                                                                            for 5 irnnjlet
                                                                              1  I/  I  M  1
                                                                    0  2   4   6  8   IO  30  *O
                                                                        BLENDING TIME. s*condi
                                                                                                               -
                                                                                                           o e
                                                                                                           u. o
                                                                                                           _l O
                                                                                                           8S
                                                                                                           Sg «>•,
                                                                                                                          >
                                                                         0  2   4   6  8   IO  30  *0

                                                                             BLENDING TIME,  seconds
-O-O- UNSCREENED
     SCREENED
     (23 MICRON)
               Figure 5.  Effect  of blending  on  bacterial  counts

-------
MPN technique was used, five tubes of each dilution of 10 ml,
1 ml and 0.1 ml of sample were carried through the confirma-
tory test.  It is intended to use the MPN procedure periodi-
cally in the remaining phase of this study as a check on the
MF procedure.

Virology

Viruses to be Studied—

As mentioned previously, the presence or absence of indica-
tor bacteria does not absolutely confirm or deny the pres-
ence of viruses.  It was, therefore, decided to study the
effects of disinfection on several viruses in order to
compile more information as to the validity of using indi-
cator bacteria.  Viruses are normally present in sewage and,
therefore, in overflows in such low numbers that their
detection requires a large amount of sample.  This should
not imply that virus levels are sufficiently low that sewage
may be considered safe.  There is no minimum amount of
virus that may be considered safe and no one has ever deter-
mined viral standards.  The ingestion of even one viral
particle is sufficient to be an infectious dose for some
viruses.

The effects of disinfection on viruses are best assessed by
determining the net reductions in population as measured in
logarithms.  The viral counts in most sewage are such that
a five log reduction can be assumed to yield zero virus.
In order to observe these five log reductions, the SCSO is
seeded to give an approximate initial count of 10^ organ-
isms/ml.  It is assumed that a reduction from 10$ to zero
organisms/ml can be accomplished under the same disinfectant
conditions as a reduction from 10^ to 10-^ organisms/ml.
This is based on the fact that the number of organisms is
negligible compared to the number of disinfectant molecules
generally observed under no-demand conditions with adequate
mixing.  Therefore, this is an idealized case of Chick's Law
in that the rate of kill is only a function of disinfectant
concentration and time.  The initial concentration of the
species to be disinfected has no effect.  The assumption is
valid for viruses in polluted water because: a) the viruses
remove a negligible amount of active disinfectant even at
concentrations of 10^2 organisms/ml, b)  the inactivation of
one virus particle is independent of the inactivation of a
second, c) the same proportion of the population is inacti-
vated regardless of the starting population, and d) the dis-
infectant demand is the same regardless of initial viral
population.
                             24

-------
The majority of the virus studies were performed on three
viruses, bacteriophage f2, bacteriophage 0X174 and polio-
virus-1.  The former two viruses were chosen because of
their relative ease in preparation and assay, their lack of
effects on humans and previous work that indicated these
viruses were similar to pathogenic viruses in resistance to
disinfection22.  Poliovirus-1 was chosen because it is a
typical infectious, pathogenic organism.  For other aspects
of the study, several other viruses were used.  The names,
strains, common abbreviations and associated diseases of
the viruses are listed in Table 6.

Table 6. Viruses Considered for Disinfection Study	
Name
Bacteriophage
Bacteriophage
Poliovirus
Herpes Simplex
Virus
Strain
f2
0X174
Sabin
Type 1
L2
Abbreviation
f2
0X174
Polio-1
HSV-L2
Disease
Bacterial Virus
Bacterial Virus
Polio
Fever Blisters
Newcastle Disease
  Virus
Vaccinia Virus
California
Sharp
Yellow Fever Virus  17D
Coxsackie Virus   B-3
Echovirus         7
NDV           Chicken Virus
vaccinia      Smallpox
YFV           Yellow Fever
Coxsackie B-3 Encephalitis
ECHO-7        Meningitis
Growth ,  Purification  and Assay  Procedures —

Bacteriophage  f2 was  grown  on a RNase  1" mutant, Escherichia
coli  (E.coli)  A19  ^'   with yields of  approximately  1 X lO^2
plaque~~forming  units  (PFU)/ml.  Optimum  conditions  for the
production and  assay  of  f2  required  two  media  containing
 (per  liter):  (A)  tryptone-salt medium — 10  g tryptone,
5 g NaCl; and  (B)  complete  medium — 10  g  tryptone, 10 g
NaCl, 5 g yeast extract,  1  g  glucose.  The A19  cells were
grown for 17 hours  in tryptone-NaCl  at 37°C with vigorous
aeration; 5 ml  of this overnight culture was used to
inoculate 100 ml  of complete  medium.   After 1  to 2  hours
aeration  (optical density = 0.065),  f2 phage were added at
a multiplicity  of infection of 5  (approximately 1 X 109
PFU/ml final concentration) ,  and the culture was incubated
for an additional 3 to 4  hours.  The cells were then lysed
by the addition of  50 ml  of chloroform and 20  ml of 10~3
molar (M) EDTA, pH  7.0,  and stored at  4°C  overnight.  The
lysate was centrifuged in sterile bottles  to remove cell
debris, decanted  into sterile containers and stored at 4°C.
                             25

-------
Consistent f2 plaque formation could be obtained conveniently
with A19 cells which had been stored at -65°C.  The cells
were grown for 17 hours in tryptone-NaCl .and preserved by
the addition of 20 percent glycerol (wt/vol) and freezing at
-65°C in 5 ml aliquots.  After thawing at 37°C, 0.25 ml A19
was added to 2.5 ml of 0.7 percent soft agar overlay
containing dilutions of f2.  Assay plates and overlay
required the complete medium for plaques.

Bacteriophage 0X174 was grown in broth cultures of E.coli C
according to the method of Sinsheimer^S with a yield of
5 X IQlO PFU/ml.  Purified virus was prepared by differential
centrifugation followed by isopycnic banding in CsCl
(P = 1.42).  CsCl was removed from virus containing fractions
by dialysis against 0.01 M phosphate buffer, pH 7.0.

The polio-1 was passed two times in human HEp-2 cells to
yield titers of 107 PFU/ml.  The virus was grown and stored
in Eagle's Minimum Essential Medium with Earl's balanced
salt solution, containing twice the prescribed amount of
glutamine, 100 units penicillin, 100 ug dihydrostreptomycin
and 50 ug nystatin per ml.  Agar overlay for plaque assays
contained the same medium plus 5 percent calf serum and 0.7
percent ion agar #2 (Oxoid).  The virus was purified by
differential centrifugation and by gradient density
centrifugation in 11 to 45 percent (wt/wt) continuous
sucrose gradients.  The other viruses used in this study
were prepared in a similar manner to the above, but grown
in the appropriate tissue culture.

Recovery of Virus from SCSO by Membrane Filtration—

The method of R.N. Rao (unpublished data) was used success-
fully to recover viruses from SCSO (1:1; tapwater:sewage)
as well as actual CSO.  These waters were clarified by  (1)
centrifugation in 250 ml bottles at 3000 rpm for 30 minutes
on an IEC centrifuge, (2) filtration through a membrane
prefilter, or '(3) filtration through a 23 micron aperture
screen.  The pH was adjusted to 3.0 with HC1 and 120 mg
MgCl2/100 ml was added.  The water was filtered through a
0.45 micron filter.  The filters were removed, cut into
pieces and placed in a beaker with 30 ml three percent beef
extract, pH 8.0.  The sample was allowed to stand 30 minutes
at room temperature with stirring or 20-60 second sonication.
Eluted samples could be preserved several days by shaking
with a few drops of chloroform and transferring the super-
natant solution to sterile tubes.  Storage at 4°C in beef
extract-MgCl2 rather than freezing prevented losses in virus
due to clumping and osmotic shock.
                             26

-------
Table 7 summarizes typical experiences with this technique
using three enteroviruses as test subjects.  All three were
recovered with high efficiencies.  Sonication and prefil-
tration commonly result in recoveries greater than 100
percent due to dissolution of virus aggregates.  Prefil-
tration consistently improved recoveries on the membranes as
well as increasing the volume which one filter can handle.
Reproducibility of these relatively high PFU counts was
acceptable.

Table 7.  Recovery of Viruses from SCSO with Membrane Filters
Treatment
Virus inoculum (1 ml)
Virus sonicated in beef
extract eluant
Virus concentrated on MF
without prefilter, soni-
cated, eluted with beef
extract
PFU/ml Recovered
Coxsackie B-3 Polio-1
7.6 X 103 1.2 X 104
8.1 X 103 0.7 X 104

7.2 X 103 0.7 X 104
Virus concentrated on MF
with prefilter, sonicated,
eluted with beef extract 2.4 X 104 1.1 X 104

Echo-7
U.U ^ J.W


1.58 X 10
6.4 X 103
5.6 X 103
Extraction  of  the  prefilter with beef extract at pH 8.0 did
not  recover much of  the  virus.  However, alkaline extraction
at pH  11.0  for 10  minutes  did  increase  the yield.

The  limitation of  this direct  method of adsorption of virus
is that  the MF eventually  clogs with sediment.  Recovery of
viruses  showed that  a 142  mm MF could effectively concentrate
virus  in a  gallon  of water before  clogging if properly pre-
treated.  Table 8  shows  isolations of virus  from 30 ml raw
sewage diluted to 3  liters.  Similarly  3.5 liters of this
same material  provided good recoveries  of polio-1 when the
sewage was  artifically seeded.

The  filtration characteristics of  actual CSO showed that CSO
clogged  MF  as  quickly as raw sewage regardless  of when the
CSO  was  collected.   Figures 6  and  7 compare  the filtration
of CSO collected early and late in a storm.  Water was
collected within the first half hour of overflow and 6 hours
later  after intermittent hard  showers and continuous over-
flow.  Surprisingly,  the  later  flow was  appreciably harder to
filter than the early flow.  Prefiltration improved the
capacity of filters  with both  kinds of  water.   The experiment
may  have been  affected by  the  fact that heavy rains had been
                              27

-------
            2200 -
            2000 -
                                      PREFILTERED


                                      WITHOUT PREFILTRATION
                                                                     PREFILTEREO

                                                                     WITHOUT PREFILTRATION
                      10
                          15  20  25  30  35  40
                                                  45  0    5    10
                                                   TIME, minutes
            15   20  25  30
35
    40  45
Figure  6. Membrane  filtration  of
            early  storm  CSO
Figure  7.  Membrane  filtration  of
            late  CSO

-------
Table 8. Filter Recovery Tests on Virus in Diluted Sewage
Material Total Volume (ml) PFU/ml Recovered
Inoculum
Sewage diluted 1:100 with
tap water



Sewage diluted 1:100 with
tap water and seeded with
the inoculum



3



3
3
3
3
1

,000



,500
,500
,500
,500
1.





7.
5.
7.
1.
27





3
7
1
1
X 10*

60
40
117
115
X 103
X 103a
X 103
X 104
-a No MgCl2 used in the eluant

experienced almost daily for the two weeks previous to the
sample collection and consequently few TSS could be expected
to have accumulated.  Thus there was very little flushing of
the sewers by the early CSO; the later flow was more turbid
with clay or soil particles in suspension.

Other Procedures—

One ml portions of stock virus solutions were added to the
SCSO before any screening or other pretreatment.  The entire
supply of SCSO was thoroughly shaken and allowed to stand for
one hour to insure adequate dispersion of the virus.  The
lack of any information in the literature on the inactivation
of virus with C102 necessitated some preliminary work in no-
demand systems of virus in various buffered solutions.
Citrate-phosphate, phosphate and borate buffers  (0.01 M) at
various pH from 3.5 to 9.0 were used to determine the effect
of pH on the antiviral activity of C1O2-  Fifty ml samples
of buffer at each pH containing 0.4 mg/1 C102 were inoculated
with 0.1 ml of CsCl purified 0X174 to a final titer of
5 X 10^ PFU/ml.  Samples were taken at regular intervals,
neutralized and controls were maintained in buffer.

In addition to free organisms, viruses may exist in sewage
while trapped within cells or bound to cells which have been
excreted from the intestinal tract.  To tests the effectiveness
of C102 treatment on such viruses, the following experiment
was performed.

Monolayer cultures of HEp-2 cells were infected with polio-1
at a multiplicity of infection of 5 PFU/cell.  The cells
which had rounded after 36 hours incubation at 37°C were
shaken off the glass, washed once by centrifugation, resus-
                              29

-------
pended in glucose potassium nutrient (GKN) buffer and counted
in a hemocytometer.  Ten ml of 9.2 X 10 6 cells/ ml were
recovered.  At the beginning of the experiment 0.1 ml of cell
suspension was pipetted into a tube containing 5 . 0 ml
tryptosephosphate broth (TPB) and 1 . 0 ml chloroform.  Since
the chloroform lysed any cells present, titration of this
sample revealed the total amount of virus present in the cell
preparation.  Another 0.1 ml of cell suspension was pipetted
into 5 ml of TPB  (no chloroform) which was centrifuged in a
clinical desktop centrifuge.  Titration of the supernatant
revealed the free virus that was in the cell preparation.
The remaining infected cells were then added to 500 ml SCSO.
The total virus in the water sample was determined by
pipetting 5 ml of the inoculated water in 5 ml of 2 percent
TPB and 1 ml chloroform.  After passing the remaining
infected SCSO through the 23 micron screen, another sample
for total virus was taken.  The remaining infected SCSO was
divided into three 100 ml samples which were treated with 4,
8  and 12 mg/1 C102, respectively.

Adenosine Triphosphate  (ATP)

General —

The quantitative  determination of ATP using the bioluminescent
firefly reaction  potentially offers a rapid alternative to
the bacterial measurements as an indicator of microbial
content of  a water sample^ , 27 , 28 _  ATP is universally
present in  all known living cells^, and  it has been shown
that non-living sources of ATP occur only under rare and
contrived conditions^.  Furthermore, ATP released by dying
microorganisms is quickly acted on by other organisms, and
converted to dissimilar phosphate forms.  Thus, ATP can be
considered  a measure of the total amount  of living matter
within a
 The  hypothesis  tested  in this study was that ATP  measurement
 yields  similar  information to the measurement  of  the
 traditional  bacterial  indicators of microbial  contamination,
 such as total coliforms, with respect to disinfection  effi-
 ciency.   It  is  generally accepted that the presence of
 coliform bacteria  is an indicator of contamination by  human
 or animal wastes,  and  subsequently the possible presence of
 an infectious viral or bacterial species.  There  are two
 questions to be considered:

      1.   Are the coliforms present in large enough quantities
          with respect  to the total number of microorganisms
          to  allow  correlation between ATP and  TC, or
                              30

-------
     2.  Might not ATP be a better indicator of microbial
         contamination than coliforms?

It is beyond the scope of this study to test the above
questions with sufficient thoroughness to make any definitive
statements.  At this time only the possibility of such
correlations can be shown.

Analytical Techniques —

The determination of ATP is accomplished by simulating the
conditions within the firefly and measuring the amount of
light emitted.  The entire process can be simplified by the
injection of a sample into the reactants contained within an
instrument suitable for quantitative light detection such as
the Dupont Model 760 Luminescence Biometer used in these
experiments.  All reagents and supplies were obtained from
Dupont with the exception of morpholinopropane sulfanilic
acid  (MOPS) which was obtained from the Aldrich Chemical
Company .

The biometer was calibrated using ATP standards following the
detailed instructions in the operations manual-^.  ATP was
extracted by filtering five ml of sample through 0.45 micron
membrane filters (Millipore Corp.) and adding one ml of a
mixture of 80% 1-butanol (Fisher Scientific) and 20% MOPS
buffer to the filter.  The pad was then rinsed twice with
2.5 ml of MOPS buffer and the total filtrate (six ml) was
stored in ice.  After twenty samples had been extracted in
this manner and stored in ice, the ATP assays were made.  The
instrument was calibrated at the beginning and end of each
run to check stability.  The entire process for twenty samples,
including reagent preparation, required about two hours.

Interference by Disinfectants —

In order for ATP to serve as a measure of disinfection, the
disinfecting agents, in this case Cl2 and C102 , must not
interfere with the firefly reactions.  To study this possi-
bility, Cl2 and C102 were added to an ATP solution to give
concentrations ranging from 1 to 100 mg/1 .  The ATP content
of each solution was measured after 60 and 300 seconds
contact time.  The results presented in Tables 9 and 10
indicate a definite reaction between ATP and disinfectants,
particularly
At the doses used for normal or high-rate disinfection, there
would appear to be only a slight interference.  In wastewater
this effect should be even less than shown in the tables,
                              31

-------
Table 9 . Effect of C±2 on ATP Standards
ATP (yg/ml X 10"1)
Contact Time
Cl2 Dose (mg/1)
0
5
10
15
25
50
100
0 seconds
1.71
1.71
1.71
1.71
1.71
1.71
1.71
60 seconds
1.36
0.980
0.517
0.104
0.0841
0.0249
0.0290
300 seconds
1.05
0.900
0.318
0.0632
0.0966
0.0155
0.0143
Table 10.   Effect of C102 on ATP Standards
                              ATP (ucf/ml X 10 1)
Contact Time
CIO? Dose (mg/1)
0
5
10
15
25
50
100
0 seconds
1.01
1.01
1.01
1.01
1.01
1.01
1. 01
60 seconds
0. 863
0.887
0.995
0.950
0.940
0.711
0.653
300 seconds
1. 06
0.886
0.925
0.574
0.482
0.236
0.354

since the ATP would become exposed only to the residual  Cl2
and C102-  However, to completely answer any questions,  it
is recommended that sufficient thio be added to the MOPS
buffer-1-butanol extraction mixture to eliminate up to
100 mg/1 of residual disinfectant.  The easiest way to
accomplish this is to add 2500 mg/1 thio to the MOPS  buffer
used in all stages of the analyses.  To test the effects of
thio on ATP, various amounts of thio were added to an ATP
solution and the amount of ATP measured after 60 and  300
seconds.  The results are shown in Table 11.

The effect of thio on ATP can be considered negligible,  and
this procedure was routinely adapted for all future ATP
measurements in disinfected wastewaters.

Correlation with Bacteria--

The correlation of bacteria with ATP was done in  two  stages
first, on untreated samples of CSO to determine the
relationship between bacteria and ATP, and second  on
                             32

-------
Table 11.  Effect of Thio on ATP Standards

                             ATP  (vig/ml X 10"1)
Thio Dose (mg/1)
0
500
1000
2500
5000

0 seconds
1.54
1.41
1.62
1.56
1.60
Contact Time
60 seconds
1.48
1.39
1.46
1.44
1.49

300 seconds
1.22
1.16
1.20
1.24
1.21
disinfected samples to determine if ATP varies in a way
similar to bacterial counts upon disinfection.  The first
stage was accomplished during the preliminary characterization
phase in which over 1000 samples of untreated CSO were
collected at different times and locations.  Whenever time
permitted, ATP, TC, FC and FS were measured in addition to
many chemical and physical parameters.  The results of this
work appear in Appendix A.  There was very little direct
correlation between ATP and either of the three bacteria as
shown in Table 12.

Table 12.  Correlation of ATP and Bacteria in Untreated CSO

Bacteria	Correlation Coefficients
   TC                                       0.033
   FC                                       0.035
   FS                                       0.034
The correlation  coefficients were  derived  from  a  least
squares  fit of over  250 points  assuming  a  linear  relationship.
The lack of correlation can possibly be  attributed  to the
variations in sources  of ATP other than  the  three measured
bacteria and to  the  difficulty  in  measuring  high  ATP values.
Future work in this  direction should include correlation
with total bacteria  counts and  pathogenic  organisms.

This lack of correlation should not be a discouraging factor
in the application of  ATP measurement as an  indicator of
microbial contamination.  The primary concern is  that ATP
can be used to differentiate between the high bacterial
counts encountered in  untreated CSO and  the  low bacterial
counts required  by effluent limitations.   If the  coliforms
represent only a small fraction of the bacteria in  untreated
CSO, poor correlations with ATP are to be  expected.  However,
if the coliforms are more resistant to disinfection than other
                              33

-------
bacteria, the correlations between TC and ATP should increase
in disinfected samples.  In fact, the more resistant the
bacteria, the higher the expected correlation with ATP  in
disinfected samples.  This implies that if ATP is used  as
a measure of residual bacteria in disinfected samples,  any
decisions about the water quality would be at least as  safe
as those based on coliforms.

Correlations in Disinfected Samples--

The second stage of the demonstration involved the measure-
ment of ATP and bacteria during the disinfection process.
It was felt that the previously observed sample-to-sample
variations in bacterial species would not be a factor pro-
viding that all bacteria had a susceptibility to disinfection
similar to or less than those observed.  Any decrease in
bacteria would then lead to a corresponding decrease in ATP,
which could then conceivably be used as a measure of disin-
fection.

Several preliminary trials were performed to determine  if
these decreases could be observed.  The results given in
Table 13 indicate that parallel decreases occur upon dis-
infection.  At this point in time the bench-scale studies
had progressed to the two-stage disinfection studies to be
discussed later, and ATP was added to the parameters to be
measured in this phase of the study.   The results of this
extensive work will be presented and discussed at length in
the next section.

Chemical Parameters

The side effects of disinfection were determined by studying
the changes in certain chemical parameters.   pH, TOG, NHoN
and TKN should be most susceptible to change in the presence
of C12 and C102.  pH was measured with a Beckman Zeromatic
pH meter that was standardized with pH 'buffers 7.0 and 10.0
(Harleco Chemical Company).   TOG was measured with a Beckman
Model 915 Total Organic Carbon Analyzer, and TKN and NH3N
were measured with a Technicon Autoanalyzer II, using the
hypochlorite reaction, with and without digestion,
respectively.  The results of the chemical studies are given
and discussed in the next section.

STUDY OF VARIABLES INFLUENCING DISINFECTION

Introduction

The main purpose of the bench-scale study was to determine
                             34

-------
Table 13 .  ATP and Bacterial Reductions Upon Disinfection
Disin-
fectant
Cl2
C12
ci2
Cl2
C12
C12
C12
Cl2
C12
Cl2
C12
C12
cio2
C102
C102
C102
C1O2
C102
C102
C102
C102
C102
C102
C102
Dose
(mg/1)
0
5
10
15
25
100
0
5
10
15
25
100
0
5
10
15
25
100
0
5
10
15
25
100
Contact
Time
(seconds)
30
30
30
30
30
30
300
300
300
300
300
300
30
30
30
30
30
30
300
300
300
300
300
300
ATP
(ug/mlXlQ-1
632.
125.
75.
73.
6.
1.
632.
73.
22.
7.
3.
0.
512.
110.
74.
21.
4.
3.
512.
45.
27.
5.
1.
0.


0
0
80
10

0
4
00
90
910


4
2
67
93

8
6
58
76
625
TC
(Counts/
) 100 ml)
4,700,
4,900,
4,000,
250,
1,

4,700,
3,400,
44,
1,


33,000,
1,150,
300,
103,
6,

33,000,
760,
40,
34,


000
000
000
000
300
0
000
000
000
010
5
0
000
000
000
000
900
300
000
000
000
800
700
0
FS
(Counts/
100 ml)
328
326
154
13


328
59
9



322
39
21
8


322
17
5



,000
,000
,000
,300
260
0
,000
,000
,700
590
5
0
,000
,000
,800
,000
730
41
,000
,600
,300
576
36
0
the effects of  contact time, disinfectant dose and type,
temperature, method of addition of disinfectant and mixing
on high-rate disinfection.   Each  of  the  above  factors with
the exception of  the  latter  was varied in a  range that might
be found  in routine field operations while the remainder were
held  constant.  A standard procedure was established for the
addition  of disinfectants and  collection of  samples as shown
in Figures 8 and  9.

The SCSO  was prepared as previously  described  and passed
through a fine  mesh screen if  necessary  for  a  particular
trial, as  further  described.  The  SCSO was then split into
the requisite number  of one  liter portions which were kept
in 1500 ml beakers or jars,  hence the name jar tests.  In
all cases one jar was reserved for a control sample.  The
desired disinfectants in the various dosages were added to
the jar and rapidly mixed for  fifteen seconds.  The jars
                             35

-------
                                                                                                         RAW SEWAGE
OJ
                                      RAW SEWAGE



                                       DISTILLED  WATER

                                       SIMULATED  OVERFLOW
                                     23 MICRON SCREEN
                                           OR
                                         NO SCREEN
  »     V    T    V     T    V     T    DISINFECTANT
0 mg/I  4 ma/I emg/l I2mg/l  I6mg/l 20ng/l 25mg/l     DOSAGE
                1     i    L    I   1 1 1
                <*>    ^   <*    i   | «t ]
1    T—}
     >0n




      I
                         SAM PL E S  TAK EN  AT
                     0,15,30,60,120,180,300  SECONDS
                                        15 SECONDS
                                         MIXING
                                         ' UTER
                                        BEAKERS
                                      1     1      1
                                                                                                   T
                                                                                                  5 gait
                                                                                    T
                                                                                                         DISTILLED  WATER

                                                                                                         SIMULATED  OVERFLOW
                                                                                         23 MICRON  SCREEN
                                                                                               OR
                                                                                             NO SCREEN
                                                                  fVtttTTY   DISINFECTANT
                                                                4 rog/I  4 mg/l  8mg/[  8mg/l  2mg/l 2mg/l  4mg/l 4 mg/l   DOSAGE
                                                                 e»    Jo    Jo    tk    <}=>    JL

                                                                I	 CI2	1 OR I	••	
                                                                                                                           15 OH 30
                                                                                                                         SECONDS MIXING
                                                                               Zing/1
      T    T     T    T     T     »     Y  DISINFECTANT
     I mg/l  2mg/l  4mg/l  2mg/I  4 mg/l  2mo/l  4mg/l  DOSAGE
JLl |JL|   J.I  11
I

I

L
                                                                                                                          CONTINUOUS
                                                                                                                           MIXING
                                                                             SAMPLES  TAKEN  AT
                                                                         O,J5,3O,6O,I20,[8O,3OO  SECONDS
                                                                               Rgure 9.  Two-stage disinfection
              Figure 8.  Single-stage disinfection outline                    .outline

-------
were maintained at the desired temperature and samples were
withdrawn at specified intervals for chemical, viral,
bacterial and ATP measurements.  These samples were collected
in sterile, 125 ml bacteria bottles to which 12.5 ml of
0.025 N sterile thio had been added to halt the action of the
disinfectant at the desired time.  The resulting concentration
of thio in the sample is approximately 600 mg/1 which is
sufficient for the neutralization of the maximum dosages5.
The neutralization reaction is instantaneous upon mixing and
the excess thio does not interfere with any of the parameters
to be measured.

A volume correction was made on all values to compensate for
the addition of thio.  However, in the case of disinfectants,
the volumes were sufficiently small  (1-2 ml) that no cor-
rection was necessary.

A portion of the sample was taken from the bacteria bottle
and prepared for the virus determinations.  For virus assay,
5 ml of each neutralized sample was pipetted into a screwcap
tube containing 5 ml of 2 percent tryptone broth and 1 ml
chloroform.  The tubes were shaken, the phases were allowed
to separate, and the tryptone broth was pipetted aseptically
into a sterile screwcap tube.  These samples were refrigerated
until virus titration could be performed.  A portion was
taken for chemical analysis and the remainder used for
bacteria and ATP measurements.

Selection of Contact Times

The contact times for disinfection were selected after a
review of the overflow conditions that exist within the
Syracuse Interceptor System had been completed.  Dye studies
were performed on actual overflows under various flow condi-
tions to determine the time between overflow regulator
diversion point and the discharge of overflows to the
receiving water body.  Since the area available adjacent to
the various overflow points is limited, the size of contact
tanks for disinfection will be necessarily limited in size
thereby limiting total detention time.  The following range
of contact times were chosen as being representative of
possible field conditions: 0, 30, 60, 120, 180 and 300
seconds.

Disinfectant Dosages

It was felt that given the short contact times in high-rate
disinfection, Cl2 dosages up to 25 mg/1 might be necessary
to achieve the desired levels of 1,000 TC/100 ml and
                              37

-------
200 FC/100 ml.   The values chosen for C12 were 0, 4, 8, 12,
16, 20 and 25 mg/1.  The same dosages of C1C>2 were chosen to
facilitate comparison of the two disinfectants.  It is
recognized that other disinfectants including ultraviolet
light (UV), bromine (Br2)  and ozone (O3) do exist, but the
scope of this project includes only Cl2 and C1O2.

Screen Mesh Size

Fine-mesh screening prior to disinfection may act in two
ways.  First, a fraction of the particulate matter, and
microorganisms absorbed therein, will be removed on the
screen.  Subsequently, disinfection may be enhanced by
decreasing disinfectant demand and reducing the size of
protective solids.  The bacteria and TSS values obtained
by screening through 23, 74 and 149 micron aperture are
shown in Table 14.

Table 14. Effect of Screening on Bacterial Counts	
Screen
Opening
Trial (microns)
1 unscreened
149
74
23
2 unscreened
149
74
23
TC
(counts/
100 ml)
2,770,000
3,750,000
2,920,000
2,720,000
5,600,000
6,100,000
5,200,000
4,000,000
FS
(counts/
100 ml)
53,000
67,000
62,000
69,000
	
	
	
___
Total Suspen-
ded Solids
(mg/1)
270
206
166
142
161
112
62
58
Although the removal of TSS is a desirable effect,  the  main
purpose in screening in this study is to aid  in  disinfection.
The  role of screen size in disinfection is covered  in detail
in the discussion section.

The  actual screening process was carried out  after  the  raw
sewage was diluted 1:1 with distilled water.   The entire
batch of SCSO was screened under 20 inches vacuum prior to
any  disinfection studies.

Temperature Effects

The  climatic conditions in Onondaga County are such that the
influent temperature at the existing Metropolitan Treatment
Plant varies from 8°C to  21°C3.  The CSO may  be  subject to
even larger extremes of temperature.  Therefore, it was
                             38

-------
decided to study disinfection at three different temperatures,
2°, 22° and 30°C.  Unscreened samples were brought to
constant temperature at 2°, 22° and 30°C before being
disinfected with C3-2 and C1C>2.  Procedures for sampling and
analysis were identical to those used in the other jar
tests as described in this section.

The temperature of SCSO samples was lowered to 2°C by
refrigeration of one liter portions.  The jars were
maintained at 2°C while performing the disinfection and
sampling by keeping the sample in ice.  The tests run at
22°C were performed at room temperature and were monitored
for fluctuations.  The remaining tests were performed by
keeping the sample in a water bath and maintaining the
temperature at 30°C.

Additional studies were performed using only C1O2 and virus.
For these studies one ml of poliovirus-1 was pipetted into
50 ml SCSO in duplicate, pre-warmed bottles.  The bottles
were shaken and then one bottle was treated with 4 mg/1 C102•
Dilutions of samples from both bottles were made and titered.
Inactivation curves were obtained at 4°, 12°, 22° and 30°C.

Addition Techniques

The effects of sequential or two-stage addition of one or
both disinfectants were compared to single-stage disinfection.
The initial concept was to use Cl2 as the first disinfectant
to reduce the demand and then add C1C>2 which could act
primarily as a disinfectant, as indicated by previous studies
on multi-stage addition of disinfectants33.  The two-stage
studies were expanded to investigate this phenomenon more
thoroughly.

A series of two-stage reactions using Cl2 in 4 and 8 mg/1
dosages and C1C>2 in 2 and 4 mg/1 dosages were designed and
performed as illustrated in Figure 9.  A sample of simulated
overflow was screened (23 micron), divided into 1500 ml
portions, and an initial sample taken.  The first disin-
fectant was added.  After the appropriate interval, the
second disinfectant was added.  Samples were withdrawn
into bottles containing an excess of sodium thiosulfate for
analysis of TC, FS and ATP.  The parameters TOG, NH3N, NC>2N,
N03N, TKN, pH and temperature were measured for a few trials,
but when none showed more than a 5% change from initial
values, they were dropped from further trials.
                              39

-------
Mixing

Mixing is recognized to have an important effect on the rate
of disinfection3^.  However, it is difficult to simulate  the
different methods of mixing that can be used during a  full-
scale operation in a bench-scale study.  Therefore, the
effects of mixing will be assessed in a later phase of this
project.  In the jar tests, mixing was carried out very
rapidly  (within 15 seconds) by a magnetic stirrer.

