United States
Environmental Protection
Agency
Municipal Environmental Research  EPA-600/2-80-102
Laboratory           August 1980
Cincinnati OH 45268
Research and Development
Ultraviolet
Disinfection  of
Municipal
Wastewater  Effluents

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further deveJopment and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental Health  Effects Research
      2.  Environmental Protection Technology
      3.  Ecological Research
      4.  Environmental Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special" Reports
      9,  Miscellaneous Reports

This report has  been assigned  to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research  performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution-sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia  22161.

-------
                                      EPA-600/2-80-102
                                      August  1980
        ULTRAVIOLET DISINFECTION  OF
      MUNICIPAL WASTEWATER EFFLUENTS
                    by

          Albert C.  Petrasek,  Jr.
              Harold W.  Wolf
             Steven  E. Esmond
             D.  Craig Andrews
          Texas  A &  M University
       Col 1ege Stati on,  Texas   77843
            Grant No.  R-803292
              Project Officer

             Albert D. Venosa
       Wastewater Research Division
Municipal Environmental Research Laboratory
          Cincinnati, Ohio  45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
    OFFICE OF RESEARCH AND DEVELOPMENT
   U.S. ENVIRONMENTAL PROTECTION AGENCY
          CINCINNATI, OHIO  45268

-------
                                 DISCLAIMER
     This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion.  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.

-------
                                  FOREWORD
     The U.S.  Environmental  Protection Agency was created because of increas-
ing public and government concern about the dangers of pollution to the health
and welfare of the American  people.   Noxious air, foul water, and spoiled land
are tragic testimonies to the deterioration of our natural environment.  The
complexity of that environment and the interplay of its components require
a concentrated and integrated attack on the problem.

     Research and development is that necessary first step in problem solution;
it involves defining the problem, measuring its impact, and searching for
solutions. The Municipal Environmental Research Laboratory develops new and
improved technology and systems to prevent, treat, and manage wastewater and
solid and hazardous waste pollutant discharges from municipal and community
sources, to preserve and treat public drinking water supplies, and to minimize
the adverse economic, social, healtlu'and aesthetic effects of pollution.  This
publication is one of the products of that research and provides a most vital
communications link between  the researcher and the user community.

     This study concerns .the evaluation of ultraviolet light as a potential
candidate for replacing chlorine as a wastewater effluent disinfectant.  Devel-
opment of a process such as  ultraviolet light disinfection is important, since
it is nontoxic to aquatic life, does not form potentially toxic halogenated •: •
compounds in the treated water, and provides excellent disinfection at a
reasonable cost.
                                       Francis T. Mayo, Director
                                       Municipal Environmental Research
                                          Laboratory
                                      iii

-------
                                  ABSTRACT

     This research project was conducted at the Dallas Water Reclamation
Research Center by the Dallas Water Utilities Department, Texas A&M Univer-
sity, and the United States Environmental Protection Agency to evaluate the
feasibility of using ultraviolet light irradiation as a disinfection process
for municipal wastewater effluents.

     During this project two different UV exposure and irradiation systems
were studied.  The first system investigated was the Kelly-Purdy Unit, which
consisted of a shallow-tray exposure chamber with 13 UV lamps mounted
horizontally  10 cm above the bottom of the chamber.  This unit was operated
under varying conditions of both flow and depth and generally provided
inadequate disinfection, although fecal coliform densities were usually
reduced by approximately three logs.

     The second UV system used during the project was the Model EP-50
manufactured by Ultraviolet Purification Systems, Inc.   This exposure cham-
ber consisted of a stainless steel  pressure vessel  with nine UV lamps
running longitudinally through the chamber.  Each lamp was isolated from
the effluent being disinfected by a quartz sleeve.   The disinfection observed
on any given run was shown to be a function of the  UV dose,  and greater
than four log reductions in fecal  coliform densities were observed at times.


     During this project three special  virus  studies were conducted.   The
influent to the EP-50 was seeded with an F2 coliphage and an attenuated
Type I poliovirus.   During the three virus runs, viral  and phage densities in
the influent and effluent of the UV irradiation  chamber were monitored.
The observed reductions in viruses  were correlated .with UV doses,  and the F2
coliphage response to the UV disinfection process was similar to the
response of the-Type I poliovirus.

     This report was submitted in fulfillment of Grant No. R-803292 by
Dallas Water Utilities under the sponsorship of the U.S. Environmental Pro-
tection Agency.
                                     i v

-------
                                CONTENTS

Foreword	  i i i

Abstract	-	   iv

Figures	vii

Tables	'	•   xv

Acknowledgment	xix


             1.  Introduction   	    1
                       Background of the Dallas program  	    4
                       Research presented in this report .....    5

             2.  Conclusions	    7

             3.  Recommendations	  ......

             4.  Description of Research Facilities   	   11
                       City of  Dallas  collection system  	   11
                       Treatment  facilities   	 .....   14

             5.  Sampling  and Analytical Procedures   .......   29
                       Sampling procedures  	   29
                       Analytical procedures  .  	   29

             6.  Operation of the Kelly-Purdy  Unit  .  .	   34
                       •Hydraulic  characteristics	;  ... 34
                        Intensity  and dose determinations  	   .38
                        Run No.  Kl   	   49
                        Run No.  K2	   53
                        Run No.  K3	 .   65
                        Run No.  K4  .  . .  .	-  .   75
                        Run No.  K5  ....;....	   82

              7.  Operation  of  the  UPS Unit   .	   96
                        Hydraulic  characteristics  	   96
                        Determination of dose .	103
                        Run No.  Ul	109
                        Run No.  U2	119

-------
                            CONTENTS (continued)

               Run No. U3	119
               Run No. U4	  130
               Run No. U5	140
               Run No. U6	158
               Run No. U7	167
               Run No. U8	*.	174

       8. Virus Studies	  191
               First UV virus run	193
               Second UV virus run	  199
               Third UV virus run	213

       9. Summary of Operating Experience 	  230

      10. Photoreactivation 	  252

References	260
                                   VI

-------
                                  FIGURES
Number
  1
  2
  4
  5

  6

  7

  8

  9

 10

 11

 12

 13

 14
                                                           Page
Existing City of Dallas water supply network .......   2
Possible sources of influents for the Pilot
  Plant at the White Rock Sewage Treatment Plant	17
Flow schematic of the Dallas Water Reclamation
  Research Center Demonstration Plant  .-....•	18
Perspective drawing of the Kelly-Purdy unit	26
Theoretical residence time as a function of flow and
  depth for the Kelly-Purdy system	35
Residence time distribution functions obtained at a
  water depth of 2.54 cm in the Kelly-Purdy system ....  36
Residence time distribution functions obtained at a depth
  of 6.35 cm in the Kelly-Purdy system	,	37
Ratio of actual to theoretical residence time for the
  Kelly-Purdy unit as a function of flow	39
Dimensionless, cumulative residence time distribution
  function for the Kelly-Purdy unit (2.54 cm water depth).  40
Dimensionless, cumulative residence time distribution
  function for the Kelly-Purdy unit (6.35 cm water depth).  41
Measured UV light intensities at various locations in the
  Kelly-Purdy irradiation chamber	  42
Average intensity of UV light as a function of water
  depth and the extinction coefficient 	
44
Typical data used to determine the extinction
  coefficient	45
Comparison between the observed and theoretically predicted
  relationship of the extinction coefficient and trans-
  mi ttance	  47
                                    vii

-------
                             FIGURES (continued)
Number                                                               Page
 15       UV dose as a function of flow and the extinction
            coefficient for the Kelly-Purdy unit	  48
 16       Demonstration Plant configuration during Run No. Kl,
            August 6, 1974	50
 17       Results of Run No. Kl, August 6, 1974	  52
                                                               j
 18       Pilot Plant configuration for Run No. K2	54
 19       Standard plate count data for Run No. K2	58
 20       Total coliform data for Run No.  K2  ,	59
 21       Fecal coliform data for Run No. K2	:  .  .  21
 22       Extreme-value frequency distribution for standard plate
            count data, Run No.  K2	62
 23       Extreme-value frequency distribution for total  coliform
            data, Run No.  K2	63
 24       Extreme-va1ue frequency distribution for fecal  coliform
            data, Run No.  K2	64
 25       Pilot Plant process configuration during Run No. K3  ...  25
 26       Standard plate count data for Run No. K3 .  . .	69
 27       Total coliform data for Run No. K3	70
 28       Fecal coliform data for Run No. K3	  28
 29       Extreme-value frequency distribution for standard plate
            count data, Run No.  K3	  72
 30       Extreme-value frequency distribution for total  coliforms,
            Run No.  K3	73
 31        Extreme-value frequency distribution for fecal  coliform
            data, Run No.  K3	74
 32       Pilot Plant process configuration during  Run  No. K4  ...   76
 33       Standard plate count data for Run No. K4	   79
 34       Total coliform data for Run No.  K4	80
                                   vm

-------
Number
 35
 36

 37

 38

 39
 40
 41
 42
 43

 44

 45

 46
 47

 48
 .49
 50
 51
 52
 53

  54
                  FIGURES (continued)
                                                           Page
Fecal  col i form data for Run No.  K4	  81
Extreme-value frequency distribution for standard plate
  count data, Run No. K4	83
Extreme-value frequency distribution for total coliform
  data, Run No. K4 ....._	84
Extreme-value frequency distribution for fecal coliform
  data, Run No. K4 . . . -.	85
Pilot Plant configuration during Run No. K5	  86
Standard plate count data for Run'.No. K5	 .'  .  89
Total coliform data for Run No. K5 . .  .	90
Fecal coliform data for Run No. K5	  .  91
Extreme-value frequency distribution for standard plate
  count data, Run No. K5	93
Extreme-value frequency distribution for total coliform
  data, Run No. K5	• •  •	94
Extreme-value frequency distribution for fecal coliform
  data, Run No. K5 .  .  .  .	95
Schematic flow diagram of UV light system  .........  97
Theoretical residence time of UPS UV disinfection unit as
  a function of liquid flow rate	98
RTDF's for the UPS UV unit at 0.63 liters/sec	 .  .  99
RTDF  for UPS unit at  1.89  liters/sec	100
RTDF's for the UPS unit at 3.47  liters/sec	.101
Cummulative  dimensionliess  RTDF's  for the UPS unit   ....  102
Isointensity patterns  for UPS exposure  chamber ......  105
The  effect of  flow  and  extinction  coefficient on the
   calculated UV  dose	  108
Pilot Plant  configuration during Run No.  Ul	'.  .  HO

-------
                             FIGURES (continued)
Number                                                               page
 55       Standard plate count data for Run No. Ul	113
 56       Total coliform data for Run No. Ul	 .  .  .  .  .114
 57       Fecal col i form data for Run No. Ul .  .	115
 58       Extreme-value frequency distribution  of standard plate
            count data for Run No. Ul.  .  .	116
 59       Extreme-value frequency distribution  of total  coliform
            data for Run No.  Ul	 117
 60       Extreme-value frequency distribution  of fecal  coliform
            data for Run No.  Ul .	118
 61        Pilot Plant configuration  during  Run  No. U2	120
 62       Standard plate count data for Run  No.  U2 .  .	123
 63       Total  coliform data for Run No. U2	  124
 64       Fecal  coliform data for Run No. U2	125
 65        Extreme-value frequency distribution  of standard plate
            count  data  for  Run  No.  U2	126
 66        Extreme-value frequency distribution  of total  coliform
            data for Run No.  U2   .  . ,	127
 67        Extreme-value frequency distribution of fecal  coliform data
            for Run  No.  U2	128
 68        Pilot Plant process configuration  for  Run No.  U3	129
 69        Standard plate count data for Run  No.  U3 . . .	133
 70        Total coliform data for Run No.  U3	134
71        Fecal coliform  data for Run No.  U3	135
72        Extreme-value frequency distribution of standard plate
           count data for Run No. U3	136
73       Extreme-value frequency distribution of total  coliform
           data for Run No. U3	 137

-------
                             FIGURES (continued)
Number                                                               Page ;
 74       Extreme-value frequency distribution of fecal coliform
            data for Run No. U3	138
 75       Pilot Plant configuration for Run No. U4, March 6 & 7,
            1975 . . . .	  .139
 76       Extreme-value frequency distribution of standard plate
            count data for Run No. U4	  .143
 77       Extreme-value frequency distribtuion of total coliform
            data for Run No. U4	144
 78       Extreme-value frequency distribution of fecal coliform
            data for Run No. U4	145
 79       Pilot Plant configuration for Run No. U5, March 8 to
            July 14, 1975	146
 80       Standard plate count data for Run No. U5	  .  .150
 81       Total coliform data for Run No. U5	151
 82       Fecal coliform data for Run No. U5	152
 83       UV intensity versus log reduction of total coliforms
            for Run No. U5	153
 84       UV intensity versus log reduction of fecal coliforms
            for Run No. U5	154
 85       Extreme-value frequency distribution of standard
            plate count for  Run No. U5   . .	  .155
 86       Extreme-value frequency distribution of total coliform
            data for Run No. U5	 .  .  .	156
 87       Extreme-value frequency distribution of fecal coliform  -
            data for Run No. U5	157
 88       Pilot Plant configuration for Run No. U6, July  15 to
            August 3, 1975.	159
 89       Standard plate count data for Run No. U6	  .  .162
 90       Total Coliform data for Run No. U6   .	  .  .163
 91       Fecal' coliform data for Run No. U6   ...........  .164
                                     xi

-------
                             FIGURES  (continued)
Number                                                               Page
 92       UV intensity versus log reduction of total coliform data
            for Run No. U6	165
 93       UV intensity versus log reduction of'fecal coliform data
            for Run No. U6	166
 94       Extreme-value frequency distribution of standard plate
            count for Run No. U6	168
 95       Extreme-va1ue frequency distribution of total coliform
            data for Run No. U6	169
 96       Extreme-value frequency distribution of fecal coliform
            data for Run No. U6	170
 97       Pilot Plant configuration for*Run No. U7; August 8 to
            October 5, 1975	:	171
 98       Standard plate count data for Run No. U7	175
 99       Total coliform data for Run No. U7	,. 176
100       Fecal coliform data for Run No. U7	177
101       Extreme-va1ue frequency distribution of standard plate
            count data for Run No.  U7	'.	178
102       Extreme-value frequency distribution of total coliform
            data for Run No. U7	 179
103       Extreme-value frequency distribution of fecal coliform
            data for Run No. U7	.180
104       Pilot Plant configuration for Run No. U8; October 7
            to November 31, ,1975	 181
105       Standard plate count data for Run No. U8	185
106       Total coliform data for Run No. U8	186
107       Fecal coliform data for Run No. U8	187
108       Extreme-va1ue frequency distribution of standard plate
            count data for Run No.  U8	 188
109       Extreme-va1ue frequency distribution of total coliform
            data for Run No. U8	189
                                    xn

-------
                             FIGURES (continued)
Number
110       Extreme-value frequency distribution of fecal  coliform
            data for Run No.  U8	.190
111       Location of sampling stations, UV Virus Runs	192
112       Meter readings with tap water from cold start 	203
113       Meter readings after five minutes of purging with tap
            water	204
114       Meter readings on activated sludge effluent	205
115       Total coliform log reduction as a function of the
            calculated dose	-225
116       Fecal coliform log reduction as a function of the
            calculated dose .	226
117       F2 bacteriophage log reduction as a function of the
            calculated dose . .		227
118       Poliovirus log reduction as a function of the
            calculated dose .	228
119       Virus reduction versus detention time  .	 .  .229
120       The effect of turbidity on percent transmittance	231
121       Correlation between percent transmittance and total
            suspended solids  	232
122       Correlation between transmittance at 254 nm and  COD  .  . .  .233
123       Correlation between transmittance at 254 nm and  TOC  .  . .  .234
124       Correlations between ammonia-N and  COD and TOC	236
125       Correlation between NhU-N and percent  transmittance
            at  254  nm	T  .	237
1.26       Correlation between organic nitrogen and percent
            transmittance at 254 nm	238
127       Effect  of cleaning the quartz sleeves  on UV dose	239
128       Effluent  suspended solids versus  effluent fecal  coliforms .240
                                   xm

-------
                              FIGURES (continued)
 Number
 129
 130
 131
 132
137

138

139
                                                           Page
Turbidity versus fecal coliforms 	 241
UV intensity versus percent transmittance.' ......... 243
  i.
Observed UV intensities versus  log  reduction  in total
  coliforms	244
Observed UV intensities versus  log reductions in fecal
133
134
135
136
Effluent fecal coliforms as a function of UV dose . . .
Relationships between observed log reductions in fecal
coliforms and the calculated and indicated UV dose . .
Relationships between observed log reductions in total
coliforms and the calculated and observed UV dose . .
Fecal col i form densities with an exceedence probability
. 246
. 248
. 249
  greater than or equal  to 5 percent versus indicated
  dose	250
  Process  configuration  for  light-dark study: August
    1-19,  1975	253
  Time-series  plots of fecal  coliform data  for the light-
    dark study	   257
  Fecal coliforms  in open  basin  versus covered basin  .  . . 258
                                    xiv

-------
                                  TABLES
Number
  1
  2
  3
  4
  5
  6
  7
  8
 10
 11

 12
 13

 14
 15

 16
Summary of existing water supply for year 2020  	
Influent metals concentration for fiscal  year 1975  .  .  .
Page
  3
 12
Characteristics of-raw wastewater at the White Rock STP
  for the period of October 1, 1974 through September
    30, 1975 .  . . .	.  . .	    13
Technical data for the No. 1 aeration basin	.    20
Technical data for the No. 1 final clarifier	    22
Media specifications  for the No.  1 mixed-media filter.  .    24
Results of Run No. Kl, August 6,  1974	    51
Performance of No. 1  activated sludge system during
  Run No. K2	    55
Effect of flow rate on reduction of total coliforms,
  fecal coliforms, and standard plate count by ultra-
    violet light (Special Study of Sept. 14, 1974) ...    56
Results of Run No. K2	    57
Summary of No. 1 activated sludge system performance
  during Run No. K3	    67
Results of Run No. K3	    68
Performance of No. 1 activated sludge system and No. 2
  multimedia filter during Run No. K4	    77
Results of Run No. K 4 .  .	    78
Performance of the No. 1  activated Sludge System during
  Run  No. K5 .	    87
Results of Run No. K 5 .  .  .  .	 .    88
                                     XV

-------
                             TABLES (continued)
Number                                                               page
                                                                  i
 17       Performance of the No.  1  activated sludge system during
            Run No. Ul	  .  •	  T11
 18       Results of Run Ul   	"112
 19       Performance of activated  sludge and tertiary settling
            basin during Run No.  U2  .  .	121
 20       Results of Run U2	•  •  •  •  •  122
 21       Performance of the activated  sludge system and multimedia
            filter during Run No. U3 .	131
 22       Results of Run U3	132
 23       Performance of the No.  1  activated sludge system and
            upflow clarifier during Run No.  U4	141
 24       Results of Run U4	142
 25       Performance of the No.  1  activated sludge system during
            Run No. U5	147
 26       Results of Run U5	149
 27       Performance of the activated  sludge system and multimedia
            filters during Run No.  U6	150
 28       Results of Run U6	161
 29       Performance of the No.  1  activated sludge system  and
            multimedia No. 1  filter during Run  No.  U7	172
 30       Results of Run U7	173
 31       Performance of the No.  1  activated sludge system
            during Run No. U8	182
 32       Results of Run U8	  183
 33       Operating summary  of the  first  UV  virus run	194
 34       Results of chemical  analyses  of grab  samples averaged
            over all  four flow rates in the  first virus  run   ....  195
 35       Summary of the bacteriological  data from  the first virus
            run	196
                                    xvi

-------
                             TABLES (continued)
Number

 36


 37


 38


 39

 40

 41

 42


 43


 44


 45


 46


 47

 48

 49

 50


 51


 52


 53
                                                           Page

Steady-state titers of F2 bacteriophage controls in
  first virus run . ..... .......... ..... 197

Titers of F2 bacteriophage exposed to ultraviolet radiation
  at different flow rates in first virus run

Titers of poliovirus type I exposed to UV radiation at
  different flow rates in first virus run ...... .  .   .20°
Summary of results from first virus run .  .  .  .......

Summary of dose-related data for the first virus run  .  .  .

Operating summary of second UV virus run  	

Results of chemical analyses of grab samples averaged over
                                                           202
  all four flow rates in the second virus run
Summary of the bacteriological data from the second
  virus run	,	
                                                           208


                                                           209
Steady-state titers of F2 bacteriophage controls in first
  virus run	 . .210

Titers of F2 bacteriophage exposed to ultraviolet radiation
  at different flow rates in 'second virus run	 .211

Titers of poliovirus type I exposed to UV radiation at
  different flow rates in second virus run	.212

Summary of results from second virus run  	214

Summary of dose related data for the second  virus run  . ; .215

Summary of operations of virus run no . 3	 . .216

Results of chemical analyses of grab samples averaged over
  all four flow rates in the third virus run  .	217
Summary of the bacteriological data from the third
  virus run	."  .  .
                                                           .219
Steady-state titers of F2 bacteriophage controls  in
  third virus run	220

Titers of F2 bacteriophage exposed to  ultraviolet radiation
  at  different flow rates in  third virus run   	22'
                                    xvn

-------
                             TABLES (continued)
Number
 54
 55
 56
 57
 58
 59
 60

 61
                                                           Page
liters of poliovirus type I exposed to UV radiation at
  different flow rates in third virus run ....... .  .222
Summary of results from third virus run . .  .	223
Summary of dose-related data for the third virus run  . .  .224
Summary of the data for the Kelly-Purdy disinfection
  unit	251
Summary of the data for the UPS disinfection unit	251
Process control and operation summary light-dark study:
  August 1-19, 1975	  .254
Chemical and bacteriological summary of the light-dark
  study, August 1-19, 1975	255
Analysis of mean coliform densities in light-dark study .  .259
                                  xvm

-------
                             ACKNOWLEDGEMENT

      This report is the product of cooperative effort by so many people  ;•.
 that individual  mention becomes unfeasible.  To the entire group of Dallas
 Water Utilities  personnel  and the Texas A&M Research Foundation members
 we want to convey our gratitude for their support and cooperation.

      In particular we want to offer our thanks to Mr. Henry J.  Graeser,
 Director, Dallas Water Utilities, who originated the plan and inspired the
 effort; and Dr.  I.M. Rice, Assistant Director for Engineering and Planning
 during the program and presently the successor to Mr. Graeser as Director,
 for grant preparation and financial planning in addition to general grant
 administration.

     We gratefully acknowledge the cooperation offered by Dr. Gerald Berg
and his staff from the Environmental Protection Agency, Cincinnati,  Ohio,
including especially Dr. Robert Safferman and Mr. Donald Berman for design-
ing and helping conduct the poliovirus experiments.
                                     XTX

-------

-------
                                  SECTION 1
                                INTRODUCTION
     As shown in Figure 1, the City of Dallas in North Central  Texas is
situated in the upper Trinity River Watershed.  This region must be classified
as naturally water deficient; in fact, there is only one natural lake in the
whole of the State of Texas.  In the upper Trinity River Basin, average annual
precipitation ranges from about 100 centimeters per year in the eastern
portions to 50 centimeters per year in the westerly sections.  In general,
average annual precipitation and evaporation rates are about equal, and
drought periods in excess of 60 days are not uncommon.  These factors
combine to make water a very valuable resource in North Central Texas.

     The City of Dallas derives its drinking water supply from the extensive
reservoir network shown in Figure 1, and estimates indicate that an
adequate supply of fresh water exists to meet demands anticipated to the
year 1995.  The water supply is derived from six reservoirs on three
watersheds, with estimated safe yields for the year 2020 in Table 1.

    , During the middle 1950's North Central Texas experienced a protracted
drought.  The City of Dallas investigated two alternate sources^of water
to augment its dwindling supply.  One was to pump water from the highly
mineralized Red River into the existing reservoir system, and the other
was to  utilize the waters of the West Fork of the Trinity River which
carry a considerable amount of pollution from the wastewater effluents of the
City of Ft. Worth and of the mid-cities between Ft. Worth and Dallas.  In
1955 the thought of  "drinking someone else's sewage"  did not meet with wide
acceptance from either the  general public or the City Council and the
decision was  made t9 import water from the Red River.  Consequently,  the.City
survived the  drought by utilizing the Red River over  a three-year period —
but not without effect  (1). A most important  point  emerging  from this  period
is  that as early as  1955  a  major American city had  seriously evaluated the
possiblity1 of an  indirect wastewater  reuse as an alternate supply of  drinking
water.

      With  respect to total  water resources management in the upper  Trinity
River Basin,  indirect  but intentional wastewater  reuse  is  of considerable
 importance to the City of Dallas.  Both the  Bachman and  the  Elm Fork  Water.
 Purification  Plants  of the  City of Dallas withdraw  water from  the  Elm Fork
of the Trinity River.   Estimated water  supply for the year 2020 includes
 329,000 cu m per  day (86.8 MGD) from  Garza-Little  Elm Reservoir, 38,000 cu  m

-------

-------
        TABLE 1.  SUMMARY OF EXISTING WATER SUPPLY FOR YEAR 2020
Reservoir and Basin
Estimated Safe^Yield
MGD          MVday
Trinity River Basin
     Lewisville (Garza-Little Elm)
     Grapevine
     Ray Hubbard
     Lavon
     Return flows
86.8
10.0
55.4
10.0
41.3
329,000
38,000
210,000
38,000
156,000
Sabine River Basin
     Tawakoni
163
616,000
Neches River Basin
     Palestine
 102
386,000
TOTAL AVAILABLE SUPPLY
468
1,773,000

-------
 per day (10.0 MGD)  from Grapevine  Reservoir,  and  156,000  cu m per day
 (41.3 MGD)  in return  flows.   The return  flows  are composed exclusively of
 wastewater  effluents  discharged approximately 33  kilometers (20 miles)
 upstream from the water treatment  plant  intake structures.  The return flows
 constitute  30 percent of the  523,000  cu  m  per  day (138 MGD) that will be
 used as a source  of drinking  water, and  represent a valuable water resource.
 However, travel time  from the point of discharge  to the most distant intake
 structure is  less than one day, allowing little time for  significant natural
 purification  to occur.   Under these circumstances, the City of Dallas has
 considerable  interest in water and wastewater  treatment technology since it
 is  imperative that  the public health  be  safeguarded.

      Bearing  in mind  the occasional shortages  of  rainfall and the conse-
 quences of  a  prolonged drought, and knowing also  that new sources of supply
 are disappearing  rapidly, the City of Dallas,  has  felt for some years that
 the building  of new reservoirs on  streams  to augment the  potable water
 supply would  become virtually impossible.  The fact that eastern cities in
 particular  have been  able to  produce  potable,  palatable waters from rivers
 that  have been increasingly polluted  has led to consideration that some
 day the recycle of  wastewaters, properly treated, may actually be a means
 of  survival in the  semiarid Southwest and  West.

      Dallas has for several years  pursued  studies of wastewater and water
 treatment methods that might  lead  to  the production of potable water
 completely acceptable for all  uses.  Because there are technological, legal,
 and esthetic  considerations to be  satisfied, Dallas has followed a cautious
 policy that more  than is  known now must  be learned about constituents of
wastewater proposed for  reuse.  The three  large areas still  relatively unsat-
 isfied are viruses, heavy metal,   and organics.  Any or all  of these may be
 found  to  some degree  in  recycled waters, and enough work must be done to
 satisfy all parties that  no significant  risk attaches to either health or
 comfort from  the  prolonged use of  such waters.  Dallas will  want to know
 that the most stringent standards  can be met before proposing recycle.

 BACKGROUND OF THE DALLAS  PROGRAM
     As a result of the previously described circumstances,  namely,  the
 consideration of using the heavily polluted West Fork of the Trinity in
 1955,  the present indirect use of  upstream effluents, however minor,  and the
economics dictating that maximal  amounts of wastewaters be salvaged if
practical, the Dallas Water Utilities have pursued an active,  viable waste-
water  reuse research program since June 1970.

     The Demonstration Plant of the Dallas Water Reclamation Research
Center, which is described in detail  in a subsequent section of this  report,
was built and brought.on-line in late July 1969.   Equipment  check-out
and/or modification consumed at least nine months, and the facility  was  in
service for almost a year before the staff considered it to  be truly  opera-
tional.  Additionally, a laboratory and administration building was  con-
structed to provide laboratory capabilities for the research program
and office space for several  different Water Utilities Department  activities.
The laboratory facilities were occupied in the Spring of 1971.
                                     4

-------
     The funds for construction of the laboratory and administration build-
ing were provided by Federal  Grant No. WPC Tex 588, to the amount of
$571,093 of which the City provided 52 percent.  The Demonstration Plant was
constructed under Grant No. 17080 EKG from the Environmental  Protection
Agency (EPA) for a total cost of $689,015 of which the City of Dallas pro-
vided 45 percent.

     From start-up until January 1971, the Dallas Water Reclamation Research
Center operated on additional funds provided in Grant No. 17080 EKG for
the study of various water reclamation unit processes.  The initial phase
of the project was directed at evaluating different sequences of unit
processes to expand and up-grade the City's Central Wastewater Treatment
Plant. '                             '

     At the conclusion of these studies additional unit processes were in-
vestigated by utilizing a number of different wastewaters as influents to
the unit processes at the Demonstration Plant.  For the most part the
investigations were short term  (2 to  4 weeks), and these data have been
presented in the  Final Report for the project  (2).

     From June 1972 through  February  1974 the  Research Center evaluated the
removal of metals and viruses through three different advanced wastewater
treatment sequences.  Concentrations  for over  twenty different metals were
evaluated in the  influent  and effluent of each unit process over the pro-
tracted study period.  This  project was funded through Grant No. S-801026
from EPA for a total cost  of $200,287, of which the City shared 41 percent
 (3, 4).

RESEARCH PRESENTED  IN THIS REPORT
     The present project,  Grant No.R-803292, was  started  in June 1974  and
 concluded during November  1975.  This project  was conducted in two  phases,
 both of which had distinctly different research objectives.

 Characterization for Potable Reuse

      In the first part  of this research  effort the selected treatment  train
was  operated to  produce the  best possible quality of effluent.   This product
 water  was  then  examined in the light of  present and proposed  standards
 for  its  potential reuse as a potable water.   The  objectives were:

                   1.  Production of a high-quality product water.

                   2.  Monitoring of the effluent for parameters  to  indicate
                      whether potable quality according to available criteria
                      had been achieved.

                   3. Evaluating the reliability of the sequence of unit
                      processes utilized in the study as it related to
                      wastewater reuse.

-------
Ultraviolet Disinfection of Municipal Effluents

     Phase two of this project was a study of the feasibility of utilizing
ultraviolet light as a disinfection process for municipal wastewater
effluents.  The objectives of the ultraviolet disinfection project were
as follows:

                    1.  Evaluate the feasibility of using ultraviolet light
                        as a disinfection process for municipal wastewater
                        effluents.

                    2.  Determine-whether any additional treatment is
                        required to achieve a consistent fecal coliform
                        density less than 200 per 100 mis.

                    3.  Evaluate the effect of photo-reactivation.

     This research effort was conducted for a total  cost of $262,754 of which
the EPA provided 27 percent of the funds and the City of Dallas the
remainder.

     Since the two portions of the study had such different objectives, this
report discusses only the research effort conducted during the second phase
of the project.  The work performed during the first phase of the reuse
portion of the project has already been reported (5).

-------
                                SECTION 2

                               CONCLUSIONS

     This research effort leads to several conclusions  that bear upon
disinfection technology.

     This effort was principally a feasibility study to evaluate ultraviolet
light irradiation as an alternate disinfection technique for secondary
effluents.  The conclusion of all  personnel  associated with the project
is that UV light does present a viable disinfection process for
secondary effluents.  This research indicates that an activated sludge
effluent can be adequately disinfected with UV light to comply with the
200 fecal coliforms per 100 mis disinfection criterion..
         *                                          -
     Shallow tray exposure chambers similar to the Kelly-Purdy system  used
in this research should not be used for secondary effluents.  Solids
deposition in the bottom of the exposure chamber constituted a problem of
major proportions, and no economical solution to this problem is apparent at
this time.  Additionally^ large surface areas are required.  For instance,
a flow of 43.8 liters per second (1 MGD) would require an exposure chamber
with a surface area of approximately 58 sq. meters (625 sq.ft.).  Finally,
the shallow-tray chambers are not capable of taking advantage of light
emmission. from the full 360 degrees around the lamps.  This single factor
results in the shallow-tray design being less energy efficient than the
exposure chambers which utilize submerged lamps.

