'¦
i	i '	'	;
^	MAY 15	!
— — — — — —	_ __^__	_; ' ' ' "-V'	I |






ITTLETON
BELLEV1EVV i,
of j
> \ Vy
4 % >
\ \ V UNION sr. \
\vJw \ A
Denver
!	1RONMEN
SURVEILLANCE AND ANALYSIS DIVISION
TECHNICAL SUPPORT BRANCH
!	
AUGUST 1972

-------
ft 32M,:lM
90%-<-73.-00(0	(Z\i
ot^i
S&A/TSB-3	C ' I
EVALUATION OF THE EFFECTIVENESS OF CHLORINATION
AT THE LITTLETON WASTEWATER TREATMENT PLANT
LITTLETON, COLORADO
May 15-23, 1972
U S EPA Region 8 Library
80C-L
999 181 h SI , Suite 500
Denver, CO 80202-2466
TECHNICAL SUPPORT BRANCH
SURVEILLANCE AND ANALYSIS DIVISION
U. S. ENVIRONMENTAL PROTECTION AGENCY
REGION VIII
August 1972

-------
TABLE OF CONTENTS
Page No.
I. Introduction 		1
II. Description of Plant 		1
III. Applicable Water Quality Criteria 		1
IV. Sampling Procedure 		3
A.	Method for Determination of Length of Contact
T i'me		3
B.	Method for Evaluation of Effectiveness of
Chiorination 		4
V. Analysis of Results 		6
A.	Relationship of Col iforms to Flow		6
B.	Relationship of Chlorine Residual to Chlorine
Dosage 		6
C.	Relationship of Coliform Concentration to
Chlorine Dosage 		13
D.	Relationship of Coliform Concentration to
Chlorine Dosage and Contact Time 		13
VI. Summary and Conclusions		17
VII. Recommendations		17
Appendix A - Survey Data		21
Appendix B - References		26
i

-------
LIST OF TABLES
Page No.
Table
1. Recommended Chlorine Loadings for Various Flows
with Present Treatment Plant Design 	 20
A-l Raw Data for Contact Time Measurement	 21
A-2 Tabulation of Chlorination and Bacteriological Data . 22
i i

-------
LIST OF FIGURES
Figure	Page No.
1.	Plant Flow Schematic		2
2.	Flow vs. Contact Time		5
3.	Concentration of Total Coliforms vs. Flow 		7
4.	Concentration of Fecal Coliforms vs. Flow 		8
5.	Typical Breakpoint Chlorination Curve 		10
6.	Chlorine Residual vs. Chlorine Dosage 		11
7.	Concentration of Total and Fecal Coliforms vs.
Chlorine Dosage		14
8.	Concentration of Total and Fecal Coliforms vs.
Chlorine Dosage Times Contact Time 		16
i i i

-------
I. Introduction
Region VIII of the United States Environmental Protection
Agency developed an Accomplishment Plan for the Metropolitan
Denver-South Platte River Basin areas. As a result of implen-
tation of this plan a study was made at the Littleton, Colorado,
Wastewater Treatment Plant. The purpose of this study is to
evaluate the effectiveness of chlorination at the Littleton
plant in providing satisfactory disinfection before discharge
to the South Platte River. An evaluation was also made of chlorine
residuals downstream from the Littleton outfall.
11. Description of Plant
A schematic diagram of Littleton's high rate trickling
filter wastewater treatment plant is shown in Figure 1. The
two 30 foot primary clarifiers and the No. 1 and No. 2 digesters
are operational but are not in service. The average daily flow
from January through May, 1972, was 4.25 MGD. During the survey
the highest recorded flow (peak flow during any one day) was
6.19 MGD.
Disinfection with chlorine is used at the Littleton
facility to meet the Colorado Water Quality Standards. The
plant effluent passes through a Parshall flume and chlorine
(the amount of chlorine varies with flow) is applied. The
effluent enters the outfall line, passes through an inverted
siphon, and flows to the river. The outfall is approximately
2,000 feet downstream from the treatment plant.
Presently the operational mode at Littleton, with respect
to chlorination, is to adjust the chlorine dosage to obtain a
chlorine residual at the outfall of 0.2 mg/1. In most cases a
chlorine dosage of about 3.5 mg/1 is needed to maintain the
desired 0.2 mg/1 chlorine residual at the outfall.
III. Applicable Water Quality Criteria
The bacteriological standard that applies to the South
Platte River for the reach that receives wastewater effluent
from the Littleton facility requires the log mean of fecal
coliform organisms to be less than 1000 per 100 ml. and no
more than 2000 per 100 ml. in greater than ten (10) percent
of the samples collected in any thirty (30) day period. In
addition, the effluent must be free from biocides, toxic,
or other deleterious substances in levels, concentrations,
or combinations sufficient to be harmful to aquatic life.
1