IMPACT STUDIES

Aftergrowth in Receiving Waters

The assessment of the effects of the discharge of treated CSO
into the receiving waters, in this case, Onondaga Lake and
several of its tributaries, is an important part of the
study.  The discharge of residual disinfectants and their
by-products can have a toxic effect on plant and animal
life35.  Although the disinfection process will reduce the
bacterial and viral population to an acceptable level, the
phenomenon of regrowth in the receiving water is well  known3^.

In order to determine these effects in Onondaga Lake,  a
series of tank tests were conducted.  A sample of disin-
fected SCSO was added on a 1:1 ratio to Onondaga Creek water
taken immediately before the Syracuse city limits.  This  is
the ratio of total overflow volume, from an average storm
for the  entire city, to minimum dry-weather creek flow.   This
mixture was vigorously agitated for one hour which represents
one-half of the transit time in the creek for an average  wet-
weather  flow.  Finally, the mixture was added to Onondaga
Lake water at a 1:14 ratio that simulates the ratio of the
above creek flow to the epilimnion of the southern end of the
lake.  The calculations that support these estimated ratios
can be found in Appendix C.  These conditions were chosen to
maximize the impact of the treated SCSO and therefore  maximize
the safety margin.

The actual tests were performed on SCSO, screened  (23  micron)
and unscreened, and disinfected with 25 mg/1 Cl2 and 12 mg/1
C102 for 30 and 60 seconds.  Additionally, a non-disinfected
sample of SCSO was used for control purposes.  The contact
times selected for design considerations indicated that these
would be the maximum times available in full-scale demonstra-
tion facilities.  The dosages were selected to reduce  the
bacterial counts to extremely low levels  (<100) in order  that
the difference between initial and final aftergrowth counts
could be more easily discerned.
                              40

-------
The chemical composition of the different samples of SCSO,
creek water and lake water was found to be essentially
constant throughout the tank tests.  All work was conducted
at temperatures between 20°C and 25°C, the dissolved oxygen
in the tanks was kept at four mg/1 or greater and normal
room lighting conditions were maintained.  These were chosen
to optimize aftergrowth.  The disinfectant was added to the
SCSO and mixed for fifteen seconds.

Samples were then taken from the overflow before disinfection
and immediately before addition to the creek water according
to the outline in Figure 10.  Prior to the addition to the
lake water, samples were taken from the overflow and creek
mixture, including the control, at 20-minute invervals.  A
second control consisting of creek water was sampled and
added to lake water.  These four mixtures plus a third control
of lake water were sampled at various times up to three days,
at which point the bacterial regrowth was assumed to have been
completed.  The three indicator bacteria, TC, FC and FS,
several viruses and Cl2 and C1C>2 residuals were measured on
all samples.

Nitrogen Removal

The nitrogen and phosphate compounds in CSO represent a
significant percentage of the total contribution of nutrients
to Onondaga Lake-^.  The complete reaction between Cl2 and
ammonia  (NH3N), called breakpoint chlorination, is one way to
remove NH3N from wastewaters.  Since a full-scale application
of breakpoint chlorination could be accomplished with approxi-
mately the same equipment as is used for disinfection with
chlorine, a bench-scale investigation was initiated to eval-
uate this possibility.

If the pH is maintained between 6.0 and 7.0, a theoretical
dose of 7.6 gm Cl2 per gm of NH3N results in the following
reaction37:

     3 C12 + 2NH3	— N2+ + 6H + 6 Cl~

The products of this reaction change drastically if the pH
varies from the above range:

     pH <6.0 3 C12 + NH3 	»-NCl3 + 3H + 3 Cl"

     pH >7.0 3 H20 + 4 C12 + NH3	~N°3~ + 8 cl~ + 9H+
                              41

-------
                                                  BEAKER
                                                                           JAR
t\J
                                                                                    DISCHARGE TO
                                                                                                     TANK
OVERFLOW

\
SCREENING 8 /
DISINFECT ON /
<4 CONDITIONS)





0
CONTACT
TIME
0 seconds
CONTINUOU
30
SECONDS
CONTACT
TIME
30 seconds
CONTINUOU
60
SECONDS
CONTACT
TIME
ENTRY INTO CREEK
1:1

s
1:1

S
1:1 to



20,40,60mlnute
CONTINUOUS

20, 40, 60 minute
CONTINUOUS
LAKE EPIUMNION
KI4

9
1:14 _

t
1:14


0,l9mlnutes, 1,4,12 h
,2,3 doys
0,19 minutes, 1,4, 12
,2,3 doys
our*
hour*
                                                60 seconds


                                                CONTINUOUS
                                                                                                                  SAMPLINQ INTERVAL*

                                                                                                                  MIXING
                                                                                                                  SAMPLING INTERVALS

                                                                                                                  MIXING
20,40,60m,Bu...              f.'iV,,,1""' *'*' " *"""* SAMPLING INTERVALS
                                                                        CONTINUOUS
CREEK
WATER
CONTROL

n4 to


                                                                       20,40,60mlnul»


                                                                        CONTINUOUS

CONDITION




A
•
C
D
DISINFECTANT
CI2
CI2
CI02
ClOg
SCREENING.
0
23 MICRONS
0
23 MICRONS
                                                                                                                  MIXING
                         19 minutes, 1,4, 12 hours  SAMPLING INTERVALS
                         1,2,3 days
                                           MIXING
                                                                                                19minutes, 1,4,12 houn  SAMPLING INTERVALS
                                                                                                I, 2, 3 doys
                Figure   IO.  Aftergrowth  studies   for   maximum  flow   conditions

-------
Both of these reactions are undesirable since nitrogen tri-
chloride  (NC13) is an odorous, toxic material while nitrate
(NC>3N) is another nutrient.  The natural buffering capacity
of CSO may maintain the pH in the desired range, but if not,
some form of pH control will be necessary.  In tertiary
wastewaters the weight ratio required  for breakpoint is
greater than the theoretical value of  7.6:1, rising to nine
or ten to one in secondary and primary effluents, respectively.
In screened CSO, the ratio is expected to be somewhere between
these latter two figures.

The average NH3N concentration ranged  from 5 to 10 mg/1, so
Cl2 dosages were selected up to 100 mg/1.  The experiment was
performed in a manner similar to the jar tests except for the
higher dosages.  Samples were taken at 30 and 120 seconds for
each dose and analyzed for TKN, NH3N,  (N03N + N02N) total
alkalinity  (TALK) and residual C12.  The thio used to termin-
ate chlorination interfered with the test for N02N, so that
separation of the NO3N + N02N measurement into its two compo-
nents was impossible.  All the above tests were performed on
a Technicon AutoAnalyzer II using standard Technicon Method-
ologies38 except for residual C12.  The DPD method was used
to analyze for free C12 and mono-, di-, and tri-chloramine8.
The pH was measured continuously and occasional samples were
examined  for TC, FC and FS.
                               43

-------
                          SECTION V

                   RESULTS AND DISCUSSION

SINGLE-STAGE DISINFECTION

Chlorine

Presentation of Results—

Bacteriology—The results of disinfection with Cl2 using TC
and FS as a measure of efficiency are graphically presented
in Figures 11 and 12,  respectively.  The tabulated data from
which the graphs were  drawn is contained in Appendix B.  The
same four trials are presented in the same order in each
figure.  The individual graphs in each figure represent a
separate trial for which a different SCSO was used.  The
effects of contact time, Cl2 dosage, and screening can be
determined from these  figures.

A detailed examination of Figure 11 shows that within the
two minutes specified  for high-rate disinfection, a dosage of
16 mg/1 Cl2 achieved the required TC level (1,000 counts/ 100
ml) for only 25 percent of the trials.  A dosage of 20 mg/1
showed no improvement  in the results.

The results for FS showed a similar variability.  A standard
of 200 FS/100 ml was chosen as a target level since the
characterization study showed that FS levels in CSO were
approximately 20 percent of the TC levels.  This arbitrary
approach was taken since published standards for FS levels
could not be found. With the two-minute contact time, a
dosage of 25 mg/1 Cl2  or greater was necessary to reach the
target value.

The variations observed in duplicate trials is of the same
magnitude as the difference between screened and unscreened
trials.  Based on this limited data, the direct effects of
screening as performed in this study are negligible.

These results are in rough agreement with the limited amount
of previous work on high-rate disinfection of CSO.  In a
study at Grosse Point  Woods, Michigan39, four log reductions
of TC and FS were obtained in CSO with dosages of 8.0 and
10.8 mg/1 C12, respectively, in 6.4 minutes.  These compare
favorably with the observed three to four log reductions
                             44

-------
obtained in two and five minutes.  The lower dosages were most
likely successful because of a lower Cl2 demand.  Glover and
co-workers achieved three to four 4°f41 iog reductions with
1 mg/1 C124U in one study and 5 mg/1 Cl2 in another42 both in
a two-minute contact time.  Again the lower dosages are attri-
buted to a lower Cl2 demand.  Although no information on this
subject is given, the  studies cited used actual CSO and not
necessarily a first flush.  Therefore, this study, in which
the SCSO was chosen to represent a first flush, should require
higher dosages.  Disinfection of treated CSO with Cl2 in
Milwaukee resulted in  three to five log reductions in TC in
ten to twenty minutes43.

There is considerable  literature on the other aspects of dis-
infection of bacteria with Cl2 but the above summarizes most
of the studies on high-rate disinfection of CSO.  In order to
comprehend the nature of disinfection with Cl2 it is necessary
to consider other aspects.  In buffered solutions of pure
cultures or in drinking water supplies, five log bacterial
reductions are observed in two minutes or less with no
decrease in free C12 residual44.  In conventional sewage
treatment effluent disinfection, a common approach is to
maintain a total C12 residual of one to five mg/1 for a
contact time of fifteen minutes to several hours4^'4^.  It
would not be practicable to list the numerous references to
disinfection with Gig/ but two recent review articles sum-
marize the field47'  .

Virology--The inactivation of 0X174 was measured in order to
determine how much variation in kill would occur when different
batches of SCSO were used.  Six replicate experiments were
performed over a period of five weeks with Cl2 as the disin-
fectant.  Background counts of phage were determined in each
experiment and the background was always less than 0.005
percent of the 0X174 test particles.  Figure 13 shows the
survivors obtained with 8 and 20 mg/1 Cl2-  Approximately the
same degree of variation was observed at all sampling periods
for both concentrations.  Despite the fact that a fresh
sewage sample was used to make the SCSO in each experiment,
the variation observed was well within acceptable limits.
Furthermore, the variations were random and did not seem
related to any particular sewage sample.

The antiviral activity of Cl2 at several concentrations is
demonstrated in Figure 14.  The techniques used readily
distinguish a 4 mg/1 difference in C12 concentration.  The
inactivation curves in this crude water are definitely bi-
phasic (two-stage).   In the first phase (0 to 60 seconds), a
rapid inactivation of virus occurred which appeared to be
                              45

-------
                       SAMPLE INFORMATION.

                        BLCNTJCD
                        SCREENED   23m.cra
                        TEMPERATURE Z( O'C
                        »M      710
                        CH DEMAND  60O
                                                                     [to9
                                                                     j ro4
    BLENOED
    SCREENED   2 3 micro*
    TEMPERATURE 220*C
    pH      7.10
    Clz DEMAND  &00
                      120  I9O  ISO   ZIO
                       TIME, B«cofidt
                                                                                        TIME, seconds
                   Trial   I
Trial  2
                                                   SCREENED
Figure  II.  Single-stage disinfection of TC  with  CI2

-------
                     SAMPLE INFORMATION-

                      •LENOED
                      UNSCREENED
                      TEMPERATURE 22CTC
                      pH      T. 10
                      Cll DEMAND  600
                                                                      E I0»
                                                                      g

                                                                      2
  IAMPU INFORMATION'
  BLENDED
  UNSCREENED
  TEMPERATURE 225"C
  pH      720
  Cl) DEMAND  aOO
                                                                                                            J	I
                      0  ISO  ISO  210
                      TIME, ittcondt
 120   ISO  180
  TIME, stcondt
                    Trial
Trial  2
                                                 UNSCREENED
Figure  II. Continued

-------
                              SAMPLE INFORMATION'
00
                                                                         10*
  •LENDED
  SCREENED   23iMcran
  TEMPERATUfiE 21 O'C
  pH      710
  Cft DEMAND  6.OO
                                 ISO  100
                               TIME , «*
 120  ISO   ISO  2IO
  TIME, seconds
                             Trial
Trial  2
                                                        SCREENED
           Figure 12.  Single-stage disinfection of  FS with CI2

-------
                                      SAMPLE INFORMATION'
                                       BLENDED
                                       UNSCREENED
                     10* -
                     10s
VD
                                               Z2O-C
                                               T. IO
                                               &OO
                                                                                        10'
I    I    I    I    I    I    I    I
                                                                                SAMPU {NFORMATION'
                                                                                 BI.ENOEO
                                                                                 UNSCREENED
                                                                                 TEMPERATURE 22.3'C
                                                                                 pH      7.20
                                                                                 Clt DEMAND ftOO
                                                                                                              I	I
                          3O   60   9O   120  ISO   ISO   2IO   24O  27O  3OO
                                        TIME, lecondi
                                                               0   SO   M   90   I2O   ISO   IBO   2IO   240  ZTO  .3OO
                                                                              TIME, ••condl
                                     Trial  I
                                                                              Trial  2
                                                                  UNSCREENED
               Figure 12.   Continued

-------
            X D*
Ul
O
                      120
                       T»
                                                                        TIME.
         Figure  13. Effect of  sample-to-sample
                   variations on  disinfection
                   of  0X174  with CI2
Figure  14. Single-stage  disinfection
        of  0X174  with  CI2

-------
concentration-dependent as evidenced by the slopes of the
lines.  In the second phase  (60 to 300 seconds), disinfection
occurred at a definite, but  lower rate.  Regression co-effi-
cients for the slopes of the second phase of each line ranged
between 0.3 and 0.7, indicating only a very questionable
concentration dependence.  This type of low grade second-
phase inactivation was also  reported by Shuval49 in experi-
ments on Cl2 inactivation of polio-1 and ECHO-7 in which the
exposure time was extended to four hours.  It is known that
when Cl2 is applied to water the primary disinfectant formed
is hypochlorous acid (HOC1), and that chlorine-demand sub-
stances (CDS) in wastewater  compete with the microorganisms
for it46.  Since the break in the biphasic curves is sharp
(especially at the higher Cl2 concentrations), it seems logical
to conclude that after that  point no more HOC1 exists in the
water and that the remaining disinfectant action comes from
those combined Cl2 products which are commonly measured as
Cl2 residual in wastewaters50.  Kelly and Sanderson5! indicated
that combined Cl2 is viricidal for polio-1 but at a much lower
rate than HOC1.  Therefore,  a longer disinfection time would
be required.  Lathrop and Sproul5
-------
that the free Cl2 is rapidly converted to combined C±2 which
becomes the disinfecting agent and disinfection is relatively
much slower.

A brief study of the speed of the reaction between Cl2 and CDS
is presented in Table 15.

Table 15. Free Cl? Concentration in SCSQa	

                                 Initial Cl? Dosage  (mg/1)
Contact Time (seconds)
0
30
60
90
120
180
300
5
5.
0.
0.
0.
0.
0.
0.

0
9
2
0
0
0
0
10
10.
1.
0.
0.
0.
0.
0.
15
0
8
7
1
0
0
0
Ib
4
1
0
0
0
0
.0
.3
.0
.5
.2
.1
.0
20
20.
8.
2.
0.
0.
0.
0.

0
7
4
8
5
2
0
25
2b.
9.
2.
0.
0.
0.
0.

0
4
3
9
5
4
1
   as determined by DPD

Since the time of existence of the free Cl2 is the same  as the
duration of the steep initial drops in Figures 11, 12 and 14,
the implication is that free Cl2 is the primary disinfecting
agent.  The gradual decrease after the initial drop is assumed
to be due to the reduced disinfectant capability of combined
C12.

The exact mechanism through which free Cl2 attacks bacteria
and viruses is not known.   However, based on the results of
this study, the following empirical observation can be made.
It is felt that disinfection is somehow related to the lipid
(fat) concentration in the cell envelope or membrane.  FS
which has a low lipid concentration in the cell membrane5 ,
is more resistant to disinfection than TC, which has a higher
lipid concentration54.  Also, polio-1, ECHO-7 and Coxsackie B-3,
which have no envelope and therefore relatively few lipids
exposed to Cl2, are several times more resistant to disin-
fection than the enveloped viruses such as HSV and YFV.
Therefore, the observation is that the greater the lipid
content in the outer layer of the cell, the greater the  sus-
ceptibility to disinfection.

Effects of Screening—

As illustrated in Figures 11 and 12,  there appeared to be
little or no predictable effects on disinfection with C12 by
the fine-mesh screening process.  In fact, the variability in
duplicate trials was greater than any observed effect of
                              52

-------
screening.  This somewhat surprising result parallels the
findings of Glover42.  In order to understand the reasons for
this behavior, the effect of screening without disinfection
was investigated.

Screening without disinfection--The effects of screening alone
are summarized in Table 16.  Although significant TSS removals
were accomplished with a 23 micron aperture, only slight
reductions in BOD5 and bacteria were observed.  This is con-
sistent with Glover's work.  The reasons for these phenomena
are as follows42:

     1.  Natural predators for bacteria are largely removed
         by screening and are thus not present in large
         numbers on the discharge side.

     2.  Large clumps of bacteria are broken up into numerous
         smaller clumps or singlets by passage through the
         screen.

     3.  The bacterial food supply is made more available
          (more surface area is produced on the escaping
         solids) by the screening process and growth kinetics
         are enhanced.  This is perhaps reflected in some of
         the BOD5 measurements in which increases were ob-
         served across the fine-mesh screen.

Although these statements are strictly hypothetical, these
results have been observed in both full- and bench-scale
studies.  The need for future work on the mechanisms of
screening is apparent.

Screening followed by disinfection--It is somewhat surprising
that screening does not enhance disinfection in view of the
fact that primary settling, a method of TSS removal that is
inferior to screening, does enhance disinfection in domestic
sewage^b.  A possible explanation of this observation is that
in high-rate screening many of the solid particles are removed,
but many are shredded and pass through the screen.  Only a
relatively small fraction of the particles need to be shredded
to cause the total surface area to equal that before screening.
The vacuum filtration under 20 inches of Hg used to simulate
screening should have a similar shredding effect.

The rates of reaction of Cl2 with bacteria and CDS are both
quite rapid.   Therefore, if the ratio of bacteria to CDS is the
same inside a particle as on the surface, screening will equally
enhance both reactions.   The result will be no net observable
                               53

-------
       Table  16.  Effect of Screening on Actual CSO
en
Parameter
TSS
Unscreened
Screened
% Reduction
BOD5
Unscreened
Screened
% Reduction
TC (unblended)
Unscreened
Screened
% Reduction
FS (unblended)
Unscreened
Screened
% Reduction
1

161
64
60

100
90
10

6000000
2130000
64

312000
141000
55
2

139
62
55

150
153
-2

99000000
85000000
14

340000
486000
-43
3

64
14
78

22
15
32

207000
169000
18

29000
74000
-188
4

34
6
82

15
24
-60

2930000
1360000
54

36000
48000
-33
5

54
14
74

19
10
47

5000000
6300000
-26

196000
138000
30
6

270
142
47

120
117
2

17000000
12200000
22

478000
780000
-62
7

184
130
29

118
117
1

22000000
23500000
-7

497000
760000
-53

-------
increase in disinfection.  Only a method of TSS removal that
does not shred the particles can be expected to enhance high-
rate disinfection.  This is not to say that screening is not
a useful technique in wastewater treatment.  Obviously, many
microorganisms and TSS are removed by screening.  Even though
the impact of these removals is not felt in subsequent high-
rate disinfection, the long-lasting or downstream adverse
effects of screened effluents should be less than for un-
screened effluents.  Any aftergrowth of bacteria in the
receiving waters should also be decreased by screening.

The effects of screening on virus removals could not be
determined because of the extremely low levels of natural
virus.  As might be expected there was no appreciable effect
of screening on the seeded viral samples as shown in Table 17.
This was true even if the sample was stirred slowly for
twenty hours after seeding and before commencement of a test.
Therefore, it seemed reasonable to assume that free virus
particles will not be absorbed by particulate matter in such
a situation.

Table 17. Effect of Screening on Virus in SCSO
74 y
aperture
Time of
Treatment
0

24

hour

hour

Screening
Unscreened
Screened
Unscreened
Screened
PFU/ml
2.
2.
2.
2.
8 X10'
4X10
4X10
4X10
7
7
7
Percent
Recovery
100
85
100
100
2

3 y aperture

PFU/ml
2.
2.
2.
2.
2X10'
7X107
2X10?
3X107
Percent
Recovery
100
122
100
104
Disinfection tests also showed no difference in screened and
unscreened samples.

Chlorine Dioxide

Presentation of Results--

Bacteriology--Figures 15 and 16 present the results of single-
stage disinfection with C102 on screened and unscreened
samples.  The tabulated data may be found in Appendix B.

Adequate high-rate disinfection with C102, as measured by TC,
was achieved with 8 mg/1 C102 50 percent of the time and with
16 mg/1 75 percent of the time.  Satisfactory disinfection
for the remaining trial was not achieved even with 25 mg/1.
Whatever the level of disinfection, there was a rapid initial
                              55

-------
                          SAMPLE INFORMATION'
                           BLENDED
                           SCREENED   23 micron
                           TEMPERATURE 213'C
                         I   ISO   180  210
                          TIME, seconds
                                       24O   27D  SCO
SAMPLE INFORMATION
 BLEKOED
 SCREENED 23m
 TEMPERATURE 235 "C
 pH      720
 CIO? DEMAND «OO
 120  ISO   180
   TIME.ucandi
                                                                                                   210  140  ETC  9OO
                        Trial   I
                                                     SCREENED
 Trial  2
Figure  15.  Single-stage  disinfection  of  TC  with  CIO2

-------
(Jl
                  O   SO   «O  »0   I2O  ISO  160   2IO  24O  Z70  5OO
                                                                                                 SAMPLE INFORMATION-
                                                                                                  •LENDCO
                                                                                                  UNSCREENED
                                                                                                  TEMPERATURE 2U'C
                                                                                                  p H      7.10
                                                                                                  ClOi DEMAND 5.0O
                                                                                  O   5O  61
                                                                                                  TIME, seconds
                                                                                                               24O ZTO  3OO
                                Trial I
                                                          UNSCREENED
Trial  2
             Figure   15. Continued

-------
                                SAMPLE INFORMATION.
                                BLENDED
                                SCREENED  21 rr,
                                TEWPEHHTUFIE 23 3* C
CO
                               I2O  130  ISO  21O  24Q 270 3OO
                                TIME, seconds
                                                                                                 I BO  Z>0  240  270
                                                                                            TIME.itconds
                              Trial I
Trial 2
                                                          SCREENED
            Figure 16.  Single-stage  disinfection of  FS  with  CIO2

-------
Ln
                                   SAMPLE INFORMATION-
                                    BLENDED
                                    UNSCREENED
                                    TEMPERATURE 21VC
                                    P M      T.2O
                                    ClOi DEMAND 6.OO
                       3°  «O   9O  (20  130  iao  2IO  240  27O  300
                                                                                      10*
     INFOAMAT1ON <
 •LCNOEO
 UNSCREENED
 TtUWtRATUflE 21 3-C
 »H      7.10
 OOt 0£M&NO  3.00
                                                                                       0   30  «0   *0
                                                                                                       IZO  190   mo   2(O  24O   270   3OO
                                                                                                        TIME, stconds
                                  Trial  I
Trial  2
                                                               UNSCREENED
             Figure  16.  Continued

-------
dropoff that was complete at approximately 30 seconds, with
very little kill obtained after that point.

The results for FS showed a pattern similar to TC, with
8 mg/1 C102 sufficient to meet the arbitrary levels of 200
FS/100 ml 50 percent of the time.  The initial rapid drop was
also observed.  The implication is that C1C>2 itself is the
disinfecting species and that the decomposition product, C1C>2~
has little disinfectant ability.  The results in Figures 1
and 2 indicate a corresponding rapid drop in C1C>2 concentra-
tion.

There have been no reports previously published concerning
disinfection of  CSO with C1C>2, and only a limited number of
papers dealing with C1C>2 as a disinfectant57"60.  There has
been some work done outside this country, but it was difficult
to assess the validity of the data.

Duplicate trials showed wide variations similar to the trials
with Cl2-  This may be taken as an indication that the varia-
tions in chemical composition of the sample have a greater
effect on disinfection than any of the controlled factors
except time and dose.   Based on the limited data, there may
be a slight effect of screening on disinfection with C102.
Both TC and FS showed adequate disinfection with 8 mg/1 on
screened samples while the unscreened samples required 16 rag/1
and greater than 25 mg/1.

Virology—The inactivation of 0X174 by C1C>2 in SCSO is shown
in Figure 17.  Biphasic curves similar to Cl2 inactivation
curves  were evident at the lower concentrations, and the
major part of the inactivation occurred in the first 120
seconds.  Four mg/1 C102 reduced the count by 3.1 logs in the
first 120 seconds and only 3.15 logs in 300 seconds.  This
suggests that, like C12, C102 was the primary disinfectant
and reacted completely with substances in the SCSO in 120
seconds.  This same conclusion can be formed from the esr and
bacteriological studies.  Virus samples were not taken at
longer time intervals so little can be said of the residual
antiviral activity of the reaction products, presumably C1C>2~.

The results of the inactivation of polio-1 in the same SCSO
are shown in Figure 18.  Four mg/1 C102 reduced the count by
1.5 logs in 120 seconds.  Eight mg/1 C102 caused a two log
reduction in the first 30 seconds.  Higher dosages caused
reductions to below the reliable limits of the plaque assay
system.  No background virus was detected in the SCSO.
                              60

-------
             10* -
(Ti
                                                                   1-4 10
                                                                   o

                                                                   £
                                                                    of
                                  leo  210  z«o  rro  300
                             TIME. lecondi
                                                                                  120  150  tao  210
                                                                                   TIME, »«conOj
                                                                                              Z4O  zro  300
           Figure 17. Single-stage disinfection
                      of  0X174  with  CI02
Figure  18.  Single-stage
            disinfection of
            polio-l with CIC>2

-------
Clearly the polio-1 was much less sensitive to C102 than was
0X174 (compare Figures 17 and 18): 4 mg/1 C102 caused 3.0 log
reduction of 0X174, but only a 1.5 log reduction of polio-1.
No long-term residual C102 activity was noted for either virus.

These results represent the average for three trials.  The
trial-to-trial variations were no more than twenty percent for
each time-dose point.  As mentioned previously, screening has
no effect on virus reductions.  The studies were carried out
using only 0X174 and polio-1 since it was felt that these
viruses were representative of most pathogenic species in
their response to disinfection.

The lack of references to disinfection of virus with C102 in
any media indicated that some preliminary studies were
necessary.  The data in Table 18 shows the relative sensitivity
of several other viruses to C102 in SCSO.  The experiments
revealed an apparent difference in sensitivity between envel-
oped viruses (HSV, NDV, YFV and vaccinia) and the two other
viruses.  Four mg/1 C102 reduced the former by 2 logs and the
latter by less than one log.  Eight mg/1 reduced HSV, NDV and
YFV by 5 logs or more under conditions which reduced ECHO-7
and possibly Coxsackie B-3 by less than a half log.

The relative sensitivity of various water-borne indicator
organisms in SCSO was compared (Figure 19).   It is apparent
that the polio-1 was least sensitive and FS were less sensi-
tive than TC.  0X174 was considerably more sensitive than
either of the bacteria or the polio-1.   Thus, although 0X174
may be an excellent virus for studying the kinetics of dis-
infection by C102, it is not a very good indicator for dis-
infection of CSO.  C1C>2 has shown very similar rates of inac-
tivation against almost all of the animal viruses studied:
Coxsackie B-3,  ECHO-7, HSV-L2, NDV, vaccinia and polio-1.
Viricidal activity against all of them stopped immediately
upon addition of thio.

The f2 phage was found to be a better indicator of viruses
than 0X174.  Figure 20 shows typical inactivation curves
obtained when an f2 stock was diluted approximately 1:10~16
in SCSO and treated with C102.  One can predict that 4 mg/1
C102 will inactivate about the same amount of f2 as polio-1
(1.3 logs).

In No-Demand Systems—C102 inactivation of virus in a no-
demand system serves as a useful baseline for studying the
kinetics of viral inactivation and for comparing C1O2 to
other disinfectants.  It has the virtue of complete repro-
ducibility and is the most practical way to divorce the
                              62

-------
U)
       Table 18.  Survival of Viruses in SCSO Treated With C102a
C102
Dosage
(mg/1)
0
0.2
0.5
1.0
2.0
4.0
8.0
HSV-L2
(PPU/ml)
2.8X105
2.6X105
1.4X105
2.2X105
11.2X104
2.2X103
0
NDV
Calif.
(PFU/ml)
5.4X10°
8.6X106
5.0X106
4.2X106
1.1X106
0
YFV
(PFU/ml)
4.0X105
2.4X105
1.6X105
8.2X104
7.2X10^
4.4X103
0
Vaccinia
(PFU/ml)
4.2X10^
1.6X102
2.4X102
32
0
0
ECHO-7
(PFU/ml)
4.2X10°
4. 8X106
5.5X106
4.8X106
1.16X106
1.21X106
1.04X106
Coxsackie
B-3
(PFU/ml)
1.0X10°
8.4X105
1.0X106
1.2X105
       ^^ one minute contact time

-------
CTi
              0   30  tO  K>  IZO  ISO  ISO  2W 24O 27O 3OO
                                                                             TIME, MCOOdt
         Figure  19.  Single-stage disinfection
                     of  several bacteria and
                     viruses with CIO2
Figure 20.  Single-stage
            disinfection  of
            f2 with CIO2

-------
direct effect of C102 on the virus from indirect effects via
the virus environment.

Polio-1 was purified by differential centrifugation and rate-
zonal centrifugation in sucrose in order to remove C102
demanding substances.  All no-demand disinfection experiments
were run in sterile 0.01 M phosphate buffer at pH 7.0.  The
resultant plots  (Figure 21) show that  for different concen-
trations of C1C>2, virus counts are an  exponential function
of time.  This suggests a pseudo first-order reaction as
would be expected from Chick's Law.  Considerably less C102
was required to achieve virus levels comparable to those in
SCSO.  Application of only 0.4 mg/1 ClC>2 for 300 seconds
inactivated 3.5 logs of polio-1.  When all these individual
treatments from several experiments were converted to
standard dose (mg/1 x time of exposure) of C102 and plotted
against log numbers of survivors (Figure 22) , a straight-line
function was observed.

It is quite clear from these experiments that CSO rapidly
removes C1C>2 and there is little or no residual activity left.
While there is a considerable difference in the sensitivity
of viruses to C1C>2, even the most resistant viruses succumb
to tenths of a mg/1 under no-demand conditions.  Furthermore,
the inactivation appears to be exponential with no apparent
drop in activity over long periods of  time.  In one experiment
under no-demand conditions, 0.2 mg/1 C102 continued to inac-
tivate 0X174 exponentially for 90 minutes before the experi-
ment was terminated.  In SCSO, however, activity of up to 25
mg/1 C102 rarely lasted more than 30 to 60 seconds even at
the highest concentrations.