     Total suspended solids within the range 5 to 50 mg/1 had very little or
no effect on the ability of the water to transmit UV light.  The effect of ,
TSS on UV light transmission outside of this range was not'investigated.

     The studies conducted with attenuated Type I Poliovirus indicate  that
ultraviolet light disinfection can also be effective against animal
viruses.  The data indicate that the virus kill obtained is a function of
the UV dose used, and the correlation is excellent.

     Virus,inactivation studies conducted with an F2 coliphage, and in con-
junction with the poliovirus research, indicates that the coliphages are
good indicators of animal virus response to UV disinfection processes.

     Light transmission measured at 254 nm with an ultraviolet spectrophoto-
meter provides data which are absolutely essential to determination of UV
dose.  Additionally, of the two principal factors affecting the .actual UV
dose, UV transmissibility and flow, flow is the least significant.  Several

-------
 efforts were made  to quantify  the effect of flow on observed kill, and the
 results were generally disappointing.   For hydraulic transients to be of
 much  real  significance the  actual residence time must be changed by about
 one log, and in  practice  this  is very  unlikely to occur.


      The turbidity of the effluents being disinfected did not affect the
 ability of the liquid to  transmit UV light.  This conclusion is valid
 within the range of turbidity  values encountered during this research
 effort (0.5 to 12  NTU), and no effort  was made to validate this conclusion
 at higher  turbidity values.


      Ultraviolet light disinfection systems, employing submerged lamps
 represent  the preferred exposure chamber configuration.  These systems
 are capable of utilizing  all radiation  emitted by the lamps, and are
 therefore, more  energy  efficient than  the shallow-tray design.


      Over  the ranges of the many variables encountered in the project, UV
 dose  was the one factor that had the most pronounced effect on the observed
 kill  and the number of organisms found  in the effluent.  In order to
 achieve adequate kills on a typical secondary effluent, the UV dose neces-
 sary  appears to  be in the range of 30,000 to 35,000 microwatt-seconds per
 square centimeter.  As used in this report, dose is the average UV light
 intensity  in the water being disinfected multiplied by the theoretical
 residence  time of  the exp'osure chamber  used.


     The monitoring of UV intensity at  some point in the exposure chamber
 is absolutely essential for process control.  This measurement can be
made  easily and  routinely and  is directly  related to the UV dose the
effluent is receiving.


     The development of slime around the UV lamps must be controlled to
prevent a decrease in UV dose.  Routine UV intensity measurements can be
used  to determine when the  unit must be cleaned.   This did not represent
a major problem  during the  project, and a cleaning interval  of about two
weeks seems reasonable, although the cleaning frequency will  be dependent
on water quality to a certain degree and perhaps  temperature.


     As with any other disinfection process, the  efficiency of UV light
is a function of the quality of the water to be disinfected.   This
research had demonstrated conclusively  that UV light is more effective
on a nitrified activated sludge effluent than on  a non-nitrified effluent.
*he presence of  ammonia nitrogen should have nothing to do with the UV
kill mechanism,  and one must conclude that the higher quality effluent
                                    8

-------
produced as a direct result of nitrification has improved UV transmission
characteristics, which results in a higjier effective UV dose.

-------
                                  SECTION 3

                               RECOMMENDATIONS

     Additional studies should be performed with larger'systems to permit
accurate estimation of the cost associated with the operation of UV
disinfection systems.  The manufacturer of the lamps estimates the bulb
life at 7000 hours (about 10 months), but the UV output decreases as lamp
operating time increases, and the effective life is unknown.  The effect
on dose of using different materials of construction is also not known to
the authors.  Also, the actual amount of operator time required by these
systems can only be determined more accurately by large scale operation
over determined extended periods of time, since seasonal  effects on water
quality may be a factor.

     The UV dose data in this report have been presented in units of energy
per unit area (pwatt-sec/cm2).  In an enclosed chamber such as the UPS
unit, there is no plane surface upon which .the UV energy impinges.  With
this type of geometry it is not possible to directly measure the average
UV intensity in the water being irradiated, and it is equally impossible
to make an accurate dose calculation.  A very real  need exists to develop
a reliable procedure for directly measuring UV dose that is relatively
simple and rapid.

     Nitrified operation of a biological wastewater treatment facility has
again been shown to produce a "superior" effluent — in this case, an
increased transmissibility for UV light.  The longer sludge ages associated
with nitrified operation apparently result in a more diverse biological
population which, in turn, breaks down more of the organic materials, rend-
ering the medium more transmissible to UV.  Certainly more information
should be learned about these phenomena, so that greater use-can be made
of the many advantages they offer in an environmental context.
                                     10

-------
                                SECTION 4

                     DESCRIPTION OF RESEARCH FACILITIES
CITY OF DALLAS COLLECTION SYSTEM

     The sanitary sewer collection system serving the City of Dallas
consists of 5074 kilometers (3,137 mir) of gravity mains.   During the
fiscal year covered in this report the City abandoned 2.54 kilometers
(1.58 mi) of sanitary sewer main, and added 35.1 kilometers (21.8 mi)
of mains to the collection system.  The best available estimates indicate
that the total length of laterals is between one and two times the
length of the gravity mains, indicating a total collection system length
between 10,000 and 15,000 kilometers (6215 and 9323 miles).  The
City has no combined sewers.

     In addition to the normal domestic wastes discharged to the collection
system, the City has significant contributions from industrial and
commercial establishments.  During fiscal year 1974 the industrial dis-
charges represented 12.1 percent of the total flow received at the
Central Plant, or 21.9xl06 cu m  (5.80x109 gal) per year.  The 221 sign-
ificant industries monitored by the Water Utilities Department
discharged a total of 1.70xl07 kg of BOD. (3.74x10? Ibs) and 1.38xlO/ kg
(3.03x107) of total, suspended solids(TSS) .to the collection system
during fiscal year 1974.  When expressed in terms of concentration,
the BOD5 and TSS of the industrial discharges were 773 mg/1 and 626 mg/1,
respectively.  The BODs discharge represented 37.7 percent of the total
load entering the Central Plant, while the industrial TSS discharges
represented 30.7 percent of the total TSS load.

     The activities of commercial establishments, which include restaurants,
wholesale food preparation' facilities, and service facilities (principally
car washes) substantially impact on wastewater characteristics.   The
predominant effect of commercial activities that one would normally expect
would be to increase the organic and solids loadings; however, certain
of the service activities  (car washing) can have appreciable impact on
metals concentrations.

     The concentrations of certain metals in the influents of the Dallas
and White Rock plants are given in Table 2.  The column headed  "combined"
is a calculated, flow-weighted concentration for all wastewaters arriving
at the Central Plant.  Typical data for the raw wastewater entering the
White Rock STP are presented  in Table 3, and upon inspection,the waste-
                                    11

-------
TABLE 2.  INFLUENT METALS CONCENTRATIONS FOR FISCAL YEAR  1975

Metal
Arsenic
Ban' urn
Boron
Cadmi urn
Chromi um
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Silver
Zinc
	 Q
Dallas STP
0.005
0.7
0.51
0.035
0.34
0.23
0.44
0.10
0.0016
0.21
0.017
0.580
oncentration (mg/1)-
White Rock STP
0.020
0.7
0.39
0.016
0.11
0.14
0.19
0.08
0.0012
0.09
0.016
0.227

Combined
0.016
0.7
0.42
0.020
0.17
0.16
0.25
0.08 .
0.0013
0.12
0.016
0.311
                                  12

-------
TABLE 3.  CHARACTERISTICS OF RAW WASTEWATER AT THE  WHITE  ROCK  STP  FOR
          THE PERIOD OF OCTOBER 1,  1974 THROUGH SEPTEMBER 30,  1975*.
          Flow
          Grit
          Total Solids
409,000 cu.m./day
(108.0 mgd)

0.0127 liters/cy.m.
(1.69 cu.ft./106 gal)

844 mg/1
 n
365
300
350
Settleable Solids
Total Suspended Solids
COD
BOD5 . ?
' NH3-N

Org. N
N02 & N03-N
pH
8.4 mlVl
209 mg/1
417 mg/1
191 mg/1
16.4 mg/1

13.5 mg/1
0.3 mg/1
7.2
350
350
350
350
350
t
350
350
350

*Arithmetic means of 24-hour flow-weighted, composite samples  collected
 daily.
                                    13

-------
water appears to be representative of domestic wastewaters.  Most of the
industrial waste discharges enter the Dallas STP, and for this reason
the wastewaters entering the White Rock plant are more suitable for
wastewater reuse studies.

TREATMENT FACILITIES

     The demonstration plant of the Dallas Water Reclamation Research Cen-
ter is colocated with the White Rock Sewage Treatment Plant (STP) at
the City of Dallas' Central Wastewater Treatment Facility.  The Central
Plant complex is situated on the south bank of the Trinity River
approximately five kilometers (3 miles) south of the City's central
business district.  The Central Plant actually consists of the three
treatment facilities described below, two of which are trickling filter
plants.  The third facility is completely-mixed activated sludge
followed by tertiary mixed-media filtration.

Dallas Sewage Treatment Plant      -

     The Dallas STP is the oldest wastewater treatment facility operated
by the City of Dallas.  This single-stage, standard-rate, trickling
filter facility consists of four bar screens and grit channels, twenty-
four Imhoff tanks which are operated as primary clarifiers, two rec-
tangular primary clarifiers, sixteen standard-rate trickling filters
which are 53 meters (174 feet) in diameter, and three final clarifiers.

White Rock Sewage Treatment Plant

     The White Rock STP is a two-stage, high-rate trickling filter facility
without intermediate clarification.  The plant consists of two bar
screens and grit channels, six rectangular primary clarifiers, four
first-stage, high-rate trickling filters, eight second-stage, high-rate
trickling filters, and four rectangular final clarifiers.  All trickling
filters are 53 meters (174 feet) in diameter, and contain a maximum of
2.3 meters (7.5 feet) of media.

Tertiary Treatment Complex

     Under normal flow conditions the effluents from both the Dallas and
White Rock facilities are discharged to the tertiary treatment complex
prior to discharge into the Trinity River.  The tertiary complex consists
of twelve completely-mixed, activated sludge aeration basins and twelve
final clarifiers, followed by fourteen mixed-media gravity filters.  The
physical relationship between the individual facilities, present at the
Central Plant complex can be seen in Plate 1.

Demonstration Plant
     All influents to the demonstration plant are pumped from the White
Rock STP and all effluents and sludges from the pilot pi ant-are returned to
                                   14

-------
                                                         o
                                                         in
                                                         cn
                                                         £=
                                                          O
                                                          O
                                                          O
                                                          fO
                                                          0)
                                                          O)
                                                          S-
                                                          0)
                                                          4J
                                                          (C
                                                          t/)
                                                          fO
                                                          (O
                                                          
-------
 the headworks of the White Rock plant.  As indicated in Figure 2, there
 are a total of five possible influents which can be supplied at a maximum
 flow of 47.3 liters/SRC. (75f)_cmm), with the exception of the raw sewage
 pump that is rated at 18.'9 liters/sec(300 gpm).   The discharges from
 all pumps are routed to a valve station at the White Rock plant, from
 which the flow is directed to the pilot plant through one of three main
 influent lines.   Each influent line services one of the major treatment
 modules (biological, chemical, or physical) at the demonstration plant.

      As indicated in Figure 3, which is a piping diagram of the major
 components of the demonstration plant, the facility is relatively complex
 and very flexible.  The following unit processes are present at the pilot
 plant,  and many are identified  in Plate 2 (aerial  photograph of the
 demonstration plant).
      1.  No.
1
completely-mixed activated sludge system.
      2.  No.  2  completely-mixed  activated  sludge  system.

      3.  Upflow clarifier.

      4.  Gravity-flow, mixed-media  filter.

      5.  Gravity-flow, dual-media filter.

      6.  Gravity-flow activated  carbon contactors  (2 each).

      7.  Chlorine contact basins (2 each).

      8.  Ozone  generator and contacting system.

      9.  Reverse osmosis deminerialization unit.

    10.  Ultraviolet light  disinfection units (2 each).

    11.  Chemical storage and feeding equipment.


 No.  1 Activated Sludge System—
      The No. 1 activated sludge system consists of the No.l aeration basin
and the  No. 1  final clarifier, and return sludge and effluent pumps.'
The return sludge pump has a practical operating range of 6-.3 to 47.3
liters/sec.  (100 to 750 gpm), while the effluent pump has an operational
range of 3.2 to 20.5 liters/sec. (50 to 325 gpm).

No. 1 Aeration Basin--
     The No. 1 aeration basin is a circular mild-steel tank, erected above
ground;  technical  data for this unit are summarized in Table 4.  Several
different types of mixing and oxygen transfer equipment have been
evaluated in this basin.   The original equipment was supplied by Dorr-
Oliver, which consisted of an 18.6 kw (25 HP) surface/submersed  turbine
variable-speed mixer and an 11.2 kw (15 HP) variable-speed air compressor.
                                    16

-------
oo
UJ
o
o
00
                                                      CD
                                                      o:
                                                                                       (C

                                                                                       51
                                                                                       D-


                                                                                       O)
                              00
                                               17

-------
                                                               C
                                                               rtS
                                                               -4-5
                                                               03
                                                               0)
                                                               S-
                                                               O)
                                                               •M
                                                               C
                                                               OJ
                                                               O
                                                               O
                                                               to
                                                               
-------
                                                       (C
                                                      r—
                                                      O-
                                                       O
                                                       £
                                                       in
                                                       E
                                                       O
                                                       a;
                                                      ca
                                                       a)
                                                      .c
                                                      4J
                                                       Q-
                                                       fO

                                                       01
                                                       o

                                                       o

                                                       Q-

                                                       i—
                                                       CO
                                                       a>
                                                       eC
                                                       OJ

                                                       O)

                                                       R3

                                                       D.
19

-------
 TABLE 4.   TECHNICAL  DATA  FOR  THE  NO.  1 AERATION BASIN
 Diameter


 Depth


 Volume
Theoretical Residence Time at
   6.3 liters/sec  (100 gpm)

Total Nameplate Power Input
Maximum Power Level
Air Flow
7.62 m
(25.0 ft.)

3.66 m
(12.0 ft.)

560 m3
(45,000 gal)
7.5 hours

29.8 kw
(40 HP)

0.053 kw/m3
(0.89 HP per 1000 gal)

118 or 235 liters/sec*
(250 or 500 scfm)*
*Depending upon number of compressors used.
                        20

-------
 This  system was  removed  and  replaced with  Fiscal in equipment.  Aquarius,
 Inc.  is  the manufacturer of  the  Fiscalin system, which was operated
 throughout  most  of the reuse portion of this  grant.   During the
 UV phase of this research effort the Aquarius equipment,was replaced  by
 a "jet aeration" system  manufactured by Pentech/Houdaille.

 No.  1  Final Clarifier—
      The No.  1  clarifier is  a circular, mild-steel basin  erected  above.
 ground.   As initially provided by Rex  Chainbelt the  unit  had  peripherial
 feed and square effluent weirs in the  center  of the  tank.  The unit
 was not  originally equipped  with a surface skimmer,  and the center
 effluent weir configuration  made addition  of  a skimmer quite  complex.
 Therefore,  the basin was modified by  removing the  center  effluent weirs
 and bolting a new peripherial effluent weir to the inside of  the  existing
 influent baffle skirt.   A skimmer and  scum collector were then* fitted to
 the basin.

      Sludge is removed by the head differential between  the water surface
 in the clarifier and the return sludge pump well,  via a  single armed
 header.   Technical data  for this clarifier are presented  in  Table 5.

 No. 2 Activated Sludge System—                             3         _
      The No.  2 activated sludge system consists of the 28.4  m  (7500  gal)
 completely-mixed aeration basin, and a three-hopper Smith and Lovelace
 final clarifier.  The aeration basin has a diameter of 3.14  m (10.3 ft.)
, and a side water depth of 3.66 m  (12.0 ft.).  Mixing and oxygen transfer
 is affected by diffusers, and the maximum air flow is 53.8 liters/sec.
  (114  scfm).

 .Upflow  Clarifier—
       An  Infilco Densator is  the  chemical treatment unit used at the
 demonstration plant.  The unit  consists of the main tank, 5.5 m in
 diameter and 5.5 m deep, and an  inner cylinder that serves as the rapid
 mixing  and flocculation  zones.   Influent enters the top of the inner
 cylinder and the main tank  serves as the  upflow clarification compartment.

       Energy  input  for mixing and flocculation  is supplied via independent
  turbine-type mixers, each of which is equipped with a U.S. Electric
  Varidrive  that  has  a 10 to.l turndown capability.

       A  2.54  cm  x 10.2 cm (1  in.  x 4 in.) .steel  fluidizer bar  is  used
  to prevent the  sludge from  over compacting.

  Chemical Storage and Feed Equipment-

       Facilities are present at the demonstration  plant to store  and  feed
  the following  chemicals:

            1.  Hydrated lime                 ,

            2.  Hydrated aluminium sulfate
                                     21

-------
 TABLE  5.   TECHNICAL  DATA  FOR  THE  NO.  1  FINAL  CLARIFIER
 Diameter


 Depth


 Volume


 Surface Area


Weir Length


Surface Overflow Rate
at 6.3 liters/sec (TOO gpm

Weir Loading at
6.3 liters/sec (100 gpm)
  9.14 rri .  •
  (30.0 feet)

  3.66 m  '
  (12.0 feet)

  240 m3
'  (63,500 gal.)  '

  65.7 m2  9
  (707 feetr)

  27.1 m
  .(89 feet)

  8.50 m3/m2/day
 (2.4 gpd/ft.z)

  20.1 m3/m/day
 (1620 gpd/ft.Z)
                              22

-------
          3. Ferric chloride

          4. Dry polyelectrolytes

          5. Liquid polyelectrolytes

          6. Activated silica

          7. Powdered activated carbon

          8. Chlorine

     All coagulants and coagulant aids, with the exception of lime, can be
fed. to either aeration basin or final clarifier, to the upflow clarifier,
or in front of the filters or carbon contactors for use as filtration aids.
The lime slurry can be pumped to either of the activated sludge systems
or the upflow clarifier.

No. 1 Mixed-Media Filter—
     The No. 1 filter initially consisted of 0.91 meters (3.0 feet) of
sand overlayed with 0.30 meters (12 inches) of anthracite.  The influent
flow was split equally between the top and bottom of the filter bed,
and the effluent was withdrawn through a mid-bed collector located 15 cm
(6 inches) below the sand-anthracite interface.  The filter performance
can be characterized as having been generally good; however, structural
deficiencies with the mid-bed collector resulted in frequent maintenance
and the unit was converted to a conventional gravity-flow filter.

     When the filter was rebuilt media supplied by Neptune Microfloc was
utilized and the media specifications are summarized in Table 6.  The
filter has a nominal diameter of 1.21 meters (4.0 feet) and a surface area
of T.17 sq.m. (12.6 sq.ft.).  At a flow of 2.37 liter/sec. (37.5 gpm)
the filtration rate if 175 cu  m  per sq. m. per day (3 gpm/sq.ft.).
Filter backwashing is conventional, and utilizes a surface wash in lieu
of an air scrub.

No. 2 Dual-Media Filter—
     Structually,.the No. 2 filter is almost identical to the No. 1
filter, and this unit is also operated in the conventional gravity-flow
mode.  Media consist of 30.48 cm (12 inches) of sand with a 60.96 cm
(24 inches) anthracite cap.  The filter sand has an effective size of
0.57 mm and a uniformity coefficient of 1.6.  Air scour is normally used
prior to backwash at a rate of 20.2 liters per sec  per sq m  (4 scfm per
sq ft ).

Chlorine Contact Basins—
   ^  The demonstration plant has two- chlorine contact basins which may
be operated in parallel or in series.  Each basin is 5.49 meters, (18.0 feet)
long, 2.26 meters  (7.41 feet) wide, and 0.48 meters (1.58 feet) deep.
                                     23

-------








DC
UJ
5
HH
U_
•a:
»-4
o
UJ
E:
o
UJ
X
£

r~"~

o
z

UJ
.X.
I—
o;
o
u.
CO
0
P
^f
o
t-H
U.
f-H
O
UJ
o.
CO

l^£
1— 1
UJ
E:
10

UJ
_J
CD
*^c
I™"



































































>>1=
4->  «?i-

i — i — Or— ^%

O O O O EC


r— "^J* f*-*. CM ^f* ^~
CT^ **11*^ CM **^» CO "*"x.
• CO • r— • r—
_« +tmmfff ^^ *^_^ f"^ ^_ ^








CO
>
ra

CD


r^ CM
r— CM

!••• n-




^

r**» ^^
00 CM
* • •
O 0

•"*•
E
•
E LO
^J-
co —
^_^
E O
O
CO
CM •*
(£}

r-- r—


















co r>-

CO
«\ y
CU
c/j »
i- O)
fO E
•<-> O •!-
(DO Uu
E

res
CD


CM
CO
•
r~




^

^M
^;
0

'— >
E
•r—

LO
cn


o

CO

•
^j-
CM





















CO

1
co
^*
VI
•o
E
rtJ
CO


LO
CO

r~




£

LO
O
•—


E

CM
CM


O

CO
00
*
LO
LO


















CO
5?*

n
CU
^_)
•r-*
O
s-
r^
[ ^
E
•=C










1/>
CU

O
E
•r—
cr>
CO
v^'
o
ID
o

CD
CD
II

fC
•r-
T3
CU

S-
+J
^~
LJU
4_
O

(— ;
t \
Q.
CU
Q

r—

^_5
.2













24

-------
Each basin has a volume of 5.95 cu.m.  (1570 gal.),  which  results  in  a
theoretical  residence time of 15.7 minutes  at a  flow of 6.31  liters  per  sec.
(100 gpm).  Eleven fiberglass baffles  were  installed in each  basin,  such
that plug flow would be closely approximated.  Dye  studies  have been used
to quantify the hydraulic characteristics of the end-around baffling system,
and observed residence time distribution functions  closely  approximate
theoretical  values.  The chlorination  equipment  is  capable  of a maximum
feed of 22.7 kg (50 Ibs.) of chlorine  per day.

Ultraviolet Light Disinfection Units—
     Kelly-Purdy UV Unit— The Kelly-Purdy  ultraviolet light  disinfection
unit consisted of a shallow-tray exposure chamber of 185  cm (6 feet)
long, 92 cm (3 feet) wide, and 9.2 cm (3-5/8 inches) deep.  Flow
entered and exited the unit on the narrow ends of the unit  as shown
in Figure 4.

     Thirteen 30-watt ultraviolet lamps (G30T8)  were mounted  above
the exposure chamber. The influent flow was measured with an  undulating
disc meter, and water depth in the unit was controlled by changing
the elevation of the effluent weir.  A photograph of the  unit is
shown as Plate 3.

     U.P.S. Model EP-50-- A model EP-50 ultraviolet light disinfection
unit manufactured by Ultraviolet Purification Systems, Inc.,  of Scarsdale,
New York was used during this study.  The unit consisted  of a 53.6  liter
(14.2 gallons) stainless steel cylindrical  chamber  that housed nine
longitudinally-mounted 40-watt UV lamps.  Each lamp was enclosed  by a
quartz sleeve and had an individual ammeter on the  control  panel.  The
effective arc length of the lamps was 76 cm.  The UPS unit  is shown
in Plate 4.

     The UPS unit, as shipped from the factory,  was equipped  with a  water
quality meter.  This meter measured gross ultraviolet light intensity on a
unitless scale.  Through the course of the  investigations,  the water quality
meter proved unreliable and was used only as a general indicator  of UV
intensity.

     All UV intensity measurements during the course of the project were
made with an  IL 500 radiometer manufactured by International  Light, Inc.,
a division of Dexter Industrial Green, Newburyport, Massachusetts.
                                    25

-------
                                                                  Inf
Figure  4.   Perspective drawing of the Kelly-Purdy Unit.
                            26

-------
                                                           -l->
                                                           •r-
                                                           £=
                                                           o
                                                           •£
                                                           o
                                                           t/>

                                                           O
                                                           O)

                                                           cu
                                                           M-
                                                           O
                                                           Q-
                                                           «

                                                           CD
                                                           O
                                                           4J
                                                           O
                                                           CO

                                                           O)

                                                           res
                                                           51
27

-------
Plate 4.  Photograph of the Kelly-Purdy and U.P.S. disinfection units.




                                     28

-------
                                 SECTION 5

                     SAMPLING AND ANALYTICAL PROCEDURES
SAMPLING PROCEDURES

     The sampling procedures described below were utilized for the duration
of this research effort.  Samples for routine wet chemistry analyses were
collected by the operators on duty at the demonstration plant.  Samples
for microbiological analyses were collected by the staff microbiologist or
microbiology laboratory assistants.

Routine Chemistry_Samples

     Samples, for routine wet chemistry analyses were collected by the plant
operators seven days a week at 1 AM, 5 AM, 9 AM, 1 PM, 5 PM, and 9 PM.
Wide-mouth, half-pint plastic bottles were used for sample collection.
These sample bottles were placed in a refrigerator until being transported
to the laboratory, at which time they were composited by the staff chemists.
Since the demonstration plant was operated at hydraulic steady-state, equal
volumes (400 ml) of each of the six grab samples were used for the 24-hour
composite sample.

Mi crobi ologi cal^Sarnples

     Either the staff microbiologist  or the microbiology laboratory
technicians  collected all samples for microbiological evaluation.  The
samples were collected in 125 ml, wide-mouth glass bottles with glass
.stoppers that had been previously dry sterilized at 350°F for one hour.

ANALYTICAL PROCEDURES
     The analytical  procedures used  in  this research effort followed the
 13th Edition  of  Standard Methc
-------
maintained.  These problems resulted in the installation of several
physical flow measuring elements  (orifices, weirs, and venturi sections)
such that accurate flow measurements could be obtained.

Biochemical Oxygen Demand     ~~

     The procedure used for determining the five-day biochemical oxygen
demand was as given in Section 219 of Standard Methods, and the dilution
technique described in paragraph 4 (ii) was utilized.

Chemical Oxygen Demand

     The following procedures were utilized to determine COD values on
the routine samples.  The low level technique was employed for those
samples where the COD was expected to be less than 50 mg/1.

High Level Technique—
     The procedure used was as described in Section 220 of Standard Methods.

Low Level Technique--
     The COD of low level samples was determined by using the procedure
given on page 19 of Methods for Chemical Analyses of Water and Wastes
1971^  Two modifications were made  to the procedure.  The amount of
mercuric sulfate was reduced from 1.0 to 0.4 grams, and the ferrous
ammonium sulfate solution was 0.01 N instead of 0.025 N.

Total Organic Carbon

     All total organic carbon determinations were made by using a
Beckman Model 915 Total Organic Carbon Analyzer.

Total Residue

     Total solids determinations were made in accordance with the pro-
cedure in Section 224A of Standard Methods.

Nonfi1terable Residue

     Total suspended solids determinations were made by employing the
procedure in Section 224C of Standard Methods and using 2.4-cm diameter
glass-fiber filters and Gooch crucibles.

Total Dissolved Solids

     Total dissolved solids were computed by subtracting the nonfil-
terable residue from the total residue.

Total Phosphorus
                    N
     The single reagent method given in Methods for Chemical Analyses of
Hater and Wastes 1971 was used for all total phosphorus determinations.
                                   30

-------
The amount of ammonium persulfate used was increased from 0.4 to 0.5
grams, and the amount of combined reagent was increased from 8 ml to 10 ml.

Ammonia Nitrogen                                       .

     Ammonia nitrogen determinations were made by using an ion-specific
electrode, and the Known Addition Method.  The electrode used was an Orion
Model 95-10.

Total Kjeldahl and Organic Nitrogen

     Total Kjeldahl nitrogen was determined by using an ion-specific
electrode and the Known Addition Method after completing the digestion
phase of the procedure given in Section 216 of Standard Methods.
Organic nitrogen was determined by subtracting the ammonia nitrogen
from the total Kjeldahl nitrogen.

Nitrite Nitrogen

     Nitrite nitrogen determinations were made by using the procedure
described in Section 134 of Standard Methods.

Nitrate Nitrogen

     The phenoldisulfonic acid method, Section 213D of Standard Methods,
was used to determine combined nitrite-nitrate nitrogen.  Nitrate
nitrogen was computed by subtraction of the nitrite nitrogen.

Sulfate

     Sulfate was determined by an indirect atomic absorption spectroscopy
method by adding a known concentration of barium chloride to form a
barium sulfate precipitate.  The barium concentration in solution was then
determined by atomic absorption, and the sulfate concentration determined
by subtraction.

Chloride

     Chloride concentrations were determined by the mercuric nitrate
method described in Section 112B of Standard Methods.

Alkalinity                                               .

     Total and phenolphthalein alkalinity  were determined by using the
procedures given in Section 102 of Standard Methods.

Turbidity

     Turbidity was determined by the  nephelometric method described in
Section 163A of Standard Methods with  a Hach Model 21OOATurbidimeter.
The  standard references were formazin  polymer  suspensions.
                                    31 .

-------
 Color                                          ,

     Color determinations were made  by  the plant operators by  the use of
 a  Hellige Aqua Tester  and platinum-cobalt color disk.

 Microbiological Determi nations

 Standard Plate Count--
     Standard  plate counts were performed as described in section 406 of
 Standard Methods.   Plates were poured with Plate Count Agar  (Difco)
 and incubated  at 35°C  ±0.5° for 24  hours.  Plates were counted on a
 Quebec colony  illuminator with an-American Optical electric  counter.

 Total Coliforms--
     Total  coliforms were determined by the following procedure:
 (a) During the Kelly-Purdy portion of the project the membrane filter
 (MF) procedure was  utilized with Gelman GN6 membranes.  After a sample
 was filtered the membrane was placed on an absorbent pad saturated
 with M-ENDO Broth (Difco).

 (b) During the remaining portion of  the study the Most Probable Number
 (MPN) procedure was used, as described in Section 407E of Standard Methods.
 Lauryl Tryptose Broth  (Difco) was inoculated and incubated at 35QC for
 24 hours.  Positive tubes were confirmed with Brilliant Green Bile Broth
 (Difco).

 Fecal Coliform—
     Fecal coliforms were measured by the following procedures:
 (a) During the Kelly-Purdy portion of the study the MF method was used
 with Gelman GN6 membranes.  Growth medium was MFC Broth (Difco).

 (b) During the remainder of the project period a 5-tube, 4 .dilution MPN
 procedure was  used.

 Virus Determinations

 Stock—
     The stock cultures used for the virus study were the Poliovirus type
 I (vaccine strain), f2 Coliphage, and E^.  coli K12  (f+) .indicator cells.

 Poliovirus Assay—
     One- to four-liter samples were collected during the experimental runs
 in gas-sterilized flexible Cubitainers, capped, and placed in ice.
The samples were shipped by air to Cincinnati  in insulated boxes containing
 ice pre-frozen in water-tight quart-size Cubitainers.   The shipments were
 picked up at the airport and taken to the National  Environmental Research
 Center Cincinnati laboratory by EPA personnel.  All  samples  (except sludge)
were Swinny-filtered with 0.45 u Mi Hi pore filter membranes treated with
Tween 80 and then inoculated onto BGM (Barren  Green Monkey Kidney tissue)
 cell lines using 0.5 ml in .each of 4 bottles for each dilution.  For
 sludge samples, approximately 200 ml  of sample was centrifuged and a 15 gm
                                    32

-------
portion of the centrate placed in a beaker.  To this residue was added
40 ml of 10-percent buffered beef extract (Oxoid, Lab Lemco Powder, Flow
Laboratory, Rockville, Maryland) which was mixed for 30 minutes on a
magnetic stirrer, and then Swinny filtered with a 0.45 y membrane.  All
of the filtrate was then inoculated onto BGM cell lines using 1 ml per
bottle (approximately 40 bottles).

Coliphage Media—
     The coliphage media consisted'of three substances: Tryptone broth,
tryptone overlay agar, and tryptone plating agar.  The tryptone broth
consisted of 10 gm/1 of Tryptone(D,ifco 0123), 1.0 g/1 of yeast extract
(Difco 0127), 1.0 g/1 of glucose, 8.0 g/1 Nad, and 0.22 g/1 of CaCl2.
Tryptone overlay agar was the same as tryptone broth with the addition of
7.0 g/1 of agar (Difco 0140).  Tryptone plating agar was the same.as
tryptone broth with the addition of 15 g/1 of agar.  The salt diluent was
8.5 g/1 NaCL and 0.22 g/1 CaClo.  Media and diluent sterilized by autoclav-
ing at 15 psi and 121°C for 15 minutes. Glassware sterilized in a hot air
sterilizer at 170°C for two hours.