-------
Figure 1
Evaluation of Chlorination at the
Littleton Wastewater Treatment Plant
May 1972
Plant Flow Schematic
2

-------
IV. Sampling Procedure
The primary purpose of disinfection is the destruction of
all pathogenic organisms. Since pathogenic organisms are few
in number and very difficult and time-consuming to isolate and
identify the coliform group of organisms are used as indicators
of contamination. Total coliform organisms include those found
in the intestinal tract of human beings and other warmblooded
animals, and in plants, soil, air, and the aquatic environment.
Fecal coliforms include only those found in the intestinal tract
of humans and warmblooded animals. Bacteriological samples were
collected during this study and were analysed for both the total
and fecal coliform group of .nicroorganisms.
Two factors which are extremely important in disinfection
are length of contact time and concentration of chlorine. At
Littleton, the length of contact time is equal to the time after
the chlorine is added until the flow is discharged from the out-
fall line. This contact time varies inversely with flow (i.e.,
a higher flow yields a shorter contact time).
Concentration of chlorine could be represented by either
the amount of chlorine added to the plant effluent (chlorine
dosage) or the amount of chlorine remaining at the end of the
contact time (chlorine residual). Normally chlorine residual
is used to represent concentration of chlorine. However, for
this report chlorine dosage was chosen to represent concentration
of chlorine since this discussion is directed to operation of the
disinfection unit and chlorine dosage is the parameter most
easily controlled by plant personnel.
A. Method for Determination of Length of Contact Time
The length of contact time in Littleton's outfall line
was approximated by conducting several dye studies. Approxi-
mately two (2) milliliters of rhodamine-WT dye was added to
the outfall line at the point of chlorine application to the
effluent flow. Grab samples were taken at the outfall every
fifteen (15) seconds. The fluorescence of each sample as
measured by a fluorometer was attained. The contact time
was assumed to be equal to that time from when the dye was
added until the peak fluorescence was achieved. The exact
peak fluorescence and thus contact time was calculated
through interpolation.! (See Table A-l in Appendix A for
data.)
^ It was assumed that the fluorescence, if it was continually
monitored, would follow a normal distribution. Thus, standard
interpolation methods were followed.
3

-------
The relationship between length of contact time and
flow is shown in Figure 2. A line of best fit was drawn
through the data and shows that for an average daily flow
of 4.25 MGD, the corresponding contact time would be about
18 minutes. When discussing disinfection, high flows are
most critical because contact time is shortest (as shown
in Figure 2), hence emphasis will be on the higher flows
in this discussion.
B. Methods for Evaluation of Effectiveness of Chlorination
To determine the effectiveness of chlorination, the
following factors were evaluated: flow, chlorine dosage,
chlorine residual, and total and fecal coliform concentra-
tion. Two (2) sampling locations were chosen, one at a
point just prior to where the chlorine was applied and one
at the outfall. Sampling locations upstream and downstream
from the outfall were also monitored.
Bacteriological samples of the plant effluent were taken
just prior to chlorination. Plant effluent flow measure-
ments were made at the Parshall flume located near the
chlorination building. Chlorine loadings (lb./day) were
read at the chlorinator located in the chlorination build-
ing and were recorded.2
The chlorine dosage was calculated using the actual chlorine
loading and flow data. To determine the effect of chlorine
dosage on the kill of total and fecal coliform organisms, the
chlorine dosage was varied.
The time of sampling at the outfall was equal to the time
of the initial sample (the sample just prior to chlorination)
plus the time of flow (contact time) as determined by the dye
studies. Bacteriological samples and total chlorine residuals
were taken simultaneously at the outfall. Total and fecal
coliform concentrations were determined from the bacteriologi-
cal samples using the membrane filter test. Chlorine residuals
were measured with the color comparator.
Periodic bacteriological samples and total chlorine residuals
were taken downstream from the outfall. Bacteriological samples
were also taken at an upstream station and examined for total
and fecal coliform concentration. (See Table A-2 in Appendix
A for data.)
^ The actual amount of chlorine used each day is measured by plant
personnel by determining the daily change in weight of the two
(2) ton chlorine gas cylinders. The chlorine loading which was
read at the chlorinator was compared to the actual amount of
chlorine used each day. It was determined that the chlorine
loading read at the chlorinator was too high by a factor of 0.878.
Therefore, the actual chlorine loading is equal to the chlorine
loading read at the chlorinator times 0.878.
4