Effect of pH—In order to compare effects of different pH
values it was necessary to run the experiments under no-
demand conditions.  Figure 23 is a detailed illustration of
pH effects on C1C>2 viricidal activity  against 0X174.  The pH
vs. PFU/ml of 0X174 survivors was plotted at three standard
doses of C102: 12 (0.4 mg/1 X 30 seconds), 24  (0.4 mg/1 X
60 seconds) and 72  (0.4 mg/1 X 180 seconds).  The active pH
range lies between 4.5 and 7.5.  The maximum antiviral
activity occurred at pH 4.5.  Similar  effects were observed
when the pH of SCSO was adjusted and seeded with 0X174.

The role of pH in Cl2 disinfection has been thoroughly
studied and is well documented47/48.   It is thought that HOCl
is the disinfecting agent rather than  the dissociated hypo-
chlorite ion (OC1~), or dissolved Cl2  gas.  HOCl is a weak
acid (K=2.95 X 10~8) so a lower pH will favor the undissociated
form and, therefore, promote increased disinfection.  This has
                              65

-------
CTi
                          TIME, seconds
                                                                         STANDARD DOSE. sec. « mg/l
          Figure  21. Single-stage disinfection
                     of  polio-l  with  CIO2 in
                      a  no-demand  system
Figure 22.  Standard dose
            curve for
            disinfection  of
            polio-l with CIO2

-------
                                                        SO  6O  »O  12O  ISO ISO  2IO  24Q 2 TO 3OO
Figure 23. Effect of pH on disinfection
           of  0X174 with  CI02
Figure 24. Effect of CI02 on
            cell-associated  polio-I

-------
been verified experimentally by a number of workers.

It requires four times as much Cl2 as C102 to inactivate
viruses in SCSO but this may be due to the fact that the pH
of SCSO is outside the optimum value of 2.5 for Cl2 disinfec-
tion.  This is one of the main reasons for employing C1C>2 as
a disinfectant.  It is a neutral compound and is effective
at neutral or slightly acidic pH.  It is also more slowly
absorbed by the particulate fraction of SCSO than is
Cell-Associated Viruses_ — In addition to free virions, water-
borne animal viruses may be transported either inside or on
the surface of cells which have been excreted from the
intestinal tract of man and animals.  The possibility that
under natural conditions these cell-associated viruses are
protected during C1O2 disinfection was tested by the following
experiment.

An infected cell inoculum was prepared which contained 4.9 X
106 PFU/ml of polio-1; 75 percent of this (3.7 X 106 PFU/ml)
was found to be cell-associated and 25 percent (1.2 X 10^
PFU/ml)  was free, extracellular polio-1.   Ten ml of the
inoculum was poured into 500 ml of SCSO,  passed through the
23 micron screen, treated with C102/ neutralized and titered
for cell-bound and extracellular virus.  Neither free or
bound virus was lost by this filtration.

Figure 24 shows the surviving cell-associated virus after
C102 treatment and chloroform extraction.  An increase in the
expected number of survivors was noted.  For example, 8 mg/1
C102 normally reduces free virus titers in SCSO by 2.2 logs
in 30 seconds: in Figure 24 the cell-associated virus was
only reduced by 1.15 logs.  Thus, the presence of the cellular
material clearly interferred with the effectiveness of C102
treatment.

Source of Chlorine Dioxide Demand — Under no-demand conditions,
C102 shows linear inactivation curves for polio-2 and 0X174.
Under high-demand conditions found in SCSO, the activity
ceases after 30-60 seconds.  The component' in SCSO which is
responsible for this neutralization of C102 is unknown.  An
effort was made to detect this neutralizing component.  A
sample of the highly contaminated first flush of a CSO was
treated in various ways to reduce demand: (1) autoclaving for
30 minutes at 121°C; (2)  filtration through 0.45 micron
membrane filters and; (3) centrifugation at 40,000 rpm for
30 minutes.  0X174 purified in sucrose gradients was then
added to the samples and controls.  Various concentrations of
C102 were added for 180 seconds and neutralized with thio.
The survivors are listed in Table 19.
                             68

-------
'-0
       Table 19.  Single-stage Disinfection With C102 in CSO Treated to Remove Demand5
Centrifuged 40
C102
(mg/1)
0.0
0.2
0.5
1.0
2.0
4. 0
6.0
Untreated
CSO
100
47
11.
5.
0.
0.

4
3
001
0005
Filtered
0.45U MF
100
62
32
0.62
0.002
0.008
,000 rpm, 30 min
15
psi
Pellet Resuspended 30 min.
Supernatant in Phosphate Buffer Autoclaved
100
45
14
0.
0.
0.


13
0001
0006
100
82
67
1.
0.
0.


3
012
007
100
46
13
0.
0.
0.


11
001
0005
Phage in
Phosphate
Buffer
100
33.
15.
0.
0.
0.
""
7
9
2
0015
00007

        a values in percent 0X174survivors

-------
These data indicate that the small particulate fraction of
CSO exerted a strong demand on C1C>2, especially at concen-
trations of C102 higher than 2 mg/1.  For example, at 4 mg/1
C102 the inhibitory activity was reduced in the supernatant
fraction but was not removed greatly by filtration.  The
resuspended pellet strongly interfered with the activity of
C102, again showing the particulate nature of the inhibitory
component.  Autoclaving destroyed a large amount of inhibitory
component, perhaps by precipitating particulates or reducing
their surface area.  It is; not known what particulates are in
CSO, but it is possible they are inorganic materials.  An
effort will be made in the future to collect them by high-
speed centrifugation after filtration, to make a chemical
analysis, and to examine them by electron microscopy.

Other Parameters—The preliminary results of the effect of
C102 on ATP were given in Table 13.  The majority of the ATP
study is discussed in a later section dealing with two-stage
disinfection.  The effects of C102 on pH,  TOC and NH3N were
found to be negligible in the dosages used in this study.

Mechanism of Disinfection—

In a manner analogous to Cl2, C102 is a much more powerful
disinfectant than its decomposition product C102~-  Therefore,
the same type of biphasic curve, with a rapid initial drop
followed by a gradual decrease, is observed.  In this case,
this gradual decrease is almost flat, reflecting the weak
bactericidal and viricidal properties of C102~.

The chemical mechanism(s)  involved in the disinfection action
of C102 in water is unclear.  Ridenour and Ingols61 reported
that C102 (1) reacts with the large colloidal molecules of
peptones following the laws of surface absorption, (2) does
not react with NH3N, urea, glucose, or alcohols, and  (3)
reacts with phenols to about the same extent as does Cl2 but
much more slowly.  They hypothesized that as a bactericide,
C102 has an advantage over Cl2 in that it is less reactive
to substances in the water.

As is the case with Cl2, the exact mechanism of C102 attack
on bacteria and viruses is unknown.  Bernarde58 showed that
C102 does not rupture the bacterial cell walls but this is
all that is known.  The same observations about lipid con-
centrations and disinfectant sensitivities that were made for
C12 can be made for C102.   The fact that on a molar basis
C102 is more effective than Cl2 implies that there is an
increased sensitivity of the lipids when C102 is used.  This
could also explain the observation that the spread in bacteri-
                             70

-------
cidal and viricidal properties for high and low lipid contain-
ing organisms is greater with C102 than Cl2.

Ejffects of Screening—

Since disinfection with C1C>2 was affected by screening while
disinfection with Cl2 was not affected, there must a differ-
ence in the nature of the disinfection reactions.  If the
same materials that made up the Cl2 demand also made up the
C102 demand, as has been previously hypothesized, identical
screening effects on disinfection should be observed.  To test
this hypothesis, a sample of SCSO was split into two portions
with the Cl2 and C1C>2 demands determined on each portion.
Within experimental error the values came out approximately
the same.  The sample for which the Cl2 demand had been
determined was now dechlorinated by titration with thio to
the starch-iodine endpoint.  The C1C>2 demand of this sample
was now found to be 80% of the C102 demand of the original
sample.  Thus, the two demands must be partially due to the
different materials.

It is surmised that the C1C>2 demand is due to dissolved
organics, which unlike the CDS, are not found encapsulated
within the solids.  Thus, screening does little to the C102
demand but only exposes more bacteria to the disinfection
process and thus promotes C102 disinfection.  As further
proof, settling actually reduces the Cl2 demand but leaves
the C102 demand unchanged, as shown in Table 20.

Table 20. Effect of Solids Removal on Disinfection with
                     C12 and CIO?
Sample
SCSO
scsoa
SCSO-settled
Cl2 Demand
6.5
0.0
3.5
C102 Demand
6.0
5.0
5.2
 a carried to breakpoint and dechlorinated

Temperature Effects

In addition to the climatic reasons for studying the effect
of temperature on disinfection, previous work on this sub-
ject has shown definite temperature effects62'63

Bacteriology--

The bacterial results of this study are given in graphical
form in Figure 25 and in tabular form in Appendix B.  There
                              71

-------
NJ
                                       SAMPLE INFORMATION
                                        BLENDED
                                        UNSCREENED
                                        PH     6.89
                                        Clt DEMAND 8OO
                             TIME, seconds
   12O ISO I8O 210 24O 2TO
     TIME, seconds
                       TC  VS  CI2
FS  vs
            Figure  25.  Effect  of temperature  on  disinfection  of bacteria

-------
                                              SAMPLE  INFORMATION'
                                                BLENDED
                                                UNSCREENED
                                                PH      &69
                                                CIOJ DEMAND B.OO
-0
OJ
                       I   I   I   I   I   I   I
                      SAMPLE INFORMATION*
                       BLENDED
                       UNSCREENED
                       PH       cas
                       CIO, DEMAND  *OO
                               120 I9O IflO 21O 2«O  2TO  SOO
                                  TIME, second*
      120 190 180 210 240 2TO SOO
         TIME.itconds
                         TC   vs.  CI02
FS  vs.  CI02
              Figure   25.   Continued

-------
may be a slight increase in efficiency with increasing temp-
erature but the effects of time,  dose and sample composition
are much greater.   Previous studies of no-demand systems and
drinking water supplies show a much greater enhancement of
disinfection at higher temperatures with both C±2   and
C10263.  In SCSO these effects were still measurable, but
notably decreased.   The much shorter contact times used in
this study and competition from disinfectant demands should
have this effect.

Virology--

The results of the  temperature variations on viral inactiva-
tion are presented  in Table 21.   A 40 percent difference in
activity was observed between the 60 second and 300 second
exposures to C102 at 4°C;  no difference was seen at the higher
temperatures.   This indicates that although the initial rate
of disinfection may be reduced at colder temperatures, the
total disinfection  achieved is the same.

Table 21. Effect of Temperature on Disinfection of Polio-1
	with C102a	

                              Percent Reduction of Polio-1
Temperature (°C)
4
12
22
35
60 seconds
39.5
63.7
66.1
72.8
300 seconds
82.5
68. 6
78
78.8
-a 4 mg/1

A more detailed study was conducted with 0X174.  Figure 26
is a graph of phage survivors vs.  temperature at a constant
dose of 4 mg/1 C102 .   Although the points show considerable
scatter, a trend line fitted by the method of least squares
supports the above observations with polio-1 that there is
only a slight increase in C102 activity with increasing
temperature.  The magnitude of the temperature effect is
seen from the fact that there was  less than one log differ-
ence in the number of phage survivors treated at 0°C and 45°C.

The most profound temperature effect was seen with the inacti-
vation of f2 under no-demand conditions (Figure 27) .  At a
concentration of 2 mg/1 C102 for 180 seconds, an apparently
exponential increase  in activity of C102 was observed.  From
an initial concentration of 1 X 107 PFU/ml, the virus titer
dropped to 3.9 X 10^  PFU/ml at 45°C, but at 0°C, 1.6 X 105
PFU/ml were recovered.  Although this experiment has been
performed only once,  it suggests that the reaction mechanism
                             74

-------
             1C* I
Ul
                                                               4...
«*fl/l CI02

FOR flO SECONDS
                                                                         2i*ig/l CI02
                                                                         FOR 180 SECONDS
                   10   20    JO    *0
                    TEMPERATURE. -C
                                                                             TEMPERATURE. 'C
        Figure  26. Effect  of temperature  on
                    disinfection of  0X174  in
                    SCSO
                                           Figure  27.  Effect of  temperature
                                                       on disinfection  of ^
                                                       in  a  no-demand
                                                       system

-------
or the sensitive sites of the virus are markedly different.
Recent experiments have shown that under the same no-demand
conditions, temperature and exposure time, only  0.8 mg/1  C102
are required to inactivate the same number of infective  units
of polio-1 that 2 mg/1 C102 does with f2 phage.  Under demand
conditions, the competition between the organic background
and the virus is so great that the temperature  effects are
largely unseen except at 4°C  (Table 21).

Comparison of Chlorine and Chlorine Dioxide

Bacteriology--

Disinfection, as measured by TC kills, reached  acceptable
levels 50 percent of the time with 8 mg/1 C1C>2  and 25 mg/1
Cl2-  There is no comparative information in the literature
for CSO.  Generally, it has been shown in previous
studies58'60 that C102 is a better bactericidal agent than
Cl2 in no-demand systems and drinking water.

Virology--

The relative inactivation strength of the Cl2 and C1C>2 toward
virus is illustrated more or less directly in Figure 28.  A
dosage of 4 mg/1 C1C>2 inactivated as much f2 phage as 20 mg/1
Cl2, suggesting that on an equimolor basis the C1C>2 is about
ten times as active a viricide as Cl2 when they are diluted
with water which has both a Cl2 demand and a C1C>2 demand

In these studies an attempt was made to discover whether it
is feasible to substitute C102 for Cl2 as a viricide and to
determine the best method of utilizing C102 as a viricide in
the treatment of CSO.  On the basis of these data it appears
that C102 is a satisfactory alternative to Cl2 for disinfec-
tion of viruses.  The C102 activity is relatively constant
over a wide range of changes in pH, temperature and ionic
requirements.   It inactivates viruses very rapidly and has a
wide spectrum of viricidal activity.   No truly resistant
virus has been found to date and no resistant phage mutants
have been observed.   The C102 demand of heavily contaminated
water is slightly less than the Cl2 demand.  This lengthens
the working life of the disinfectant and reduces the dosage
needed.  Although the degradation products of C102 in CSO
and other crude waters are not known, there is no evidence
that appreciable quantities of chloramines are found.

Discussion--

From these studies it would appear that C102 is a better dis-
                             76

-------
   I06
   105
c:  io4
Q.

C\J
   10'
                                       I
                                            I
          30    6O   90    120    150    180   210


                              TIME, seconds
                                                 240
270
300
Figure  28. Comparison  of disinfection of  ^2  w'th  CI02

             and  CU
                                 77

-------
infectant than free C3-2 whenever disinfection is to be carried
out by these two species, i.e., in pure cultures, drinking
water or high-rate disinfection of sewage.  In contaminated
waters with a significant disinfectant demand, the Cl2 and
C102 are rapidly converted to combined Cl2 and C1O2~, respec-
tively.  Thus for long-term disinfection in these waters, the
comparison is between these latter two species.  Since C102~
has less disinfection ability than most forms of combined C12,
chlorine is the choice under these long-term conditions.

Economic considerations of the operating costs alone indicate
that Cl2 is the cheaper of the two, even employing the higher
doses.  Cl2 as the gas or as sodium hypochlorite costs much
less than C1C>2 on a per pound basis.   Even with the higher
dosages necessary in this study, disinfecting with Cl2 would
be less expensive.  This subject will be dealt with in
greater detail in a later section of this report.

The effects of the decomposition products on the receiving
waters must also be compared.  Cl2, either free or combined,
is known to be toxic to many species of fresh-water fish at
the microgram per liter (ug/1)  level^^'"^.  However, when
compared to the dangers to human health of untreated CS065
and given the relatively short half-life of Cl2 in most
receiving waters, the risk of fish kill may be the best alter-
native.  It may not be feasible to dechlorinate the CSO prior
to their discharge since the outfall pipeline often serves as
part of the contact chamber..  Little is known of the toxic
effects of C1C>2~ on marine life, although low concentrations
(less than 1 mg/1 C102~) have caused difficulties in some
studies"".

The proper weighing of these three factors, disinfection
efficiency, cost and toxicity of receiving waters, requires
careful evaluation than can better be made after the full-
scale facilities have been operated.

TWO-STAGE DISINFECTION

The two-stage disinfection study consisted first of sequential
addition of two doses of C102 and then a single dose of Cl2
followed by a dose of C102.   Reductions in ATP were measured
in parallel with bacterial reductions to determine the validity
of using ATP as an indication of disinfection.

Sequential Addition of Chlorine Dioxide

Bacteriology--

The sequential addition of more than one dose of Cl2 is known
                              78

-------
to have the same effect as a single dose equaling the cumula-
tive amount of Cl267.  Therefore, two-stage addition of C1C>2
was studied to determine if C1C>2 behaved similarly to Cl2.
The bacterial results are presented in Figures 29 and 30,
with the corresponding tabulated data in Appendix D.  For the
present, the discussion of ATP is deferred but will be taken
up in detail later in this section.  Ideally, there should
be a stepwise reduction in bacteria at the times corresponding
to the addition of C102.  However, the plots tend to be
smooth curves that are quite similar to the single-stage
curves.  A possible ''explanation for this is that the intervals
between dosages, either 15 or 30 seconds, are short enough
that complete mixing of the first dosage cannot be completed
by the time the second dosage is added.  Stepwise reduction
should be observed if longer intervals between dosages were
used, but in high-rate disinfection, this is not feasible.

The comparison of the results in Figures 29 and 30 with the
single-stage results in Figures 15 and 16 show the same
reductions with either one- or two-stage addition of equal
amounts of disinfectants.  This observation is consistent
with the proposed mechanism for C1C>2 disinfection and indicates
that the rate of disinfection is a linear function of dosage.

Virology--

Figure 31 compares the inactivation of 0X174 by addition of
five mg/1 C102 to the addition of five sequential one mg/1
dosages at 30-second intervals.  It is apparent that much of
the effective activity of the large single dosages is lost
before it can be mixed completely.  It should be noted that
the mixing for the single dose was less vigorous than for
the other additions of disinfectants in this and other experi-
ments.  This further illustrates the need for thorough mixing.
It is likely that in the small region of disinfectant contact
there was a rapid, complete viral kill that resulted in the
steep initial drop in Curve 1 of Figure 31.  However, there
also may have been enough C102 demanding substances to con-
vert all the C102 to C1C>2~.  The experiment was not repeated
with a more rapid mixing of a single, large dose of C102-
Multiple small doses with thorough mixing may be a more
economical way to treat the water.

Although the plaque counts at 150, 180 and 210 seconds are
only approximations due to the small number of surviving
viruses, it appears that there was a second factor which
provided some protection for the virus.  Each additional dose
was less effective than the preceding one.  A variety of
explanations are possible: clumping of viruses, genetic rescue
                              79

-------
    10'
       2mg CIO at0 seconds
        I    z
       2m«
Znvg CI OjatO seconds
2tng CI02" 15  "
2mg CIO at 0 seconds
2 gig CIQg ' IS
2 mg CIO at 0 second*
2rnq CI02 '= 15
         3O  6O  90   120
          TIME, seconds
  30   60   90   120
   TIME, seconds
  30   60  90   120
   TIME,seconds
  30  60  90  120
   TIME,seconds
                                                 TOTAL  COLIFORM
                                                 FECAL  COLIFORM
                                            -•--  FECAL  STREP
Figure  29.  Two-stage  disinfection  of  bacteria  and  ATP
                with  CI02  (I)
                                        80

-------
     4 tng/l ClOg at 0 seconds    4 mg/l CI02 at 0 saconds   4 mg/l CI02 at 0 seconds   4 mg/l ClOg at 0 seconds
     0   30   60   90  120
         TIME, seconds
0   30  60  90  120
   TIME, second
0  30  60  90  120
   TIME, seconds
0  30  60  90  120
   TIME,seconds

TOTAL COLIFORM
FECAL COLIFORM
FECAL STREP
Figure   30.   Two-stage   disinfection  of  bacteria   and   ATP
                with  CI02 (II)
                                        81

-------
                          ARROW INDICATES ADDITION OF I mg/l CI02
                                    *.
                                        KEY
                                      CI02 DOSAGE
                                   I  mg/l SEQUENTIAL ADDITION
                                   5 mg/l SINGLE ADDITION
             30    60    90    120    150    180    210    240    27O    300
Figure  31,  Addition  of  multiple  doses of  CI02  to  0X174
                                 82

-------
as a result of interaction of two disabled particles,
membrane barrier effect, reinfection by viruses  in a pro-
tected corner of the vessel, etc.  The most likely explanation
seems to be that the curve represents the progressively  less
efficient mixing in which the same proportion of viruses is
out of reach of the C1C>2 each time disinfectant  is added.

Chemical Parameters—

Two-stage addition of any combination of disinfectants had
no effect on any of the chemical parameters.  All of this work
was conducted on screened samples.

Chlorine Followed by Chlorine Dioxide

Bacteriology—

The second part of the two-stage study was the sequential
addition of different disinfectants.  This could enhance dis-
infection in two ways.  The first disinfectant could pre-
condition the waste so that the second disinfectant could
work more efficiently.  This possibility has already been
suggested when Cl2 is used as the first agent, followed by
C1C>2^.  The second possibility is that interactions between
the two disinfectants could lead to increased efficiency by
either or both agents.

The bacterial reductions are given in Figures 32 and 33, with
tabulated data in Appendix D for Cl2 followed by C1C>2.  Varying
the interval between dosages from 15 to 30 seconds did not
affect the results, presumably an effect of incomplete mixing
as explained previously-  The three test bacteria groups
(TC, FC, FS) behaved in a similar manner.  Although the TC
levels were reduced to an acceptable value, as was also found
in single-stage studies, some of the initial values were one
log less than in the single-stage studies.  Therefore, the
final values tend to be somewhat misleading and  the incremental
log reductions at corresponding times should be  compared.

In order to determine if disinfection has been enhanced beyond
the additive effects of the different agents, the log reduc-
tions must be compared.  Table 22 includes the log reductions
in TC accomplished by 4 mg/1 and 8 mg/1 Cl2, and 2 mg/1 and
4 mg/1 C1O2 as is determined in the single-stage study from
Figures 11 and 15, respectively.  The sums of the reductions
are compared to the observed log reductions in the two-stage
studies for the appropriate dosages for TC bacteria in
Table 23.
                               83

-------
     I07
       4 mg/l CI2 ot 0 Mcondi   4mg/l Clg ot 0 Mcoodi
       2   "  ClOg " 15   "     2  "  ClOg " 30 "
                4mg/ICIg ot 0 taconds  4 mg/l Clg at 0 seconds
                4  "  CI02 "15   "     4  "  CI02 " 30  "
   o
   X
   g I04
     I03
                                                                         j	I
          30   60   80
          TIME, seconds
SO  SO  90  120
TIME, seconds
SO   60   90
TIME, seconds
                                                             120
JO   60   90  120
TIME, seconds
                                                               TOTAL COLIFORM
                                                               FECAL COLIFORM
Figure   32.  Two-stage  disinfection  of  bacteria  and  ATP  with
          CI2   and  CI02  (I)
                                       84

-------
      8mg/l CI2 at 0 seconds
      2 "  CI02" 15  "
8 mg/l CI2 ol 0 seconds
2  "  CI02 " 30  "
Bmg/l Clg at 0 seconds
4  "  ClO" 15   "
                 8 mg/l CI2 at 0 seconds
                 4  "  CI02" 30  "
         3O  60  90
         TIME, seconds
   30  6O  9O
   TIME, seconds
                                        120
 30   60  90
TIME, seconds
                   0   30  6O  90  120
                      TIME, seconds
                   TOTAL COLIFORM
                   FECAL COLIFORM
                   FECAL STREP
Figure   33.  Two-stage  disinfection  of  bacteria and  ATP
                with  C\2  and   CI02  (H)
                                       85

-------
Table 22. Average Bacterial Reductions for Single-stage
          Disinfection
Disinfectant
C12
C12
C102
C102
Dosage
(mg/1)
4
8
2
4
Time
(Seconds)
120
120
90-105
90-195
Log Reduction
0.5
1.0
1.0
1.0
Table 23. Average Bacterial Reductions for Two-Stage
	Disinfection	

                                            Log Reduction
   First     Dosage    Second    Dosage    at 120 Seconds
Disinfectant (mg/1) Disinfectant (mg/1)  Predicted  Observed
ci2
Cl2
C12
Cl2
4
4
8
8
C102
C102
C102
C102
2
4
2
4
1.5
2.0
2.0
2.5
1.5
1.5
3.0
4.0
At low reductions there is no significant difference between
observed and predicted values, most likely because of diffi-
culties in measuring small differences in large numbers.  As
the reliability in reported differences increases, the
enhancement of disinfection becomes more easily discernible.
Because of the variations in disinfection in different  samples
of SCSO observed in the single-stage disinfection, these
results can only be considered as indicators of a two-stage
effect.  Many such observations must be made before any
definitive statements can be issued.

Virology—

Samples of SCSO containing 0X174 were treated with 8 mg/1 C12
for 120 seconds and then 2 mg/1 C102 was added.  As shown in
Figure 34, both disinfectants were tested separately and in
combination as controls.  Eight mg/1 C12 and 2 mg/1 C1O2
both reduced the titer of 0X174 by 2.2 logs in 300 seconds.
When both were added simultaneously, the virus titer was
reduced by 4.4 logs in 300 seconds.   However, when C12  was
added first and then C102, the virus titer was reduced  by
5.3 logs.   Thus, there would appear to be an advantage  to
sequential as  opposed to simultaneous addition of disinfec-
tants.   This work was confirmed repeatedly and the experiment
was enlarged as shown in Figure 35.   C102 and C12 were  added
                             86

-------
              10*
                                               KEY      INTERVAL

                                            SYMBOL °°%ffi'   wcoods

                                              0  8CI2:2CI02  130


                                             -*—  2CI&2    -
                                             -*-  8 CI2: JCIOj  0
                                120   ISO   180

                                 TIME, seconds
                                                    270
Figure  34. Comparison  of single
             disinfection  of 0X174
-stage and  two-stage

 with  CI2 and  CI02
                               87

-------
       I09
       10'
       10s-
    U.
    QL
       I03
       10'
       10
               12  REPLICATES
               UNTREATED
          Cl2 8
          CI02 2 mg/l

         INTERVAL 30 seconds
       CONTACT TIME 300 seconds
Tt
                                   REPLI-
                                   CATES
                                          -TTh
        REPLI-
        CATES
                TT
REPLI-
CATES
REPLI-
CATES
                                  I. CI2'.CI2  CI2'-
                                       2-STEP TREATMENT
Figure  35.  Repeatability  of  disinfection of 0X174 with
              and  CI02
                                 88

-------
in dosages sufficient to provide identical kills of 0X174
(8.0 mg/1 Cl2, 2.0 mg/1 C102) when added to SCSO.  The dis-
infectants were added sequentially at 30-second intervals to
seeded SCSO and neutralized with thio for subsequent virus
enumeration.  Four combinations were studied: Cl2 followed by
C102, C102 followed by Cl2/ C102 followed by C102, and Cl2
followed by Cl2-  Despite the fact that single-stage addition
of Cl2 and C102 gave identical amounts of inactivation, two-
stage addition of Cl2 was less effective than two-stage addi-
tion of C102.   A possible advantage of Cl2 followed by C1C>2
as compared to C102 followed by Cl2 may be seen in Figure 35.

Mechanism of Disinfection

A possible interpretation of these data suggests that the
bactericidal and viricidal properties of both Cl2 and C1C>2
may depend on the length of a free residual.  If Cl2 is more
rapidly neutralized or absorbed than C102, it would be more
difficult to maintain residual free Cl2 by multiple dosage
techniques.  The two-step inactivations in Figure 36 show
that the rate of inactivation for the second viricide is very
much reduced,  regardless of whether it is the same as the
primary reagent.  The second dose continues to kill for a
longer period of time, 60-90 seconds, than the primary dose
for both Cl2 and C1C>2 •  This suggests that reactive sites or
sensitive particles become exposed in the resistant population.
If the free residual is maintained, these exposed particles
are killed.

As indicated in Table 20 the Cl2 and C102 demands are due to
an interaction between the two disinfectants.  The interaction
could take place through the following mechanism.  One of the
reactions for the preparation of C102 involved the addition
of Cl2 to a solution of C102~-7

     H+ + HOC1 + 2NaC102	— 2C102 + NaCl + Na+ + H20

It can be hypothesized that after the C102 has been oxidized
to C1O2~ in SCSO, any free Cl2 also present might reduce C102~
back to C102-   This process would prolong the existence of the
more potent disinfectant, C1O2/ and thus enhance disinfection
beyond that expected by the sum of the respective concentra-
tions of Cl2 and C1O2.  It might be argued that the reduction
in free Cl2 would compensate for the extra C102 but it has
been shown that C102 is a more powerful disinfectant when
taken on a weight basis.

If combined Cl2 is also capable of reducing C102~ to C102, the
process would be a cyclic one continuing until all the Cl2,
                              89

-------
       10* -
KEY
SYMBOL DOSAGE
mg/l
— X — BC\2- 2 CI02
—A— 2 ClOj: 8 Cl2
— 0- 2CI02' 2 CI02
B 8 CI2'.8CI2
INTERVAL
seconds

30
30
30
30
               3O   60    9O    120   150   180   2IO
                               TIME, seconds
                                                24O
                                                     27O
                                                          3OO
Figure  36. Two-stage  disinfection  of  0X174  with
              and   CIC>2
                             90

-------
free or combined, was converted to chloride  (Cl~).  However,
it is not expected that the combined Cl2 has sufficient
reducing power to accomplish this reaction.

If this hypothesis is correct, the free Cl2 will undergo
competitive reactions between the C102~ and the Cl2 demand.
Thus the greater the amount of Cl2/ the greater the expected
enhancement.  Although this was observed, the numbers of
combinations were not sufficiently great to prove the issue.
The enhancement effect would be increased by any means that
would increase the free Cl2 concentration.  In the bacterial
studies, Cl2 was added 15 to 30 seconds prior to the addition
of C1C>2/ allowing the Cl2 demand to be exerted for 30 seconds,
thus decreasing the free Cl2 concentration.  This minimized
the regeneration effect, but some regeneration should have
taken place since it is known that the free Cl2 requires
longer than 30 seconds to disappear.

The virus studies do not verify this theory.  The simultaneous
addition of Cl2 and C1C>2 should have given greater enhancement
than delayed addition of C102-  However, the result was only
an additive effect of both dosages.  It is suggested that esr
be used to detect an increase in C102 concentration after the
addition of Cl2 as an absolute verification of the concept.
Little clarification of the issue would result from studies
of bacteria in no-demand systems.  Any free Cl2 added to such
a system would remain as free Cl2-  Also the C102 would not
be converted to C102~ so regeneration is impossible.  If the
regeneration of C102 cannot be substantiated, the explanation
of enhanced disinfection must lie in some other interaction
of Cl2 and C102«  Whatever the reason, this two-stage effect
should most certainly be studied in greater detail in future
bench- and full-scale operations.

ATP CORRELATIONS WITH BACTERIA

The measurement of ATP was added to the two-stage disinfection
studies to determine if this parameter would yield the same
bactericidal information as was obtained through direct bac-
terial measurements and, therefore, yield a reliable control
signal for disinfection.  Examination of Figures 29, 30, 32,
and 33 shows that for all trials in which the TC count was
greater than 1000 per 100 ml after two minutes contact time,
the ATP value was greater than one arbitrary unit.  Identical
ATP standards were used to calibrate the instrument for each
trial, so the arbitrary units were the same for all trials.