Coliphage seed--
     An overnight culture of £. coli K12  (f+) was diluted 1:100 in one
liter of tryptone broth.  The culture was grown on a shaker at 37°C to
an optical density of 0.2-0.3 which was approximately 108cells/ml.  The
culture was infected with f2 coliphage at a multiplicity of infection
(MOI) of 3, and grown for 4-6 hours longer on the shaker.  Twenty to thirty
milliliters of chloroform were added and  it was refrigerated overnight.   The
following day, the:culture was centrifuged at 16,000 G for 20 minutes at
4°C to remove cellular debris.  The supernatant yielded a stock suspension
with a titer of at least 1 x 10'' pfu/ml.

Coliphage Assay—
     Samples of 1-0 ml were collected and  0,5 ml chloroform was added
immediately.  The samples were stored in  a refrigerator overnight.  The
next day, the following procedure was utilized.  To sterile aluminum-capped
tubes  in a 47°C water bath, the following mixture was added: 2.5 ml molten
tryptone. overlay agar, 2.0 ml of E_. coli  K12 indicator cells diluted in
tryptone broth to a concentration of lO^cells/ml, and 0.5 ml of^sample
containing the phage or, if necessary,  0.5 ml of a 10-fold serial
dilution.  Salt diluent was used to make  the sample dilutions.  The tube
contents were mixed on a vortex mixer.  Contents were poured onto a petri
plate  containing 20 ml of solidified tryptone plating agar.  The  plate was
swirled  to evenly distribute the overlay  agar and then allowed to solidify.
 The plates were  -inverted and  incubated  18 hours .at  370C.   Plaques were  then
 counted and  the  titer calculated.
                                     33

-------
                                    SECTION  6

                          OPERATION  OF  THE KELLY-PURDY UNIT
 HYDRAULIC  CHARACTERISTICS

      Like  other radiation  processes, the  kill obtained in the UV irradiation
 chamber was  expected  to  be functionally dependent on the dose of ultra-
 violet light applied  to  the liquid flow.  The theoretical UV dose can be
 computed by  multiplying  the theoretical residence time of the flow in the
 irradiation  chamber by the average (arithmetic mean) UV intensity in the
 water being  irradiated.  Since considerable variation can exist between
 the theoretical and actual residence times in any unit process, the hydraulic
 characteristics of the UV  irradiation chambers used during the study were
 of considerable importance.  The occurrence of extensive short circuiting
 within the chamber would have had the effect of reducing the actual residence
 time  at a  given flow, which would have reduced the actual UV dose and resulted
 in higher  bacterial densities in the radiated water.

      Figure  5 is a graph that presents the theoretical residence time in the
 Kelly-Purdy  irradiation  chamber as a function of both flow and liquid depth.
 During most  of this project the theoretical residence time for the Kelly-
 Purdy unit ranged from 0.4 to 2.0 minutes, while the depth was varied from
 2.54  cm (1 inch) to 6.35 cm (2.5 inches).  The majority of the research
was conducted with the water depth at 2.54 cm and with flows of about 1.94
 I/sec. (30 gpm), which resulted in an approximate exposure of 20 seconds.

      Dye studies were performed at water depths of 2.54 cm (1  inch) and
6.35 cm (2.5 inches) to evaluate the residence time distribution functions
 (RTDF) of  the system.  Slugs of Rhodamine WT dye were injected into the
influent flow to the Kelly-Purdy system, and samples of the effluent were
collected  and  analyzed  for relative dye concentration by using a Turner
Model  111   Fluorometer.  Flows of 0.31, 0.63, and 1.26 liters per second
were used  for the RTDF studies.
                                                                       lar
     Figure 6 presents the RTDFs obtained at a water depth of 2.54 cm and
the corresponding theoretical residence times (theoretical T), and similar
data for a flow depth of 6.35 cm(2.5inch) are shown in Figure 7.   Actual
residence time values can be evaluated by locating the centroid of the RTDF,as
suggested by Merrill  (7),  and that procedure was used in the analysis of
these data.

     The residence times obtained from the RTDF's are significantly less
                                      34

-------
o
•=3-
                 md6
o
oo
o
>

                                                           >
                                                           0) •—
                                                          -a r—
                                                          •i- O)
                                                           
-------
       o  o
       O)  01
       tO  CO
       0)  CU
       r— LO
       00 10

       O O
                                                                                      O

                                                                                      CO
                                                                                                  CD
                                                          LO


                                                          CM
                                                                                      O
                                                                                       •

                                                                                      CM
                                                                                           CO
                                                                                           CU
                                                                                      LO
                                                                                       •  o
                                                                                      r—  LU
                                                                                          OO
                                                                                          CL.
                                                                                     un

                                                                                     CD
O
              o
              o
CD
CO
O
CM
                                                         O

                                                        CD
                                                                                                 LO


                                                                                                 CM
                                                                                                  O.
                                                                                                  CD
                                                                                                  i.
                                                                                                  O)
                                                                                                 •4J
                                                                                                  to


                                                                                                  
                                                                                                 3
                                                                                                 _Q
                                                                                                 •r~
                                                                                                 s-
                                                                                                    O)
                                                                                                    -i->
                                                                                                    1/5
                                                                                                 O)
                                                                                                 O
                                                                     O)
                                                                                                 (/) r—
                                                                                                 CU CU
                                                                                                 CU

                                                                                                 =1
                                                                                                 cn
                           NOIIVcllN33N03  3AIlV13cl
                                                 36

-------
                                                                                                                   M-
                                                                                                                   O
o o o
cu a) cu
tO V) (/5

to 01 to
i- S- S-
O)  -M 4J
•i~™ "i™ •p™

to co i—
CM U3 CO
• • •
r— 0 0

in

"^ T
<0 r- ' /
o oo o
•r- • CD
-(-> O tO
OJ -^
S-- 
to
•^^^
H to
&-.
r— CU
(O 4->
O •!-
•J-J """
oj co
S_ l£>.
O •
OJ 0
/ J=
J H- ca

^ ^


A
r
/
/
/
^^J
fPrl

// i
/. II
VB •«
/I
V / I
^* i\
                                                                                                 ur>
                                                                                                         to
                                                                                                         0)
                                                                                                         Q
                                                                                                         LU

                                                                                                         D_
                                                                                                    Q.
                                                                                                    <1>
                                                                                                   T3

                                                                                                    (C

                                                                                                   4->
                                                                                                    (O

                                                                                                   -o
                                                                                                    QJ
                                                                                                    E
                                                                                                   • r~
                                                                                                    (Q
                                                                                                   •4-5
                                                                                                   XD
                                                                                                    O

                                                                                                    to
                                                                                                    E
                                                                                                    O

                                                                                                   +J E
                                                                                                    O O)
                                                                                                    C 4J
                                                                                                    3 10
                                                                                                   M- >i
                                                                                                                   J3 Q.
                                                                                                                   •i-   I
                                                                                                                   CO • —
                                                                                                                   •r-  
-------
 than the theoretical residence times.  Figure 8 shows the ratio of the
 actual residence time to the theoretical Residence time as a function of
 flow.  It is both interesting and significant to notice that the difference
 between observed and theoretical residence times decreases as the flow
 increases.  The most logical explanation for this is that the
 localized turbulence produced in the influent section of the unit at higher
 flows produces more uniform distribution across the transverse section,
 thus improving the hydraulic performance of the irradiation chamber.

      It is important to notice that the Kelly-Purdy unit performs better
 hydraulically at the shallow depth, where the actual and theoretical resid-
 ence times are in much better agreement than at the 6.35 cm depth.

      Cumulative, dimensionless residence time distribution functions were
 computed for the dye studies as suggested by Thomas and Archibald (8).
 These curves are shown for the Kelly-Purdy unit in Figures 9 and '10.  A tank
 which exhibited ideal plug-flow characteristics would produce a cumulative,
 dimensionless RTDF that was a horizontal line with zero recovery until  the
 actual  and theoretical residence times coincided.   At [t(l/r)]  equal  to
 ';°iJhe curve would be Vert1cal»  and become horizontal  with a  dye recovery
 °1 • °S Percent-  Therefore good  performance is  indicated by a vertical  line
 at 1.0;  the more the curve exhibits an S-shape,  the poorer the  hydraulic
 performance of the system.

      A  comparison of the curves  in  Figures  9  and 10 indicates better  perfor-
 mance at  the shallower depth of  2.54 cm.  The dye  appears  sooner and  persists
 longer  in the unit when the flow is  at a depth of  6.35 cm.            Persists

 INTENSITY AND DOSE DETERMINATIONS
                            *

 .  *  Ultraviolet  light intensity and dose data must be available  if UV dis-
 infection data  are to  be quantified.  The intensity of the UV light at 254
 nm was measured in the Kelly-Purdy unit with  a Model IL500 Research Radiomet-
 er manufactured  by  International Light, Inc.  The photodetector was cali-
 brated by International Light, Inc., and a copy of the calibration certifi-
 cation was provided with the instrument.
                                  W!re Permed at various locations in the
    chnwn      -          Tu    °f the three sePa™te intensity measurements
are shown in Figure 11   The data in Figure 11 indicate the UV intensity at
nan" atelLT ffe I" t"e Kelly-Purdy unit.  The computed mean intensity was
1190 ^watts/cm*.  Variation in these values was due to the geometry of the
unit.  The highest intensities were measured in the middle of the chamber.

of nlK? In?, 3aS*a9* thr°u9h water UV light is attenuated, and the amount
?L iS;+a   *yfn\ P-   1S a*sumed to be some function of the water depth, f(D)
2nJr S  °n/(D) ls.usually taken as an exponential function of both water-
depth, D, and an extinction coefficient, k, and has the form shown below.

                 f(D)  = I  e ~kD
                         o
                                      38

-------
 O
 LjJ
 O
 
-------
   100
    80  -
    60  -
0)
o


-------
-p

O)
o
  .
LU
>•
O
O
LU
                                                0.3 I/sec

                                                0.63 I/sec

                                                1.26 I/sec
20 -
                                                          2.0
  Figure 10.   Dimensionless, cumulative residence time
              distribution function for the Kelly-Pur'dy unit
              (6.35 cm water depth).
                             41

-------
^
<


       o
       ta
       -Q
         o
         O
        CM
        CM

                                       O)


                                                                     c:
                                                                     o

                                                                    t-
                                                                     o

                                                                    q-
                                                                     o

                                                                    LO
                                                                    0
                                                                    CO
                                                                         LO
                                                                         CO
                                                                         o
                                                                         co
                                                                         0
                                                                         OJ
                                                                         o
                                                                         O
                                                                         0
                                                                         co
                                                                         0
                                                                         to
                                                                                    i
                                                                                   >>

                                                                                   r~~
                                                                                   0)


                                                                                   (L>
                                                                                     t/1
                                                                                     c
                                                                                     o
                                                                                   O
                                                                             E    O
                                                                             O   r—

                                                                               •»    t^
                                                                             D;    =s
                                                                             l—i    o
                                                                               LU
                                                                             U.
                                                                             u_
                                                                             uu
                                                                              UJ
                                                                              C_5
                                                                              oo
                                                                         0
                                                                         CM
                                                                                   (O
                                                                                   (O
                                                                                     to
                                                                                      -a
                                                                                    =D CC

                                                                                    -a s-
                                                                                    O) -r-

                                                                                    = >,
                                                                                    CO T3
                                                                                    n3 S-
                                                                                    O) 3
o    o
o    o
o    LO
OJ    r—
O    O
O    O
o    LO
               o    o
               o    o
               O    LO
               CM    i—
                                 o    o
                                 o    o
                                 O    LO
                                         o
o  o     o
o  o     o
O  LO     O
CM  i—     i—
CD
o
LO
                                                                            O
                                                                          o
                                                                                    CD
                   uiu 1792  iv A1ISN31NI 1H9I1
                                     42

-------
     I0 is the surface intensity in watts/cm , k is the extinction coeffici-
ent with units of cnH, and D is the depth below the water surface in cm.

     An average intensity in the water can be computed by integrating  f(D)
between the surface and the depth in question, and dividing by the water
depth.

                       i    -fD
                       I...- -Jo
avg
                                   D
                        /-D
                         •0
                                -kD
            dD
                                             T  D
     The above integration yields the equation shown,below, which can easily
be used to compute average intensity values.
                        avg
                            = I0  (1 -e
                                        -kD>
                                    kD
     Average intensity values were computed for the Kelly-Purdy Unit as a
function of both depth and the extinction coefficient, k.  These values.
are shown in Figure 12 as a family of curves for various values of k.
A wastewater effluent with an extinction coefficient of about 0.5 cm~T
would cause a fifty-percent reduction in the average light intensity in a
water depth of three centimeters.

     A numerical value for the extinction coefficient, k, can be obtained
experimentally by measuring the UV light intensity at different depths.
Typical data are shown in Figure 13 for tap water, an activated sludge
effluent, a filtered activated sludge effluent, and a filtered activated
sludge effluent that had been passed through two activated carbon contactors.
When these data are plotted on semi-log paper, the slope is the extinction
coefficient.

     The change in ultraviolet light transmissibility as a function of water
quality is evident.  The steeper the slope the higher the extinction
coefficient, and thus the greater the attenuation.  The activated sludge
effluent had the highest extinction coefficient, while tap water had the
lowest.

     The direct measurement of an extinction coefficient is not always
possible, and in many installations this could be a time consuming problem.
It should be apparent that a relationship between the extinction coefficient
                                     43

-------
Cvl
 o
 -p
 +J
 
-------
           A  A
TAP WATER
CARBON COLUMN EFFLUENT
FILTERED ACTIVATED SLUDGE EFFLUENT
ACTIVATED SLUDGE EFFLUENT
    C\J
     o
     10
     •p
     CO
     p.
     CO
     •Z.
     UJ
                  123456789
                  DISTANCE ABOVE  BOTTOM  OF K-P  UNIT,  cm
                                               10
Figure  13.   Typical data used to determine  the  extinction  coefficient.
                                 '   45 ,

-------
and the percent light transmittance exists,
relationship:

                       r_T  . -kD
                                             Using a previously defined
but JMs the transmittance, T.  Therefore,
    I(J                     T      -kn
and
                       In (T) = -kD
                           Jo

     Most UV spectrophotometers used for measuring the transmittance of
water samples have a path length of 1 cm.  Therefore, the transmittance
and extinction coefficient are related as shown below.
In
                         T = In (y) = -k
                                  o
                                              D=l
     The above relationship is important in that transmittance data can be
readily obtained on water samples if a spectrophotometer is available that
can be used in the UV range.  Several waters of varying quality were run
through the Kelly-Purdy Unit, and the extinction" coefficient and transmittance
values determined.  These data are shown in Figure 14 in conjunction with
the theoretically predicted curve.  The agreement between the experimental
results and theory is excellent.

     Dose determinations (i.e., the product of light  intensity  and
residence time) can be made relatively easy if the average light intensity
in the water is known, and if the actual residence time in the irradiation
chamber is known.  The previous discussion explained the relationships
between intensity, transmittance, and the extinction coefficient.  Addition-
ally the RTDF's have defined the actual residence times as a function of
flow.  This information can be simplified and expressed graphically as
shown in Figure 15.
                                      46

-------






e
L
h
e
C
c.
h
1






^
0 C
0  c
> a


ul
u_
U.I
LU
CD
0
_J
00 LL.
UL
Q LU
1— LLJ
eC CD
> a
»— i Z3
t— _l
o co
<
C3
O LU
LU |—
o: -=c
LU >
1— I-H
_1 1—
i— i O
U- eC
0 •





 c
-. 
- •=>/
1 7
V






0
CO





./
/
/








CD C
CM r-

1

/•
/





/





D C
0 -C
co c:
o
CO •£>
CM i—
>
i — •
r- a>
O i — to O
• 1 O C
CM E -i- (O
O 4-> -P
CU 4->
CO » S- T-
^/ r-» 1=
i — OJ t/>
•> x: c
1— -P (C
vo z • &.
i— M C
O CO -O
1— I £=
•^ LL. -O (0
,— LU > 4J
O S- C
O (DO)
CM CO -r-
• '*£. -O O
r- O O •!-
i— i M-
01— -H 0
H- c:
CO X OJ E
LU (DO
0 S •!-
-p -p
O) O
co -Q c:
o c: -P
O X
to 
-------
      140
      '. 120
      100
  CO
   o
   X
   O
   O
   (O
       80
       60
   g   40
       20
        0
                        FLOW, gpm
                   10       20        30
                                       I
                       DEPTH = 2.54 cm
          0      0.5    1.0     1.5     2.0    2.5
                       FLOW, I/sec
Figure 15.  UV dose as a function of flow and the extinction
            coefficient for the Kelly-Purdy unit.
                              48

-------
 RUN  NO.  Kl
     Run No.  KT  was a preliminary investigative activity that was conducted
to provide basic data for project planning.   The demonstration plant was
configured as shown in Figure 16 for this study, in which 0.95 liters per
second (15 gpm) of effluent from the No.  1  completely-mixed activated
sludge system was pumped to the Kelly-Purdy UV Disinfection Unit.   The
remaining unit processes shown in Figure 16 were being utilized on other.
research projects.

     On 26 July 1974  the twenty-four-hour composite sample for the activated
sludge effluent had the following characteristics: TSS = 10 mg/1,  COD = 42
mg/1, NH3-N =1.05 mg/1.Alkalinity = 105 mg/1 as CaC03, soluble TOC = 9 mg/1,
and pH = 7.1  units.  These water quality criteria are typical  of a relatively
good quality activated sludge effluent., and microorganism  reductions observ-
ed during this run should have been representative of the unit's disinfection
capabilities.

     During this study the Kelly-Purdy unit was operated with a water depth
of 6.35 cm (2.5 inches), which resulted in a theoretical residence time of
1.8 minutes.   Results of the disinfection study are shown in Table 7 and
Figure 17.

     The apparent increase in TSS is doubtless the result of sampling or
analytical error, and is not supported by the more numerous turbidity
data which indicate no significant change.   A 2.29 log reduction was observed
for the total coliforms, and a 2.25 log reduction was observed for the fecal
coliforms.
                                                                        2
     The mean UV light intensity at the water surface was 1190 u watt/cm ,
as previously described.  During this period a UV spectrophotometer was not
available; however, by assuming an average extinction coefficient of 0.5
the mean UV intensity in the irradiated water can be readily calculated
to be 359 uwatts/cm2. with a theoretical residence time of 1.8 minutes
the resulting theoretical UV dose was 38,800 uwatt-sec/cm  .  The data
presented in Figure 8 indicate that the ratio of actual to theoretical
residence times was 0.77; therefore, the actual residence time and dose
were 83 seconds and 29,900 ywatt-sec/cm2.

     Inadequate disinfection was obtained during this run.  The quality
of the activated sludge effluent being treated was good; yet disinfection
was poor.  Only a 2.25-log reduction in fecal coliforms was observed and
at least 3.5-log reduction is usually necessary to  satisfy the  fecal
coliform standard of 200 organisms per 100 milliliters.

     UV transmittance data were not available at this stage of the project,
so it is possible that the extinction coefficient was much higher than
the estimated 0.5.  The poor hydraulic characteristics of the unit when
operated at the 6.35 cm depth did make a very significant difference in
the UV dose, and this factor certainly affected the observed kills.
                                      49

-------
50

-------
                                       I—          O
                                                                                 8
S3
 Ol   -r-    t. :
 E    T3    m
—   -r-    ~a -
                                              O
                                             o c
                                                -
                                                §    "g
                                             +j  s-     o  s_
                                              o    
-------
    10'
    10'
    10'
o
o
I  io5
CO

CO
CC     A
0   104
    io2-
    10
               Standard Plate Count
           Fecal Coliforms
                                         INFLUENT
                                               4
              • -©"
       Standard Plate Count
Total Coliforms    >
                                        EFFLUENT
                        Fecal  Coliforms
       I
I
I
I
J
      9:00 AM  9:30 AM  10:00 AM  10:30 AM 11:00 AM  11:30 AM

                           TIME OF DAY

     Figure 17.   Results  of Run No.  Kl, August 6, 1974.
                              52

-------
RUN NO.  K2

     During Run No.  K2 the Kelly-Purdy unit was operated on an activated
sludge effluent at a flow of 0.72 I/sec, as shown in Figure 18.  The depth
of effluent in the exposure chamber was changed from 2.54 cm (1 inch)  on
September 28 to 5.08 cm (2 inches) to determine the effect of depth on
disinfection.  Additionally, the mean flow was varied from 0.32 I/sec (5 gpm)
to 1.58 I/sec (25 gpm) to evaluate the effect of exposure time on disinfec-
tion.

     The performance of the No. 1 activated sludge system during Run No. K2
is summarized in Table 8.  Complete nitrification was desired at this time,
but was not consistently achieved, as indicated by the 3.2 mg/1 of ammonia
nitrogen in the effluent.  Effluent quality was very good if evaluated in
terms of secondary treatment limitations.  The computed reductions are based
on the activated sludge influent stream, and not on the data for the raw
wastewater entering the White Rock STP.  The average BODs and TSS of the
raw wastewater were approximately 260 mg/1, which would yield reductions
of 93 percent and 92 percent, respectively.

     On 14 September 1974, the unit was operated at a depth of 2.54 cm (1
inch) and at flow rates of 1.6 I/sec (25 gpm) and 0.6 I/sec (10 gpm).
Results of this operation are presented in Table 9.  Examination of these
data indicates that no significant difference in kill was observed at the
two flow rates studied.  Moreover, no significant difference in mean log
reduction of total coliforms, fecal coliforms, and standard plate count
organisms by ultraviolet light was observed.

     The data for all of Run No  K2 are summarized in Table 10.  These data
are for analyses performed on the grab samples collected for microbiological
examination, and are not the routine process control samples for the Pilot
Plant.  During this run turbidity, total suspended solids, and color were
relatively low, as were observed COD and TOC values.

     The results of this portion of the disinfection study are presented
graphically  in Figure  19,  20,  and 21, which are  time-series plots of three
microbiological water  quality  parameters.  Examination  of the  plots
indicate that  the influent plate count  densities were relatively constant
during this  run.  Furthermore,  no significant  difference was evident in
the  disinfection obtained  at the two depths studied.

      Slightly  less  than  a  three-log  reduction  was observed for both total
and  fecal  coliform  organisms.   This was  inadequate  for  complying with the
usual effluent limitation  of 200 fecal  coliform  organisms  per  100 ml, since
the  geometric  mean  fecal  coliform density for  Run  No. K2 was  290/100 ml.

      Most  microbiological  data obtained during the  study of  disinfection
processes  exhibit  considerable variability,  and  more often  than  not an
                                      53

-------
 I   S-
•r-  C1J
 o
-Q
 S-
103
O E

•* =5
                                                        o
                                                       _
                                                        S-
                                                        n3
                                                       o

-------
TABLE 8.     PERFORMANCE OF NO. 1 ACTIVATED SLUDGE SYSTEM DURING RUN NO.
             K2'

Activated Sludge
Parameter Influent (mg/1)
TSS
BOD5
COD
Sol. TOC
NH3-N
Organic Nitrogen
N02 & N03-N
Total Nitrogen
Total P
Turbidity, NTU
Color, Pt-Co Units
127
126
294
-
16.0
11.0
0.3
27.3
8.2
-
-
Activated Sludge
Effluent (mg/1)
21
18
42
10
3.2
4.5
6.7
14.4
6.1
3.9
39
Reduction
(Percent)
83.5
85.7
85.7
-
80.0
59.1
-
47.3
25.6
-
-
                                      55

-------

m
<^J«
r^»
p_

A
«3*
*"""'
C*
UJ

S-"
**LU
COK
3?* i*v
fV* til
O CO
u.
t-4 IL.
1 r—i
O
f *i >•
O
_J ID
O CO
UJ
ej;
A H"4
CO CO
EUJ
ceo.
O CO
P » 	 „
i— •
— 1 1—

O CD
1-4
_J —J
1— f—
O UJ

CD
U.I-1
O £»
2: 2
CD t—
HH_J
H- =D
O
=D >-
O CO
UJ
Ctft—
z

oo
O
UJ
I— UJ

**1
so.
0
— 1 O
U- C£.
i^r
U. 0
o z

cb co
UJ
u, a
u-z
UJ cC

•
cr>
LJ
f
CO
i^r"
1—








O
CO

«*^
^M

vo
•
o









o

UL.

























O
CO

^^^
|««>

VO
k
*"~












S
c
(
£
n
S
r
a

+J
E «;
CO

t{~
(i«.
UJ









•P
c u
CO
,_
tf«
E
I— i








0)
3
^^
fB









-P

CO
3
r^
t|«
q-
UJ









4j
E U
CO
3

EI_
f~
1— 1



*


[3
j
J

5

3
L.



o o o o
r™ i™- r— n~ CO
x- x x x LO
i— CO CM CO
• • • • ^^
«xj" ^t" CO CO










D VO VO VO
o o o o
XXX X CO
t^. LO CM «* •

LO r«I vo vo





E
ta
CO
s:

ra 4-
CO O

CJ>
O







• «=f «* «*
CD CD O O
r— i— i— i— VO
XXX X LO
vo o -sj- vo
. «3-
CO CO «3- CO










D VO VO VO
O CD O O
r- i— i— •— en
XXX X CO
LO CO CO f"» •
. vo
CO 1^ VD r^.
•




 i— S-
O O CO
t— o a-












co
CM

CM










E
o

''43
CJ
3
•a
CO
o;

en
o

E
rei
CO
s:















CO
CO
•
CM























co co co co
o o o o
XXX X CO
1— 1— «=t- «*
. co
LO CO VO VO







6

vo vo vo
o o o
i— r— i— en
XX X i—
VO CM LO •
I— I— LO VO

CM r-^ r-^





E
ta
CO

E
rei •+-
CO 0
s:
en .
o
•






co co co co
o o o o
r— i— r— r— CM
xxx x r->.
CO LO CO CO
. co
^J- VO «^" LO









vo vo vo vo
O CD CD CD

XXX X i—
vo CM en co •
CO i— CO, CM VO

i— i — i— r—





v> *E
G
s: o
o o
r- q- •—
(O *r~
O i— S-
CO O CO
U_ 0 Q.











CO
co
•
CM











E
o
•r—
4J
o
3
-a
CO


CD
o

E
03
CO
s:














<
CTi
co
•
CM























CM CO CO CO
o o o . o
XXX X i—
vo r^ ^" co •
• • • • oo
en r— i— i—










LO LO LO
O 0 O
*x "x "x 5-
CM i—, VO
• • • LO
CM CO CM





^
ta
CO
s:
E
res 1-
CO 0
s:
CD
o






CO
OO OO CD CO
O O i— O
•— r- X i— i—
XX-* X r—
^d~ ^i* , i~ ~ CO •
• • • • CO
I— i— i— r—










LO LO LO LO
o o o o

XXX' X f-^
r— en vo i—
• • • • • LO
•* <* VO LO





1=
-o
S- 0
rei o
T3 CO 4-> r—
E H-> E
ta 03 3 S~
+J i— O CO
CO Q- O Q.
t










0
oo
•
CM











E
o
•r"

VJ
3
•a
CO


CD
o

E
rei
CO
















o
vo
•
CM






















-------
o
                   £      &
            m      t—      i—     i—
                           in      r—     i—
                   i-      CVJ
                                  o
                                  i—
                                   X
                                  o
                                                            •s
            U5      i—      i—     t—
            O      i—      i—     ^
             •
                                                            !   J
                                                            •s   -s
a

IL.
o
      •—i—      I--      CO
                                                                 f
                                                                 o
            £   "2;
                   .«-   •— E   i— :
                            .

                           0)   0) <
                           Q.   U_
                       57

-------
    10
      8
    10'
                         DEPTH=2.54  cm
    10C
O)

Q.
CO


to

-------
     10
       8i
     10'
     106
o
o
S-
OJ
Q.
oo
o
o
o
     10
     10'
     10'
                  DEPTH=2;54  cm   -*
                                                 DEPTH=5.08 cm
GM
           GM
       I
                                  _t	I
I
        10   '    15       20       25        30  |l       5



                 September 1974                   October  1974



         Figure 20.  Total Coliform data for Run No.  K2.
                                 59

-------
      10
        8
o
o
S-

O)

CL
o
o
O
      10'
icr
      10'
      10'
                    DEPTH=2.54
           GM
           GM
                                          _>. DEPTH=5.08 cm
                                             INFLUENT
         10       15       20        25



                  Septe>ber  1974
                                         30   1     ,5



                                              October 1974
         Figure 21.   Fecal  coliform  data  for Run  No.  K2.
                                60

-------
occasional high value will be observed for no readily explainable reason.
One must assume that the data are representative of the process performance,
and not the result of poor laboratory technique.  The occurrence of sporadic
high numbers in microbiological data is the reason that the geometric mean
is used as a relative indicator of the representative organism density
instead of the arithmetic mean.

     During the analysis of these data several  different statistical  pro-
cedures were employed to evaluate the consistency or reliability of the
ultraviolet light disinfection process.  All of these data were characteris-
tically skewed toward the higher side, and histograms clearly showed
tailing toward higher values.

     One particular statistical procedure that has been used by the authors
and many other researchers to evaluate process reliability is the plotting
of the frequency distribution, or the probability distribution function of
the data.  In this type of data analysis, the individual data are ranked
from highest to lowest, and the probability of each element being equalled
or exceeded by the remaining data is computed.   The value of each element
in the data file is then plotted as the ordinate, and the corresponding
probability is plotted as the abcissa.

     The objective of this analysis is to obtain a straight-line plot,
such that the prediction of events with a very low probability of occurrence
is facilitated.  Several different probability distribution functions can
be used to establish the  scale for the abcissa.  The normal or Gaussian
distribution is most commonly used, and this type of graph paper is
commercially available! however, only those data which are normally
distributed plot as a straight line.

     The log normal distribution will tend to straighten those data skewed
to the higher values, and this function has been used with considerable
success in plotting effluent data from wastewater, treatment plants.

    , The Type I asymptotic extreme-value distribution of largest values, most
commonly called the Gumbel distribution  ( 9,10), has been extensively used
by hydrologists to predict rare flood events.  For an in-depth discussion
of the statistics involved»the reader should consult a good hydrology text,
or a civil engineering statistics text.

     During this project  the normal, log normal, and extreme-value frequency
distributions were all tested, and the Gumbel extreme-value frequency
distribution consistently provided the most satisfactory graphical analysis.
The Gumbel plots are shown for Run K2 in Figures 22 through 24.

     The one percent probability level indicates the value which will be
equalled or exceeded approximately four times per year, while the 0.1 percent
probability value will be equalled or exceeded once every 2.7 years,
in theory.  Additionally, Figure 24 indicates that the fecal coliform density
of 200 per 100 mis will., be equalled or exceeded  in sixty  (60) percent of
the samples collected.
                                      61

-------
     10y
     10"
S-
0)
O.
to

10
     10'
     10
     10J
      10
       0
                                    EXTRAPOLATED
                     EFFLUENT
         99  90  80  60 40   20   10   532   1   0.5  0.2 0.1

                  PROBABILITY OF BEING EXCEEDED
    Figure  22.   Extreme-value frequency distribution for standard
                plate count data, Run No. K2.
                                62

-------
    10y
     10
     10-
o
o
S-

O)

Q.
£
f-f
o
o
     10
     10
     10J
     10
       0
                                 EXTRAPOLATED
              INFLUENT
                         EFFLUENT
                                            t
                                                    t
        99 90 80 60 40   20   10   532    1    0.5   0.2 0.1


                 PROBABILITY OF BEING EXCEEDED
    Figure 23.  Extreme-value frequency distribution for total

                coliform data,  Run  No.  K2.
                              63

-------
    10'
    10"
    ixr
    10
o
o

-------
RUN NO. K3

     The Pilot Plant was configured as shown in Figure 25 for Run No. K3.
The objective of this run was to obtain additional data on UV disinfection
of a nitrified activated sludge effluent with the Kelly-Purdy Unit.  The
unit was operated at a depth of 2.54 cm (1 inch) and an average flow of
1.13 I/sec (17.9 gpm).

     During this run the No. 1 activated sludge system performed reasonably
well.  An average of only 1.1 mg/1 of ammonia nitrogen was present in the
effluent, as can be seen in Table 11.  The observed BODc, COD and TSS are
typical for a municipal wastewater treatment facility; therefore, the
observed disinfection should have been indicative of anticipated performance
at other facilities.

     Table 12 presents a summary of the water quality data which were
obtained from analyses on the grab samples collected for microbiological
examination.  Turbidity values and observed total suspended solids concen-
trations were somewhat higher than in Run No. K2; however, color and soluble
TOC concentrations were comparable.