-------
14"30	15-0	15-30	16-0	16-30	17-0	17-30	18-0
Contact Time - Minutes - Seconds

-------
V. Analysis of Results
Analysis of the data led to the establishment of various
relationships between the measured parameters. These relation-
ships are outlined below.
A.	Relationship of Coliforms to Flow
The number of coliform organisms remaining after disinfection
is the indicator used to determine the effectiveness of the
chlorination system. The primary factors which influence the
effectiveness of chlorination are length of chlorine contact
time and chlorine dosage.
The concentration of coliform organisms that are to be
disinfected would influence the effectiveness of chlorination.
Although a specific relationship between total coliform con-
centration and effluent flow was not determined because of
the limited data available, a general trend is indicated by
the dotted line in Figure 3. As flow increases, the concen-
tration of total coliforms also increases requiring destruction
of more total coliforms at higher flow rates to achieve an
equally low coliform concentration after chlorination.
Figure 4 shows that the concentration of fecal coliform
organisms neither significantly increases nor decreases with
increased flow, although a slight decrease may be interpreted
as indicated by the dotted line. Since there is not a signifi-
cant decrease in the concentration of fecal coliform organisms
with increased flow, nearly the same number of fecal coliforms
would require destruction for both low and high flows to achieve
an equally low coliform concentration after chlorination.
The greatest concentration of total coliforms prior to
chlorination exists with higher flow rates. The concentra-
tion of fecal coliforms remains nearly the same for both low
and high flow rates. It is concluded that the most critical
time to achieve effective chlorination occurs at higher flow
rates because both high concentrations of coliforms and short
chlorine contact times exist.
B.	Relationship of Chlorine Residual to Chlorine Dosage
As chlorine is applied to the plant effluent, various
reactions take place between the chlorine and bacteria,
inorganic and organic compounds (except ammonia and other
nitrogenous compounds), and many other substances in the
water. These reactions tie up chlorine making it ineffective
for further disinfection. The chlorine that is tied up and
no longer useful for disinfection purposes is called chlorine
demand. When the chlorine demand of the effluent is satisfied,
further chlorine dosages remain in the effluent in the form of
chlorine residual.
6

-------

-------

-------
The chlorine residual exists in various forms depending
upon the chlorine dosage applied. At lower chlorine dosages,
below what is called breakpoint chlorination, the chlorine
dosage reacts with the ammonia and other nitrogenous compounds
in the effluent and forms monochlorimines, dichlorimines , and
trichlorimines. These chlorimine compounds are disinfectants,
with the monochlorimines and dichlorimines having most of the
disinfectant power, and are called collectively combined avail-
able chlorine. At higher chlorine dosages, above breakpoint
chlorination, the chlorine residual exists in the form of
hypochlorite ions (0CI~) or hypochlorous acid (HOC!) and is
called free available chlorine.
Figure 5 shows a typical relationship between chlorine
dosage and chlorine residual. Initially the chlorine dosage
is expended as. chlorine demand. After the chlorine demand
is satisfied, the chlorine reacts with ammonia and other
nitrogenous compounds and forms combined available chlorine.
As the chlorine dosage nears the breakpoint, all ammonia is
converted to trichlorimines or further oxidized to free
nitrogen and other gases and the chlorine residual decreases.
Increased chlorine dosages past the breakpoint yields chlorine
residuals in the free available form (1).
Figure 6 shows the relationship between the amount of chlorine
applied to Littleton's treatment plant effluent (chlorine dosage)
and the chlorine residual concentration at the plant outfall.
Also, the relationship between the chlorine applied to the plant
effluent and the chlorine residual in the South Platte River at
a distance of 270 yards downstream from the plant outfall is
shown. The chlorine residual data shown in Figure 6 was grouped
according to the chlorine dosage applied. Increments of 0.5 mg/1
of chlorine dosage was chosen. The arithmetic average chlorine
residual within each 0.5 mg/1 increment of chlorine dosage was
plotted.
Figure 6 shows that as the chlorine dosage increased the
chlorine residual at the plant outfall also increased except
at chlorine dosages of between 4-. 75 and 5.25 mg/1 (average
chlorine dosage of 5.0 mg/1). At this average chlorine dosage
of 5.0 mg/1, the chlorine residual decreased as in the typical
breakpoint chlorination curve.
Figure 6 also shows that as the chlorine residual in the
plant outfall increased or decreased, the chlorine residual
in the river also increased or decreased, except at the
7.5 mg/1 dosage. The lower chlorine residual for this dosage
9