Similar observations can be made using 200 counts/100 ml for
FC and FS.  Table 24 gives the correlation between ATP and
TC and FC.  The large fluctuations in initial ATP values led
                              91

-------
to the conclusion  that better correlations might be obtained
with lower values.  As would be expected, the correlations
are better for disinfected samples in which the bacterial
counts and ATP concentration are relatively low.  This is
most likely due to the fact that TC were chosen as indicator
bacteria because they are relatively resistant to disinfection.
Therefore, in disinfected samples, the TC comprise a higher
percentage of the total bacterial population than in non-dis-
infected samples.  At ATP levels less than one arbitrary unit,
the lower correlations most likely result from the lack of
precision in low bacterial counts.  However, for all ATP less
than 0.5 unit,  the TC were, with one exception, less than
1000 counts/100 ml.  Thus, some ATP level can be selected as
an indicator of successful disinfection.

Table 24. ATP Correlation Coefficients
Range of ATP (Arbitrary Units)
0-50
15-50
0-15
0-1
TC
0.700
0.206
0.842
0.387
FC
0.759
0.515
0.794
0.346
The coefficients were calculated from the points in Figures
29, 30, 32 and 33.   It is recognized that this data is only
a small fraction of the amount of work that must be performed
in order to reach any definitive conclusions about ATP and
bacteria correlations.  However, there would appear to be
sufficient justification to study the possible use of ATP
as a monitor to control disinfection.

Application as a Control for Disinfection

The most common method of controlling disinfectant (usually
Cl2)  dosages is the monitoring of residual disinfectant to
maintain a fixed level in the treated effluent, a method that
admittedly has its  limitations68.  Other methods include
addition based only on flows and, in larger plants, periodic
bacterial counts, but only as feedback information.  These
methods may be acceptable in sewage treatment plants in which
the microbial variations are somewhat predictable, but totally
unacceptable for CSO treatment facilities in which the varia-
tions are large and unpredictable.  The objection is that
simply maintaining  a fixed volumetric residual does not
guarantee that the  bacteria and virus have been reduced to a
safe level.  ATP is a more direct reflection of microbial
activity than residual disinfectants.
                              92

-------
Before the development and testing of an in-line ATP monitor,
there are several difficulties that must be overcome.  These
include the stability of reagents, adequate sampling tech-
nique, availability of the necessary electronics and the auto-
mation of the chemistry.  However, the feasibility of such a
system is an accepted fact69.  A continuous flow monitor is
envisioned with periodic switching from CSO to ATP for stan-
dardization.  The control signal for addition of disinfectants
could be generated from an ATP monitor located after disin-
fection.  This arrangement would require that the ATP con-
centration be kept below some pre-determined safe level.
Another possibility would be to measure ATP before and after
disinfection to detect sudden increases in degree of conta-
mination as well as residual ATP levels.  The electronics
necessary to continuously record the light intensity from
the bioluminescent reaction have already been developed^.
The procedures for a continuous, automated extraction are
already in use in several of the "Technicon Methodologies"^8.
The expense of the reagents is likely to be quite significant
because of the high cost of removing the necessary enzymes
from actual firefly lanterns.  The cost factor should most
definitely be studied in greater detail before attempting to
develop a monitor.

Stability Tests

The stability or storage life of the reagents must be at
least one week to realistically consider monitoring.  Accord-
ingly, the stability of an ATP standard was tested under a
variety of conditions; namely, in the dark, under refrigera-
tion and with buffers added.

Eight sets of ATP standards were prepared as follows:

Cuvette Number (see Table 25)
     1.  ATP = 1.0 X 1CT1 ug/ml; EDTA=0; MgS04=0
     2.  ATP = 1.0 X 1CT1 ug/ml; EDTA=1Q-2M; MgS04=0
     3.  ATP = 1.0 X 10"1 ug/ml; EDTA=10~3M; MgSC>4=0: recommended
         by DuPont32 for added stability
     4.  ATP = 1.0 X 10"1 ug/ml; EDTA=10-4M; MgS04=0
     5.  ATP = 1.0 X 10"1 ug/ml; EDTA=0; MgS04 = l()-4M: not pre-
         pared due to difficulty in adequately dissolving MgS04
     6.  ATP = 1.0 X 10"1 ug/ml: EDTA=0; MgS04=10~2M:
         recommended by DuPont32 for added stability
     7.  ATP = 1.0 X 10"1 ug/ml; EDTA=0; MgS04=10~3M
     8.  ATP = 1.0 X 10"1 ug/ml; EDTA=10~3M; MgS04=10-2M

The Biometer was calibrated each day with fresh ATP prepared
within one hour of use.  Frozen (1°C) and unfrozen  (ambient,
                               93

-------
       Table  25. ATP Stability Experimenta
VD
Cuvette
Number
1
2
3
4
6
7
8
Cuvette
Number
1
2
3
4
6
7
8
Day 0
Frozen
1 .
1 .
n.
n.
1 .
1 .
0.

OOX10~X
94X10"3
71X10"1
81X10"!
44X10-1
54X10"!
96X10"!

Standingn
1.
1.
0.
0.
1.
1.
0.
Day
Frozen
0.
1 .
n.
n.
1 .
1 .
0.
89X10"-1-
07X1Q-3
82X1Q-1
88X10"1
27X10"!
53X10"1
73X10-1







00X1 (J^-
94X10"3
71X10-1
81X10-1
44X10"1
54X1Q-1
96X10"1
3
Day 1
Frozen
0.95X10 -1
0.76X10-3
0.89X10-1
0.90X1Q-1
1.00X10"1
1. 16X10-1
0. 83X1Q-1



Standing
0.
0.
0.
0.
0.
1.
0.

Standing
0.
1.
0.
0.
0.
0.
0.
98X1Q--1-
20X10~3
19X1Q-!
88X1Q-!
87X10"1
97X10-1
78X10-!







87X1Q--1-
87X10~3
75X1Q-!
74X10-!
95X10"1
02X10-1
80X10"!


Day 2
Frozen
0

8
1
0
1
0

Frozen
0.
0.
0.
0.
0.
0.
0.
95X10
97X10
73X10
97X10
93X10
91X10
65X10
.85X10"1
b
.02X10"!
.09X10"1
.98X10-1
.68X10-1
.76X10-1
Day 4


-3
-1
-1
-1
-1
-1


Standing
0.
0.
0.
0.
0.
0.
0.

88X1Q--1-
97X10~3
77X10-1
92X10"!
80X10"!
83X10-1
81X10-1

Standing
0.

0.
0.
0.
0.
0.
79X10"^
c
66X10"1
79X10"1
81X10"!
81X10"!
73X10-1
       a ATP values expressed in ug/ml
       b inaccuracies of maximum number of injections  (6) show values vary by a factor of
         100 or greater
       c Sample cuvette destroyed in storage

-------
22°C)  samples were both kept in the darkness until  use.   The
frozen samples were thawed in ambient temperatures  and  dark-
ness  45 minutes prior to use.  All eight  sets of ATP stan-
dards had respective frozen and unfrozen test samples.

The sample ATP standards were measured in  accordance with
instructions in Luminescence Biometer Instruction Manual-^ .
All values (Table 25) are the average of two consecutive
readings of + 0.15, or that of a maximum of six  readings  of
the same exponential factor which did not  fall within such
a range.

Cuvettes containing sample ATP standards #1 and  #8  seemed to
show markedly greater stability under both frozen and ambient
temperature conditions.  Sample #1 contained no  buffers,
while sample #8 contained 10~3 EDTA and 10~2M MgSC-4.  Both
values were recommended as optimal for stability in the DuPont
Instrument Manual.

The stability of the luciferin-luciferase  reagent was investi-
gated in a similar manner.  It was found that liquid reagent
could not be kept for more than two days under any  conditions,
as shown in Table 26.

Table 26. Enzyme Stability Experiment	

                            ATP Activity (ug/ml  X 10"1)
Storage of Reagent	Day 0  Day 1  Day 2  Day 3  Day  4
Fresh
Dark,
Dark,
Dark,
Light,
(Control)
Frozen
Refrigerated
Ambient
Ambient
0.
0.
0.
0.
0.
93
93
93
93
93
0.
0.
0.
0.
0.
89
86
61
68
80
0
0
0
0
0
.93
.71
.60
.74
.64
0.
0.
0.
0.
0.
85
84
14
31
03
0.
0.
0.
0.
<0.
85
98
07
14
01
Although this is a negative finding ,  the enzyme reagent  as
prepared in this study was in its purified dissolved  form.
The crude extract which should be suitable for an  automated
procedure may be more stable.  The technology required  for
an automated ATP monitor has already been demonstrated.   The
primary concern at this point is the correlation of ATP and
bacterial reductions on a wide scale.

IMPACT STUDIES

Aftergrowth

Bacteriology—

The results of the tank tests to determine bacterial  aftergrowth
are presented in Figures 37 and 38 and
                               95

-------
                                                         5
                                                         cc
                                                         o
                                                                              23 MICRON SCREEN




                                                                             — 23 mg/l CI2 fof


                                                                             — 25  "  ' ' 6O
                 TIME, noun
                                                                       TIME, houri
Figure  37.  Aftergrowth  of  TC  in  receiving  water

-------
                               UNSCREENED
                             12 mg/l CIO 2 to 30 nco™«
                             12 "  "  •  60  •
                       TIME. hour«
Figure   37.  Continued
                                                                                          23 MICRON SCREEN
                                                                                              12 mg/l CIO2 for 30
                                                                                              12 "   •   '60

-------
00
                                  Z5 mg/l CI2 for  30
                                  25mg/ICI2 tor  60
                          UNSCREENED
                               TIME, hourt
                                                                                   10001-
                                                                                                       25 mg/l CI2 'or 30 ucondi
                                                                                                      • 25 mg/ I Clj for GO Mcondt
                                                                                                 MICRON SCREEN
                                                                                                24    56
                                                                                                                  «0     Tt
                                                                                                    TIME, hourt
             Figure  38.  Aftergrowth  of  FS   in  receiving  water

-------
                           12 m0/l OOj for 30 aank

                           12 mg/l ClOj lor SO Mcandl
                       TIME. hour.
Figure   38.   Continued
                                                                          i
                                                                          i
                                                                          o
       12 mg/l CIO2 'or JO

       12 mo/C ClOj 'or 6O
                                                                                      23 MICRON SCREEN
I     M
 TIME, hour.

-------
Appendix E.  The initial bacterial counts at time 0 in
Figures 37 and 38 were calculated by using the dilution
effects of the lake and averaging the appropriate lake and
creek  counts.  All values in Appendix E at time 0 in the
lake and creek were obtained by calculations.  FC were run
in the first trial but were dropped from the remaining three
trials because all samples had zero counts.

Figure 37 shows a rapid die-off of most of the TC that enter
the lake.  Some regrowth is observed but the counts never
reach the levels initially present in the lake control.  The
die-off is hard to explain since the disinfectant residuals
were below 10 ug/1, the minimum detectable level, and every
attempt was made to promote bacterial life.  The fact that
similar die-offs were observed for the controls raised
serious doubts as to the reliability of the studies.

One cannot determine if the die-off in the controls occurs in
the lake because of the continuous input from the treatment
plant.  In view of the dilution factor of greater than 100:1
and the lake residence time of 50-200 days, the numbers
in the lake outlet (100 to 200 counts TC/100 ml)  imply that
the lake supports bacterial life^.  However, this claim could
not be verified experimentally.

The FS showed a similar die-off, but no aftergrowth of these
organisms was noted.   Aside from giving different initial
lake values, there appeared to be little effect on aftergrowth
from screening, dosage, or type of disinfectant.

Virology--

The creek and lake water used in these studies were assayed
for background viruses at the same dilutions as in the dis-
infected samples, and none were detected.  A dose of 25 mg/1
Cl2 effectively removed all of the virus seed in the SCSO
(7.7 logs of 0X174 and 3.7 logs polio-1),  as did 12 mg/1 C102
(8 logs of 0X174 and 4.6 logs polio-1).   If numbers 4 and 5
of Table 27 (see Figure 39) are compared with sample numbers
4 and 5 of Table 28,  it is clear that the C1C>2 inactivated
most of the virus in the first 60 seconds,  whereas Cl2
inactivation took longer.  Twenty minutes after the addition
of creek water, samples 8 and 9 of Table 27, Cl2 had inacti-
vated all of the virus added, indicating that Cl2 residual
probably continued to have some viricidal action even after
dilution with creek water.  No statement can be made concerning
the C102 residual since very few viruses survived the first
60 seconds of treatment (sample numbers 5 and 6 of Table 28).
                              100

-------
Unfortunately, the experiment required a large initial volume
of SCSO and high concentrations of polio-1 could not be at-
tained.  It is possible that the 1:1 dilution performed after
treatment (i.e., dilution upon entrance into the stream - see
sample numbers 4, 5, 8 and 9 of Table 27) may have reduced
the polio-1 below detectable levels.

This illustrates one of the major problems in bench-scale
aftergrowth studies.  If the low initial virus level has been
diluted by addition to the streams or lakes  (1:28 total dilu-
tion in the case of Onondaga Lake), the question remains
whether the virus had been efficiently inactivated by the
disinfectant, or the dilution has simply put the virus below
conventional detection techniques.

Even at 25 mg/1 Cl2/ some virus may have reached the lake,
as indicated by the small titers of 0X174 picked up in the
20 and 60 minute samples (Table 27, sample numbers 8, 9, 12
and 13).   No indication of virus getting to the lake was
seen in the 12 mg/1 C102 experiment.

Samples 10 and 14 of Table 27 and 28 indicated no viricidal
action by creek water, and sample 19 showed no viricidal
action by lake water.  Therefore, one cannot depend on the
natural water into which CSO discharge for the inactivation
of any virus which has escaped disinfection.

Both disinfectants were effective at the levels used.  The
Cl2 efficiency, however, was partly dependent upon residual
activity which occurred after entrance of the CSO into the
receiving water.  This is probably less dependable than the
C102 treatment in which almost all of the virus was inacti-
vated before entrance into the stream.  It is anticipated
that the use of the resistant f2 phage may give a more reli-
able index to the possible number of pathogenic viruses which
could possibly enter the lake.

Piscussion--

It can be concluded that aftergrowth may occur, but it is
difficult to assess this phenomena in a laboratory experiment.
It will also be difficult to determine aftergrowth on a full-
scale test in the demonstration phase because of the large
numbers of bacteria entering Onondaga Lake from untreated CSO,
and from the existing treatment plant.  Research of this type
is bes.t conducted in rivers and streams where the bacterial
growth can be observed in a natural environment free from
interferences and other bacterial inputs.
                             101

-------
                                    scso
                                SEEDED WITH VIRUS
                                     #2
                                AFTER SCREENING
                                                  _T I 2 mg/l CI02 or
                                                  T 25 mg/l CI2
                    DILUTION WITH CREEK WATER
                 1:14 DILUTION WITH LAKE WATER
Figure  39.  Key  for  samples  collected  in  viral  aftergrowth
              study
                                  102

-------
Table 27. Aftergrowth of Virus; Di'slnfected with C3-2
Sample Number
From Figure 39
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Table 28. Aftergrowth
Sample Number
From Figure 39
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Polio-1
(PFU/ml)
7.4 X 103
6.6 X 103
6.4 X 103
9.2 X 101
4.0 X 10°
— _ —
1.4 X 103
0.0
0.0
_ __
5.8 X 103
0.0
0.0
5.0 X 101
0.0
2.2 X IQl
0.0
0.0
4.0 X 10°
0.0
of Virus Disinfected with
Polio-1
(PFU/ml)
5.8 X 104
5.8 X 104
5.9 X 104
2.0 X 10°
3.2 X 101
0.0
2.7 X 104
0.0
0.0
0.0
1.8 X 104
0.0
0.0
0.0
0.0
1.0 X 103
0.0
0.0
1.2 X 101
0.0
0X174
(PFU/ml)
7.3 X 10/
8.6 X 107
7.9 X 107
6.0 X 105
5.0 X 104
__«
1.1 X 108
3.0 X 10°
3.0 X 10°
4.0 X 104
5.8 X 107
6.0 X 10°
3.0 X 10°
5.0 X 104
0.0
2.7 X 106
7.3 X 101
0.0
2.5 X 103
0.0
C102
0X174
(PFU/ml)
1.6 X 108
2.0 X 108
1.3 X 108
3.6 X 102
1.5 X 103
0.0
1.1 X 108
0.0
1.4 X 102
0.0
5.8 X 107
0.0
0.0
0.0
0.0
1.6 X 106
0.0
0.0
3.3 X 103
0.0
                             103

-------
Breakpoint Chlorination

The results for ammonia removal by breakpoint Chlorination
are given in Table 29.

Table 29. Breakpoint Chlorination in SCSO	

                                 Time  (seconds)
C12 (mg/1)
0
5
20
30
35
40
45
50
60
Cl?:NH3Na
0
1:1
4:1
6:1
7:1
8:1
9:1
10:1
12:1
0

5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
5.35
30
NH^N
5.35
5.35
4.87
3.51
1.68
1.08
0.79
0.50
0.37
120
(mg/1)
5.35
5.35
4.59
3.23
1.50
0.29
0.14
0.10
0.11
300

5.35
5.35
4.48
3.06
1.54
0.36
0.09
0.06
0.08
a mg/l:mg/l

The ORGN, T-ALK, pH and NC>3N, N02N values changed by less than
10% of the original values.  The results support other experi-
ments^7 showing ammonia removal at C12:NH3N weight ratios
greater than 7.6:1 in wastewaters.  In practice, the break-
point Chlorination may be difficult to carry out since strict
pH control at 6.8 + 0.5 is necessary.  If the natural buffering
capacities of wastewaters is exceeded, pH adjustments are
necessary.  Although this did not occur in these studies, this
possibility must be investigated for more dilute CSO.  Failure
to maintain proper pH results in the substitution of one pro-
blem  (NC13 or N03N) for the original  (NHsN)  as shown in the
procedures section.

The expense of a minimum 7-6:1 ratio excess of Cl2 to NH3N
must be evaluated in view of other means of ammonia removal.
The other common possible methods are biological nitrification,
stripping and ion exchange processes.  The first of these
requires long detention times of several hours, and the second
is not practical in northern regions of the country for climatic
reasons and is not considered in this report.  The use of
conventional synthetic ion exchange resin has been criticized
for its expense and lack of selectivity.  However, the possi-
bility of using a naturally occurring material, clinoptilolite,
is presently being pursued under U.S. EPA Grant No. S-8024007"1
The final report on this project will present detailed costs
of this method.
                              104

-------
The operating costs of breakpoint chlorination, considering
only Cl2, is calculated as follows:

     1.  Laboratory Results:

         8.5 mg Cl2/mg Nt^N is required to reach the breakpoint
         in CSO

     2.  Assumptions:

         Average maximum NH^N in CSO is 5 mg/1  (Appendix A).
         Number of treatment stations on CSO system is 601-
         Average station capacity is 15 mgd1-
         Average duration of one CSO is 2.5 hr1.
         Average number of overflow occurrences is 8.8/mol.
         Cost of Cl2 is $0.10/lb  (ton cylinder)       ] 1974
         Cost of Cl2 is $0.07/lb  (liquid hypochlorite)} prices

     3.  Conversion Factors:

         1 Ib = 454 gm
         1 gal = 3.785 1
         1 mg/1 = 8.34X10-6 Ib/gal
         1 hr = 4.17X1Q-2 day

     4.  Calculations:

         a.  Ib of Cl2 required per gal CSO to reach breakpoint

             (5 mg NH3N/1 CSO) X  (8.5 mg Cl2/mg NH3N) =
             (42.5 mg C12/1 CSO)

             (42.5 mg C12/1 CSO) X (8.34X1CT6 lb/gal/mg/1) =
             (3.54X10~4 Ib Cl2/gal CSO)

         b.  gal CSO to be treated in one year

             (60 stations) X  (1.5X107  gal CSO/day/station) X
             (4.17X10"2 day/hr) X (2.5 hr/occurrence) =
             (9.36 X107 gal CSO/occurrence)

             (9.36 X 107 gal CSO/occurence) X  (8.8 occurrence/mo)
             X (12 mo/yr)  = (9.88X109 gal CSO/yr)

         c.  Ib Cl2 required per year

             (9.88X109 gal CSO/yr) X (3.54X10~4 Ib Cl2/gal
             CSO)  = (3.50X106 Cl2/yr)
                               105

-------
          d.   cost of Cl2 per year

              ton cylinders
              (3.50X106 Ib Cl2/yr)  X ($0.10/lb)  = $350,000/yr

              liquid hypochlorite
              (3.50X10^ Ib Cl2/yr)  X ($0.07/lb)  = $245,000/yr

Thus, for ammonia removal in CSO, the cost for Cl2 for one
year is $245,000 or about $0.025 per thousand gallons of
treated CSO.   This compares with a cost of approximately
$135,000 for disinfection alone based on 25 mg/1 Cl2-  For
two-stage disinfection with 8 mg/1 C102 and 2 mg/1 Cl2 / the
cost would be $45,000 plus the cost of C102.   The additional
cost of ammon removal beyond the cost of disinfection is
$110,000 or about $0.01 per thousand gallons of CSO.  This
cost compares very favorably with stripping which has been
estimated at $0.02 per thousand gallons.  Other methods are
generally thought to have even higher costs.

DESCRIPTION AND OPERATION OF DEMONSTRATION FACILITIES

Introduction

The discussion of the results of the bench-scale studies indi-
cates that because of the difficulty in simulating field con-
ditions, particularly mixing, a complete assessment of the
factors affecting disinfection was not possible.  Therefore,
the demonstration of full-scale facilities hereafter outlined
will address the following questions: 1) What degree of confi-
dence should be placed on the results from bench-scale disin-
fection studies when projecting the designing of full-scale
tankage and disinfectant-generating equipment? 2) What degree
of confidence can be placed on the selection of operating para-
meters such as disinfectant dosages by assessing bench-scale
results? 3) What can be considered a feasible method for
achieving a high-rate removal of microbial contamination? 4)
What is the effect of various levels of high-rate suspended
solids removal prior to high-rate disinfection?

To address each of these questions by subjecting the demonstra-
tion facilities to various modes of operation is the major
objective of the field program.  This program will have a
major impact on the overall CSO abatement program for
Onondaga County and many of the results and observations will
have an impact on a national basis.  This section describes
the demonstration facilities, their subsequent operation and
the relation of proposed evaluation to the results of the
laboratory studies.  Also included in this section is a
description of the sampling and analysis program designed to
provide data with which the feasibility of process concepts
                              106

-------
will be assessed.

Description of Facilities

The site plan for each of the demonstration facilities is
shown in Figures 40 and 41 for Maltbie Street and West Newell
Street, respectively.  The Maltbie Street facility was
designed around the concept of high-rate, fine-mesh screening
followed by high-rate disinfection.  This station includes
several pieces of mechanical equipment and automated control
instrumentation.  The West Newell Street facility was designed
around the swirl concentrator conceived in England by
Smisson^ , and further developed by the U. S. Environmental
Protection Agency in cooperation with the American Public
Works Association, General Electric and LaSalle Hydraulic
Laboratory, in Montreal, P.Q.73.  The swirl flow regulator-
solids concentrator concept of first controlling or regula-
ting storm flow and simultaneously treating the flow was
installed to demonstrate the feasibility of an installation
relatively free of mechanical equipment.

The design of the facilities was initiated before the bench-
scale studies had been completed.  Therefore, hydraulic and
process flexibility were maximized wherever possible in anti-
cipation of variations in process concepts that would be
precipitated by an analysis of the bench-scale results.  Also
adding to the consideration of design flexibility were the
results of a literature search of the previously completed
CSO characterization programs which indicated a wide varia-
bility of microbial and organic discharges.  This variability
was verified by a characterization program of three different
CSO within the Onondaga Lake drainage basin (Appendix A).

Maltbie Street Prototype Treatment Facility—

Pumping Station—As shown in Figure 40, the Maltbie Street
Pumping Station is located at the discharge of a CSO into
Onondaga Creek.  Figure 42 illustrates that the pumping
station structure is divided into three compartments, namely,
the influent chamber, the metering chamber and the wet well.
The chamber is equipped with a bar screen and an emergency
bypass to provide the removal of large particulates and
allow the bypassing of the total CSO in the event of pump
failure.

The metering chamber was designed with a magnetic flow meter
(Brooks Instrument Division, Emerson Electric Company, Hatfield,
Pa.) which will measure the total CSO emanating from the over-
flow regulator.  Maximum capacity of the existing outfall has
                              107

-------
                                                                                                        ONONDAGA
                                                                                                         CREEK
                                                          DISINFECTION  CONTACT
                                                                TANKS
O
OO
                                   TOTAL DISINFECTED
                                      EFFLUENT
                    OVERFLOW  REGULATOR
                      W/ LEAPING  WEIR
TRANSPORT
  TRUNK
                                                               SOLIDS  CONCENTRATE -
                                                                     LINE
                                        SIPHON TO MAIN-
                                          INTERCEPTOR
              Figure   40.  Maltbie   Street   site  plan

-------
        EXISTING  COMBINED  SEWER
                   *= SAMPLING POINT
~v
\
\
( H
\ i i
\ i
\\ ' /







\ \ SOLIDS i /
\ \ CONCENTRATE ' '
\ \UNE 1
> ' II COMBINED
\ \ ' 1 SEWER
! . , || INFLUENT
\ \ ' '
I ' '
\\ ,
\\
\\ '
U 1
PUMP (g) / |
MANHOLEVr^/ I .

»1 II
MAGMETER— ^^f j |
MAW
; i
fit:


i


SOLE ' ' '1
\ \ » 1 i y
\\> < | |
— ~J //'*~~~**\\\l DISIN-
-J (( <~>&\\ I CI2 or FECTICN
\ U ^ ) Cl°2 EOUIP-
\ \V_^X7 / MENT
< V^ 	 x X HOUSING
SWIRL "V'
CONCENTRATOR
Figure  41.  West  Newell Street site  plan
                               109

-------
                           AIR  FLOATATION CELL
                               *rSOLIDS CONCENTRATE LINE
           SWECO
           WASTE-
           WATER
           CONCENT-
           RATOR
 CRANE
 "MICRO-
STRAINER"
         SCREENING  BUILDING AND DISINFECTION TANKS
                  KEY
            M - FLOW METER
            X - SAMPLING LOCATION
           ---FLOW DIRECTION
                                                                    '-I—L^ —
                                     x
                                                                                       o  I
                                                       o

                                                                    PUMPING STATION
figure  42.   Maltbie  Street  process  orientation

-------
been calculated as 30 mgd  (1314 I/sec.).  The flow signal
from the magnetic flowmeter will be used to activate a circuit
to which all sampling equipment will be wired.

The third chamber in the pumping station acts as a wet well
for the overall pumping system which is located above the
chamber.  The pumping system was designed as three parallel
pumping systems each consisting of a 2.5 mgd  (109.5 I/sec.)
constant speed and a 2.5 mgd  (109.5 I/sec.) variable speed
pump.  The three 5 mgd  (219 I/sec.) pump combinations serve
to feed three parallel high-rate screening and disinfection
systems.  These systems are described later in this section.

The pumping systems will provide a variable or constant flow
to the screening units.  Under variable flow operation, the
variable speed pump will be activated by a level sensor in
the wet well.  It will operate until its capacity of 2.5 mgd
 (109.5 I/sec.) is reached at which time the constant speed
pump will be activated.  The variable speed pump will continue
to operate until the 5 mgd  (219 I/sec.) capacity of the single
pumping system is reached.  To provide an operation under
constant flow conditions, the variable speed pump will be
manually inactivated, and upon the occurrence of an overflow,
the constant speed pump will automatically start when the wet
well level is sensed.

Located on the discharge side of each pumping system is a
magnetic flowmeter which will provide a signal to activate
a screening unit and the disinfection equipment associated
with that screening unit.  These flow signals will permit
unattended operation of the entire facility, thereby decreas-
ing manpower requirements and overall operating costs.  This
factor will be taken into consideration in the evaluation of
the facility.

Screening Capabilities—The screening facilities illustrated
in Figure 42 were hydraulically designed to provide flexibility
in modifying treatment processes.  All influent and effluent
piping is of steel fabrication which will simplify any modi-
fications in the piping arrangement.  Modifications will be
made during the demonstration program if results indicate
that a change or changes may optimize the high-rate process.

The parallel screening units consist of a Crane Microstrainer
(Crane Co.,' Chicago, Illinois) with a screen aperture of 23
microns (Mark O), a Zurn Micro-Matic ® (Zurn Industries, Inc.,
Erie, Pa.)  with a screen aperture of 71 microns and a Sweco
Wastewater Concentrator ®  (Southwestern Engineering Co.,
Massilon, Ohio)  with a screen aperture of 105 microns.  The
                              111

-------
specific screening units were selected on the basis of
results obtained from previous high-rate operation of the
screens41'74'75 and the variability in process design
afforded by these units.

The Crane and Zurn units are alike in that both remove solids
through a finely-screened drum which rotates relatively
slowly (4.5 to 6.5 rpm) around an horizontal axis.  The Crane
unit has had the higher number of operating hours on CSO
between the two units; however, it is expected that each will
provide similar suspended solids removal capabilities.  The
Sweco Wastewater Concentrator removes TSS by the use of a
combination of finely-meshed screens fastened to a cage which
revolves at a high speed (55 rpm)  around a vertical axis.
The concentrator is followed by an air flotation cell which
takes advantage of entrapped air in the screened effluent for
a higher degree of suspended solids removal.

Initially, the loading rates on the screening units will be
varied throughout the duration of the overflow by operating
each pumping system in the variable state.  The loading rates
on the Zurn and Crane units may vary from approximately 10 to
45 gpm/ft2 (0.41 to 1.84 1/min/m2) during any given runoff
period.  Because of the larger screen aperture on the Sweco
unit, the loading rates on it may vary from approximately 30
to 45 gpm/ft2 (1.23 to 1.84 1/min/m2).  Some evaluation of
the range of loading rates at which each unit can effectively
operate will be attempted along with an assessment of the
operational efficiency of the units under varying hydraulic
conditions.

In order to better evaluate the operating range of each unit,
however, the flow to the units will be held constant during
several trials to provide loading rates of 30 to 45 gpm/ft2
(1.23 to 1.84 1/min/m2) on the Crane and Zurn units and
loading rates of 85 to 145 gpm/ft2 (3.48 to 5.94 1/min/m2)
on the Sweco unit.  This assessment of operating ranges will
hopefully provide enough information from which to project
the necessary number of screening units, and thus tl\e
building size for various point-source treatment facilities.

Each screening unit will be evaluated for its operating effi-
ciency and degree of maintenance.   Data will be accumulated
on the ability of the screening units to start and stop after
both long and short periods in and out of operation, on the
effectiveness of their backwash cycles, and on the durability
of the screening material by assessing screen life.

Aside from the operating efficiencies of each unit, the efflu-
ent TSS quality from each screen will be evaluated as to  its
                             112

-------
effect on the disinfection process.  No conclusive findings
could be made during the bench-scale studies regarding this
phenomena due to the difficulty in simulating a screening
process in the laboratory.  Therefore, since each screening
unit is equipped with screens of different aperture sizes
and it is anticipated that the effluent emanating from each
will differ in TSS quality, the effect of aperture size on
the disinfection process can be further evaluated.

During the demonstration period, the screens on two of the
screening units will be changed to produce different quality
effluents.  The aperture size selected will be dependent upon
the results of the initial evaluations outlined above.  How-
ever, the physical size and shape of the TSS may be different
due to the different methods of removal provided by the Sweco
Wastewater Concentrator and the Crane and Zurn microscreens.

An hypothesis formed from the results of the bench-scale
studies purports that the method of TSS removal may have an
effect on the bactericidal and viricidal efficiency of the
disinfecting agent.  The reliability of this hypothesis will
be evaluated through the results obtained during the demonstra-
tion program.