     Table 12 also summarizes the results of disinfection for Run No. K3.
Figures 26, 27, and 28 present data for standard plate counts, total
coliforms, and fecal coliforms, respectively.  Although relatively good
bacterial reductions were achieved during this run  (3.1 logs for fecal
coliforms), the effluent would not have met a 200 fecal coliforms per 100
ml standard.

     The intensity of the UV light at the water surface was spot checked
periodically during all of the Kelly-Purdy runs.  These checks indicated
that no change in UV intensity occurred; therefore, the previously reported
mean UV intensity (at the water surface) of the 1190 uwatts/cm2 has been
used for all dose calculations.  Additionally, during this portion of the
project activated sludge effluent transmittance data at a wavelength of 254
nm were not available.  An extinction coefficient, k,of 0.5 cm"' was
assumed for the purposes of calculating the average UV intensity in the
water flowing through the Kelly-Purdy unit.

     During this run the average flow of 1.13 liters per second resulted
in an average theoretical residence time of 36 seconds. With the depth of
flow set at 2.54 cm, and an assumed extinction coefficient of 0.5 cm~', the
mean UV intensity in the irradiated activated sludge effluent was 674 ywatts/
cm2.  The computed UV dose was 24,500wwatt-sec/cm2.

     The observed reductions in bacterial populations were not sufficient
to satisify a typical disinfection limitation of 200 fecal coliforms per.
100 ml.  The fecal coliform data shown in Figure 28 indicate that relatively
consistent reductions were obtained during this run; however, the extreme-
value frequency distribution (EVFD) in Figure 31 indicates that 38 percent
of the samples exceeded 400 fecal coliforms per 100 ml. .The EVFD for the
total coliform data indicates that 24 percent of the samples exceeded 2000
                                     65

-------
                           •
                            (1)
                             I  S-
                           •r- OJ
       O
       O
       s-
       (C
       o
       •3-

       o
                            E

                           r—
                            O
                           O
                                                                            O
                                                                            .£3

                                                                            (B
                                                                            O
O
                         O)
                         cu
                         Ol
                        •a
                        oo
                        -a
                         ai
                        •P
                         O
                                                 O
                                                    O)
CO
 IT

o-
                                                            a:
                                                            Z3
                                                            Q-
                  o
                  UJ
                  U_
                                                                                            O

                                                                                            •P
                                                                                            o
                                                                                            o
                                            1/5
                                            0)
                                            o
                                            s
                                            CL
                                            c
                                            (O
                                           r—
                                            Q.


                                            O


                                           Cu
                                                                                            O)

                                                                                            3
                                                                                            CD
                                                                                            •n~
                                                                                            u_
               II

              o-
                                       66

-------
TABLE 11.   SUMMARY OF NO. 1  ACTIVATED SLUDGE SYSTEM
           PERFORMANCE DURING RUN NO. K3

Activated Sludge
Parameter Influent
mg/1
TSS
BOD,
0
COD
TOC, soluble
NH3-N
Org. N
N02 & N03-N
Total N
Total P
Turbidity,NTU
Color, Pt-Co units
126
161
328
-
16.6
8.7
0.4
25.7
10.8
-
-
Activated Sludge
Effluent
mg/1
27
12
43
11.8
1.1
3.3
7.8
12.2
7.7
8.4
44
Reduction
percent
78.6
92.5
86.9
-
93.4
62.1
-
52.5
28.7
-
-
                           67

-------
                   OCVI

                   CTS
                   O  O)
                                       I     I      I
                                CO    ID    CM
                                                                   CM      i—
                                                  r-      00       r—
                                                                                       •S
                                                          O

                                                          "x
                                                                                       O

                                                                                       "o
                                                          CVJ      r—
                                                 4_    -ry

                                                 •a    i—
                          i—    CM    CM
                                                          f—      CNJ
                                                                                       o.    o
                                                                                       OJ   ,—
                                                                                       s-    o
i
                           01
                          4J
                          s
OO    r—
t/>    O
h~    O
                                                                         O S-
                                                                         0>
                                                 i
                                                 o
                                                 s
                                                 68

-------
10
  8
10'
10C
10-
10
10'
     - 6M
       _L
1
                             INFLUENT
                                 EFFLUENT
                                                 25
       5        10        15       20
                     October 1974
Figure 26.  Standard plate count data for Run No.  K3.
30
                            69

-------
o
o
co

cc
o
o
CJ
     10
       8
     10'
     10"
    10
    10
      2  _
    10'
             GM
              15FT
                         i
i
i
        1      5        10       15         20        25


                             October 1974


        Figure 27.  Total coliform data for Run  No.  K3.
                            30
                                70

-------
     10C
     10'
o
o
 
-------
    10'
     10
O)
Q.
az
o
     10'
     10'
               V
                              EXTRAPOLATED
                        INFLUENT
                       EFFLUENT
        99  90 80 60  40   20  10   5  3      1   0.5  0.2  0.1


                     PROBABILITY OF BEING EXCEEDED


        Figure 29.   Extreme-va'lue frequency distribution for standard

                    plate count data, Run No.  K3.
                                  72

-------
                                EXTRAPOLATED
o
o
s-
•
                              73

-------
                             EXTRAPOLATED
10'
   99  90 80 60 40   20   10  5  3      1   0.5  0.2 0.1  ,

                 PROBABILITY OF BEING EXCEEDED

   Figure 31.  Extreme-value frequency distribution for fecal
               coliform data, Run. No. K3.
                            74

-------
coliforms per 100 ml.

     Although 3.09 and 3.55 log reductions were observed for fecal  coliforms
and total coliforms, respectively, the effluent counts were higher than
desired. The dose of 24,300 ywatt-sec/cm2 was insufficient to provide
adequate disinfection.

RUN NO. K4

     Inasmuch as the previous UV disinfection runs had indicated that the
Kelly-Purdy unit was not capable of adequately disinfecting a nitrified
activated sludge effluent to the 200 fecal coliform per 100 ml effluent
limitation, a filtered nitrified activated sludge effluent was studied.
The Pilot Plant was configured such that a portion of the effluent from the
No. 1 activated sludge system was pumped to the No. 2 multimedia filter
(sand and anthracite).  The flow then proceeded by gravity to the Kelly-
Purdy unit, which was operated at a water depth of 2.54 cm.  The process
configuration is shown in Figure 32.                   .

     During the time Run No. K4 was being conducted, complete nitrification
was achieved as indicated in Table 13, where the average ammonia nitrogen
concentration was 0.2 mg/1.  Overall wastewater quality during this period
was excellent, as indicated by the low levels of TSS and COD in the filter-
ed effluent.

     The water quality data obtained from the grab samples collected for
microbiological analyses are shown in Table 14, and are indicative of a
relatively high-quality secondary effluent.

     The average flow rate of the filtered, activated sludge effluent to the
Kelly-Purdy unit was 0.96 liters per second.  At the operating depth of
2.54 cm, the 0.96 liters per second flow resulted in a theoretical  residence
time of 43 seconds.  In the absence of transmissability data, an assumed
extinction coefficient of 0.4 cm'1 was used to compute an average UV
intensity in the water of 747 ywatts/cm2.  The lower value for k was select-
ed due to improved effluent quality, when compared to the unfiltered activat-
ed sludge effluent.  This combination of residence time (43 seconds)
and UV intensity results in a theoretical UV dose of 32,000 ywatt-sec/cm2.

     The observed log reductions reported in Table 14 were only slightly
higher than the reductions observed in previous runs, and the influent
counts of all three microbiological water quality indicators were about
one log lower than in Runs K3 and Kl.  Both of these factors partially
explain the reason the fecal coliform data complied with the 200 F.C./100 ml
standard.

     Time-series plots of the microbiological data are shown in Figures
33 through 35.  All three graphs indicate that the respective influent pop-
ulation densities were reasonably constant, but the effluent counts are
significantly lower during the middle of November.  The examination of all
available water quality data failed to identify any reason for the mid-month
                                     75

-------
r
                                                                                                                                         c

                                                                                                                                        3.
                                                                                                                                         CJ)
                                                                                                                                         fO
                                                                                                                                         cr>
                                                                                                                                         O
                                                                                                                                         O
                                                                                                                                         O)
                                                                                                                                         O
                                                                                                                                         O

                                                                                                                                         a.
                                                                                                                                         fO


                                                                                                                                        Q_



                                                                                                                                         O


                                                                                                                                        •r~
                                                                                                                                        Q.
                                                                                                                                        CM
                                                                                                                                        CO
                                                                                                                                         (1)
                                                                              76

-------
TABLE 13.    PERFORMANCE OF NO.  1  ACTIVATED SLUDGE SYSTEM
            AND NO. 2 MULTIMEDIA FILTER DURING RUN  NO.  K4

A.
Parameter
TSS
COD
TOC, soluble
NH3-N
Org. N
N02 & N03-N
Total N
Total P
Turbidity, NTU
Color, Pt-Co units
S. Inf.
mg/1
146
287
-
11.6
9.1
0.3
21.0
6.4
-
_
A. S. Eff.
mg/1
15
31 ^
9.5
0.2
2.4
8.5
11.1
4.2
2.6
25
No. 2 Filter
Effluent
mg/1
5
20
9.7
0.2
1.8
8.7
10.7
4.3
2.5
31
Reduction
percent
96.6
93.0
-
98.3
80.2
-
49.0
32.8
3.8
_
                          77

-------
               -3-6
              jot tit
               01 to
               o >


1=   §
                      CO


                      o
                                                           Cd     f—
                                                                            m
                                                                                -
                                             OJ     r-
                                                    CO     r—
§
a:
                                                  *O r—   O
                                                  •—  E   1-
                                                  CL.      't- r—

                                                     S-   i— £
                                                  •O  (U   O
                                                  s-  a.   c_> CD

                                                  *        o



                                                  1§   SI
                                                  4->  O   O 0}
                                           78

-------
10
10
                   10
              15
20
25
   Figure 33.
       November 1974
Standard plate count data  for  Run  No.  K4.
                            79

-------
S
    10'
                        I
I
I
        1       5         10        15        20        25
                            November 1974
       Figure  34.  Total coliform data for Run No. K4.
                               80

-------
    10'
o
o
s-


CL

 «%

co


a;
o
o
o
3
LU
U_
    10
    10'
    10'
                         INFLUENT
                       EFFLUENT
                                             GM
                                  I
                        10        15       20


                            November 1974
25
       Figure  35.   Fecal  coliform data for Run No. K4.
                                81

-------
 decrease.

       The EVFD's  for  the microbiological water quality data are presented
 in  Figures  36  through 38.   These  curves  indicate that a fecal coliform
 density  greater than  200 per  100  ml should be expected in 2.5 percent of1 the
 samples, and that 0.6 percent of  the samples exceeded 400 F.C./100 ml.  The
 EVFD's indicate that  even with a  high quality effluent, which can be
 disinfected to meet the effluent  limitation of 200 F.C./100 ml (based on
 a geometric mean), a  significant  probability exists that the treatment plant
 will  fail to comply with the 400  F.C./100 ml. effluent limitation.

 RUN NO.  K5

      The Pilot Plant  was configured as shown in Figure 39 during Run No. K5.
 The Kelly-Purdy UV Disinfection Unit was operated at a water depth of 2.54
 cm  (1  inch) and a mean flow of 1.4 I/sec (22.2 gpm). The effluent  from the
 No.  1 activated sludge system was used  in  this run.

      The performance  of the activated sludge process during Run K5 is
 summarized  in  Table 15.  The  biological process suffered a substantial  upset
 in early December when a plant operator inadvertantly left the waste sludge
 valve open  over-night.  The result was a reduction in MLSS from about 3500
 mg/1  to 250 mg/1, and almost a complete breakdown in the capability of the
 activated sludge process to treat wastewater.  During this time effluent
 quality deteriorated  significantly and the remainder of the month was spent
 recovering  from the process upset.

      Data obtained from the analyses of grab samples collected for microbio-
 logical examination are presented in Table 16.   Color and turbidity were
 substantially  higher  than in previous runs, as were  TSS, COD,  and  TOC.

      Figures 40, 41, and 42 present time-series plots of the standard plate
 count, total coliform, and  fecal  coliform data, respectively.   During
 Run No.  K5 the effluent counts for the microbiological  parameters were
 higher than the values observed during the previous runs.   The coliform
 log reductions obtained during Run No.  K5 were not substantially different
 than  those obtained during  earlier runs.

     The theoretical  residence time during this run was 29 seconds.   If an
 extinction coefficient of 0.55 cm~' is assumed, due to the poor effluent
quality,  the average UV intensity in the water can be calculated to be
641  ywatt/cm .   This combination of residence time and intensity resulted in
a theoretical  UV dose of only I8,800ywatt-sec/cm2.

     Examination of the time-series plots of the microbiological data
indicated an apparent  decrease in  population densities for both the activated
sludge effluent and the disinfected effluent as the biological  process
recovered.   The fecal  coliform data presented in  Figure 42 indicate an  almost
constant  three-log reduction  regardless  of the influent population density.
                                     82

-------
                             EXTRAPOLATED
     99 90 80 60 40   20   10   532   1   0.5  0.2 0.1
              PROBABILITY OF BEING EXCEEDED

Figure 36.  Extreme-value frequency distribution for standard
            pi ate, count data, Run No.  K4.
                           83

-------
                                  EXTRAPOLATED
    10*
•5  104
o
o
 s_
 OJ
 Q.
io
I"
S.
a
O
g  IO2
    10J
    10
      0
                             EFFLUENT
                                                    i

       99 90 80  60  40    20   10   5 ,  3 2   1   0.5  0.2 0.1

                 PROBABILITY  OF BEING  EXCEEDED
    Figure 37.  Extreme-value  frequency  distribution ,for total
                coliform data,  Run  No. K4.
                              84

-------
     10'
     10
     10"
o
o
QJ
Q.
DC

O
tt-
O
     10
     10
     10J
     10
       0
                                EXTRAPOLATED
                                  INFLUENT
                             EFFLUENT
        99 90 80 60 40    20    10    5   3  2    1    0.5  0.2 0.1


                 PROBABILITY  OF BEING  EXCEEDED
    Figure  38.   Extreme-value frequency distribution for fecal

                 coliform data, Run No. K4.
                             85

-------
                                        to
                                        •I—
                                        "O

                                        Is-
                                    I— T- CU
                                     O -S T-
                                                             «3- O  S
                                                                .0  3
                                                              • S_ r—
                                                             O fO  O
    c. c
CO  O E
   _Q 3
  •  i. r—
 O  rtS O
2: O C_3
POOO
                                                                                                                LO
                                                                                                                 i-

                                                                                                                •a

                                                                                                                 c'
                                                                                                                 o
                                                                                                                •I—
                                                                                                                •4-5
                                                                                                                 03
                                                                                                                 E
                                                                                                                 O
                                                                                                                 O
                                                                                                                 03
                                                                                                                r—
                                                                                                                Q_


                                                                                                                +->

                                                                                                                 O

                                                                                                                •r—
                                                                                                                D.
                                                                                                                CT>
                                                                                                                CO
                                                                                                                 
-------
TABLE 15.   PERFORMANCE OF THE NO.  1  ACTIVATED SLUDGE
            SYSTEM DURING RUN NO.  K5

Activated Sludge
Parameter Influent
mg/1
TSS
COD
TOC, soluble
NH3-N
Org. N
N02 & N03-N
Total N
Total P
Turbidity, NTU
Color, Pt-Co units
98
217
-
13.9
7.6
0.3
21.8
7.5
-
_
Activated Sludge
Ef.fl uent
mg/1
39
68
14
10.1
3.6
2.0
15.7
3.8
10.2
>70
Reduction
percent
60.2
68.7
-
27.3
52.6
-
28.0
49.3
.
_
                         87

-------








































in
B
g
g
£

£
|j
IX


10
LU














+J
1
C
» ^ r— ^- H-
^ ^ E m m r--
§S * » - 3
k c/) ca cj *—
r- 3 C/) CD O O
U- h- I-. 0 H- V)




1



1


o

A

O
f-.
A




in
in

cn
CM


i


o
A




g
A


O
in


cn
CM

42
I
O
O

a.
0
8


oo
oo


00

*
"L

X
cn


CO
o
5

to
CO
o
*x


to
cn

to
***
to
to
CD
X
CO
CM


r-.
o


in
r—
X
CO



o



O)
rd r—
CL, E

Standard
Count pe


I--.

CM

in

*
*o
x
^
in

in
o
5
in

CO
o
X
CO

to
-

CM
cn
to
to
o

CO




X
o
in

to
°
• r^


CO
r-~


1
O
^•t—

w— r*-
43,.
•sg.


O
CD
CO

o
CM
CO
CO
o

H

r-*
-t
S
CO

CO
o
*x
0


-,

o
CJ
«5
«o

-^




I

in
0
S



CM
r—


E
o
^i—

si
fig.





























1

S
C7>
O

O
5
2
O
1
^
^
TS
S
R
2
2
(O

. CO
i
[ *-*
S-

'





































>•
§
43
n
^
u
'g>
o
!o
2
u

,0


-------
     10C
    10'
    10'
 S-

 O)
 Q.
oo
s:
00
cc
o
    10s-
    10'
          GM
GM
                         I
                         I
               10       15       20

                   December 1974
                                                     25
30
        Figure 40.   Standard plate count data for  Run  No.  K5
                                89

-------
     10
       8
     10'
     10C
o
o
GO



o
0
     io
    TO
    icr
    10
      i
-  GM
 6M
       I
        1
                  I
I
                                            I
I
                                           25
        5        10        15       20


                     December 1974



Figure 41.  Total coliform data for Run No. l<5.
                                                              30
                               90

-------
    10L
    10'
    10
o
o
s.
CD
Q-
o;
o
c
CJ
   10
   •io3
   102
   .10'
          6M
         6M
     EFFLUENT
I
I
I
I
         I       5        10        15        20        25



                               December  1974



         Figure 42.  Fecal coliform data  for Run No.  K5'.
                                91

-------
     The EVFD's presented in Figures 43 through 45 indicate that process
performance was much less consistent during Run No.  K5  than during Run
No.K4.    During Run No. K5, 80 percent of the UV disinfected samples  exceed-
ed 200 F.C./100 ml, and 72 percent exceeded 400 F.C./100 ml.
                                      92

-------
    10y
    10
    10
o  T n3
a:  10
o
    10
    10J
    10
                              EXTRAPOLATED
           /
              /
                  X  EFFLUENT
                   INFLUENT
                                     X
                                         I
0

 99 90 80 60 40   20   10   532   1   0.5  0.2 0.1

          PROBABILITY OF BEING EXCEEDED
Figure 43.  Extreme-value frequency distribution for standard

            plate count data, Run No. K5,
                             93

-------
    10
      8
    10'
    10
o
o
S-

0)
O-
    10'
     10
                               EXTRAPOLATED
        99  90 80 60 40   20   10   532   1   0.5  0.2 0.1


                 PROBABILITY OF BEING EXCEEDED
    Figure 44.  Extreme-value frequency distribution for total

                coliform data. Run No. K5.
                              94

-------
o
o
S-
0)
Q.
CO
1
£
o
o
5
LU
U-
    10y
    10
    io
    io
                               EXTRAPOLATED
                  INFLUENT

                        EFFLUENT

                                                   X'
       99 90 80 60 40   20    10    532   1   0.5  0.2 0.1

                PROBABILITY  OF BEING  EXCEEDED




  Figure 45.   Extreme-value  frequency distribution for fecal

              coliform data, Run No. K5.
                            95

-------
                                 SECTION 7

                         OPERATION OF THE UPS UNIT
HYDRAULIC CHARACTERISTICS        ,

     The model EP-50 UV disinfection system, manufactured by Ultraviolet
Purification Systems, Inc., arrived in late 1974 and was brought on-line
in January 1975.  When the plumbing for the UPS unit was completed, the
entire UV light system schematic flow diagram was as shown in Figure
46.  The final study system was capable of testing any of twelve different
wastewaters in either or both of the UV disinfection units.

     The conceptual design of the UPS unit differed significantly from the
Kelly-Purdy unit, in that the UV lamps were surrounded by the process flow
instead of being suspended above a relatively .thin layer of the wastewater
effluent.

     Theoretical residence time for the UPS unit is shown as a function of
flow rate in Figure 47.  Several dye tests were performed with the UPS
unit so that the RTDF's for the unit could be evaluated.  The resulting
RTDF's are shown in Figures 48 through 50.  Figure 48 presents the two
RTDF's obtained at a flow rate of 0.63 liters/sec. (10 gpm).  The
theoretical residence time at this flow was 1.49 minutes, and the average
residence time obtained from the centroids of the RTDF's was 1.83 minutes.
These data indicate that the actual residence time was about 24 seconds
longer than the theoretical residence time.  The pronounced tailing in the
RTDF's that was observed during these dye studies shifted the centroids
to the right.  The net effect was to significantly increase the apparent
residence time.

     The RTDF obtained at 1.89 liters/sec. (30 gpm) is shown in Figure 49.
The theoretical residence time is 29.8 second and an actual residence time
of 22 seconds was observed for this dye study.  For this flow the actual
residence time was eight seconds (27 percent) shorter than the theoretical
residence time.

     The two RTDF's obtained at a flow of 3.47 liter/sec.  (55 gpm) are
shown in Figure 50.  The actual residence time determined from the centroids
of the dye studies was 13.1 seconds, which is 3.2 seconds  (20 percent)
shorter than the theoretical residence time of 16.3 seconds.

     Cummulative, dimensionless RTDF's for the UPS unit are shown in Figure
51 .   The lower flows exhibited more tailing, which is particularly
                                     96

-------
     O)
               CU




S—
CU
.{_>
^— I
•r—
U_

«—
.
O
2:




S-
CD
^j
r—
•I—
LJ_

CNJ
,
O
2:
!_
O
O

. s=
0
_Q
S-
03
CJ

OO
.
O
~z.
,«
0


c
o
r^
S-
03
C_3

"*
.
O
~ZL



I ^
C
CU
3

^_
q-
LU

oo
•
•sC



j ^
E
CU

p_
tj^n
q-
LU

0
*
CO





- • •
^_>
c
(0

CU

a:
•
3


CU
CO
03
S
^
S-
ra
£••
•r"
 ^j
oo oo



CU
3
(«•
'1
f-
LU

r—
03

>r~
U_



            a.
            rtS
            a)
           i—   CU   S-
            Q.   >   O)
 res
oo
                   .
                 res
                      a;
                     s:
                            I/I

                           4->
                                      q-
                                       o
                                       (O
                                       s-
                                       cn
                                       03
                                       O
                                      r—
                                      4-

                                       O
                                      •r-
                                      -U
                                       03

                                       CU
                                      .c
                                       o
                                       a>
97

-------
o
O)
in
o

u_
      3.5
      3.0
      2.5
      2.0
      1.5
     1.0
     0.5
                                                                50
                                                                40
                                  30   w
                                       n


                                      o







                                  20
                                                                10
                  20
40
60
80
100
                           RESIDENCE TIME,  sec
       Figure 47.  Theoretical  residence  time of  UPS  UV  disinfection

                   unit  as a  function of  liquid flow  rate.
                                  98

-------
                                                    to
                                                    
-------
                                                         o
                                                         f-.
                                                         o
                                                         10
                                                         o
                                                         LO
                                                              O
                                                              O)
                                                         o
                                                         CO
                                                         o
                                                         CM
                                                                   o
                                                                   CD
                                                                   CO
                                                                  -p
                                                                  CT>
                                                                  CO
                                                                  fO
                                                                  =3



                                                                  O-
                                                                  S-
                                                                  o
                                                                  o;
                                                                  cu
O
to
o
CO
o
CM
         NOIl\/cUN30NOO  3AQ 3AI1V13H
                              TOO

-------
                                                                                o
                                                                                CO
                                                                                CM
                                                                               CM
                                                                               CO
                                                                               CM
                                                                               cn
LO
ro
o
oo
LO
CM
o
CM
                                            LO
                                              LO
                                                                                      o
                                                                                      CD
                                                                                      QJ
                                                                                   O)
                                                                                   to
                                                                                   UJ  CO
                                                                            D-
                                                                            ZD


                                                                            O)
                                                                                      O
                                                                                     <*-

                                                                                      
-------
                                                              i.
                                                              en
                                                              LO
                                                              CO
                                                              LLJ
                                                              Q
                                                              LU
                                                              O
                                                              Q
                                                                       CO
                                                         O

                                                         co
                                                                       ur>
                                                                        •

                                                                       CM
                                                                            OO
                                                                            UJ
                                                                            O
                                                                                 co
                                                                                 Q.
                                                                            •=c
                                                                            O   S-
                                                                            i—i   o
                                                                            I—   4-
                                                                            LU
                                                                       _   a;   «/>
                                                                       0   o
                                                                        •   LU   U.
                                                                       CM   -i-   <—\
                                                                            u,
                                                                            CD
                                                                       LO
                                                                   o:

                                                                   to

                                                                   cu
                                                              O    i—
                                                                                 to

                                                                                 OJ
                                                                                 T3

                                                                                 QJ
                                                                       LT)


                                                                       O
                                                                                 O
                                                                                 CD


                                                                                 3
                                                                                 CD
O
o
o
CO
o
CM
                        Q3cl3A033cl  3AQ  JO  39VlN30H3d
                                       102

-------
evident  in  the  0.63  liter/sec.  RTDF's.

     The hydraulic performance  of  the UPS unit was generally good.  The dye
studies  indicated that at  a  flow of  3.47 liter/sec  an organism could
possibly exit the UV exposure chamber after only a 3-second exposure. This
is  indicative of short circuiting.   Tailing was observed at all flows.  In
theory plug flow was not approached  very closely; however, in practice the
unit performed  about as well as can  be expected from a very short residence
.time unit.                   •      .
 DETERMINATION  OF  DOSE                                ,

      Most  UV dose data  reported  in  the  literature are in units of ywatt-sec
 per  cnr, and considering  the experimental procedures employed in early re-
 search  efforts  the dose units,are reasonable.  The majority of the commercial
 UV irradiation  chambers presently manufactured are totally enclosed.  The
 absence of a free water surface  that  is relatively uniformly irradiated
 makes the  determination of  the average  UV intensity in the water almost
 impossible.  In an enclosed chamber with more than one UV lamp, the UV
 intensity  within  the water  being irradiated will vary from point to point.
 This  variation  could not  be quantified with the  instrumentation available
 during  the time period  in which  the project was  conducted.

      With  a totally enclosed irradiation chamber it is very easy to measure
 the  energy applied per  unit volume  of water per  unit time, but it is not
 possible to measure the average  UV  energy reaching a unit volume of water
 per  unit time.  During  this project UV  dose within the UPS unit was not
 directly measured.  In  an effort to overcome this deficiency, several
 different  approaches were taken  in  estimating the UV dose under differing
 experimental conditions.

      An actinometry procedure described by Parker (11), and by Calvert and
 Pitts (12) was  investigated to check  the output  of the UV lamps, and pro-
 vided excellent results.  In this procedure the  iron (III) in potassium ferr-
 ioxalate is reduced to  iron (II), while the oxalate ion is oxidized.  The
 resulting  color change  is measured  by a conventional spectrophotometer.
 Data  from  actinometric  experiments  performed in  distilled water indicated
 that 117.6 joules per second of  UV  energy were being produced by the nine
 40-watt lamps.  These data  indicated  that approximately one-third of the
 energy  supplied to the  lamps was emitted in the  UV region.  This value
 agrees  reasonably well  with the  37  percent efficiency claimed by the manu-
 facturer.

      The only  dose related  parameter  that could  be measured easily and
 on a  routine basis was  the  UV light,intensity at' a quartz window on the side
 of the  UPS unit.   The window was surrounded by a 2.54 cm (1 inch) diameter
 316  stainless  steel pipe  that was 6 cm  (2.4 inches) long.  Intensity measure-
 ments were made with the  previously described IL 500 radiometer.  Since the
 sensor  used for the IL  500  was larger than the pipe projecting
                                     103

-------
r
             from  the quartz window, some of the radiation reaching the window was block-
             ed  from reaching  the  IL 500 sensor.  As a result of this geometry, the UV
             intensity readings taken with the  IL 500 were low.

                  An experiment was conducted to determine the effect of the system
             geometry on  the intensity values measured.  Data indicated that the
             measured values should be multiplied by a factor of 8.9 to obtain correct
             intensity values.

                  In order  to  compute dose in units of energy per unit area it was
             necessary to know the average UV intensity within the exposure chamber.
             With  the technology then available it was not possible to directly determine
             or  measure the average Intensity (or dose) in a totally enclosed exposure
             chamber.  The  procedure described  below was used to compute a mean UV
             intensity within  the  UPS unit.

                      1. A relative intensity  of 1.0 at the surface of the quartz
                         sleeves, and no attenuation of the UV light were assumed.
                         The  UV intensity at various distances from each lamp was
                         then computed.  The computation assumed the lamp functioned
                         as a line source, and the result was a set of concentric lines
                         of equal intensity around each lamp; however, the intensity
                         patterns represented  in this manner indicate the effect of
                         only one lamp.

                      2. Several  lamps  influenced the actual intensity at any point
                         within the contactor. To consider this effect the intensity
                         values for each lamp, at any given point, were summed.  The
                         resultant number was  the intensity at a given point relative
                         to the intensity at the surface of the quartz sleeves.  There-
                         fore a relative intensity value of 1.3 would indicate that
                         the  calculated intensity at  that point was 130 percent of the
                         UV intensity at the surface of the quartz sleeves.  The
                         effects  of shadowing  were considered in the computation of
                         relative intensity values.

                      3. Relative intensity values were computed for several hundred
                         discrete points on the transverse cross section of the
                         exposure chamber.  Points of equal intensity were connected
                         such that isointensity lines were developed as shown in Figure
                         52.

                      4. The  average, relative UV intensity within the exposure
                         chamber  was then easily computed by planimetering the areas
                         between  adjacent isointensity lines, and multiplying by the
                         average  UV  intensity  between the isointensity lines.  This
                         resulted in an area-intensity product.

                         The  area-intensity products were computed for the total cross
                         section, and summed.  The sum of the area-intensity products
                         divided  by  the total  area yielded the mean relative intensity

                                                 104

-------
Figure 52.    Isointensity patterns for UPS exposure chamber.
                                  105

-------
              within the exposure chamber.

              The mean relative intensity computed in this manner
              was 1.1, and the intensity measured at the surface
              of a quartz sleeve (lamp and  sleeve removed from unit)
              was 4.6xlO~^watts per cm2.  The resultant average intensity
              within the exposure chamber was 5.06xlO~3watts per
              cm2, but this value assumed that there was no attenuation
              occurring.
     In order to compute the actual average UV intensity within the exposure
chamber attenuation was considered by utilizing the following equation.
                       I =
                               -kd
                       where I  = 5060 uwatts/cm
                                                            -1
                             k  = extinction coefficient (crrf ),

                             d  = mean light path length (cm)

     The extinction coefficients for various runs were determined from
the transmittance values measured at 254 nm, and the mean path length of
the UV light within the exposure chamber was estimated to be 4.0 cm.   Dose
can be simply computed by multiplying the average intensity value obtained
by the procedure just described by the theoretical residence time.

                       example: Given - flow = 2.0 liters/sec (32 gpm)
                                        transmittance @ 254 mm = 62 percent

           1. Calculate the average intensity, I:

                       I = I e~kd

                         = 5060   e -°'5 (4'°>

                         = 5060 (0.135)
                                        2
                         = 683 ywatts/cm

           2. Calculate theoretical residence time, T:

                           v
                       T = q
                         =   53.6 liters
                            2.0 liters/sec
                         =  26.8 sec
                                     106

-------
           3. Calculate dose, D-J-:
                       DT =  I x T

                             683 ywatts

                                cm
                          =  18,300 ywatt-sec/crri
x 26.8 sec.

      2
   ,  When the UV dose was calculated in the above manner the effect of
slime accumulation on the quartz sleeves surrounding the UV lamps was not
considered; furthermore, the theoretical residence time was used in lieu
of the actual residence time.  The only two variables affecting the UV dose
which enter the above calculations are the flow rate and the extinction
coefficient.  Figure 53 presents the calculated dose, Dj, as a function of
both of these variables.