-------
Predominantly Combined
Available Chlorine Residual
7 \
/ \
/
Figure 5
Evaluation of Ghlorination at the
Littleton Wastewater Treatment Plant
May 1972
Typical Breakpoint Chiorination Curve
/
Predominantly Free Available
Chlorine Residual
/
/


-Breakpoint
Chiori nati on
mq/1

-------
Figure 6
Evaluation of Chiorination at the
Littleton Wastewater Treatment Plant
May 1972
Chlorine Residual vs. Chlorine Dosagel
Reading at the Outfall
Reading in South Platte
River 270 yards downstream
from Outfal1
Reading taken during
higher river flow
+->
03
O
o
Q.
rO
0J
S-
CG
~Z>
aj
+->
n3
E
"O
O)
a>
o
c_j

rc
Q
"O
aj
u
CD
o
03
Q
o
2:
"O
a)
u
aj
o
4->
O
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
Chlorine Dosaqe - mg/1
11

-------
is due to the fact that the river flow was higher during
the time that this reading was taken due to heavy rains;
therefore, the residual concentration was affected by
dilution. At a chlorine dosage of 7.54 mg/1 there was a
trace chlorine residual a distance of one mile downstream
from the Littleton outfall.
Using chlorine to disinfect treated wastewater effluents
necessitates the existence of a chlorine residual at the
outfall, unless the chlorine is purposely removed by special
treatment. This chlorine residual at the outfall causes a
chlorine residual in the river which may be deleterious to
fish life under conditions of high effluent flow and corres-
ponding low river flow.
Studies indicate that the lethal concentration of chlorine
varies with different species of fish. Free available chlorine
concentrations of 0.03 mg/1 have been reported to have killed
rainbow trout, whereas concentrations of 0.1 mg/1 have been
reported to have not harmed trout. Concentrations of 0.15 to
0.2 mg/1 have killed the more tolerant fish species, carp,
whereas, concentrations of 1.0 mg/1 have been reported to have
not harmed carp. The wide discrepancy in the above examples
for each species of fish can be attributed to other factors
such as pH, temperature, dissolved oxygen, and the synergism
and antagism of other pollutants markedly affecting the toxic-
ity of chlorine toward fish. The examples do show that the
less tolerant species of fish are affected by lower concen-
trations of chlorine residual. Studies also indicate that in
some instances combined available chlorine is more toxic toward
fish than free available chlorine and other studies show the
opposite is true (2). In any event, it may be said that rel-
atively small concentrations of chlorine can be detrimental
to fish life; hence, every effort should be made to maintain
as low a chlorine residual as possible in the outfall and still
maintain the bacteriological water quality standards.
Operating personnel at Littleton adjust the chlorine
dosage to obtain a chlorine residual of about 0.2 mg/1 at the
outfall. This 0.2 mg/1 residual produces a residual in the
river a distance of 270 yards downstream, of about 0.05 mg/1.
It should be noted that these samples were taken during,
relatively high river flow which was due to heavy rains.
The exact quantity of river flow was not measured, but was
approximately one (1) inch over the top of the face of the
Englewood Water Supply Dam at Union Avenue. It would be
12