Throughout the entire period of demonstration, the total
amount of TSS removed will be assessed as to its effect on
interceptor sewer loadings, assuming all the collected TSS
are returned to the main intercepting sewer.  Also projected
will be the effects of these TSS on the treatment and TSS
handling capacity at the local sewage treatment plant.  These
evaluations will have to be made in order to project the
requirement of on-site solids handling at each point-source
treatment facility.

Disinfection--As noted above, various evaluations will be
made on the effects of solids on the disinfection process.
The disinfection facilities available for the demonstration
project were designed to provide high-rate inactivation of
bacteria and viruses.  The levels to which the bacteria
must be reduced in order for the disinfection process to be
judged sufficient is discussed earlier in Section V.   It
should be noted that the standard of 1,000 TC per 100 ml is
a maximum level to be measured in the receiving waters for
this demonstration, but the standard will be used as a goal
for bacterial quality in the effluent.

The disinfection detention facilities consist of three
parallel tanks, one for each of the effluents from the para-
llel screening processes.   The tanks were sized by using the
                              113

-------
one-minute contact period that was indicated as sufficient
for reaching the bacterial standard from the results of the
bench-scale tests.  Each tank was designed with a proportional
weir which will provide a constant velocity and, therefore,
a constant detention time regardless of flow.  This constant
contact period will facilitate trial-to-trial comparisons of
disinfection techniques despite variability of flow rates.

In order to demonstrate the effectiveness of different mixing
applications, a common header to each tank from a single
screening unit is now being proposed.  This will insure that
the influent to each tank will be as equal as possible, thus
maximizing the validity of any comparison of mixing procedures.

Initially, two tanks will be equipped with three flash mixers,
one located at the point of disinfectant injection and the
others located near the downstream end of each longitudinal
baffle.  The third tank will be equipped with a single flash
mixer at the point of disinfectant injection.  This tank will
be available for the installation of Glover's corrugated
labyrinth design (discussed below).   This demonstration study
will include single-flash mixing in each tank and sequential-
flash mixing.

Bench-scale studies used complete mixing in a way that may
not be duplicated in a full-scale facility.  Therefore, a
portion of the project will be devoted to demonstrating the
influence that various techniques of mixing have on the dis-
infection process.   The detention tanks were designed to allow
the addition of several types of facilities for this purpose.
Those that have been selected for demonstration include flash
mixing, sequential-flash mixing, and increased turbulence
through the use of corrugated labyrinths.  Each of these
mixing techniques will be assessed against a no-mix condition.

Single-flash mixing will provide for a nearly instantaneous
mixing of the disinfectant and the CSO at the point of appli-
cation.  The effect on disinfection efficiency of this
technique will be compared with that of sequential-flash
mixing.  During the laboratory studies, two-stage disinfection
was performed to determine whether or not sequential mixing
should be considered as a demonstration possibility.  Although
the results were far from conclusive, they indicated that
this concept warranted consideration during the demonstration
program.   Positive results were observed by Kruse, et.al.,45
during a full-scale sewage treatment plant study of sequen-
tial-flash mixing,  and by Dow Chemical39 during a high velo-
city, narrow flow-through tube study.
                              114

-------
Pilot work by Glover42 indicated that an increase in
turbulance throughout the length of the detention tank will
increase the efficiency of the disinfection process.  The
design objective for Glover's high-rate pilot contact chamber
was to achieve a G't factor of 10,000; G1 being the velocity
gradient and t being the detention time in the tank.  Since
detention time is relatively constant the only manner in
which G't can be increased is by increasing G'.  Glover did
this by inserting corrugated, closely-spaced baffles parallel
to the flow.  A similar installation will be made in one of
the parallel demonstration tanks to compare the efficiency
of this disinfection technique to the others listed above.

In addition to assessing the physical mechanics of disinfec-
tion, the type of disinfection (single- or two-stage) and the
optimum dosage quantities of Cl2 and/or C1C>2 will be assessed.
The bench-scale studies evaluated the optimum dosages, here-
after referred to as minimum dosages which reduce the bacter-
ial levels below the defined goals, of Cl2 and C102 on a
single-stage basis and also the optimum combination of Cl2
and C102 dosages on a two-stage basis.  The results and a
subsequent discussion of these studies are presented in
Section V.  The bench-scale studies concluded that single-
stage Cl2 and ClC>2 disinfection will be optimal at 25 and
12 mg/1, respectively, and that two-stage disinfection with
Cl2 followed by C102 will be optimal at 8 and 2 mg/1,
respectively (it should be noted here that these dosages are
all based on contact periods of two minutes or less).  It
was also concluded that two-stage disinfection will produce
an effluent of acceptable bacterial quality at a lower cost
than single-stage will produce.

The one major drawback in projecting laboratory results to
full-scale application is the inability to simulate actual
field conditions in the laboratory.  Anticipated conditions
such as temperature fluctuations and TSS variability were
studied and accounted for when analyzing the results.
However, it was impossible to simulate the various physical
shapes of solid particles  that may exist in the effluents
of different solid removal processes.  Simulation of the
incomplete mixing process in the contact tanks was also very
difficult.  Therefore, the bench-scale results for both
single-stage and two-stage disinfection must be verified by
full-scale testing.

As stated above, the disinfectant agents that will be com-
pared or combined sequentially will be Cl2 and C102.  These
agents, however, will be generated in a different manner in
the demonstration program than they were in the bench-scale
                              115

-------
tests.  The C3-2 used in the laboratory was obtained as a  5%
solution of sodium hypochlorite (Section IV), whereas_C12
will be generated at the demonstration site by combining  Cl2
gas stored in ton cylinders with water to form liquid HOC1.
The C102 used in the laboratory was generated according to the
technique described in Standard Methods5, whereas C102 will
be generated by means of a Nitrosyl Chloride generation
system  (U. S. Patent 375079)6 at the demonstration site.

Particular attention will be given to the C1C>2 generation
method since it is a new process for full-scale application
and the concentration of C1C>2 produced is critical.  The pro-
cess consists of pumping two chemicals, a NaCl03-NaNC>2  (sodium
chlorate-sodium nitrite) slurry and HNC>3 (nitric acid) , which
will mix in a specially designed lucite reaction chamber.
The resulting reaction is expected to produce a 12 percent
solution of C102 that will be fed directly to the disinfection
contact tank.  The product will be sampled as often as possible
to determine the consistency of C102 concentration.

The capacity of each generator was selected on the basis of
the generation of an amount of disinfectant sufficient for
treating a 5 mgd (219 I/sec.)  flow (maximum capacity of one
contact tank) at the dosages that were determined as optimum
from the bench-scale results.   However, with the hydraulic
flexibility of both the pumping system and the screening units,
it will be possible to decrease total flow, increase screen
loading capacity (decrease drum submergence) and thus, increase
the dosage rate of Cl2 and/or C1C>2.

Sampling and Analytical Program—Aside from influencing the
design and operation of the demonstration treatment facilities
and related instrumentation, the results of the bench-scale
tests will also have an effect on the type and amount of
sampling and analysis that will be performed during the demon-
stration program.  Table 30 presents a tentative outline  for
the operation of the Maltbie Street facility.  The various
hydraulic and process alternatives that will be demonstrated
have been discussed in the previous sections describing this
facility.  For each trial, samples will be taken at the influ-
ent to the screening units, the effluent to the units and the
flotation cell, and at key locations in the disinfection
contact tanks.  These locations will be selected on the basis
of the mixing application used in the particular tank.  For
single-flash mixing and Glover's corrugated baffle system,
the samples will be collected immediately after the mixer,
midway, and at the end of the tank.  For sequential-flash'
mixing, the samples will be collected after each mixer and at
the end of the tank.  The data is to be evaluated as follows:
                              116

-------
Overflows Single-stage C102 disinfection verification and
 (1-3)    the effect of various amounts of TSS on disinfec-
          tion.  Preliminary evaluation of various screen
          loading rates.

Overflows Single-stage Cl2 disinfection verification and
 (4-6)    comparison of TSS effect  (C102 vs. Cl2).  Increase
          data on screen  loading rates.

Overflows Intense evaluation of high screen-loading rates.
 (7-10)   Attempt to assess validity of hypothesis on effect
          of physical shape of solid particle on disinfection.

Overflows Two-stage disinfection verification and comparison
 (11-14)  of single- and  two-stage in parallel tanks with
          near equal influent suspended solids.

Overflows Using optimum method of disinfection, evaluation of
 (15-20)  mixing techniques.

The samplers to be used are Sigmamotor refrigerated, sequen-
tial samplers (Sigmamotor, Inc., Middleport, N.Y.).  As
previously stated, they will be activated when a flow is
sensed in the 30-inch flowmeter located in the pumping station,
The sampling interval will initially be set at 15 minutes
which will allow for a high accumulation of data throughout
the critical period of operation (first flush).  Therefore,
if the CSO extends over a six-hour period, there will be a
possibility of collecting 24 samples at each location.  An
analytical schedule is presented in Table 31.

West Newell Street Prototype Treatment Facility—

Swirl Concentrator—Figure 41 illustrates the relationship
of the swirl concentrator to the existing service trunk sewer
and combined sewer overflow located at West Newell Street.
The background and experimental work completed on the swirl
concentrator is documented in U.S.  EPA reports73"76.  The
design adopted for this installation closely follows the
recommendations cited in the first report.  A pump was added
to the solids concentrate line in order to allow the unit to
operate under forced as well as gravity conditions.  Due to
existing hydraulic conditions in the interceptor system, the
pump is needed to maintain dry-weather flow in the primary
gutter.  During a rainfall,  the pump will shut off when the
flow through the swirl chamber reaches the calculated capacity
of  the interceptor.   The chamber will then fill to the weir
level and begin to overflow.  As opposed to the Maltbie Street
facility, the swirl concentrator operates with relatively
                              117

-------
Table 30. Maltbie Street Operation  Schedule
Screen Loading
Over- Rate (gpm/ft2)
flow
1
2
3
4
5
6
7
8
9
10
11
, 12
13
14
15
16
17
18
19
20
c -
v -
F -
FS -
G -
Sweco Crane
V
V
V
V
V
V
73
73
73
73
c
c
c
c
c
c
c
c
c
c
Constant
variable
V
V
V
V
V
V
30
30
45
45
c
c
c
c
c
c
c
c
c
c


Single-flash
Screen Aperture
Size (microns) Cl2 Dosage (mg/1)
Zurn Sweco Crane Zurn Tank #1 Tank #2 Tank #3
V
V
V
V
V
V
30
30
45
45
c
c
c
c
c
c
c
c
c
c


mix
Sequential- flash
Glover1 s
con
rucrat
105
105
no scr.
105
105
105
105
105
105
105
common
H
11
"
11
"
"
11
11
"



mix
:ed labvri
23
23
23
23
23
23
23
23
23
23
71
71
71
71
71
no scr.
71
71
71
71
header
ii
"
11
11
11
11
11
"
"




.nth
t
1
1
I










0
0
0
25
25 +
25 +
25 +
0
25 +
0~
0
0
0
0
8+
8 +
8 +
8 +
8+
8 +





0
0
0
25
25+
25+
25+
0~
25 +
0
25+
25 +
25+
25 +
8+
8+
8+
8 +
8+
8 +





0
0
0
25
25+
25+
25+
0
25+
0~
8
8+
8+
8 +
8+
8 +
8+
8+
8+
8+





 + - Vary dosage up or down
NM - No mix
                              118

-------
Table 30. Maltbie Street Operation Schedule  (Cont'd)
C102 Dosage (mg/1)	
Tank #1Tank #2Tank #3
   12       12       12
   12+      12+      12+
   12+      12+      12+
    000
    000
    000
    000
   12+      12+      12+
    000
   12+      12+      12+
   12+       0        2
   12+       0        2+
   12+       0        2+
   12+       0        2+
    2+       2+       2+
    2+       2+       2 +
    2+       2+       2+
    2+       2+       2+
    2+       2+       2+
    2+       2+       2+
 c - Constant
 v - Variable
 F - Single-flash mix
FS - Sequential-flash mix
 G - Glover's currugated labyrinth
 + - Vary dosage up or down
NM - No mix
Mixing
Tank #1
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
NM
F+
F+
F+
F+

Tank #2
F-
F
F
F
F
F
F
F
F
F
F
F
F
F
NM
G
G+
G+
G+
G+

Tank #3
F
F
F
F
F
F
F
F
F
F
F
F
F
F
FS
FS
FS +
FS+
FS+
FS +
Cost
(dollars)
9,000
9,000
9,000
9,000
4,000
4,500
4,500
4,500
4,500
4,500
4,500
4,500
4,500
4,500
4,500
4,500
6,750
6,750
6,750
6,750
                               119

-------
Table 31.  Maltbie Street Analytical Schedule


Total
Screen
Parameters Influent
BODs
TOC
COD
TKN
NH3N
N02N
N03N
T-IP
Cl
T-Alk
TSS
Sett. Solids
VSS
TSa
vsa
TC
FC
FS
PH
Oil & Grease"
Metalsc
Flowrate
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Samples
Each
Screen
Effluent
X
X
X
X






X
X
X
X
X
X
X
X
X


X
to be taken
Flotation
Effluent
X
X
X
X






X
X
X
X
X
X
X
X
X


X

Disinfec-
tion Tank
Locations
X
X
X
X
X
X
X
X
X
X





X
X
X
X
X
X
X
 5* Analyses on composites or every six samj:
   Analyses on selected composites
 c Includes Hg, Pb, Cr, Fe, Cu, Cd, Ni, Zn
little mechanical equipment, thus providing a marked contrast
in anticipated operating and maintenance expenses.

The influent to the unit is comprised of the entire flow  (dry
and wet-weather) from the service trunk sewer.  This flow
averages approximately 1.0 mgd (43.8 I/sec.) during dry-
weather conditions and ranges from an average of  3 mgd  (131.4
I/sec.) to a maximum of 8.9 mgd (398.8 I/sec.) during wet-
weather conditions.

Under complete gravity service, dry-weather conditions  will
produce a flow which will be kept within a channel located in
the floor of the swirl unit.  The flow will be discharged
through the solids concentrate line which will carry the  flow
back to the interceptor.  During wet-weather or runoff
                             120

-------
conditions, the  flow  in  the  dry-weather  channel will  increase
until it spills  out onto the floor of the  swirl chamber.
The water level  in the tank  will then continue to  rise  to
the level where  it will  spill over the overflow weir  and
discharge to the receiving stream.

A swirling action will be produced by the  momentum of the
influent flow and the geometry  of the tank during  an  overflow
condition.  This swirling pattern will cause  the solid
particles suspended in the flow to move  toward the outside
portion of the tank and  downward.  Many  of these particles
will then become trapped in  the dry-weather channel and
subsequently will be  removed through the solids concentrate
line.

The pump that has been installed in the  solids concentrate
line will provide the flexibility of testing  the operation of
the swirl concentrator under both gravity  and pumping condi-
tions.  It is anticipated that  the swirl concentrator may
not be feasible  in many  locations in Onondaga County  as well
as nationwide due to  the  lack of available hydraulic  head to
direct the dry-weather flow  back into the  main sewer  line.
Although pumping will increase  operating and  maintenance
costs, it may be  necessary to overcome hydraulic restrictions.

Disinfection—The portion of the flow which will spill  over
the weir will be  disinfected with ClC>2 generated in the same
manner as outlined for the Maltbie Street  facility.   There
is no specially  designed  disinfection contact chamber follow-
ing the swirl concentrator,  only a 30-second  to one-minute
detention time in the effluent  piping.   A  flash mixer will
be available at  the point of the C1C>2 injection to  enhance
the probability  of ClC>2/microorganism  contact.  There  are
also provisions  to prechlorinate and take  advantage of  the
mixing action inherent in the swirl chamber.

During the initial operation of the facility, an Englehard
Chloropac sodium hypochlorite generating system^  (Englehard
Industries Division,  Englehard  Minerals  and Chemicals
Corporation, East Newark, N.J.) will be  installed  and will
be available for demonstration  toward the  middle of the demon-
stration period.   The system will utilize  an  existing brine
supply as its influent,  and  through the  use of Englehard's
electrolytic process, sodium hypochlorite  will be  generated,
stored on site,  and fed  to the  swirl concentrator  effluent
during wet-weather conditions.

By installing the Englehard  system at West Newell  Street, an
evaluation of the use of  sodium hypochlorite  in the two-stage
                             121

-------
disinfection process will be possible.  It will also provide
an opportunity for assessing the comparative reliability of
C102 and sodium hypochlorite generation techniques.

A summary of the intended operational schedule is given in
Table 32.

Table 32. West Newell Street Operation Schedule _

                Disinfection               Englehard
           _
           C102 (mg/1)  C102 (mg/1)  Cl2 (mg/1)  C102
Storm Flow Pre-disin-  Post-disin- Pre-chlori- Post-chlori
No.   Rate fection     fection     nation      nation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
V
V
V
V
V
V 15+
V 15+
V 15 +
V 15 +
V 15 +
V 15+
V
V
V
V
V
V
V
V
V
V
12+
12 +
12 +
12 +
12 +

2 +
2 +
2+






4 +
4 +
4 +
4 +








8 +
8+
8 +
8+
8 +
8+
8 +





25+
25 +
25 +
20+
20+
20+




+_ - Vary flow up or down

Sampling and Analytical Program—The sampling equipment,
sampling intervals and analytical parameters for West Newell
Street are ^listed in Table 33.   They are similar to those
outlined for Maltbie Street in Table 31.  The samples will be
collected from the influent,  the effluent prior to disinfec-
tion, and the effluent before it is discharged to Onondaga
Creek.  Samples will also be  collected from the C1O2 and
sodium hypochlorite generators on a random basis to monitor
the strength of the disinfectants and the consistency of the
generation process.  The removal of TSS by the swirl concen-
                             122

-------
trator is a function of the settling properties of the
particles as well as other parameters.  Therefore, the
settling velocities of the TSS in the West Newell Street
overflow will be measured for several storms.  Periodic grab
samples will be taken at all locations to check on the
reliability of the Sigmamotor automatic samplers.  Particular
carewill be taken to verify the TSS removals.

Table 33. West Newell Street Analytical Schedule	

            	Samples to be taken	    	
Parameters  Influent
Pre-Disinfected
   Effluent
Post-Disinfected
   Effluent
BOD5a
TOC
COD
TKN
NH3N
NO 2
NO 3
T-IP
Cl
T-ALK
TSS
Sett. Solids
VSS
TSa
TVSa
TC
FC
FS
PH
Oil & Grease^
Metals0
Flowrate
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X





X
X
X
X
X
X
X
X
X


X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
 a Analyses on composite of every six samples
 k Analyses on selected composites
 c Includes Hg, Pb, Cr, Fe, Cu, Cd, Ni, Zn

Special Instrumentation

The flow data that will be generated will be telemetered to
a central location, identified and stored on punch tape.  The
tapes will then be processed and the information premanently
stored, along with coinciding analytical and rainfall data,
on computer disc files.  The purpose of the telemetry and
computer system is to minimize man-hours in data handling
and to establish a basis for a central control station for
                             123

-------
the overall CSO abatement program.

Also included in the demonstration program will be an evalua-
tion of the reliability of the Badger Meter Suspended Solids
Monitor (Badger Meter Company, Milwaukee, Wis.).  This
monitor is based upon an optical depolarization method which
was developed earlier by American Standard80.  Badger Meter
has carried on further development of the solids meter,
refining its optical and electronic design and adapting the
meter for unattended service in a sewer environment.

The meter will be installed in several flow streams at both
Maltbie Street and West Newell Street.  It will be tested
under various suspended solids conditions starting with a
low concentration of less than 5 mg/1 and ranging up to
10,000 mg/1.  Initial correlations will be made between meter
readings and laboratory measurements on specially collected
samples.

All chemical and operation data collected over the entire
period of demonstration at both Maltbie and West Newell
Street will be assessed from a standpoint of individual
process reliability and efficiency as well as overall CSO
treatment capabilities.   This latter assessment will be made
by making use of the U.S.  EPA Stormwater Management Model,
EPA Project No.  11024 DOC81"84 to project the feasible
alternatives for alleviating the CSO problem that exists in
Onondaga County.   The information generated from this demon-
stration will also assist in the planning and implementation
of abatement programs throughout the nation.
                             124

-------
                          SECTION  VI

                            SUMMARY

In order to prevent bacterial  and viral  contamination  of
Onondaga Lake, it was  determined  that  in addition  to new
tertiary treatment facilities  for municipal  and  industrial
discharges, combined sewer  overflows  (CSO) would have  to  be
treated.  Various methods of transmission and  centralized
treatment of these CSO were found to be  relatively expensive^,
therefore, point-source treatment is to  be investigated
through U.S. Environmental  Protection  Agency demonstration
grant, EPA No. S-802400  (11020 HFR) awarded  to the County of
Onondaga.  The treatment  is to consist of different methods
of suspended solids removal followed by  disinfection with
chlorine (Cl2) and chlorine dioxide  (C102),  either separately
or in various combinations.  The  contact times for disinfec-
tion of CSO are generally less than two  minutes  because of
the short distances and rapid  flows from the point of  disin-
fectant application to the  point  of CSO  discharge.  These
constraints are a result  of geographic and economic limita-
tions that are imposed by the occurrence of  CSO  in highly
developed urban areas.  In  order  to obtain approximate
dosages for field demonstration,  it was  necessary  first to
perform high-rate disinfection utilizing a two-minute  contact
time in the laboratory.   Total coliform  (TC),  fecal coliform
(FC)  and fecal streptococci '(FS)  bacteria and  polio-1,
Coxsackie B-3, ECHO-7, HSV-L2, NDV, YFV,  Vaccinia,  f2  and
0X174 viruses (see Table  6) were  the indicator organisms  used
to assess disinfection efficiency-  In addition  to  bacterial
and viral indicators, adenosine triphosphate (ATP)  was also
evaluated as an indicator of efficiency  of disinfection and
as a measure of microbial population.  In order  to  minimize
the sample-to-sample variations,  a simulated combined  sewer
overflow (SCSO)  which  consisted of 50  percent  municipal
sewage and 50 percent distilled water  was used for  all bench-
scale studies.

The properties of C102 and  its reactions in  SCSO were  deter-
mined using electron spin resonance (esr).   In general,
results showed that in two  minutes C102  reacts with various
substances in SCSO to yield chlorite ion (C102~).   These
results were further supported by the  DPD colorimetric
method8 which showed the  sum of C102~  and C102 to  be constant
in each sample to which C1O2 was  added.
                              125

-------
The membrane filter (MF)  technique for bacterial enumeration
was utilized after it was determined that heavy metals, chlori-
nated hydrocarbon pesticides, and suspended solids, in the
levels normally found in CSO, caused no significant difference
from TC levels measured by the MPN technique.  The suspended
solids present in SCSO were shown to have a protective effect
on the bacteria so that counts were not representative of the
true bacterial populations.  Therefore, tests were conducted
to determine the time for which a sample of SCSO should be
blended to give more representative bacterial counts.   Sonica-
tion was disregarded as a method of solids dispersion because
previous studies20'21 showed that bacterial kills result from
the high frequencies employed in the procedure.  Using a con-
ventional household blender, Hamilton Beach Model 50,  a
blending time of six seconds yielded the maximum numbers of
TC, FC and FS.

An attempt was made to evaluate the effects of temperature and
mixing on the disinfection process.  Raising the temperature
from 2°C to 30°C resulted in only a slight increase in high-
rate disinfection of bacteria with Cl2 and C1C>2 in SCSO.  Other
studies in no-demand situations showed marked temperature
       a o a *3
ef f ectsD/:'" .  The difference may be due to the rapid competi-
tive reactions with disinfectant demands which are not pre-
sent in no-demand solutions.  The results of temperature vari-
ation on high-rate disinfection of viruses in SCSO with C102
showed a 40 percent reduction in disinfection efficiency from
22°C to 2°C, but no change was observed at temperatures higher
than 22°C.  This is most likely due to the fact that the con-
version of C102 to C102~ is more temperature dependent than
disinfection with C102.  Other tests with C102, conducted with
0X174, showed only a slight increase in disinfection efficiency
between 2°C and 45°C.   No temperature studies were conducted
for disinfection of viruses with Cl?/ since this subject is
well documented in previous studies^3.  A greater degree of
mixing was accomplished in the bench-scale studies than is
normally obtained in full-scale operations.  Since it would
have been difficult to measure the degree of mixing, rapid
and complete mixing was performed for all studies.

Since the removal of suspended solids will precede disinfec-
tion in the demonstration of full-scale prototype treatment
facilities, the effect of solids removal was evaluated in the
bench-scale studies.  The results of disinfection studies
performed on samples that had been pre-screened through a 23
micron aperture were compared with similar studies on unscreened
samples.  Although total suspended solids were reduced 60
percent with a 23 micron screen, only slight reductions in
BOD5 (4 percent) and bacteria (20 percent) were observed.  In
                              126

-------
some cases the bacterial counts  (unblended) in screened
samples were greater than in unscreened samples as has been
observed by other workers42>  Possible reasons for increased
bacterial counts upon screening  are the removal of natural
bacterial predators, the dispersion of clumps of bacteria
and increased bacterial growth rate due to the dispersion of
solids.  Thus, in addition to the physical removal of bacteria
and solids, screening may affect the efficiency of disinfec-
tion in two opposing ways:

     1.  an increased exposure of bacteria that facilitates
         rapid disinfection, and

     2.  an increased exposure of disinfectant-demanding
         substances, that accelerates the reactions with
         these substances.

On the basis of target levels of 1000 TC per 100 ml, 200 FC
per 100 ml and 200 FS per 100 ml, and target reductions of
five logs of viruses, tests were conducted to determine the
disinfectant dosages required to achieve satisfactory high-
rate disinfection.  To reach the target levels of TC, a
dosage of 20 mg/1 Cl2 was required in 50 percent of the trials,
and little improvement was obtained with dosages of 25 mg/1.
A dosage of 25 mg/1 Cl2 was necessary to reach the desired
value of FS within two minutes in 25 percent of the trials.
It was found that the variations observed in duplicate trials
were of the same magnitude as the difference between screened
and unscreened trials.  Thus, the effect of screening on high-
rate disinfection was considered negligible.  Reductions on
the order of three logs of 0X174 virus were obtained in high-
rate disinfection of SCSO with 25 mg/1 Cl2-  In general, both
bacterial and viral disinfection curves were biphasic as a
result of the disappearance of free Cl2 and subsequent forma-
tion of less potent combined Cl2/ primarily chloramines.  The
target five log reduction of viruses was not reached with con-
centrations of Cl2 as high as 20 mg/1.  Other work has shown
that viruses are not completely  inactivated without achieving
breakpoint chlorination, in which the Cl2 demand is satisfied
to the extent that a free Cl2 residual can be maintained.

Dosages of 8 mg/1 and 16 mg/1 of C102 yielded target levels
of TC in 50 percent and 75 percent of the trials, respectively.
Similar results were obtained using FS, while a dosage of
4 mg/1 of C102 achieved target reductions in virus levels.
Viruses and bacteria showed considerable variability in sensi-
tivity to the C1O2 in the following order: polio-1, FS, TC and
0X174.   Single-stage disinfection of bacteria and viruses with
C102 showed a rapid decrease in  C102 concentration in SCSO,
                              127

-------
with the major disinfection activity completed within 120
seconds.  Although screening had little effect on disinfec-
tion with Cl2, disinfection with C1C>2 was slightly improved.
On the basis of comparative tests, it was surmised that much
of the C102 demand is due to dissolved organics, which are
not so encapsulated within the suspended solids as the Cl2
demanding substances.  Thus, screening does less to affect
C102 demand than Cl2 demand, but does expose more bacteria,
thus enhancing C102 disinfection efficiency.  This was further
verified by an experiment in which suspended solids removal
by settling reduced Cl2 demand, but did not significantly
reduce CIO2 demand.

It was determined that C102 is preferred to Cl2 for high-rate
disinfection of viruses because C102 is effective over more
of the pH and temperature ranges normally found in CSO.  C102
has shown viricidal activity toward representative species of
the major groups of animal and bacterial viruses.  Slight
increases in disinfection of viruses were observed when ap-
plying C102 in stages, however, this was felt to be the result
of mixing phenomena.  The application of Cl2 followed by C102
after a period of 15 or 30 seconds showed significant increases
in bacterial and viral reductions over the additive reductions
from comparable single-stage tests.  The application of C102,
followed by Cl2 after 15 or 30 seconds, was performed using
only virus as an indicator and showed less significant increa-
ses.  Since the demand substances have been shown to be dif-
ferent for Cl2 than for C102, the enhanced disinfection when
these materials were used in combination was not due to a
reduction in demand but rather an interaction between the two
disinfectants.  It was surmised that upon the oxidation of
C102 to C102~ in SCSO, free Cl2 reduced the C102~ back to CK>2,
thus prolonging the existence of the more potent disinfectant,
C102-  Such a process is particularly feasible in high-rate
disinfection in which free Cl2 is available for regeneration
of C102-  In order to achieve target levels for TC, FC and FS
bacteria and target log reduction of viruses, a concentration
of 8 mg/1 Cl2 followed by 2 mg/1 C102 in 30 seconds was the
minimum effective combination.

Concurrent measurements of ATP, bacteria and viruses in CSO
and SCSO indicated the feasibility of using ATP as a reliable
and rapid method to control the disinfection process and as a
parameter of interest.  On untreated CSO the linear correlation
of ATP with the indicator bacteria TC and FC gave coefficients
of 0.03 and 0.04, respectively.  On disinfected SCSO similar
correlations gave 0.70 and 0.76 including all samples.  At low
ATP  (^1 arbitrary unit) and bacteria (a-1000 TC/100 ml) levels,
the coefficients were 0.84 and 0.79 for TC and FC, respectively.
At high levels (ATP -\,25 arbitrary units, TC 0,1000,000 counts/
100 ml) the coefficients were 0.21 and 0.51, respectively.
                              128

-------
results and trends are to be expected since ATP is a measure
of all organisms, and the numbers and diversity of organisms
is greatest in untreated samples.  As the level of disinfec-
tion increases, the diversity of organisms decreases because
each species has a different susceptibility to disinfection.
Since the indicator bacteria are chosen as such because they
are among the most resistant fecal organisms, their predomi-
nance increases with the level of disinfection.  Therefore,
at the low bacterial levels required for discharge of CSO,
ATP should be an acceptable indicator of fecal contamination.
The results showed that an ATP level could be chosen such
that all samples which had higher ATP values had bacterial
counts higher than target levels.  All samples which had ATP
values below the specified level had bacteria counts below
the target levels.  The rate of disinfection for TC, FC and
FS was similar to the rate of decrease of ATP.  Thus, it
would seem that the use of ATP to control disinfectant addi-
tion is a valid possibility.

The first few steps were taken toward the development of an
ATP monitor for the above purpose.  The stability of ATP
reagents used in an automatic monitor was evaluated.  It was
found that the luciferin-luciferase reagent could not be
stored more than two days unless frozen.  Further investiga-
tions on this question will be conducted under subsequent
phases of this program under U.S. Environmental Protection
Agency Grant No. S-802400.

Aftergrowth tests, conducted to determine the ultimate
bacterial and viral counts that might result in the receiving
water from the discharge of untreated and disinfected CSO,
showed no measurable increases during and up to three days.
These results were felt to be more indicative of the inability
to simulate receiving water conditions in the laboratory rather
than a lack of aftergrowth.  Breakpoint chlorination was
shown to give target bacteria and viral reductions and ammonia
removal, but the costs may be a prohibitive factor.