     The dose variation as a function of flow rate, within  the practical
operating range of the UPS unit (1.5 to 3.2 liters per second), is insignifi-
cant relative to the change in dose as a function of the extinction coeffi-
cient-  The water and its associated impurities appears  to  be the most
important factor affecting UV dose in an operating system.
                                      107

-------
O
O)
trt
 I

LlJ
CO
O
UJ
O

S
     10C
10'
               10
                  20
                    1.0
FLOW, gpm •

30      40
50     60
                                      k = 0.0  (100% T)
                                      k = 0.5  (62% T)
                                      k =  1.0  (37% T)
                           2.0
              3.0
          4.0
        Figure  53.
                          FLOW,  I/sec

               The effect of  flow  and  extinction  coefficient
               on the  calculated ,UV dose.
                               108

-------
RUN NO. Ul                             •

     The first run with the Ultraviolet Purification Systems,,Inc. UV dis-
infection unit was made between 23 January and 5 February  1975.  The
Demonstration Plant was configured as shown in Figure 54 for this study,
with effluent from the No. 1 activated sludge system being pumped directly
to .the UPS unit at 1.8 liters/sec (29 gpm).

     The nitrified activated sludge effluent during this period was of
excellent quality, as the data presented in Table 17 indicate.  Effluent TSS
and COD concentrations averaged 17 and 33 mg/1, respectively.  Nitrification
was almost complete with only 2.2 mg/1 of ammonia nitrogen in the effluent,
and the reduction in total nitrogen averaged approximately 45 percent.

     The data obtained from the grab samples collected for microbiological
analysis are presented in Table 18.   The percent transmittance values were
determined at 254 nm on a Beckman DU spectrophotometer using a quartz
cuvette with a 10-mm light path.  The effluent geometric mean total and
fecal coliform densities were less than 4.7 per 100 ml and 2.3 per 100 ml,
respectively, indicating that excellent disinfection was achieved. Greater
than four log reductions in coliform numbers were achieved, while only a 2.26
log reduction in standard plate count occurred in Run Ul.   Figures 55, 56,
and 57 are time series plots of the microbiological data.

     During this run the average flow rate of 1.8 liters/sec resulted in a
theoretical residence time of 29.8 sec.    The mean transmittance of 70.0
percent corresponded to an extinction coefficient of 0.35 cm"'.   The
theoretical calculated dose, based on these operational  parameters, was
37,200 ywatt-sec/cm .  This was one of the highest doses utilized during
the project and the UPS unit provided excellent disinfection under these
conditions.

     Extreme value frequency distributions for the microbiological data
are presented in Figure 58, 59, and 60.  Examination of the fecal coliform
data in Figure 60 indicates a probability of 0.1 for finding a fecal
coliform count exceeding 200 per 100 ml under the conditions of this run.
Expressed in another way, the curve indicates that a fecal coliform count
greater than 200 per ,100 ml could be expected once in 1000 days, or once
in about 2.7 years assuming the same wastewater quality and UV dosage
conditions.
                                    109

-------
    10
        •r- O)
        3 «r-
         CO
        a 
O


«r~
Q.




•


LO
                    a>
                    s-
                    3
                    CD
         no

-------
TABLE 17.   PERFORMANCE OF THE NO. 1  ACTIVATED SLUDGE
            SYSTEM DURING RUN NO. Ul

Parameter Influent
mg/1
TSS
COD
TOC, soluble
NH3-N
Org.-N
NO &N03-N
Total N
Total P
Turbidity ,
NTU
Color,
Pt-Co units
75
203
37.6
14.4
7.1
0.6
22.0
6.6
3.4
32.5
Effluent
mg/1
17
33
9
2.2
2.4
7.6
12.1
5.3
3.5
24
Removal
percent
77
84
76
85
66
-
44.5
20
-
.
                          Ill

-------
                                                                                     *J-      *—
                                              I      I     I
00
C\J

CO
                  *
                           i    to    to      co   to    in
                                                                             f—             o
                OCM

                mra
                o     SI
                           E   £
                           u
CO


Ul
                          r-    t£>   ID       CO    IO   to
                                     ^-            cn    cn   4J o       s_  o.   o o   O o
                                      E     *       E    E   c c E    m          o      o
                                            S_ tf)                      CCtO       (0,
                                     co   r- v-    o   a   s. 4-> ^-    «j  3    -MS-    (J J-
                                     co    o c    o   o   
-------
        10'
     lilO4
     s_
     cu
     Q.
     oo
     00    -

     ^  103
     o;
     o
        10'
                     INFLUENT
EFFLUENT

  OFF
                I
                  I
                25       30  |       5

                January        February


Figure 55. Standard plate  count  data  for Run No. Ul
                        113

-------
     10'
    10*
 o
 o
 0.
 o
 o
 o
    10
    10'
    10
      0
               INFLUENT
EFFLUENT
                       I
                        I
             25       30  I       5


           January          February



Figure 56.  Total coliform data for Run No. Ul
                    114

-------
       10'
       icr
  o
  o
   CL
  00

  oc

  £
  o
  o
  
-------
     10y
     10"
  s-
  O)
  CL
  CO
      10
      10
      10
        0
                                EXTRAPOLATED
                         INFLUENT
                             EFFLUENT
         99  90  80  60  40   20   10   532   1   0.5  0.2 0.1

                  PROBABILITY OF BEING EXCEEDED
Figure 53.  Extreme-value frequency distribution of standard plate
            count data for Run No. Ul.
                               116

-------
     107
                                 EXTRAPOLATED
     10*
o
o
£-
O)
O-
  «\
oo
10
O
     10'

          99  90 80 60 40   20   10  5  3  2

                  PROBABILITY OF BEING EXCEEDED
                                             0.5  0.2 0
  Figure  59.   Extreme-value frequency distribution of total
               coliform data for Run Ul.
                             117

-------
                                 EXTRAPOLATED
    10'
    icr
CD

O
 cu
 O.
  n

CO
s
    10'
                              EFFLUENT
        99 90 80 60 40 '  20   10   532   1   0.5    0.2  0.1



                    PROBABILITY OF BEING EXCEEDED



        Figure  60.   Extreme-value frequency distribution of

                      fecal col i form data for Run 111.
                             118

-------
 RUN NO. U2

     Run No. U2 was conducted during 7-17 February 1975.  As shown in Figure
 61, the effluent from the No. 1 activated sludge system was pumped to the
 upflow clarifier (Infilco Densator) at a flow of 6.30 liters/sec (100 gpm).
 The upflow clarifier was operated as a tertiary settling basin only, and no
 coagulants or coagulant aids were utilized during the run.  The process
 configuration was selected so that the effect of additional treatment,
 obtained by the tertiary settling of the activated sludge effluent, could
 be evaluated.

     The data presented in Table 19 summarize the performance of the
Demonstration Plant during Run U2.  During this time the activated sludge
 process was producing a good quality effluent.  Effluent COD, soluble TOC,
 and ammonia-N concentrations averaged 35, 8, and 1.6 mg/1, respectively.
 The tertiary settling did very little to enhance effluent quality.  COD
 was reduced from 35 to 24 mg/1, color was reduced by 10 Pt-Co units, and
 turbidity was unaffected.

     During the run the UPS unit was operated at an average flow.of 1.9
 liters/sec (30 gpm), which resulted in a theoretical residence time of 28.2
 seconds.  The transmittance measured at 254 nm was 71.8 percent, which
 corresponds to an extinction coefficient of 0.34 cm"1.  The theoretical,
 calculated dose, DT, for this run was 36,600 uwatts-sec/cm2.  The mean inten-
 sity at the quartz window was 1200 uwatts/cm2, after the reading had been
 corrected for geometry.  This value (1200 ywatts/cm2) multiplied by the
 actual residence time of 25 seconds yields a dose related value of 30,500
 uwatt-sec/cm2.  Since this value is a measured intensity value multiplied by
 the actual detention time it is very closely related to the actual UV dose.
 This term will be referred to as the indicated dose, Dj;.

     The data obtained from the grab samples collected for microbiological
 analyses are presented in Table 20.  Excellent disinfection occurred.  The
 geometric mean fecal and total coliform densities were less than 3.1 and
 7.5 per TOO ml, respectively.  Total and fecal coliform log reductions
 exceeded 4.0, while the mean standard plate count reduction was less than
 2.0.

     The time series plots of the microbiological data presented in Figures
 62, 63, and 64, are indicative of consistently good disinfection.  Extreme-
 value frequency distributions for standard plate counts, total coliforms,
 and fecal coliforms are presented in Figures 65, 66, and 67, respectively.
RUN NO. U3

     Run No. U3 was conducted from 19 February 1975 to 3 March 1975.
The Demonstration Plant was set-up as shown in Figure 68 for this run.
treatment was identical to Run No. U2, except the tertiary settled,
activated sludge effluent was filtered prior to UV disinfection.
The
                                      119

-------
         o
                  ,1  S-
                  *r™ QJ
                  4-1 4J


                  S Lu

                  i— 
O    n-
                         120

-------
       TABLE 19.    PERFORMANCE OF ACTIVATED SLUDGE AND TERTIARY
                   SETTLING  BASIN DURING RUN NO.U2
Parameter A
TSS
COD
TOC, soluble
NH3-N
• Org-N
N02 & N03-N
Total N
Total P
Turbidity,NTU
. S. Inf.
mg/1
97.4
199
31
12.2
7.3
0.1
19.6
7.3
_
A. S. Eff.
mg/1
21.2
35.0
8.1
1.6
2.5
6.6
10.7
4.5
4.1
Tertiary
Settled
Effluent
mg/1
40.3
24.4
7.9
1.8
2.0
6.7
10.5
2.5
4.2
Removal
percent
59.0
88.0
75.0
85.0
73.0
-
46.4
65.8
_
Color, Pt-Co
units
19
16
                               121

-------
                          CO      r—
                                                                       to      't—
s
c
                                                              in      r—      i—
                                                              ^-      o
                                                               p—      CJ
                                                                                CO       i—
                                                                                        g
                                                                                        o
                                                                                        'i
                                                                                             RO1
                                                                                             o
£
                                                               •—      CO
                                                       r-      r-      CM
      O 4J     *
CO   F- -t-   O
CO    O C   O
I-   0=3   h-
                                             s
s- •(-> «d-    ro 3
 S-
O 0)
h- Q.
                                            122

-------
       10'
 •s     10
 i-
 OJ
 Q.
 00
 sn
 oo
 CIS
 D;
 o
       10
       10
       10
                           INFLUENT
          15        1.0       15


                      February


Figure 62.  Standard  plate count data for Run No.  U2.
                         123

-------
  10'
  10C
  10V
  10
  10'
       GM
      _ GM
                  INFLUENT
                  EFFLUENT
             I	I	I
     1        5       10        15
                February
Figure 63.  Total coliform data for Run No.  U2.
                      124

-------
     10y
o
o
&   10
Q.
Ci
o
S
«=c
O
UJ
    •10
    10
    10
      0
         GM
                          INFLUENT
                         EFFLUENT
        1      5         10        15


                  February


     Figure 64.  Fecal coliform  data  for  Run  No.  U2.
                        125

-------
                                 EXTRAPOLATED
         99 90 80 60 40   20   10   5,32   1   0.5  0.2 0.1

                  PROBABILITY OF BEING EXCEEDED
Figure 65.  Extreme-value frequency distribution of standard  plate
            count data for Run No.  U2.
                             126

-------
      10y
                                EXTRAPOLATED
      10
  g   10
  01
  Q.
 IS)


 o
 P   io2
      10J
                          INFLUENT
      10'
         99 90 80 60 40   20   10   532    1   0.5   0.2  0.1

                  PROBABILITY OF BEING EXCEEDED


Figure 66.  Extreme-value frequency distribution of total  coliform
            data for Run No. U2.
                                127

-------
o
o
S-
Ol
OL
£
o
o
_
LU
U_
     10'
                                EXTRAPOLATED
     10
10
10
     10J
     10
       0
                             INFLUENT
        99 90 80 60 40   20   10   532    1    0.5   0.2  0.1

                .PROBABILITY OF BEING  EXCEEDED
Figure  67.   Extreme-value frequency distribution of fecal coliform
             data  for Run No.  U2.
                                128

-------
                       • -a
                      o  
-------
     Water quality data for the various untt processes are summarized in
Table 21.  The No. 1 activated sludge system continued to achieve good
nitrification and produce a high quality effluent,as the effluent TSS, COD,
and ammonia-N concentrations of 28, 42, and 1.5 rng/1, respectively, indicate.
As in the previous run the tertiary settling process did little to improve
effluent quality; however, the filtration process did significantly enhance
effluent quality.  The filter lowered the mean TSS concentration from 32 mg/1
to 4 mg/1, an 88-percent reduction, and effected a 46'percent reduction in
effluent COD, from 42 to 23 mg/1.

     During Run No. U3 the flow.to the UPS unit averaged 1.6 liters/sec
(25 gpm), which was approximately the same as in the two previous runs.
This flow rate resulted in ,a theoretical residence time of 33.5 seconds,
and an actual residence time of 29.5 seconds.  The mean measured transmit-
tance value of 68.9 percent was equivalent to an extinction coefficient of
0.38 cm"'.  The calculated mean dose, Dj, during the run was 37,100 ywatt-
sec/cms and the indicated dose, Dj, was 57,800 uwatt-sec/cnr.   The
average reading on the IL 500 .radiometer was 220 ywatts/cnr, which corres-
ponds to a mean intensity at the exposure chamber wall of 1960 ywatts/cm^
when the original reading is corrected for system geometry.

     Table 22 summarizes the data obtained from the microbiological
samples.  The geometric means for both the total and fecal coliforms were
less than 2 per 100 ml.  Greater than a five-log reduction was observed for
total coliforms, and fecal coliforms were reduced by 4.22 logs.  The
standard plate count reduction was 2.32 logs.

     The time series plots of the microbiological data presented in Figures
69, 70, and 71, graphically depict the excellent disinfection obtained
during this run.  Figures 72, 73, and 74, are the extreme-value
frequency distributions.

     The best disinfection observed during the project occurred during Run
No. U3.  Two factors contributed to the excellent results obtained; first,
the quality of the wastewater effluent being disinfected was very good.
However it should be noted that the lower mean TSS and COD concentrations
obtained by filtration had no effect on the transmissability of UV light,
since the precent transmittance at 254 nm observed during runs Ul, U2, and
U3 were practically identical.  Second, the calculated doses, Dj, were
highest during runs Ul., U2, and U3 and the highest radiometer readings were
obtained during Run U3.

RUN NO. U4

     Run No. U4 was conducted 6-7 March 1975 to evaluate the effect of
chemical coagulation/flocculation on UV disinfection.  The Pilot Plant was
configured as shown in Figure 75   The activated sludge effluent was pumped
to the upflow clarifier, where ferric chloride was added as a coagulant.  The
chemically treated water then flowed by gravity to the UPS UV Unit.
                                      130

-------
TABLE 21.   PERFORMANCE OF THE ACTIVATED SLUDGE SYSTEM AND
           MULTIMEDIA FILTER DURING RUN NO.  U3

A.S.
Parameter Influent
mg/1
TSS
COD
TOC, soluble
NH3-N
Org.-N
N02 & N03-N
Total N
Total P
Turbidity
NTU
Color
Pt-Co units
95
247
41.4
14.0
8.0
0.4
22.4
9.2
-

A.S.
Effluent
mg/1
28
41.6
9.3
1.5
3.2
6.2
10.9
7.2
-

Tertiary
Settled
Effluent
mg/1
32
42.1
9.4
1.5
2.4
6.2
10.1
7,8
2.0
17.5
No-.l Filter
Ef f 1 uent
mg/1
4
22.8
8.2
1.1
1.6
5.1
7.8
7.5
1.9
17.9
Removal
percent
96
91
80
92
80
-
65
18
_

                         131

-------
                o
              g-s
              ££
              OCM


              0,11
              00
               £
        I
        c
             Cv-
             cs: to
             sa
£3
DC
                      1   r-  CJ
                                    oo   o  at
                                                      I     r—     CM     CM
                     in   o

                     r-   CM
                                                           r—     CM
                                                     i—    O

                     CO   r-   CO
                                                                        O



                                                                        O
                                                            ^- r-   i— r

    5i   Ife    si
CM  CO O   I— O.  U_ I
                                                                                 •8
                             o

                            [o
                            IS

                             2
                                                                                 •S   °
                                                                                 t   S
                             u

                            5
                                                                                 §    a
                                                                                 s   &
                                            132

-------
          10'
          10
          10
       s-
       O)
       Q.
       CO
          10-
           10'
                 INFLUENT
                      EFFLUENT
                                 GM
                      I      I	1
                      25    28  I     4
                 February       March


Figure 69.  Standard plate count data for Run No.  U3.
                        133

-------
   o
   o
   O)
   o.
   CO
   o:
   o
  o
  o
       10'
       10'
       10
         0
             INFLUENT
            GM
             I
                  EFFLUENT
J
I
             20       2528l     4



            February           March


Figure 70.  Total coliform data for Run  No.  U3.
                     134

-------
        10'
    o
    o
    o    4
    -
    s-
    
-------
     10y
     icr
     10*
     io
 oo
 i
 o
     10
      10J
      10
        0
                               EXTRAPOLATED
                             INFLUENT
                                   EFFLUENT
         99  90  80  60 40   20   10   532   1   0.5  0.2 0.1

                  PROBABILITY OF BEING EXCEEDED
Ftgure 72.  Extreme-value frequency distribution of standard plate
            count data for Run No. U3.
                               136

-------
    10y
    icr
    10"
    10
 s-
 
-------
        10y
        10"
        10"
    o    4

    °  10


     CD
     CL
    £
    O


    H  io2
       10J
        10
         0
                                  EXTRAPOLATED
                  /
                                 INFLUENT
                                 EFFLUENT
          99  90  80  60 40   20   10   532   1   0.5  0.2 0.1

                    PROBABILITY OF BEING EXCEEDED
Figure 74,  Extreme-value frequency distribution of fecal coliform

            data for Run No. U3.
                                138

-------
        o
                  s: u-

                  i— 03
                                                                LO
                                                                o

                                                                to
                                             o
                                             UJ
                                             u.
                                          to 2:
                                          O_ 1—I
                                          :o oo
O)
C/J
I.
C71
                                         OJ
                                        CVJ
                                               t
                                               en
                                              CO
                                                                £=
                                                           s_
                                                           o
                                                           M-


                                                           O
                                                           •r-
                                                           4-5
                                                           03

                                                          • 3
                                                           cn
                                                                o
                                                                o

                                                                -M

                                                                fO

                                                                Q.


                                                                O
                                                                r—•
                                                                •r—
                                                                Q_
                                                                O)
                                                                CD
                         139

-------
     During this run the UPS unit was operated at a mean flow rate of 1.95
liters/sec (31 gpm) which resulted in a theoretical residence time of 27.5
seconds and an actual residence time of 24.5 seconds.   The mean transmittance
measured at 254 nm was 66.2 percent, which corresponds to an extinction
coefficient of 0.42 cm~'.  The calculated dose, Dy, was 25,900 uwatt-sec/cm2.
The average of the intensities measured with the radiometer was 110 ywatt/
cm2, which is equal to 980 ywatts/cm2 at the quartz window in the exposure
chamber wall after correction for geometry.  The indicated dose,  DI, was
24,000 ywatt-sec/cm2.

     Since this run lasted only two days the data shown m Table  23 are very
limited, but it is clear that the activated sludge process was achieving
excellent nitrification and producing a good effluent.  Table 24  summarizes
the water quality data from the microbiological samples.

     The observed log reductions were not as high as in previous  runs;
however, the bacterial densities in the chemically treated effluent entering
the UV unit were lower than in previous configurations.

     The extreme-value frequency distributions are shown in Figures 76, 77,,
and 78.
RUN NO. U5

     Run No. U5 was one of the longest runs of the project, and during this
effort the activated sludge effluent was pumped directly to the UPS UV unit
as shown in Figure 79.  This run was initiated on 8 March 1975, and terminat-
ed on 14 July 1975.  The average flow rate through the UPS unit was 2.5 I/sec
(39 gpm).

     Water quality data for the No. 1 activated sludge system are presented
in Table 25.  It is apparent that effluent quality was lower than in previous
periods.  Effluent TSS averaged 29 mg/1, and the effluent COD increased to
an average of 54 mg/1.  Nitrification was only achieved intermittently,
resulting in an average effluent ammonia nitrogen concentration of 7.1 mg/1.

     The major factor responsible for the decrease in effluent quality was
mechanical failure of the mixing and oxygen transfer equipment in the No.  1
aeration basin, which occurred in early April.  From that time until mid-  .
July temporary air diffusers were used, and the diffusers were, at best, of
marginal utility.  The result was the operation of the system with very
limited oxygen transfer capability.

     During Run U5 the average flow of 2.5 liters/sec resulted in a theoret-
ical residence time of 21.4 seconds and an actual residence time of 19.5
seconds.  The mean transmittance value measured at 254 nm was 68.4 percent,
which corresponds to an extinction coefficient, k, of 0.39 cm.  The
average radiometer reading obtained during this run was 71 ywatts/cm2, which
results in an indicated dose, DT, of 12,300 ywatt-sec/cm^ after correction
for system geometry and multiplication by the actual residence time.

                                     140

-------
         TABLE 23.  ,  PERFORMANCE OF THE NO.  1  ACTIVATED SLUDGE SYSTEM
                   .  AND UPFLOW CLARIFIER DURING RUN NO.  U4
Parameter
A. S. Inf.
  mg/1
A. S. Eff.
  mg/1
Upflow
Clarifier
Effluent
  mg/1
Removal
percent
   TSS           96
   TOC,(soluble) 56
   NH.-N         19
     O
   Org-N         10.3
                  14
                  14
                  0.2
                  2.3
                   22
                   12
                   0.8
                   1.7
                77
                79
                96
                83
                                  141

-------
CCM
01 to
s
                                                          s
                                                     r—    r-    O
                                                           i—    CO
                                                           •-^    **•
                                                     O
                                                     "x
                                                     r—     CM
                                                          O
                                                           O
                                               CM     r—     •—   - t—
                                                                        I
                                                                         -
                                                                         a
                                               00
                                               en
                                          o
                                          "x
                                                   £1   £^
                  r—   T-   D)

                      3   E
                  5   5
                                            SO) T- I
                                           Q. E <
                                       142

-------
     10y
     10"
     itr
 g.   10

  «\
GO

CO
o;    , r,
o    10
     10
     10J
     10
       0
                                 EXTRAPOLATED
                           r  TM
                              INFLUENT
                                EFFLUENT
        99 90 80 60 40   20   10   532   1   0.5   0.2  0.1


                 PROBABILITY OF BEING EXCEEDED
  Figure 76.   Extreme-value frequency distribution of standard plate

               count data for Run No. U4.
                             143

-------
s_
(1)
a.
8
     ioy
     10"
     10*
     10
     10
     10J
X
                              EXTRAPOLATED

                                   INFLUENT
                                   9
                            EFFLUENT
                                            /
                                              f-

        99 90 80 60 40    20    10    532   1   0.5  0.2 0.1


                 PROBABILITY  OF BEING EXCEEDED



   Figure  77.  Extreme-value  frequency distribution of total coliform

                data for  Run No. U4.
                              144

-------
    10'
    10
0   in4
o   lu



s-


Q.
2 .10'
o
o
«=c

a    2
^  10
    10J
    10
      0
                               EXTRAPOLATED
                       EFFLUENT
                                    INFLUENT
       99 90 80 60 40   20   10   532    1   0.5   0.2  0.1


                PROBABILITY OF BEING EXCEEDED
 Figure  78.   Extreme-value  frequency distribution of fecal  coliform

              data  for  Run No.  U4
                           145

-------

-------
TABLE 25.   PERFORMANCE OF THE NO. 1  ACTIVATED SLUDGE
            SYSTEM DURING RUN NO. U5

Parameter
TSS
BOD5
COD
TOC.soluble
NH3-N
Org-N
N02 & N03-N
Total N
Total P
Turbidity, MTU
Influent
mg/1
84
91
229
42.5
16.6
8.4
0.26
25.3
7.5
4.8
Effluent
mg/1
29
24
54
13.0
7.1
4.0
2.4
13.5
5.8
4.8
Removal
percent
65
74
77
69
57
52

47
23
0
                        147

-------
     The data from the microbiological grab samples are summarized in
Table 26.  The total and fecal coliform reductions were not as high as in
previous runs.

     The time series plot of the standard plate counts, Figure 8d, indicates
a sudden increase in influent and effluent counts during early April.  Total
coliform densities also increased sharply in April (Figure 81), followed by
gradual increases in late June and early July.  The same trends were
noted in the fecal coliform data (Figure 82).  Included in Figure 82 is a
time-series plot of NHs-N concentrations.  It is evident that changes in
Nhh-N concentrations were paralleled by corresponding changes in fecal
coliform numbers during this period.

     Figures 83 and 84 are plots of the UV intensity measured by the IL 500
radiometer (at the quartz window) and observed log reductions in total
coliforms and fecal coliforms, respectively.  Good correlations exist, which
one would expect since the measured intensity and the actual  dose are relat-
ed.  The relationship between the intensity measured with the IL 500 and the
indicated dose, DT, has been previously discussed.  Several factors can be
responsible for the variations noted in the measured UV intensity, but
changes in transmissability at 254 nm and slime accumulation on the quartz
sleeves surrounding the UV lamps are probably the most significant variables.
Both of these factors are discussed in Section 9 of this report.

     The estimating equations and correlation coefficients for total and
fecal coliform densities and measured UV intensity are presented on Figures
83 and 84,  respectively.

     Figures 85, 86, and 87 are the extreme value frequency distributions
for Run No. U5.  The fecal coliform data indicate that 280 organisms per 100
ml could be expected 50 percent of the time, worse than in any previous run.
Additionally, the log reductions of organisms observed during this run were
not as good as in previous runs.

     The decrease in disinfection efficiency observed during this run can
be attributed to a number of different factors.  The operational difficulties
with the activated sludge system resulted in the production of an effluent
of lower quality that was somewhat more difficult td disinfect.  The
effluent transmissability at 254 nm was lower than in all previous runs
and the influent organism densities were higher than in all previous runs,
as were the mean UV influent COD and ammonia-N concentrations.

     Furthermore, the flow to the UPS unit was intentionally increased to
stress the disinfection capability of the system.  The result was a calcul-
ated dose, D-r, that was approximately 40 percent lower than in all previous
runs.  In addition the measured mean UV intensity (uncorrected) of 71 ywatt/
cm^ was substantially lower than the 137, 220, and 110 ywatts/cm^ measured
during runs U2, U3, and U4, respectively.
                                      148

-------
g-i
.


                         :=    i
                          <&
                               (O
                          S-    -t-J
                          o    (d
                         M-    XJ

                         •o    ^~
                          O)    (O

                         t    £
                          o    o>
          f—   i—   CM
                                                          S.   r- E   ^- :
                           i- at

                           O 4->    •*
                          t—-i-   CJ

                           O C   <=>
                                  149

-------
       I	I
                                             QJ
                                             E
                                             •"3
                                             s_
                                             Q.
                                             O
                                             (O
                                                               If)
                                                                a;
                                                                S-
                                                                o
                                                                M-
                                                                tO
                                                                -(->
                                                                (0
                                                    3
                                                    O
                                                    O
                                                    OJ
                                                    (O
                                                    r—
                                                    Q.
                                                    5-
                                                    fO
                                                    T3
                                                    (O
                                                    OO
                                                   o
                                                   CO
                                                                OJ
                                                                CD
oo   r-.
  o    o
10   un
 o    o
*    OO   OJ   r—
 O   O    O    O
           Liu aad 'SWSINVDHO
                               150

-------
                                  o
                                 H-

                                  fl3
                                  O
                                  O
                                  CO

                                  13
                                  3
                                  cr>
Liu 00 L
'SWHOJIIOO 1V101
        151

-------
        L/6iu  {N-
                 CM
                   O
                        t—     O
                          O    O
                                       3
                                       •"3
                                       O)
                                                    LO
                                       s-
                                       Q.
                                       O
                                       s-
                                                      s-
                                                      o
                                                     q-

                                                      rO
                                                     •P
                                                      fO
                                                     T3
                                                    S-
                                                    o
                                                    O
                                                    O
                                                    (C
                                                    O
                                                    
-------
         10°
                         Log Y = 0.45  X+ 0.29
                                 r = 0.74
            0     12     3     4     5     6     7

                    , LOG,REDUCTION TOTAL COLIFORMS
Figure 83.  UV intensity versus log reduction of total  coliforms
            for Run No. U5.                            •
                              153

-------
O

in


(O

3.

 n
>-

>—I
to
     10
o;
=3
00
eC
LU
                         Log Y = 0.43  X+  0.36

                                r =0.73
         01       23456       7

                    LOG REDUCTION FECAL COLIFORM
Figure 84.   UV  intensity versus log reduction of  fecal
             coliforms for Run No. U5.
                         154

-------
                           EXTRAPOLATED
     9990806040   20   10   532   1   0.5  0.20.1

              PROBABILITY OF BEING EXCEEDED

Figure 85.   Extreme-value frequency distribution, of standard
            plate count for Run No. U5.
                           155

-------
      10y
       10"
       10"
   8   10
   $-
   O)
   CL
   o
   o
       102
       10J
       10
         0
                                 EXTRAPOLATED
X
                 /
      /
           INFLUENT
          1
                      EFFLUENT
          99  90 80 60 40   20   10   532   1   0.5  0.2 0.1

                   PROBABILITY OF BEING EXCEEDED
Figure 86.  Extreme-value frequency distribution of total coliform
            data for Run No. U5.
                               156

-------
   o
   o
   O)
   Q.
   o;
   o
   IX.
   o
   o
   UJ
   u..
        10'
        10
        10-
        10
        10
        10J
        10
          0
                                    EXTRAPOLATED
                          7
                         INFLUENT
                                 EFFLUENT
           99 90 80 60 40   20   10   532   1   0.5   0.2  0.1


                    PROBABILITY OF BEING EXCEEDED
Figure 87.  Extreme-value frequency  distribution of fecal coliform

            data for Run No.  U5.
                                157

-------
RUN NO. U6

     Run No. U6 was conducted on filtered activated sludge effluent from
15 July to 3 August 1975, as indicated in Figure 88.  The water quality data
for this run are summarized in Table 27.  It is clear that effluent quality
was poor, due to the absence of adequate oxygen transfer capacity in the No.
1 activated sludge system.  This deterioration in effluent quality caused
a marked reduction in UV disinfection efficiency.

     The gross organic water quality parameters presented in Tables 27 and
28 indicate clearly the poor quality of the effluent that was being produced
by the activated sludge process.  Filtered COD and BODc concentrations
averaged 69 and 18 mg/1, respectively, and under normal conditions a COD and
BOD of 25 and 5 would have been anticipated.  Effluent color averaged
>48 Pt-Co units,as compared to previous values of 25.  The high TSS
concentrations in the filtered effluent are the result of insufficient
backwash water flow, and thus, incorrectly backwashed filters.

     During this run the filtered activated sludge effluent was pumped to
the UPS UV unit at an average flow rate of 2.2 liters/sec  (35 gpm), which
resulted in a theoretical residence time of 24.4 seconds and an actual
residence time of 22.0 seconds.  The effect of reduced effluent quality
on UV  transmisslbility was qtearly evident.   During this run the average
extinction coefficient, k, was 0.61 cm~', which corresponds to 54.1 percent
transmittance.  In all previous runs the transmittance ranged from 66 to 72
percent.  The calculated dose, Dj,, was the lowest observed during the pro-
ject, and averaged 10,800 uwatt-sec/cm2.
                                                                      2
     The mean intensity value measured with the IL50Q was 39 uwatts/cm .
The average for the four previous runs was 135 ywatts/cm2, 246 percent
higher than the value measured for this run.  The indicated dose, Dj, for
Run U6 was only 7640 uwatt-sec/cnv1-.

     Microbiological data are summarized in Table 28.  During this run the
geometric mean fecal coliform density was greater than 2000 per 100 ml,
which was two to three logs higher than previous runs when there was
adequate oxygen transfer capacity.,

     The mean log reduction in fecal and total coliform organisms averaged
2.69 and 2.74, respectively, while the mean standard plate count reduction
was 2.42 logs.

     Figures 89, go, and 91 are time series plots of the microbiological
data.  The large decrease in effluent counts on July 23 coincided with the
cleaning of the unit on that date.  The very significant variation in
effluent bacterial densities are important  if one must achieve consistently
good disinfection.