-------
expected that the chlorine residual in the river would be
higher if the river flow was lower and would consequently
increase the chances of having conditions {higher chlorine
residual) deleterious to fish, especially the less tolerant
species.
C.	Relationship of Coliform Concentration to Chlorine Dosage
Disinfection with chlorine is used by Littleton to meet
the bacteriological requirement established by the Colorado
Water Quality Standards, Figure 7 shows the relationship
between chlorine dosage and the concentration of total and
fecal coliform organisms. For Figure 7 the coliform data
was grouped according to the chlorine dosage. Increments
of 0.5 mg/1 of chlorine dosage increment was grouped and
the logarithmic average coliform concentration was deter-
mined. Colorado state standards for coliform concentrations
are based upon logarithmic averages.
Figures 6 and 7 show that the concentration of coliform
organisms corresponds inversely with the chlorine residual.
For example, Figure 6 showed that at breakpoint chlorination
there was a decrease in the chlorine residual. Correspond-
ingly, Figure 7 shows that the concentration of coliform
organisms increased at breakpoint (i.e., 5 mg/1 chlorine
dosage), as well as at other points of lower chlorine
residual (i.e., 3.5 mg/1 chlorine dosage).
Figure 7 also shows that at a chlorine dosage of about
3.5 mg/1, the log average fecal coliform concentration is
300 per 100 ml. (NOTE: 3.5 mg/1 is the approximate normal
chlorine dosage provided at Littleton.) This concentration
of 300 fecal coliforms per 100 ml at the outfall decreased
to 52 fecal coliform per 100 ml in the stream at the sampling
station 270 yards downstream, meeting the state's bacterio-
logical stream standards. If the stream standards are changed
to effluent standards and the limit set at 100 fecal coliforms
per 100 ml, additional disinfection will be required. The
amount of additional disinfection required is discussed in the
following section (Section D).
D.	Relationship of Coliform Concentration to Chlorine Dosage
Times Contact Time
Both chlorine contact time and chlorine dosage are very
important in disinfection. Where other factors are constant,
the disinfecting action or the kill of harmful organisms and
coliform organisms is directly proportional to the chlorine
dosage times the chlorine contact time.
13

-------
,000
900
800
700
600
500
400
300
200
100
0

-------
Figure 8 shows the relationship between the product of
chlorine dosage times contact time and total and fecal
coliform concentrations at the Littleton outfall. Data was
grouped in increments of 10 according to the product of
chlorine dosage times contact time. Coliform data within
each increment of 10 was grouped and the logarithmic
average coliform concentration was determined. Figure 8
shows that more coliforms are killed (i.e., fewer coli-
forms are present at the outfall) as the product of chlorine
dosage times contact time is increased until the breakpoint
is achieved.
Using Figure 8, various combinations of chlorine dosages
and contact times can theoretically be determined to obtain
a given coliform count at the outfall. For economic reasons
the lowest possible factor to obtain the desired or required
coliform count should be used. The lowest factor would re-
quire a minimum amount of chlorine and a minimum contact
time resulting in a lower operating cost and a smaller chlor-
ine contact basin.
If the present Colorado Water Quality Standards are
modified to effluent standards and are upgraded to require
a logarithmic average fecal coliform concentration at the
outfall of less than 100 per 100 ml., the product of the
chlorine dosage times contact time must be at least 65.
However, the following limitations apply: the chlorine
dosage must be greater than the chlorine demand and the
chlorine dosage must be either above or below that required
for breakpoint chlorination.
To obtain the product of 65 under present operating
conditions (i.e., with a minimum contact time of 14
minutes) the chlorine dosage would have to be 4.64 mg/1.
This high chlorine dosage would raise the chlorine res-
idual at the outfall to about 0.75 mg/1. This residual at
the outfall would raise the chlorine residual in the river
(270 yards downstream) to about 0.1 mg/1 which would be
more detrimental to fish life than the present river chlorine
residual of 0.05 mg/1.
A better combination of chlorine dosage and contact time
to obtain a product of at least 65, and a fecal coliform
concentration of less than 100 per 100 ml., would be to
use the present chlorine dosage of 3.5 mg/1 and increase
the chlorine contact time to 19 minutes. This combination
would require construction of a chlorine contact basin at
Littleton since a 19 minute contact time is not available
with present facilities. With even longer contact times,
it may be possible to use lower chlorine dosages. Lower
15

-------
55	65	75	85	95	105
Chlorine Dosaqe Times Contact Time - mq/1 X Minutes
16