The results of the bench-scale studies were meant to serve as
a basis to design the full-scale demonstration facilities.
However, because many questions remained unanswered, the
demonstration facilities were constructed with as much flex-
ibility as possible in order to verify the bench-scale find-
ings.   The effectiveness of Cl2 and C102 in single- and two-
stage disinfection is to be determined with particular emphasis
on mixing.   Suspended solids removal prior to disinfection is
to include micro-screening (23 micron aperture), fine-mesh
screening (71 and 105 micron aperture)  and swirl concentration.
The different methods of suspended solids removal are to be
                              129

-------
evaluated on a full-scale for simplicity of operation, effec-
tiveness of solids removal and minimization of disinfection
requirements.
                            130

-------
                         SECTION VII

                         REFERENCES

1.  Comprehensive Sewerage Study for Onondaga County. Camp,
    Dresser & McKee Consulting Engineers. Boston, Mass.
    Project Number WPC-CS-169. Dept. of Public Works, Div.
    of Drainage. & 'Sanitation, County of Onondaga, N.Y.
    November 1968. 300 p.

2.  1973 Industrial Wastes Study. O'Brien & Gere Engineers,
    Inc. Syracuse, New York. Dept.  of Public Works, Div. of
    Drainage & Sanitation, County of Onondaga, N.Y. August
    1974. 400 p.

3.  Onondaga Lake Study. Water Quality Office, U.S. Environ-
    mental Protection Agency. Washington, D.C. Project Number
    11060 FAE 4/71. Dept. of Public Works, Div. of Drainage
    & Sanitation, County of Onondaga, N.Y. April 1971. 460 p.

4.  Onondaga Lake Monitoring Program 1973. O'Brien & Gere
    Engineers, Inc. Syracuse, New York. Project Number
    115.233. Dept. of Public Works, Div. of Drainage &
    Sanitation, County of Onondaga, N.Y. December 1974. 150 p.

5.  Standard Methods for the Examination of Water and Waste-
    water, 13th Edition. New York,  American Public Health
    Association, 1971. 874 p.

6.  Callerame, J. Nitrosyl Chloride Generation of Chlorine
    Dioxide. U.S. Patent Number 375079. Chemical Generators,
    Inc., Rochester, N.Y. August 1973.

7.  Gordon, G.V., R.G. Kieffer, and D.H. Rosenblatt. Chemistry
    of Chlorine Dioxide. In: Progress in Inorganic Chemistry,
    Vol. 15, Lippard, S.J. (ed.). New York, Wiley Interscience.
    1972. p. 201-286.

8.  Palin, A.T. Determining Chlorine Dioxide and Chlorite.
    J Amer Water Works Ass. 62:483-485, August 1969.

9.  Bennett,J.E., and D.J.E. Ingram. Paramagnetic Resonance
    Spectrum of Chlorine Dioxide. Phil Mag (London). 8(1):
    109-111, January 1956.
                              131

-------
10.   Drago,  R,S.  Electron Paramagnetic Resonance Spectroscopy,
     In:  Physical Methods in Inorganic Chemistry, Drago, R.S.
     (ed.).  New York,  Reinhold,  1965.  p.  328-361.

11.   Carrington,  A.  Electron Spin Resonance Spectra of
     Aromatic Radical  and Radical-Ions.  Quart Rev (London).
     17(1) :67-69, March 1963.

12.   Berg,  G. Microbiology—Detection  and Occurrence of
     Viruses. J Water  Pollution  Control  Fed.  44:1193-98,
     June 1972.

13.   Brezinski, F.T. Estuary Water Quality and Salmonella.  In:
     Proceedings  of  the National Specialty Conference on
     Disinfection. New York, American  Society Civil Engineers,
     1970,  p. 481-493.

14.   Berg,  G.P. Integrated Approach to the Problem of Viruses
     in Water.  In: Proceedings of the  National Specialty
     Conference on Disinfection. New York, American Society
     Civil  Engineers,  1970, p. 339-364.

15.   Cohen,  J., and  H.I.  Shuval. Coliforms, Fecal Coliforms
     and Fecal Streptococci as Indicators of Water Pollution.
     Water,  Air and  Soil Pollution.  2:85-95,  1973.

16.   Ludovici,  P.P., R.A. Phillips,  and  W.S.  Jiter.   Effective-
     ness of Chlorine  in Eliminating Bacteria and Viruses from
     Tertiary Treated  Wastewater.  University of Arizona.
     (Presented at American Chemical Society Meeting, Chicago,
     August 26-31, 1973).

17.   Mara,  D.D. Coliform Counts  of Polluted Waters:  A Compari-
     son of Media and  Methods. Water Res. 7:1899-1903, 1973.

18.   Lin, S. Evaluation of Coliform Tests for Chlorinated
     Secondary Effluents. J Water Pollution Control Fed.
     45:498-506,  March 1973.

19.   McKee,  J.E., R.T.  McLaughlin, and P. Lesgourgues.
     Applications of Molecular Filter  Techniques to the
     Bacterial Assay of Sewage.  III. Effects of Physical and
     Chemical Disinfection. Sewage Works. 30 (3) : 245-252 ,
     March  1958.

20.   Dalzell, R.C.,  H.  Kinsloe,  J.j. Reid, and E. Ackerman.
     Exposure of Micro-organisms to Focussed and Unfocussed
     Sound  Fields. J Bacteriol.  73:499-503, April 1957.
                              132

-------
21.  Harvey, E.N., and A.L. Loomis. High-Speed Photomicro-
     graphy on Living Cells Subjected to Supersonic Vibra-
     tions. J. Gen Physiol. 15:147, 1932.

22.  Bunos, N., and Y. Kott. A New Approach in E_._ coli;
     Identification. In: Advances in Water Pollution Research,
     Jenkins, S.H.  (ed.). New York, Pergamon Press, 1973.
     p. 73-81.

23.  Gesteland, R.F. Isolation and Properties of E. coli
     strains mutant for RNase 1. Fed Proc.  24:293, Abstr. 905,
     1965.

24.  Gesteland, R.F. Isolation and Characterization of Ribo-
     nuclease 1 Mutants of Escherichia coli. J Mol Biol.
     16:67-84, 1966.

25.  Sinsheimer, R.L. Purification and Properties of
     Bacteriophage 0X174. J Mol Biol. 1:37-42, 1959.

26.  McElroy, W.D. Energy Source for Bioluminescence in  an
     Isolated System. Proc Nat Acad Sci U.S. 33:345, September,
     1947.

27.  Chappelle, E.W., and G.V. Levin. Use of the Firefly
     Bioluminescent Reaction for Rapid Detection and Counting
     of Bacteria. Biochem Med. 2:41-52, February 1968.

28.  Dhople, A.H., and J.H. Hanks. Quantitative Extraction of
     Adenosine Triphosphate from Cultivable and Host-Grown
     Microbes: Calculation of Adenosine Triphosphate Pools.
     Appl Microbiol. 26(3):399-403, September 1973.

29.  Heunnekens, F.M., and H.R. Whitely. Phosphoric Acid
     Anhydrides and Other Energy-Rich Compounds. In: Compara-
     tive Biochemistry, Volume I, Florkin M., and H.S. Mason
     (ed.). New York, Academic Press, 1960. p. 129.

30.  Ponnamperuma, C., C. Sagan, and R. Moriner. Synthesis of
     Adenosine Triphosphate Under Possible Primitive Earth
     Conditions. Nature (London). 199 (4890):222-226,  July
     1963.

31.  Levin, G.L., C. Chen, and G. Davis. Development of  the
     Firefly Bioluminescent Assay for the Rapid, Quantitative
     Detection of Microbial Contamination of Water. Hazleton
     Laboratories, Inc., Falls Church, Virginia. Aerospace
     Medical Research Laboratories Report Number AMAL-TR-67-71,
     July 1967. 73 p.
                              133

-------
32.   Instruction Manual  Luminescence Biometer.  E.I.  duPont
     de Nemours  and Co.  (Inc.),  Instrument Products  Division.
     Wilmington, Delaware.  Catalog Number 760018,  April 1970.
     p. 150.

33.   Berdt,  H.,  and H.J.  Linneweh.  Comparative  Quantitative
     Studies  on  the Bactericidal Activity of Chlorine and
     Chlorine Dioxide. Arch Hyg  Bacteriol (Germany).
     153(1):41-46,  1969.  fromChemAbs.  71:29528e, 1969.

34.   White,  G.C. Disinfection Practices  in the  San Francisco
     Bay Area. J Water Pollution Control Fed.  46:89-101,
     January  1974.

35.   Zillich, J.A.  Toxicity of Combined  Chlorine Residuals
     to Freshwater   Fish.  J Water Pollution Control  Fed.
     44:212,  1972.

36.   Shuval,  H.I.,  J. Cohen,  and R.  Kalodney-  Regrowth of
     Coliforms and  Fecal  Coliforms  in Chlorinated  Wastewater
     Effluent. Water Res.  7:537-546, 1973.

37.   Pressley, T.A., D.F.  Bishop,  and S.G.  Roan. Ammonia
     Removal  by  Breakpoint Chlorination.  Environ Sci  Technol.
     6:622-628,  July 1972.

38.   Automating  Manual Methods Using Technicon  AutoAnalyzer
     II System Techniques.  Technicon Industrial Systems.
     Tarrytown,  New York.  Publication Number TN1-0170-01,
     1972.  p. 34-39.

39.   Chemical Treatment  of Combined Sewer Overflows.  U.S.
     Environmental  Protection Agency, Water Quality  Office.
     Washington, D.C. Report Number 11023 FOB.  September  1970,
     p. 181.

40.   Glover,  G.E.,  and P.M.  Yatsuk.  Microstraining and Disin-
     fection  of  Combined Sewer Overflows. U.S.  Federal Water
     Quality  Administration.  Washington,  D.C.  Report  Number
     11023  EVO.  June 1970.  76 p.

41.   Diaper,  E.W.J., and G.E. Glover. Microstraining  of
     Combined Sewer Overflows. J Water Pollution Control  Fed.
     43:2101-2113,  October 1971.

42.   Glover,  G.E.,  and G.R.  Herbert. Microstraining and Dis-
     infection of Combined Sewer Overflows-Phase II.  U.S.
     Environmental  Protection Agency, Office of Research and
     Monitoring. Edison,  New Jersey. Report Number 11023  FWT.
     Philadelphia Water  Department.  January 1973.  116 p.
                             134

-------
43.   Screening/Flotation Treatment of Combined Sewer Overflows.
     U.S. Environmental Protection Agency, Office of Research
     and Monitoring. Washington, B.C. Report Number 11020 FDC.
     January 1972. 172 p.

44.   Kruse, C.W.,  Y. Hsu, A.C. Griffiths, and R.  Stringer.
     Halogen Action on Bacteria, Viruses, and Protozoa.  In:
     Proceedings of the National Specialty Conference on Dis-
     infection, American Society Civil Engineers, July 8-10,
     1970. p. 113-136.

45.   Kruse, C.W.,  K. Kawata,  V.P. Olivieri, and R.E. Longley.
     Improvement in Terminal  Disinfection of Sewage Effluents.
     Water Sewage  Works. 120:57-64, June 1973.

46.   Chambers, C.W. Chlorination for Control of Bacteria and
     Viruses in Treatment Plant Effluents. J Water Pollution
     Control Fed.  43:228-241, February 1971.

47.   Kinman, R.N., and H.A. Farber. Disinfection. J Water
     Pollution Control Fed. 44:972-976, June 1972.

48.   Davis, E.M.,  and G.B. Jackson. Disinfection. J Water
     Pollution Control Fed. 45:1088-1093, June 1973.

49.   Shuval, H.I.  Detection and Control of Enteroviruses in
     the Water Environment. In: Developments in Water Quality
     Research, Shuval, H.I. (ed.). Ann Arbor, Michigan,  Ann
     Arbor-Humphrey Science,  1970. p. 47-72.

50.   Lathrop, F.L., and O.L.  Sproul. High-Level Inactivation
     of Viruses in Wastewater by Chlorine.  J Water Pollution
     Control Fed.  41:567-575, 1969.

51.   Kelly, S., and W.W. Sanderson. Density of Enteroviruses
     in Sewage. J  Water Pollution Control Fed. 32:1269-1273,
     1960.

52.   Marais, A., E. Nupen, G. Stander, and J. Hoffman.  Compari-
     son of the Inactivation  of E_._ coli and Poliovirus  1 in
     Polluted and  Unpolluted  Waters by Chlorination. Presi-
     dential Conf. Water for  Peace, Washington, D.C. 3:670-674,
     1967.

53.   Butterfield,  C.E. Bactericidal Properties of Free  and
     Combined Available Chlorine. J Amer Water Works Ass.
     40:1305-1312, December 1948.
                              135

-------
54.   Asselineau,  Jean.  Bacterial Lipids.  San Francisco,
     Holden-Day,  1966.

55.   Krueger,  Robert G. ,  Nicholas W.  Gillham, and Joseph H.
     Coggin,  Jr.  Introduction to Microbiology.  New York,
     Maxmillan,  1973.

56.   Monroe,  D.W.,  and  B.C.  Phillips.  Chlorine  Disinfection
     in Final Settling  Basins.  In: Proceedings  of the National
     Specialty Conference on Disinfection,  American Society
     Civil Engineers, July 8-10, 1970.  p.  163-177.

57.   Bernard!, M.A., B.M.  Israel, V.P-  Olivieri,  and M.L.
     Granstrom.  Efficiency of Chlorine Dioxide  as a Bactericide,
     Appl Microbiol. 13(5):776-780,  September 1965.

58.   Bernard!, M.A., W.B.  Snow,  V.P.  Olivieri,  and B. Davidson.
     Kinetics and Mechanism of Bacterial  Disinfection by
     Chlorine Dioxide.  Appl Microbiol.  15 (2) : 257-265 , March
     1967.

59.   Ridenour, G.M., and  E.H.  Armbruster.  Bactericidal Effect
     of Chlorine  Dioxide.  J Amer Water Works Ass. 41:537-550,
     June 1949.

60.   Ridenour, G.M., and  R.S.  Ingols.  Bactericidal Properties
     of Chlorine  Dioxide.  J Amer Water Works Ass. 39:561-567,
     1947.

61.   Ridenour, G.M., and  R.S.  Ingols.  Chemical  Properties of
     C1O2 in  Water Treatment.  J  Amer  Water  Works  Ass. 40:1207-
     1227, 1958.

62.   Reid, L.C.,  Jr. Effect of Temperature  on Chlorine Disin-
     fection  of Water.  Diss Abstr B.  33:3680-3681, 1973.

63.   Bernardi, M.A., W.B.  Snow,  and V.P.  Olivieri. Chlorine
     Dioxide  Disinfection Temperature Effects.  J  Appl Bacter-
     iol. 30:159-167, January 1967.

64.   Brungs,  W.A.  Effects of Residual Chlorine  on Aquatic
     Life. J  Water Pollution Control  Fed.  45:2180-2193,
     October  1973.

65.   Camp, T.R.,  and R.H.  Culver. Disinfection: Objectives
     and Standards.  Water Sewage Works. R147-151, 1962.

66.   Musil, J.,  z.  Knotek,  J.  Chalupa,  and P. Schmidt. Toxi-
     cology Aspects  of  C102 Application for the Treatment of
     Water Containing Phenols.  Sb. Vysoke  Skoly Chem Technol
     Froze Technol Vody.  8:327-346,  1965.  Chem Abstr. 65:1960d.
                              136

-------
67-  Morris, J.D. Aspects of the Quantitative Assessment of
     Germicidal Efficiency. Harvard University (Presented at
     American Chemical Society Meeting, Chicago,  August 26-31,
     1973).

68.  Baker,  R.J. Engineering Considerations in Disinfection.
     In: Proceedings of the National Specialty Conference on
     Disinfection, American Society Civil Engineers.  New York,
     July 8-10, 1970. p. 683-697.

69.  Lewis,  R.N. Proceedings of the First Microbiology Seminar
     on Standardization. U.S. Environmental Protection Agency-
     Washington, D.C. Publication Number R4-73-022. March
     1973. p. 137.

70.  Van Dyke, K., R. Stitzel, T. McClellan, and  C.
     Szutkiewicz. Automated Analysis of ATP; Its  Application
     to On-Line Continuous-Flow Incubations and Measurement
     of ATP-Coupled Enzyme Systems. In: Advances  in Automated
     Chemistry, Volume I. Tarrytown, New York, Technicon
     Corporation. 1969. p. 47-53.

71.  Nutrient Removal Using Existing Combined Sewer Overflow
     Treatment Facilities. O'Brien & Gere Engineers,  Inc.,
     Syracuse, New York-Project Number 115.238, Dept.  of
     Public  Works, Div. of Drainage & Sanitation, County of
     Onondaga.

72.  Smisson, B.S. Design, Construction and Performance of
     Vortex  Overflow; Symposium on Storm Sewage Overflows.
     Institution of Civil Engineers, 1967. p. 99.

73.  Swirl Concentrator As a Combined Sewer Overflow Regulator
     Facility. U.S. Environmental Protection Agency,  Office
     of Research and Monitoring. Washington, D.C. Report
     Number  EPA-R2-72-008. September 1972. 179 p.

74.  Field,  R. Dual Functioning Swirl Combined Sewer  Overflow
     Regulator/Concentrator. U.S. Environmental Protection
     Agency, National Environmental Research Center,
     Cincinnati, Ohio. Report Number EPA-670/2-73-059.
     September 1973. 49 p.

75.  Sullivan, R.H., M.M. Cohn, J.E. Ure, F.E. Parkinson, and
     G. Galiana. Relationship Between Diameter And Height For
     The Design of A Swirl Concentrator As A Combined  Sewer
     Overflow Regulator. U.S. Environmental Protection  Agency,
     Research Center, Cincinnati, Ohio. Report Number  EPA-
     670/2-74-039. July 1974. 44 p.
                             137

-------
76.   Sullivan,  R.H.,  M.M.  Cohn,  J.E.  Ure,  and F.E.  Parkinson.
     Swirl Concentrator As A Grit Separator Device.  U.S.
     Environmental Protection Agency,  National Environmental
     Research Center,  Cincinnati, Ohio.  Report Number EPA-
     670/2-74-026. June 1974.  93 p.

77.   High-Rate  Filtration  of Combined  Sewer Overflows. U.S.
     Environmental Protection Agency,  Office of Research and
     Monitoring.  Washington, B.C. Report Number 11023 EYI 04/72,
     April 1972.  339  p.

78.   Demonstration of Rotary Screening for Combined Sewer
     Overflows.  U.S.  Environmental Protection Agency, Office
     of Research and  Monitoring.  Washington, D.C.  Report
     Number 11023 FDD 07/71. July 1971.  55 p.

79.   Hypochlorite Generator for  Treatment  of Combined Sewer
     Overflows.  U.S.  Environmental Protection Agency, Office
     of Research and  Monitoring.  Washington, D.C.  Report
     Number 11023 DAA 03/72. March 1972. 89 p.

80.   Liskowitz,  J.W.,  and  G.J. Franey.  Measurement of Suspended
     Solids Concentrations in Sewage by a  Depolarization Method.
     Env Sci Tech. 6(l):43-47, January 1972.

81.   Storm Water Management Model,  Volume  I-Final  Report. U.S.
     Environmental Protection Agency,  Water Quality Office.
     Washington,  D.C.  Report Number 11024  DOC 07/71.  July 1971.
     352 p.

82.   Storm Water Management Model,  Volume  II-Verification and
     Testing.  U.S. Environmental Protection Agency,  Water
     Quality Office.  Washington,  D.C.  Report Number 11024 DOC
     08/71. August 1971. 172 p.

83.   Storm Water Management- Model,  Volume  Ill-User's Manual.
     U.S.  Environmental  Protection Agency,  Water Quality Office.
     Washington,  D.C.  Report Number 11024  DOC 09/71.  September
     1971.  359  p.

84.   Storm Water Management Model,  Volume  IV-Program Listing.
     U.S.  Environmental  Protection Agency,  Water Quality Office,
     Washington,  D.C.  Report Number 11024  DOC 10/71.  October
     1971.  249  p.
                              138

-------
                        SECTION VIII

                    LIST OF PUBLICATIONS

1.    Murphy, C. B., Jr., and E. C. Tifft.  A Novel Application
     of Electron Spin Resonance;  The Analysis of Chlorine
     Dioxide in Waste-water.  O'Brien & Gere Engineers, Inc.
     (Presented at the 5th Northeast Regional Meeting,
     American Chemical Society, Rochester, N. Y., October
     14-17, 1974.)  13 p.

2.    McVea, J. L.  Virus Inactivation of Chlorine Dioxide
     and Its Application to Stormwater Overflow.  Syracuse
     University, Master's Thesis.  November 1972, 90 p.

3.    Smith, J. E., and J. L. McVea.  Virus Inactivation by
     Chlorine Dioxide and Its Application to Stormwater
     Overflow.  Syracuse University.  (Presented at the 166th
     American Chemical Society National Meeting, Chicago,
     August 26-31, 1973.)  9 p.

4.    Smith, J. E., and J. L. McVea.  Virus Inactivation by
     Chlorine Dioxide and Its Application to Stormwater
     Overflow.  Water Res.  (Submitted for publication 1974.)

5.    Tifft, E. C., S. L. Richardson, P.  E. Moffa, and R. Field.
     Disinfection Techniques For Point-Source Treatment of
     Combined Sewer Overflows.  O'Brien & Gere Engineers, Inc.
     (Presented at the 1974 Spring Meeting of the New York
     State Water Pollution Control Association.  Kiamisha
     Lake, N. Y., June 21-23,  1974.)  11 p.
                             139

-------
                         SECTION IX

                  LIST OF ABBREVIATIONS
CSO
SCSO
C102
Cl~
cio-
C102~
cio3-
HOC1
CDS
Dosage
Br2
Cd
Cr
Cu
Fe
Hg
Ni
Pb
Zn
03
UV
esr
g
G
G'
GHZ
K
fg
M
biphasic
fps
cfs
I/sec
mg/1
ug/1
rpm
mgd
l/min/m2
Combined Sewer Overflow
Simulated Combined Sewer Overflow
Chlorine
Chlorine Dioxide
Chloride Ion
Hypochlorite Ion
Chlorite Ion
Chlorate Ion
Hypochlorous Acid
Chlorine Demanding Substances
Quantity sufficient to yield an initial
concentration (often used interchangeably
with Dose)
Bromine
Cadmium
Chromium
Copper
Iron
Mercury
Nickel
Lead
Zinc
Ozone
Ultraviolet
Electron Spin Resonance
Magnetogyroscopic Ratio
Gauss
Velocity Gradient
Megahertz
Kilocycles per second
                  10
                    -15
gram)
femtogram (1 fg1 =
Molar
two-phase
feet per second
cubic feet per second
liters per second
milligram per liter
microgram per liter
revolutions per minute
million gallons per day
liters per minute per square meter
(application rate)
                           140

-------
 e
TS
TSS
VS
vss
Sett-S
TOC
COD
T-ALK
T-IP
NH3N
TKN
ORGN
N02N
NQ3N
N02N03
N
OT
P
Thio
DPD
MF
MPN
TC
FC
FS

E. coli
ATP
MOPS
PFU
Bacter iophage ,
phage
f2
0X174
polio-1
HSV-L2
NDV
Vaccinia
Coxsackie B-3
YFV
ECHO- 7
GKN
TPB
CsCl
MgS04
EDTA
HEp-2
Density
Total  Solids
Total  Suspended Solids
Volatile  Solids
Volatile  Suspended Solids
Settleable Solids
Total  Organic Carbon
Chemical  Oxygen Demand
Five-Day  Biochemical Oxygen Demand
Total  Alkalinity in mg/1 Calcium Carbonate
Total  Hydrolyzable and Orthophosphate in mg/1 P
Ammonia in mg/1 N
Total  Kjeldahl Nitrogen in mg/1 N
Organic Kjeldahl Nitrogen in mg/1 N
Nitrite in mg/1 N
Nitrate in mg/1 N
Nitrite plus Nitrate in mg/1 N
Nitrogen
Orthotolidine
Phosphorous
Sodium Thiosulfate (Na2S203•5H20)
N, N-Diethyl-p-phenylenediamine
Membrane  Filter
Most Probable Number
Total  Coliform Bacteria, T Coli,
Fecal  Coliform Bacteria, F Coli,
Fecal  Streptococcus Bacteria, F.
Strep
Escherichia coli
Adenosine Triphosphate
Morpholinopropane Sulfanilic Acid
plaque forming units

Bacterial Virus
Bacteriophage f2
Bacteriophage 0X174
Poliovirus Sabine Type 1
Herpes Simplex Virus-L2
Newcastle Disease Virus
Vaccinia Virus
Coxsackie Virus B-3
Yellow Fever Virus
Echovirus-7, Enteric Cytopathogenic Human Orphan
Glucose Pottai~ium Nutrient
Tryptose Phosphate Broth
Cesium Chloride
Magnesium Sulfate
Ethylene Diamine Tetraacetic Acie
Human Epithelial Cell Line
Total Coli
Fecal Coli
Strep, Fecal
                              141

-------
gal            gallon
U              micron
hr             hour
mon            month
yr             year
mgd            million gallons per day
ft             feet
v              velocity
HC1            Hydrochloric Acid
                            142

-------
                          SECTION X

                         APPENDICES

                                                       Page

A.  Characterization of Combined Sewer Overflows       144

B.  Tabulation of Data for Single-stage Disinfection
    Studies                                            153

C.  Derivation of Dilution Factors and Residence Times
    for Aftergrowth Studies                            164

D.  Tabulation of Data for Two-Stage Disinfection
    Studies                                            166

E.  Tabulation of Data for Aftergrowth Studies         171
                            143

-------
                         APPENDIX A

        CHARACTERIZATION OF COMBINED SEWER OVERFLOWS

The data is a partial listing of the information obtained in
a characterization study of CSO in the City of Syracuse.
The purpose of their inclusion is to demonstrate the
variable nature of CSO.   The key to identification of the
computer printouts is as follows:

     STNO - Storm Number-approximately thirty storms were
            sampled.

     PLOC - Primary Location-several different overflow
            sites were sampled.  The two included here are

            1 - Maltbie  Street
            2 - Rowland  Street

     SLOG - Secondary Location-different locations within
            an overflow  site, intended primarily for use in
            the demonstration of treatment facilities.

            1 - Untreated Overflow
            2 - Receiving Water Upstream of Overflow
            3 - Receiving Water Downstream of Overflow
           10 - Lithium  Solution used for flow measurement

     TYPE -

            0 - Grab
            1 - Sequential or simple composite
            N - Composite - Number tells actual number of
                sequentials included in sample

     SQNO - Sequence Number-gives the order in which
            samples were taken during a storm; 0 indicates
            a time of pre-overflow rain

     SAMP - Sample Number-an arbitrary number assigned to aid
            in analytical bookkeeping

     FLOW - Flowrate in  mgd

     RAIN - Accumulated  rain in inches

     RAININT - Rain intensity during the previous interval in
               inches per hour
                             144

-------
            Table  A-l.
                                                                                           REPORT  PRINTED   2/28/73  PAGE    7  -  1
                                             '»YKAC1)SE COHiilNI-ri SEUEM. UVEI'.PI.OUS  PRELIMINARY  PHASf
            STNO  PLOC  SLCC  TVPF.  SQNO  SAMP
                                                  DATE
                                                          TIW
                                                                  FLOU   RAIN  RAININ1
                                                                                         PH
                                                                                                T-CQU
                                                                                                           TDC   CCD
TSS-,   -VSS
tft.
Ln
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1
1
1
1
1
I
1
I
I
1
1
1
I
1
1
1
1
1
1
1
1
, , ......
1
1
1
1
1
1
1
1
1
1
1
1.
1
1
1.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
I
I
1
1
1
1
1
1
1
1
1
I
1
1
I
1
1
1
0
0
0
0
0
0
1
1
I
1
1
1
I
1
1
1
1
I
I
1
\
t
1
1
1
1
1
I
1
1
1
1
1,
1
1.
I
1
0
0
0
0
0
0
0
1
2
3
4
5
6
7
P
9
10
11
1?
13
l'i
I1)
16
17
1 «
1')
20
21
2,n
?!
I'd
?7
,'P
30
3,!
1.377
1 373
UdO
1 3 4 ,!
233
2:>'5
•'57
75!)
20'.
26 5
E
?7Q
2T>
?.r,
"'74
HI ft
307
ni
M <>
6/30/73
6/30/72
6/30/72
6/.10/7?.
6/30/72
6/30/72
6/30/72
$?/'i!
2'!?/"
6/3C/7'.'
6AW77
ft/'iC/?;'
6/30/72
6/-30/7?
'./.ID/72
n/ J /72
6/10/77
I1-/.!:1/ 7">
;>/M./I?
df >''/ 7'>
'i/ v/7;>
<,/ ';<•/ v
Zl'i
no
245
300
j :i o
.345
4 0 "
415
430
500
600
630
fi't1?
700
71r>
730
74$
300
' ft ) 0
90'
915
930
1030
IJJ'I
] ',3')
17 VI


14.900
11.130
l'i!900
14.900
14. '700
le!i?o
3.i! 70
2.470
1. 710
1.3'JO
0.9 ,U
0..170
0. HO
0 . * •' 0
n.oo
O.'l ?0
O.fi'iO
U. 3 70
11.? )0
•J.I 0
0.
0.
0.
0.
0.
0.
0.
(1.
0.
0.
0.
0.
0.
0.
1.
1.
1.
1.
1.
1.
I.
1,
u
1.
1 .
I.
1 .
I.
1.
1.
1,
1.
I.
1 .
1 .
0.)
05
1'J
15
70
25
30
35
45
50
5*
70
PO
90
00
10
12
15
15
15
15
ie)
15
15
15
15
15
13
30
30
30
V
3f
0 . 2 0
0.2 J
0.7,0
0.70
0.20
0.20
C.40
0.2o
0.20
0.60
0.40
0.4.)
0.4U
0.20
0 . ? 0
0.08
0.12
O.Oi
0.0(j
0,00
0.00
o.oc
0.00
0.0')
C . 0 0
C.Ou
C.Oil
r.oo
O.O'l
O.I *
0.1)0
0.00
. 0.00
C.'l'J
0.0.)
C; . 0 '1
ll.Od


6.9
7.2
7.4
7.5
7.6
7.5
7.7
7.7
7.6
7.7
7.7
7.7
7.1 --
7.4
7.5
7. 1
7.5
7.0
r. i
7.3
7.5
7.6
7.P
7.1
7. 3
7.7
i. ;)
•I..'
7..1.


4'iOO(.0.
2000(,'0.
340000.
33000o!
800.10.
6300!'o!
1600 !C.
.'OO'.'O.
70001T.
S630i;o.
3100L.O.
5600.:0.
61000o!
9SOO '0.
4300 V'!
4Ui.(!.
IOC( ^,
'(.'III ().


14.
9.
6.
7.
6.
4.
U.
6.
12.
14.
20.
lo.
14.
12.
15.
21.
33.
15.
21.
17.
23.
21.
1').
24.
4?..
36.
?7
"Jf. 	


?7.
10.
1C.
5.
4.
7.
5.
5.
9.
fl.
14.
15.
13. *
15.
is!
36.
32,
»!
i:
30.
3?.
4'^.

f
90.0
60.0
75.0
110.0
B'llo
75.0
75.0
40.0
375.0
160.0
7!iO.O
540.0
310.0
ZT'Jff.O
lin.o
170.0
1 T',1 . 0
K5.C
2^0.0
18*3.0
17670
1.36.0
17.0
T3.0
3 04..")