     Figures 92 and 93 are plots of the UV  intensity measured by the  IL 500
radiometer  (at the quartz window) and observed log  reduction  in total
coliforms and fecal coliforms, respectively.  Again, good correlations
                                     158

-------
                                 I  S-
                                i— ta
                                 • -o
                                O O)
                                                           I—
                                                         o
                                                        CL. UJ
                                                                 co



                                                                 3
                                                                 05

                                                                 >

                                                                 3
                                                         Q.
                                                         CO
CO
               O
               CD
O

00
                    Q.
                    CD
                   VO
                   CM
                                                                                3

                                                                               CXL
                                                                                s-
                                                                                o
                                                                                (O

                                                                                3
                                                                                O
                                                                                u
                                                                 
-------
TABLE 27.   PERFORMANCE OF THE ACTIVATED SLUDGE SYSTEM AND
            MULTIMEDIA FILTERS DURING RUN NO.  U6

A.S.
Parameter Influent ,
mg/1
TSS
BOD,-
b
COD
TOC, soluble
NH3-N
Org-N
N02 & N03-N
Total N
Total P
Turbidity, NTU
Colon Pt-Co units
78
119
239
52.8
15.7
8.1
0.2
24.0
9.3
-
-
A.S.
Ef f 1 uent
mg/1
58
31
107
21.2
12.8
6.7
0.3
19.8
6.7
4.7
46
Filter No. 1
Ef f 1 uent
30
18
69
19.6
12.6
5.2
0.2
18.0
7.3
4.6
48
Removal
percent
62
93
71
63
20
36

25
22
-
—

                           160

-------
  cn o
  .25
     o
  5-8
                                                        cn
                                                        to

                                                        CM
                            o
                            r^
                             A
                                                        ^-      CM
                00

                O
                                                       CM      f—     t—
                                                        i—      CM
                cn   • to     r*-
                                 to
                                 CM
                                                                                  -s
V
 OCM
                                                                        8
          O
          CM

          CM
                     i—      **•
          10

          CM
to   co
r—   CO
CM
1^.
          s
          i—   r—   CM
                                                                   i   |
                                                                   o
                                                                  5   1
                                                                   O    i.
                            -
                    C/)    •— *f
                    OO    O
                    1—    O
                           . (rt .
                            -M
                            O) (O C
                        •,   u 4->
                       O   $- -W VJ-
                       O   O) ••- to
                       0   Q. ECM
Standar
Count p
                                                                     i- m
                                                                     So

                                  161

-------
     10'
     icr
 s_
 QJ
 O.
 £
 S
 o;
 o
     10'
     10"
                           INFLUENT
             GM
           15        20         25

                        July
30
      August
Figure 89.  Standard plate count data for  Run  No.  U6.
                        162

-------
    10
    icr
o
o
Q.
Di

O
    10
    TO
    10
    10
      0   I   -  -    I          I
          15       20       25       30    I   3

                        July    .            August


    Figure 90.   Total coliform data for Run No. U6.
                       163

-------
    10'
    10*
o
o
s-
01
CL
00
p     o

    10
    103
    10'
    10
      p  I
         INFLUENT


          6M
I
J	I
        1         15       20       25   |  3


                    July                  August


   Figure 91.  Fecal coliform data for.Run No. U6,
                       164

-------
                                    Log Y = 0.54  X- 0.05

                                           r = 0.86
                                                5
                     LOG REDUCTION TOTAL COLIFORMS
Figure 92.  UV Intensity versus log reduction of total coliform data
            for Run No.  U6.                 -
                                 165

-------
O

tS)
+J
•P
rd
CO

LU
o
LU
CC
   10'
   10
      0
   10
     -1
                                      Log  Y  =  0.54 X+ 0.01
                                              r =  0.84
 Figure 93.
                1        23       45       6        7

                         LOG REDUCTION FECAL COLIFORMS

              UV Intensity versus log reduction .of  fecal  coliform
              data for Run No. U6.
                               166

-------
existed and the estimating equations and correlation coefficients are pre-
sented on Figures 92 and 93..

     Extreme- value  frequency distributions for the microbiological data
are shown on Figures 94,  95  and 96.  A fecal coliform density of 200/100 ml
would be exceeded over 70 percent of the time.  Additionally, the steep
slope of the effluent line is indicative of extreme variability in the data,
a situation that should be expected with a poor effluent and a low UV dose.


RUN NO.  U7

     Run No. U7 was initiated immediately after the installation of new
aeration -equipment on 8 August and terminated on 5 October 1975.  As shown
in Figure 97 the process configuration consisted of mixed-media filtration
of the activated sludge effluent followed by UV disinfection with the UPS
unit.

     Water quality data for the unit processes are summarized in Table 29,
and indicate a substantial improvement in treatment over runs U5 and U6.
Nitrification was not complete; however, the effluent ammonia nitrogen
averaged only 4 mg/1 , indicating much improved treatment.  The BOD5 and
COD in the filter effluent averaged 3 and 35 mg/1, respectively, in contrast
to the BOD5  of 18 mg/1 and the COD of 69 mg/1 observed during the previous
run.

     The data obtained from, the microbiological samples are summarized
in Table 30, and reflect the general improvement in treatment.  The mean
color value was reduced from the greater than 48 Pt-Co units observed during
Run No.  U6 to 19 Pt-Co units during this run.  Most importantly, the
transmissability increased from 54 percent to 64 percent.
     During Run No. U7 the mean flow rate to the UPS unit was 3.1 liters/sec.
The resulting theoretical residence time was 17.3 seconds, and the average
actual residence time was 15.5 seconds.  The transmissability at 254 nm of
64.1 percent corresponded to an extinction coefficient, k, of 0.45 cm~^
which resulted in a calculated dose, DT, of 14,500 ywatt-sec/cm2.

     An average intensity of 123 uwatts/cm2 was measured with the IL 500
radiometer.  When this value was multiplied by 8.9 to correct for geometry,
the resulting UV intensity at the quartz window in the exposure chamber
wall was 1100 ^watts/cm2.  This intensity value is 300 percent greater than
that measured during the previous run when effluent quality was poor.  The
resulting indicated dose, D, , was 17,000 ywatt-sec/cm2.

     It is important to note that the flow was increased from 2.2 liter/sec.
in Run U6 to 3.1 liters/sec, during Run U7.  This resulted in a 30-percent
decrease in residence time within the exposure chamber, which should have
reduced the UV dose proportionately.  However, the improved effluent quality,
which resulted in enhanced UV transmissability  actually increased the UV
dose.
                                     167

-------
                            EXTRAPOLATED
     99 90 80 60 40   20   10   532    1   0.5   0.2  0.1
              PROBABILITY OF BEING  EXCEEDED

Figure  94..  Extreme-value frequency distribution of standard
            plate count data for Run No. U6.
                           f!68

-------
                                EXTRAPOLATED
         99 90 80 60 40   20   10   532   1   0.5  0.2 0.1

                  PROBABILITY OF BEING EXCEEDED


Figure 95.   Extreme-value frequency distribution of total coliform
            data for Run No. U6.
                               169

-------
                                 EXTRAPOLATED
           99  90 80 60 40   20   10   532   1   ,0.5  0.2  0.1

                    PROBABILITY OF BEING EXCEEDED

Figure  96.  Extreme-value frequency distribution  of fecal coliform
            data for Run No. U6.
                                170

-------
 • "O
O 
-------
TABLE 29.  PERFORMANCE OF THE NO. 1  ACTIVATED SLUDGE SYSTEM AND
           MULTIMEDIA NO. 1 FILTER DURING RUN NO.  U7


Parameter


TSS
BOD5
COD/sol COD
TOC, soluble
NH3-N
Org-N
N00&NO.-N
2 3
Total N
Total P
Turbidity NTU
A.S.
Influent
mg/1

83
127
283
58.3
19.6
9.7
0.2

29.5
10.1
__
A.S.
Effluent
mg/1

37
9
61.9/30.2
14.3
4.6
5.0
4.7

14.3 '
8.1
1.7

No. 1 Filter
Effluent
mg/1
23
3 -
35.3
13.8
4.0
2.8
4.6

11.4
7.1
1.6

Removal
percent

72
98
88
76
80
71
_

61
30
_
   Color,
    Pt-Co units
17
18
                                 172

-------
-if
 E CC
                                              i—     CM
                                             i—     CM
                                              CO     i-     ,-
              10

              o   •—
                                                                  2
                                                                  O
                                                                  'i
                                       O    i—    r—    i—
                  r-    O
                                                                  2  o
          .—   Or—
     ^o
CO     r—
«     2S
CO     CO

      CVJ
                                                                     R
                                                                     0)
                                   •4-* O
                                   ss
                 {2
tJ   C3
o ,  o
1—   O
                                           =*
                                            .           ^
                                             i-  i— E   i—
                                           •gg.
fO 3
4-> O
(/) CJ>
4-» S-
O Q*
t— O-
                O S-
                OJ O)
                U- Q.
                              173

-------
     The improved biological treatment and the resulting higher quality
effluent had a pronounced effect on the UV disinfection process.  The
geometric mean fecal  and total coliform densities were less than 100  per  100
ml and 190 per TOO ml,  respectively, greater than one-log improvement
over the results obtained during the previous run.

     The coliform organisms were reduced slightly more than 3.5 logs.  This
is not as good as the greater than 4-1og reductions observed during Runs
Ul, U2, and U3; however,the UV doses used during these runs were approximate-
ly double the dose used during this run.

     Time-series plots of the microbiological data are shown in Figures 93,
99,and 100, while Figures 101, 102, and-103 present the extreme-value
probability distributions.

     The very significant improvement in disinfection that resulted from
improved treatment is clear in the time-series plots.  The effluent coliform
densities decreased approximately two logs on August 9.

     UV intensity, as measured by the IL 500 radiometer at the quartz window,
was plotted versus observed log reductions in total and fecal coliforms.
Poor correlations were obtained.  For total coliforms the equation of the
line obtained by linear regression analysis was log y = 0.064 x - 4.14.
The correlation coefficient obtained was 0.39.  For fecal coliforms,  the
equation obtained was log y = 0.057 x - 4.11.  The correlation coefficient
was 0.28.

RUN NO. U8
     The last UV disinfection run was started on 7 October and terminated
on 31 November 1975.  As shown in Figure 104, activated sludge effluent
was pumped directly to the UPS unit.  Water quality data for the No.  1
activated sludge system are summarized in Table 31.  Treatment was adequate,
as indicated by the moderately low mean COD and TSS concentrations.
Nitrification was incomplete, with an average of 5.4 mg/1 of ammonia nitrogen
being found in the effluent.

     The objective of this run was to stress the disinfection capabilities
of the UPS unit as much as possible.  For this reason the unit was operated
at the maximum obtainable flow of 3.2 liters/sec (51 gpm).  The resulting
theoretical and actual residence times were 16.8 and 15.6 seconds,
respectively.  An extinction coefficient, k, of 0.50 cm"' resulted from the
mean transmissability at 254 nm of 61.2 percent.  The theoretical,calculated
dose, Dy, was 11,500 ywatt-sec./cm2.  The average intensity measured with
IL 500 radiometer was 100 uwatt-sec./cm2 which resulted in an indicated dose,
Dj, of 13,400 uwatt-sec/cm2 after being corrected for geometry and multipli-
ed by the actual residence time.

     Data from the microbiological samples are presented in Table 32.
During this run the mean coliform reductions were slightly less than three
logs, which is not as good as had been observed in previous runs when the
                                     174

-------
                                                            CJ3
I—
  O
IO
 O
LO
  o
oo
  o
CM
 o
                                                                                          LO
                                                                                          CD
                                                                                          CO
                                                                                          LO
                                                                                          CM
                                                                                          O
                                                                                          CM
                                                                                          LO
                                                                                  S-
                                                                                  cu
                                                                                  _a
                                                                                  o
                                                                                  4->
                                                                                  O
                                                                                  CD
                                                                                               S-
                                                                                               OJ
                                                                                               O)
                                                                                              •4->
                                                                                               Q.
                                                                                               CD
                                                                                              OO
                                                                                          O
                                                                                          CO
                                                                             LO
                                                                             CM
                                                                                          O
                                                                                          CM
                                                                                              CO
                                                                                              3
                                                                                              CD
                                                                             LO
                                                                                                  •M
                                                                                                   to
                                                                                                  -a
                                                                                      o
                                                                                      o

                                                                                      OJ
                                                                                      -M
                                                                                      (C
                                                                                      r—
                                                                                      Q.


                                                                                      "2
                                                                                      
                                                                                      OO
                                                                                                  00
                                                                                                  CTi


                                                                                                  O)
O
 O
                                       Liu  aad 'SWSINV9HO
                                                175

-------
Liu COL
'SWHOdllOO 1V101
         176

-------
                                            u_
Liu 00 L
'SWUOJIIOO 1V03J
         177

-------
    10'
 s-
 
-------
                                  EXTRAPOLATED
      :  10
   o
   o
   S-
   C1J
   Q-
   o;
   o
   o
   O
           99 90 80 60 40   20   10   532    1    0.5   0.2  0.1
                    PROBABILITY OF BEING EXCEEDED
Figure 102.  Extreme-value frequency distribution'of  total coliform
             data for Run No. U7.
                                 179

-------
                                   EXTRAPOLATED
  o
  o
  CD

  O.
  o:
  s.
  o
  o
  s
  LU
           99 90 80 60 40   20   10   532   1   0.5   0.2  0.1


                    PROBABILITY OF BEING EXCEEDED
Figure T03.   Extreme-value frequency distribution of fecal  col.iform

             data for Run No.  U7.
                                180

-------
                                                                             co

                                                                              S-
                                                                              CD

                                                                              O
                                                                             O
                                                                             -M
                                                         M  O
                                                         Q_  UJ
                                                             co
                                                             I—<
                                                             Q
 O
 CD
 C/J
co    i—
                                                         O
                                                         cu
                                                         t/5
                                                        CVJ
                                                          •
                                                        CO
                                                                              C1J
                                                                             -Q
                                                                              O
                                                                             •M
                                                                              O
                                                                             O
                                                                             oo
                                                             Q.
                                                             C75
                                                                              i.
                                                                              o
                                                                             H-

                                                                              E-
                                                                              O
 3
 CD
                                                                             O
                                                                             O
 ea

O-

4->   •
 O LO

M^ en
Q- i—
                                                                             "vf
                                                                             O
                                                                             CU
                                                                             cn
                                                                             •r-
                                                                             U_
                                       181

-------
TABLE 31.    PERFORMANCE OF THE NO.  1  ACTIVATED SLUDGE
             SYSTEM DURING RUN NO.  U8

Parameter
TSS
COD/COD sol
TOC, soluble
NH3-N
Org.-N
N02 & N03-N
Total N
Total P
Turbidity, NTU
Influent
mg/1
91
310
61.2
22.7
11.8
0.4
34.9
10.0
7.4
Ef f 1 uent
mg/1
17
63.4/33.9
1319^
5.4
3.7
6.7
15.8
8.0
7.7
Removal
percent
81
80
77
76
69
-
55
20
•
                        182

-------
                                                      i—     Ol
OCM
   c
en to
                    t       111
                                                             •S
                            I    10   CO
                                                    CO      i—     i—
                                                      o     o      o

                                                      x     "x      "x
                                                      Ol     CD      O
                                5    S
                                                      i—      in
O) ttt
oo;
                                                      s
                                              OJ     ^-
                                              O


                                              CO
                                                      
                                          
-------
quality of the activated sludge effluent was good.   Additionally,  the
effluent total and fecal coliform densities were somewhat higher than in
previous runs with.effluent of similar quality.

     The combination of decreased residence time and the relatively low
percent transmittance resulted in one of the lowest UV doses  used  during
the project.  The calculated dose., Dy, during Run U8 was only about one-third
of the dose used during Run Ul through U3, and was  about one-half of the
dose used during Runs U4 and U5.  The dose used during Run U7 was approxi-
mately 30 percent higher than the 11,500 ywatt-sec./cm2 used during this run.
The low dose used was the reason for the decrease in disinfection efficiency.

     Figures 105, 106, and 107 are time-series plots of the microbiological
data.  The apparent variation in bacterial densities in the UV disinfected
effluent is approximately one log either side of the geometric mean.  The
effect of the variability can be seen in the extreme-value frequency ,
distributions in Figures 108, 109, and 110.  The indication is that 50
percent of the time a fecaT coliform density of about 320 per 100 ml will be
exceeded, and that a density of 400 fecal coliforms per 100 ml will be
exceeded approximately 45 percent of the time.

     As in Run U7 poor correlations were observed when UV intensity was,
plotted versus observed log reductions in total and fecal coliforms.  When
plotted versus log reduction of'total coliforms the equation of the line
obtained by linear regression analysis was log y = 0.048 x - 3.79.  The
correlation coefficient for this line was - 0.15.  When plotted versus log
reduction of fecal coliforms the equation obtained was log y = 0.035 x
- 3.83.   The correlation coefficient for the line was - 0.09.
                                     184

-------
                                     1551.
vo
 o
                                                                             o
                                                                             CO
                                                                             LO
                                                                             CM
                                                                             o
                                                                             CM
                                                                                 s-
                                                                                 O)
                                                                             LO
                                                                                 CD
                                                                                 o
                                                                                ^  CO
                                                                             r—      O
                                                                             LO
                                                                             O
                                                                             CO
                                                                 uo
                                                                 OJ
                                                                                 S-
                                                                                 
                                                                                     03

                                                                                    "o.
LO
  o
CO
  o
CM
  o
o
 o
                                                                                     cu
                                                                                     S-
                                                                                     C7)


                                                                                    Lll
         Liu
                               'SWSINVQciO
                                    185

-------
                                       CO
LIU oo L
•*swaojnoo  ivioi
      186

-------
                                                             Lrt
                                                             CM
                                                             O
                                                             CM
                                                                S-
                                                             LO CO
                                                                .a

                                                                cu

                                                                O
                                                             LT>
                                                             O
                                                             CO
                                                             un
                                                             CM
                                                              O  S-
                                                              CM  CO
                                                                _Q
                                                                 O
                                                                •(->
                                                                 O
                                                                O

                                                              LO
                                             §
                                              •
                                             O
                                             s:


                                             a;
                                             s-
                                             o

                                             (C

                                             -a
                                              O
                                              O
                                              (O
                                              O
                                              CO
                                                                     r-.
                                                                     O
                                                                      cu
 un
  O
oo
 O
CM
 O
O
  O
Liu 00L  ^d.'SWHOdHOO  1V03J
                        187

-------
     10y
     10*
     10
 Q.



60


&
t-t
     In
DC   10
o
     10
     10J
     10
                               EXTRAPOLATED
                          INFLUENT
                       1 EFFLUENT
        99 90 80 60 40  ' 20    10    532    1    0.5   0.2  0.1

                 PROBABILITY OF BEING  EXCEEDED
Figure  108.  Extreme-value frequency distribution of standard

              plate  count data for Run No.  U8.
                             . 188

-------
     10'
     10
o
o
DC

£
o,
     1(T
     10
       2.
     10J
     10
       0

                                 EXTRAPOLATED
                        INFLUENT
                              EFFLUENT
        99  90 80 60 40   20   10   532   1   0.5  0.2 0.1


                 PROBABILITY OF BEING EXCEEDED
   Figure 109.  Extreme-value frequency distribution of total

                coliform data for Run No. U8.
                             189

-------
   10y
   10
   io
o   10
o
O)
Q.
C£.

£
o -  i n'
UJ  10
    IO
    10'
                                EXTRAPOLATED

                  /
                       INFLUENT
z
  EFFLUENT
       99 90 80 60 40   20   10   532    1    0.5   0.2  0.1

                PROBABILITY OF BEING EXCEEDED
 Figure  110.  Extreme-value frequency distribution of fecal
              coliform data for Run No. U8.
                             190

-------
                                 SECTION 8

                               VIRUS STUDIES
     One of the objectives of the project was to evaluate the comparative
resistance of poliovirus, F2 bacteriophage, and total and fecal coliform
bacteria to ultraviolet radiation.  The poliovirus strain used was an atten-
uated Type I culture obtained from Dr. 6. Berg, U.S. E.P.A., Cincinnati,
Ohio.

     A total of three replicate experiments were conducted on the following
dates: (1) 22 April 1975, (2) 13 May 1975, and (3) 26 June 1975.  Initial
volume of poliovirus seed was approximately 3.2 liters, containing approxi-
mately 8x1O7 plaque forming units (pfu) per ml.  Bacterial virus F2 was
added to the poliovirus culture at a titer of 3xl08 pfu per ml.

     Figure m  illustrates schematically the flow diagram of a typical run.
Activated sludge effluent was used as a wastewater source.  The virus/phage
seed was pumped into the effluent stream at a rate of 100 ml/minute on the
suction side of a centrifugal pump,  where the seed became well  mixed
with the effluent prior to entering the UPS UV unit.  After emerging from
the UV unit the effluent flowed into chlorine contact basins, where a very
high chlorine residual was maintained to destroy any poliovirus which
might have survived the UV irradiation.  Three sample taps were located as
shown in the diagram.

     After collecting the control samples the UV lamps were energized and
allowed to reach their peak output.   Ultraviolet intensity readings were
made with the IL 500 radiometer, and the water quality meter provided with
the UPS unit.  After the warm-up time, three separate test samples were
collected for virus enumeration  (1) prior to virus/phage addition (for
indigenous virus),  (2) after the.pump, but before the UV unit,  and (3)
after the UV unit.    Three duplicate samples were collected one minute
later, and another duplicate set, one minute after that.  This entire pro-
tocol was repeated at all four wastewater flow rates.

    In order to protect the chemists and biologists from poliovirus exposure,
the bacteriology and chemistry samples were collected prior to addition of
the virus/phage inoculum to the wastewater, but with the UV unit at steady
state and the lamps on.  No chemical analyses were .performed on the seeded
samples.
                                    191

-------
                      Q.
                      (O
 CU
i—   CU
 Q-  >
 £  r-

CO  >
                                S-
                                CU
                               4-5
                                CU
                                                       c:
                                                       3
in
3
                                                       CO
                                                       £1
                                                       O
                                                       Q.
                                                       q-
                                                       o
                                                       (O
                                                       O
                                                       O
                                                       CU
                                                        en
 CD
 S-  CO
 O  CO
i— 03
                     192

-------
FIRST UV VIRUS RUN

     An operating summary of the UV unit for the first virus run appears on
Table 33.  The hydraulic characteristics .and start-finish meter readings are
presented for each of the flow conditions investigated.

   ,  A summary of the chemical characteristics of the wastewater before and
after exposure to UV radiation averaged over all four flow rates is given
in Table 34.  No significant change in any wastewater quality variable, ex-
cept possibly BOD, was observed.  A summary of the bacteriological data is
given in Table 35.  It is clear that, as wastewater flow rate decreased,
total and fecal coliform reduction increased.  This was expected, since a
decrease in flow rate corresponds to an increase in residence time, and
thus, an increase in exposure time to UV irradiation.  It also appears that
total and fecal coliforms displayed approximately the same relative resis-
tance to UV irradiation.

     Results from the phage seeding experiments are given -in Table 36
(control sampling) and 37 (test sampling).  The purpose of the control
sampling was to establish the attainment of steady-state conditions at
each flow rate prior to the test sampling.  The UV lamps were turned off
during control sampling.  Steady-state conditions were established when
influent and effluent titers were the same.

     No  F2 phages, were detected in either the Control 2 samples (before UV)
or the Control 3 samples (after UV).  The phage/poliovirus seed suspension
was conveyed from a cubitainer 3 meters above the floor of the filter
gallery through a metering pump  and tygon tube.  The 3-meter length of
tygon tubing had been flushed with sterile distilled water prior to the run,
and was still full of the water when the virus feed pump was activated.
The virus/phage suspension had not completely displaced this water at the
commencement of the first run (4.7 I/sec.), and as a result the influent
titers were negative.  This effect was not repeated in the three successive
runs, because the line was not flushed between runs.

     Steady state conditions were established at 3.2 and 1.9 I/sec (Table
36).  At 0.6 I/sec, two of the Control 3 samples were negative for phage.
This was presumably because the UPS UV unit had not been completely flushed
following the last test run  when the phage titer had been reduced by
several logs.  As a result the phage titer at the beginning of the control
run was lower than expected, and thus the phage numbers were undetected
because the dilutions were carried out too far.

     Examination of Table 37 reveals that the number of indigenous phage
present in the wastewater was insignificant relative to the F2 titer after
seeding.  It is also evident that F2 phage were undetected in the samples
before and after the UV unit during the 4.7 I/sec run probably for the
same reasons mentioned for the control samples.  In general, it appears
that F2 bacteriophage may be slightly more resistant to UV than total  or
fecal coliforms, as indicated by the mean log reduction values (Table 37).
                                    193

-------
TABLE 33.   OPERATING SUMMARY OF THE FIRST UV VIRUS RUN

Flow Rates, I/sec
Theoretical t, sec.
100 percent flushing time,
sec.
WQ Meter (Start), Dimension-
less units - scale (0-40)
WQ Meter (Finish), Dimension-
less units (scale 0-40)
Radiometer (Start) yw/cm
Radiometer (Finish), yw/cm
4.7
11.4

28

32

27
550
530
3.2
16.8

42

29

24
550
540
1.9
28.2

70

<20

20
550
540
0.6
85.1

210

30

25
550
540
                          194

-------
TABLE 34.  RESULTS OF CHEMICAL ANALYSES OF GRAB SAMPLES AVERAGED OVER ALL
           FOUR FLOW RATES IN THE FIRST VIRUS RUN.

Parameter Influent to UV Effluent from UV
COD, .mg/1
TOC, mg/1
SOC, mg/1
BOD, mg/1
TSS, mg/1
Turbidity, NTU
Color, Pt-Co units
NH3-N, mg/1
Org-N, mg/1
N02-N + N03-N
N02-N
D.O. , mg/1
pH
Sp. .Cond.
Total Alk. , mg/1
p Alk. , mg/1
TDS, mg/1
Cl~, mg/1
%Transmittance at
254 nm
46.2
15.5
10.0
10
20
2.8
30
1.8
4.4
6.0
0.3
3.4
7.4
727
131
0
522
58
67.2
47.9
15.2
10.8
4.5
22
2.8
30

• -
. -

3.3
*v
7.3
750
-
-
-
-
66.6
                                 195

-------
a:
to
ui
uu

•=C
g
i—i
CD
O
a;
LU
CO

LU
:n
o
>-
          O
          (U
          to
         CD
         o
         LZ
LO
CO
LU
_1
CO




vo

o






cr>










CM

CO






.
'*






,
<*-
i . i
LU






n *
1 — I
LU



1-
E
i— (


•
^ —
M-
LU





•
4-
C
1—4

•
^^
"x
o
CO

LO
0
f~_
X
^
CM
CD
"x
CD
LO"



vo
o

"x
co
o
r—
X
CM


ID
O

"x
co
^_
CO
FZ
S-
O r—
r— «t- E
(O T-
+-> r— O
O 0 O
I— CJi—




0
ID

LO
A



o
o
LO










CO
LO
co





CD
F-—
CM





c
0
•r—
.f^
O
=3

0 0}



















































^0
*x
o
CM
V


LO
CD
"x
co
•
0
CD
'x
O
CM

^
O
^—
X
r-.
CM
o
'x
0
CM



LO
O
i— •
X
CM
O
"x
0
«*

LO
o
^—
X
CM

CM
V> E

S- O
0 0

n3 -r-
Or- S-
CD O O)
U. 0 0-




CO
•
^«
A




0
lO












r~~
CM





•3-
^
CM




>. .
J-
0
•r—
•»->
O
3
o> -o
O CD
_J C£
                                         196

-------
    TABLE 36.     STEADY-STATE TITERS OF F2 BACTERIOPHAGE CONTROLS IN
                 FIRST VIRUS RUN.

Phage titers, pfu/ml
Detention time, sec
(Flow Rate, I/sec)
Sample

Control 2
30 sec
60 sec
90 sec
Geometri c
Mean
Control 3
30 sec
60 sec
90 sec
11.4
(4.7)
Before UV
< 2. 0x10°
< 2. 0x10°
< 2. 0x10°
2.0x10°
After UV
< 2. 0x10°
< 2. 0x10°
< 2. 0x10°
16.8
(3.2)

4.3xl04
3.6xI04
8.3xl04
S.OxlO4
5.3xl04
7.1xl04
1.2xl05
28 85
(1.9) (0.6)

1.4xl05 5.0xl04
7.6xl04 l.OxlO4
1.2xl05 4.2xl05
l.lxlO5 5.9xl04
7.8xl04 <2.0xlO°
7.1xl04 <2.0xlO°
3.9xl04 T.2xl06
Geometric    <2.0x10
Mean
Control 1 SEED CULTURE
          3.2x10

          2.8x10
                8
8
              7.7x10
6.0x10
                                                  ,4
<1.7xlO
                                     197

-------
      TABLE 37.    TITERS OF F2 BACTERIOPHA6E EXPOSED TO ULTRAVIOLET
                   RADIATION AT DIFFERENT FLOW RATES IN FIRST VIRUS RUN.



Phage Titers,
pfu/ml

Detention Time, sec
(Flow Rate, I/sec)
Sampl e
INDIGENOUS PHAGE
0 sec
60 sec
120 sec
Geometric Mean
F2 Phage
Before UV Unit*
0 sec
60 sec
120 sec
Geometric Mean
F2 Phage
After UV Unit*
0 sec
60 sec
120 sec
Geometric Mean
Log Reduction
11.4
(4.7)

1.7xl02
2.7xl02
1.2xl02
1..8xl02

lx!03
>,lx!03
>lx!03
-
16.8
(3.2)

9.5X101
9.8X101
l.lxlO2
l.OxlO2

2.8xl04
5.9xl04
2.8xl04
3.6xl04

3.9xl02
3.6xl02
6.8xl02
4.6xl02
1.90,
28
(1.9)

4.9X101
6.3X101
6.2X101
B.SxlO1

__
4.4xl04
9.6xl04
6.5xl04

l.lxlO2
1.6xl02 ,
l.SxlO2
1.5xl02
2.63
85
(0.6)

l.lxlO2
6.5X101
.l.lxlO2
9.2X101

7.7xl05
6.0xl04
2.9xl05
2.4xl05

5.4X101
8.0x10°
l.lxlO1
1.7X101
4.15
*UV light had been on for two minutes.
                                    198

-------
     Results from the poliovirus seeding experiments are summarized in Table
38.  Poliovirus was undetectable  in the control samples during the 4.7 I/sec
run due to the incomplete flushing of sterile distilled water from the feed
line.  It is clear that steady-state flow conditions were established<-at the
other three flow rates, as indicated by the comparable virus titers between
the Control 2 and Control 3 samples at each flow condition.  Mean log
reduction of poliovirus increased with decreasing flow rate (or increasing
detention time).  Also, poliovirus appears to have elicited a similar re..^
Sponse to UV inactivation as coliform bacteria. A complete summary of the
first virus run is given in Table 39.

     A summary of the dose-related data for the first UV run appears in
Table 40.  Dosages were calculated as enumerated in Section 7.  The very
high indicated doses, Dj, were a direct result of the high intensity read-
ings.  The highest doses, Dj, were obtained during this run, as well as the
best disinfection, for both viruses and coliforms.


SECOND UV VIRUS RUN

     The second poliovirus run with the UPS unit was conducted on 13 May
1975.  On the afternoon of 12 May, the UV unit was turned off, taken out of
service, cleaned with a proprietary cleaning solution provided by UPS, Inc.,
rinsed with tap water, and drained.  Early the following morning it was
refilled and flushed again with tap water.   The lamps were then turned on,
and intensity readings with the UPS water quality meter and IL 500 radio-
meter were recorded every 30 seconds for a period of 5 minutes.  Figure 112
shows the responses.  Maximum output was attained approximately 1.5 minutes
after startup.  The lamps were then turned off for a.period of 5 minutes,
but tap water flushing continued.  Then the lamps were turned on, and again
the readings were recorded, as shown in Figure 113.   This time, the response
was much quicker, as the maximum output was attained after only 1 minute.
Following this, the lamps were turned off,  and activated sludge effluent was
valved into the unit for the first time since the cleaning.  About 1,5 min-
utes later the lamps,were turned on, and meter readings were recorded, as
shown on Figure 114.  The response was significantly different from the tap
water readings.  First, the UV output reached a maximum level  after only
20 seconds, and the maximum output was higher than that recorded on tap
water.  Then, between 2 minutes and 4 minutes the intensity declined from
110 to 100 uwatts/cm2.  At the same time, the water quality meter responded
much differently, reaching a maximum value of only 10 after 30 seconds, but
remaining unchanged thereafter.