-------
chlorine dosages would reduce the chlorine residual going
to the river. In turn the chlorine residual in the river
would be reduced to a level that would be less detrimental
to fish and other aquatic life. A reduction in the daily
cost of chlorination would also occur. The construction
of a chlorine contact basin would apparently add much more
flexibility to the operation of the disinfection unit at
the Littleton treatment plant.
Summary and Conclusions
For the average daily flow of 4.25 MGD at the Littleton
wastewater treatment plant, the length of chlorine contact
time is about 18 minutes. During the survey, the highest
recorded flow during any one day was 6.19 MGD and the resulting
contact time was about 14 minutes. Since length of contact
time is one of the two most important factors (chlorine dosage
is the second factor) in disinfection and since high flows
yield shorter contact times, the effectiveness of chlorination
at Littleton was studied during periods of hiqh plant effluent
flow.
The concentration of total coliform organisms after
secondary treatment and prior to chlorination increased as
the effluent flow increased. The concentration of fecal
coliform organisms did not significantly increase or decrease
as the flow increased. The concentration of coliform organisms
that are to be disinfected would directly influence the effective-
ness of chlorination. The most critical time to achieve effective
chlorination occurs at higher flow rates because both high concen-
trations of coliforms and short chlorine contact times exist.
Presently the operational mode at Littleton, with respect to
chlorination, is to adjust the chlorine dosage to obtain a chlorine
residual at the outfall of 0.2 mg/1. In most cases a chlorine
dosage of about 3.5 mg/1 is needed to maintain the desired 0.2 mg/1
chlorine residual.
During this survey, the chlorine dosage was adjusted from
3.41 mg/1 to 7.54 mg/1 to determine the effect of various chlorine
dosages on total and fecal coliform concentrations at the plant
outfall. The chlorine residuals were monitored at the outfall and
in the South Platte River 270 yards downstream from the plant out-
fall. The chlorine residual at the plant outfall and in the river
increased as the chlorine dosage increased except at breakpoint
chlorination which occurred at a chlorine dosage of about 5.0 mg/1.
17

-------
The present operational mode at Littleton with respect to
chlorination (a chlorine dosage of about 3.5 mg/1) gives a
chlorine residual at the outfall of about 0.2 mg/1 and a
chlorine residual in the South Platte River 270 yards down-
stream of about 0.05 mg/1. The chlorine residual in the
river increased or decreased inversly as the flow of the
river increased or decreased. The 0.05 mg/1 chlorine residual
occurred when the river flow was relatively high. Literature
reveals that a chlorine residual of 0.05 mg/1 may have a
detrimental effect on fish life, especially the less tolerant
species of fish. If the chlorine dosage would be increased,
thus increasing the chlorine residual at the outfall, or the
river flow would decrease, the chlorine residual in the river
would be raised thus increasing the possibility of an even
greater detrimental effect on fish life.
To determine the effectiveness of chlorination with varied
chlorine dosages, bacteriological samples were taken at the
plant outfall and total and fecal coliform tests were run on
these samples using the membrane filter test. The total and
fecal coliform concentration at the outfall varies inversely
with the chlorine residual at the outfall. The present opera-
tional mode at Littleton with respect to chlorination gave a
logarithmic average of 300 fecal coliforms per 100 ml. This
average effluent discharge did not cause a violation of the
Colorado Water Quality Bacteriological Standard of 1000 fecal
coliform per 100 ml in the South Platte River. For the chlorine
dosages that were studied, dosages of about 5.0 mg/1, breakpoint
chlorination, yielded the highest total and fecal coliform
concentration (10,000 per 100 ml. and 350 per 100 ml. respectively)
in the effluent.
With the ever increasing emphasis on upgrading the quality
of our rivers and streams, it is possible that the bacteriological
water quality standards may be modified to include effluent stand-
ards which may require a logarithmetric average of 100 fecal coli-
forms per 100 ml. At Littleton, additional disinfection would be
required to meet this higher standard. Two parameters, chlorine
dosage and contact time, may be adjusted to obtain the additional
disinfection. With the present design at Littleton (i.e., a mini-
mum chlorine contact time of 14 minutes) the chlorine dosage
required to reach the 100 fecal coliform per 100 ml. concentration
would be about 4.64 mg/1. This high chlorine dosage would raise
the chlorine residual at the outfall to about 0.75 mg/1. The
residual of 0.75 mg/1 at the outfall would raise the chlorine
residual in the river (270 yards downstream from the outfall) to
about 0.1 mg/1 which would be more detrimental to fish life than
the present river chlorine residual of 0.05 mg/1.
18