?0 0
24.0
5.0
0..)
0.0
0.0
0.0
0.0
15.0
iO.O
21.0
P.0.0
70.0
4'j.O
40.0
30.0
0.0
U.P
i.O
0.0
0.0
•:i.o
o.o
/ rt . 0
8.0
12.0
4..1
C.O
7 k) . 0

-------
Table A-2.
                                                        115201517
                                                                               REPORT PRINTED  2/28/73 PAGE
                                                                                                              7-2
                                 SYF'.ACUSE COMBINED SEl.EP. rvERF-LOUS PRELIMINARY PHASE
STNO  PLOC  SLOC  TVPtf  SONQ  SAMP
                                      DATf
                                              TIMI:
                                                     TKN
                                                           NH3N  HfJGN  N07N03
                                                                                N07M  NOT-!
                                                                                             TALK
                                                                                                    T-IP
                                                                                                            CL
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
•t
4
4
4
4
4
4
4
4
4
4
I
I
1
1
1
1
1
1
1
1
1
I
1
I
1
1
I
I
1
1
1
1
I
I
1
I
I
1
1
I
1
I
]
I
1
I
1
1
1
1
I
1
1
1
1
1
1
I
1
1
1
1
1
1
1
1
1
1
I
1
1
1
I
1
1
1
1
1
1
1
I
]
1
1
1
1
I
1
C
0
C
0
0
n
0
1
1
1
1
1
1
1
I
1
1
1
1
1
1
]
1
1
1
1
1
1
1
1
1
]
1
I
I
1
1
1
0
0
0
0
0
0
C
1
2
3
4
t
6
7
9
10
11
12
13
14
15
16
17
IP
19
20
.?!
22
23
24
?5
26
27
.V'J
'JO
31
J7
1377
1378
1379
1380
1311
I.3B2
13J3
?5L
252
25 1
?54
;'55
256
757
25a
259
260
761
?6?
26!
264
261
266
267
263
?6 J
270
?71
272
273
^>74
-il':'»
.106
307
30
30 1
310
31'
-.12
6/30/72
6/30/72
6/3C/72
6/3C/72
6/30/7?
6/3C//'2
6/3C/72
'./3C/7?
-/ :s./12
6/30/72
fc/3P/72
A/30//7
6/}
ft/ V/7;1
6/30/72
CV3J/72
•S/ (:V7?
fJ/3iV72
'>/ 10/72
6/30/72
V3C/77.
6/30/72
6/30/72
6/3.'V72
6/3.'./7->
ft/30/72
fi/3C/7?.
6/30/72
6/ir/7;>
6/30/72
6/3J/72
-,/3T/72
'/'..,/ I Z
6/.30/?2
. / VV7?
i/":r /72
f/ i-V7?
f./ir/7?
•./TV.'.1.
2 on
215
23^
.?45
300
315
330
345
400
415
430
445
son
51 r>
530
545
600
614
630
645
70H
7Ti
730
745
800
,315
330
845
90"
9lri
930
103 '.
113fl
123n
I 33'!
1 4 3 ' ;
1 55>i
1MO
1M<.


0.9
0.9
1.9
0.9
0.4
1.7
0.4
C. 5
0.7
0.7
1.7
I. J
1.4
£ * 1
2.^
3.7
2. 7
2.7
4.?
4.0
4.3
3.3
3.9
4.8
5.3
6.11
5.7
7.7
7.V
•5 . -'-.
5.'i


0. JO
•!.25
0.25
0.15
0.1 i>
0.15
0. 30
0.15
0.35
0.46
0. 35
0. J5
0.50
O.f;0
1.40
1.6S
1.2H
2.30
1.90
1.55
2.50
1.05
O.OS
1.40
l.'>5
I.,'. 5
0.30
0.35
0.91?
0.75
0.75


0.7
0.6
1.6
0.7
0.2
1.0
0.1
0.7
J.3
0.2
0.8
0.9
0.9
1.6
1.5
1.5
1.4
0.4
2.3
2.4
l.C
2.2
3.C
3.4
3.C
5.1
5.4
7.3
6.2
4.*
4.S


0.01
O.Q3
0.08
0.09
0.07
0.06
0.06
C.07
0.06
o.oa
(J.IO
0.21
Cull
0.18
0.26
:>. 19
0 . 2 5
0.25
0.31
0.35
0.31
0.33
0.32
0.33
0.35
1.47
C.95
."• . 'i I
!"..C, 4
O.M)
O.-'O


0.013
0.026
0.076
0.016
0.014
0.016
0.021
0. )21
0.071
0.036
0.031
0.031
0.036
0. J44
0.060
C . 04 P
0.073
0.094
0.060
0.060
0.057
0.063
(1.081
0.087
0.063
0.^50
0.266
0. 106
0. 11 !
0.097
f : . ; rf f-j


0.00
0.00
0.05
0.06
0.06
0.04
0.04
0.05
0.04
0.04
0.07
0.18
0.07
0.14
0.20
0.15
0.1V
0.16
0.25
0.2=5
n.?->
0.27
0.24
0.24
0.2-
1.2?
0.6S
0.',"
0.5.^
0.5!!
U.51


40.
41.
41.
41.
41.
42.
49.
53.
53.
57.
R3.
86.
01.
116.
135.
149.
163.
194.
19C.
192.
192.
157.
168.
180.
192.
197.
185.
193.
212.
198.
203.


0.19
0.13
0.12
0.10
0.09
0.11
0.09
0.10
0.09
0.1R
C.ll
0.37.
0.33
0.50
0.49
0.44
0.25
0.60
2.49
0.65
1.40
0.72
1.21
0.70
C.47
1.74
0. 35
0.76
1.31
0.7ft
0.79


8.
10.
10.
7.
6.
7.
4.
10.
11.
16.
20.
23.
25.
32.
36.
44.
46.
52.
54.
53.
53.
46.
5?.
57.
5-3.
64.
54.
66.
69.
77.
Ti.

-------
Table  A-3.
                                                       115201517
                                                                              REPORT  PRINTED  2/28/73 PAGE
                                                                                                            9-1
                                 SYRACUSE COHaiNED SEWER OVERFLOWS PRELIMINARY  PHASE
 srr40  PLOC  sure  TYPT-  soro
                                      DATF
                                                     FLOW
                                                            RAIN  RAININT
                                                                            PH
                                                                                   T-CULt
                                                                                             TOG   COD
                                                                                                           TSS
                                                                                                                  vss
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
3
3
T
1
3
3
3
3
3
-?
T
3
3
3
3
3
J
J
3
3
3
3
7
j
3
3
3
3
^
3
3
^
3
3
T
3
3
3
3
1
1
1
I
1
1
1
1
1
1
1
1
1
]
1
1
1
1
1
1
1
1
1
1
1
1
J
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
]
1
1
1
I
1
1
1
]
1
I
1
1
1
1
1
1
1
I
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3
4
5
6
7
n
9
in
11
12
13
14
15
16
17
1°
10
"MJ
21
22
.-3
24
i c^
,'6
21
S°.
2 1
3il
31
32
33
14
1-i
"V,
'7
"i^
3T
41!
41
280
?n]
2 fi 1
2 Lii
'->R4
r (i 5
2:i6
?R7
?e-:
••3')
?-90
?in
29 2
29 <
2'34
29'-
106
Vy7
"9>1
29 )
300
301
'. 1?
33 1
334
33-1
336
3W
338
33 )
Mt-:~.
"t \
?'l J
14 1
i4'i
1'l'j
146
347
)4.!
6/30/72
6/30/7?
j/Jf/72
6/3C/7?
'./•J.fl/7,">
6/ )n/72
6/3C/ /?
6/30/72
6/ *C/72
6/TV7?
6/3I-/7?
L/T,/ f2
6/3C/7?
6/1C/7'
6/3C/72
4/ 30/7?
6/3C/72
••••/JC/7.?
6/ iT/72
6/W72
i,/ JO/72
6/3'5/7?
6/3'/ ?2
6/31 /7?
6/K. /72
•'. / 3 0 / 7 2
,',/!C/72
',/ !"./7J
(./1(. /7?
;./30/7?
6/ 1C/72
-./ 1' /7P
6 / 1 r / 7 .'
6/3T/7?.
• / 3 r. / / '
' / '. i ' / 7 ,">
7 / 1 / f ?
7/ 1/7?
7 / 1 / / ?
446
son
515
530
545
600
615
630
645
700
715
730
745
noo
.115
(130
S45
90n
915
910
94^
1000
110"
] 200
1300
1400
I SOU
1601
1700
UOO
190 )
'00')
MO''.
2?or
:'3.i '
?4'J 1
KM
70'
301
11.530
10.1150
10.100
ti.OI'.O
10. If."
10.ll\)
8.9«0
8 . o •: o
6. 730
5.780
5.730
'i.C'jO
4.040
3. 3;.o
3.360
'.7iO
3. 110
3.110
3.110
3.) 10
1.110
3. 110
?.OVl
l.BCO
l.SCO
1 .300
1. 700
1.700
i ..icn
1.700
1 .I'OO
1 .7-.')
1.910
1. 750
1. J- "l
1.7 •')
l.'it-O
1 . 7-3(1
1 . S >0
0.95
l.OS
1.05
1 .05
1.10
1.10
1.10
1.10
1. It!
1.10
l.Ui
1.1"
1.10
1 .10
1. 10
1.15
1.15
1.15
1 .15
1.15
1.15
1.15
1.15
1.15
1.15
1.15
1.20
i.?n
1.20
1.20
1.20
1.20
1.20
1.20
1.20
1.20
1.20
1 .20
1.21
c.oo
0.40
o.oc
O.Ol!
0.20
O.QC
0.00
C.O.I
O.no
0.00
0.00
o.ou
0.00
0.00
0.00
0.20
0.00
0.00
o.oo
0.00
0.00
0.00
O.Ou
0.00
0.00
o.oc
0.05
0 . O'O
c.oc
0.00
o.oo
o.ou
C.O 1
c.oo
0.00
0.00
0.00
r.oo
o.oo
7.6
7.6
7.7
7.9
7.4
7.7
7.9
M. 1
b.O
7.4
7.1
7.9
7.9
'{.0
7.6
7.5
7.6
7.2
7.7
7.M
7.9
7.3
7.-1
6.9
7.5
7.6
7.6
7.5
7.5
7.6
7.H
7.9
7.7
7.6
7.7
a.o
H.O
7.C
670000.
500000.
1C-JOOOO.
8100PO.
130000.
leoot.i).
14001 0.
700CO.
1800CO.
1000'JT.
70000.
15 oo no.
40000.
22001;.').
240000.
180CO.
120CO.
19000.
900n.
,'5000.
70CO.
i2CO';o.
1 20000.
90000.
1100'JO.
I30oi;o.
60000.
vgoo^o.
3 10300.
260000.
200000.
2 U1000 .
330000.
2 60'! 00.
1900CO.
400000.
-; loooo.
10.
B.
10.
10.
12.
12^.
11.
10.
11.
13.
11.
13.
14.
13.
12.
16.
15.
14.
12.
13.
12.
14.
17.
1«.
18.
23.
22.
21.
24.
26.
23.
2b.
26.
25.
1^.
16.
12.
20.
5.
3.
3.
4.
5.
4.
3.
7.
4.
8.
8.
11.
12.
11.
13.
15.
18.
12.
18.
16.
17.
19.
17.
17.
15.
19.
16.
17.
20.
17.
19.
21.
18.
17.
14.
12.
12.
115.0
135.0
125.0
120.0
125.0
182.0
110.0
175.0
185.0
165.0
145.0
140.0
135.0
115.0
110. 0
105.0
125.0
120.0
100.0
140.0
145.0
33.0
?3.0
20.0
20.0
15.0
27.0
19.0
13.0
?7.0
21.0
32.0
10.0
10.0
17.0
14.0
12.0
0.0
5.0
0.0
0,0
0.0
120.0
0.0
45.0
40.0
40.0
0.0
35.0
25.0
30.0
35.0
35.0
40.0
35.0
25.0
85.0
55.0
13.0
2ij.O
18.0
7.0
10.0
20.0
10.0
11.0
13.0
12.0
11.0
3.0
5.0
11.0
4.0
5.0

-------
                Table  A-4.
                                                                        115201517
                                                                                               REPORT PRINTED  2/28/73 PAGE
                                                                                                                              9-2
                                                 SYRACUSE COMB 1 NED SCWE1! uVEitFLDHS PRTLIMINARY PHASl"
                STNO  PLOC  SLOG  TYPfi  SQi\0
                                                      DATE
                                                              T1 V f
                                                                     TKN
                                                                                 IJRGN
                                                                                                N02N  N03t-i
                                                                                                             TALK
                                                                                                                    T-tP
                                                                                                                            CL
00
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
3
3
3
3
n
3
3
3
3
3
3
3
3
3
3
3
3
3
3
1
3
3
3
3
3
3
5
3
3
3
3
1
3
3
3
3
3
3
)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
I
1
1
1
1
I
1
1
1
1
1
1
I
1
1
I
I
1
1
1
1
1
I
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
I
1
1
1
1
1
I
I
1
1
1
1
1
I
t
I
1
3
i,
5
6
7
b
9
in
11
12
13
14
15
16
i 7
19
19
?0
21
??
?3
24
25
76
27
7 ^
2')
30
!1
32
33
34
35
36
37
T '
39
£<-
41
280
?« I
?fl?
2,'. 3
7P4
794
?S6
2F.7
i1 3 .•)
.? a 9
290
291
2-J2
293
,-•'94
2 95
796
297
vy,:
79')
300
3.11
33''
33 i
334
33 5
336
337
33a
339
340
341
.J42
34 3
144
340
Ui6
?4 7
< '. -i
6/30/72

o/TJ/72
7/ I//:'
7/ !/T'
?/ 1/7'
445
500
51-1
530
5 4 5
600
615
630
645
700
71=5
730
745
300
nis
830
845
90^
91b
930
945
1000
lion
1200
130J
140"
1500
1600
1 70.1
1.-100
1900
?'i 1C
2100
2200
v-jnr,
?4( .;
1' ''
T 1 i
3'
1.2
1.2
1.1
1.?
1 .0
1 .4
1.0
1 .3
1.9
2.0
Z.1!
2.5
3.5
4.6
5.2
4.5
3.9
4.2
4.0
4. a
4.4
4.4
4.7
4.1
3.9
3.4
3.'i
4.0
4.1
?.:j
2.3
2. 7
3 . ~i
2.'
3. .
'.4
1. i
0.25
0.25
0.30
0. i:;
0 . ••' 8
0 . ? 5
0.25
0 . 4 !)
0.55
0.65
O.'iS
1.0 -j
2.02
3..!3
3.52
2.30
2.70
2.70
2.62
3.70
3.12
3.02
2. '8
l.')3
1.43
1 .30
l.:?6
1.15
1 . ft r-
1 .65
1.68
2.?u
2.5C
1.30
1 .7fi
?.45
1 . i: 5
•;.7'j
0.9
0.9
o.a
0.9
0.7
1.1
0.7
•>.9
1.3
1.3
1 .6
1.4
1.5
0.9
1.7
2.2
1.?
1.5
1.4
1.1
1.3
1.4
2.4
2.2
?.5
2.1
2.1
2.13
2.4
1.1
0.6
•1. 5
1.1
0.9
1 .0
0.9
1 .7
'.' • W'
0.11
0.11
0.11
C.14
(1.15
0. 14
0.17
n.l8
0.21
0.22
0.23
C.26
0.28
0.30
0.33
0.34
0.34
0.39
0.38
0.39
0.46
0.52
0.31
0.34
0.34
0.34
U. 3'J
o.3a
0.35
0.5,3
0.41
0.3'5
0.36
0.34
''. 35
(..34
'.34
r.. 14
0.016
O.:ll4
0.016
0.016
0.01 9
0.019
0.019
0.0?1
0.021
0.021
0.026
0.029
0.0,79
0. i31
0.031
0.031
0.0 6 5
0 . D 5 0
0.060
O.GflS
0. 117
0.105
0. 14')
0.036
0.:336
0.036
0.054
0.054
n.r.31
0. 104
0.052
0.^4 7
0.700
0.041
0. !31
o . v,; 1
O.'TM'
l).0?0
n.O-')
0.10
0.09
0.12
0.13
0.12
0.1S
0.16
O.H
0.20
0.2''
0.23
0 . 2 'j
0.27
0.3'i
0.31
0.27
0.34
0.32
0.30
0.34
0.41
0.27
0.30
0.30
0.3J
0.33
0.33
0.32
0.40
0.36
0.30
0.16
0.30
0.3;'
0.31
0.3"
•)-32
125.
126.
149.
159.
150.


274.
785.
232.
230.
?34.
236.
?38.
247.
246.
253.
255.
253.
255.
259.
257.
259.
259.
259.
257.
257.
249.
247.
261.
?cS .
,'60.
255.
0.12
0.11
0.13
0.11
0.12
0.12
0. 13
C.16
0.19
0.22
0.?9
0.27
0.53
0.53
0.57
0.51
O.i>l
0. 73
0.54
l.Cfl
0.95
1.26
1.19
1.22
0.66
1.06
1.77
0.94
0.53
0.48
0.54
0.50
0.37
0.52
0.73
0.39
0.19
J.29
23.
27.
33.
37.
32.
36.

78.
82.
81.
85.
85.
85.
86.
SB.
83.
83.
83. •
89.
92.
9?.
92.
93.
92.
93.
88.
9 J.
89.
89.
92.
93.
92.
91.

-------
Table  A-5.
                                                       115201517
                                                                             REPORT PKINTED  2/28/73 PAGE  11 - 1
                                SYPACUSE COMBINED SEhPR L'VEftFLOWS  PRELIMINARY  PHAStf
STNO  PLOC  SLOC   TYPE   SO;tO   SAM0
                                     DATt
                                             T IM. l;
                                                            rt.MN  KA1NINT
                                                                                  T-CCLI
                                                                                             TOG
                                                                                                   COO
                                                                                                           TSS
                                                                                                                  VSS
5
5
5
5
'j
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
1
I
I
1
]
1
I
1
1
I
I
1
1
1
1
I
1
I
1
1
1
1
1
1
1
1
i
I
1
1
1
1
1
1
1
1
1
1
1
2
3
10
0
0
1
1
1
1
1
1
1
I
1
1
I
I
1
1
1
15
1
1
0
0
1
2
3
A
5
6
7
8
q
10
11
12
13
14
15
1
1
1
1
13.;4
1385
35)
360
361
36?
363
364
3ob
J66
367
3<>8
36-7
37"
371
37 '
373
3tl3
33'i
3 '1 -j
11
7/
7/
r/
7/
7/
7/
7/
7/
7/
7/
?/
7/
7/
7/
7/
7/
7/
7/
7/
7/
7/
7/72
7/72
7/72
7/72
7/72
7/72
7/72
7/72
7/7?
7/72
7/72
7/72
7/7?-
7/7?
7/72
7/72
7/72
7/7?
7/72
7/72
7/72
1530
1545
160 ;
1615
163P
1645
17CK:
1715
1730
174S
1HOO
1R15
1 330
1 H45
190i!
1915
193-1
193
193C
1930
1930

11.600
U.70U
4.970
2.900
i.sto
0.9 '50
0.760
0.390
0.?YO
0.220
0 . 1 ' C
0.140
0. l?0
O.CllQ
O.O.TO

150.000
0.00
0.20
0.40
0.40
0.40
',.40
C.40
0.40
0.4!)
0.40
0.40
0 . 4 0
0.40
0.40
0.40
0.40
0.40
0.4T

(J . H 0
0.:J(1
0.01'
0.00
O.OU
O.OU
C.or.
0.00
c.ou
0.0.1
o.oo
0.00
0.00
o.oo
0 . 0 J
0.0';
0 . H f ;


6. a
7.0
7.?
7.1
7.0
7.2
7.]
7.3
6.7
6.0
6.5
6.6
6.6
6.6
C
7.4
7.6


1950000.
2120000.
17HOOOO.
360000.
490000.
4i omo.
V600CJO.
100000.
""0000.
4BOCO!-0.
6300000.
170GOOC.
8.-JOOOOO.
^toooooo.
850COO.
loooooo.


11.
16.
63.
10.
40.
16.
17.
13.
250.
555.
50.
41.
32.
91.
5.
6.


30.
36.
116.
37.
37.
47.
36.
34.
1060.
1750.
129.
64.
67.
150.
24.
24.


715.0
615.0
325.0
275.0
IB5.0
175.0
110. 0
185.0
175.0
170.0
li,5.0
82.5
23?. 5


2W.O
230.0
160.0
142.0
116.0
110. 0
80.0
180.0
165.0
170.0
16U.O
70.0
100.0


-------
                Table  A-6.
                                                                         I15?01517
                                                                                               REPORT PRINTED  2/28/73 PAGE   11-2
                                                  SYRACUSE  CUMBINED  StjHER  UVERH.OHS  PKKLIMINARY  PH/\SI5
                STNQ  PLOC  SLOG  TYP!-.   SONO
                                                       DATE
                                                              Tit*1-
                                                                            MM IN   Cl.tGN   N02M03
                                                                                                 N02N   N03N    TALK    T- 1 P
                                                                                                                             CL
Ul
o
5
5
5
tj
5
F^
5
5
5
*>
5
r,
!)
3
5
5
ti
5
5
1 1 0
1 1 0
1 I 1
I 1 1
I 1 1
1 1 1
1 1 1
I 1 1
1 1 1
1 I 1
1 1 1
1 1 1
1 I I
1 1 1
1 1 1
1 1 1
I 1 I
1 1 15
I ? 1
1 1 1
1100
0
r
1
2
3
4
•5
6
7
R
•V
10
11
12
13
14
15
1
1
1
1
13H4
\ '.'15
,i5T
360
3hl
361:
36 1
36-i
11 (V 5
36o
167
368
36')
.570
371,
37',!
371
3>13
IK 4
305
11
7/
7/
//
7/
7/
11
!/
7/
//
7/
It
'It
It
7/
7/
7/
7/
7/
7/
7/
I/
ItlZ
7/72
7/72
?/72
7/7?
7/72
7/72
in?.
7/7?
7/72
7/72
7/72
7/72
7/7.1
7/72
7/7?
7/72
7/72
7/7?
7/72
7/7,?
1, 531'
I 5/i 5
1600
161 r>
163 i
164 r'
170')
1713
173'1!
17Vi
). 100
ism
1130
I.IA5
l')0'i
I'll
193 1
1 <) 3 i 1
1 9 1 1 )
I93'i
1.93')

2.3
"J.0
1.6
1. 1
2. 1
2.3
.1 . '1
1.7
6.')
12.3
2,2
J.I
2. )
2.0
1.4
1.9


0.40
0.40
n . 4 p
0 . ') t!
o.-n
0.96
1 . V)
1.72
1.26
? . 20
0.66
0.56
0,
-------
            Table  A-7.
                                                                   115201517
                                                                                         REPORT PRINTED  2/28/73 PAGE  26-1
                                            SYRACUSE CObHlNEO SfWER  nVGUFLm.'S PRELIMINARY PHASH
             STNO  PI.or,  su:c  TYPK   so NO  S\MI>
                                                 DATE
                                                         Tl MI-
                                                                 FLOW
                                                                        RAIN  WAlNINf
                                                                                       PH
                                                                                              T-CULI
                                                                                                        TOC
                                                                                                              cno
                                                                                                                      TSS
                                                                                                                             VSS
-
20
20
20
20
?n
20
P 0
20
20
20
20
20
20
2')
20
?0
20
20
2:1
3
3
3
3
3
3
.3
3
1
3
3
3
3
3
3
3
3
3
3
3
3
3
3
1
I
1
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
I
•7
'3
10
0
0
c
1
1
I
1
1.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
20
1
1
(:
r
0
i
•7
3
4
•j
7
'3
10
11
I 2
13
14
IS
16
17
1 •'
JO
20
1
1
1
1
139.)
604
60'
60 .
607
60,1
600
610
611.
/ 13
61 5
616
61 7
61 )
62,'
6 ' 1
6!' 4
ti ,"* "•)
6?7
9/1 V 72
0/13/72
0/11/72
9/13/72
9/1 )/??
•V 11/72
"/I 1/72
9/13/72
"/13/72
'J/13/72
0/13/72
9/U/72
5/11/72
9/13/72
9/1 i/ 12
0/13/72
9/1 V72
9/13/72
0/1 3/VP
9/1 3/7>
0/1.3/72
« / 1 i / I ?
V/U/72
1200
1 10 >
140 i
ISIS
1541
1600
L6lt>
1630
] 645
170n
171-j
1730
174b
1800
1030
190"
1930
1945
1945
194'.'
10 4 "5

3
2
0
1
1
0
0
0
0
3
1

140

.120
.440
.910
Mi
.AXO
,C60
.560
.370
.370
.2'iO
.220
.65,0

.00 I
0.00
0,04
G.Of
0.13
0. 13
0.17
0.24
0.28
0.30
0.3B
0.38
o . i a
0.3')
0. U)
0.38
0.38
'1. 30
0 . 3 f)
U,3f>
o.in
o . 3 a
0 . 1 H
11.3(1
O.Hfl

0
0
0
0
0
0
r
0
0
c
0
0
0
0
0
c
0
c
0
0
0
0

.04
. o 4
.O'j
.00
.16
.211
.16
.Oil
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.03


6.9
6.6
7.0
7.2
7.3
7.2
7.0
6.6
6.9
7.0
7.1
7.1
7.1
7.1
7. 1
7.1
7.?
tv
7.5
7.9


100000.
60000000.
92000000.
66000000.
R4COOOOO.
.aoooooo.
15600000.
19oOOOCO.
17400000.
46800000.
61600000.
44000000.
VOOOOOQ.
4070000.
1040000.


127.
35.
34.
153.
16.
73.
32.
23.
25.
25.
31.
-35.
50.
44.
53.
16.
22.


440.
41.
51.
71.
50.
48.
40.
3ft.
36.
43.
30.
Ifl.
39.
131.
47.
49.
24.
35.


134.0
190.0
596.0
300.0
144.0
80.0
70.0
54.0
60.0
60.0
60.0
84.0
7S.O
78.0
46.0
82.0


68.0
85.0
360.0
168.0
72.0
3C..O
36.0
28.0
29.0
32.0
32.0
54.0
4tt.O
52.0
16.0
34.0

in
H

-------
                 Table  A-8.
                                                                          115701517
                                                                                                 REPORT PRINTED  2/23/73 PAGE  26-2
                                                  SYRACUSE COMBINED  SEWER  UVE.V LOUS  PRELIMINARY PHASE
                 STNO  f'LOC  SLUG  TYPE  SONO
                                                       C.ATI:
                                                               TINT
                                                                       TKM
                                                                                         N02A03
                                                                                                  NC2N  N03'-l   TALK   T-IP
                                                                                                                              CL
Ul
to
20
r.
70
30
?n
7.)
20
70
.•Ti
-) , ^
?0
?()
20
?n
?:i
;>o
in
no
70
?o
?o
20
?c
20
3
3
3
3
3
3
1
3
T
3
3
3
3
3
3
3
1
3
3
3
3
3
3
3
3
T
3
L
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
3
111
Q
0
L-
1
1
1
1
1
1
I
I
1
I
1
I
1
1
1
1
1
1
1
1
70
1
1
0
I;
0
0
1
~t
3
't
5
6
7
,M
)
10
11
17
13
14
I 5
16
17
if;
19
20
1
1
1
1
' 39.'
1 3 T 3
1394
r: 1 1 'l
C.D-,
f.tj-,
6:)7
"l i :!
60',
.-JO
All
o!2
<-l 'i
•j 1 'i
61 r.
f>\(,
617
ft I?.
619
fi :• o
621
622
623
('>/' j
62't
t .">
6, '!
v// 13/72
•VI 3/72
•i/ I 3/72
•V1V72
-i/1 i/72
1/1 ;/72
;l/13/72
'.VH/77
• / 13/7,'
-f/13/7-'
')/ 13/72
J/13/72
9/13/72
T/13/72
•5/13/77
•)/l i/72
T/l 3/77
y/i ;./7?
9/13/72
T/13/77
9/13/77
-)/l '/72
4/1 i/72
9/13/72
J/ 13/7 2
;.V 13/77
9/13//2
170.i
1 30n
I'lOO
15u:l
15l«i
1 53(1
15'i5
l 6n^
161:'.
1.S3I-
1 o ', •-.
17P'J
1715
1 730
1 7--, 5
1800
1 8 1 'i
1.13°
I
19/, s

9.T
7.4
6.'-
7.0
fi.9
7.f.
7.4
3.4
C.-V
9.3
10.1
10.7
11.3
14. 1
16.0
1"5.4
;-.4
2.3


1.16
7.9f!
4.52
2.03
3.66
2.24
2.52
3.64
3.90
4.52
4 . 3 r;
4.91
5.35
7.17
7.95
7.49
0.31
0.40


;J.i
4.4
2.0
5.0
5.2
5.6
4.9
4.8
4.5
4.8
5.3
5.3
•5.9
6.9
fl.O
7.9
2.1
1.9


3.-'iJ
1.02
0.15
0.07
1.30
0.35
0.44
0.53
0.52
0.96
0.37
0.16
C.03
o.ni
0.01
iJ.Ol
0.46
0.44


1 .190
0.558
0.070
0.054
0.496
0.146
0.130
0.125
0.287
0.576
0. (47
0. 16J
0.077
0. Ill
0 . 0 1 1
0.011
0.060
0..127


2.21
0.4ii
0.03
U.O-1
0 . b 0
0.20
0.31
0.40
0.23
0.43
0.07
0.0"
o.n i
o.o ;
o.oo
O.O'i
0.40
0.41


299.
244.
159.
46.
36.
86.
117.
153.
170.
193.
203.
200.
213.
718.
?15.
209.
275.
733.


0.42
2.46
2.27
1.26
0.36
I. 01
0.9fc
1.43
1.08
1.70
2.26
2.25
2.89
3.95
2.815
3.19
0.03
0.15


77.
83.
57.
31.
19.
37.
48.
60.
6cJ.
64.
611.
77.
78.
77.
73.
71.
96.
79.