     The ultraviolet intensity readings of 100 to 110 ywatt/cm2 on tap. water
were unusual in that in previous disinfection runs intensity readings of
approximately 550 uwatts/cirr on wastewater effluent had been obtained.  There
are several possible reasons for the low intensity readings.  First,
film build-up on the quartz sleeves might have been inadequately removed.
Second, the proprietary cleaning compound used may absorb UV light.   If the
                                    199

-------
TABLE 38.     TITERS OF POLIOVIRUS'TYPE I EXPOSED TO UV RADIATION AT DIFFERENT
              FLOW RATES IN FIRST VIRUS RUN.
Virus Titers, pfu/ml
Detention Time, sec
(Flow Rate, I/sec)
Sample
Control 2-before UV
30 sec
60 sec
90 sec
Geometric Mean
Control 3-after UV
30 sec
60 sec
90 sec
Geometric Mean
Before UV Unit* ,
0 sec
60 sec
120 sec
Geometric Mean
After UV Unit*
0 sec
60 sec
120 sec
Geometric Mean
Log Reduction
11.4 16.8
(4.7) (3.2)

0 9.2xl03
0 8.2xl03
0 7.9xl03
0 8.4xl03

0 6.0xl03
0 S.OxlO3
6 l.lxlO4
0 S.OxlO3

5.8xl03 1.2xl04
6.8xl03 l.OxlO4
8.6xl03 l.OxlO4
7.0xl03 l.lxlO4

2.8X101 9.0x10°
7.2X101 l.OxlO1
5.9X101 4.0x10°
4.9x10 7x10
2.15 3.23
28
(1.9)

1.2xl04
l.SxlO4
l.lxlO4 '
l.SxlO4

1.2xl04
1.3xl04
l.SxlO4
l.SxlO4

9.7xl03
l.lxlO4
l.SxlO4
l.lxlO4

1
0
1
<1
>4.04
85
(0.6)

3.9X104
S.OxlO4
S.lxlO4
4.6xl04

2.6xl03
,1.4xl04
,2.2xl04
9.3xl03

3.7xl04
5.4xl04
4.5xl04
4.5xl04

0
3
0
<1.4xlO°
>4.49
*UV lights had been on for 2 minutes.
Other Results:
    All samples from the C12 contact basin were 0 .
    All samples of indigenous virus were  0.
    Control 1 (virus feed):  start = 3.8xl07 PFU/ml
                             end   = 2.9x1O7 PFU/ml

                                       200

-------
TABLE 39.  SUMMARY OF RESULTS FROM FIRST VIRUS RUN.

MEAN LOG REDUCTIONS
Organism
11.4
(4.7)
Total
Coliforms 2.70
Fecal
Coliforms 2.74
F2
Bacteriophage
Polio virus
Type 2 . 2.15
Detention Time, sec
(Flow Rate, I/sec)
16.8
(3.2)
3.53
2.74
1.90
3.23
28
(1.9).
5.00
4.60
2.63
4.04
85
(0.6)
>5.60
>4.81
4.15
>4.49
                                     201

-------
TABLE 40.  SUMMARY OF DOSE-RELATED DATA FOR THE FIRST VIRUS RUN

Flow Rate,! /sec
Theoretical Detention
Time, sec
Actual Detention
Time, sec
Transmittance @
254 nm, percent
Extinction Coefficient
cm"!
UV Intensity,
uwatt/cm2
Calculated Dose,0
Dp wwatt-sec/cm2
Indicated Dose, Dj,
ywatt-sec/cm2
4.7
11.4
10
66.9
0.41
540
11,200
48,100
3.2
16.8
15.5
66.9
0.41
545
16,500
75,200
1.9
28
25
66.9
0.41
545
27,500
121,300
0.6
85
75
66.9
0.41
545
83,400
363,800
                                     202

-------
"S   80
   ,  60  -
o   40
a

-------
-o
0)
(O
 W)
 to
oo
CD
UJ
c;
o:
UJ
                           RADIOMETER, ywatt/cm'
WQ METER, dimension!ess
                              TIME, minutes


   Figure 113.  Meter  readings after five minutes of purging
                with tap  water.
                              204

-------

 
-------
disinfection unit was not adequately flushed after it was cleaned, the
residual cleaning compound may have caused the low intensity readings.
Another possibility is that the tap water might have contained some UV
absorbing materials that the wastewater did not.  Although this is improbable,
it cannot be eliminated as a possibility since the percent transmittance of
the tap water was not measured.

    The second virus run commenced approximately one hour after the above
intensity experiments.  As in the first virus run sampling was done at
four flow rates: 4.7, 0.6, 1.9, arid 3.2 I/sec., respectively.  The sampling
protocol was not  substantially different from the first run.  The most sign-
ificant changes involved the timing at the 0.6 I/sec flow rate,  which was to
permit a sufficient amount of time to pass for flushing of the lines, and UV
unit.

     A brief operating summary of the second virus run is presented in Table
41.  Start and stop readings were noted on both the International Light
radiometer and the UPS water quality meter.  Analytical summaries of the
chemical and bacteriological data follow on Tables 42 and 43.  Some addition-
al chemical analyses were performed on the initial and final influent
samples (4.7 and 3.2 I/sec, respectively), including N-series, TDS, and
chlorides.  It is evident from Table 42 that there were no significant dif-
ferences in the chemical characteristics between influent and effluent sam-
ples in the second virus run.

     The data in Table 43 indicate that the total and fecal coliform log
reductions were lower than in the first virus run, probably reflecting the
effect of reduced ultraviolet dosages as determined by I6wer intensity mea-
surements.  It is difficult to determine the cause of the diminished UV
intensity readings in the second virus run. Comparison of Table 42 and'with
Table 34 reveals that the only wastewater quality variable which increased
in the  second virus  run was color.  This may have been responsible for the
decreased percent transmittance observed in the latter run.  Another possi-
bility, discussed earlier, was film build-up on the quartz sleeves surround-
ing the UV lamps, thereby partially attenuating the UV output from the lamps.

     Results from the phage seeding experiments are given in Tables 44 and
45.  It is clear from Table 44 that the displacement problem encountered
with the control samples of the first virus run was resolved, since phage
were present in all  control samples.  Moreover, steady-state flow conditions
were established at  all flow rates, since influent and effluent samples
both  contained comparable phage titers.  It is evident from Table 45  that the
titers of indigenous phage were negligible relative to the seeded samples.
Results of exposing . F2 phage to ultraviolet light indicated that the phage
again were more resistant than total and fecal coliform  bacteria.

     Results from the poliovirus seeding experiments are summarized in Table
46.  Steady-state flow  conditions were established, as evidenced  by the com-
parable virus titers between the influent and effluent control samples at
each flow condition.  Again, mean  log  reduction of poliovirus increased
with detention  time,  as  in  the first virus run.   It appears  that  poliovirus

                                     206

-------
      TABLE 41.    OPERATING SUMMARY OF SECOND UV VIRUS RUN
Flow (I/sec)
4.7
3.2
1.9
0.6
WQ Meter (start)
Dimension!ess Units
(scale 0-40)            10
WQ Meter (finish)
Dimension!ess Units
(scale 0-40)            10
Radiometer (start),
yw/cnr                  93
              97
              97
              97
Radiometer (finish),
yw/cnr                  107
              97
              93
              97
                                    207

-------
TABLE 42.  RESULTS OF CHEMICAL ANALYSES OF GRAB SAMPLES AVERAGED OVER ALL
           FOUR FLOW RATES IN THE SECOND VIRUS RUN.
Parameter
COD, mg/1
TOC, mg/1
SOC, mg/1
BOD, mg/1
TSS, mg/1
Turbidity, NTU
Color, Pt-Co Units
NH3-N, mg/1
Org-N, mg/1
N02-N + N03-N, mg/1
N02-N, mg/1
PH
Sp. Cond.
Total Alk., mg/1
TDS, mg/1
Cl", mg/1
%Transmittance at
254 nm (as received)
%Transmittance at
254 nm (centrifuged)
Influent to UV
37.4
12
8
14
23
3.6
40
4.4
3.6
0.8
0.11
7.6
782
172
481
64
59.3
58.7
Effluent from UV
39.1
13
8
9
24
3.7
42
(
7.6
782
172
,
-
65.6
67.6
                                    208

-------
O
O
LiJ
CO

LJLJ
in




I
uu
CO
CD
O
O
o;
LLJ

          o
CO
•*

LiJ
_J
CO

4-
4-
LU
<0
CD


4-
E




4-
4-
UJ


°">
, — •


4-
C

q_
4-
LjJ
CM
co
4-
c
I — 1


4-
4-
LU

1*^
'

4-
i-H
0)
1 \
QJ
re
S-
(0
a.
,_
o

p
«o
o
X
co
•"•
CM
O
r—
X
CO
•
CM



LO
O
"x
en
***
CM
o
'x
o
CM
vo
o
"x
•*
I—
0
"x
p—


o
'x
r—
r^ S-
0 0)
CO Q.
"re E
-t-> s- o
o o o
l"~ ' *I— ' ^^


to
OJ

*4"










^^*
LO
CO




„



co
CO







01
1 —
CM

z:
o
1—)
1—
co
CD a
O UJ
o
'x
o
K


LO
O
^
co
co
.""o

'x
o
oo"




LO
o
x
CO
CM
CM
O
X
o
CM
LO
O
f i ii
X
CO
CM
CM
O
"x
o
CO


LO
o
"x
£ o
0 0
r— 4- r—

-------
TABLE 44.  STEADY-STATE TITERS OF F2 BACTERIOPHA6E CONTROLS IN FIRST
           VIRUS RUN. '


Phage Titers
, pfu

Detention Time, sec
(Flow rate, I/sec)
Sample
11.4 16.8
(4.7) (3.2)
28
(1.9)
85
(0.6)
Control 2-Before UV Unit
30 sec
60 sec                   2.0xl02
90 sec                  .  -
120 sec
Geometric Mean           2.0xl02
Control 3-After UV Unit
30 sec
60 sec                   2.0xl02
90 sec
120 sec
1	•	  '                    -... •"-  f
Geometric Mean           2.0x10^
Control 1-Seed Culture
Start  6.2xl06
End    S.OxlO6
4.0x10'
2.4x10'
2.6x10^
1.4x10
6.0x10'
7.2x10];
9.0x10^
      3
1.6x10
      3
2.8x10-
2.3xl03
2.5x10-
5.8x10
3.0x10-
1.4x10-
1.4x10-
                           2.0x10^
2.0x10
                  3
                                    210

-------
TABLE 45. TITERS OF F2 BACTERIOPHA6E EXPOSED TO ULTRAVIOLET RADIATION AT
          DIFFERENT FLOW RATES IN SECOND VIRUS RUN.

Detention Time
(sec)
Flow (I/sec)
INDIGENOUS PHAGE
60 sec
120 sec
180 sec
Geometric Mean
Before UV Unit
60 sec
T.2Q sec
180 sec
Geometric Mean
After UV Unit
60 sec
120 sec
180 sec
Geometric Mean
LOG REDUCTION

Phage Titer,
pfu/ml

Detention Time, sec
(Flow rate, I/sec)
1 1 .4
(4.7)

6.2X101
3.6X101
l.SxlO1
i
3.4x10'

g.oxio2
l.OxlO3
1 . 2x1 03
l.OxlO3

6.0x10°
l.OxlO1
8.0x10®
7.8xl(#
2.13
16.8
(3.2)

4.0x10°
2.0x10°
2.0x10°
n
2.5xlOU

5.0xl02
e.oxio2
2.0xl02
3.9xl02

2.0x10°
l.QxlO1
4.2X101
1.2X101
1.52
28
(1.9)

1.2xlOV
l.OxlO1
6.0x10°
n
9.0xlOu

1.2xl03
S.OxlO2
1 . 8x1 O3
1.2xl03

4.0x10°
1,0x10°
l.SxlO1
* 3.7x10°
2.51
85
(0.6)

8.0x10°, ;
l.OxlO1
l.OxlO1
n
9.3xlOu

1.2xl03
9.8xl02
3.2xl02
7.2xl02

8.0x10°
8.0x10°
l.OxlO1
8.6x10°
1.93
*Sampling Times for the 0.6 I/sec flow were 3 minutes, 4 minutes and
 5 minutes.
                                    211

-------
TABLE 46.  TITERS OF POLIOVIRUS TYPE I EXPOSED TO UV RADIATION AT DIFFERENT
           FLOW RATES IN SECOND VIRUS RUN.
Virus T.iters, pfu/ml
Detention Time, sec
(Flow rate, I/sec)
Sampl e
Control 2-Before UV
30 sec
60 sec
90 sec
Geometric Mean
Control 3-After UV
30 sec
60 sec
90 sec
Geometric Mean
Before UV Unit**
0 sec
60 sec
120 sec
Geometric Mean
After UV Unit**
0 sec
60 sec
120 sec
Geometric Mean
LOG REDUCTION
*The sampling times
11.4
(4.7)

5.9xl03
4.4xl()3
4.6xl03
4.9xl03

4.9xl03
5.2xl03
4.7xl03
4.9xl03

5.3xl03
S.OxlO3
4.8X103
S.OxlO3

7-OxlO1
1 .2xl02
l.SxlO2
l.OxlO2

for C-2 and C-3 were
16.8
(3.2)

6.7xl03
6-OxlO3
6.9xl03
6.5xl03

3.6xl03
6.5xl03
6.8xl03
5.4xl03

5.4xl03
4.0xl03
3.6xl03
4.3xl03

2.5X101
2.4X101
2.6X101
2.5X101
2.23
1 , 2, and 3
28
(1.9)

8.8X103
l.lxlO4
7.6xl03
8.3xl03

l.lxlO3
5.8xl03
7-lxlO3
3.6xl03

l.OxlO4
l.lxlO4
6.5xl03
9.0xl03

1.6X101
1.2X101
1.4X101
1.4X101
2.81
minutes for
85
(0.6*)

2.6xl04
2.3xl04
2-OxlO4
2.3xl04

l.SxlO4
1.9xl04
1.7xl04
1.6xl04

l.SxiO4
1.4X104
l.SxlO4
1.4xl04

S.OxlO1
1.7X101
7.0x10°
l.SxlO1
2.97
this flow
  rate.
** U.V.  lights  had  been on for  two minutes.
Other results:
     All  samples from  the  C12  contact  basin were 0.
     All  samples for indigenous  virus  were 0.
     Control  1  (virus  feed):   start =  4.0 x 107 PFU/ml
                              end   =  2.95, x  107 PFU/ml

                                   212

-------
may have been slightly less responsive to UV than total and fecal coliforms
during this run.  A complete summary of the second virus run is given in
Table 47.

     A summary of the dose-related data is given in Table 48.  Due to the
much lower intensity readings, much lower indicated doses, Dj, were observed,
Contrary to the first virus run the theoretical dosages were a good approx-
imation of the indicated dosages.


THIRD UV VIRUS RUN

     The third programmed UV virus run with the UPS Ultraviolet Disinfection
Unit was performed on the morning of 26 June 1975.    Preparations for the
run began on 24 June when the unit was taken out of service and cleaned with
UPS cleaning solution.  After draining the unit, fresh cleaning solution was
added and allowed to recirculate through the contactor all night by means of
a small externally-mounted booster pump.  Early on the morning of 25 June,
the cleaning solution was flushed with tap water, and the radiometer read-
ing was 760 ywatt/cm2.  However, the reading fell to 20 ywatts/cm2 follow-
ing introduction of activated sludge effluent to the unit.

     It was originally planned to conduct the virus run on the 25th;
however, in view of the UV intensity on that morning it was decided to
postpone the run until the next day, and the cleaning sequence was re-
peated.  Although the poliovirus stock culture had been thawed by the time
this decision was made,  the F2 phage had not been added.  The virus solu-
tion was,stored overnight at 5°C, and F2 phage were added to it on the
morning of 26 June.

     Performance of the demonstration plant was monitored very closely prior
to the run.  Grab samples of activated sludge effluent were taken at 11 AM,
1:30 PM, and 3:35 PM on 25 June for COD analyses.  The results were: 122,
124, and 129 mg/1 COD, respectively.  Average DO in the aeration basin was •
1.2 mg/1 in spite of efforts to increase it.  On 26 June, the day of the
run, the same average DO was measured.  MLVSS was 2170 mg/1.  SVI was 306 ..
mg/1.  The F/M ratio was 0.34 kg COD/day-kg MLSS.  The aeration rate was
0.48 m3/m3. — more than adequate for the high rate operating mode with good
oxygen transfer equipment.

     On the morning of 26 June the virus run commenced, in spite of the fact
that the initial radiometer reading on activated sludge effluent was only
30 ywatt/cm2.  The process configuration and operating protocol were identi-
cal to the second virus run.  Start-finish meter readings at each of the
four flow rates are listed on Table 49.  The lowest UV intensity measured
by the IL 500 radiometer was 26 ywatt/cm2.  The highest was 29 ywatt/cm2.

     Analytical results of the chemical analyses are presented in Table 50.
It is clear that the influent wastewater quality for the  third virus run had
substantially deteriorated.  Although the level of TSS was similar to the
first two^virus runs, organic matter in the form of COD,  TOC, and BOD was
                                     213

-------
TABLE 47.  SUMMARY OF RESULTS FROM SECOND VIRUS RUN

MEAN LOG REDUCTIONS
Organism
Total Coli forms
Fecal Coli forms
F2 Bacteriophage
Poliovirus Type 1
Detention Time, sec
(Flow rate, I/sec)
11.4
, (4.7)
2.19
2.33
2.13
1.69
16.8
(3.2)
3.85
3.06
1.52
2.23
28
(1.9)
3.54
3.46 '
2.51
2.81
85
(0.6)
4.26
3.67
1.93
2.97
                                     214

-------
TABLE 48.  SUMMARY OF DOSE RELATED DATA FOR THE SECOND VIRUS RUN

Flow rate, I/sec
Theoretical Detention
Time, sec
Actual Detention
Time, sec
Transmittance @
254 nm, percent
Extinction Coefficient,
cnr'
2
UV Intensity ,,uwatt/cm
Calculated Dose, DT,
ywatt-sec/cnr
Indicated Dose, Dj,
viwatt-sec/cm2
4.7 '.

11.4

10

62.5
0.48
100
8,500
8,900
3.2

16.8

15.5

62.5
0.48
97
12,500
13,400
1.9

28.2

25

62.5
0.48
95
20,900
21,100
0.6

85.1

75

62.5
0.48
97
63,100
64,700
                                    215

-------
TABLE 49.  SUMMARY OF OPERATIONS OF VIRUS RUN NO. 3
Flow (I/sec) 4.7
WQ Meter (start) 0
WQ Meter (finish)
Radiometer (start),
yw/cm2 26
Radiometer (finish),
uw/cnr
3.2 1.9
0 0
0 0
29 29
28.5 27
0.6
0
0
29
26.5
                                     216

-------
TABLE 50.  RESULTS OF CHEMICAL ANALYSES OF GRAB SAMPLES AVERAGED OVER
           ALL FOUR FLOW RATES IN THE THIRD VIRUS RUN.

Parameter Influent to UV
COD, mg/1
TOC, mg/1
SOC, mg/1
BOD, mg/1
TSS, mg/1
Turbidity, NTU
Color,Pt-Co Units (filtered)
NH3-N, mg/1
Org.-N, mg/1
N02-N + N03-N, mg/1
N02-N, mg/1
PH
Sp. Cond.
Total Alk. , mg/1
TDS, mg/1
Cl~, mg/1
%Transmittance at 254 nm
75.2
28
11
22
22
11
30
10.3
6.2
0.3
0.05
.7.6
900
182
436
70
51.6
Effluent from UV
77.0
30
12
15
23
11
30..
-
-
"
.
7.5
. 900
184
- - • - • •
'
51.6
                                    217

-------
almost double previous levels.  NHs-N and organic-N were markedly higher,
and UV transmittance was significantly lower.  Turbidity was three times
higher than previous runs.

     Results of the microbiological samples are given in Table 51. . Influent
MPN's were 1 to 2 logs higher than other virus runs, again reflecting the
generally deteriorated wastewater quality conditions.  Log coliform reduc-
tion were substantially lower and more erratic.

     Results of the control sampling for F2 phage are shown in Table 52.
The influent controls in the 4.7, 1.9, and 0.6 I/sec, runs were close to the
expected values.

     The respective effluent controls, however, were significantly lower.
The only explanation which can be offered is the generally poorer wastewater
quality, which may have adversely affected phage survival and recovery.
Results from exposing the seeded phage to ultraviolet disinfection are given
in Table 53.  It is evident that the titers of indigenous phage were not
insignificant relative to the seeded numbers.  Thus when evaluating the log
reduction values, it is difficult to ascertain which phage population con-
tributed more to the effect.  In general, the log reductions were lower and
more erratic than in previous runs.

     Table 54 presents the results of the poliovirus analyses of the third
virus run.  As can be seen the effluent counts were higher than the previous
two runs.  This was clearly due to the substantially reduced UV dosage as
a result of the marked attenuation of the UV energy by the poorer quality
wastewater.  Examination of the results of the control samples before and
after the UV unit indicates that steady-state conditions were readily
achieved.  A complete summary of the third virus run is given in Table 55.

     Table 56 summarizes the dose-related data for the third virus run.
The dosages obtained in this run were much lower than the other two runs.
This is also evident in the disinfection data, since the worst disinfection
occurred during the third virus run.

     Figures 115-118 summarize the microorganism reduction data from the
three virus runs.  The runs were made during a period of equipment diffi-
culties when effluent COD values ranged from 37 to 75 mg/1.  The highest
COD waters gave the poorest microbiological results, but little difference
was observed between phages and poliovirus.  The phages were possibly a
little more resistant than the poliovirus, and the fecal coliforms were a
little more resistant than both the viruses, but the equations are really
not that different particularly in view of the low correlation coefficients
(0.70-0.72).  Figures 116 and 117 show the individual data points and
resultant curve for fecal coliforms and coliphages, respectively, and
Figure 118 compares the curves for both these organisms with that of polio.
The no-effect dose (determined by settling y = o and solving for x) ranges
from 148 u watt-sec/cm2 for fecal coliforms to 871 y watt-sec/cm2 for
coliphages.
                                    218

-------
oo

D:
a
D;
O
D;
s
J-H
cs
CD
a:
UJ
o
•=c
oa
D;
oo
ca























o



CO
O
"x

•
CM
A
in
o
r—
X-
^j-
•
CM
A

10
O
"x
LO
CO
LO
0
"x

•
CM
A


r--.
o
'x
«*•
LO
(/)
0
q-
•i— p— .
r-— £
O
o o
o
r— r-~
fO
•(-> S-
0 OJ
I— Q.







O
o
*
CO
V








CM

co














<£>
•
V
'








LO
CO
CM
V



O
1 — 1
I—
0
•=}
Q
LU
o;

o
o


































































^j-
0
"x
1 —

p—



VO
0
*x
en
^!

o
"x
r—
,_!


r>.
0
^
r--
•
P_

«.,.
o
P«
X
CM

ai


vo
o
"x
r-.
r~
•*0
"x
r--.

P_«



^
o
"x
CO
*~
o
q-
•r- 0
•— E
O
0 0
0

to
0 S-
Qi 
-------
TABLE 52 .   STEADY-STATE TITERS OF F, BACTERIOPHAGE CONTROLS IN THIRD
            VIRUS RUN.



Phage Titers
,, pfu/ml

Detention Time, sec
(Flow rate, I/sec)
Sample
Control 2-Before
30 sec
60 sec
90 sec
Geometric Mean
Control 3-After
30 sec
~-*fc
60 sec
90 sec
11.2
(4.7)
UV Un.it
l.SxlO3
l.OxlO3
l.OxlO3
1.2xl03
UV Unit
1.6xl02
2.2xl02
2.2xl02
16.8
(3.4)
6.5xl02
6.5xl02
3.7xl02
5.4xl02
3.2xl02
1.6xl02
l.SxlO2
28
(1.9)
3.3xl03
4.6xl03
4.1xl03
4.0xT03
,6.0xl02
**
4.0xl02
84
(0.6*)
8.3xl04
8.5xl04
l.OxlO4
4.1xl04
2,2xl03
2.8xl03
l.SxlO3
Geometric Mean
             2.0x10"
2.1x10
                                          2
4.9x10
                                               2
2.2x10^
Control 1-SEED CULTURE

Start   4.6xl06
End
2.6xlOl
*Sampling on the 0.6 I/sec flow was performed at 1 minute, 2 minutes,  and
 3 minutes.                                                   ,

** Data not available.
                                     220

-------
 TABLE  53.  TITERS  OF  F?  BACTERIOPHAGE  EXPOSED TO  ULTRAVIOLET RADIATION
            AT DIFFERENT  FLOW  RATES  IN  THIRD VIRUS RUN.
Phage Titers, pfu/ml
Detention Time, sec
(Flow Rates, I/sec)

Samp! e
INDIGENOUS PHAGE
60 sec
120 sec
180 sec
Geometric Mean
Before UV Unit
60 sec
120 sec
180 sec
Geometric Mean
After UV Unit
60 sec
120 sec
180 sec
Geometric Mean
LOG REDUCTION
11.2
(4.7)

4.4xl02
3,.9xl02
2.1xl02
3.3xl02

1.4xl03
1.5xl03
4.6xl02
9.9xl02

7.4X101
T.2xl02
8.7X101
9.2X101
1.04
16.8
(3.2)

3.6xl02
1.4xl02
2.0xl02
2.2xl02

7,lxl02
1.4xl03
l.lxlO3
l.OxlO3

^.OxlO1
6.0xlO]
2-OxlO1
<2.3xlO]
>1.64
28
0-9)
i .
1.3xl02
1.9xl02
S.OxlO1
l.lxlO2

2.0xl02
S.OxlO2
6.0xl02
3.3xl02

l.OxlO1
2.0X101
4.0x10° .
9.3x10°
1.55
84
(0.6*)
-
, 2.2xl02
2.4xl02
4.6xl02
2.9xl02

S.OxlO3
2.1xl03
2.5xl03
2.5xl03

S.OxlO1
l.SxlO1
l.SxlO1
S.OxlO1
1.92
*Sampling for the 0.6 I/sec flow was performed at 3 min.  4 m1n,  and
 5 min.
                                 221

-------
TABLE 54.   TITERS OF POLIOVIRUS TYPE  I EXPOSED TO UV RADIATION AT
           DIFFERENT FLOW RATES IN THIRD VIRUS RUN.
                              Virus Titers, pfu/ml
                               Detention Time, sec
                                (Flow Rate, I/sec)
Sampl e
11.2
(4.7)
16.8
. (3.2)
28
(1.9)
84
(0.6*)
Control 2-Before UV Unit
6.7X10'

8.5x10'
30 sec

60 sec

90 sec

Geometric Mean       7.2xlOv

Control 3-After UV Unit
5.2xlO
y.exio
7.9xio

6.8xlOc
1.5x10^

1.6xl04

1.6xl04

1.6xl04
5.9x10

6.2x10^

8.0x10^

6.6x10^
                                                                   ,4
30 sec
60 sec
90 sec
Geometric Mean
Before UV**
0 sec
60 sec

120 sec
Geometric Mean
After UV**
0 sec
60 sec
120 sec
Geometric Mean
Log Reduction
6.8x10°
7.0xl03
4.3xl03
5.9xl03

6.1xl03
7.2xl03
o
5.3x10°
6.2xl03

4.2xl02
S.lxlO2
7.5xl02
5.4xl02
1.06
7.3x10°
Q.lxlO3
7.2xl03
7.8xl03

9.6xl03
6.9xl03
o
7.5x10°
7.9xl03

l.SxlO2
l.SxlO2
2.2xl02
1.5xl02
1.72
1.6x10^
1.4xl04
1.7xl04
1.5xl04

1.5xl04
1.9xl04
n
1.5x10*
. 1.6xl04

1.5xl02
1.4xl02
1.7xl02
1.5xl02
2.02
2.5x10^
3.6xl04
5.3X104
3.6xl04

7.1xl04
4.3xl04
4
6.1x10*
5.7xl04

S.SxlO2
l.SxlO2
l.SxlO2
l.SxlO2
2.50
"*The sampling Time for 0.6  I/sec  flow was  1

 **The UV lights had been  on for 2 minutes.
 Other Results:
 All  samples from the C12  contact  basin were
 All  samples for indigenous  virus  were <1.  7
 Control  1  (virus feed);  Start   7.55 x 107
                                  8.45 x 10X
                       min, 2 min, and 3 min.
                        PFU/ml
                        PFU/ml
                                 222

-------
TABLE 55.  SUMMARY OF RESULTS FROM THIRD VIRUS RUN.
                                    MEAN LOG REDUCTIONS
Detention Time, sec.
(Flow Rate, I/sec.)
Organism
Total Col i forms
Fecal Col i forms
F2 Bacteriophage
Poliovirus Type 1
11.4
(4.7)
<2.35
2.88
1.04
1.06
16.8
(3.2)
<1.16
1.27
Si. 64
1.72
28
(1.9)
3.42
3.19
1.55
2.02
85
(1.9)
<3.00
2.65
1.92
2.50
                                     223

-------
TABLE 56.  SUMMARY OF DOSE-RELATED DATA FOR THE THIRD VIRUS RUN

Flow Rate, I/sec
Theoretical Detention
Time, sec
Actual Detention
Time, sec
Transmittance @
254 nm, percent
4.7
11.4
10
51.6
Extinction Coefficient,
cm-1 0.65
UV Intensity,
ywatt/cm
Calculated Dose, DT
pwatt-sec/cnr '
Indicated Dose, Dj
vwatt-sec/cmz
26
4,300
2,300
3.2
16.8
15.5
51.6
0.65
29
6,300
4,000
1.9
28.2
25
51.6
0.65
28
10,600
6,200
0.6
85.1
75
51.6
0.65
28
32,000
18,700
                                     224

-------
                                    to
                                    •xj-

                                    CM
                                   LO


                                    'o
                                        LO
 X


to  °

•*";  u

CO  i.
                                                             O
                                                             10
                                                             O

                                                             LO
                                                               ro
                                                                o
                                                            o
                                                            ro
                                                            o
                             o
                                                                 o
                                                                 OJ
-p


                                                                        o
                                                                        ta
                                                                        u
                                                                        a>
                                    q-
                                    o
                                                                        c
                                                                        o
                                    o
                                    c:
                                                                        res
                                    in
                                    to
                                                                        O
                                                                        3
                                                                        -a
                                                                        01
                                                                        o
                                    o
                                    o
                                                                        (B
                                                                        4J
                                                                        o
                                                                        3
                                                                        01
LO
                        co
                  DOT wyojuoo ivioi
                             225

-------
o
UJ
or

CD
O
O
O
•a:
O
     Figure 116.
          LOG UV DOSE

Fecal coliform reduction  vs.  UV dose
                               226

-------
     5
O
I—1
h-
O
LU
CC
CJ3
O
LU
C_J
=3;
CQ
                              LOG UV DOSE
       Figure 117.  Coliphage reduction vs.  UV dose

                                227

-------
     4 ••
O
l-H

O


LU
CO
O
     3 •-
     2 ..
                               LOG UV DOSE


       Figure 118.  Comparison ,of dose-response  for three organisms

                               228

-------
  10
 lo3
 t-
 Q_
 co
 LU
 co
'UJ

CO
  10'
\
                                                 RUN No. 1  22 April 1975


                                                 RUN No. 2  13 May 1975


                                                 RUN No. 3  26 June 19775
AVERAGE PERCENT TRANSMITTANCE =51.6
                           AVERAGE PERCENT TRANSMITTANCE =62.5
                              AVERAGE PERCENT TRANSMITTANCE = 66.9
             10       20       30      40      50        60       70  -    80


                              DETENTION TIME IN  UPS  UNIT, sec





     Figure  119.   Virus  reduction versus  detention time.
                                     229

-------
                                  SECTION 9

                        SUMMARY OF OPERATING EXPERIENCE

     During the sixteen months in which the ultraviolet disinfection project
was conducted an extensive amount of data was collected.  It would be quite
impossible to present the,results of every analytical test performed during
the project in this report.  In this section the authors would like to
summarize and review some of the more significant observations made during
this research effort.

     Previous data have shown that UV dose is the controlling factor in
microorganism kill, and that the percent transmittance at 254 nm is very
important to the actual dose received by the microorganisms in the exposure
chamber (13, 14).  At the start of this project, we expected UV transmission
to be dependent on the turbidity of the water.  The data presented in Figure
120, which is a plot of turbidity versus percent transmittance, indicate
lack of correlation.

     There would similarly seem to be a potential relationship between
suspended solids and transmittance at 254 nm.  Figure 121 presents these
data (averages for each of the UPS runs), and correlation is clearly lacking,
since a correlation coefficient of only 0.16 resulted.  Additional efforts
to improve on these data by plotting daily values, instead of run averages,
proved fruitless.  There does not appear to be any correlation between TSS
and transmittance at 254 nm, at least within the ranges observed on the
project.