-------
A better combination of chlorine dosage and contact time to
achieve the additional disinfection and decrease the chlorine
residual in the river would be to increase the chlorine contact
time and decrease the chlorine dosage. This combination would
reduce the chlorine residual at the outfall. In turn, the chlorine
residual in the river would be reduced to a level that would be
less detrimental to fish and other aquatic life. Additionally,
a reduction in the daily cost of chlorination would occur. However,
the longer contact time necessitates the construction of a chlorine
contact basin.
VII. Recommendations
The following recommendations are made:
1.	If the present Colorado Water Quality Standards are
modified to include effluent standards which require
100 fecal coliforms per 100 ml. in the plant effluent,
a chlorine contact basin would be necessary to provide
the most satisfactory disinfection at the Littleton
plant.
2.	Although higher chlorine dosages (up to breakpoint
chlorination and after breakpoint chlorination) would
achieve better disinfection, present objectives, to
include meeting present Colorado Water Quality Standards
and maintaining a low chlorine residual in the river,
will be best achieved by maintaining a chlorine dosage
of 3.5 mg/1. Therefore, a dosage of about 3.5 mg/1
should be continued at the Littleton treatment plant.
To obtain a chlorine dosage of 3.5 mg/1 the following
chlorine loadings (lb./day) at various flow rates should
be followed: See Table I.
3.	The scale for measuring chlorine load released by the
chlorinator at Littleton gives a reading which is
0.878 times larger than the actual chlorine load.
The chlorine load scale should be corrected to show
the actual chlorine loading.
4.	At Littleton a chlorine dosage between 4.75 and 5.25 mg/1
is not as effective for disinfection purposes as the
lower chlorine dosage of 3.5 mg/1. A chlorine dosage
between 4.75 and 5.25 mg/1 is in the breakpoint chlorina-
te range. For more effective disinfection the chlorine
dosage should not be set between 4.75 and 5.25 mg/1.
19

-------
TABLE I
Evaluation of Chlorination at the Littleton
Wastewater Treatment Plant
May 1972
Recommended Chlorine Loadings for Various
Flows with Present Treatment Plant Design
*Requi red
Recommended	Setting On
Chlorine	Chlorine	Chlorinator
Dosage	Flow	Loading	at Littleton
(mg/1)	(MGD)	(lb./day)	(lb./day)
3.5	3.5	102	116
3.5	4.0	117	133
3.5	4.5	131	150
3.5	5.0	146	166
3.5	5.5	160	183
3.5	6.0	175	200
* The chlorine load setting on the chlorinator at Littleton must be
greater than the recommended chlorine loading by a factor of 0.878.
20

-------
APPENDIX A
Survey Data

-------
TABLE A-l
Evaluation of Chlorination at the
Littleton Wastewater Treatment Plant



May 1972





Raw Data for
Contact Time
Measurement


Date
Time
Parshall
F1 ume
Head
Parshall
Flume
Width
Flow

Contact
Time


(Ft.)
(In.)
(MGD)
(Min.) (Sec.)
May 16
1040
1.27
18
5.60
15
9
May 16
1160
1.27
18
5.60
15
9
May 16
1355
1.19
18
5.07
15
36
May 16
1510
1.14
18
4.74
17
1
May 17
0800
1.05
18
4.18
17
45
May 17
1100
1.30
18
5.81
15
0
May 17
1411
1.11
18
4.55
16
20
May 18
1007
1.33
18
5.97
14
18
May 18
1100
1.27
18
5.60
15
0
May 22
1026
1.35
18
6.12
14
7
May 22
1320
1.20
18
5.13
15
1
May 23
0922
1 .39
18
6.19
14
19
May 23
_ 1044
1.24
18
5.38
14
55
May 23
1411
1.09
18
4.44
16
45
May 24
0945
1.20
18
5.13
14
30
y
May 24
1015
1.31
18
5.85
14
33
May 24
1110
1.23
18
5.32
15
4
May 24
1306
1.16
18
4.85
16
4
21

-------
ro
ro
TABLE A-2
EVALUATION OF CHLORINATION FOR THE
LITTLETON WASTEWATER TREATMENT PLANT
MAY 1972
Tabulation of Chlorination and Bacteriological Data
Chlorine Chlorine Contact	Total	Fecal
Station
Date
Time
Flow
Dosaqe
Residual