-------
                         APPENDIX B

  TABULATION OF DATA FOR SINGLE-STAGE DISINFECTION STUDIES

This data is the basis for the graphs of bacteria shown in
Figures 11, 12, 15, 16 and 25.
                            153

-------
Table B-l.  Effect Of
                                       On An Unscreened Effluent

Dosage (mg/1)

Contact
time
(sec)
0
30
60
120
180
300


Contact
time
(sec)
0
30
60
120
180
300

0

TCa
5,580,000
4,800,000
6,160,000
7,150,000
9,240,000
5,940,000

16


TC
5,580,000
715,000
209,000
187,000
2,200
3,300



FS
98,500
97,900
103,400
85,800
100,000
93,500




FS
98,500
34,100
55,000
30,800
11,700
9,460

4

TC FS
5,580,000 98,500
1,760,000 83,500
3,540,000 103,000
1,254,000 73,700
4,900,000 72,600
2,750,000 49,500
Dosage (mg/1)
20


TC - FS
5,580,000 98,500 5
616,000 24,200
11,000 23,000
5,940 12,540
660 6,930
320 3,740



TC
5,580,000
4,620,000
1,450,000
1,045,000
1,023,000
418,000




TC
,580,000
37,400
16,500
6,820
5,170
4,500

8

FS
98,500
108,900
74,800
46,200
41,800
13,200

25


FS
98,500
1,210
990
913
143
460

12

TC FS
5,580,000 98,500
583,000 77,000
275,000 55,000
242,000 38,000
74,800 20,300
29,000 12,000











a all bacterial data represent counts/100 ml

-------
                     Table  B-2.  Effect  Of  Cl2  On An Unscreened Effluent
Ul

Dosage (mg/1)

Contact
time
(sec)
0
30
60
120
180
300


Contact
time
(sec)
0
30
60
120
180
300





TCa
5,000
6,050
15,400
4,620
6,050
5,280




TC
5,000
3





,000
,000
,000
,000
,000
,000





,000
,850
220
860
120
120

0



FS
105
96
126
96
79
101

16



105
1





,000
,800
,500
,800
,200
,200




FS
,000
,100
110
20
66
27



TC
5,000,000
3,960,000
3,500,000
3,400,000
1,140,000
920,000
Dosage



TC
5,000,000
6,820
5,500
210
198
11

4



FS
105,
78,
73,
48,
44,
38,
(mg/1)
20


000
000
700
400
000
500




FS
105,
2,





000
200
110
99
11
0



TC
5,000
5,060
1,639
440
369
196




TC
5,000
11





8


,000
,000
,000
,000
,000
,900

25



,000
,000
560
590
980
85



FS
105,000
78,000
27,700
13,300
10,200
5,800




FS
105,000
2,750
170
120
88
46

12

TC FS
5,000,000 105,000
15,400 2,200
5,900 1,067
800 627
910 250
760 10











       a all bacterial data represent counts/100 ml

-------
        Table B-3. Effect Of Cl2 On An Effluent From A 23 Micron Screen



Contact
time
(sec)
0
30
60
120
180
300




0

TCa
3,970
5,390
2,860
5,280
8,390
3,090
,000
,000
,000
,000
,000
,000



FS
96,700
97,900
103,400
82,500
105,600
101,200



TC
3,970
4,950
3,861
1,199
946
165




,000
,000
,000
,000
,000
,000
Dosage

Contact
time
(sec)
0
30
60
120
180
300




TC
3,970
913
165
5



16



,000
,000
,000
,500
330
209




FS
96,700 3
49,500
30,800
5,610
1,980
485




TC
,970,
693,
4,
3,
3,
6,





000
000
000
500
400
900

Dosage
4

FS
96,700
89,100
84,700
68,200
79,200
64,900
(mg/1)
20


FS
96,700
34,100
15,730
4,400
1,155
220

(mg/1)


TC
3,970
1,870
297
531
50
2




TC
3,970,
326,
3,
2,
1,
1,





8



12

FS TC
,000
,000
,000
,400
,600
,420




96
59
31
17
22
2

25


,700 3,970,000
,400 539,000
,130 440,000
,930 22,000
,220 11,000
,915 220





FS
96,700
57,200
33,000
17,600
5,390
770




FS
000
700
400
200
200
430

96,
17,
2,
1,



700
900
200
012
209
121







a all bacterial data represent counts/100 ml

-------
               Table  B-4.  Effect Of Cl2 On An Effluent From A 23 Micron Screen
Ln
Dosage (mg/1)

Contact
time
(sec)
0
30
60
120
180
300




TCa
3,130
3,100
450
1,880
3,300
2,300
,000
,000
,000
,000
,000
,000
0

FS
93,800
97,900
115,500
84,700
83,600
91,300


TC
3,130,
2,970,
1,320,
1,210,
188,
82,



000
000
000
000
100
500
Dosage

Contact
time
( sec)
0
30
60
120
180
300



TC
3,130
460
83
5






,000
,000
,600
,610
880
350
16


FS
93,800
19,580
15,620
7,040
5,610
110



TC
3,130,
6,
1,







000
380
200
198
110
270
4

FS
93,800
53,900
42,900
74,800
30,800
20,900
(mg/1)
20


FS
93,800
3,440
209
181
47
11


TC
3,130
2,300
380
120
3





TC
3,130
1
1






,000
,000
,000
,000
,700
660

2



,000
,980
,100
66
110
190
8


93
62
48
13
13
9

5



93





12

FS TC
,800 3,130,000
,700 1,870,000
,400 230,000
,200 34,000
,200 4,700
,460 490




FS
,800
220
220
33
22
22

FS
93,800
33,000
26,400
10,780
11,500
5,720











         all bacterial data represent counts/100 ml

-------
              Table B-5. Effect Of C102 On An Unscreened Effluent
Dosage (mg/1)

Contact
time
(sec)
0
30
60
120
180
300


TC
3,120
2,200
3,400
4,730
2,860
2,970
0

a
,000
,000
,000
,000
,000
,000


FS
126,000
99,000
119,900
101,000
114,000
114,400


TC
3,120
77
84
83
120
119


4



FS
,000
,000
,700
,000
,000
,000
Dosage

Contact
time
(sec)
0
30
60
120
180
300




TC
3,120
38
2
1



16



,000
,500
,300
,012
430
560




FS
126,000
418
230
66
33
11




TC
3,120
110
2
1

1




126 ,
4,
5,
2,
5,
5,
(mg/1)
20


000
400
720
860
060
600




FS
,000
,000
,640
,078
297
,180

126,






000
605
66
66
11
33



TC
3,120
44
27
31
5
1




TC
3,120
4
1




8


,000
,000
,000
,900
,500
,650

25



,000
,840
,430
440
649
480

12

FS TC
126,000 3,120,000
6,160 22,000
2,200 18,700
1,650 4,400
440 1,650
110 2,200




FS
126,000
437
495
16
11
16


FS
126,000
880
550
220
44
11











a all bacterial data represent counts/100 ml

-------
             Table B-6. Effect Of C102 On An Unscreened Effluent
Dosage (mg/1)

Contact
time
(sec)
0
30
60
120
180
300


0

TCa
9,230
4,950
4,290
6,270

6,050
,000
,000
,000
,000

,000




FS
163
165
220
159

189
,000
,000
,000
,000

,200


TC
9,230,
6,380,
3,960,
2,640,

3,300,
4


000 163
000 135
000 94
000 67

000 95


FS
,000
,300
,600
,100

,700


TC
9,230
510
440
550

440
8


,000
,400
,000
,000

,000


FS
163,000
33,000
33,000
24,640

3,432
12

TC FS
9,230,000 163,000
144,100 10,560
166,.000 10,000
104,500 836

31,900 1,000
Dosage (mg/1)

Contact
time
(sec)
0
30
60
120
180
300




TC
9,230
51
44
35

60

16



,000
,700
,000
,800

,700





163
5
1







FS
,000
,280
,232
770

451




TC

No
20




data



FS


Available















TC
9,230
24
17
13

9

2



,000
,100
,600
,640

,350

5


FS
163,000
682
418
340

748











a all bacterial data represent  counts/100 ml

-------
      Table B-7. Effect of C102 On An Effluent From A 23 Micron Screen
Dosage

Contact
time
(sec)
0
30
60
120
180
300


0

Tca
5,390
6,000
4,950
6,050
4,950
6,700
,000
,000
,000
,000
,000
,000


FS
169,900
134,200
132,000
101,200
148,500
162,800


TC
5,390
340
220
169
144
123


4

(mg/1)


FS
,000
,000
,000
,400
,000
,000
Dosage

Contact
time
( sec)
0
30
60
120
180
300



TC
5,390
2
2



16



,000
,200
,310
286
154
99



FS
169,900
450
88
77
44
33



TC
5,390
2
2
3





169,
50,
74,
62,
53,
58,
Cmg/1)
20


900
600
800
700
900
300




FS
,000
,200
,500
,850
330
220
169,





900
340
132
429
11
44
8

TC
5,390,000
4,510
1,100
330
440
154

25


TC
5,390,000
6,160
1,375
506
520
231
12

FS TC
169,900 5,390,000
990 3,850
660 660
47 440
132 220
77 220




FS
169,900
605
320
132
55
22

FS
169,900
748
143
55
16
33











all bacterial data represent counts/100 ml

-------
      Table B-8. Effect Of C102 On An Effluent From A 23 Micron Screen
Dosage

Contact
time
(sec)
0
30
60
120
180
300


0

TCa
5,382
4,180
3,850
5,720
5,830
3,960
,000
,000
,000
,000
,000
,000




FS
110
94
89
102
94
83
,600
,600
,100
,300
,600
,600


TC
5,382
99
47
40
34
47


4

(mg/1)


FS
,000
,000
,300
,700
,000
,300
Dosage

Contact
time
(sec)
0
30
60
120
180
300




5,382
7
5
2
1

16



,000
,700
,390
,285
,056
990




110









,600
330
11
22
11
11




5,382
5
4
2

1




,000
,060
,070
,200
946
,012
110,
5,
2,
1,
2,
1,
Cmg/1)
20



110,





600
060
200
980
400
870





600
110
22
0
0
00
8

TC
5,382,000 110
2,200
1,760
66
11
33

25



5,382,000 110
8,800
540
760
550
590
12

FS TC
,600 5,382,000
220 990
110 940
99 320
44 120
5 66





,600
286
9
11
0
0

FS
110,600
220
99
88
11
11











a all bacterial data represent  counts/100 ml

-------
Table B-9.  Temperature Effects On Bacteria During Disinfection With CLj
Cl_2 Dosage (mg/1)
Temp .
(°C)
2
23.5
30
Contact
time
(sec)
0
30
60
120
180
300
0
30
60
120
180
300
Q
30
60
120
180
300
8
Tca
7,300,000
3,400,000
2,100,000
1,100,000
560,000
61,000
1,025,000
4,400,000
1,610,000
200,000
100,000
15,000
13,800,000
11,600,000
2,400,000
290,000
60,000
75,000

FS
178,000
147,000
180,000
102,000
66,000
35,000
207,000
106,000
68,000
56,000
12,000
14,400
213,000
75,000
34,000
13,000
11,200
5,800
25
TC
7,300,000
106,000
46,000
19,000
189,000
800
10,250,000
70,000
9,000
2,100
450
200
13,800,000
65,000
71,000
3,900
5,000
1,020

FS
178,000
34,000
194,000
20,000
11,400
1,900
207,000
16,000
5,300
1,700
500
40
213,000
4,400
320
100
250
10
 a  all  bacterial  data represent counts/100 ml

-------
           Table B-10.  Temperature Effects On Bacteria During Disinfection With  C102
CTi
u>
C102 Dosage (mg/1)
8
Temp Contact
(°C) time
(sec)
2 0
30
60
120
180
300
23.5 0
30
60
120
180
300
30 0
30
60
120
180
300


TCa
8,600,000
135,000
127,000
120,000
244,000
40,000
7,700,000
400,000
216,000
168,000
100,000
30,800
10,950,000
280,000
96,000
84,000
100,000
27,000


FS
155,500
73,000
28,000
9,300
7,000
3,200
179,500
17,000
8,900
6,200
7,300
2,600
208,000
41,800
9,500
7,200
5,600
2,280
25


TC
8,600,000
50,000
25,000
7,400
18,400
20,000
7,700,000
24,000
19,000
24,000
10,000
3,600
10,950,000
61,000
39,000
12,800
12,200
13,100


FS
155,500
3,870
1,600
1,300
10,600
9,500
179,500
3,050
1,250
670
360
1,250
20,800
2,520
940
970
1,800
1,100
             a  all bacterial data represent counts/100 ml

-------
                         APPENDIX C

    DERIVATION OF DILUTION FACTORS AND RESIDENCE TIMES
                  FOR AFTERGROWTH STUDIES

1.  Calculation of ratio of total overflow to dry-weather
    Onondaga Creek flow

    Qb = Average dry-weather Onondaga Creek flow at point of
         entry to the City of Syracuse;  i.e., before
         occurrence of any CSO.   From Reference 3, this is
         taken to be 4500 cfs.

    Qo = Total flow for those CSO discharging into Onondaga
         Creek.  For a maximum flow condition as taken from
         Reference 1.

         Qo = ciA1 where c = coefficient of runoff
                          = 0.20
                        i = rainfall intensity
                          =0.50  in/hr
                        A" = area  of drainage basin
                          = 53,000 acres

    Q0 - 5300 cfs

    R! = ratio of Qb/Qo
       = 4500 cfs ^
         5300 cfs   1:

2.  Calculation of maximum CSO  residence time in Onondaga
    Creek

    T = residence time  = distance from discharge of first
        overflow into  creek to  entry into Onondaga Lake (D)
        divided by creek velocity (v).

    From Reference 3,  D = 33,000  ft

    v is calculated from the equation Q/A where
    Q = Q0 + Qb/  and A  is the average cross-sectional area of
        the creek at time of maximum CSO.

    From the Manning Equation.

    Q = 1.49 X A X R2/3 X S1/2

    Q = 9800 cfs
                             164

-------
Appendix C  (Cont'd)

    n = Manning's coefficient of friction = 0.025

    A = trapezoidal cross-sectional area of creek with a
        24-foot base as taken from Reference 1.

      = 24 h + h2 where h = average creek depth

    R = hydraulic radius
      = A/PW where PW = wetted perimeter = 24 + 2.84h from
        Reference 1.

    S = slope of creek bed = 0.0015 in/ft from Reference 1.

    Solving from h and subsequently A yields

    A = 880 ft2

    Therefore,  v = 11.15 fps

               T = 0.82 hour

                 - 1 hour

3.   Calculation of ratio of total creek flow, including CSO
    to Onondaga Lake

    The total creek volume consists of 9800 cfs for the
    duration of an average storm which is 400 minutes from
    Reference 1.   This quantity Vj_ is 1.76 x 109 gallons.

    The volume  (V^I  of the epilimnion of the southern basin
    of Onondaga Lake is 24.7 x 109 gallons.  This quantity
    is taken because this is the area of immediate impact on
    the lake and consequently its selection will maximize
    any effects.

    R2 = dilution factor from creek to lake

       = V2/V1  = 1/14.2 =1:14
                             165

-------
                         APPENDIX D

    TABULATION OF DATA FOR TWO-STAGE DISINFECTION STUDIES

This data is the basis for the graphs of bacteria and ATP in
Figures 29, 30, 32 and 33.
                            166

-------
Table D-l. Two-Stage Disinfection Results
   Including ATP-Bacteria Correlations
Dosage
(mg/1)
C102


2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
C102


0
2
2
2
2
2
2
0
0
2
2
2
2
2
0
4
4
4
4
4
4
0
0
4
4
4
4
4
Contact
time
(sec)
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
Bacteria
(counts/100 ml)


TC
6,380,000
407,000
165,000
22,000
49,500
22,000
5,000
4,840,000
1,430,000
1,056,000
16,500
14,300
11,000
5,500
2,860,000
715,000
29,500
7,700
3,300
1,100
770
4,180,000
1,045,000
836,000
16,500
2,640
1,430
480


FC
615,000
63,500
11,000
4,620
3,300
2,200
2,300
594,000
231,000
116,600
7,240
3,520
2,200
1,540
308,000
41,800
1,850
460
2,200
660
405
322,000
97,000
52,000
1,100
600
250
60


FS
99,000
92,400
77,000
39,600
33,000
16,500
24,900
110,000
86,900
77,000
40,700
18,700
8,580
2,970
77,000
67,100
36,300
1,100
900
770
220
114,300
91,300
83,600
11,000
3,310
540
385
ATP
(arbitrary
units)
50.82
22.44
10.89
7.89
5.98
4.56
3.00
50.27
25.30
30.25
7.77
5.20
3.19
1.64
19.25
8.40
1.54
0.65
0.54
0.40
0.44
18.48
10.78
9.92
0.78
0.48
0.33
0.27
                  167

-------
Table D-2.  Two-Stage Disinfection Results
   Including ATP-Bacteria Correlations
Dosage
(mg/1)
C102


4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
C102


0
2
2
2
2
2
2
0
0
2
2
2
2
2
0
4
4
4
4
4
4
0
0
4
4
4
4
4
Bacteria
(counts/100
ml)
Contact
time
(sec)
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120

TC
1,870,000
340,000
9,999
8,800
9,900
2,200
999
1,430,000
80,300
18,700
4,070
2,970
2,899
2,899
803,000
28,600
880
880
580
410
275
1,122,000
126,000
19,250
1,100
460
250
120

FC
39,600
1,100
999
110
110
330
77
29,700
1,760
360
66
77
77
55
24,750
550
20
10
10
10
0
880
440
440
0
20
0
0

FS
121,000
30,800
10,000
2,530
2,640
1,430
1,100
129,800
57,200
14,850
2,860
1,590
1,100
350
130,900
14,300
1,180
820
600
300
280
124,000
28,300
6,999
1,200
500
385
120

ATP
(arbitrary
units)
31.57
5.00
1.13
0.83
0.56
0.46
0.46
19.47
4.33
1.89
0.86
0.52
0.45
0.36
24.97
3.19
1.21
0.75
0.49
0.00
0.00
20.24
5.21
0.57
0.00
0.00
0.00
0.00
                  168

-------
Table D-3. Two-Stage Disinfection Results
   Including ATP-Bacteria Correlations
Dosage
(mg/1)
C12


4
4
4
4
4
4
4
4
-4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
ClOo
£


0
2
2
2
2
2
2
0
0
2
2
2
2
2
0
4
4
4
4
4
4
0
0
4
4
4
4
4
Contact
time
Csec)
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
Bacteria
(counts/100 ml)


TC
3,430,000
572,000
113,300
473,000
123,000
1,100
52,800
2,220,000
924,000
506,000
187,000
62,700
132,000
75,900
1,870,000
1,870,000
406,500
176,000
187,000
80,300
59,400
2,640,000
1,430,000
1,045,000
143,000
96,800
60,500
13,200


FS
308,000
84,700
59,400
82,500
40,700
38,500
12,100
110,000
86,900
74,800
38,500
39,600
40,700
13,200
112,600
84,700
18,700
8,800
5,500
8,910
6,160
109,900
47,300
56,100
102,300
8,800
4,180
3,410
ATP
(arbitrary
units)
45.21
35.31
32.01
31.35
26.84
29.37
26.73
38.72
29.37
27.72
27.28
23.32
20.02
19.80
25.96
18.15
10.34
6.82
6.83
6.07
5.34
13.31
16.83
10.23
4.33
2.02
2.92
2.26
                  169

-------
Table D-4.  Two-Stage Disinfection Results
   Including ATP-Bacteria Correlations
Dosage
(mg/1)
C12
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
cio2
0
2
2
2
2
2
2
0
0
2
2
2
2
2
0
4
4
4
4
4
4
0
0
4
4
4
4
4
Contact
time
(sec)
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
Bacteria
(counts/100 ml)
TC
187,000
121,000
90,000
1,100
1,210
660
110
220,000
61,600
4,400
1,100
330
99
99
660,000
999
99
99,999
18,700
110
99
FC
12,100
1,430
99
99
99
99
9
9,900
2,420
2,250
320
165
9
11
30,250
220
9
22,500
3,630
90
9
FS
82,500
38,500
2,200
3,300
3,850
1,980
1,870
72,900
17,600
3,300
999
2,750
1,980
1,320
11,000
999
99
74,800
8,800
110
44
ATP
(arbitrary
units)
7.22
10.30
2.75
2.93
1.24
0.70
0.55
9.60
6.06
1.45
1.46
0.74
0.73
0.65
10.01
7.11
0.89
0.49
0.27
0.39
0.27
8.17
10.21
5.20
0.52
0.35
0.25
0.19
                   170

-------
                         APPENDIX E

         TABULATION OF DATA FOR AFTERGROWTH STUDIES

This data is the basis for the graphs of bacteria shown in
Figures 37 and 38.
                           171

-------
 Table E-l. Total Coliform Results From Tank Tests Without Screening Using Cl2'
Sample
Cl2 Dosage
(mg/1) " 0
Composition Contact
time (sec) 0

Overflow

Overflow

+

Creek
Overflow

+

Creek

+

Lake
Time of
Sample
Before
Screen
After
Screen
After C12
Initial
20 min
40 min
60 min

Initial
15 min
1 hr
4 hr
12 hr
24 hr
48 hr
72 hr


3,100,000
2,400,000
-
1,200,000
1,480,000
990,000
1,760,000

117,000
34,300
27,000
30,200
13,700
1,120
110
10

25
30

3,100,100
2,400,000
2,900
18,600
1,100
2,800
5,500

820
250
10
10
0
11
5
4

25
60

3,100,000
2,400,000
51,400
29,800
330
130
660

495
240
10
4
22
60
185
8


Creek
Control

-
-
8,200
1,160
2,300
4,400

744
690
400
470
250
365
44
24


Lake
Control

-
-
-
-
-
-

484
594
270
22
56
88
16
4

a all bacterial data represent counts/100 ml

-------
        Table E-2.  Fecal Coliform Results From Tank Tests Without Screening Using
OJ
Sample
Cl2 Dosage
(mg/1)
0
25
25
Composition Contact



Overflow




Overflow

+

Creek
Overflow

+

Creek

+

Lake
time (sec)
Time of
Sample
Before
Screen
After
Screen
After C12
Initial
20 min
40 min
60 min

Initial
15 min
1 hr
4 hr
12 hr
24 hr
48 hr
72 hr

0



630,000

500,000
539,000
270,400
302,000
390,000
280,000

18,700
11,200
5,390
5,890
2,770
132
0
0

30



630,000

500,000
69,000
35,400
440
910
110

73
0
0
0
0
0
0
0

60



630,000

500,000
330
1,065
88
120
150

76
25
0
0
0
0
0
0


Creek
Control



-

-
-
1,800
440
1,600
550

103
71
66
135
11
0
0
0


Lake
Control



-

-
-
-
-
-
-

71
22
16
5
16
5
0
0

       a all bacterial data represent counts/100 ml

-------
  Table E-3. Fecal Strep Results From Tank Tests Without Screening Using

Sample
Cl2 Dosage
(mg/1)
0
25
25
Composition Contact



Overflow




Overflow

+

Creek
Overflow

+

Creek

+

Lake
time (sec)
Time of
Sample
Before
Screen
After
Screen
After Cl2
Initial
20 min
40 min
60 min

Initial
15 min
1 hr
4 hr
12 hr
24 hr
48 hr
72 hr

0



161,000

152,000
154,000
76,000
51,700
75,900
89,000

5,940
4,800
3,200
2,400
1,180
115
0
0

30



161,000

152,000
1,920
1,015
550
340
330

36
20
2
0
0
0
0
0

60



161,000

152,000
1,160
635
55
33
22

16
11
0
0
0
3
0
0


Creek
Control



-

-
-
110
660
110
220

29
15
11
0
0
5
5
0


Lake
Control



-

-
-
-
-
-
-

15
0
0
0
0
0
0
8

a all bacterial data represent counts/100 ml

-------
Table E-4. Total Coliform Results From Tank Tests With A 23 Micron Screen Using
Sample
Cl2 Dosage
(mg/1)
0
25
25
Composition Contact



Overflow




Overflow

+

Creek
Overflow

+

Creek

+

Lake
time (sec)
Time of
Sample
Before
Screen
After
Screen
After Cl2
Initial
20 min
40 min
60 min

Initial
15 min
1 hr
4 hr
12 hr
24 hr
48 hr
72 hr

0



750,000

880,000
860,000
431,000
390,000
620,000
920,000

61,00
23,00
18,400
18,700
3,400
1,200
1,023
52

30



750,000

880,000
1,100
1,450
1,360
17,500
0

10
0
0
0
8
48
260
150

60



750,000

880,000
250
1,025
350
0
0

10
0
0
0
0
0
0
0


Creek
Control



-

-
-
1,800
550
2,200
1,370

101
170
236
85
12
8
16
16


Lake
Control



-

-
-
—
—
-
-

11
11
22
16
16
16
8
12

  a  all bacterial data represent counts/100 ml

-------
       Table  E-5.  Fecal Strep Results From Tank Tests With A 23 Micron Screen Using
CTl
Sample
Cl2 Dosage
(mg/1)
0
25
25
Composition Contact



Overflow




Overflow

+

Creek
Overflow

+

Creek

+

Lake
time (sec)
Time of
Sample
Before
Screen
After
Screen
After C12
Initial
20 min
40 min
60 min

Initial
15 min
1 hr
4 hr
12 hr
24 hr
48 hr
72 hr

0



93,000

77,000
84,000
42,000
60,000
58,000
47,000

3,140
43,000
14,000
1,100
110
22
0
6

30



93,000

77,000
490
330
66
120
0

4
0
0
0
0
0
0
0

60



93,000

77,000
330
250
25
0
0

4
0
0
0
0
0
16
0


Creek
Control



-

-
-
170
170
230
130

13
11
0
8
4
0
8
0


Lake
Control



-

-
-
-
-
-
-

4
0
0
4
0
0
0
0


-------
 Table  E-6.   Total  Coliform Results  From Tank Tests Without Screening Using C102'
Sample
C102 Dosage
(mg/1) 0
Composition Contact
time (sec) 0
25
30
25
60

Creek
Control

Lake
Control
Time of
Sample
Overflow
Before
Screen
After
Screen
3,900,000
6,700,000
After C102 6,300,000
Overflow

+

Creek
Overflow

+

Creek

+

Lake
Initial
20 rain
40 min
60 min

Initial
15 min
1 hr
4 hr
12 hr
24 hr
48 hr
72 hr

3,150,000
4,200,000
3,300,000
1,760,000

117,500
94,000
91,000
30,000
4,900
1,490
182
504

3,900,000
6,700,000
22,000
12,850
5,700
880
1,300

498
300
220
184
233
74
132
25

3,900,000
6,700,000
2,200
2,850
660
280
300

431
180
204
114
92
88
26
33

-
-
3,700
2,400
110
350

418
350
420
200
57
67
78
12

-
-
-
-
-
-

440
300
470
376
252
127
132
148

a all bacterial data represent counts/100 ml

-------
        Table E-7. Fecal Strep Results From Tank Tests Without Screening Using Cl02a
00
Sample
C1C>2 Dosage
Cmg/1)
0
25
25
Composition Contact



Overflow




Overflow

+

Creek
Overflow

+

Creek

+

Lake
time (sec)
Time of
Sample
Before
Screen
After
Screen
After .C10_2_
Initial
20 min
40 min
60 min

Initial
15 min
1 hr
4 hr
12 hr
24 hr
48 hr
72 hr

0



172,000

146,000
160,000
81,850
102,000
82,000
79,000

5,270
36,000
26,000
1,290
360
60
8
4

30



172,000

146,000
1,500
1,245
270
44
44

3
0
0
0
0
0
0
0

60



172,000

146,000
280
635
92
30
29

2
0
0
0
0
0
0
0


Creek
Control



-

-
-
990
130
126
88

6
24
8
12
0
0
0
0


Lake
Control



-

-
-
-
-
-
-

440
300
470
376
252
127
132
148

       a all bacterial data represent counts/100 ml

-------
Table E-8. Total Coliform Results From Tank Tests With A 23 Micron Screen Using C102a
Sample
Composition



Overflow




Overflow

+

Creek
Overflow

+

Creek

+

Lake
C102 Dosage
(mg/1)
Contact
time (sec)
Time of
Sample
Before
Screen
After
Screen
After CIO?
Initial
20 min
40 min
60 min

Initial
15 min
1 hr
4 hr
12 hr
24 hr
48 hr
72 hr

0

25

25



Creek
0



12,100

1,800
3,000
1,503
1,900
680
690

50
49
27
38
4
1







,000

,000
,000
,200
,000
,000
,000

,750
,000
,000
,000
,400
,960
550
170

30



12,000

1,800
5
5
2
1
1

4
2
4
1









,000

,000
,500
,950
,000
,560
,230

,832
,700
,800
,500
850
50
132
--

60



12,100

1,800
3
5




4
5
9
4





Control



,000

,000
,900
,150
530
550
390

,776
,200
,500
,500
599
250
170
150








6
7
6
4

b
5
4
2








-

-
-
,400
,100
,000
,440

,420
,600
,700
,400
700
55
88
66


Lake
Control



-

-
-
-
-
-
-

5,100
4,900
6,200
3,600
2,260
270
160
88


  a  all bacterial data represent counts/100 ml

-------
      Table E-9.  Fecal Strep Results From Tank Tests With A 23 Micron Screen Using Cl02a
00
o
Sample
C102 Dosage
(mg/1)
0
25
25
Composition Contact



Overflow




Overflow

+

Creek
Overflow

+

Creek

+

Lake
time (sec)
Time of
Sample
Before
Screen
After
Screen
After C102
Initial
20 min
40 min
60 min

Initial
15 min
1 hr
4 hr
12 hr
24 hr
48 hr
72

0



111,000

70,000
80,000
40,195
37,000
36,000
38,000

2,610
3,300
2,700
730
75
20
0
0

30



111,000

70,000
770
580
110
57
92

83
22
39
36
4
0
0
-

60



111,000

70,000
, 420
455
22
8
20

78
134
24
20
0
4
0
0


Creek
Control



-

-
-
390
- 290
310
206

91
57
52
44
8
0
0
0


Lake
Control



-

-
-
-
-
—
-

83
79
76
35
8
0
0
0

       a all bacterial data represent counts/100 ml

-------
 1. REPORT NO.
                             TECHNICAL REPORT DATA    .
                       (Please read Infractions on the reverse before completing)	
                             	'	13. RECIPIENT'S ACCESSION-NO.
    EPA-670/2-75-021
                         2.
 4. TITLE ANDSUBTITLE
 BENCH-SCALE  HIGH-RATE DISINFECTION OF
 COMBINED  SEWER OVERFLOWS; With  Chlorine
 and Chlorine Dioxide
                                                 5. REPORT DATE        .
                                                 April 1975; Issuing  Date
                                                 6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
 Peter E. Moffa,  Edwin C. Tifft,  Jr.,
 Steven L.  Richardson, and James  E.  Smith
                                                  8. PERFORMING ORGANIZATION
9. PERFORMING ORG "kNIZATION NAME AND ADDRESS

 O'Brien & Gere Engineers,  Inc.
 1304  Buckley Road
 Syracuse, New York   13201
                                                  10. PROGRAM ELEMENT NO.
                                                  1BB034; ROAP 21-ASY; Task 124
                                                  11. cto^l^WC^/GRANT NO.

                                                  S-802400  (11020  HFR)
 12. SPONSORING AGENCY NAME AND ADDRESS
 National  Environmental Research  Center
 Office of Research and Development
 U.S. Environmental Protection Agency
 Cincinnati,  Ohio  45268
                                                 13. TYPE OF REPORT AND PERIOD COVERED
                                                 Interim,  Nov. 71-Mar.  73
                                                 14. SPONSORING AGENCY CODE
 15. SUPPLEMENTARY NOTES
                        Division of Drainage and  Sanitation
Joint  Sponsorship with Department of Public Works
	County of Onondago, New York	
 16. ABSTRACT
 A bench-scale  study of high-rate disinfection of combined  sewer over-
 flows with chlorine and chlorine dioxide was performed to  aid  in the
 design and operation of full-scale  prototype treatment facilities.
 Four logarithm reductions in three  inidicator bacteria and  several
 common viruses were obtained with 25  mg/1 chlorine or 12 mg/1  chlorine
 dioxide in two-minutes contact time.   Sequential addition  of disinfec-
 tants enhanced the  process such that  only eight mg/1 of chlorine fol-
 lowed in 15 to 30  seconds by two mg/1 chlorine dioxide were necessary
 to obtain similar  reductions.  The  removal of suspended solids by
 microscreening through a 23 micron  aperture had little effect  on dis-
 infection efficiency.   Disinfection increased slightly with increased
 temperature, and a  study of the receiving waters indicated no  bacterial
 or viral aftergrowth.   Adenosine triphosphate (ATP) was found  to be a
 possible alternative to the bacterial indicators  of disinfection
 efficiency and microbial contamination.   Electron spin resonance (esr)
 was used as a  primary standard method for quantitative measurement  of
 chlorine dioxide residuals.
                                             Bibliography, 84 references
                           KEY WORDS AND DOCUMENT ANALYSIS
               DESCRIPTORS
 Urban areas, Runoff,
 Sewage, Overflows,
 Tests, *Screenings,
 *Disinfection,  Bacteri-
 cides, Combined sewers,
 Chlorine
                                       b.IDENTIFIERS/OPEN ENDED TERMS
                            *High-rate treatment, Combined
                            sewer  overflows, Suspended
                            solids removal, Storm runoff,
                            *Chlorine dioxide, Wastewater
                            treatment, Laboratory tests,
                            Viricides, County of Onondago
                            (New York)
                                                             c. COSATI Field/Group
13B
 3. DISTRIBUTION STATEMENT
 RELEASE TO PUBLIC
                                      19. SECURITY CLASS (This Report)
                                          UNCLASSIFIED
                                                              !1 . NO. OF PAGES
                                                                   193
                                      20. SECURITY CLASS (This page)
                                          UNCLASSIFIED
                                                             22. PRICE
EPA Form 2220-1 (9-73)
                                     181
                                            U. S. GOVERNMENT PRINTING OFFICE: 1975-657-592/5360 Region No. 5-I I

-------