     Attempts were made to correlate the observed transmittance values,
which were available only after the start of the UPS unit, and several
additional water quality parameters.  There is no correlation evident
between either total phosphorus or combined nitrite-nitrate nitrogen and
transmittance.

     Figure 122 is a plot of percent transmittance versus the COD of the
water being processed by the UPS disinfection unit.  The correlation
coefficient obtained was - 0.76, which is good for an experiment of this
type.  Transmittance is plotted as a function of total organic carbon in
Figure 123.  The correlation coefficient is - 0.95, which indicates very
good correlation.  The fact that COD and TOC correlate well with the observ-
ed transmissibility at 254 nm indicates that the amount of soluble organic
material present in a wastewater will affect the efficiency of an ultraviolet
light disinfection process used to treat that wastewater.
                                     230

-------
    10
CQ
Cf.
                        25                50


                            TRANSMITTANCE, percent
75
90
      Figure  120.   The effect of turbidity on percent transmittance.
                                    231

-------
 0)
 o

 0)
 n.
LU
o


1
CO
z
«=C
    72
    70
    68
    66
62
    60
    58
    56
    54
    52
     50
        0
                              Y = 0.07 X+ 64.64

                                    r = 0.16
         10
        Figure  121.
20      30     40     50


 TOTAL SUSPENDED SOLIDS, mg/1

  Correlation between percent  Transmittance and
  Total Suspended  Solids.
                                232

-------
-p

CD
O

OJ
Q.
CVJ


•a:

UJ
s:
oo
    100
     90
80
     70
     60
     50
    40
    30
                10
T = -0.31 COD + 75


    r = 0.76
                     20        30


                       COD, mg/1,
                    40
50
 Figure 122.  Correlation between transmittance at 254  nm  and COD.
                                 233

-------
  o
  o
  s_
  O)
  Q.
  CM
  UJ
  o
      100
       90
       80
       70
60
       50
       40
       30
               i     I
                          T = - 1.50 TOC + 81


                              r = 0.95
                       I    I     1	I     I
          0    2    4    6   8   10   12   14   16  18


                               TOC, mg/1




Figure 123.  Correlation between transmittance at 254 nm and TOC.
                              234

-------
     The amount of organic material present in the effluent of a biological
treatment process is related to the degree of nitrification occurring;
therefore, a well-nitrified effluent should be more amenable to disinfection
with ultraviolet light than a non-nitrified effluent. In Figure 124 ammonia-
nitrogen concentrations are plotted as a function of both COD and TOC.   The
correlation coefficient for the NH3-N versus TOC plot is 0.88, while the
NH3-N versus COD graph yields a correlation coefficient of 0.97.  Both
correlations are excellent, and indicate that reduced concentrations of
organic compounds can be achieved by utilizing a nitrifying activated sludge
system.

     Since both COD and TOC correlated well with transmittance, a correlation
between transmittance and the concentration of unoxidized nitrogen compounds
was anticipated.  Figure 125 is a plot of transmittance as a function of
ammonia-N concentration.  The correlation is quite good (r = -0.81), as is
the correlation between organic nitrogen and transmittance ( r = -0.77) shown
in Figure 126.

     The quartz sleeves housing the UV lamps in the UPS unit were subject to
a certain amount of slime build-up since they were continuously submerged in
a secondary effluent.  Slime accumulation will decrease the amount of UV
energy radiated to the surrounding water resulting in a decrease in the
unit's disinfection efficiency.  Figure 127 is a time series plot of the UV
dose calculated from intensity readings.  There is clearly a decrease in the
UV dose, but the initial rate of decrease observed in early October is  quite
gradual.  The rate of decrease observed in early November was considerably
greater than the October rate.  These data indicate that the sleeves remained
clean for several weeks before significant slime development occurs but that
the slime will accumulate rapidly once started.

     The UPS unit was cleaned with a proprietary compound provided by
Ultraviolet Purification Systems, Inc. in mid-November, and the UV dose
increased to the early October values.  The cleaning frequency required to
keep the system operating at peak efficiency can be expected to vary with the
quality of the effluent, but intervals of about two weeks seem  reasonably
consistent with these data and temperatures.

      At the start of this project the factors expected to be responsible
for the number of organisms found in the effluent from a UV disinfection
system  were total suspended solids, turbidity, transmissability at 254 nm,
and UV dose, among others.  Figure 128 is a graph of effluent total suspended
solids versus effluent fecal coliforms.  No readily apparent correlation
exists between these two data sets, or the correspondence is only poorly
defined.  Attempts to quantify the suspected correlation between TSS and
coliform densities were unsuccessful.  Plots using data at a constant dose
for any given run exhibited considerable scatter.

     Figure 129 is a graph of effluent fecal coliforms versus turbidity, and
no significant interdependence is evident.  The fact that neither turbidity
nor suspended solids have any demonstrated significant impact on the
observed effluent fecal coliform counts is a significant finding, expecially
                                     235

-------
      18
      16
      14
      12
      10
       8
       6
       4
       2
             NH3-N = 0.27 COD - 3.8
                     r = 0.97
0  10
   20
   18
   16
   14
,_  12
I* 10
T   8
 oo
i   6
    4
    2
                   20
                        30    40
                        COD, mg/1
50
60
                  j-N = 0.96 TOC -  5.3
                        r = 0.88
                       T   t  .    .   .    .
Figure 124.
           .4      8      12     16     20
                       TOC, mg/1
          Correlations between ammonia-N and  COD  and TOC.
                            236

-------
 o
O)
Q.
LU
O
                          %T =  (1.18)(NH0-N)  + 70.83
         Figure 125.
                              6      8       10

                               NH3-N, mg/1
Correlation,between NhL-N and percent
transmittance at 254 nm.
                                237

-------
C
QJ
U


-------
                                                         O)
                                                         t/1
                                                         O
                                                         -a
                                                         c:
                                                         o
                                                         C/5
                                                         O)
                                                         OJ
                                                         O)
                                                         S-

                                                         (O


                                                         cr


                                                         


                                                         CD


                                                         •r—


                                                         (13
                                                         q-
                                                         o

                                                         4-5
                                                         o
                                                         
-------
  10C
   10'
o
o
i.
O)
CL
_
8 10
o
   10
   10'
            ••
            ..*<
                20
                                                             120
                         40       60       80      100
                             TOTAL SUSPENDED SOLIDS
Figure 128. Effluent suspended solids versus effluent fecal coliforms.
140
                                    240

-------
  10C
  10'
810
a;.
a.
o
      0
         2468

                         TURBIDITY

Figure 129.   Turbidity versus fecal coliforms.
10
12
14
                                  241

-------
since the result is contrary to what one intuitively would expect.   One must
note that the range of turbidity arid suspended solids values observed
during this project was not great; therefore, this finding cannot be extra-
polated to very turbid water, or waters with high suspended sblids  concen-
trations with any degree of confidence.  However, the range of TSS  concentra-
tions and turbidity values encountered during the project are likely to be
typical of municipal secondary treatment plants.  And over this range,
turbidity and TSS are not particularly important water quality parameters
with respect to UV disinfection.

     The only practical method that can be used at this time to routinely
monitor UV dose is the direct measurement of UV intensity at some point on
the wall of the exposure chamber.  Two factors should be primarily  respon-
sible for changes observed in the measured UV intensity - the light trans-
missability of the water and slime accumulation on the quartz sleeves.
Figure 130 is a plot of measured UV intensity versus percent transmittance.
There is no apparent correlation.  The graph includes data for ten  months,
and better results were expected.  At this time no reason for the lack  of
correlation can be offered except for slime growth on the quartz sleeves.

     The routine measurement of UV intensity does constitute an excellent
process control procedure.  Measured UV intensities and observed log reduc-
tions in fecal and total coliforms are shown in Figure 131 and 132, respec-
tively.  The data plotted are from the ten month period starting on February
1, 1975 and ending November 30, 1975.  Visual inspection indicates  good
correlation.

     The number of organisms present in an effluent disinfected with UV light
is a function of not only the applied UV dose, but also the number  of
organisms present in the influent.  Data"plotted for the individual runs
illustrate this point clearly.  In particular Figures 99 and 100 show the
change in coliform densities in the influent and effluent observed  during
Run U7.

     The singularly most important factor in determining the number of
coliform organisms to be found in a UV disinfected effluent is the  UV dose.
Figure 133 presents a plot of effluent fecal coliforms per 100 ml versus
UV dose, and the correlation is good (r = -0.61).  These data are from  Run
No. U8, and indicate that if an effluent fecal coliform count of 200 per
100 ml is desired a UV dose of at least 15,500 uwatt - sec per sq.  cm.  is
required.

     During the discussion of individual UPS runs two dose  values  have been
reported, the calculated dose, Dy, and the indicated dose, Dj.  A least-
squares curve fit of the average dose values for each run resulted  in the
following estimating equation,

          Dj = 1.27XDT - 5670

and the resulting correlation coefficient was 0.83.
                                     242

-------
     10
       "2
     ID
       '3
CM
 CJ
 -M
 IO
 P V5
     io
       -6
     10
       -7
                                                     »
                                              •  « •  **
                                               «•••••
                                                          •••*
         20      30        40       50       60,       70

                                 TRANSMITTANCE,  percent

         Figure 130.  UV intensity versus percent transmittance.
80
90
                                    243

-------
     10
       -1,
     10
       -2
     10
       -3
CM
 01
  (t)
 LU
     10
       -5
     10
       -6
                                  • *
«  •   «.
  ••   •

7<5  ^:
t*.A • .*..
                                       •.•%  ;-
                                       * •• * «* * *
                                       : Jt • 12'
                                       »«^
                                      k-r"
         0        1        23         4         56        7

                            LOG REDUCTION TOTAL  COLIFORMS

         Figure 131. Observed UV intensities versus  log reduction in total

                     coliforms.
                                     244

-------
ID-'
, -2
10
in'3
10
j
o
i/)
S
n
>-
1 — I
t/)
z: _A
1±110 H
•z.
i— r
:==>


•m-5
10
in-6










•.






.
.
• ..:
• ••
• •
4 • •
•



.
•
* * • * * *
»ff« ••* •
t . .'*.,
•'.•.•••'"• I

*




' """'.**•••

T'r"J •}•
• • t • ( •«•
• • .
• • .• •.
•-• • ••!•







'I
t * •
.*•
• .
• •
•





.
•
.
















         0        12        3         45         6          7
                               LOG REDUCTION FECAL  COLIFORMS

Figure 132.  Observed UV intensities versus log reductions  in  fecal  coliforms,
                                    245

-------

"e
° 10
o
s-





,







^^

• • •



•


•
.
• • •
^tf • •
\
• ^^.



•
•
*



. . ^

«• •

\



*
LOG y =






S.
\
\
\
•
-1.21x10"
r = -0.6








*
9
4x+4.19
1










5000 1000 ' 12000 14000 16000 18000 20000
Figure 133.  Effluent fecal coliforms as a function of (N dose.
                          246

-------
     Figure 134 presents two graphs, one of the calculated dose, DT, versus
log reductions in fecal coliforms, the other of indicated dose, Dj, versus
log reduction in fecal coliforms.  The values plotted are means for each
of the UPS runs, and the estimating equations and correlation coefficients
are shown with the appropriate graph.,  Very good correlation was observed
(r = 0.83 for DT and 0.86 for DT) between dose and observed log reduction
value.

     The same data analyses are presented for the total coliform and dose
data in Figure 135.  Estimating equations and correlation coefficients
are included on the figure.  Excellent correlation was observed, especially
between the indicated dose, Dj, and the observed log reductions in total
coliforms which had a correlation coefficient of 0.96.  Both total and fecal
coliform log reductions correlated better with the indicated dose than the
calculated dose.  This is to be expected since the indicated dose is a
better indicator of the amount of UV energy actually reaching the organisms.

     Figure 136 is a plot of the log-jn of the indicated dose, Dj, versus the
log of the coliform density whose probability of exceedence is greater than
or equal to five percent.  The fecal coliform densities were obtained from
the extreme value frequency distributions for each of the UPS runs, and the
dose values are means for each run.  The correlation coefficient is -0.96,
which is excellent.  The curve indicates that a fecal coliform density of
200/100 ml can be expected to be exceeded five percent of the time if the
indicated dose is 24,400 pwatt-sec/cm2.

     Similar analyses with exceedence probabilities of one percent and ten
percent using a semi-log form resulted in correlation coefficients of
-0.88 and -0.90, respectively.

     Table 57 and 58 summarize the data for the Kelly-Purdy unit and the
UPS unit respectively.  Even a cursory scan of the log-reduction columns
will indicate that the UPS unit exhibited vastly superior performance.
This should be expected since the UPS unit is a modern design, and several
factors contributed to the Kelly-Purdy unit's inadequacy.

     The major operational difficulty encountered with the old Kelly-Purdy
unit was sludge accumulation in the bottom of the exposure chamber.  The
chamber functioned as a "shallow-tray" sedimentation basin, and the rate
of sludge accumulation was appreciable.  As a result of the system
geometry, it was possible to take advantage of radiation from only 180
degrees of the lamps in the Kelly-Purdy unit, while emission from the full
360 degrees could be used in the UPS unit, making the UPS configuration
more effective.
                                    247

-------
 X
      IO
      03
 O   CD
 O
CO
«*
CM


 II
      II


      S-
                                                          oo

                                                          Q;   -o
                                                          O   CU
                                                      oo
                                                          o
                                                          o
                                      O

                                      U_
                                                              =3
                                                              o
                                                              O

                                                              Ol

                                                             •4-5
                                                          O   -O
                                                          i—i   E
                                                               td
                                                      CM  o
                                                          CI5   T-
                                                      r—  O   i—
                                                          _1   O
                                                               O
0
o
o
0
LO
0
• o
o
CD
«^1*
0
0
o
o
oo
o
o
0
CD
CM
O
O
0
O
r—
O




                                                      O
                       '3SOQ  C!31\/3IC1NI
                                                              rd
                                                              O
                                                              OJ
                                                             M-r
                                                               t/)
                                                               E
                                                               O
                                                               O
                                                               3
                                                               -a
                                                      oo
                                                               CD
                                                               o
                                                               OJ
                                                               >

                                                               aj
                                                               to
                                                               J3
                                                               o
                                                             .£= to
                                                          cz   co o
                                                          LlJ   C. -r-
                                                          a:   o "a
                                                               •r- E
                                                          CD   +J •!-
                                                          o   
-------
                                                       LO
                                                            CO


                                                            C£
                                                            O
                                                            O
                                                            O
                                                       oo    O
                                                                > -o

                                                                0) >
                                                                to ^
                                                                -Q
                                                                o -a
                                                                OJ <1)
                                                                S to
                                                                •4-> -Q
                                                                
                                                                o res
                                                                •r" i—
                                                                4-> 3
                                                                ro O

                                                                O) "to
                                                                Di O
                                                                          LO
                                                                          OO
                                                                          a>
o
o
o
o
LO
O
o
o
o
o
o
o
o
OO
O
o
o
o
CM
o
o
CD
O
                               249

-------
                        LOG UV DOSE, D
                                      I
Figure 136.
Fecal coliform densities with an exceedence
probability greater than or equal to 5 percent
versus indicated dose.
                          250

-------













INFECTION UNIT.
CO
l»-4
o
s'
d£
S
1
>-
_J
_J
Ul
•*£.
Ul
re
I—
oc
£
1
S
LU
=c
1—
u.
o
S


CO
r-*
to
Ul
_J
CO

CO
O£ O
^£
K?|
AJ
=>
_1
LI- r-
z a — .
"Si1
co •»-*.
cog,

c
cn o
O 1-
—1 <<->
u
!U -0
£
O CU
IE

Conforms
jent
r-, t_«
to *f-
tu *"*
^E
«0 "x.
J U
•> 0) c/)
1 1
So

4-> O
11"
4- -M
°g
Q.r—
1— C
O
C
3
oc

t*+ cn *t to
CO 'CO CO CM r-
i i ^ cn co
«a- CM
i i i r— cn
f-.
co to • cn
r- r-* CM to in

tn o cn o o
CM cn o ^- o
CM CM CO CO CO

CO CM CM r- CO
o o o o o
X X X X X
CM CM CO CM r-
V
in in m •& to
o o o o o

«3- CO CO CM to
"d" CM CO IO t-~
o o o o
0 O O O
CO 1 tO CD CO
CO -<3- CM CO
CO CM CO i^»
CO
1
tn in -cj- in  -0
uj uj uj to uj §» la

	 f— 4->
«C «i  *"
o ^













ti
5 DISINFECTION
a.
UJ
UJ
h-
U.
O
§
£
CO
00
in
Ul
_j
CO




















1 r—
HH Z E
c o cn
4- O -^
c o cn
^.o s
i
u
3 C
O O) •r-
-ja: -M
C
, 0)
i E
O UI
4-
"o W
0 CU
O S- 4-
U. Q-t-t
~e
*o o
•M O WJ
S -J.
•f- O O to
r— 1
*-" l/> tO
(002
CJ Q 3-
S 0) 0
•f—  «
8
(O 4->
*s
i "
|&
*-«
o a>

o
c
3
oc
CD
CM CO r— CO i— -O^-
cMi— r-or^f— -^-in
r— f- CM CM
CO 1O CO Cn r— r— r— f—
V
CM r-* cn co
r*.^-CMCOCMp-^-^-
r-cMcocn^-incor-.
r— CM r- i— *d- UO CM CO
^-^»-cMin^:cnrx.^-
CMOCMr— O VO in CO
^•^•^•COCOCMCOCM
A A A A A
OOOOCMCOCMCM.
OOOOOOOO
xxxxxxxx
COf— O CO CO O O CM
CMCOCM^J-CMCMr-'«3-
V V V V A ,- V V V
^^^^"o1^"©"^
XXXXXXXX
r*« ^
COCOCOCMCMr— r- *—
•M
cocMintntn^j-coco g
cn oo co r^. r— ^- r^» 10 -^
CMCJCOCMCMCMi— t— JT
**-
«>
4J *
' = -g>
§ ^
r*.cocncM^tf— r-cM !£w
Or-COCOCO^x^i— 'ft'?,
r*» r>* to to to in to to • "
S S, g
co -o .>
CM CO CM CM ^ *« iJ
• • • • g
<« *a r-
•M 
-------
                                 SECTION 10
                                                                         t

                              PHOTOREACTIVATION


     In October 1974, a photoreactivation study was performed using  clear
and amber glass sample bottles.  The results indicated that exposure of UV
effluent samples to 30 minutes of sunlight produced a 1.1  log rise in
total coliforms, and a 0.6 log rise in fecal coliforms.

     A study was conducted during August 1975 to see if similar observations
could be made on a pilot scale.  The plant's two chlorine  basins,  into
which the UV unit normally discharged, were set up to operate in parallel.  ••
A wooden frame covered with a black poly-vinyl chloride (PVC) sheet
was constructed and placed over the No, 2 chlorine basin to black out
sunlight.  The No. 1 basin remained open to the atmosphere.  In this manner,
UV effluent was split continuously to an open and a covered basin.  No
chlorine was fed during this period.  Figure 137 shows the arrangement  of
the Demonstration Plant during this special 20-day study,  which ran  from
1 to 19 August 1975.  Although this period overlapped two  UV runs, the
operation of the UV exposure chamber was not altered for this study.

     Table 59 outlines the major operating parameters of interest.  Half the
flow from the UV contactor entered the No. 1 (open) basin, while the other
half entered the No. 2 (covered) basin.  At flows of 1.5 I/sec, each basin
had a theoretical detention time of 65 minutes.  Previous  dye tests  indicat-
ed the actual time to be very close to theoretical.  It was assumed that
half of the irradiated effluent was subjected to 65 minutes of sunlight,
the other half to 65 minutes of darkness.

     Samples of the effluent from each basin were collected over the 20-day
study period in coordination with the normal UV sampling program.  In this
manner  UV effluent samples corresponded to the influent of both basins.
All light-dark samples were collected in dark glass sample bottles.   Samples
were taken only during daylight hours.  It became evident after several
days that algae were present in both basins, although much more so in the
open (light) basin than in the covered one.  The covered basin was not
truly dark as originally intended  because  a  small  amount  of  light seeped
around the edges of the cover.

     A chemical and bacteriological summary of  the light-dark sampling
appears on Table 60.  Bacteriological data are  reported as geometric means.
Substantial increases in MPN's occurred in both basins, although  the basin
exposed to sunlight showed a larger increase.   Total coliforms increased
0.7  log in the light basin, but only 0.3 log in the dark basin.  Likewise,
                                     252

-------
* -

J
D
T £=
•)-*
r— C
. si
0 O
<_)
1
= <— ^— ^*
I/i a;
ed Q.
CQ O
	 O
A v
•WH^LMM nl m
Hi
£ I-1
en
LO
r-^ OJ

CM C C S-
• S- tO >
O O 'O O
2: i— ca c_>
x: - —
0
.CM
                                                                               LO
                                                                               r-.
                                                                               CTl
                                                                                CT)
                                                                               I
                                                                               4J
                                                                                to
                                                                                S-
                                                                                eo
                                                                               -a

                                                                               +->
                                                                               ^:
                                                                                en
                                                                               I

                                                                                O
                                                                                res
                                                                                S-
                                                                                3
                                                                                cn.
                                                                                O
                                                                                O
                                                                                
-------
TABLE 59.  PROCESS CONTROL AND OPERATION SUMMARY LIGHT-DARK STUDY:  AUGUST
           1-19, 1975
          UV DISINFECTION UNIT
          Flow
          Theoretical detention time
          Water temperature
          UV intensity
          Water quality
          Transmittance (@2537 A°)
3.0 I/sec
18 sec.
28 degrees C
122 ywatts/cnr
Out of Service
63.3 percent
          NO. 1 CHLORINE BASIN
          Flow
          Theoretical detention time
          Conditions:
1.5 I/sec
65 minutes
Open to atmosphere
          NO. 2 CHLORINE BASIN
          Flow
          Theoretical detention time
          Condi ti ons:
1.5 I/sec
65 minutes
Surface covered
                                     254

-------
TABLE  6Q.  CHEMICAL AND BACTERIOLOGICAL SUMMARY  OF  THE  LIGHT-DARK STUDY,
           AUGUST 1-19, 1975

PARAMETER
Total MPN*
Fecal MPN*
DO, mg/1
Color
Turbidity, FTU
TSS, mg/1
VSS, mg/1
BOD,-, mg/1
COD, mg/1
TOC, mg/1
pH
*Geometric Mean
INFLUENT
(UV EFF)
(N)
4,600
(11)
370
(9)
1.1
(15)
23
(15)
1.3
(12)
6
(14)
4
(13)
2.3
(12)
28.8
(14)
13
04)
7.4
(15)

LIGHT
(OPEN)
(N)
21 ,000
(13)
3,800
(15)
1.9
(15)
28
(15)
1.3
(8)
17
(15)
14
(14)
5
' (12)
45.4
(15)
13
(15)
7.3
(14)

DARK
(COVERED)
(N)
9,300
(15)
2,000
(15)
1.7
(15)
27
(15)
1.1
(8)
17
(15)
16
(13)
5
(13)
42.8
(15),
14
(15)
7.3
(14)

                                    255

-------
fecal coliforms increased by 1.0 log and 0.7 log in the light and  dark
basins, respectively.  The increases in D.O., VSS,  and COD are indicative
of the algal masses which were present in each basin.

     Figure 138 is a time-series plot of fecal coliform populations  in
the: a) influent (solid line), b) open basin effluent  (broken line),  and
c) covered basin effluent (dotted line).  In general,  both basin effluents
had higher MPN's than the influent* and the coliform densities in  the open
basin were consistently higher than the covered basin.  (The gap from
4 August through 8 represents down-time for installation of new aeration
equipment in the No. 1 aeration basin).  Figure 139 is a plot of the paired
fecal coliform populations in the open and covered basins.  Clearly,  the
open basin had higher population in 10 out of 14 samples.

     A two-tailed t-test (15) was used to compare the  log mean coliform
values between sampling points.  Under the null hypothesis, H0 : y]  = y2>
where y-j and y2 are tne population means being compared, the t-statistic
is calculated as follows:
      t =
                1
                 1/N-,  +
                        1/N
                                  ,  where
s

a
                     9         9

                     I   +  N2S2
                           - 2
    = first sample mean with N-, data points,
    s second sample mean with N2data points,
    = sample standard deviation
    = population standard deviation
     A significance level of 0.05 was chosen to test the hypothesis.  Results
are presented in Table 61.  It is evident that two of the six tests failed
the null hypothesis, i.e., log influent total coliforms were significantly
different from log open basin total coliforms.  All other differences
were not statistically significant.  Thus, increases in total coliform
numbers likely occurred in both basins, but the increases could not be
attributed to photo reactivati on £er se.  It should be emphasized, however,
that sample sizes were relatively small and that a considerably  different
conclusion may have resulted if larger sample sizes had been used.  Further-
more, since the covered basin was not completely opaque to visible light,
some of the increase in coliform numbers in the covered basin may still  have
been due to photoreactivation.
                                     256

-------
   icr
o
o
 s_
 O)
 Q.
GO
s:
o;
o
CJ)
O
UJ
u_
-\
'.\
 :\
            V.
             \
   TO'
    10'
                                                       INFLUENT


                                                  — LIGHT BASIN



                                                   ••- DARK BASIN
                                10
                                    AUGUST
                                     15
20
        Figure  138.  Time-series plots of fecal coliform  data  for the

                     light dark study.
                                   257

-------
     10
     10
 in
 to
ca

-o
 OJ



 I  io5
O
o
o
t.
g.
8

-------
TABLE 61.   ANALYSIS OF MEAN COLIFORM  DENSITIES  IN LIGHT-DARK STUDY

Log Mean Coliform Density
Parameter n influent
Total 11 2.66
Col i forms 11 2.66

Fecal 9 2.57
Col i forms 9 2.57

n Covered n Open Std. Dev.
15 3.97 1.08
13 4.31 1.18
15 3.97 13 4.31 1.09
15 3.29 1.15
15 3.58 1.25
15 3.29 15 3.58 1.22
t
3.05*
3.43*
0.91
1.48
1.92
0.66

*Significant at 95% level.
                                    259

-------
                                 REFERENCES

 1.    Wolf, H.W.  and S.E.  Esmond;   Water Quality for Potable  Reuse  of
       Wastewater.  Water and Sewage Works, 121  (2):  48,  1974.

 2.    Wastewater  Reclamation Studies in Dallas,  Texas,  Final  Report on
       Project No. 17080 EKG, Dallas Water Utilities  Department  and
       Texas A&M Research Foundation, March 1973.

 3.    Research and Development Division Projects Receiving  Federal
       Assistance, 1969-1976, Report to the Director, Dallas Water Utilities,
       1976.

 4.    Wolf, H.W., R.S.  Safferman,  A.R. Mixson, and C.E. Stringer. -  Virus
       Inactivation during Tertiary Treatment.   Virus Survival in Water  and
       Wastewater  Systems,  J.F.  Malina Jr. and  B.P.  Sagik,  Ed.  Center
       for Research in Water Resources, The University of Texas  at Austin
       (1974).

 5.    Petrasek, A.C., Jr.  Wastewater Characterization and Process
       Reliability for Potable Wastewater Reclamation.  EPA-600/2-77-210,
       U.S. Environmental Protection Agency, Cincinnati,  Ohio, 1977, 114pp.  ,

 6.    Standard Methods  for the Examination of Water  and Wastewater, 13th  ed.,
       American Public Health Assoc., Inc., Washington,  D.C.,  1971.

 7.    Morrill, A.B.  Sedimentation Basin Research and Design.   American
       Waterworks Association Journal. 24 (9):  1442-1463,  1932.

 8.    Thomas, H.A., Jr. and R.S. Archibald. Longitudinal Mixing Measured
       by Radioactive Tracers.  Transactions of  the American Society of
       Civil Engineers,  117:839-855, 1952.

 9.    Gumbel, E.J.  Statistics of Extremes. Columbia University Press,
       New York, 1958.

10.    Benjamin, J.R. and C.A. Cornell.  Probability,Statistics, and Decision
       for Civil Engineers.   McGraw-Hill Book  Company,  New York, 1970.

11.    Parker, C.A. A New Sensitive Chemical Actinometer,  I.  Some Trials
       with Potassium Ferioxylate.   Proc. of the Royal Society of London.
       A220:104-116, 1953.

12.    Calvert, J.G. and J.N. Pitts Jr.  Photochemistry.   John Wiley and Sons,
       Inc., New York, 1967.  pp 780-813.
                                     260

-------
                              REFERENCES (continued)

13.    Roeber, J.A.  and P.M.  Hoot.   Ultraviolet Disinfection of Activated
       Sludge Effluent Discharging  to Shellfish Waters,  EPA-600/2-75-060,
       U.S.  Environmental  Protection Agency,  Cincinnati, Ohio,  1975.   85 pp.

14.    Witherell, I.E., R.  Solomon, and K.M.  Stone.   A Demonstration Project
       to Determine  the Feasibility of Using  Ozone and Ultraviolet  Radiation
       Disinfection  for Small  Community Water Systems.  In:  Proceedings  of
       AWWA Disinfection Seminar, American Water Works Association,Anaheim,
       California, 1977.

15.    Spiegel, Murray R.  Schaum's  Outline of Theory  and Problem of Statis-:
       tics.   McGraw-Hill  Book Company, New York, 1961.
                                     261

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1, REPORT NO.
 EPA-600/2-80-102
                                                           3. RECIPIENT'S ACCESSION-NO.
"'ULTRAVIOLEfDISINFECTION OF MUNICIPAL
 WASTEWATER EFFLUENTS
             5. REPORT DATE
               August 1980  (Issuing  Date)
             6. PERFORMING ORGANIZATION CODE
 f. AUTHOR(S)
 Albert  C.  Petrasek, Jr., Harold W. Wolf,
 Steven  E.  Esmond, and D. Craig Andrews
                                                           8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  Dallas  Water Utilities
  Dallas,  Texas  75201
             10. PROGRAM ELEMENT NO.
               35B1C,  D.U.  B-124, Task A/06
             11. CONTRACT/GRANT NO.
                 Grant #R-803292
12. SPONSORING AGENCY NAME AND ADDRESS
  Municipal  Environmental Research Laboratory—Cin.,OH
  Office of  Research and Development
  U.S. Environmental  Protection Agency
  Cincinnati,  Ohio  45268
                                                           13. TYPE OF REPORT AND PERIOD COVERED
              Final,  6/74-8/76
             14. SPONSORING AGENCY CODE

                 EPA/600/14
15. SUPPLEMENTARY NOTES
     Project Officer:  Albert D. Venosa (513)  684-7668
16.ABSTRACT  Dur.|ng ^s project two  different UV exposure and irradiation  systems were
studied.  The first system investigated  was the Kelly-Purdy Unit, which consisted of a
shallow-tray exposure chamber with  13  UV lamps mounted horizontally  10 cm above the
bottom  of the chamber.  This unit was  operated under varying conditions of both flow
and depth and generally provided inadequate disinfection, although fecal  coliform
densities were  usually reduced by  approximately three logs.
     The second UV system used during  the project was the Model EP-50 manufactured- by
Ultraviolet Purification Systems, Inc.   This exposure chamber consisted of a stainless
steel pressure vessel with nine UV  lamps running longitudinally through the chamber.
Each lamp was isolated from the effluent being disinfected by a quartz sleeve.   The
disinfection observed on any given  run was shown to be a function of the  UV dose, and
greater than four log reductions in fecal coliform densities were observed at times.
      During this project three special  virus studies were conducted.  The influent
to the  EP-50 was seeded with an F2 coliphage and an attenuated Type I poliovirus.
During  the  three virus runs, viral  and phage densities in the influent and effluent
of the  UV irradiation chamber were  monitored.  The observed reductions in viruses were
correlated  with UV doses, and the F2 co'liphage response to the UV disinfection  process
was similar to the response of the  Type  I poliovirus.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS  C. COS AT I Field/Group
 Disinfection
 Ultraviolet radiation
 Coliform bacteria
 Wastewater, waste,  treatment
 Microorganism control, bactericides
 Dallas, TX;
 UV absorbance;  Wastewatei
 quality; Nitrified
 effluent; Viruses;
 Poliovirus;  Bacteriophagfi
         13B
18. DISTRIBUTION STATEMENT

  Release to Public
19. SECURITY CLASS (ThisReport)'
  Unclassified
21. NO. OF PAGES

    282
20. SECURITY CLASS (Thispage)
  Unclassified
                                                                        22. PRICE
EPA Form 2220-1 (9-73)
                                            262
                                                          U.S. GOVERNMENT PRINTING OFFICE: 1980--657-165/0122

-------