Time
Coli form
Coli form



(MGD)
(MG/L)
(MG/L)
(Min
) (Sec)
Col/100 ml
Col/100 ml
I
5/16
1025
5.58
3.78
-

-
8,600,000
1 ,100,000
II
II
1039
5.58
-
1.2
14
50
1,500
50
I
II
1145
5.58
4.15
-

-
4,900,000
310,000
II
II
1200
5.58
-
0.5
14
50
1 ,200
57
I
II
1455
4.78
4.41
-

-
4,600,000
940,000
II
II
1515
4.78
-
0.8
15
50
4,200
40
III
II
1520
-
-
0.1

-
970
15
I
5/17
0955
5.65
4.48
-

-
6,100,000
520,000
II
II
1010
5.65
-
0.9
14
44
1 ,300
100
I
II
1100
5.78
4.73
-

-
4,400,000
870,000
II
II
1115
5.78
-
0.5
14
36
3,000
40
I
II
1145
5.25
5.02
-

_
5,600,000
1 ,200,000
Remarks

-------
TABLE A-2 (Cont.)
Chlorine Chlorine Contact
Station Date Time Flow Dosage Residual	Time
TMGDJ (MG/1J	(MG/L) (Min) (Sec)
II 5/17 1200 5.25	-	0.25	15 12
I	11 1345 4.71 5.14
II	" 1400 4.71	-	TR	16	4
I
5/17
1555
4.38
5.53
-

-
II
5/17
1615
4.38
-
0.4
17
30
III
5/17
1620
-
-
0.1

-
I
II
2047
5.18
5.48
-

-
II
II
2105
5.18
-
1.5
15
17
I
II
2130
5.25
5.82
-

-
II
II
2150
5.25
-
1.4
15
11
III
II
2155
-
-
0.5

-
I
5/18
0910
5.72
6.08
_


2.
Total	Fecal
Coliform Col iform	Remarks
Col/100 ml Col/100 ml
10,000	350
4,500,000 1,200,000
300	30 Questionable
Bacteriological
Data
5,300,000 1,000,000
10,000	180
6,100,000 1,400,000
660	20
6,000,000	580,000
760	20
7,300,000
540,000

-------
TABLE A-2 (Cont.)
Chlorine Chlorine	Contact Total	Fecal
Station Date Time Flow Dosage Residual	Time	Coliform Coliform	Remarks
TMGD) (MG/L) (MG/L) (Min) (Sec) Col/I00 ml Col/I00 ml
II	5/18 0930 5.72	-	1.5	14 41	2,000	60
III	" 0935 -	-	0.25	-	Stream Flow Higher
Than Previous Days
I	" 1007 5.92 7.11	-	- 7,700,000 640,000
II	" 1023 5.92 -	1.7	14 27	570	20
I	5/18 1100 5.58 7.54	-	- 6,300,000 1,000,000
II	5/18 1120 5.58 -	2.0	14 48	640	100 Questionable
Bacteriological
Data
III	" 1200	0.2
IV	" 1215	TR	-
I	5/22 0927 5.92 3.46	-	- 7,200,000 950,000 Stream Flow Higher
Than May 18
II	5/22 0942 5.92	-	0.6	14 28	5,200	270
V	" 0950 -	-	-	200	170
III " 1000 TR - 830 160
I	" 1041 6.12 3.45	-	- 11,000,000 1,500,000
II	" 1055 6.12 -	0.3	14 16	44,000	890

-------
TABLE A-2 (Cont.)
Station	Date	Time	Flow
TmgdT
I	5/22	1405	4.78
II	"	1421	4.78
I	5/23	0920	6.19
II	"	0935	6.19
III	"	0955
I	"	1131	5.38
II	"	1145	5.38
V	"	1152
IV	"	1211
I	"	1411	4.45
II	"	1424	4.45
Chlorine
Dosage
(MG/L)
3.86
3.41
3.52
4.26
Chlorine
Residual
(MG/L)
0.4
0.3
0.1
0.3
TR
0.8
Contact	Total
Time	Co1iform
(Min) (Sec) Col/100 ml
15
55
14 12
15 3
17 0
4,000,000
52,000
5,500,000
15,000
860
5,200,000
1,800
170
770
5,600,000
760
Fecal
Coli form
Col/100 ml
950,000
50
750,000
470
30
740,000
80
40
100
1 ,200,000
40
Remarks
Stream .Flow Same
As May 22
Station I - Sewage Treatment Plant Effluent - Prior to Chlorination
Station II - Sewage Treatment Plant Effluent - At the Outfall
Station III - South Platte River - 270 Yards Downstream from Littleton Outfall
Station IV - South Platte River at Oxford Street Bridge (1 Mile Downstream from Littleton Outfall)
Station	V - South Platte River - Upstream from Littleton Outfall

-------
APPENDIX B
References

-------
1.	Sawyer, Clair N., Chemistry for Sanitary Engineers, McGraw Hill
Book Company, Inc., New York, New York, (1960) pp. 246-256.
2.	McKee, Jack E. and Wolf, Harold W., Water Quality Criteria,
California State Water Resources Control Board, Sacramento,
California.
26

-------